73068 Jobs About the cover In almost every language there is a range of words related to jobs, each emphasizing a different angle. Some words hint at the nature of the activity being performed, evoking the skill or expertise that is required. Others refer to the volume of human inputs used in production, bringing images of effort and con- veying a sense of physical exertion. There are also words asso- ciated with the sheer numbers of people engaged in economic Arabic Mapuche Bulgarian Malagasy activity, which are more easily associated with aggregate sta- Indonesian/Malaysian Lithuanian Welsh Bahasa Tagalog tistics. In other cases, what seems to be at stake is a contrac- French Finnish Icelandic Ga Chinese Korean tual relationship, involving mutual obligations and a degree of Italian Thai stability. In some languages, there are even words to designate Urdu Swedish Greek Guaraní Georgian the place where the person works, or at least a slot in a produc- Burmese German Hebrew Tswana tion process. This multiplicity of words clearly shows that jobs English Croatian Maori Basque are multi-dimensional and cannot be characterized by a single Portuguese Bengali Mohawk Farsi term or measured by a single indicator. Swahili Romanian Afrikaans Hindi Yorùbá Words related to jobs do not always translate well from one Russian Shona Portuguese Tibetan Polish language to another, as the range of options available in each Tamil Hungarian Aymara Ukrainian Zulu Kirundi Dutch case can be different. If languages shape thinking, there are Quechua Turkish Vietnamese times when the ways in which people refer to jobs seem to be Romansh Albanian Juba Arabic Roma Gaelic Tajiki at odds. Gaps probably arise from the different characteristics Amharic Spanish Dinka Japanese Roma of jobs being emphasized in different societies. They also sug- gest that jobs’ agendas can differ across countries. Galician In many languages, words related to jobs serve not only as common nouns but also as proper nouns. Throughout his- tory family names have been associated with specific skills or trades: Vankar in Hindi, Hattori in Japanese, Herrero in Span- ish, or Mfundisi in Zulu, just to mention a few. The use of job-related words as household identifiers shows that people associated themselves with what they did. Nowadays, people aspire to choose their jobs based on what motivates them and on what could make their lives more meaningful. In almost every language there are also several words to express the lack of a job. Almost invariably these words have a negative connotation, close in spirit to deprivation; at times they even carry an element of stigma. In all these ways, language conveys the idea that jobs are more than what people earn, or what they do at work: they are also part of who they are. Jobs © 2012 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved 1 2 3 4 15 14 13 12 This work is a product of the staff of The World Bank with external contributions. Note that The World Bank does not necessarily own each component of the content included in the work. The World Bank therefore does not warrant that the use of the content contained in the work will not infringe on the rights of third parties. The risk of claims resulting from such infringement rests solely with you. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 Unported license (CC BY 3.0) http://creativecommons.org/licenses/ by/3.0. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for com- mercial purposes, under the following conditions: Attribution—Please cite the work as follows: World Bank. 2012. World Development Report 2013: Jobs. Washington, DC: World Bank. DOI: 10.1596/978-0-8213-9575-2. License: Creative Commons Attribution CC BY 3.0. Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. All queries on rights and licenses should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Softcover ISSN: 0163-5085 ISBN: 978-0-8213-9575-2 e-ISBN: 978-0-8213-9576-9 DOI: 10.1596/978-0-8213-9575-2 Hardcover ISSN: 0163-5085 ISBN: 978-0-8213-9620-9 DOI: 10.1596/978-0-8213-9620-9 Photo credits: page 67: Garimpeiros (independent prospectors) at the Serra Pelada gold mine, in Brazil © Sebastião Salgado/Amazonas—Press Images. Used with permission of Sebastião Salgado/Amazonas—Press Images. Further permission required for reuse. page 91: Day laborer in a pineapple plantation in Pontian, Malaysia © Justin Guariglia/Redux. Used with permission of Justin Guariglia/ Redux. Further permission required for reuse. page 145: Shopkeeper and a friend at a foodstuff shop in Mpape, Nigeria © Ayemoba Godswill/World Bank; Rural migrants working in construction in China © Curt Carnemark/World Bank. page 222: Farmers in a pomegranate field in Tajikistan © Gennadiy Ratushenko/World Bank; Wage worker at a garment factory in Vietnam © Lino Vuth/World Bank; Street vendor in Kabul, Afghanistan © Steve McCurry/Magnum Photos; Drying peppers in the street in Mexico © Curt Carnemark/World Bank. page 248: Employees at a call center in Poland © Piotr Malecki/Panos Pictures. page 318: Worker at a construction site in Jakarta, Indonesia © Sebastião Salgado/Amazonas—Press Images. Used with permission of Sebastião Salgado/Amazonas. Further permission required for reuse. Cover design: Will Kemp, World Bank Interior design: Debra Naylor Contents Foreword  xiii Acknowledgments  xv Abbreviations and data notes   xvii Overview  Moving jobs center stage   2 Jobs wanted  3 Development happens through jobs   8 Valuing jobs  14 Jobs agendas are diverse . . . but connected   17 Policies through the jobs lens   21 Jobs are center stage, but where are the numbers?   34 Questions  When is the conventional wisdom right?    36 Notes  39 References  41 1 The jobs challenge   48 A job, but not always a salary   49 Youth bulges, aging societies, and migrant nations   51 Cities, wages, and women   52 Jobs are changing in surprising ways   54 Prosperity, but a changing distribution of earnings   56 The role of the private sector   58 Vulnerability on a global scale   58 Question 1  What is a job?    63 Notes  68 References  69 v vi  CONTENTS Part 1  Jobs are transformational   74 2 Jobs and living standards   76 Jobs improve material well-being   76 Jobs are more than just earnings   82 Jobs and life satisfaction   84 Question 2  Growth strategies or jobs strategies?   87 Notes  92 References  93 3 Jobs and productivity   98 Employment turbulence, not jobless growth    98 Most jobs are in very small farms and firms    104 In farms, uneven technological progress   106 Among firms, much churning and few gazelles   107 Question 3  Can entrepreneurship be fostered?   114 Notes  119 References  121 4 Jobs and social cohesion   126 Jobs can help manage social tensions   127 Jobs (or the lack of jobs) can shape social interactions    134 Question 4  Can policies contribute to social cohesion?    140 Notes  146 References  147 Part 2  What are good jobs for development?   152 5 Valuing jobs   154 Rights as the foundation   155 The value of jobs to individuals and society   158 Spillovers from jobs   159 Can the development payoffs from jobs be quantified?   162 Question 5  Skills or jobs—which comes first?   174 Notes  179 References  182 6 Diverse jobs agendas   190 Agrarian economies  190 Conflict-affected countries  193 Urbanizing countries  197 Resource-rich countries  199 Small island nations   203 Contents  vii Countries with high youth unemployment   206 Formalizing economies  210 Aging societies  213 Question 6  A targeted investment climate?   217 Notes  223 References  225 7 Connected jobs agendas  232 Migration of workers   232 Migration of jobs   237 Question 7  Competing for jobs?    243 Notes  249 References  250 Part 3  Policies through the jobs lens   256 8 Labor policies revisited   258 Labor regulations: A “plateau� effect   260 Collective representation: New forms of voice   263 Active labor market programs: Effective within limits   267 Social insurance: The challenge of expanding coverage   272 Question 8  Protecting workers or protecting jobs?   277 Notes  281 References  284 9 Beyond labor policies   292 Establishing the fundamentals   293 Setting policy priorities for jobs   298 Diverse jobs agendas, diverse policy priorities   301 Connected jobs agendas: Global partnerships for jobs   305 Jobs are center stage, but where are the numbers?   311 Question 9  How to accelerate labor reallocation?   313 Notes  319 References  321 Appendixes  328 Glossary  329 Bibliographical note  332 Background papers and notes   334 Selected indicators  337 Index  381 viii   CONTENTS Boxes 1 How does women’s labor force participation increase?   30 5.3 The concept of Decent Work and the Decent Work 1.1 The nature of work and leisure change as cities Agenda  158 develop  53 5.4 Economics and the social sciences deal with spillovers from 1.2 Jobs bring earnings opportunities to women, but also new jobs, under different names   160 difficulties  54 5.5 Several data sources can be used to quantify the development 1.3 The temporary staffing industry is growing in developing payoffs from jobs   163 countries  57 5.6 International definitions of green jobs can be too narrow for 1.4 Responses to the crisis went beyond income support for the developing countries  170 unemployed  62 5.7 How skills are formed, and how they can be 1.5 Few countries produce statistics on informality   64 measured  175 1.6 Not all child work is child labor    66 5.8 Manpower planning has given way to dynamic skills 2.1 There are many dimensions of living standards and many development  177 ways to measure them   77 6.1 Can agrarian Ethiopia compete in manufacturing?    194 2.2 Most poor people work   80 6.2 Conflict can increase labor force participation among 2.3 The value of job attributes can be quantified through hedonic women   194 pricing  83 6.3 Solving jobs challenges is urgent in South Sudan   195 2.4 Work can pose risks to health and safety   84 6.4 Development pessimism about Bangladesh was 2.5 The relationship between growth and employment is not understandable, but has been proven wrong    197 mechanical  88 6.5 The entrepreneurs of Bangladesh are local   198 2.6 Korea went from a growth to a jobs strategy, and Singapore 6.6 Landowner companies can build capacity while spreading the oher way around   90 the wealth    203 3.1 What drives economic growth?   99 6.7 The debate on how to reduce informality is intense 3.2 Microenterprises account for most job creation and Mexico  212 in  destruction  106 6.8 In Ukraine, the impact of aging is compounded by migration 3.3 Most microenterprises are in rural areas and engage in and declining fertility   216 commerce  110 6.9 Once again, the debate rages over industrial policy   218 3.4 What explains the boom in the garment industry in Bangladesh?   117 6.10 Caution is needed when interpreting results from surveys  219 enterprise  4.1 What is social cohesion?    128 6.11 Special economic zones have a mixed record   221 4.2 Do jobs cause trust? Analysis of Eurobarometer and Latinobarómetro Surveys  132 7.1 Why do multinationals locate where they do?   240 4.3 Displacement and unemployment can lead to the erosion of 7.2 E-links create job opportunities in developing countries, trust and ties   133 but the scale is still modest   240 4.4 Jobs, motivation, and identity in Risaralda, 7.3 Globalization is often viewed as jobs migrating Colombia  134 abroad  244 4.5 Voice can be extended to the self-employed: The case of 8.1 Employment protection legislation covers more than SEWA  135 firing  rules   260 4.6 Some jobs connect people across ethnic boundaries   136 8.2 Are bargaining councils the cause of unemployment in 4.7 Measuring inequality of opportunities in access to South Africa?  265 jobs  138 8.3 New forms of collective bargaining are emerging 4.8 Domestic workers: The journey to an ILO China  266 in  convention  141 8.4 Recicladores forced changes in Bogotá’s solid waste 4.9 From laws on the books to laws in action in Cambodia’s management policies  267 garment sector  142 8.5 E-links to jobs: New technologies open new frontiers   268 4.10 In post-conflict settings, well-designed programs reduce 8.6 The Mahatma Gandhi National Rural Employment social tensions  143 Guarantee Act launched the biggest public works program 5.1 Children do perilous work in artisanal gold mines in in the world   271 Mali  155 8.7 Modern technology can reduce social protection costs, 5.2 Compliance with core labor standards is partial    157 leakage, and corruption   276 Contents  ix 8.8 Kurzarbeit has become a new word in labor market 9.3 Improving business practices facilitates compliance with policies  280 labor standards  307 9.1 How does women’s labor force participation 9.4 Knowledge gaps on jobs and development chart the research increase?  300 agenda  312 9.2 There have been successes in tackling jobs challenges around 9.5 China’s hukou system has been partially liberalized   315 the world  302 Figures 1 A job does not always come with a wage   5 1.6 The skills mix changes with economic development    56 2 Among youth, unemployment is not always the issue   6 1.7 Jobs provide higher earnings and benefits as countries 3 In China, employment growth is led by the private grow   57 sector  8 1.8 Wages in developing countries are catching up    59 4 Jobs are transformational   8 1.9 Returns to education are higher in poorer countries    60 5 Jobs provide higher earnings and benefits as countries 1.10 In China, employment growth is led by the private grow  9 sector  60 6 Jobs account for much of the decline in extreme 1.11 In developing countries, the crisis affected earnings more poverty  10 than employment  61 7 Simultaneous job creation and destruction characterize all 1.12 A majority of countries have ratified the core labor economies  11 standards  65 8 Larger firms pay higher wages   12 2.1 Working hours vary across ages    78 9 The employment share of microenterprises is greater in 2.2 Women spend more time in activities not directly generating developing countries  13 income  79 10 People who are unemployed, or do not have motivating jobs, 2.3 Jobs are the most important source of household participate less in society   14 income  80 11 Views on preferred jobs and most important jobs 2.4 Jobs take households out of poverty, especially in developing differ  16 countries  81 12 Some jobs do more for development   17 2.5 Jobs account for much of the decline in extreme poverty  82 13 The individual and social values of jobs can differ   17 2.6 Workers often care more about job security than about 14 Good jobs for development are not the same income  85 everywhere  20 2.7 Life satisfaction is lower among farmers and the 15 Manufacturing jobs have migrated away from high-income unemployed  86 countries  22 3.1 Economic growth does not occur at the expense of jobs in the 16 Three distinct layers of policies are needed   23 medium term  99 17 Finance and electricity are among the top constraints faced 3.2 Simultaneous job creation and destruction characterize all by formal private enterprises   24 economies   100 18 Combining work and training increases the success rates of 3.3 Labor reallocation across sectors was a driver of productivity programs  27 growth in East Asia   101 19 A decision tree can help set policy priorities   28 3.4 Efficiency gains at the firm level are the main driver of 20 Which countries succeeded at addressing their jobs challenges productivity growth  102 and how?  32 3.5 Efficiency gains and employment growth can go 1.1 A job does not always come with a wage   50 together  103 1.2 Among youth, unemployment is not always the issue   51 3.6 Smallholder farming is dominant outside Latin 1.3 Employment growth is needed to cope with population America  104 growth  52 3.7 The employment share of microenterprises is greater in 1.4 Moving from farms to cities does not always bring economic developing countries    105 growth   53 3.8 Crop yields have diverged vastly across regions   107 1.5 Labor productivity remains low in developing 3.9 The dispersion of productivity in manufacturing is greater in countries   55 developing countries    108 x  CONTENTS 3.10 Large firms tend to perform better and to pay better than 6.4 Migration matters for small island nations, even more so in small ones  109 the Pacific  205 3.11 Young firms are more likely than old ones to engage in 6.5 Youth unemployment rates are extremely high in some innovative activities  109 countries  207 3.12 Surviving firms were born larger and grew less in Ghana than 6.6 Having higher education does not bring better employment in Portugal  111 chances in Tunisia    208 3.13 The majority of firms grew little in India and Mexico   112 6.7 Labor regulation may not be the biggest obstacle to 3.14 Some among the self-employed have the potential to become formalization  211 successful entrepreneurs  115 6.8 The labor force will shrink if age-specific participation rates 3.15 Management scores vary widely across small enterprises in remain constant    214 Sub-Saharan Africa  116 6.9 Labor productivity has to increase to avoid declines in living 4.1 Trust and civic engagement go together with peaceful standards  215 collective decision making   129 6.10 The assessment of constraints to business varies across 4.2 People who are unemployed trust and participate enterprises  220 less  130 7.1 Manufacturing jobs have migrated away from high-income 4.3 People with motivating jobs trust and participate countries  238 more  131 7.2 The global number of manufacturing jobs has not varied 4.4 Having a job means more community participation in much  239 Indonesia  132 7.3 Policies for jobs may or may not harm other 4.5 Inequality of job opportunities varies across countries  247 countries  139 8.1 The mix of labor policies and institutions varies across 5.1 Views on preferred jobs and most important jobs countries  259 differ  159 8.2 The coverage of collective bargaining is low in developing 5.2 Some jobs do more for development   160 countries  264 5.3 The individual and social values of jobs can differ   162 8.3 Combining work and training increases the success rates of programs  269 5.4 Some earnings gaps decrease with the level of development; some do not    165 8.4 In Romania, public works programs have the lowest placement rate and highest placement costs    272 5.5 A higher women’s share of household income raises food expenditures in the Republic of Congo   166 8.5 Labor taxes and social contributions vary across different countries facing different job challenges   275 5.6 Who gets the jobs matters for poverty reduction in Bulgaria and Latvia  166 8.6 Workers are willing to give up earnings for access to health insurance and pensions   275 5.7 Agglomeration effects vary across industrial sectors in Taiwan, China  167 8.7 Decoupling between job creation and job destruction was massive in the United States during recessions   278 5.8 Knowledge spillovers from foreign direct investment increase domestic productivity  168 9.1 Three distinct layers of policies are needed   293 5.9 High emissions per worker can go hand in hand with low 9.2 Finance and electricity are among the top constraints faced emissions per unit of output   169 by formal private enterprises   295 5.10 Proximity of garment factories stimulates schooling among 9.3 The rule of law is associated with development    297 young girls in Bangladesh   171 9.4 A decision tree can help set policy priorities   299 5.11 Not all jobs provide social identity, networks, or a sense of 9.5 Chile reduced its dependence on mineral exports   304 fairness  172 9.6 Unemployment rates for youth have fallen in 5.12 Gender and father’s education account for a large share of Slovenia  305 inequality of opportunity in access to jobs   173 9.7 Offers to liberalize services are generally modest   309 5.13 Relative to other obstacles, skills have become more severe 9.8 Is there a “missing middle� in the distribution of constraint to business   175 manufacturing firms in India?    314 6.1 In the absence of a Green Revolution, poverty remains high 9.9 Export processing zones were a driver of foreign direct in agrarian economies    192 investment in Sri Lanka   316 6.2 Instability and poor infrastructure are severe constraints on 9.10 Restrictions to hukou conversion increase with city size and business in conflict-affected countries   196 income  317 6.3 Small island nations are located far away from economic centers  204 Contents  xi Maps Tables 1 Only in some countries are migrants a substantial share of 3.1 Few small firms grew in Mexico   113 the population  21 6.1 Projects in extractive industries are capital intensive and 3.1 Manufacturing activities are sprawling out of the main urban create few jobs    200 centers in the Republic of Korea    102 6.2 Cities in resource-rich developing countries are among the 7.1 Only in some countries are migrants a substantial share of most expensive in the world    201 the population  233 8.1 There is a wave of new empirical evidence on the impacts 7.2 Many migrants are highly skilled   235 EPL  261 of  8.1 Coverage of social insurance remains low in many 8.2 The impacts of minimum wages are a favorite research topic countries   274 in labor economics   262 Foreword Today, jobs are a critical concern across the globe—for policy makers, the business community, and the billions of men and women striving to provide for their families. As the world struggles to emerge from the global crisis, some 200 million people—includ- ing 75 million under the age of 25—are unemployed. Many millions more, most of them women, find themselves shut out of the labor force altogether. Looking forward, over the next 15 years an additional 600 million new jobs will be needed to absorb burgeoning working-age populations, mainly in Asia and Sub-Saharan Africa. Meanwhile, almost half of all workers in developing countries are engaged in small-scale farming or self-employment, jobs that typically do not come with a steady paycheck and ben- efits. The problem for most poor people in these countries is not the lack of a job or too few hours of work; many hold more than one job and work long hours. Yet, too often, they are not earning enough to secure a better future for themselves and their children, and at times they are working in unsafe conditions and without the protection of their basic rights. Jobs are instrumental to achieving economic and social development. Beyond their critical importance for individual well-being, they lie at the heart of many broader societal objec- tives, such as poverty reduction, economy-wide productivity growth, and social cohesion. The development payoffs from jobs include acquiring skills, empowering women, and stabilizing post-conflict societies. Jobs that contribute to these broader goals are valuable not only for those who hold them but for society as a whole: they are good jobs for development. The World Development Report 2013 takes the centrality of jobs in the development pro- cess as its starting point and challenges and reframes how we think about work. Adopting a cross-sectoral and multidisciplinary approach, the Report looks at why some jobs do more for development than others. The Report finds that the jobs with the greatest development payoffs are those that make cities function better, connect the economy to global markets, protect the environment, foster trust and civic engagement, or reduce poverty. Critically, these jobs are not only found in the formal sector; depending on the country context, informal jobs can also be transformational. Building on this framework, the Report tackles some of the most pressing questions policy makers are asking right now: Should countries design their development strategies around growth or focus on jobs? Are there situations where the focus should be on protecting jobs as opposed to protecting workers? Which needs to come first in the development process— creating jobs or building skills? The private sector is the key engine of job creation, accounting for 90 percent of all jobs in the developing world. But governments play a vital role by ensuring that the conditions are in place for strong private sector–led growth and by alleviating the constraints that hinder the private sector from creating good jobs for development. The Report advances a three-stage approach to help governments meet these objectives. First, policy fundamentals—including macroeconomic stability, an enabling business envi- ronment, investments in human capital, and the rule of law—are essential for both growth and job creation. Second, well-designed labor policies can help ensure that growth translates into employment opportunities, but they need to be complemented by a broader approach to job creation that looks beyond the labor market. Third, governments should strategically identify xiii xiv  F O R E WO R D which jobs would do the most for development given their specific country context, and re- move or offset the obstacles that prevent the private sector from creating more of those jobs. In today’s global economy, the world of work is rapidly evolving. Demographic shifts, tech- nological progress, and the lasting effects of the international financial crisis are reshaping the employment landscape in countries around the world. Countries that successfully adapt to these changes and meet their jobs challenges can achieve dramatic gains in living standards, productivity growth, and more cohesive societies. Those that do not will miss out on the trans- formational effects of economic and social development. The World Development Report 2013 is an important contribution to our collective under- standing of the role of jobs in development. Its insights will provide valuable guidance for the World Bank Group as we collaborate with partners and clients to advance their jobs agendas. Working together, we can foster job creation and maximize the development impact of jobs. Jim Yong Kim President The World Bank Group Acknowledgments This Report was prepared by a team led by Martín Rama, together with Kathleen Beegle and Jesko Hentschel. The other members of the core team were Gordon Betcherman, Samuel Freije-Rodriquez, Yue Li, Claudio E. Montenegro, Keijiro Otsuka, and Dena Ringold. Research analysts Thomas Bowen, Virgilio Galdo, Jimena Luna, Cathrine Machingauta, Daniel Pala- zov, Anca Bogdana Rusu, Junko Sekine, and Alexander Skinner completed the team. Addi- tional research support was provided by Mehtabul Azam, Nadia Selim, and Faiyaz Talukdar. The team benefited from continuous engagement with Mary Hallward-Driemeier, Roland Michelitsch, and Patti Petesch. The Report was cosponsored by the Development Economics Vice Presidency (DEC) and the Human Development Network (HDN). Overall guidance for the preparation of the Report was provided by Justin Lin, former Senior Vice President and Chief Economist, Development Economics; Martin Ravallion, acting Senior Vice President and Chief Economist, Develop- ment Economics; and Tamar Manuelyan-Atinc, Vice President and Head of the Human Development Network. Asli Demirgüç-Kunt, Director for Development Policy, oversaw the preparation process, together with Arup Banerji, Director for Social Protection and Labor. Former World Bank President Robert B. Zoellick, President Jim Yong Kim, and Manag- ing Directors Caroline Anstey and Mahmoud Mohieldin provided invaluable insights during the preparation process. Executive Directors and their offices also engaged constructively through various meetings and workshops. An advisory panel, comprising George Akerlof, Ernest Aryeetey, Ragui Assaad, Ela Bhatt, Cai Fang, John Haltiwanger, Ravi Kanbur, Gordana Matkovic, ´ and Ricardo Paes de Barros, contributed rich analytical inputs and feedback throughout the process. Seven country case studies informed the preparation of the Report. The case study for Bangladesh was led by Binayak Sen and Mahabub Hossain, with Yasuyuki Sawada. Nelly Agu- ilera, Angel Calderón Madrid, Mercedes González de la Rocha, Gabriel Martínez, Eduardo Rodriguez-Oreggia, and Héctor Villarreal participated in Mexico’s case study. The study for Mozambique was led by Finn Tarp, with Channing Arndt, Antonio Cruz, Sam Jones, and Fausto Mafambisse. For Papua New Guinea, Colin Filer and Marjorie Andrew coordi- nated the research. The South Sudan study was led by Lual Deng, together with Nada Eissa. AbdelRahmen El Lahga coordinated the Tunisian work, with the participation of Ines Bouassida, Mohamed Ali Marouani, Ben Ayed Mouelhi Rim, Abdelwahab Ben Hafaiedh, and Fathi Elachhab. Finally, Olga Kupets, Svitlana Babenko, and Volodymyr Vakhitov con- ducted the study for Ukraine. The team would like to acknowledge the generous support for the preparation of the Report by the Government of Norway through its Ministry of Foreign Affairs, the multi- donor Knowledge for Change Program (KCP II), the Nordic Trust Fund, the G ­ overnment of Denmark through its Royal Ministry of Foreign Affairs, the Swiss State Secretariat for Eco- nomic Affairs (SECO), the Canadian International Development Agency (CIDA), the Gov- ernment of Sweden through its Ministry for Foreign Affairs, and the Government of Japan xv xvi  AC K N OW L E D G M E N T S through its Policy and Human Resource Development program. The German Ministry for Economic Cooperation and Development Cooperation (BMZ) through the German Agency for International Cooperation (GIZ) organized a development forum that brought together leading researchers from around the world in Berlin. Generous support was also received for the country case studies by the Australian Agency for International Development (AusAID), Canada’s International Development Research Centre (IDRC), the Government of Denmark through its Royal Ministry of Foreign Affairs, the Japan International Cooperation Agency (JICA) through the JICA Institute, and the United Nations University World Institute for Development Economics Research (UNU-WIDER). The United Kingdom’s Overseas Development Institute (ODI) assisted the team through the organization of seminars and workshops. A special recognition goes to the International Labour Organization (ILO) for its contin- ued engagement with the team. José Manuel Salazar-Xiriñachs and Duncan Campbell coor- dinated this process, with the participation of numerous colleagues from the ILO. Interagency consultations were held with the International Monetary Fund (IMF), the Organisation for Economic Co-operation and Development (OECD), and the United Nations Economic and Social Council (ECOSOC). The team also benefited from an ongoing dialogue with the Inter- national Trade Union Confederation (ITUC). Country consultations were conducted in Bangladesh, Canada, China, Denmark, Finland, France, Germany, India, Japan, the Republic of Korea, Mexico, Mozambique, Norway, Papua New Guinea, Singapore, Sweden, Switzerland, Tunisia, Turkey, Ukraine, and the United King- dom. All consultations involved senior government officials. Most included academics, business representatives, trade union leaders, and members of civil society. In addition, bilateral meet- ings were held with senior government officials from Australia, the Netherlands, South Africa, and Spain. Consultations with researchers and academics were arranged with the help of the African Economic Research Consortium (AERC) in Kenya, the Economic Research Forum (ERF) in the Arab Republic of Egypt, and the Latin American and Caribbean Economic Associa- tion (LACEA) in Chile. The Institute for the Study of Labor (IZA) organized special work- shops with its research network in Germany and Turkey, coordinated by Klaus Zimmerman. Forskningsstiftelsen Fafo in Norway undertook a household survey in four countries, which this Report draws on. The production of the Report and the logistics supporting it were assured by Brónagh Murphy, Mihaela Stangu, Jason Victor, and Cécile Wodon, with a contribution by Quyên Thúy �inh. Ivar Cederholm coordinated resource mobilization. Irina Sergeeva and Sonia Joseph were in charge of resource management. Martha Gottron, Bruce Ross-Larson, Gerry Quinn, and Robert Zimmermann participated in the editing of the Report. The Development Data Group, coordinated by Johan Mistiaen, contributed to the preparation of its statistical annex. The Office of the Publisher coordinated the design, typesetting, printing, and dissemina- tion of both the hard and soft versions of the Report. Special thanks go to Mary Fisk, Stephen McGroarty, Santiago Pombo-Bejarano, Nancy Lammers, Stephen Pazdan, Denise Bergeron, Andres Meneses, Theresa Cooke, Shana Wagger, Jose De Buerba, and Mario Trubiano, as well as to the Translations and Interpretation Unit’s Cecile Jannotin and Bouchra Belfqih. The team also thanks Vivian Hon, as well as Claudia Sepúlveda, for their coordinating role; Merrell Tuck-Primdahl for her guidance on communication; Vamsee Krishna Kanchi and Swati P. Mishra for their support with the website; Gerry Herman for his help with the prepa- ration of the movie series associated with the Report; and Gytis Kanchas, Nacer Mohamed Megherbi, and Jean-Pierre S. Djomalieu for information technology support. Many others inside and outside the World Bank contributed with comments and inputs. Their names are listed in the Bibliographical Note. Abbreviations and data notes ABBREVIATIONS ADB Asian Development Bank ALMP active labor market program ARB Asociación de Recicladores de Bogotá (Bogotá Association of Recyclers) BPO business process outsourcing CAFTA Central America Free Trade Agreement CASEN La Encuesta de Caracterizacíon Socioeconomica Nacional (Chile National Socioeconomic Characterization) CIRAD Centre de coopération internationale en recherche agronomique   pour le développement (Center for International Cooperation in Agronomic Research for Development) CFA Committee on Freedom of Association COSATU Confederation of South African Trade Unions CSR corporate social responsibility ECLAC Economic Commission for Latin America and the Caribbean ECOSOC United Nations Economic and Social Council EMBRAPA Empresa Brasileira de Pesquisa Agropecuária (Brazilian Enterprise for Agricultural Research) EPL employment protection legislation EPZ export processing zone EU European Union FAO Food and Agriculture Organization of the United Nations FAFO Forskningsstiftelsen Fafo (Fafo Research Foundation) FDI foreign direct investment FACB freedom of association and collective bargaining GATT General Agreement on Tariffs and Trade GATS General Agreement on Trade in Services GDP gross domestic product GNP gross national product HOI Human Opportunity Index I2D2 International Income Distribution Database IC Industrial Council ICLS International Conference of Labour Statisticians ICTWSS Institutional Characteristics of Trade Unions,   Wage Setting, State Intervention and Social Pacts IDA Industrial Disputes Act (India) IDRC International Development Research Center IEA International Energy Agency IFC International Finance Corporation xvii xviii  A B B R E V I AT I O N S A N D DATA N O T E S IFPRI International Food Policy Research Institute ILO International Labour Organization IMF International Monetary Fund IPCC International Panel on Climate Change ISSP International Social Survey Programme IT information technology IZA Forschungsinstitut zur Zukunft der Arbeit (Institute for the   Study of Labor) KILM Key Indicators of the Labor Market KUT Korea University of Technology and Education MDG Millennium Development Goal MERCOSUR Mercado Común del Sur (Southern Cone Common Market) MFA Multi-Fiber Arrangement MGNREGA Mahatma Gandhi National Rural Employment Guarantee Act MIS Management Information System NASSCOM National Association of Software and Service Companies NEET not in education, employment, or training NGO nongovernmental organization ODI Overseas Development Institute OECD Organisation for Economic Co-operation and Development PISA Programme for International Student Assessment PPP purchasing power parity R&D research and development SEWA Self Employed Women’s Association SEZ special economic zone SME small and medium enterprise SNA System of National Accounts SOE state-owned enterprise TEWA Termination of Employment of Workmen Act TFP total factor productivity TVE technical and vocational education UN United Nations UNDP United Nations Development Programme UNECE United Nations Economic Commission of Europe UNEP United Nations Environment Programme UNESCO United Nations Educational, Scientific and Cultural Organization WDR World Development Report WTO World Trade Organization WIEGO Women in Informal Employment: Globalizing and Organizing DATA NOTES The use of the word countries to refer to economies implies no judgment by the World Bank about the legal or other status of territory. The term developing countries includes low- and middle-income economies and thus may include economies in transition from central plan- ning, as a matter of convenience. Dollar figures are current U.S. dollars, unless otherwise specified. Billion means 1,000 million; trillion means 1,000 billion. OVERVIEW Moving jobs center stage J obs are the cornerstone of economic and social development. Indeed, development ­ happens through jobs. People work their way out of poverty and hardship through bet- • While jobs can contribute to social cohesion, is there anything governments can do about it, apart from trying to support job creation? ter livelihoods. Economies grow as people get •  Are greater investments in education and training a prerequisite for employability, or better at what they do, as they move from farms can skills be built through jobs? to firms, and as more productive jobs are cre- ated and less productive ones disappear. Soci- • Should efforts to improve the investment cli- eties flourish as jobs bring together people mate target the areas, activities, or firms with from different ethnic and social backgrounds greater potential for job creation? and nurture a sense of opportunity. Jobs are thus transformational—they can transform •  What is the risk that policies to foster job creation in one country will come at the ex- what we earn, what we do, and even who we pense of jobs in other countries? are. No surprise, then, that jobs are atop the •  When confronted with large shocks and ma- development agenda everywhere—for every- jor restructuring, is it advisable to protect jobs one from policy makers to the populace, from and not just people? business leaders to union representatives, from activists to academics. Looking to seize oppor- •  How can the reallocation of workers be ac- celerated from areas and activities with low tunities for job creation presented by massive productivity to those with greater potential? demographic shifts, technological innovations, global migrations of people and tasks, and deep Individuals value jobs for the earnings and changes in the nature of work, policy makers ask benefits they provide, as well as for their contri- difficult questions: butions to self-esteem and happiness. But some jobs have broader impacts on society. Jobs for •  Should countries build their development women can change the way households spend strategies around growth or should they money and invest in the education and health rather focus on jobs? of children. Jobs in cities support greater spe- •  Can entrepreneurship be fostered, especially cialization and the exchange of ideas, making among the many microenterprises in devel- other jobs more productive. Jobs connected oping countries, or are entrepreneurs born? to global markets bring home new technologi- Moving jobs center stage    3 cal and managerial knowledge. And in turbulent tries grow richer, the policy environment environments, jobs for young men can provide must be conducive to growth. That requires alternatives to violence and help restore peace. attending to macroeconomic stability, an en- Through their broader influence on living abling business environment, human capital standards, productivity, and social cohesion, accumulation, and the rule of law. these jobs have an even greater value to society than they do for the individual. But some jobs •  Labor policies. Because growth alone may not be enough, labor policies need to facilitate can have negative spillovers. Jobs supported job creation and enhance the development through transfers or privilege represent a bur- payoffs from jobs. Policies can address labor den to others or undermine their opportunities market distortions while not being a drag on to find remunerative employment. Jobs damag- efficiency. But they should avoid distortion- ing the environment take a toll on everybody. ary interventions that constrain employ- Thus it is that some jobs do more for develop- ment in cities and global value chains—and ment, while others may do little, even if they are provide voice and protection for the most appealing to individuals. vulnerable. Which jobs have the greatest develop- ment payoffs depends on the circumstances. •  Priorities. Because some jobs do more for Countries differ in their level of development, development than others, it is necessary to demography, endowments, and institutions. identify the types of jobs with the greatest Agrarian socie­ ties face the challenge of making development payoffs given a country’s con- agricultural jobs more productive and creat- text, and to remove—or at least offset—the ing job opportunities outside farms. Resource- market imperfections and institutional fail- rich countries need to diversify their exports, ures that result in too few of those jobs being so that jobs are connected to global markets created. rather than supported through government The centrality of jobs for development transfers. Formalizing countries need to de- should not be interpreted as the centrality of sign their social protection systems in ways labor policies and institutions. Nearly half that extend their coverage without penalizing the people at work in developing countries employment. are farmers or self-employed and so are out- A vast majority of jobs are created by the side the labor market. And even in the case of private sector. Governments, though, can sup- wage employment, labor policies and institu- port—or hinder—the private sector in creat- tions may or may not be the main obstacle to ing jobs. The idea that development happens job creation. Often, the most relevant obstacles through jobs sheds new light on the strategies, lie outside of the labor market. The catalysts policies, and programs governments can pur- for job creation may be policies that make cit- sue. Strategies should identify which types of ies work better, help farmers access and apply jobs would have the highest development pay- appropriate agricultural techniques, or allow offs, given a country’s circumstances. Policies firms to develop new exports. Jobs are the cor- should remove the obstacles that prevent the nerstone of development, and development private sector from creating jobs. Programs for policies are needed for jobs. generating employment may also be warranted, for instance, in conflict-affected countries. But the costs and benefits of these policies and pro- Jobs wanted grams have to be assessed, taking into account the potential spillovers from jobs, both positive To many, a “job� brings to mind a worker with and negative. an employer and a regular paycheck. Yet, the At a more practical level, this jobs lens on majority of workers in the poorest countries devel­opment leads to a three-layered policy are outside the scope of an employer-employee approach: relationship. Worldwide, more than 3 billion •  Fundamentals. Because jobs provide higher people are working, but their jobs vary greatly. earnings and broader social benefits as coun- Some 1.65 billion are employed and receive reg- 4  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 1.6 billion people working for a wage or a salary 1.5 billion people working in farming and self-employment 77% labor force participation by women in Vietnam 28% labor force participation by women in Pakistan 39% of the manufacturing jobs are in microenterprises in Chile 97% of the manufacturing jobs are in microenterprises in Ethiopia 2x employment growth in a firm in Mexico over 35 years 10x employment growth in a firm in the United States over 35 years 115 million children working in hazardous conditions 21 million victims of forced labor 600 million jobs needed over 15 years to keep current employment rates 90 million people working abroad 621 million youth neither working nor studying 22x  the productivity gap between manufacturing firms in the 90th and 10th percentiles in India 9x the productivity gap between manufacturing firms in the 90th and 10th percentiles in the United States 10 million entrants to the labor force per year in Sub-Saharan Africa 30 million postsecondary students in China 3% international migrants as a share of the world population 60% foreign-born population in Kuwait, Qatar, and the United Arab Emirates Moving jobs center stage    5 ular wages or salaries. Another 1.5 billion work rights as the boundaries of what is unacceptable. in farming and small household enterprises, or Among them are the United Nations Universal in casual or seasonal day labor. Meanwhile, 200 Declaration of Human Rights (1948) and the million people, a disproportionate share of them International Labour Organization Declaration youth, are unemployed and actively looking for on Fundamental Principles and Rights at Work work. Almost 2 billion working-age adults, the (1998), which further specifies core labor stan- majority of them women, are neither working dards. Combining these different perspectives, nor looking for work, but an unknown number jobs are activities that generate income, mone- of them are eager to have a job. Clarifying what tary or in kind, without violating human rights. is meant by a job is thus a useful starting point. The meaning of the words used to de- Different places, different jobs scribe what people do to earn a living varies across countries and cultures. Some words re- The world of work is particularly diverse in de- fer to workers in offices or factories. Others are veloping countries. This variety refers not only broader, encompassing farmers, self-employed to the number of hours worked and the number vendors in cities, and caregivers of children and of jobs available, the usual yardsticks in indus- the elderly. The distinction is not merely seman- trial countries, but also to the characteristics of tic. The varied meanings hint at the different jobs. Two main aspects stand out. One is the aspects of jobs that people value. And views on prevalence of self-employment and farming.2 what a job is almost inevitably influence views The other is the coexistence of traditional and on what policies for jobs should look like. modern modes of production, from subsistence For statisticians, a job is “a set of tasks and agriculture and low-skilled work to technology- duties performed, or meant to be performed, driven manufacturing and services and highly by one person, including for an employer or skilled knowledge work. in self-employment.�1 Jobs are performed by While nearly half of the jobs in the developing the employed. These are defined as people who world are outside the labor market, the shares of produce goods and services for the market or wage work, farming, and self-­ employment differ for their own use. But the statistical definition greatly across countries.3 Nonwage work repre- is mute about what should not be considered sents more than 80 percent of women’s em- a job. International norms view basic human ployment in Sub-Saharan ­ Africa—but less than A F I G U R E 1   job does not always come with a wage men women 100 wage employment share of total employment, % 80 self-employment 60 nonwage 40 employment 20 farming 0 Europe and Latin America South Middle East East Asia Sub-Saharan Central Asia and the Asia and and Paci�c Africa Caribbean North Africa Source: World Development Report 2013 team. Note: Data are for the most recent year available. 6  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 for both men and women in Tanzania and Viet- Among F I G U R E 2   youth, unemployment is not always the nam. Beyond these stark contrasts in participa- issue tion, women continue to earn significantly less than men, and the differences are not fully ex- not in school or at work plained by education, experience, or sector of work. While a growing share of youth between not looking for work looking for work ages 15 and 24 allocate most of their time to Pakistan women schooling and training, youth unemployment is 2008 men still alarming in some countries (above 40 per- Turkey cent in South Africa since early 2008 and above 2005 50 percent in Spain in early 2012).5 Even in India countries where it is low, youth unemployment 2009 is twice the national average or more. In addi- Indonesia 2010 tion, 621 million young people are “idle�—not in school or training, not employed, and not Chile 2009 looking for work. Rates of idleness vary across Brazil countries, ranging between 10 and 50 percent 2009 among 15- to 24-year-olds (figure 2).6 Many Ukraine youth work in unpaid jobs; if paid, they are less 2005 likely to have social insurance.7 Ghana 2005 Tanzania The changing world of work 2009 This complex picture is compounded by mas- 0 10 20 30 40 50 60 sive demographic shifts. To keep employment share of population ages 15–24, % as a share of the working-age population con- stant, in 2020 there should be around 600 mil- Source: World Development Report 2013 team. lion more jobs than in 2005, a majority of them in Asia and Sub-Saharan Africa. While some countries have experienced very large increases in their labor force—nearly 8 million new en- 20 percent in Eastern Europe and Central Asia trants a year in China since the mid-1990s and 7 (figure 1). million in India—others face a shrinking popu- Work across the developing world is also lation. Ukraine’s labor force, for example, is es- characterized by a high prevalence of informal- timated to fall by about 160,000 people a year.8 ity, whether defined on the basis of lack of firm Rapid urbanization is changing the com- registration, lack of social security coverage, or position of employment. More than half the lack of an employment contract. Informal em- population in developing countries is expected ployment is not under the purview of labor reg- to be living in cities and towns before 2020.9 ulations, either because of their limited scope or As a result, the growth of the nonagricultural because of deliberate avoidance or evasion. Re- labor force will vastly exceed the growth of the gardless of the specific definition used, informal- agricultural labor force. This structural change, ity is generally associated with lower productiv- which in industrial countries took decades, now ity. However, this does not necessarily mean that transforms lives in developing countries in a formalization would result in greater efficiency. generation. Structural change can bring about Informality can be a symptom of lower produc- remarkable improvements in efficiency, and tivity as much as it can be a cause of it.4 some developing countries have narrowed the Gender and age differences are striking. productivity gap with industrial countries rap- Worldwide, fewer than half of women have jobs, idly. But others have failed to catch up.10 Over- compared with almost four-fifths of men. In all, the gap between developing and developed Pakistan, 28 percent of women but more than regions remains wide. 82 percent of men participate in the labor force, Globalization is also changing the nature whereas participation rates are above 75 percent of jobs. Industrial countries are shifting from Moving jobs center stage    7 primary and traditional manufacturing indus- Between 1995 and 2005, the private sector ac- tries toward services and knowledge-intensive counted for 90 percent of jobs created in Bra- activities.11 At the same time, technological zil, and for 95 percent in the Philippines and improvements and outsourcing to developing Turkey.21 The most remarkable example of the countries are leading to a decline in medium- ­ expansion of employment through private sec- skilled jobs.12 Production tasks have been splin- tor growth is China. In 1981, private sector em- tered so that they can be performed in different ployment accounted for 2.3 million workers, locations.13 Transnational companies have built while state-owned enterprises (SOEs) had 80 integrated value chains to tap into national skill million workers.22 Twenty years later, the private pools around the world.14 Outsourcing is oc- sector accounted for 74.7 million workers, sur- curring in services as well as in manufacturing. passing, for the first time, the 74.6 million work- The share of developing countries in exports of ers in SOEs (figure 3). world services nearly doubled to 21 percent be- In contrast to the global average, in some tween 1990 and 2008.15 countries in the Middle East and North Africa, Technology is changing the way workers and the state is a leading employer, a pattern that can firms connect, through their access to much be linked to the political economy of the post- larger, even global, employment marketplaces. independence period, and in some cases to the Some of the new marketplaces operate through abundance of oil revenues.23 For a long period, the internet; others use mobile phone technol- public sector jobs were offered to young college ogy.16 Part-time and temporary wage employ- graduates. But as the fiscal space for continued ment are now major features of industrial and expansion in public sector employment shrank, developing countries. In South Africa, tempo- “queuing� for public sector jobs became more rary agency workers make up about 7 percent of prevalent, leading to informality, a devaluation the labor force; the temporary staffing industry of educational credentials, and forms of social provides employment to an average of 410,000 exclusion.24 A fairly well-educated and young workers a day. In India, the number of tempo- labor force remains unemployed, or underem- rary workers that employment agencies recruit ployed, and labor productivity stagnates.25 grew more than 10 percent in 2009 and 18 per- Overall, countries have been successful at cent in 2010.17 creating jobs. More people have jobs now than This changing landscape of global produc- ever before, and those jobs provide generally tion has also brought about shifts in skill en- higher earnings. Indeed, amid rapid social and dowments and in the world distribution of top economic change, poverty has declined in de- talent. China and India rank high in perceived veloping countries. The share of the popula- attractiveness as outsourcing hubs because of tion of the developing world living on less than their exceptionally high ratings in the avail- US$1.25 a day (in purchasing power parity) fell ability of skills.18 India has close to 20 million from 52 percent in 1981 to 22 percent in 2008, students in higher education, nearly as many as or from 1.94 billion people to 1.29 billion.26 the United States; both countries are outpaced This reduction is the result of multiple factors, by China, with 30 million postsecondary stu- but the creation of millions of new, more pro- dents.19 The United States still accounts for a ductive jobs, mostly in Asia but also in other large share of top scores in international student parts of the developing world, has been the assessments, but the Republic of Korea has the main driving force.27 same share as Germany, and both are closely fol- Jobs are vulnerable to economic downturns, lowed by the Russian Federation. The number though, much more so in the private sector than of high-performing students in Shanghai alone the public sector. Short-term crises may wipe is one-fifth that of Germany and about twice out years of progress. They may start in a single that of Argentina.20 country but now, through globalization, spread over entire regions or to the world. The recent financial crisis created 22 million new unem- The role of the private sector ployed in a single year. Growth in total employ- In such rapidly changing times, the private sec- ment, hovering around 1.8 percent a year before tor is the main engine of job creation and the 2008, fell to less than 0.5 percent in 2009, and source of almost 9 of every 10 jobs in the world. by 2011 had not yet reached its pre-crisis level.28 8  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 In F I G U R E 3   China, employment growth is led by the private sector 110 100 90 number of workers, millions 80 70 60 50 40 30 20 10 0 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 20 state-owned enterprises private rms (8 workers or more) individual rms (fewer than 8 workers) foreign-owned companies Source: Kanamori and Zhao 2004. Note: Data for foreign-owned companies in 2002 and for non-state-owned enterprises in 2003 are not available. Policy responses to prevent and mitigate the im- Demography, urbanization, globalization, pact of crises involve different combinations of technology, and macroeconomic crises bring instruments, with potentially diverse implica- about formidable jobs challenges. Countries tions for jobs.29 that fail to address them may fall into vicious circles of slow growth in labor earnings and job-related dissatisfaction affecting a sizable Jobs F I G U R E 4   are transformational portion of the labor force.30 Youth unemploy- ment and idleness may be high, and women may have fewer job opportunities, leaving po- tential economic and social gains untapped.31 A repeating pattern of small gains in living DEVELOPMENT standards, slow productivity growth, and erod- ing social cohesion can set in. In contrast, countries that address these jobs challenges can develop virtuous circles. The results—pros- perous populations, a growing middle class, LIVING PRODUCTIVITY SOCIAL COHESION increased productivity, and improved oppor- STANDARDS tunities for women and youth—may then be self-reinforcing. Development happens through jobs Jobs are more than just the earnings and benefits JOBS they provide. They are also the output they gen- erate, and part of who we are and how we interact Source: World Development Report 2013 team. with others in society. Through these outcomes, Moving jobs center stage    9 jobs can boost living standards, raise productiv- not refer to identical workers. But growth also ity, and foster social cohesion (figure 4). improves the living standards of workers whose skills have not changed. More than two decades of research on pov- Jobs are what we earn erty dynamics, spanning countries as different Jobs are the most important determinant of as Canada, Ecuador, Germany, and South Africa, living standards. For most people, work is the show that labor-related events trigger exits from main source of income, especially in the poorest poverty.33 These events range from the head of countries. Many families escape or fall into pov- a household changing jobs to family members erty because family members get or lose a job. starting to work and to working family mem- Opportunities for gainful work, including in bers earning more. Conversely, a lack of job op- farming and self-employment, offer households portunities reduces the ability of households the means to increase consumption and reduce to improve their well-being.34 In a large set of its variability. Higher yields in agriculture, ac- qualitative studies in low-income countries, get- cess to small off-farm activities, the migration ting jobs and starting businesses were two of the of family members to cities, and transitions to main reasons for people to rise out of poverty.35 wage employment are milestones on the path to Quantitative analysis confirms that changes prosperity.32 And as earnings increase, individ- in labor earnings are the largest contributor to ual choices expand—household members can poverty reduction (figure 6). In 10 of 18 Latin choose to stay out of the labor force or to work American countries, changes in labor income fewer hours and dedicate more time to educa- explain more than half the reduction in poverty, tion, to retirement, or to family. and in another 5 countries, more than a third. In Earnings from work increase with economic Bangladesh, Peru, and Thailand, changes in edu- development, and the benefits associated with cation, work experience, and region of residence jobs improve as well. The relationship is not mattered, but the returns to these characteristics mechanical, but growth is clearly good for jobs (including labor earnings) mattered most. Just (figure 5). Admittedly, as economies become having work was not enough, given that most more developed, the average skills of jobhold- people work in less developed economies. What ers increase, implying that observations across made a difference for escaping poverty was in- countries are not strictly comparable, as they do creasing the earnings from work.36 Jobs F I G U R E 5   provide higher earnings and benefits as countries grow a. Average wage b. Social security coverage 100,000 100 programs, % of total employment average wage in manufacturing, contributors to social security 80 2005 PPP US$ 10,000 60 40 1,000 20 100 0 300 3,000 30,000 300 3,000 30,000 GDP per capita, 2005 PPP US$ GDP per capita, 2005 PPP US$ Source: World Development Report 2013 team. Note: GDP = gross domestic product; PPP = purchasing power parity. Each dot represents a country. 10   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Jobs F I G U R E 6   account for much of the decline in extreme poverty 200 percentage of total change in extreme poverty 150 100 50 0 –50 –100 ia or ico ile va ica a r ay nd a il a as a l sh ru pa do az in an bi m an ad ur de gu Pe do Ch la aR ex nt m Ne na Br ua Gh m nd ai lv la ra ol lo ge M Pa st Ec Th Ro Sa ng M Pa Ho Co Co Ar El Ba family composition labor income nonlabor income consumption-to-income ratio Sources: Azevedo and others 2012; Inchauste and others 2012; both for the World Development Report 2013. Note: Family composition indicates the change in the share of adults (ages 18 and older) within the household. Labor income refers to the change in employment and earnings for each adult. Nonlabor income refers to changes in other sources of income such as transfers, pensions, and imputed housing rents. If a bar is located below the horizontal axis, it means that that source would have increased, instead of decreased, poverty. The changes are computed for Argentina (2000–10); Bangladesh (2000–10); Brazil (2001–09); Chile (2000–09); Colombia (2002–10); Costa Rica (2000–08); Ecuador (2003–10); El Salvador (2000-09); Ghana (1998–2005); Honduras (1999–2009); Mexico (2000–10); Moldova (2001–10); Panama (2001–09); Paraguay (1999–2010); Peru (2002–10); Nepal (1996–2003); Romania (2001–09); and Thailand (2000–09). The changes for Bangladesh, Ghana, Moldova, Nepal, Peru, Romania, and Thailand are computed using consumption-based measures of poverty, while the changes for the other countries are based on income measures. Beyond their fundamental and immediate are created and less productive jobs disappear. contribution to earnings, jobs also affect other These gains may ultimately be driven by new dimensions of well-being, including mental and goods, new methods of production and trans- physical health. Not having a job undermines portation, and new markets, but they material- life satisfaction, especially in countries where ize through a constant restructuring and real- wage employment is the norm and where the location of resources, including labor.38 Net job lack of opportunities translates into open un- creation figures hide much larger processes of employment rather than underemployment. gross job creation and gross job destruction. On Among those employed, the material, nonmate- average across developing countries, between rial, and even subjective characteristics of jobs 7 and 20 percent of jobs in manufacturing are can all have an impact on well-being.37 Other created within a year, but a similar proportion features such as workplace safety, job security, disappear (figure 7).39 learning and advancement opportunities, and Because economies grow as high-productivity health and social protection benefits are valued jobs are created and low-productivity jobs dis- by workers. But relatively few jobs offer these appear, the relationship between productivity advantages in developing countries. gains and job creation is not mechanical. In the medium term, employment trends align closely with trends in the size of the labor force, so Jobs are what we do growth is truly jobless in very few cases. In the Economic growth happens as jobs become more short term, however, innovations can be associ- productive, but also as more productive jobs ated with either increases or decreases in em- Moving jobs center stage    11 ployment.40 The popular perception is that pro- ductivity grows through downsizing, but some Simultaneous F I G U R E 7   job creation and destruction firms are able to achieve both productivity and characterize all economies employment gains.41 In Chile, Ethiopia, and Romania, successful “upsizers� contributed to net job gross job gross job output and employment growth substantively; creation creation destruction sometimes they are more numerous than the ECONOMY-WIDE successful “down­­sizers.�42 And the combination of private sector vibrancy and state sector re- Latvia structuring led to rapid output and employ- ment growth in transition economies and in Mexico China in the late 1990s and the early 2000s.43 Successful upsizers tend to be younger, Argentina leaner, and more innovative.44 But overall, large firms are both more innovative and more pro- Estonia ductive. They invest more in machinery. They are much more likely than small firms to de- Hungary velop new product lines, to introduce new tech- nology, to open and close plants, to outsource, Slovenia and to engage in joint ventures with foreign partners.45 These firms produce more with a Romania given amount of labor, and export more as well. industrial economies They also pay substantively higher wages than (average) micro- and small enterprises (figure 8). In de- veloping countries, however, many people work MANUFACTURING SECTOR ONLY in very small and not necessarily very dynamic Ethiopia economic units. Family farms dominate in agriculture. At 1.8 Indonesia and 1.2 hectares, respectively, average farm size is small in Sub-Saharan Africa, and especially Brazil in Asia.46 The Green Revolution has led to both higher cereal yields and more job creation be- Chile cause the new technologies are labor intensive. But progress has been uneven across regions Taiwan, China and has not taken place on a large scale in Sub- Saharan Africa. More mechanized farms have Colombia higher productivity, but constraints in land markets usually slow mechanization; without it, yields per hectare tend to be higher on smaller Venezuela, RB farms. industrial economies Outside agriculture there are massive (average) numbers of microenterprises and household ­ businesses (figure 9). These small units play –5 0 5 10 15 20 significant roles in job creation, even in high- share of total employment, % middle-income countries. They account for 97 percent of employment in the manufacturing sector in Ethiopia, but still for a sizable 39 per- Sources: World Development Report 2013 team estimates based on Bartelsman, Haltiwanger, and Scarpetta 2009b and Shiferaw and Bedi 2010. cent in Chile. In the services sector, their role is Note: The figure shows annual job flows. Data are from Argentina (1996–2001); Brazil (1997–2000); often more important. Even in Eastern European Canada (1984–97); Chile (1980–98); Colombia (1983–97); Estonia (1996–2000); Ethiopia (1997–2007); countries, where the private sector is only two Finland (1989–97); France (1989–97); Germany (1977–99); Hungary (1993–2000); Indonesia (1991–94); Italy (1987–94); Latvia (1983–98); Mexico (1986–2000); the Netherlands (1993–95); Portugal (1983–98); Romania decades old, microenterprises are the source of (1993–2000); Slovenia (1991–2000); Taiwan, China (1986–91); the United Kingdom (1982–98); the United 10 to 20 percent of employment in manufactur- States (1986–91, 1994–96); and República Bolivariana de Venezuela (1996–98). 12   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 enterprises, the gazelles, invest and earn higher Larger F I G U R E 8   firms pay higher wages returns.50 While large firms are more productive, they 6 were not all born large. In industrial countries, some of the more resounding successes, from Honda to Microsoft, started in garages. Many successful companies in developing countries also grew out of small household businesses. Thailand’s Charoen Pokphand Group, founded 4 in 1921 as a small seed shop in Bangkok by two estimates, % brothers, has grown into one of the largest multi­ national conglomerates in agribusiness, operat- ing in 15 countries and encompassing close to 100 companies. India’s Tata Group transformed 2 from a Mumbai-based family-owned trading firm in the late 19th century to a multinational conglomerate, comprising 114 companies and subsidiaries across eight business sectors on ­ several continents. Many of China’s success- 0 ful clusters, such as the footwear industry in 0 20 50 80 120 Wenzhou, also started from small family busi- ­ wage premium, relative to microenterprises (%) nesses working close to each other.51 small large Unfortunately, in many developing coun- tries, larger and older firms tend to be stagnant while smaller and younger enterprises are prone Source: Montenegro and Patrinos 2012 for the World Development Report 2013. to churning. A vibrant dynamic process is usu- Note: The figure uses 138 household and labor force surveys spanning 33 countries over 1991–2010. The horizontal axis reports the estimated wage premium of small firms (10 to 50 workers) and large firms (more ally absent. In Ghana, many firms were born than 50 workers) relative to microenterprises, controlling for worker characteristics. large and showed little growth over 15 years; in Portugal, by contrast, many firms born as mi- croenterprises grew substantially.52 The major- ing and 30 to 50 percent of employment in ser- ity of firms in India is also born small, but they vices. The large numbers of economic units are tend to stay small, without displaying much associated with a very wide dispersion of total variation in employment over their life cycle. factor productivity. In India, even within nar- A revealing comparison involves the size of rowly defined sectors, a manufacturing plant 35-year old firms relative to their size at birth. at the 10th percentile of the distribution gener- In India, the size declines by a fourth; in Mex- ates 22 times less output than a plant in the 90th ico, it doubles. In the United States, it becomes percentile would produce with the same inputs. 10 times bigger.53 The potential gains from This pattern is similar in a number of Latin greater entrepreneurial vibrancy, and from a American countries. By comparison, the ratio is more substantial reallocation of labor from 1 to 9 in the United States.47 low- to high-productivity units, are sizable.54 While microenterprises have lackluster per- But helping those gains materialize is a daunt- formance as a group, they are also very diverse. ing task. Microenterprises and household businesses are a means of survival for the poor and a way of di- Jobs are who we are versifying out of farming activities. On ­ average, their owners do not earn much.48 But in middle- Having, or not having, a job can shape how peo- income countries, many among the owners of ple view themselves and relate to others. While micro- and small enterprises are as entrepreneur- some jobs can be empowering, in extreme cases ial as their peers in industrial countries. Their a lack of job opportunities can contribute to vi- weak performance may be due to an adverse olence or social unrest. Youth may turn to gangs investment environment—for example, limited to compensate for the absence of identity and access to credit.49 Yet a small number of micro- belonging that a job might provide. In Ecua- Moving jobs center stage    13 The F I G U R E 9   employment share of microenterprises is greater in developing countries Ethiopia Egypt, Arab Rep. India Bolivia Colombia Ghana Mexico Venezuela, RB Argentina Poland Turkey Hungary South Africa Uruguay Czech Republic Slovenia Chile Romania Vietnam industrial countries (average) 0 20 40 60 80 100 share of employment, % manufacturing sector services sector Sources: World Development Report 2013 team estimates and EUROSTAT. Note: Microenterprises are firms, formal or informal, with fewer than 10 workers. Data for developing countries are from Argentina (2006–10), Bolivia (2005, 2007), Chile (2006, 2009), Colombia (2009), the Czech Republic (2005–07), the Arab Republic of Egypt (2006), Ethiopia (1999), Ghana (1991), Hungary (2007–08), India (2004, 2009), Mexico (2004–10), Poland (2005–07), Romania (2005–07), Slovenia (2005–07), South Africa (2005–07), Turkey (2006–10), Uruguay (2009), República Bolivariana de Venezuela (2004–06), and Vietnam (2009). Data for industrial countries are from Austria, Belgium, Denmark, Finland, France, Germany, Greece, Italy, Luxembourg, the Netherlands, Norway, Portugal, Spain, Sweden, and the United Kingdom over 2005–07. dor, for instance, they did so “because they were whose fathers did not have formal sector jobs searching for the support, trust, and cohesion— were significantly less likely to have such jobs social capital—that they maintained their fami- themselves.58 lies did not provide, as well as because of the lack The distribution of jobs within society—and of opportunities in the local context.�55 perceptions about who has access to opportu- The workplace can be a place to encounter nities and why—can shape expectations for the new ideas and interact with people of different future and perceptions of fairness. Children’s genders or ethnicities. Bosnians interviewed aspirations may be influenced by whether their in the late 1990s commented that “the area in parents have jobs and the types of jobs they which there is the greatest support for ethnic co- have. The Arab Spring was not merely about operation is in the workplace.�56 Business people employment. But disappointment, especially in Trinidad and Tobago reported that they in- among youth, about the lack of job opportuni- teracted with people of a wider range of ethnic- ties and frustration with the allocation of jobs ities at work than they did in their social lives.57 based on connections rather than merit echoed Networks can also exclude. In Morocco, people across countries. 14   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 F I G U R E 10   People who are unemployed, or do not have motivating jobs, participate less in society a. Active membership and unemployment b. Active membership and motivating job 0.1 –0.02 N 0 N 0.01 marginal probability N marginal probability –0.1 N N 0 –0.2 N –0.3 –0.01 N –0.4 N –0.5 –0.02 –0.6 high upper lower low high upper lower low income middle middle income income middle middle income income income income income Source: Wietzke and McLeod 2012 for the World Development Report 2013. Note: The vertical axis shows the probability of the respondent being an active member of one or more of nine types of associations, controlling for the income, education, and demographic characteristics of respondents. In panel a, the probability is linked to being unemployed, and panel b to having a job characterized as cognitive, creative, or independent. The vertical lines indicate the 95 percent confidence interval of the estimated probability. Jobs influence how people view themselves, The nature of jobs matters as well. Jobs that how they interact with others, and how they empower, build agency, and respect rights are perceive their stake in society.59 Jobs also can associated with greater trust and willingness to have collective consequences. They can shape participate in civil society. Jobs that create eco- how societies handle collective decision making, nomic and social ties may build incentives to manage tensions between diverse groups, and work across boundaries and resolve conflict. avoid and resolve conflicts. The relationship is And if people believe that job opportunities are not immediate or direct, however. Jobs are only available to them either now or in the future, one factor contributing to the capacity of socie- their trust in others and their confidence in in- ties to manage collective decision making peace- stitutions may increase. Ultimately, jobs can in- fully. And social cohesion can in turn influence fluence social cohesion through their effects on jobs by shaping the context in which entrepre- social identity, networks, and fairness. neurs make business decisions. Trust beyond one’s own group and civic en- gagement are two indicators of social cohesion. Valuing jobs Unemployment and job loss are associated with lower levels of both trust and civic engagement Not all forms of work are acceptable. Activities (figure 10). While causality is difficult to es- that exploit workers, expose them to dangerous tablish, there is more than just a correlation at environments, or threaten their physical and stake. Indonesian men and women who were mental well-being are bad for individuals and working in 2000 but not in 2007 were less likely societies alike. Child prostitution and forced la- to be participating in community activities than bor contravene principles of human dignity and those still at work. And those who were working undermine individual and collective well-being. in 2007 but not in 2000 were significantly more Today, an estimated 21 million people globally likely to be involved in the community than are victims of bonded labor, slavery, forced pros- those who were still out of work.60 titution, and other forms of involuntary work.61 Moving jobs center stage    15 In 2008, 115 million children between the ages household income contributed by women of- of 5 and 17 were involved in hazardous work.62 ten results in improvements in children’s edu- International norms of human rights and labor cational attainment and health. In Bangladesh, standards reject forced labor, harmful forms of where the garment industry employs women child labor, discrimination, and the suppression in large numbers, the opening of a garment of voice among workers. factory within commuting distance of a vil- Beyond rights, the most obvious outcome lage is seen as a signal of opportunity and leads of a job is the earnings it provides to its holder. to increased schooling for girls.64 Among dis­ These earnings can be in cash or in kind and advantaged castes in Southern Indian villages, may include a range of associated benefits. an increase of US$90 in a woman’s annual in- Other characteristics, such as stability, voice, come is estimated to increase schooling among and fulfillment at work, also affect subjective her children by 1.6 years.65 well-being. Several of these dimensions of jobs Similarly, a job created or sustained through have been combined into the concept of De- foreign direct investment (FDI) matters for cent Work, introduced by the International La- other jobs, and thus for other people. With the bour Organization (ILO) in 1999.63 Defined as investment come knowledge and know-how. “opportunities for women and men to obtain These raise productivity not only in the foreign decent and productive work in conditions of subsidiary but also among local firms interact- freedom, equity, security and human dignity,� ing with the subsidiary or operating in its vicin- this concept has been used by many govern- ity. Such knowledge spillovers are sizable in low- ments to articulate their policy agendas on and middle-income countries.66 Conversely, a jobs. The concept of Decent Work has also been job in a protected industry that needs to be sup- embraced by the United Nations and several ported through transfers (either by taxpayers international organizations and endorsed by or by consumers) generates a negative spillover, numerous global forums. even more so when the need for protection is As jobs provide earnings, generate output, associated with the use of outdated technology and influence identity, they shape the well- that results in high environmental costs. being of those who hold them—and they also Jobs can also affect other people by shaping affect the well-being of others. To understand social values and norms, influencing how groups how much jobs contribute to development, it is coexist and manage tensions. In Bosnia and necessary to assess these effects—the spillovers Herzegovina and the former Yugoslav Republic from jobs. Jobs that generate positive spillovers of Macedonia, surveys found that the number of have a greater value to society than they have people willing to work together or do business to the individual who holds the job, while the with someone of a different ethnicity was greater opposite is true when spillovers are negative. In- than the number of people in favor of intereth- tuitively, many people have notions about such nic cooperation in schools or neighborhoods.67 broader payoffs. When asked about their most And in the Dominican Republic, a program tar- preferred jobs, respondents in China, Colombia, geted to youth at risk shows that jobs can change Egypt, and Sierra Leone give different answers behaviors with positive implications for society. from those they offer when asked to identify Participation in the Programa Juventud y Em- the most important jobs to society (figure 11). pleo (Youth and Employment Program), which Working as a civil servant or as a shop owner is provides a combination of vocational and life generally preferred by individuals, while teach- skills training, reduced involvement in gangs, ers and doctors are quite often mentioned as the violence, and other risky behaviors.68 most important jobs for society. For the same level of earnings and benefits, Who gets a job makes a difference too, and the larger the positive spillovers from a job, the not just for individuals. In a society that values more transformational the job can be, and the poverty reduction, jobs that take households greater its value to society. In everyday parlance, out of hardship generate a positive spillover, good jobs are those that provide greater well- because they improve the well-being of those being to the people who hold them. But good who care. Female employment also matters be- jobs for development are those with the highest yond the individual. An increase in the share of value for society. Understanding these wider 16   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 F I G U R E 11   Views on preferred jobs and most important jobs differ a. China b. Egypt, Arab Rep. 50 50 D 40 40 social value social value 30 T C 30 F T 20 20 F D 10 10 C S S 0 0 0 10 20 30 40 50 0 10 20 30 40 50 individual value individual value c. Colombia d. Sierra Leone 50 50 D 40 40 D D social value social value 30 30 20 T 20 F F T C 10 10 CS S 0 0 0 10 20 30 40 50 0 10 20 30 40 50 individual value individual value C D F S T civil doctor farmer shop teacher servant owner Sources: Bjørkhaug and others 2012; Hatløy and others 2012; Kebede and others 2012; and Zhang and others 2012; all for the World Development Report 2013. Note: The figure shows the share of respondents who would want the job for themselves (individual value) and those who think the job is good for society (social value). pay­offs to jobs has shaped recent development tribute to common goals, such as poverty reduc- thinking.69 tion, environmental protection, or fairness. Spillovers from jobs can be identified across Because a job can affect the well-being of all three transformations (figure 12). Some di- others as well as that of the jobholder, two jobs rectly affect the earnings of others, as when a job that may appear identical from an individual is supported through government transfers, or perspective could be different from a social per- restrictive regulations that reduce employment spective (figure 13). The individual perspective opportunities for others. Other spillovers take provides a useful starting point, because it often place through interactions: in households in the coincides with the social perspective. A high- case of gender equality, at the workplace when paying job in Bangalore’s information technol- knowledge and ideas are shared, or in society ogy sector is probably good for the worker; it is more broadly in the case of networks. Spillovers also good for India because it contributes to the also occur when jobs and their allocation con- country’s long-term growth. In other cases, the Moving jobs center stage    17 two perspectives may conflict. For instance, Viet- nam’s poverty rate declined with unprecedented F I G U R E 12   Some jobs do more for development speed in the 1990s when land was redistributed to farmers and agricultural commercialization Jobs was liberalized.70 From the individual perspec- Jobs in connected to Jobs that are global markets tive, farming jobs involve difficult working con- functional environmentally cities benign ditions, substantial variability in earnings, and no formal social protection. But they can make a major contribution to development, as a ticket DEVELOPMENT out of poverty for many. Conversely, bloated Jobs for Jobs that give public utilities often offer a range of privileges to the poor a sense of fairness their employees even if the utilities themselves LIVING PRODUCTIVITY SOCIAL STANDARDS COHESION provide only limited coverage and unreliable Jobs that Jobs that services and are obstacles to economic growth empower link to women networks and poverty reduction. Such jobs may look ap- pealing from an individual perspective, but are Jobs that do not Jobs that less so to society. shift burden shape social to others JOBS identity Source: World Development Report 2013 team. Jobs agendas are diverse . . . but connected the biggest payoff may be for jobs that reduce Jobs challenges are not the same everywhere. poverty or defuse conflict. Certainly, the level Creating more jobs may be a universal goal, but of development matters. The jobs agenda is not the types of jobs that can contribute the most the same in an agrarian economy as in one that to development depend on the country context. is rapidly urbanizing. It is bound to be different Jobs that connect the economy to the world may still in countries already grappling with how far matter the most in some situations; in others, the formal economy can be extended. The F I G U R E 13   individual and social values of jobs can differ urban job connected to a global value chain for a woman job offering an opportunity to a agglomeration young person effects informal job giving a chance to social global spillover a poor person identity integration job in a protected sector using outdated social value poverty sense gender technology of fairness equality reduction individual burden value shifted individual environmental value cost Source: World Development Report 2013 team. 18   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 But the nature of good jobs for development jobs that do no environmental damage have in a particular context is not simply a function particularly positive development impacts. of income per capita. It may be influenced by conflict that is ongoing or still reverberating. • Formalizing countries. Large and growing ur- ban populations generally lead to more de- A country’s geography or its natural endow- veloped economies, where a fairly substantial ments can also be determining factors. Small is- proportion of firms and workers are covered land nations have unique jobs challenges, as do by formal institutions and social programs. resource-rich economies. Or demography may But further increasing formality to levels typi- be the key characteristic—witness the imposing cal of industrial countries involves tradeoffs but very different challenges in countries facing between living standards, productivity, and high youth unemployment and those with ag- social cohesion. There is a premium on jobs ing populations. that can be formalized without making labor too costly and on jobs that reduce the divide A typology of jobs challenges between those who benefit from formal insti- tutions and those who do not. A country’s level of development, institutional strength, endowments, and demography de- In some countries, the jobs challenge is fine where the development payoff from jobs shaped by demography and special circum- is greatest. The jobs agenda in one country will stances affecting particular groups. thus be different from that in another country, depending on their dominant features. The •  In countries with high youth unemployment young people do not see opportunities for challenges facing countries as they move along the future. Many of these countries have large the development path are illustrated by the youth bulges, which can put downward pres- agrarian, urbanizing, and formalizing cases: sure on employment and earnings. Many also • Agrarian countries. Most people are still en- have education and training systems that are gaged in agriculture and live in rural areas. not developing the kinds of skills needed by Jobs that improve living standards have a the private sector. On closer inspection, the substantial development payoff because of problem is often more on the demand side high poverty rates. Cities need to be more than the supply side, with limited compe­ functional to reap the benefits from agglom- tition reducing employment opportunities, eration and global integration, so jobs that especially in more skill-intensive sectors. In set the foundation for cities to eventually be- these settings, removing privilege in business come economically dynamic are good jobs entry and access to jobs is likely to have large for development. Even in the most optimistic development payoffs. scenario, however, it may take decades before • Aging societies also face generational issues, urbanization is complete, so increasing pro- but these stem from a shrinking working-age ductivity in agriculture is a priority. population and the high cost of providing and caring for a growing number of elderly • Urbanizing countries. Productivity growth in people. The impact of the declining working- agriculture has risen enough to free up large age population can be mitigated through pol- numbers of people to work in cities. Job icies for active aging, ensuring that the most opportunities for women, typically in light productive members of society, including manufacturing, can have positive impacts the highly skilled elderly, can work. Contain- on the household allocation of resources. ing the increase in pension, health care, and Jobs that deepen the global integration of long-term care costs can be achieved through urbanizing countries, especially in higher- reforms in program design, but these reforms value-added export sectors, are also good can be a source of social strain. for development. As countries urbanize, congestion, pollution, and other costs of Natural endowments, including geography, high density become increasingly serious, so and institutions can create unique jobs challenges. Moving jobs center stage    19 • Resource-rich countries may have substantial that would make the greatest contribution to foreign exchange earnings, but this wealth development in each case. This focus allows may not translate into employment creation for a richer analysis of the potential tradeoffs beyond the exploitation of natural resources. between living standards, productivity, and Indeed, the abundance of foreign exchange social cohesion in a specific context. It pro- can hamper the competitiveness of other ex- vides clues about the obstacles to job creation port activities. Some resource-rich countries and, ultimately, the priorities for policy makers distribute part of their wealth through trans- (figure 14). fers or subsidized public sector jobs, while relying on migrants to do menial work. This Migration of people—and of jobs approach can maintain living standards but at the expense of productivity growth and social The movement of people and jobs implies that cohesion. In those countries, jobs that sup- jobs challenges, while being country specific, port the diversification of exports can have also have a global scope. These processes have large development payoffs. implications for living standards and productiv- ity at both the sending and the receiving ends, • Small island nations, because of their size and and they can transform families and entire com- remoteness, cannot reap the benefits from munities, for better or for worse. Tradeoffs are agglomeration and global integration except inevitable, and coping with them only through through tourism. So the productivity spill- the policies of receiving countries alone may overs from jobs are limited, as are employ- prove unsatisfactory. ment opportunities outside basic services and At the turn of the 21st century, there were government. Outmigration offers an alter- more than 200 million international migrants native for improving living standards, while worldwide, nearly 90 million of them work- return migration and diaspora communities ers. Many migrants are temporary or seasonal can stimulate the diffusion of new business workers who eventually return home. Some ideas among locals. countries are mainly recipients, while others are • In conflict-affected countries, the most imme- sources, and yet others neither host nor send diate challenge is to support social cohesion. significant numbers of migrants (map 1). Some Employment for ex-combatants or young are large recipients either in absolute numbers men vulnerable to participation in violence (for instance, the United States) or in relative takes on particular importance. With fragile terms (Jordan and Singapore). Migrants from institutions and volatile politics, attracting Bangladesh, Mexico, and India represent a large private investment and connecting to global share of total migrants worldwide; Fiji, Jamaica, value chains may be out of reach for quite and Tonga have a large share of their population some time. Yet construction can boom even overseas. Figures for some of the smaller coun- in poor business environments, and it is la- tries are striking. For instance, about a fifth of bor intensive. Investments in infrastructure all Salvadorians live abroad, while more than 60 can not only support social cohesion through percent of the populations of Kuwait, Qatar, and their direct employment impact, they can also the United Arab Emirates are foreign-born.71 be a step in preparing for future private sector International migration increases the in- job creation. comes of migrants and their families through earnings and remittances. The majority of the These criteria are not mutually exclusive. studies find either no effect or a very small nega- Chad and the Democratic Republic of Congo tive effect on the labor earnings of locals in re- are both resource rich and conflict affected; ceiving countries. Migrants also contribute to Jordan and Armenia are formalizing and also global output if their productivity abroad is have high youth unemployment. Still, look- higher than it would be at home, which is usually ing through the jobs lens and focusing on the the case. They may even contribute to output in key features of the different country types can the sending country, as networks of migrants help identify more clearly the kinds of jobs and returnees channel investments, innovation, 20   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 F I G U R E 14   Good jobs for development are not the same everywhere Jobs challenge What are good jobs for development? Agrarian More productive smallholder farming economies Urban jobs connected to global markets Conflict-affected Jobs demobilizing combatants countries Jobs reintegrating displaced populations Jobs providing alternatives to confrontation Urbanizing Jobs providing opportunities for women countries Jobs moving the country up the export ladder Jobs not leading to excessive congestion Jobs integrating rural migrants Resource-rich Jobs supporting export diversification countries Jobs not subsidized through transfers Small island Jobs connected to global markets nations Jobs not undermining fragile ecosystems Countries with high Jobs not supported through rents youth unemployment Jobs not allocated on the basis of connections Formalizing Jobs with affordable social benefits countries Jobs not creating gaps in social protection coverage Aging Jobs keeping the skilled active for longer societies Jobs reducing the cost of services to the elderly Source: World Development Report 2013 team. and expertise. Social effects are more mixed. computer and information services, legal and On the positive side, migration connects people technical support, and other business services. from different cultures in ways bound to widen India was the pioneer, but other countries— their horizons. On the negative side, the separa- Brazil, Chile, China, and Malaysia, to name a tion from family and friends can be a source of few—have also seized the opportunity.72 distress and isolation. Migration may also bring The obvious winners of job migration are the racial prejudice and heighten social tensions in workers and entrepreneurs in countries to which host countries, especially when migrants are se- industries and splintered service jobs have mi- cluded in segregated occupations or neighbor- grated. This migration, along with the transfer hoods, preventing their integration in society. of new technologies and advanced management Jobs are on the move as well. The past four methods, contributes to productivity growth decades have been marked by the outsourcing of and higher living standards. The hidden win- manufacturing tasks from industrial countries ners of job migration are consumers worldwide. to the developing world, especially to East Asia The improved international division of labor in- (figure 15). More recently, the same pattern is creases the availability of goods and services and observable for service tasks. In fact, services are enhances the possibility of gaining from trade. the fastest-growing component of global trade. The clear losers are those who have seen their Developing countries are now exporting not jobs disappear because of the declining compet- only traditional services, such as transportation itiveness of their industries and services. Among and tourism, but also modern and skill-inten- the losers, many skilled workers find comparable sive services, such as financial intermediation, jobs without a substantial loss in salary, but oth- Moving jobs center stage    21 M AP 1  Only in some countries are migrants a substantial share of the population a. Immigrants, % of labor force Percent 0–1.99 2.00–4.99 5.00–9.99 10.00–14.99 15.00–100 no data b. Emigrants, % of native labor force Percent 0–1.99 2.00–4.99 5.00–9.99 10.00–14.99 15.00–100 no data This map was produced by the Map Design Unit of The World Bank. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of The World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. Sources: World Development Report 2013 team based on Özden and others 2011, and Artuc and others 2012, using census data around 2000. ers do not. Low-skilled workers or those with service is critically important for development, industry- or occupation-specific skills that are whether it is teachers building skills, agricul- no longer in demand are those who suffer most. tural extension agents improving agricultural productivity, or urban planners designing func- tional cities. Temporary employment programs Policies through the jobs lens for the demobilization of combatants are also justified in some circumstances. But as a general While it is not the role of governments to create rule it is the private sector that creates jobs. The jobs, government functions are fundamental for role of government is to ensure that the condi- sustained job creation. The quality of the civil tions are in place for strong private-sector-led 22   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 growth, to understand why there are not enough Manufacturing jobs have migrated away from F I G U R E 15   good jobs for development, and to remove or high-income countries mitigate the constraints that prevent the cre- ation of more of those jobs. a. High-income countries Government can fulfill this role through a 45 three-layered policy approach (figure 16): 40 • Fundamentals. Because jobs improve with 35 development, providing higher earnings and benefits as countries grow rich, a prerequisite percent 30 is to create a policy environment that is con- 25 ducive to growth. Macroeconomic stability, an enabling business environment, human 20 capital accumulation, and the rule of law are 15 among the fundamentals. Ensuring macro- 1970 1980 1990 2000 2008 economic stability involves containing volatil- ity and avoiding major misalignments of rela- manufacturing share of GDP tive prices. Adequate infrastructure, access to manufacturing share of employment finance, and sound regulation are key ingredi- b. Japan and the Republic of Korea ents of the business environment. Good nutri- 45 tion, health, and education outcomes not only improve people’s lives but also equip them for 40 productive employment. The rule of law in- 35 cludes protection of property rights and also the progressive realization of rights at work, to percent 30 avoid a situation where growth co­ exists with 25 unacceptable forms of employment. 20 • Labor policies. Because growth does not me- chanically deliver employment, a second layer 15 is to ensure that labor policies do not under- 1970 1980 1990 2000 2008 mine job creation and instead enhance the de- manufacturing share of GDP (Japan) velopment payoffs from jobs. But labor mar- manufacturing share of employment (Japan) ket imperfections should not be addressed manufacturing share of GDP (Republic of Korea) through institutional failures. Instead, they manufacturing share of employment (Republic of Korea) should remain on a range—a plateau—where c. Other East Asian countries negative efficiency effects are modest. Labor 45 policy should avoid two cliffs: the distortion- ary interventions that clog the creation of jobs 40 in cities and in global value chains, and the 35 lack of mechanisms for voice and protection percent for the most vulnerable workers, regardless of 30 whether they are wage earners. The first cliff 25 undermines the development payoffs from agglomeration and global integration; the 20 second leads to low living standards and a so- 15 cial cohesion deficit. 1991 1995 2000 2005 2008 • Priorities. Because some jobs do more for de- manufacturing share of GDP velopment than others, it is necessary to un- manufacturing share of employment derstand where good jobs for development lie, given the country context. More selective Sources: World Development Report 2013 team estimates based on data from the United Nations Indus- trial Development Organization (UNIDO) and the United Nations Statistics Division. policy interventions are justified when in- Note: Japan is not included in panel a. GDP = gross domestic product. centives are distorted, resulting in too few of Moving jobs center stage    23 those jobs. If this is the case, policies should remove the market imperfections and institu- Three F I G U R E 16   distinct layers of policies are needed tional failures that prevent the private sector from creating more good jobs for develop- ment. If the failures and imperfections can- not be clearly identified, or cannot be easily removed, offsetting them may be an option, Know your jobs challenge but the costs and benefits of doing so need to PRIORITIES Remove or offset the constraints be carefully assessed. Fundamentals: Ensuring the basics Stay on the efficiency plateau Avoid misguided interventions Macroeconomic stability. Volatility hurts em- LABOR POLICIES Provide voice and extend protection ployment and earnings, often immediately. Ac- cording to a recent estimate, a 1.0 percent decline in gross domestic product (GDP) is associated with an increase in the unemployment rate of Macroeconomic stability 0.19 percentage point in Japan, 0.45 percentage An enabling business environment point in the United States, and 0.85 percentage Human capital FUNDAMENTALS Rule of law and respect for rights point in Spain.73 In developing countries, where farming and self-employment are more preva- lent and income support mechanisms are more limited, the short-term impact of macroeco- Source: World Development Report 2013 team. nomic instability is less on open unemployment and more on earnings from work.74 Volatility can originate internally or be lead to an overvaluation of its currency, making caused by external shocks. Internally, it is often imports more affordable and exports less com- the outcome of unsustainable budget deficits petitive. Resource-rich countries face similar and lax monetary policy. But tight budgets and pressures for their currencies to appreciate, and rigid monetary policy rules may not be a magic the commodity booms of the last few years have wand. Budget deficits are more or less worri- only made these pressures stronger. Currency some depending on how quickly an economy is overvaluation can also happen in countries growing, whereas the independence of central where large volumes of foreign assistance are banks needs to be weighed against the overall needed to jump-start development, cope with coherence of the country’s development strat- natural disasters, or facilitate recovery after a egy. Assessing the soundness of macroeconomic conflict. An analysis of 83 developing countries management requires taking account of the between 1970 and 2004 confirms that aid fosters impact of fiscal and monetary policies on eco- growth (albeit with decreasing returns) but in- nomic growth.75 duces overvaluation and has a negative impact Volatility may also result from external shocks, on export diversification.77 including natural disasters and crises originat- An enabling business environment. Finance, ing abroad. Precautionary policies can cushion infrastructure, and business regulations set the those shocks, if and when they occur. Most often, quality of the investment climate and thus in- short-term stimulus or adjustment packages are fluence job creation by private firms. Access to needed—but these tend to be less effective in the finance, a chief constraint to business expan- developing world than in developed countries sion in countries in every development phase, because of lower multiplier effects.76 is the top constraint in low- and upper-middle- Avoiding exchange rate misalignment is nec- income countries (figure 17). Financial markets essary to sustain a vibrant export sector—and have the potential to allocate resources toward thus to create jobs connected to international more productive uses, thwart the channeling markets and global value chains. Surges in a of resources to those with political connections country’s foreign exchange earnings generally or economic power, and expand financial in- 24   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 F I G U R E 17   Finance and electricity are among the top constraints faced by formal private enterprises Firm size Income level Constraint Small Medium Large Low Lower middle Upper middle High All Access to �nance Power shortage Lack of skills Informal competition Tax rates most severe second-most severe third-most severe Source: IFC, forthcoming. Note: The analysis is based on World Bank enterprise surveys covering 46,556 firms in 106 countries. Small firms have fewer than 20 employees, medium firms have 21–99, and large firms 100 and more. clusion. But regulatory oversight is needed to takes to comply with regulations or to receive ensure transparency and competition in how permits.81 Business regulations also affect com- funds are allocated.78 The financial crisis of 2008 petition and thus the pressure to innovate and reopened heated debates on the appropriate increase productivity. Across countries, regula- regulation of the financial sector and the need tions on business entry are inversely correlated to balance prudence and stability with innova- with productivity and firm creation, with stron- tion and inclusion. ger effects in sectors that have higher rates of Access to affordable and quality infrastruc- entry.82 In Mexico, easing entry requirements ture is a prerequisite for firms to operate. Power increased business registration and employ- shortages are the number-two constraint to ment and drove down consumer prices, largely firm growth and job creation mentioned by through creating new firms rather than formal- entrepreneurs the world over—and number izing informal firms.83 one in low-income countries. Telecommunica- Human capital. Good outcomes in nutrition, tions allow for a better flow of information with health, and education are development goals suppliers and customers, and the internet and in themselves, because they directly improve mobile technology facilitate the spread of new people’s lives. But they also equip people for ideas. Roads provide greater access to markets, productive employment and job opportuni- as do ports and airports.79 The way infrastruc- ties—and through this channel, human capi- ture is regulated is important as well. Inad- tal drives economic and social advances. There equate pricing policies and regulations amplify is robust evidence from throughout the world the gap in needed infrastructure services. In that an additional year of schooling raises earn- many countries, monopolies based on politi- ings substantially, and that this earnings pre- cal connections have led to reduced quantities mium reflects the higher productivity of more of infrastructure services at higher prices and educated workers.84 Together, nutrition, health, lower quality.80 and education combine to form human skills Business regulation also affects the oppor- and abilities that have been powerfully linked to tunities for businesses to grow and create jobs. productivity growth and poverty reduction in Regulations can increase the cost of doing busi- the medium to longer run.85 Also, better health ness, in money or in time needed to comply. brings, directly, higher labor productivity. As Steps taken to meet requirements or to pay fees such, human capital is a fundamental ingredi- are a burden for businesses, as are delays or dis- ent for desirable job outcomes. cretionary decisions, such as those for permits Human capital formation is cumulative. or licenses. There is great variation across firms Of crucial importance are adequate health in the same location with regard to the time it and nutrition during “the first 1,000 days,� Moving jobs center stage    25 from conception to two years of age. Brain consistently find crime and corruption to be ob- devel­opment in this time period affects physi- stacles to conducting business.95 cal health, learning abilities, and social be- An effective judicial system is a key institu- havior throughout life.86 Ensuring adequate tion for enforcing property rights and reduc- nutrition, health, and cognitive stimulation ing crime and corruption. An independent, through a nurturing environment from the accountable, and fair judiciary can contribute womb through the first years raises returns to to private sector growth and job creation by en- later child investments significantly.87 While forcing the rules that govern transactions and foundations are laid early on, human capital by helping ensure that the costs and benefits of and skills continue to be formed throughout growth are fairly distributed. The justice system childhood and young adulthood. School- can enforce contracts, reduce transaction costs ing is fundamental for the further develop- for firms, and create a safe and more predictable ment of cognitive and social skills until the business environment.96 And effective courts in- end of the teenage life. Social skills remain crease the willingness of firms to invest.97 malleable through adolescence and the early An institutional environment that respects adult years.88 Young adults can continue rights is an important ingredient of the rule of into more specialized skill-building, includ- law and a foundation for good jobs for develop- ing at tertiary levels, but success depends on ment. The ILO’s core labor standards provide a whether the generic skills needed to learn floor in the areas of child labor, forced labor, and adapt to different tasks and problem- discrimination, and freedom of association solving environments have been acquired. and collective bargaining.98 Health and safety These general skills are especially important in at work also call for attention by governments more dynamic economic environments. and employers. Ensuring that standards are Unfortunately, the evidence shows that many applied in practice requires providing access to countries are falling short in building up the information to workers and employers. It also human capital of their children and youth. The implies expanding legal coverage to workers in quality of delivery systems has often failed to jobs that fall outside formal laws and regula- keep pace with the expansion of access to basic tions. Associations of informal workers can social services. In a large majority of develop- inform them about their rights, help them use ing countries that took part in the Programme legal mechanisms, and offer them collective for International Student Assessment (PISA) in voice.99 2009, at least one-fifth of 15-year-old students were functionally illiterate (not reaching at least Labor policies: Avoiding the two cliffs level 2 in the PISA reading assessment).89 The rule of law. Across countries, the pres- A malfunctioning labor market may prevent ence of institutions that protect property rights, economic growth from translating into more uphold the rule of law, and rein in corruption is and better jobs. Traditional analyses focus on associated with higher levels of development.90 labor supply, labor demand, and their matching Property rights foster private sector growth by to explain why there may not be enough employ- allowing firms to invest without the fear that ment, or not enough wage employment in the their assets will be stolen or confiscated.91 The case of developing countries. By not addressing ability to enforce contracts widens the circle labor market imperfections, or by creating them, of potential suppliers and customers, as per- labor policies can indeed constrain job creation, sonal connections become less important in even seriously. In many cases, however, the con- establishing trust.92 The rule of law has direct straints to creating transformational jobs are not implications for the growth of firms and jobs. connected to the labor code. The low productiv- Entrepreneurs who believe their property rights ity of smallholder farming in agrarian econo- are secure reinvest more of their profits than mies is probably more closely related to failures those who do not.93 Conversely, rampant crime in agricultural research and extension. And the and violence are likely to drive firms away and lack of competition in technologically advanced discourage domestic and foreign investment.94 activities that could boost the demand for skilled Across countries, investment climate surveys work in countries with high youth unemploy- 26   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 ment is more likely to stem from cronyism and youth, women, and the less skilled. In Colombia political favoritism. and Indonesia, minimum wage increases had There is no consensus on what the content only a modest overall effect but the employment of labor policies should be. Views are polarized, impact was stronger for young workers.101 Reg- reflecting differences in fundamental beliefs. To ulations more clearly affect job flows, creating some, labor market regulations and collective “stickiness� in the labor market and slowing the 102 bargaining are sources of inefficiency that re- pace of labor reallocation.­ While this hinders duce output and employment, while protecting economic efficiency, the evidence on produc- insiders at the expense of everyone else. In this tivity is fairly inconclusive, though admittedly view, unemployment insurance and active labor scarce.103 market programs create work disincentives and In developing countries, collective bargain- are a waste of money. To others, these policies ing does not have a major impact outside the provide necessary protection to workers against public sector and activities characterized by the power of employers and the vagaries of the limited competition, where there are rents to market. They can even contribute to economic share.104 Unions consistently raise wages for efficiency by improving information, insuring workers. Studies place this premium in the 5 to against risks, and creating conditions for long- 15 percent range in Mexico; around 5 percent term investments by both workers and firms. in Korea; and at 10 to 20 percent in South Af- Advocates of both views can find examples rica.105 The costs in terms of reduced jobs are to support their positions. Those who see labor not so clear, however. In some countries, though policies and institutions as part of the problem not all, the tradeoff seems to be lower employ- point to the impressive long-term job creation ment, but even then the magnitudes are rela- record of the United States, a country with lim- tively small. The limited evidence on union ef- ited interventions in the labor market. They also fects on productivity is also mixed.106 The main point to the protective job security rules that challenges are extending voice to those who are have impeded young people from finding work not wage earners, so that the constraints facing in many North African and Southern European their farms and microenterprises can be ad- countries. By contrast, those who see labor poli- dressed, and organizing collective bargaining in cies as part of the solution point to job-sharing a way that enhances productivity. as decisive in Germany’s relative success in Active labor market programs, such as train- weathering the financial crisis. ing, employment services, wage subsidies, and A careful review of the actual effects of labor public works, have a mixed record.107 When they policies in developing countries yields a mixed are not well grounded in the needs and realities picture. Most studies find that impacts are mod- of the labor market or when administration is est—certainly more modest than the intensity poor and not transparent, they are of little use of the debate would suggest.100 Across firm sizes or even worse. When they are well designed and and country levels of development, labor poli- implemented, they can help facilitate job match- cies and regulations are generally not among the ing, mitigate the negative impacts of economic top three constraints that formal private enter- downturns, and fill the gap when employers prises face. Excessive or insufficient regulation or workers underinvest in training (figure 18). of labor markets reduces productivity. But in Even when this is the case, though, effects tend between these extremes is a plateau where ef- to be modest, so expectations about what active fects enhancing and undermining efficiency can labor market policies can achieve need to be be found side by side and most of the impact held in check. is redistributive, generally to the advantage of Social insurance coverage is limited even in middle-aged male workers (as opposed to own- the most formalized developing countries. Un- ers of capital, women, and younger workers). employment insurance can help workers man- In most countries that have been studied, age the risks of job loss, but it can also weaken job security rules and minimum wages have a job search efforts. When unemployment insur- small effect on aggregate employment. These ance, pensions, health care, and other benefits rules offer benefits for those who are covered, are financed through the payroll, high contri- while negative effects tend to be concentrated on bution rates can create hiring disincentives. In Moving jobs center stage    27 developing countries where formal sectors are small, funding these programs through general Combining F I G U R E 18   work and training increases the taxation is increasingly discussed,108 but any success rates of programs taxes create distortions. In the end, there is no substitute for affordable social protection ben- 0.15 efits that are valued by workers. The main issue is coherently integrating social protection and 0.10 social assistance to minimize gaps and overlaps. In sum, labor policies and institutions can success indicator 0.05 improve labor market information, manage risk, and provide voice. But these advantages 0 can come at the expense of labor market dy- namism, reduced incentives for job creation –0.05 and job search, and a gap in benefits between –0.10 the covered and uncovered. The challenge is to set labor policies on a plateau—a range where –0.15 regulations and institutions can at least partially in-class workplace in-class training in-class and address labor market imperfections without training only training only and workplace workplace training training combined reducing efficiency. Labor market rules that combined plus other services are too weak or programs that are too modest or nonexistent can leave problems of poor in- Source: Fares and Puerto 2009. formation, unequal power, and inadequate risk Note: The figure shows the correlation coefficient between type of training and reported success of a management untreated. In contrast, rules that program, with success defined as improving employment or earnings and being cost-effective. are too stringent and programs that are too am- bitious can compound market imperfections with institutional failures. the health insurance program for the poor in The focus on good jobs for development of- Vietnam are encouraging in this respect.109 This fers some insights to assess where the edges of cliff may be less visible than excessive labor mar- the plateau, the cliffs, may lie. At one end of the ket rigidity, but it is no less real. plateau are labor policies that slow job creation in cities, or in global value chains, and make Priorities: Realizing the development countries miss out on jobs supporting agglom- payoffs from jobs eration effects and knowledge spillovers. Forgo- ing the development payoffs from urbanization In addition to ensuring that the fundamen- and global integration would be a consequence tals support growth and that labor policies are of falling off the cliff. This is not necessarily an adequate, decision makers can help realize the ­ argument for minimum regulation. There is also development payoffs that come from jobs. Some scope for arrangements strengthening spatial jobs do more than others for living standards, coordination, and thus increasing efficiency, as productivity, and social cohesion. What those suggested by China’s recent experience with col- jobs are depends on the country context—its lective bargaining. level of development, demography, endow- At the other end of the plateau, the absence ments, and institutions. In some circumstances, of mechanisms for voice and protection for there will be no constraints to the emergence of those who do not work for an employer, or do so good jobs for development, and no specific pol- in the informal sector, is also a concern. Extend- icy will be needed. In others, governments can ing voice for workers who are often among the support the private sector in creating more of poorest may result in higher living standards. these jobs. Sometimes this can be achieved by re- Limiting abuses by employment intermediar- moving constraints that impede the creation of ies should enhance efficiency, and building in- jobs with high development payoffs. When this clusive social protection systems can contribute is not possible, policies can be more proactive to greater social cohesion. The experience of and bypass the constraints, provided that the India’s Self Employed Women’s Association and gains to society from doing so outweigh the cost. 28   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 F I G U R E 19   A decision tree can help set policy priorities No intervention is needed. Remove the Step 4 constraints. Step 1 Can the constraints YES What are good jobs be removed? for development? O set NO the constraints. Are there enough YES Can the constraints YES Can the constraints YES of these jobs? be identi ed? be o set? Design NO NO NO engagement strategy. Step 2 Step 3 Step 5 Source: World Development Report 2013 team. A simple approach to setting policy priorities In the absence of gaps of this sort, it is difficult follows five steps (figure 19): to justify government interventions beyond establishing the fundamentals and adopting • Step one: What are good jobs for development? adequate labor policies. Assessing the development payoffs from Data and analysis can be used to iden- jobs in a particular country context is the first tify misaligned incentives, indicated by a step in identifying priorities. The nature of gap between the individual and the social those jobs varies with the characteristics of value of jobs. Several research areas deal with the country, including its phase of develop- these gaps. For instance, the tools of pub- ment, demography, endowments, and insti- lic finance can measure the tax burden that tutions. Jobs challenges are not the same in applies to capital and labor and assess the cross- agrarian economies, resource-rich countries, subsidization between individuals or firms. conflict-affected countries, or in countries The methods of labor economics can uncover with high youth unemployment. And the jobs gaps between the actual earnings of specific with the greatest development impact differ groups of workers and their potential earnings, as well, resulting in diverse jobs agendas. or between the social and individual returns to •  Step two: Are there enough of these jobs? A schooling. Poverty analyses help in identifying country may or may not face constraints in the kind of jobs that are more likely to provide creating good jobs for development. For ex- opportunities to the poor, or the locations ample, light manufacturing can offer employ- where job creation would have a greater im- ment opportunities for women, with signifi- pact on reducing poverty. Productivity stud- cant impacts on poverty. If a boom is under ies allow for quantifying the spillovers from way, the development value of new manufac- employment in foreign-owned investment turing jobs might materialize. But it might companies, or in cities. Environmental studies not if, for example, inadequate urbanization shed light on the carbon footprint and pollu- policies limit the establishment of new firms. tion created by various types of jobs. And val- Moving jobs center stage    29 ues surveys can discover which types of jobs But there are cases when constraints can nei- provide social networks and social identity. ther be removed nor offset. An engagement strat- egy involving a deeper analysis of the options •  Step three: Can the constraints be identified? and buy-in by key stakeholders is needed then. The gaps between the individual and social Policy making to remove or offset constraints values of specific types of jobs indicate un- needs to be selective and supported by good exploited spillovers from jobs. The gaps typi- public finance principles. The costs and benefits cally arise from market imperfections and in- of policy options need to be assessed, but calcu- stitutional failures that cause people to work lations are different when the overall develop- in jobs that are suboptimal from a social point ment impact is the guiding objective. An em- of view, lead firms to create jobs that are not ployment program to demobilize ex-combatants as good for development as they should be, or in a c­ onflict-affected country could be assessed connect people less through jobs than would in terms of whether the earnings gains of par- be socially desirable. But identifying those ticipants justify the program costs, but a full constraints is not always easy. For instance, ­ accounting should also incorporate the poten- a broad set of cultural, social, and economic tially positive effects from reintegration and forces may result in insufficient employment peace building. In the Democratic Republic of opportunities for women. Similarly, the ob- Congo, the cost of an integration program for stacles to more jobs in cities could be in the ex-combatants was about US$800 per benefi- land market, or in the institutional arrange- ciary.110 Such a program would likely be judged ments to coordinate urban development, as cost inefficient by traditional standards. or in the ability to raise revenue to finance Whether or not it is still worth implementing infrastructure. depends on the value policy makers attach to so- cial cohesion benefits. These benefits should be •  Step four: Can the constraints be removed? If stated for the policy decision to be transparent. the institutional failures and market imper- fections leading to misaligned incentives can be identified, reforms should be considered. It Diverse jobs agendas, diverse policy is a good economic principle to target reforms priorities on the failures and imperfections at the root Some countries have successfully set policy to of the problem. Where reforms are technically bring out the development payoffs from jobs, in and politically feasible, policy makers can di- ways that provide a model to others. rectly tackle the major constraints hindering As an agrarian country, in the 1990s Viet- the creation of more good jobs for develop- nam concentrated on increasing productivity ment by the private sector. in agriculture, freeing labor to work in rural off-farm employment and eventually support- • Step five: Can the constraints be offset? ­Reforms ing migration to cities. In 1993, more than 70 might not be feasible, technically or politically. percent of employment was in agriculture, 58 Or perhaps the constraints for jobs are not percent of the population lived in poverty, and identifiable. An alternative then is to adopt famine was still a real concern.111 Two decades offsetting policies that can restore the incen- later, Vietnam is the second-largest exporter tives for job creation. For instance, if a dif- of rice and coffee; the largest exporter of black fuse but entrenched set of norms and beliefs pepper and cashew nuts; and a top exporter of makes it difficult for women to work, efforts tea, rubber, and seafood products. Poverty has could aim at increasing their employability declined dramatically. Combined with a strong through targeted investments in social and emphasis on agricultural extension, land reform physical infrastructure (box 1). Similarly, if and deregulation led to rapidly growing agri- politically charged regulations slow down the cultural productivity on very small farm plots. reallocation of labor toward more productive These policies were part of a broader package activities, urban infrastructure and logistics of reforms, or Doi Moi, that took Vietnam from could enhance the attractiveness of jobs in central planning to a market economy with a cities and jobs connected to world markets. socialist orientation.112 Policies also aimed at 30   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 BOX 1  How does women’s labor force participation increase?  Some developing countries have experienced important increases returns to their earnings. These investments can be categorized into in women’s labor participation over a relatively short period of time. three groups. They can address shortages in the availability of ser- Nowhere has the change been faster than in Latin America. Since vices (such as lack of electricity or daycare facilities) that force women the 1980s, more than 70 million women have entered the labor to allocate large amounts of time to home production. They can force, raising the female labor participation rate from 36 percent to make it easier for women to accumulate productive assets, such 43 percent. In Colombia, the rate increased from 47 percent in 1984 as education, capital, and land, facilitating their entry into high-­ to 65 percent in 2006. By contrast, in the Middle East and North productivity market activities. And they can remove norms or regula- Africa, women’s labor force participation has only grown by 0.17 tions that imply biased or even discriminatory practices, preventing percentage points per year over the last three decades. women from having equal employment opportunities. Recent research attributes this rapid transformation to increases There are successful experiences with targeted investments in labor force participation among married or cohabiting women and interventions of each of these three sorts. Public provision or with children, rather than to demographics, education, or business subsidization of child care can reduce the costs women incur at cycles. Changes in social attitudes contributed to the transforma- home when they engage in market work. Examples include pub- tion, but this is a complex area with limited scope—and justifica- licly provided or subsidized day care such as Estancias Infantiles in tion—for direct policy intervention. For instance, women’s partici- Mexico, Hogares Comunitarios in Colombia, and similar programs pation rates are very low in the West Bank and Gaza, particularly in Argentina and Brazil. Improvements in infrastructure services— among married women. But this cannot be mechanically attributed especially in water and electricity—can free up women’s time to religion, as countries like Indonesia have high participation rates. spent on domestic and care work. Electrification in rural South Other social norms and regulations prevent women from participat- Africa, for instance, has increased women’s labor force participa- ing, despite their willingness and capacity to do so. tion by about 9 percent. Correcting biases in service delivery insti- While the scope to influence social attitudes is limited, evidence tutions, such as the workings of government land distribution and suggests that public policies and programs in other areas have an registration schemes, allows women to own and inherit assets. important role to play. It also suggests that a combination of tar- Finally, the use of active labor market policies, the promotion of geted investments and interventions in social and physical infra- networks, and the removal of discriminatory regulations are impor- structure can modify women’s labor force participation and the tant to make work more rewarding for women. Sources: World Development Report 2013 team based on Amador and others 2011, Chioda 2012, and World Bank 2011d. creating employment opportunities outside ag- expressed satisfaction with their social integra- riculture. The country opened to foreign inves- tion, and 85 percent of community members tors, first in natural resource exploitation and felt there was trust between the two groups.115 light manufacturing, and then more broadly in While ex-combatants were only a small share of the context of its accession to the World Trade Rwanda’s population of 10 million, their reinte- Organization in 2007. Registered FDI increased gration had payoffs for social cohesion. Rwanda fourfold in just two years, from 1992 to 1994; has built on this start by rejuvenating the private over the past five years, FDI inflows exceeded 8 sector through reforms of institutions and busi- percent of GDP.113 ness regulations.116 The coffee industry has cre- Rwanda, a conflict-affected country, has re- ated thousands of new jobs.117 bounded after the ethnic conflict and destruction Chile, a resource-rich country, has managed of the mid-1990s. By 2000, Rwanda’s economy its copper riches in a way compatible with job had returned to precrisis levels as a result of the creation in nonresource sectors. Home to more cessation of conflict as well as an aggressive pack- than a quarter of the world’s copper reserves, age of reforms.114 Growth has continued, reach- Chile diversified its exports and its economy ing an estimated 8.8 percent in 2011, and the while effectively managing resource-related poverty rate fell by 12 percentage points between risks such as currency appreciation and infla- 2005 and 2010. In the wake of the conflict, the tion. Unemployment fell to single digits from government supported the reintegration and de- around 20 percent in the early 1980s.118 A re- mobilization of more than 54,000 former com- source stabilization fund (since 1987) to­gether batants. In 2012, 73 percent of ex-combatants with a transparent fiscal rule (since 1999) al- Moving jobs center stage    31 lowed the country to save for difficult times and New Zealand in 2007 to provide employment avoid a loss of competitiveness. Governance re- opportunities through migration, leading to forms in all areas of public sector management higher remittances, improved knowledge of promoted accountability and transparency. An agricultural techniques, computer literacy, active export-oriented growth policy, including and English-language skills.125 Brazil provides the welcoming of foreign investment, supported an example of a rapidly formalizing country. productivity spillovers from jobs connected to Over the past decade, job creation in the for- global markets. Competitive innovation funds mal sector has been three times as rapid as in for nonmineral export sectors, especially in the informal sector. Just in the five years lead- agribusiness, have broadened the export base.119 ing up to the crisis, the formal share of total The public budget boosted education spending, employment increased by about 5 percentage which almost doubled between 1990 and 2009, points.126 Non-contributory social protection leading to an unprecedented expansion of sec- programs such as Bolsa Familia, a simplifica- ondary and tertiary education.120 tion of tax rules for small business, increased Slovenia has successfully tackled its very high incentives for firms to formalize their work- youth unemployment rate, reducing the ratio of ers, and improved enforcement of tax and youth to adult unemployment from three in labor regulations contributed to this success. the 1990s to around two today.121 The success Poland, an aging society, has seen its employ- in reducing youth unemployment cannot be ment ratio increase from 60 percent in 2006 to attributed to spending on active labor market ­ 65 percent in 2009. This was due to changes in programs (about average for transition coun- the application of eligibility rules of disabil- tries), liberalizing the labor market (rules re- ity pensions, and pension reforms adjusting main more restrictive than the average in de- the level of benefits down as life expectancy veloped countries), or low minimum wages increases. In 2012, a new wave of pension re- (still on the high side).122 Potential distortions forms raised the retirement age to 67 for men from these policies seem to be somewhat off- and women from the current 65 for men and set, however, by a model of consensus-based 60 for women.127 decision making whereby trade unions and employer organizations, with broad coverage, Connected jobs agendas: Global set wages that respond well to macroeconomic partnerships for jobs trends and sectoral productivity.123 Sustained growth before the global crisis is ultimately Policies for jobs in one country can have spill- responsible for much of Slovenia’s decline in overs to other countries, both positive and nega- youth unemployment. Taking advantage of tive. An important issue is whether international European integration, the economy success- coordination mechanisms could influence gov- fully restructured its export sector. Very good ernment decisions to enhance the positive spill- infrastructure and a fairly well-skilled work- overs and mitigate the negative. Several a ­reas force helped as well. lend themselves to more and better coordination. Examples of successful policies can actu- Rights and standards. Cross-border mecha- ally be found across the entire typology of jobs nisms exist to set standards and provide chan- challenges (figure 20). As an urbanizing coun- nels for improving compliance with rights. ILO try, Korea carefully designed and phased poli- conventions can influence domestic legislation cies to accompany the transition of jobs from and be a channel for voice and coordination in- agriculture to light manufacturing and then ternationally, as demonstrated by the process of to industries with higher value added.124 Land adopting the conventions for home-based and development programs were established first, ­ domestic workers. The support for core labor followed by a land-use regulation system, and standards in the 1998 Declaration on Funda- then by comprehensive urban planning. Hous- mental Principles and Rights at Work suggests ing and transportation policies held the dis- that countries respond to pressure from the in- economies of urbanization in check. Tonga, a ternational community.128 Yet the pressure only small island nation, is actively using the Recog- goes so far. The persistence of forced labor, chil- nized Seasonal Employer program launched by dren working in hazardous conditions, discrimi- 32   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 FI G U R E 20  Which countries succeeded at addressing their jobs challenges and how? Jobs challenge Which countries and which policies? Agrarian Vietnam economies Land reform, agricultural extension, and market incentives Conflict-affected Rwanda countries Reintegration of former combatants and business reforms Urbanizing Korea, Rep. countries Land-use policies and comprehensive urban planning Resource-rich Chile countries Fiscal stability rules and export-oriented policies Small island Tonga nations Active use of return migration agreements Countries with high Slovenia youth unemployment Competition in product markets through trade integration Formalizing Brazil countries Noncontributory programs, rules simplified and enforced Aging Poland societies higher retirement age Disability and pension reform, ­ Source: World Development Report 2013 team. nation, and lack of voice suggest that ratification nied by two projects, one to build capacity for on its own is not sufficient. monitoring working conditions in garment fac- Trade agreements are a potential instrument tories and one to support an arbitration council for international coordination on rights. They to resolve collective labor disputes.129 can incorporate incentives for attention to voice Beyond the initiatives of governments and working conditions by linking trade access through conventions and trade agreements, a to the adoption and enforcement of labor laws growing emphasis is being placed on private and standards. Whether linking rights to trade sector accountability and the broader corporate actually leads to better outcomes for workers on social responsibility (CSR) agenda, whereby either side of a trade agreement is less clear. La- companies voluntarily bring social and envi- bor clauses can be used as a protectionist tool, ronmental concerns into their operations.130 undermining trade and employment opportu- Codes of conduct are most likely to be adopted nities in developing countries. Moreover, in the by companies based in the European Union and absence of capacity and institutions to moni- North America, which then make engagement tor and enforce compliance, trade agreements with labor standards a condition of business on their own can be weak instruments. For with suppliers. But there is limited evidence on instance, Cambodia’s successful bilateral trade the extent to which codes of conduct translate agreement with the United States was accompa- into improved enforcement of standards. Be- Moving jobs center stage    33 cause of the complexity of global supply chains, can be more mixed. They are clearly positive seasonal and temporary workers are outside the when cell phones connect people (and especially reach of CSR frameworks. Workers outside of the poor) to product markets, to employment global supply chains are not covered.131 To be ef- ­ opportunities, or to government services. They fective, CSR efforts should focus more on build- can be negative when the disappearance of retail ing the capacity of local firms to comply and the trade leads to the decline of urban centers and capacity of labor inspectorates to do their work. affects the livelihoods of older shopkeepers who Trade and investment. International trade may not find alternative employment easily. An in goods has been gradually liberalized over adequate sequencing of services liberalization time, and the notion that freer trade is mutu- and domestic regulatory reform is needed to ally beneficial for the transacting parties is manage these tradeoffs, and, in doing so, ad- now widely shared. However, many developing dress the concerns of developing countries.136 countries still lack the competitiveness to har- International collaboration can fill the knowl- ness the benefits from global integration. Direct edge gaps and facilitate implementation.137 assistance to reduce logistic costs and improve International agreements can also promote the competitiveness of firms and farms is thus global public goods. One case in point is gen- a priority. Aid for trade has increased substan- der equality. Trade is not gender neutral, imply- tially and now accounts for about a third of ing that liberalization changes women’s access total aid to developing countries. But there is to jobs. Traditionally men were more likely to scope for making the assistance more effective, have “brawn jobs,� involving stronger physi- by focusing on the export activities most suited cal requirements, while “brain jobs� involving to address the specific jobs challenges that re- dexterity, attention, or communication—from cipient countries face. Increasing the involve- stitching garments to processing data—present ment by the private sector would also enhance more opportunities for women. Call centers in the effectiveness of the assistance.132 Delhi and Mumbai employ more than 1 million In contrast to trade in goods, progress in people, most of them women. Preferential ac- services liberalization has been slow, at both cess for imports from sectors with more “brain the multilateral and the regional levels. Offers jobs� can thus create employment opportunities to the Doha Round currently being negotiated for women in countries where gender equality promise greater security in access to markets is far from attained.138 But as countries move but no additional liberalization compared to up the ladder of global value chains, gender the policies in force.133 Services are character- opportunities can change. This was the case in ized by well-known market imperfections— Malaysia, where the share of women working in from network externalities in infrastructure to manufacturing declined in the mid-1980s.139 asymmetric information and moral hazard in Migration. In contrast to the movement of finance—and are thus subject to more perva- goods and services across borders, few interna- sive regulations. Liberalizing trade in services tional agreements pertain to migration in gen- requires adequate domestic regulation. Setting eral and the migration of workers in particular. up markets for electricity, or cushioning the so- Those in existence have limited coverage. ILO cial impacts of large distributors on retail trade, conventions 97 and 143, in force since 1952 and is challenging.134 Not surprisingly, liberalization 1978, refer to the prevention of discrimination of services is much less advanced in developing or abusive conditions against migrants and call countries than in industrial countries.135 for penalties and sanctions against those who The productivity gains from liberalizing promote clandestine or illegal migration. But services would be substantial. Many services they have been ratified by only 49 and 23 coun- are inputs into the production process. Elec- tries, respectively. Liberalizing the provision of tricity, finance, telecommunications, and trade services by natural persons, in line with Mode 4 have a direct impact on business costs, affecting of the General Agreement on Trade in Services the competitiveness of downstream sectors. By (GATS), is not on the agenda of many countries, boosting job creation and raising labor earn- industrial or developing. And the United Na- ings, these productivity gains should also lead tions international convention on the rights of to improved living standards. Social impacts migrant workers and their families, which en- 34   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 tered into force in 2003, has been ratified by only tionship between jobs and movements in and 22 countries—mostly sending countries. out of poverty; the dynamics of micro- and While migration occurs across borders, leg- small enterprises in the informal sector; and the islation is mostly driven by country-specific links between jobs and human behaviors and laws. This is an area where a global perspec- norms. Research on the magnitude of spillovers tive is warranted, but views on what needs to from jobs could identify good jobs for devel- be done are diverse. One view focuses on the opment tailored to country contexts. Another large earnings differentials between countries, important research area concerns the impact of suggesting that global productivity and pov- jobs on the acquisition of cognitive and non- erty reduction would accelerate enormously cognitive skills, and how this impact varies de- with the free movement of labor.140 Another pending on the characteristics of the job and the perspective focuses instead on national secu- person who holds it. Similarly, more evidence rity and the protection of communities and on productivity spillovers from jobs across cit- their cultures, implying the need for barriers ies with different characteristics would have a to contain migration. Yet another highlights high value for development policy. Estimates of the moral ­ imperative of protecting the human the environmental impacts of different types of rights of migrants, no matter their legal status, jobs are, today, scarce at best. In the area of la- and giving shelter to those who suffer any form bor policies, more empirical work is needed on of persecution.141 None of these views suffices, the boundaries of the plateau, depending on the however, because any one of them alone cannot characteristics of the country. More research is address the complex tradeoffs that migration also needed on how international trade, invest- poses for policy design. ment across borders, and migration affect the In many instances, both sending and re- composition of employment across countries. ceiving countries can benefit from migration More solid knowledge on the sequencing of in- through a collaborative approach. Most abuses ternational commitments and domestic policies perpetrated by traffickers, firms, or workers related to services could address the reluctance are associated with illegal migrant flows, so the of developing countries to make further prog- formalization of these flows is a basic tool for ress in the direction of liberalization and reap protecting the rights of migrant workers. This the gains from global integration. formalization cannot be enforced without the Setting policy priorities for jobs needs to be cooperation of institutions in both sending and based on reliable data. Given that a large share of receiving countries. That is why bilateral agree- the people at work in developing countries are ments are also needed, with provisions for quo- not wage employees, and that even fewer have a tas by occupation, industry, region, and dura- formal sector job, the measurement of employ- tion of stay.142 The agreements can distinguish ment is challenging. Determining which jobs between temporary movements of workers have the greatest payoffs for poverty reduction and steps to permanent migration, with condi- requires linking information on a household’s tions and protocols to go from one to another. income or consumption with information on They can include considerations about taxa- the employment of its members. Understand- tion, s­ocial security, and even financing higher ing which economic units create more jobs, or education—a special concern in the case of whether labor reallocation leads to substantial “talent� migration. These agreements can design growth rather than just churning, requires in- incentives so that stakeholders in both sending formation on the inputs and outputs of very and receiving countries have an interest in en- diverse production units. Assessing how the forcing the provisions.143 composition of employment affects trust and willingness to participate in society requires in- formation on individual values and behaviors. Jobs are center stage, but where are The paucity of empirical analyses on the em- the numbers? ployment impact of the global crisis in devel- oping countries and the difficulty of compar- A significant research and data agenda lies ing measures of informal employment across ahead. Further analysis is needed on the rela- countries suggest that data quality and avail- Moving jobs center stage    35 ability remain a constraint for policy making. standardized and include informal firms and Much e ­ffort goes into measuring unemploy- microenterprises in establishment surveys. Such ment rates, and measuring them often.144 But an approach would move jobs center stage. open unemployment is not a very telling in- dicator in countries where a large fraction of the labor force is not salaried. The Millennium * * * Development Goal on eradicating poverty lists four indicators to monitor progress toward the Countries have a choice in responding to the jobs employment target, defined as “achieving full challenges of demographics, structural shifts, and productive employment and decent work technological progress, and periodic macroeco- for all, including women and young people.� nomic crises. They can simply pursue growth, But these indicators only partially capture ad- ensure that the labor market functions well, and vances in the quantity and quality of jobs in the hope that jobs will follow. Or they can recog- developing world.145 nize that growth does not mechanically deliver Today’s challenges for labor statistics can be the jobs that do most for development. Jobs for regrouped into three key areas: data gaps; data women, jobs in cities and in global value chains, quality issues; and planning, coordination, and and jobs providing voice and protection for the communication issues. Data gaps concern coun- most vulnerable in society may come high on tries where labor statistics do not exist at all or the list. The precise nature of the jobs challenge are collected only sporadically. Whenever such depends on a country’s geography, endowments, statistics exist, data quality is a concern through- institutions, and level of development. What is out the statistical production chain, from the common to all is the need to remedy the insti- use of appropriate definitions to questionnaire tutional failures and market imperfections that design, from sampling frame to interviewer prevent the private sector from creating more processes, and from data entry and coding to of those good jobs for development. Coun- verification and estimation procedures. Plan- tries can then return to the difficult questions ning, coordination, and communication issues posed at the outset. For each of them, there is arise when different institutions are responsible a conventional wisdom that practitioners do for collecting and disseminating the data.146 not consider to be totally satisfactory. The jobs A quarter of a century ago, a renewed em- lens in development does not lead to a flat rejec- phasis on poverty reduction as the key objec- tion of the conventional wisdom, but to a quali- tive of development policy launched a long- fication of when it holds and when it does not. term data effort. Across the world, information In short, countries can leave themselves open on household living standards was collected to small gains in living standards, slow produc- through standardized surveys. The sampling tivity growth, and fractious societies. Or, by ad- methods and the variable definitions used were dressing their jobs challenges, they can enjoy duly documented. And the data and documen- a self-reinforcing pattern of more prosperous tation were made available to researchers and livelihoods, rising productivity, and the stron- practitioners whenever possible. For jobs, em- ger social cohesion that comes from improving ployment modules attached to household sur- ­ employment opportunities and fairness in ac- veys used for poverty analyses would need to be cess to jobs. QUESTIONS When is the conventional wisdom right? Growth strategies or jobs strategies? The conventional wisdom preneurial potential can thus make a substantial difference is to focus on growth as a precondition for continued in living standards and productivity. increases in living standards and strengthened social cohe- sion. But lags and gaps among the three transformations of Can policies contribute to social cohesion? The conventional living standards, productivity, and social cohesion are not wisdom is that a lack of jobs is detrimental to social cohesion, uncommon. The impact of growth on poverty reduction var- but other than ensuring full employment there is little that ies con­siderably across countries. And in some cases, growth governments can or should do. Yet open unemployment is is not accompanied by increased social cohesion—even not the main challenge in many countries, as the characteris- though poverty may fall and living standards improve for tics of jobs also matter. While not all jobs can positively affect some, the expectations of others remain unfulfilled. The social cohesion, those that shape social identity, build net- employment intensity of different sectors and fairness in works— particularly for excluded groups— and increase fair- access to employment opportunities matter as well. It is thus ness can help defuse tensions and support peaceful collective jobs that bring together the three transformations. decision making. Tradeoffs among improving living standards, accelerating Measures that support inclusion, extend access to voice productivity growth, and fostering social cohesion arguably and rights, and improve transparency and accountability in reflect a measurement problem, more than a real choice. If the labor market can increase the extent to which people per- growth indicators captured the intangible social benefits ceive that they have a stake in society. This perception can be from jobs, from lower poverty to greater social cohesion, a especially critical when risks of social unrest from youth growth strategy and a jobs strategy would be equivalent. But unemployment and conflict are high. Employment programs a growth strategy may not pay enough attention to female can undermine social cohesion if they have weak governance employment, or to employment in secondary cities, or to or divisive targeting, but can have positive effects when they idleness among youth. When potentially important spill- are well-designed. Jobs policies for youth at risk can incorpo- overs from jobs are not realized, a jobs strategy may provide rate counseling and training in conflict resolution. Public more useful insights. works programs can facilitate community participation and engagement between citizens and local authorities. Policies Can entrepreneurship be fostered? The conventional wisdom can thus focus not only on the number of jobs, but on expand- is that most micro- and small enterprises in developing ing job opportunities for excluded groups. countries are just forms of survivorship, with limited chances to grow. But self-employed workers account for a large share Skills or jobs—what comes first? The conventional wisdom is of employment in developing countries. Even if only a frac- that investing in skills will lead to job creation and to higher tion of them succeeded in building a viable business, the productivity and labor income. High unemployment and aggregate impact on living standards and productivity would skills mismatches are often attributed to shortcomings in be substantial. Moreover, in developing countries many large education and training systems. But in reality they can also enterprises are born large, often the result of government result from market distortions, which send the wrong signals support or privileged access to finance and information. to the education system or lead to a lack of dynamism in pri- Breaking privilege is one more reason why the success of vate firms. In such situations, massive investments in train- micro- and small enterprises is so important. ing systems, as seen in many parts of the world, might show Management practices are important in explaining firm disappointing results as hoped-for job outcomes do not productivity, even in small and medium-size firms. The materialize. capacity to acquire skills and to apply them to business is one A core set of basic skills, both cognitive and social, is of the most important characteristics of successful entrepre- necessary for productive employment, and they cannot just neurs. Yet markets fail to nurture entrepreneurship, because be acquired on the job. Without such generic skills, the knowledge spillovers imply that some of the returns to prospects of improving employment opportunities and acquiring or developing new managerial ideas and knowl- earnings are thin. Skills are also critical for countries to edge are appropriated by others. And the potential to absorb move up the value-added ladder, as they can ignite innova- management practices differs greatly among beneficiaries. tion, produce the benefits of mutual learning, and hence Observable characteristics of small business owners can pre- lead to job creation themselves. But in between, much dict entrepreneurial potential, and programs to upgrade learning can happen through work: job opportunities can their managerial capacity have been shown to make a differ- shape social skills and create demand for education and ence. Programs targeted to small business owners with entre- training. Learning on the job leads to significantly higher Moving jobs center stage    37 earnings in many settings, with the return to one year of Protecting workers or protecting jobs? The conventional wis- work experience being around a third to half of the return dom is that policies that protect people are preferable, because to an additional year of schooling. they mitigate welfare losses while at the same time allowing the reallocation of labor, hence supporting creative destruc- A targeted investment climate? The conventional wisdom is tion. Protecting jobs that are no longer economically viable that a level playing field is preferable because governments through government transfers and employment protection do not have enough information to pick winners and tar- legislation freezes an inefficient allocation of resources. Pro- geting can be c ­ aptured by interest groups. But given the tecting jobs also entails a high risk of capture. It may lead to often limited fiscal space and administrative capacity of enduringly unproductive jobs, stifle technological advance, developing countries, creating an enabling business envi- prevent structural change, and eventually undermine growth. ronment across the board can be challenging, and the rele- However, there are times when many jobs are lost or vant question is how policy priorities should be set. The threatened at once and few are being created. There are also conventional wisdom views targeting with a skepticism jobs generating substantial productivity spillovers, whose that stems from failed experiences with industrial policy. disappearance in large numbers can lead to ghost towns and However, targeting may not necessarily be aimed at indus- depressed regions. Protecting people should have primacy if trial sectors. Supporting job creation in sectors with high shocks are idiosyncratic—if the employment dislocation is rates of female employment, or productivity gains in small- local and limited and if turnover continues to be the norm. holder farming, or more jobs connected to global value Protecting jobs may be warranted in times of systemic crises chains may have high development payoffs depending on or major economic restructuring. But job protection policies country contexts. can create permanent inefficiency, especially in countries When there is clarity about where the good jobs for devel- with weak institutions, making it indispensable to establish opment are, and there is sufficient information to under- and enforce trigger rules and sunset clauses that define the stand what can be done to support the creation of those jobs, extent and size of the protection. a targeted investment climate may be warranted. But this is provided that targeted interventions can be designed in a way How to accelerate the reallocation of workers? The conven- that makes them resilient to ­ capture by interest groups. The tional wisdom is to focus policy on removing the labor risk of capture is easier to contain when the number of market rigidities that keep workers in firms or areas with ­ beneficiaries is very large, such as the case of farmers, urban low productivity. But reforms may not always be politically businesses, and female micro-entrepreneurs. It is much feasible. In India, complex and cumbersome labor market higher in the case of industrial policy. institutions have unambiguously negative effects on eco- nomic efficiency but these institutions have remained largely Competing for jobs? The conventional wisdom is that the num- untouched for 60 years. ber of jobs is not finite, so that policies for jobs in one country Tolerance for the avoidance or evasion of distortive regu- cannot be harmful to other countries. Indeed, in the medium lation can help contain their cost but not ensure dynamism. to long term, total employment is roughly determined by the In India, widespread noncompliance has been the dominant size of the labor force. But policies may alter global trade, response to cumbersome labor regulations. However, labor- investment, and migration flows, affecting the composition of intensive manufacturing sectors remain sluggish despite employment. The concern is that the share of good jobs for buoyant performance of the overall economy. Other coun- development may decline in one country as it increases in tries with similarly stringent regulatory obstacles have another. Policies aimed at capturing a larger global share of accomplished more efficiency-enhancing labor reallocation the jobs with the largest productive spillovers can reduce well- by actively taking advantage of productivity spillovers from being abroad, even if global well-being increases. jobs in industrial clusters, dynamic cities, or global value However, not all efforts to support job creation amount to chains to make the regulations less binding. In Sri Lanka, beggar-thy-neighbor policies. Whether they do so depends on the development of export processing zones (EPZs) drove the type of instruments used and the nature of the spillovers the takeoff of the garment industry. In Brazil, the surge of from jobs. A key question is what purpose policies serve. Poli- internal migration is closely associated with the country’s cies that aim to improve compliance with rights, prosecuting continuing integration into the global economy and a devel- forced labor and harmful forms of child labor, amount to opment policy that favors clusters and agglomeration. In providing a global public good. On the other hand, policies China, labor reallocation is rooted in the development of that aim at reaping the benefits from productive externalities competitive cities, supported by regional competition and may adversely affect other countries, especially when they experimentation. A strategic focus on enhancing productiv- undermine an open trading system and are not aligned with a ity spillovers from jobs, through urbanization and global country’s dynamic comparative advantage. integration, can overcome the labor rigidity. 38   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Jobs drive They should not be an development afterthought of growth It is not just the Some have greater number of jobs development payoffs Jobs are created by Public action the private sector sets the stage Many jobs in developing countries are Informal in farms and very small firms is normal Some work is Rights should not unacceptable be overlooked One size does Jobs challenges vary not fit all across countries Get the policy They have a bearing regardless fundamentals right of the jobs challenge The main constraints to Labor policies matter job creation may lie elsewhere less than assumed Set priorities for Focus on the jobs with public action greater development payoffs Data and cooperation on cross-border A global agenda investments and migration lag for jobs is needed Moving jobs center stage    39 Notes 23. Nabli, Silva-Jáuregui, and Faruk Aysan 2008. 24. Assaad 2012; Assaad and Barsoum 2007.    1. Article 2, ILO 2007. Also see UN 2009. 25. Mryyan 2012; Gatti and others 2012; Stampini    2. Ghose, Majid, and Ernst 2008. and Verdier-Choucane 2011; ILO 2011.   3. Gindling and Newhouse 2012 for the World 26. Based on an update by the WDR team of Chen Development Report 2013. and Ravallion (2010).   4. Kanbur 2009. 27. World Bank 2011b.   5. International Labour Organization, Depart- 28. ILO 2012a. ment of Statistics, http://laborsta.ilo.org/sti/ 29. ILO and World Bank 2012. sti_E.html. 30. Bell and Blanchflower 2011; Farber 2011.    6. Lyon, Rosati, and Guarcello 2012 for the World 31. World Bank 2011c. Development Report 2013. 32. Ravallion 2009.    7. World Bank 2006b. 33. Inchauste 2012 for the World Development Re-   8. World Development Report 2013 team esti- port 2013. mates based on data from the International 34. Baulch 2011; Fields and others 2003. Labour Organization, http://laborsta.ilo.org/ ­ 35. Narayan, Pritchett, and Kapoor 2009. ­applv8/data/EAPEP/eapep_E.html, and World 36. Azevedo and others 2012 for the World Devel- Development Indicators, http:/data-worldbank opment Report 2013. This report uses methods .org/data/catalog/world-development- developed by authors, including Paes de Barros indicators. and others (2006) and Bourguignon and Fer-    9. United Nations 2011. reira (2005). 10. Lin 2012; Pagés 2010; World Bank 1992. 37. Blanchflower and Oswald 2011. 11. European Centre for the Development of Voca- 38. Haltiwanger 2011; Nelson 1981; Schumpeter tional Training 2008. 1934. 12. Autor and Dorn 2011; Gratton 2011; Holzer 39. Bartelsman, Haltiwanger, and Scarpetta 2009; and Lerman 2009. Davis, Haltiwanger, and Schuh 1996. 13. Feenstra 2010. 40. Bartelsman, Haltiwanger, and Scarpetta 2009. 14. Brown, Ashton, and Lauder 2010. See Selim 41. Baily, Bartelsman, and Haltiwanger 1996. 2012 for the World Development Report 2013. 42. World Development Report 2013 team 15. Goswami, Mattoo, and Sáez 2011. estimates. 16. Examples include oDesk, https://www.odesk 43. Bartelsman, Haltiwanger, and Scarpetta 2004; .com/; Babajob, http://www.babajob.com/; Brandt, Van Biesebroeck, and Zhang 2012; Lin Google Trader (for example, http://www 2012; Rutkowski and others 2005. .google.co.ug/africa/trader/search?cat=jobs); 44. World Development Report 2013 team esti- and SoukTel, http://www.souktel.org/. mates, and Dutz and others 2011. 17. TeamLease 2010. 45. Ayyagari, Demirgüç-Kunt, and Maksimovic 18. A.T. Kearney 2011. 2011; IFC, forthcoming. 19. UNESCO Institute of Statistics, http://stats 46. South Africa is excluded from this estimate be- .uis.unesco.org/unesco/TableViewer/tableView cause it is an outlier, with average farm size of .aspx?ReportId=175. 288 hectares. 20. WDR team estimates of the top 20 percent 47. Hsieh and Klenow 2009; Pagés 2010. of ratings among countries in the 2009 Pro- 48. Banerjee and Duflo 2011; Fox and Sohnesen gramme for International Student Assessment 2012; Schoar 2010; Sutton and Kellow 2010. of 15 year-olds. See http://www.pisa.oecd.org. 49. de Soto 1989; Perry and others 2007. 21. International Labour Office database on labor 50. Grimm, Kruger, and Lay 2011; McKenzie and statistics, Laborsta, http://laborsta.ilo.org/e. Woodruff 2008. 22. Private sector employment refers here to “pri- 51. Mertens 2011; Witze 2010. vate firms� and to “individuals� according to the 52. Sandefur 2010. official Chinese classsification. The former are 53. Hsieh and Klenow 2011. defined as for-profit units invested in and estab- 54. Bartelsman, Haltiwanger, and Scarpetta 2009; lished by natural persons or controlled by per- Haltiwanger 2011; Hsieh and Klenow 2009; sons using more than seven workers. The latter Syverson 2011. includes units that hire fewer than eight workers. 55. Moser 2009, 240. Foreign-invested firms and collectives are not 56. Dani and others 1999, 3. part of the private sector in official statistics. For 57. Kilroy 2011. more details, see Kanamori and Zhao (2004). 58. Gatti and others 2012. 40   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 59. Akerlof and Kranton 2010. 98. ILO 1998. 60. Giles, Mavridis, and Witoelar 2012 for the 99. Chen and others 2012 for the World Develop- World Development Report 2013. ment Report 2013. 61. ILO 2012b. 100. Betcherman 2012. 62. ILO 2010. 101. Alatas and Cameron 2003; Arango and Pachón 63. ILO 2002. 2004; Rama 2001; SMERU Research Institute 64. Heath and Mobarak 2011. 2001. 65. Luke and Munshi 2011. 102. Haltiwanger, Scarpetta, and Schweiger 2008. 66. Alfaro and Chen 2011; Romer 1993. 103. Betcherman 2012 for the World Development 67. UNDP 2003a; UNDP 2003b. Report 2013; Freeman 2009; OECD 2006. 68. Ibarraran and others 2012. 104. Aidt and Tzannatos 2002. 69. As an example, recent World Development Re- 105. Freeman 2009. 106. Aidt and Tzannatos 2002. ports on youth (World Bank 2006b), geography 107. Card, Kluve, and Weber 2010; OECD 2006; (World Bank 2009b), conflict (World Bank Almeida and others 2012 for the World Devel- 2011a), and gender (World Bank 2011c) look at opment Report 2013. spillovers from jobs in different contexts. 108. Bird and Smart 2012; Levy 2008. 70. Glewwe 2004. 109. Bhatt 2006; Chen and others 2012 for the World 71. IOM 2010. Development Report 2013. 72. Goswami, Mattoo, and Sáez 2011. 110. World Bank 2010. Note that this unit cost is the 73. Ball, Leigh, and Loungani 2012. aggregate cost of the Multi-Country Demobi- 74. World Bank 2012, various issues. lization and Reintegration Program, including 75. Commission on Growth and Development all forms of reintegration support, not only 2008. employment. 76. Kraay 2012. 111. Glewwe 2004. 77. Elbadawi, Kaltani, and Soto 2009. 112. Rama 2009. 78. King and Levine 1993; Levine 2005. 113. World Development Indicators 2012. World 79. IFC, forthcoming. Development Indicators, World Bank, Wash- 80. Foster and Briceño-Garmendia 2010. ington, DC. http://data-worldbank.org/data- 81. Djankov, Freund, and Pham 2010; Hallward- catalog/world-developmentindicators. Driemeier, Khun-Jush, and Pritchett 2010. 114. World Bank 2007. 82. Klapper, Laeven, and Rajan 2006. 115. Rwanda Demobilization and Reintegration 83. Bruhn, 2008. Commission 2012. 84. See, for example, Psacharopoulos and Patrinos 116. Rwanda was named a top reformer by Doing (2004); Montenegro and Patrinos 2012 for the Business in 2010. World Development Report 2013. 117. Dudwick and Srinivasan, forthcoming; World 85. See, for example, Hanushek and Woessmann Bank 2011a. (2008) and Commander and Svejnar (2011) 118. World Development Indicators 2012. World on the productivity link. Structural change and Development Indicators, World Bank, Wash- poverty links are explored in Lee and Newhouse ington, DC. http://data-worldbank.org/data- catalog/world-developmentindicators. (2012) for the World Development Report 2013. 119. Consejo Nacional de Innovación 2008; World 86. Engle and others 2007; Grantham-McGregor Bank 2008. and others 2007; Heckman 2008; Walker and 120. World Bank 2006a; World Development Indi- others 2007; Young and Richardson 2007. cators 2011. World Development Indicators, 87. Engle and others 2007. World Bank, Washington, DC. http://data-world 88. Heineck and Anger 2010; Cunha, Heckman and bank.org/data-catalog/world-development Schennach 2010. indicators. 89. OECD PISA 2009, http://www.pisa.oecd.org. 121. OECD 2010. 90. IMF 2003; Rodrik 2000. 122. OECD 2009. 91. Keefer 2009; North 1981, 1990. 123. OECD 2009. 92. Acemoglu, Johnson, and Robinson 2001; North 124. Yusuf and Nabeshima 2006; Park and others 1990; Rodrik, Subramanian, and Trebbi 2004. 2011. 93. World Bank 2004. 125. See World Bank 2010; Gibson, McKenzie, and 94. World Bank 2010. Rohorua 2008. 95. World Bank 2004. 126. Fajnzylber, Maloney, and Montes-Rohas 2011; 96. World Bank 2004. OECD and ILO 2011. 97. Laeven and Woodruff 2007. 127. World Bank 2011d. Moving jobs center stage    41 128. Chau and Kanbur (2002) find evidence of a vices Location Index, 2011. Chicago: A.T. Kearney peer effect whereby ratification depends on the Global Services Location Index. number of similar countries that have already Acemoglu, Daron, Simon Johnson, and James A. Rob- ratified the convention. inson. 2001. “The Colonial Origins of Compara- 129. Adler and Hwang 2012 for the World Develop- tive Development: An Empirical Investigation.� ment Report 2013. American Economic Review 91 (5): 1369–401. 130. Levi and others 2012 for the World Develop- Adler, Daniel, and Hans Hwang. 2012. “From Law on ment Report 2013; Newitt 2012 for the World the Books to Law in Action: A Note on the Role Development Report 2013. of Regulation in the Production of Good Jobs in 131. Locke, forthcoming; Locke, Quin, and Brause Cambodia’s Garment Sector.� Background paper 2007. for the WDR 2013. 132. Hoekman 2011. Aidt, Toke, and Zafiris Tzannatos. 2002. Unions and 133. Borchert, Gootiiz and Mattoo 2011. Collective Bargaining: Economic Effects in a Global 134. François and Hoekman 2010. Environment. Washington, DC: World Bank. 135. Hoekman and Mattoo 2011. Akerlof, George A., and Rachel E. Kranton. 2010. 136. Fink, Mattoo, and Rathindran 2003; François Identity Economics: How Our Identities Shape Our and Hoekman 2010. Work, Wages, and Well-Being. Princeton, NJ: 137. Hoekman and Mattoo 2011. Princeton University Press. 138. World Bank 2011c. Alatas, Vivi, and Lisa Ann Cameron. 2003. “The Im- 139. Randriamaro 2007. pact of Minimum Wages on Employment in a 140. See, for instance, Winters and others (2002); Low Income Country: An Evaluation Using the World Bank (2005). Difference-in-Differences Approach.� Policy Re- 141. See, for instance, EFRA (2011) and Angenendt search Working Paper Series 2985, World Bank, (2012). Washington, DC. 142. Regional agreements, such as the European Alfaro, Laura, and Maggie Xiaoyang Chen. 2011. “Se- Union’s Schengen area, can also cover specific lection, Reallocation, and Knowledge Spillovers: areas such as visas or social security of migrant Identifying the Impact of Multinational Activity workers. Several Latin American countries, on Aggregate Productivity.� Paper presented at the Spain, and Portugal have been developing com- World Bank Conference on Structural Transfor- mon principles about social security rights and mation and Economic Growth, Washington, DC, regulations for migrants. October 6. 143. For a discussion on these issues, see Pritchett (2006). Almeida, Rita, David Margolis, David Robalino, and 144. Sixty-five countries produce monthly or quar- Michael Weber. 2012. “Facilitating Labor Market terly labor force surveys, whereas 116 produce Transitions and Managing Risks.� Background annual surveys. paper for the WDR 2013. 145 . The four indicators are GDP per employed Amador, Diego, Raquel Bernal and Ximena Peña person (a measure of productivity), the 2011. “The Rise in Female Participation in Co- employment-to-population rate, the propor- lombia: Fertility, Marital Status or Education?� tion of the employed population living on less Background paper for the World Development than US$1.25 a day (the so-called working Report 2012. poor), and the proportion of own-account and Angenendt, Steffen. 2012. “Migration and Social In- unremunerated workers in employed popu- clusion–Looking through the Good Jobs Lens.� In lation (also called vulnerable workers). See Moving Jobs to the Center Stage, BMZ (Bundes- United Nations Development Group 2010. ministerium fuer Wirtschaftliche Zussamenar- 146. ILO 2012c. See Kanbur and Svejnar (2009) on beit), Berlin Workshop Series. Berlin. the importance of data for labor analysis and Arango, Carlos, and Angelica Pachón. 2004. “Mini- policy mum Wages in Colombia: Holding the Middle with a Bite on the Poor.� Borradores de Economía Serie 280, Banco de la República de Colombia, References Bogotá. Artuc, Erhan, Frederic Docquier, Caglar Özden, and The word processed describes informally reproduced Chris Parsons. 2012. “Education Structure of Global works that may not be commonly available through Migration Patterns: Estimates Based on Census libraries. Data.� World Bank, Washington DC. Processed. Assaad, Ragui. 2012. “The MENA Paradox: Higher A.T. Kearney. 2011. Offshoring Opportunities amid Education but Lower Job Quality.� In Moving Jobs Economic Turbulence: A.T. Kearney Global Ser- to the Center Stage. BMZ (Bundesministerium 42   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 fuer Wirtschaftliche Zussamenarbeit), Berlin Bhatt, Ela. 2006. We Are Poor But So Many: The Workshop Series. Berlin. Story of Self-Employed Women in India. New York: ———. 1997. “The Effects of Public Sector Hir- Oxford University Press. ing and Compensation Policies on the Egyptian Bird, Richard M., and Michael Smart. 2012. “Financ- Labor Market.� World Bank Economic Review 11 ing Social Expenditures in Developing Countries: (1): 85–118. Payroll or Value Added Taxes?� International Cen- Assaad, Ragui, and Ghada Barsoum. 2007. “Youth Ex- ter for Public Policy Working Paper Series 1206, clusion in Egypt: In Search of ‘Second Chances.’ � Andrew Young School of Policy Studies, Georgia Middle East Youth Initiative Working Paper Series State University, Atlanta. 2, Wolfensohn Center for Development, Dubai Bjørkhaug, Ingunn, Anne Hatløy, Tewodros Kebede, School of Government, Dubai. and Huafeng Zhang. 2012. “Perception of Good Autor, David H., and David Dorn. 2011. “The Growth Jobs: Colombia.� Background paper for the WDR of Low-Skill Service Jobs and the Polarization of 2013. the U.S. Labor Market.� Massachusetts Institute of Blanchflower, David G., and Andrew J. Oswald. 2011. Technology, Cambridge, MA. Processed. “International Happiness.� Working Paper Series Ayyagari, Meghana, Asli Demirgüç-Kunt, and Vojislav 16668. National Bureau of Economic Research, Maksimovic. 2011. “Firm Innovation in Emerging Cambridge, MA. Markets: The Roles of Governance and Finance.� Borchert, Ingo, Batshur Gootiiz, and Aaditya Mattoo. Journal of Financial and Quantitative Analysis 46 2011. “Services in Doha: What’s on the Table?� In (6): 1545–80. Unfinished Business: The WTO’s Doha Agenda, ed. Azevedo, João Pedro, Gabriela Inchauste, Sergio Will Martin and Aaditya Mattoo, 115–44. Lon- Olivieri, Jaime Saavedra Chanduvi, and Hernan don: London Publishing Partnership. Winkler. 2012. “Is Labor Income Responsible for Bourguignon, François, and Francisco H. G. Ferreira. Poverty Reduction? A Decomposition Approach.� 2005. “Decomposing Changes in the Distribution Background paper for the WDR 2013. of Household Incomes: Methodological Aspects.� Baily, Martin Neil, Eric J. Bartelsman, and John Halti- In The Microeconomics of Income Distribution Dy- namics in East Asia and Latin America, ed. Fran- wanger. 1996. “Downsizing and Productivity çois Bourguignon, Francisco H. G. Ferreira, and Growth: Myth or Reality?� Small Business Eco- Nora Lustig, 17–46. Washington, DC: World Bank. nomics 8 (4): 259–78. Brandt, Loren, Johannes Van Biesebroeck, and Yifan Ball, Laurence, Daniel Leigh, and Prakash Loungani. Zhang. 2012. “Creative Accounting or Creative Forthcoming. “Okun’s Law: Fit at 50?� Working Destruction? Firm-Level Productivity Growth in Paper, International Monetary Fund, Washington, Chinese Manufacturing.� Journal of Development DC. Economics 97 (2): 339–51. Banerjee, Abhijit V., and Esther Duflo. 2011. Poor Eco- Brown, Philip, David Ashton, and Hugh Lauder. nomics: A Radical Rethinking of the Way to Fight 2010. Skills Are Not Enough: The Globalization Global Poverty. New York: Public Affairs. of Knowledge and the Future of the UK Economy. Bartelsman, Eric, John Haltiwanger, and Stefano Wath upon Dearne, U.K.: U.K. Commission for ­ Scarpetta. 2004. “Microeconomic Evidence of Employment and Skills. Creative Destruction in Industrial and Develop- Bruhn, Miriam. 2008. “License to Sell: The Effect of ing Countries.� Discussion Paper Series 1374, Business Registration Reform on Entrepreneur- Institute for the Study of Labor, Bonn. ial Activity in Mexico.� Policy Research Working ———. 2009. “Measuring and Analyzing Cross- Paper Series 4538, World Bank, Washington, DC. Country Differences in Firm Dynamics.� In Pro- Card, David, Jochen Kluve, and Andrea Weber. ducer Dynamics: New Evidence from Micro Data, 2010. “Active Labour Market Policy Evaluations: ed. Timothy Dunne, J. Bradford Jensen, and Mark A Meta-Analysis.� Economic Journal 120 (11): J. Roberts, 17–76. Cambridge, MA: National 452–77. Bureau of Economic Research. Chau, Nancy H., and Ravi Kanbur. 2001. “The Adop- Baulch, Bob, ed. 2011. Why Poverty Persists: Poverty tion of International Labor Standards Conven- Dynamics in Asia and Africa. Cheltenham, U.K.: tions: Who, When and Why?� In Brookings Trade Edward Elgar. Forum: 2001, ed. Nancy H. Chau, Ravi Kanbur, Bell, David N. F., and David G. Blanchflower. 2011. Ann E. Harrison, and Peter Morici, 113–56. Wash- “The Crisis, Policy Reactions and Attitudes to ington, DC: Brookings Institution. Globalization and Jobs.� Discussion Paper Series Chen, Martha, Chris Bonner, Mahendra Chetty, Lucia 5680, Institute for the Study of Labor, Bonn. Fernandez, Karin Pape, Federico Parra, Arbind Betcherman, Gordon. 2012. “Labor Market Institu- Singh, and Caroline Skinner. 2012. “Urban Infor- tions: A Review of the Literature.� Background mal Workers: Representative Voice and Economic paper for the WDR 2013. Rights.� Background paper for the WDR 2013. Moving jobs center stage    43 Chen, Shaohua, and Martin Ravallion. 2010. “The De- Children in the Developing World.� Lancet 369 veloping World Is Poorer Than We Thought, but (9557): 229–42 No Less Successful in the Fight against Poverty.� European Centre for the Development of Vocational Quarterly Journal of Economics 125 (4): 1577–625. Training. 2008. Future Skill Needs in Europe, Chioda, Laura. 2012. Work and Family: Latin America Medium-Term Forecast, Synthesis Report. Brussels: and Caribbean Women in Search of a New Balance. European Centre for the Development of Voca- Washington, DC: World Bank. tional Training. Commander, Simon, and Jan Svejnar. 2011. “Business Fajnzylber, Pablo, William F. Maloney, and Gabriel V. Environment, Exports Ownership, and Firm Per- Montes-Rojas. 2011. “Does Formality Improve formance.� Review of Economics and Statistics 93 Micro-Firm Performance? Quasi-Experimental (1): 309–37. Evidence from the Brazilian SIMPLES Program.� Commission on Growth and Development. 2008. The Discussion Paper Series 4531, Institute for the Growth Commission Report: Strategies for Sustained Study of Labor, Bonn. Growth and Inclusive Development. Washington, Farber, Henry S. 2011. “Job Loss in the Great Reces- DC: Commission on Growth and Development. sion: Historical Perspective from the Displaced Consejo Nacional de Innovación. 2008. Hacia una Es- Workers Survey, 1984–2010.� Discussion Paper trategia Nacional de Innovación para la Competitiv- Series 5696, Institute for the Study of Labor, Bonn. idad. Santiago: Consejo Nacional de Innovación. Fares, Jean, and Olga Susana Puerto. 2009. “Towards Cunha, Flavio, James J. Heckman, and Susanne Comprehensive Training.� Social Protection Dis- Schennach. 2010. “Estimating the Technology of cussion Paper Series 0924, World Bank, Washing- ton, DC. Cognitive and Noncognitive Skill Formation.� Feenstra, Robert C. 2010. Offshoring in the Global Econometrica 78 (3): 883–931. Economy: Microeconomic Structure and Macroeco- Dani, Anis, Sarah Forster, Mirsada Muzur, Dino nomic Implications. Cambridge, MA: MIT Press. Djipa, Paula Lytle, and Patrizia Poggi. 1999. A So- Fields, Gary, Paul Cichello, Samuel Freije-Rodriguez, cial Assessment of Bosnia and Herzegovina. Wash- Marta Menendez, and David Newhouse. 2003. ington, DC: World Bank. “Household Income Dynamics: A Four-Country Davis, Steven J., John C. Haltiwanger, and Scott Story.� Journal of Development Studies 40 (2): 30–54. Schuh. 1996. Job Creation and Destruction. Cam- Fink, Carsten, Aaditya Mattoo, and Randeep Rathin- bridge, MA: MIT Press. dran. 2003. “An Assessment of Telecommunica- de Soto, Hernando. 1989. The Other Path: The Invisi- tions Reform in Developing Countries.� Informa- ble Revolution in the Third World. New York: tion Economics and Policy 15 (4): 443–66. Harper & Row. Foster, Vivien, and Cecilia Briceño-Garmendia, eds. Djankov, Simeon, Caroline Freund, and Cong S. 2010. Africa’s Infrastructure: A Time for Transfor- Pham. 2010. “Trading on Time.� Review of Eco- mation. Washington, DC: World Bank. nomics and Statistics 92 (1): 166–73. Fox, Louise, and Thomas Sohnesen. 2012. “House- Dudwick, Nora, and Radhika Srinivasan, with Jose hold Enterprise in Sub-Saharan Africa: Why They Cueva and Dorsati Madani. Forthcoming. Creat- Matter for Growth, Jobs, and Poverty Reduction.� ing Value Chains in Africa’s Fragile States: Are Value Policy Research Working Paper Series 6184, World Chains an Answer? Directions in Development Bank, Washington, DC. Series. Washington, DC: World Bank. François, Joseph F., and Bernard Hoekman. 2010. Dutz, Mark A., Ioannis Kessides, Stephen O’Connell, “Services Trade and Policy.� Journal of Economic and Robert D. Willig. 2011. “Competition and Literature 48 (3): 642–92. Innovation-Driven Inclusive Growth.� Policy Re- Freeman, Richard. 2009. “Labor Regulations, Unions, search Working Paper Series 5852, World Bank, and Social Protection in Developing Countries: Washington, DC. Market Distortions or Efficient Institutions?� In EFRA (European Union Agency for Fundamental Handbook of Development Economics, Volume 5, Rights). 2011. Fundamental Rights of Migrants in an ed. Dani Rodrik and Mark Rosenzweig, 4657–702. Irregular Situation in the European Union. Luxem- Amsterdam: Elsevier. bourg: Publications Office of the European Union. Gatti, Roberta, Diego Angel-Urdinola, Joana Silva, Elbadawi, Ibrahim, Linda Kaltani, and Raimundo and Andras Bodor. 2012. Striving for Better Jobs: Soto. 2009. Aid, Real Exchange Rate Misalignment The Challenge of Informality in the Middle East and Economic Performance in Sub-Saharan Africa. and North Africa. Washington, DC: World Bank. Santiago: Universidad Católica de Chile. Ghose, Ajit K., Nomaan Majid, and Christoph Ernst. Engle, Patrice L, Maureen M. Black, Jere R. Behrman, 2008. The Global Employment Challenge. Geneva: Meena Cabral de Mello, Paul J. Gertler, Lydia International Labour Organization. Kapiriri, Reynaldo Martorell, and Mary Eming Gibson, John, David McKenzie, and Halahingano Young. 2007. “Strategies to Avoid the Loss of De- ­ Rohorua. 2008. “How Pro-Poor is the Selection of velopmental Potential in More than 200 Million Seasonal Migrant Workers from Tonga Under 44   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 New Zealand’s Recognized Seasonal Employer Heath, Rachel, and Mushfiq Mobarak. 2011. “Sup- Program.� Working Paper Series 4698, World ply and Demand Side Constraints on Educational Bank, Washington, DC. Investment: Evidence from Garment Sector Jobs Giles, John, Dimitris Mavridis, and Firman Witoelar. and a Girls’ Schooling Subsidy Program in Bangla- 2012. “Subjective Well-Being, Social Cohesion, desh.� Yale University, New Haven, CT. Processed. and Labor Market Outcomes in Indonesia.� Back- Heckman, James J. 2008. “The Case for Investing in ground paper for the WDR 2013. Disadvantaged Young Children.� In Big Ideas for Gindling, T. H., and David Newhouse. 2012. “Self- Children: Investing in Our Nation’s Future, 49–58. Employment in the Developing World.� Back- Washington, DC: First Focus. ground paper for the WDR 2013. Heineck, Guido, and Silke Anger. 2010. “The Returns Glewwe, Paul W. 2004. “An Overview of Economic to Cognitive Abilities and Personality Traits in Germany.� Labour Economics 17 (3): 535–46. Growth and Household Welfare in Vietnam in the Hoekman, Bernard. 2011. “Aid for Trade: Why, What, 1990s.� In Economic Growth, Poverty and House- and Where Are We?� In Unfinished Business? The hold Welfare in Vietnam, ed. Paul Glewwe, Bina WTO’s Doha Agenda, ed. Will Martin and Aad- Agarwal, and David Dollar, 1–26. Washington, itya Mattoo, 233–54. London: London Publishing DC: World Bank. Partnership. Goswami, Arti Grover, Aaditya Mattoo, and Sebastián Hoekman, Bernard, and Aaditya Mattoo. 2011. Sáez, eds. 2011. Exporting Services: A Develop- “Services Trade Liberalization and Regulatory ing Country Perspective. Washington, DC: World Reform: Re-invigorating International Coopera- Bank. tion.� Policy Research Working Paper Series 5517, Grantham-McGregor, Sally, Yin Bun Cheung, San- World Bank, Washington, DC. tiago Cueto, Paul Glewwe, Linda Richter, Barbara Holzer, Harry, and Robert Lerman. 2009. The Future Strupp, and the International Child Development of Middle-Skill Jobs. Washington, DC: Center on Steering Group. 2007. “Development Potential in Children and Families, Brookings Institution. the First 5 Years for Children in Developing Coun- Hsieh, Chang-Tai, and Peter J. Klenow. 2009. “Misal- tries.� Lancet 369 (January): 60–70. location and Manufacturing TFP in China and Gratton, Lynda. 2011. The Shift: The Future of Work India.� Quarterly Journal of Economics 124 (4): Is Already Here. London: HarperCollins. 1403–48. Grimm, Michael, Jens Kruger, and Jann Lay. 2011. ———. 2011. “The Life Cycle of Plants in India and “Barriers to Entry and Returns to Capital in Infor- Mexico.� Chicago Booth Research Paper 11-33, mal Activities: Evidence from Sub-Saharan Africa.� Booth School of Business, University of Chicago. Review of Income and Wealth 57 (S1): S27–S53. Ibarrarán, Pablo, Laura Ripani, Bibiana Taboada, Hallward-Driemeier, Mary, Gita Khun-Jush, and Lant Juan Miguel Villa, Brigida Garcia. 2012. “Life Pritchett. 2010. “Deals Versus Rules: Policy Imple- Skills, Employability and Training for Disadvan- mentation Uncertainty and Why Firms Hate It.� taged Youth: Evidence from a Randomized Evalu- Working Paper Series 16001, National Bureau of ation Design.� IZA Conference Paper, May 12, Economic Research, Cambridge, MA. 2012. Processed. IFC (International Finance Corporation). Forthcom- Haltiwanger, John. 2011. “Globalization and Eco- ing. IFC Job Study: Assessing Private Sector Contri- nomic Volatility.� In Making Globalization Socially butions to Job Creation. Washington, DC: IFC. Sustainable, ed. Marc Bacchetta and Marion Jan- ILO (International Labour Organization). 1998. Dec- sen, 119–46. Geneva: International Labour Orga- laration on Fundamental Principles and Rights at nization and World Trade Organization. Work. Adopted by the International Labour Con- Haltiwanger, John, Stefano Scarpetta, and Helena ference at its 86th session, ILO, Geneva, June 18. Schweiger. 2008. “Assessing Job Flows across ———. 2002. Decent Work and the Informal Econ- Countries: The Role of Industry, Firm Size, and omy. Geneva: ILO. Regulations.� Working Paper 13920. National ———. 2007. Resolution Concerning Updating the Bureau of Economic Research, Cambridge, MA. International Standard Classification of Occupa- Hanushek, Eric A., and Ludger Woessmann. 2008. tions. Adopted by the Tripartite Meeting of Ex- “The Role of Cognitive Skills in Economic De- perts on Labour Statistics on Updating the Inter- velopment.� Journal of Economic Literature 46 (3): national Standard Classification of Occupations, 607–88. ILO, Geneva, December 6. Hatløy, Anne, Tewodros Kebede, Huafeng Zhang, ———. 2010. Accelerating Action against Child La- and Ingunn Bjørkhaug. 2012. “Perception of bour. Geneva: ILO. Good Jobs: Sierra Leone.� Background paper for ———. 2011. Global Employment Trends for Youth. the WDR 2013. Geneva: ILO. Moving jobs center stage    45 ———. 2012a. Global Employment Trends 2012: Pre- Kraay, Aart. 2012. “How Large Is the Government venting a Deeper Jobs Crisis. Geneva: ILO. Spending Multiplier? Evidence from World Bank ———. 2012b. ILO Global Estimate of Forced Labour: Lending.� Quarterly Journal of Economics 127 (2): Results and Methodology. Geneva: ILO. 1–59. ———. 2012c. “What Are the Key Challenges Facing Laeven, Luc, and Christopher Woodruff. 2007. “The Labour Statistics Today?� ILO, Geneva. Processed. Quality of the Legal System, Firm Ownership, and ILO and World Bank. 2012. Inventory of Policy Re- Firm Size.� Review of Economics and Statistics 89 sponses to the Financial and Economic Crisis: Joint (4): 601–14. Synthesis Report. Washington, DC: ILO and World Lee, Jean, and David Newhouse. 2012. “Cognitive Bank. Skills and Labor Market Outcomes.� Background IMF (International Monetary Fund). 2003. “Growth paper for the WDR 2013. and Institutions.� In World Economic Outlook: Levi, Margaret, Christopher Adolph, Aaron Erlich, April 2003; Growth and Institutions, 95–128. Anne Greenleaf, Milli Lake, and Jennifer Noveck. Washington, DC: IMF. 2012. “Aligning Rights and Interests: Why, When, Inchauste, Gabriela. 2012. “Jobs and Transitions out and How to Uphold Labor Standards.� Back- of Poverty: A Literature Review.� Background pa- ground paper for the WDR 2013. per for the WDR 2013. Levine, Ross. 2005. “Finance and Growth: Theory and Evidence.� In Handbook of Economic Growth, ed. Inchauste, Gabriela, Sergio Olivieri, Jaime Saavedra Philippe Aghion and Steven Durlauf, 865–934. Chanduvi, and Hernan Winkler. 2012. “Decom- Amsterdam: Elsevier. posing Recent Declines in Poverty: Evidence from Levy, Santiago. 2008. Good Intentions, Bad Outcomes, Bangladesh, Peru, and Thailand.� Background pa- Social Policy, Informality, and Economic Growth in per for the WDR 2013. Mexico. Washington, DC: Brookings Institution IOM (International Organization for Migration). Press. 2008. World Migration Report 2008: Managing Lin, Justin Yifu. 2012. Demystifying the Chinese Econ- Labor Mobility in the Evolving Global Economy. omy. Cambridge, U.K.: Cambridge University Press. Geneva: IOM. Locke, Richard. Forthcoming. Beyond Compliance: ———. 2010. World Migration Report 2010. The Fu- Promoting Labor Justice in a Global Economy. New ture of Migration: Building Capacities for Change. York: Cambridge University Press. Geneva: IOM. Locke, Richard, Fei Quin, and Alberto Brause. 2007. Kanamori, Tokishi, and Zhijun Zhao. 2004. Pri- “Does Monitoring Improve Labor Standards? vate Sector Development in the People’s Repub- Lessons from Nike.� Industrial and Labor Relations lic of China. Manila: Asian Development Bank Review 61 (1): 3–31. Institute. Luke, Nancy, and Kaivan Munshi. 2011. “Women as Kanbur, Ravi. 2009. “Conceptualizing Informality: Agents of Change: Female Income and Mobility Regulation and Enforcement.� Indian Journal of in India.� Journal of Development Economics 94 Labour Economics 52 (1): 33–42. (1): 1–17. Kanbur, Ravi, and Jan Svejnar, eds. 2009. Labor Mar- Lyon, Scott, Furio C. Rosati, and Lorenzo Guarcello. kets and Economic Development. Routledge. 2012. “At the Margins: Young People neither in Kebede, Tewodros, Anne Hatløy, Huafeng Zhang, Education nor in Employment.� Background pa- and Ingunn Bjørkhaug. 2012. “Perception of per for the WDR 2013. Good Jobs: Egypt.� Background paper for the Maloney, William F., and Jairo Núñez Méndez. WDR 2013. 2003.“Measuring the Impact of Minimum Wages: Keefer, Philip. 2009. “Governance.� In The SAGE Evidence from Latin America.� Working Paper Se- Handbook of Comparative Politics, ed. Todd Land- ries 9800, National Bureau of Economic Research, man and Neil Robinson, 439–62. London: SAGE Cambridge, MA. Publications. McKenzie, David, and Christopher Woodruff. 2008. Kilroy, Austin. 2011. “Business Bridging Ethnicity.� “Experimental Evidence on Returns to Capital Ph.D. thesis, Massachusetts Institute of Technol- and Access to Finance in Mexico.� World Bank ogy, Cambridge, MA. Economic Review 22 (3): 457–482. King, Robert, and Ross Levine. 1993. “Finance and Mertens, Brian. 2011. “Forbes Asia’s Businessman of Growth: Schumpeter Might Be Right.� Quarterly the Year.� Forbes Asia Magazine, December 5. Journal of Economics 108 (3): 717–37. Montenegro, Claudio E., and Harry Anthony Pa- Klapper, Leora, Luc Laeven, and Raghuram Rajan. trinos. 2012. “Returns to Schooling around the 2006. “Entry Regulation as a Barrier to Entrepre- World.� Background paper for the WDR 2013. neurship.� Journal of Financial Economics 82 (3): Moser, Caroline O. N. 2009. Ordinary Families, Ex- 591–629. traordinary Lives: Assets and Poverty Reduction in 46   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Guayaquil, 1978–2004. Washington, DC: Brook- Jaime Saavedra-Chanduvi. 2007. Informality: Exit ings Institution. and Exclusion. Washington, DC: World Bank. Mryyan, Nader. 2012. “Demographics, Labor Force Pritchett, Lant. 2006. Let Their People Come: Breaking Participation, and Unemployment in Jordan.� the Gridlock on Global Labor Mobility. Washing- Working Paper Series 670, Economic Research ton, DC: Center for Global Development. Forum, Giza, Egypt. Psacharopoulos, George, and Harry Anthony Patrinos. Nabli, Mustapha K., Carlos Silva-Jáuregui, and Ahmet 2004. “Returns to Investment in Education: A Fur- Faruk Aysan. 2008. “Authoritarianism, Credibility ther Update.� Education Economics 12 (2): 111–34. of Reforms, and Private Sector Development in the Rama, Martín. 2001. “The Consequences of Doubling Middle East and North Africa.� Working Paper Se- the Minimum Wage: The Case of Indonesia.� In- ries 443, Economic Research Forum, Cairo. dustrial and Labor Relations Review 54 (4): 864–81. Narayan, Deepa, Lant Pritchett, and Soumya Kapoor. ———. 2009. “Making Difficult Choices: Vietnam in 2009. Moving Out of Poverty: Success from the Transition.� Working Paper Series 40, Growth and Bottom Up. New York: Palgrave Macmillan; Wash- Development Commission, World Bank, Wash- ington, DC: World Bank. ington, DC. Nelson, Richard R. 1981. “Research on Productivity Randriamaro, Zo. 2007. Gender and Trade: Overview Growth and Productivity Differences: Dead Ends Report (2006). Brighton, U.K.: BRIDGE. and New Departures.� Journal of Economic Litera- Ravallion, Martin. 2009. “Are there lessons for Africa ture 19 (3): 1029–64. from China’s Success against Poverty?� World De- Newitt, Kirsten. 2012. “Private Sector Voluntary Ini- velopment 37 (2): 303–13. tiatives on Labour Standards.� Background paper Rodrik, Dani. 2000. “Institutions for High-Quality for the WDR 2013. Growth: What They Are and How to Acquire North, Douglass C. 1981. Structure and Change in Them.� Studies in Comparative International De- Economic History. New York: W. W. Norton. velopment 35 (3): 3–31. ———. 1990. Institutions, Institutional Change and Rodrik, Dani, Arvind Subramanian, and Francesco Economic Performance. New York: Cambridge Trebbi. 2004. “Institutions Rule: The Primacy of University Press. Institutions over Geography and Integration in OECD (Organisation for Economic Co-operation Economic Development.� Journal of Economic and Development). 2006. OECD Employment Growth 9 (2): 131–65. Outlook: 2006. Paris: OECD. Romer, Paul Michael. 1993. “Idea Gaps and Ob- ———. 2009. OECD Reviews of Labour Market and ject Gaps in Economic Development.� Journal of Social Policies: Slovenia. Paris: OECD. ­Monetary Economics 32 (3): 543–73. ———. 2010. Off to a Good Start? Jobs for Youth. Rutkowski, Jan, Stefano Scarpetta, Arup Banerji, Paris: OECD. Philip O’Keefe, Gaëlle Pierre, and Milan Vodop- OECD and ILO. 2011. G20 Country Policy Briefs: ivec. 2005. Enhancing Job Opportunities: Eastern Brazil—Share of Formal Employment Continues Europe and the Soviet Union. Washington, DC: to Grow. Paris: OECD and ILO. World Bank. Özden, Çaglar, Christopher Parsons, Maurice Schiff, Rwanda Demobilization and Reintegration Commis- and Terrie L. Walmsley. 2011. “Where on Earth sion. 2012. Tracer: Community Dynamics and Pay- Is Everybody? The Evolution of Global Bilateral ment Verification Study. Kigali: Rwanda Demobili- Migration 1960–2000.� World Bank Economic Re- zation and Reintegration Commission. view 25 (1): 12–56. Sandefur, Justin. 2010. “On the Evolution of the Firm Paes de Barros, Ricardo, Mirela de Carvalho, Samuel Size Distribution in an African Economy.� Work- Franco, Rosane Mendoça. 2006. “Uma Análise das ing Paper Series 2010-5, Centre for the Study of Principais Causas da Queda Recente na Desigual- African Economies, Oxford. dade de Renda Brasileira.� Revista Econômica 8(1): Schoar, Antoinette. 2010. “The Divide between Sub- 117–147. sistence and Transformational Entrepreneurship.� Pagés, Carmen, ed. 2010. The Age of Productivity: In Innovation Policy and the Economy, vol. 10, ed. Transforming Economies from the Bottom Up. New Josh Lerner and Scott Stern, 57–81. Cambridge, York: Palgrave Macmillan. MA: National Bureau of Economic Research. Park, Jaegil, Daejong Kim, Yongseok Ko, Funnan Schumpeter, Joseph Alois. 1934. The Theory of Eco- Kim, Keunhyun Park, and Keuntae Kim. 2011. nomic Development: An Inquiry into Profits, Capi- “Urbanization and Urban Policies in Korea.� tal, Credit, Interest, and the Business Cycle. Cam- Korea Research Institute for Human Settlements. bridge, MA: Harvard University Press. Perry, Guillermo E., William F. Maloney, Omar S. Selim, Nadia. 2012. “Innovation for Job Creation.� Arias, Pablo Fajnzylber, Andrew D. Mason, and Background paper for the WDR 2013. Moving jobs center stage    47 Shiferaw, Admasu, and Arjun S. Bedi. 2010. “The Dy- ———. 2004. World Development Report 2005: A namics of Job Creation and Job Destruction: Is Better Investment Climate for Everyone. New York: Sub-Saharan Africa Different?� Poverty, Equity Oxford University Press. and Growth Discussion Papers 22, Courant Re- ———. 2005. Global Economic Prospects: Economic search Centre, Göttingen, Germany. Implications of Remittances and Migration. Wash- SMERU Research Institute. 2001. Wage and Employ- ington, DC: World Bank. ment Effects of Minimum Wage Policy in the Indo- ———. 2006a. Chile Development Policy Review. nesian Urban Labor Market. Jakarta: SMERU Re- Washington, DC: World Bank. search Institute. ———. 2006b. World Development Report 2007: Stampini, Marco, and Audrey Verdier-Choucane. Development and the Next Generation. Washing- 2011. “Labor Market Dynamics in Tunisia: The ton, DC: World Bank. Issue of Youth Unemployment.� Discussion Paper ———. 2007. Rwanda: Toward Sustained Growth and Series 5611, Institute for the Study of Labor, Bonn. Competitiveness, Volume I, Synthesis and Priority Sutton, John, and Nebil Kellow. 2010. An Enterprise Measures. Washington, DC: World Bank. Map of Ethiopia. London: International Growth ———. 2008. Chile: Toward a Cohesive and Well Gov- Centre. erned National Innovation System. Washington Syverson, Chad. 2011. “What Determines Productiv- DC: World Bank. ity?� Journal of Economic Literature (49) 2: 326–65. ———. 2009a. Doing Business 2010. Washington, TeamLease. 2010. Temp Salary Primer 2010. Ahmed- DC: World Bank. abad, India: TeamLease Services Pvt. Ltd. ———. 2009b. World Development Report 2009: Re- shaping Economic Geography. Washington, DC: United Nations. (UN). 2009. System of National Ac- World Bank. counts. New York: UN. ———. 2010. MDRP (Multi-Country Demobiliza- ———. 2011. World Urbanization Prospects: The tion and Reingration Program) Report. Washing- 2011 Revision. New York: United Nations, Depart- ton, DC: World Bank. ment of Economic and Social Affairs. ———. 2011a. World Development Report 2011: Con- UNDP (United Nations Development Programme). flict, Security, and Development. Washington, DC: 2003a. Early Warning Report: FYR Macedonia. World Bank. New York: UNDP. ———. 2011b. More and Better Jobs in South Asia. ———. 2003b. Early Warning System: Bosnia and Washington, DC: World Bank. Herzegovina. New York: UNDP. ———. 2011c. World Development Report 2012: Gen- United Nations Development Group. 2010. Thematic der Equality and Development. Washington, DC: Paper on MDG1: Eradicate Extreme Poverty and World Bank. Hunger, Review of Progress. New York: United ———. 2011d. Capabilities, Opportunities and Par- Nations. ticipation. Gender Equality and Development in Walker, Susan P., Theodore D. Wachs, Julie Meeks the Middle East and North Africa Region. A Com- Gardner, Betsy Lozoff, Gail A. Wasserman, Ern- panion Report to the World Development Report ersto Pollitt, and Julie A. Carter. 2007. “Child De- 2012. Washington, DC: World Bank. velopment: Risk Factors for Adverse Outcomes in ———. 2011e. “Fueling Growth and Competitive- Developing Countries.� Lancet 369 (9556): 145–57. ness in Poland through Employment, Skills, Wietzke, Frank-Borge, and Catriona McLeod. 2012. and Innovation.� Technical report, World Bank, “Jobs, Well-Being, and Social Cohesion: Evidence Washington, DC. from Value and Perception Surveys.� Background ———. 2012. Job Trends. Washington, DC: World paper for the WDR 2013. Bank. Winters, Alan, Terrie Walmsley, Zhen Kun Wang, and Young, Mary Eming, and L. M. Richardson, eds. 2007. Roman Grynberg. 2002. “Negotiating the Liber- Early Child Development From Measurement to alization of the Temporary Movement of Natural Action: A Priority for Growth and Equity. Washing- Persons.� University of Sussex Discussion Paper ton, DC: The World Bank. 87, Sussex, U.K. Yusuf, Shahid, and Kaoru Nabeshima. 2006. Post- Witze, Morgen. 2010. “Case Study: Tata.� Financial Industrial East Asian Cities: Innovation for Growth. Times, December 29. Palo Alto: Stanford University Press. World Bank. 1992. World Development Report 1992: Zhang, Huafeng, Ingunn Bjørkhaug, Anne Hatløy, and Development and the Environment. New York: Tewodros Kebede. 2012. “Perception of Good Jobs: Oxford University Press. China.� Background paper for the WDR 2013. CHAPTER 1 The jobs challenge Demographic transitions, structural change, technological progress, and global volatility are changing the world of work. Yet, traditional farming and self-employment remain dominant in many countries. W orldwide, more than 3 billion people In the next 15 years, half of the population in have jobs, but the nature of their jobs developing countries will reside in urban areas, varies greatly. Some 1.65 billion have the result of a migration that is rapidly shifting regular wages or salaries. Another 1.4 billion work from the farm to the factory or the street. work in farming and small household enterprises, The rural-urban shift generally improves indi- or in casual or seasonal day labor. The majority vidual well-being, especially for those who find of workers in the poorest countries are engaged wage employment. Still, workers’ share of global in these types of work, outside the scope of an income may be declining, a pattern attributed in employer-employee relationship. Another 200 part to globalization and technological change. million people, a disproportionate share of them Wages for the same occupation are converging youth, are unemployed and actively looking for across countries, but a higher premium is paid work. Almost 2 billion working-age adults are for more skilled occupations. Women’s earnings neither working nor look­ ing for work; the still lag behind those of men, and the fraction of majority of these are women, and an unknown them who work varies enormously across coun- number are eager to have a job.1 tries. While women’s labor force participation The jobs challenge facing the world is multi- exceeds 75 percent in Vietnam, it is only 28 per- faceted, ranging from improving aspects of the cent in Pakistan. work people do, to supporting the reallocation As the world changes, so do jobs. Despite im- of people to better jobs, to creating jobs for provements in workers’ education levels, many those who want to work. Youth bulges in some firms report that they have difficulty finding the countries are bringing in millions of new job skilled workers they seek. Part-time and tempo- seekers. Sub-Saharan Africa’s labor force grows rary work appear to be increasing. In India and by about 8 million people every year. South South Africa, for example, there has been a Asia’s grows by 1 million people every month. sharp rise in the number of temporary employ- Elsewhere, the working population is rapidly ag- ment services and labor brokers. Outsourcing ing, and more and more workers are putting off was once concentrated in manufacturing, but retirement. By 2020, more than 40 million ad- new technology is now enabling the splintering ditional jobs will be needed for people 65 years of tasks in services. Meanwhile, new platforms and older.2 on the internet and mobile phones offer innova- Structural and technological changes are tions for matching workers and employers, and moving more people from rural areas to cities. not only for highly skilled jobs. The jobs challenge   49 A job, but not always a salary people would like to work more hours, whereas others would rather not. More than 15 percent To many, the word job brings to mind a worker of those employed in Armenia, Colombia, Gua- with an employer and a regular paycheck. Yet, temala, and Peru, but less than 3 percent in this narrow definition excludes nearly 1.4 billion Hungary, Pakistan, Portugal, and the United people who work for a living. The concept of a States worked fewer hours than desired.5 Some job is actually much broader than wage employ- people who would like to work have no job. ment. Jobs are activities that generate actual or Unemployment rates vary over the business cy- imputed income, monetary or in kind, formal or cle; around 2009 and 2010, when the worst of informal. But not all forms of work can be the international crisis hit most countries in deemed jobs. Activities performed against the the world, they ranged from more than 20 per- will of the worker or involving violations of fun- cent in South Africa and Spain to less than 5 per- damental human rights should not be consid- cent in Austria, Korea, Malaysia, Singapore, Sri ered jobs. Some other activities that entail work Lanka, and Thailand.6 effort, such as cooking and cleaning at home, are The world of work is more diverse in devel- not considered jobs unless they are performed by oping countries than in developed countries. people hired and paid for the work. This diversity refers not only to the number of hours worked and number of jobs available, usual yardsticks in developed countries, but Multiple forms of work also to characteristics of jobs. Two main aspects Defining and measuring jobs is challenging be- stand out. First, there is prevalence of self- cause the ways people spend their time and work employment, which often makes measures of are diverse (question 1). Economists usually dis- unemployment and underemployment inade- tinguish between work and leisure, but the real- quate.7 Second, the coexistence of traditional ity is more complex. Time can be allocated to and modern modes of production leads to nonproduction and production activities. The large variations in the nature of work, from first category includes time spent eating, sleep- subsistence agriculture and menial work to ing, schooling, and at recreation. Production in- technology-driven manufacturing and services. cludes both market and nonmarket work. Work across the developing world is charac- Whether an activity is considered production terized by a high prevalence of informality, can be assessed based on a third-person test: “if whether defined on the basis of firm registra- an activity is of such character that it might be tion, social security coverage, or a written em- delegated to a paid worker, then that activity ployment contract. Informal employment is shall be deemed productive.�3 not under the purview of labor regulations, ei- Patterns in time allocated to production dif- ther because of their limited scope or because fer across countries and over time. Jobs that they are deliberately avoided or evaded. Re- span eight hours a day, five days a week, with gardless of the specific definition used, infor- paid vacation, are not the norm in developing mal employment is generally associated with countries. Some jobs involve a few hours of lower productivity. However, this does not nec- work during certain days of the week or certain essarily mean that firm registration, social secu- weeks of the year; others entail long hours most rity coverage, or a written contract would result days of the week almost every week of the year. in greater efficiency. Informality can be a symp- Some people have had only one job in the pre­ tom of lower productivity as much as it can be vious week, while others have engaged in two a cause of it. or more jobs. In 2011, temporary employment represented more than one-fifth of total wage Different places, different jobs employment in the Republic of Korea and Spain, but around 5 percent in Australia and the Slovak Self-employment and farming represent almost Republic.4 half the jobs in the developing world. The vast The measurement of unemployment or un- majority of those in self-employment work deremployment is equally challenging. Some in small enterprises with no paid employees.8 50   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 A F I G U R E 1.1   job does not always come with a wage men women 100 wage employment share of total employment, % 80 self-employment 60 nonwage 40 employment 20 farming 0 Europe and Latin America South Middle East East Asia Sub-Saharan Central Asia and the Asia and and Paci�c Africa Caribbean North Africa Source: World Development Report 2013 team. Note: Data are for the most recent year available. But shares of wage work, farming, and self- explained by education, experience, or sector of employment differ greatly by gender and across work. countries. Nonwage work represents more A growing share of youth, typically defined than 80 percent of women’s employment in as people ages 15 to 24, is in schooling or in Sub-Saharan Africa, but less than 20 percent in training. Still, youth unemployment reaches countries of Eastern Europe and Central Asia alarming levels in some countries (above 40 per- (figure 1.1). cent in South Africa since early 2008 and above Gender differences are also striking.9 World- 50 percent in Spain in early 2012).10 Even in wide, less than 50 percent of women have jobs, countries where it is relatively low, the youth un- whereas almost 80 percent of men do. Roughly employment rate is twice or more the national 50 percent of both working men and working average. In addition, a large share of young peo- women are wage earners, but this statistic hides ple are considered “idle�—not in education, not substantial variation across countries and re- employed, and not in training or looking for gions. Women are significantly underrepre- work (figure 1.2).11 In some countries, more sented in wage employment in low- and lower- than one-third of 15- to 24-year-olds are idle; in middle-income countries, but are more likely most countries, unemployment rates are small than men to work for wages in middle-income compared to idleness rates.12 In many cases, countries. In countries such as Pakistan, where when youth work they do so in unpaid jobs. If 28 percent of women but more than 82 percent paid, they are less likely to have access to social of men participate in the labor force, wage em- security.13 ployment is a much lower share of total employ- Although child labor is in decline, it still af- ment among women than among men. Even in fects 1 in 8 children (1 in 14 in the case of haz- countries such as Tanzania and Vietnam, where ardous work). The International Labour Orga- participation rates are above 75 percent for both nization (ILO) defines child labor as any work men and women, wage employment still lags by a child under age 12 or, for a child above behind for women. Beyond these stark con- age 12, any work that impedes education or is trasts, women continue to earn significantly less damaging to health and personal development. than men. And these differences are not fully Worldwide, 306 million children were at work in The jobs challenge    51 2008. Of these, 215 million were engaged in ac- tivities that constituted child labor, and 115 mil- Among F I G U R E 1. 2   youth, unemployment is not always lion were involved in hazardous work.14 Most of the issue these children are unpaid family workers or par- ticipate in farming. More than half live in Asia not in school or at work and the Pacific; but the share is highest is in Sub- not looking for work looking for work Saharan Africa, where child labor affects 1 in 4 children (or 65 million of them). Pakistan women 2008 men Turkey 2005 Youth bulges, aging societies, and migrant nations India 2009 Indonesia Demographic shifts can be massive, but they do 2010 not always go in the same direction. The most Chile populous countries in the world have experi- 2009 enced very large increases in their labor force: Brazil 2009 nearly 8 million new entrants a year in China and 7 million a year in India since the early Ukraine 2005 1990s. (These rates are now decelerating rapidly, Ghana particularly in China.) Many smaller countries 2005 face large relative increases, even if the absolute Tanzania numbers are less astounding. In other countries, 2009 the overall population and the labor force are 0 10 20 30 40 50 60 shrinking. For example, Ukraine’s labor force is share of population ages 15–24, % estimated to decrease by 0.75 percent annually, the equivalent of approximately 160,000 fewer people every year.15 Source: World Development Report 2013 team. A simple conceptual exercise illustrates the challenges raised by these dramatic demographic transitions. To keep the ratio of employment to working-age population constant, in 2020, there 42 million jobs will have to be generated by 2020 should be around 600 million more jobs than in to cope with the growth in the number of older 2005. More than 175 million of them, or nearly people. One-quarter of these jobs will need to be 1 million a month, would be needed in East Asia in China, even though the size of the Chinese and the Pacific as well as in South Asia (figure labor force will have started to decline in abso- 1.3). The number of jobs in Sub-Saharan Africa lute terms.16 would have to increase by about 50 percent, International migration is also changing the which translates into employment growth of 2.7 size and composition of the labor force in many percent a year. But in Eastern Europe and Cen- countries. At the turn of the century, there were tral Asia, where populations are aging, only 2.4 more than 200 million international migrants million new jobs would be needed during the worldwide, and nearly 90 million of them were same period. workers. If international migrants constituted a The age structure of the labor force, not sim- nation, theirs would be the fifth-largest in the ply its size, matters as well. Youth are staying in world, ahead of Brazil. Precise estimates diverge school longer and entering the labor market but there is agreement that migrants represent later, whereas adults are living longer and nearly 3 percent of the world population.17 healthier lives. The labor force participation rate These aggregate figures hide important dif- of people 65 years of age and older has remained ferences across countries. Some are large migra- relatively stable over the past two decades. Even tion recipients either in absolute numbers (for with a stable participation rate, however, close to instance, the United States) or in relative terms 52   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Employment F I G U R E 1. 3   growth is needed to cope with population growth a. Jobs per year b. Annual change in employment South Asia Sub-Saharan Africa Middle East East Asia and Paci c and North Africa Sub-Saharan Africa South Asia Latin America Latin America and the Caribbean and the Caribbean Middle East East Asia and Paci c and North Africa Europe and Central Asia Europe and Central Asia 0 5 10 15 0 1.0 2.0 3.0 millions percent Source: World Development Report 2013 team based on data from the International Labour Organization and World Development Indicators. Note: Estimations are for the period 2005–20, based on projected population growth assuming constant employment rates. (Jordan and Singapore). Migrants from coun- been the case.22 Population movements away tries such as Bangladesh, India, and Mexico rep- from agriculture were indeed associated with resent a large share of total migrants worldwide; rapid economic growth in East Asia; much less countries such as Fiji, Jamaica, and Tonga have a so in Sub-Saharan Africa (figure 1.4). large share of their population overseas. Figures Technological change induces families to in- for some of the smaller countries are striking. crease market production in place of home pro- For instance, nearly one-fifth of all Salvadorians duction.23 Structural change has increased the live abroad, while more than three-fifths of the time devoted to consumption or investment population in Kuwait, Qatar, and the United activities but it has also meant more work. In Arab Emirates is foreign-born.18 some cases, structural change has even led to ­reduced leisure,24 particularly for workers who shift to market jobs with low productivity.25 In Cities, wages, and women the past, this process of structural change often took decades, but in many developing coun- Economic development brings significant tries, it is transforming lives within a generation changes in the composition of the labor force, a (box 1.1). process known as structural transformation.19 The shift from home to market production is Before 2020, more than half of the total popula- not gender neutral, because women have tended tion in developing countries is expected to be to specialize in home production whereas men living in cities and towns.20 That means that the traditionally have focused more on market pro- growth of the nonagricultural labor force will duction. As women move into jobs, they often vastly exceed the growth of the agricultural la- continue to work at home. When both jobs and bor force.21 Urbanization derives from growth home activities are considered, women are gen- in agricultural productivity that sustains higher erally busier than men. This is so in rich and standards of living. It can also be associated poor countries. Evidence from Sub-Saharan Af- with rapid economic growth at the aggregate rica, Europe, and India shows that women spend level, because urban jobs tend to be more pro- more time on production activities than men ductive than rural jobs. But that has not always do.26 But when they take on jobs—especially The jobs challenge    53 Moving F I G U R E 1. 4   from farms to cities does not always bring economic growth a. East Asia and Paci c b. Sub-Saharan Africa 75 75 Malaysia Liberia 60 60 Cameroon share of urban population, % share of urban population, % Nigeria Indonesia Ghana 45 45 China Guinea- Bissau 30 30 Thailand Madagascar Zimbabwe Vietnam 15 15 Kenya Ethiopia 0 0 80 800 8,000 80 800 8,000 GDP per capita, constant 2000 US$ GDP per capita, constant 2000 US$ Source: World Development Indicators. Note: Data correspond to changes between 1985 and 2010. GDP = gross domestic product. The BOX 1.1   nature of work and leisure change as cities develop The process of structural transformation can be seen today in small are preferred. Practices surrounding recruitment have also become and medium-size cities throughout the developing world. The pro- tighter. “It is public knowledge,� he added, “that to be accepted in cess has been so fast that most people, even young ones, remem- the company, candidates must have an inside connection because ber the old rural setting. As they have embraced the traits of mod- more and more people need work, while the number of job oppor- ern urban life, their jobs and their leisure have changed so much tunities is limited.� Many factory workers in the neighborhood that they do not contemplate returning to the old mores. work on a six-month contract and hope for its renewal. The best A neighborhood of 5,000 outside the city of Tangerang, Indone- local job available is often reported to be running one’s own busi- sia, started coming to life 25 years ago, when iron, rubber, garment, ness. Even when faced with possible unemployment, a 41-year-old and thread factories sprang up in the surrounding areas. Some in noted he had “never thought of returning to the village. That is the neighborhood still work as farmhands. But most women and desperate. Don’t be desperate. Find another job and don’t get men earn their living in the factories, as well as in a variety of jobs picky.� that rose up alongside them. Many men drive ojeks (motorcycle Comparing the situation now to the early days, a local official taxis); women sell sweets and other items by the gates of the facto- described the neighborhood’s busy market as “cleaner and more ries. Residents also make a living through home-based garment strategic now, and there are more sellers or merchants, so there are piecework, handicrafts such as broom-making, and construction more options. Public transportation to the market is more accessible work; some are civil servants and teachers. Poorer women collect now. It used to be hard to find, and the streets used to be muddy.� and resell scraps of fabric. The flow of newcomers has also offered He estimated that poverty in the neighborhood had fallen by half, an income-generating opportunity to local families from renting from 20 percent at the turn of the century to around 10 percent now. rooms for lodging. Seemingly everyone in the neighborhood has a He indicated that the neighborhood had weathered the global cell phone and gets around on their own motorcycle. financial crisis well and that factory workers had been able to keep A 28-year-old steelworker and father of one said that getting their jobs. These changes to work come with changes in leisure. factory work used to be a lot easier when the community was Young men now spend time on computer chats and playing video newer. Junior high school graduates were still accepted for work games. “We used to send letters through the post office,� recalls then. Now, he said, production workers must have at least a senior a 22-year-old. “Now, nobody wants to go to the post office. . . . high school degree, and vocational school graduates under age 30 It’s beneath them. Now everyone has cell phones.� Source: World Bank 2011a. 54   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Jobs BOX 1.2   bring earnings opportunities to women, but also new difficulties Jobs can transform women’s roles in households and in society more easily than men because employers have “more confidence in more broadly. In a community of 3,000 inhabitants outside Durban, them [the women].� South Africa, 80 percent of the women were estimated to be work- Increased labor force participation is not without challenges. ing outside their home, mainly as teachers and nurses but also in The women in Santiago worry about their safety because of assaults, offices and retail outlets or with the police. Women did not have theft, and gang violence; many think it is too dangerous to work at such opportunities 10 years ago. “Women are no longer regarded night. Despite the expansion of work opportunities, mobility con- as housewives,� a young woman remarked in a Durban focus tinues to be a constraint not only because of poor transportation group. and safety risks but also because of the roles women play as income- In a bustling neighborhood in East Jakarta, in Indonesia, women earners and as caregivers in the household. In rural areas, women work as street sellers, peddling food, glasses, plates, and carpets. face difficulties in traveling for farm work and other jobs outside Young women explained that selling carpets is hard work that their villages because of traditional cultural and gender norms. women do “because they have to help support their families. Their In a semi-urban area outside Cuzco, in Peru, large numbers of husbands’ jobs don’t generate enough income.� With more educa- women are now engaged in home-based handicrafts, sewing, and tion than their elders, young women in this area of Jakarta are more diverse agricultural activities. Outside the home, they take up farm likely to be working in the nearby ceramic factory or in one of the jobs, run their own small shops or restaurants, or work for hire in shops or beauty salons in the neighborhood’s markets or at the new these places. The women explained, however, that they consider shopping malls in the city. home-based activities (such as raising livestock) to be better jobs In a poor neighborhood of Santiago de los Caballeros, the than jobs outside the home, because “it’s peaceful work, and we can second-largest city in the Dominican Republic, only a few women look after the kids.� sold clothes and sweets in the streets a decade ago. Many are now Women encounter many difficulties in commuting across cities earning incomes from activities such as selling lottery tickets or run- for work. In Lautoka, an urban area in Fiji, the local economy is stag- ning small clothing stores and beauty salons. The incomes of these nant, and women are working in much larger numbers to help their women are vitally important to their families because men are households cope. Focus group participants explained that only men struggling. Factory opportunities and other jobs associated with “can take up jobs in a different town . . . but not the wife because the city’s free trade zone have been shrinking in recent years. Par- she has responsibilities at home to look after the kids and in-laws.� ticipants in a Santiago focus group said that women found work Source: World Bank 2011a. those that are outside the household setting— Transnational companies have built integrated their economic role changes (box 1.2). value chains and can tap into national skill pools around the world.30 Outsourcing is occurring in services as well as manufacturing. The share of Jobs are changing in surprising ways developing countries in exports of world services rose from 11 percent in 1990 to 21 percent in New technologies, globalization, and structural 2008.31 India has led the way in the information transformation have brought about remarkable technology (IT) sector, but other countries, such improvements in efficiency. Some developing as the Arab Republic of Egypt, have begun to fo- countries have managed to narrow the produc- cus attention on exporting services.32 tivity gap with industrial countries in only a few This changing landscape of global produc- decades.27 But others have failed to catch up and tion has also brought about shifts in skills en- the gap remains considerable for all de­ veloping dowments and in the distribution of top talent regions (figure 1.5). across countries. India and China rank high in The nature of work is changing as well. perceived attractiveness as hubs of outsourcing Industrial countries are experiencing a sus- because of their exceptionally high ratings in tained shift away from primary and traditional people skills and availability.33 India has close to manufacturing industries toward services and 20 million students in higher education, nearly knowledge-intensive jobs. At the same time, as many as the United States; both countries are technology improvements and greater reliance outpaced by China, with 30 million postsecond- on outsourcing to developing countries is lead- ary students.34 The United States still accounts ing to a decline in middle-skilled jobs.28 Technol- for a large share of international top scorers in ogy has allowed production tasks to be splintered student assessments, but Korea has the same and therefore performed in different locations.29 share as Germany, and both are closely followed The jobs challenge    55 Labor F I G U R E 1. 5   productivity remains low in developing countries 0.40 Europe and Central Asia 0.35 Latin America and the Caribbean GDP per worker relative to that in 0.30 high-income countries 0.25 Middle East and North Africa 0.20 0.15 East Asia and Paci c South Asia 0.10 Sub-Saharan Africa 0.05 0 1991 1995 2000 2005 2009 Source: World Development Indicators. Note: GDP = gross domestic product. Ratio is measured in 2005 purchasing power parity US$. by the Russian Federation. The number of high- ­ ttractiveness as hubs for the outsourcing of ser- a performing students in the city of Shanghai vices. Ghana and Senegal, ranked 26th and 28th, alone is one-fifth of that of G­ ermany and about come in significantly ahead of emerging market double that of Argentina.35 powerhouses like South Africa or Turkey.39 Skills are not one-dimensional, however. Dif- High-skilled niches are developing the world ferent jobs require different combinations of over. They tend to be l ­ocated in close proximity manual skills (needed for physical tasks), cogni- to centers of higher education in metropolitan tive skills (needed for mental tasks), and social areas with good infrastructure, from Cairo’s skills (needed to interact with others). The dis- Smart Village Business Park to Ghana’s IT En- tribution of employment by occupation can be abled ­ Service Industry cluster. 40 Bangalore and used to estimate the skill intensity of produc- Chennai in India and Suzhou in China have tion. As incomes rise, countries tend to use fewer emerged as global research and development manual skills in production, and more nonrou- hotspots. tine cognitive skills. 36 However, even for a given Technology itself is changing the way work- level of gross domestic product (GDP) per cap- ers and firms connect, through their access to ita, countries can use nonroutine skills to vary- much larger, even global, marketplaces for em- ing degrees (figure 1.6).37 ployment. Some of these marketplaces operate Technological progress expands the possibili- through the Internet; others use mobile phone ties for emerging and even low-income coun- technology.41 These changes are affecting work- tries to create jobs in higher-skilled production ers in developing countries and not just those in activities as well as to link to international value high-skilled occupations. Babajobs, for example, chains in services and manufacturing. In other was launched in 2009 and is now the largest words, technological progress enables countries digital marketplace for blue-collar jobs in India, to diverge from a linear evolutionary path from with more than 320,000 job listings and more manual skill intensity to the use of higher-order than 80,000 job seekers. cognitive and social skills.38 India and China top With changes in technology and the organi- the list of countries in an index measuring their zation of work, permanent jobs are becoming 56   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 The F I G U R E 1. 6   skills mix changes with economic development a. Manual skills b. Nonroutine cognitive skills 130 105 Rwanda 125 Lithuania 100 index, 100 = U.S. level index, 100 = U.S. level 120 India Egypt, Arab Rep. Egypt, Arab Rep. 115 95 Turkey Peru Turkey India 110 Indonesia Peru 90 Indonesia 105 Lithuania Rwanda 100 85 200 2,000 20,000 200 2,000 20,000 GDP per capita, constant 2000 US$ GDP per capita, constant 2000 US$ Source: Aedo and others 2012 for the World Development Report 2013. Note: GDP = gross domestic product. All skill intensities are measured relative to the U.S. level. Each data point shows the skill intensity of national production, derived from tran- scribing the occupational structure to individual skills, following Autor, Levy, and Murnane 2003. less common. Part-time and temporary wage Prosperity, but a changing employment (also called nonstandard employ- distribution of earnings ment) is now a major feature of industrial coun- tries. More than half the firms in the United Earnings from work increase with economic de- States expect to raise the share of their part-time velopment, and the benefits associated with jobs and temporary employees over the next five improve as well (figure 1.7). The relationship is years.42 The trend is also evident in developing not mechanical, but growth is unambiguously countries (box 1.3). good for jobs. Part of the change in earnings and People’s jobs may not match their aspirations. benefits stems from the higher average skills that Surveys in high-income countries show that as economies gain as they become more devel- many as half of all workers—among both self- oped; part comes from workers with the same employed and wage employees—would prefer skills enjoying better opportunities. to be their own bosses.43 Percentages are lower Poverty has declined in the developing world, in low- and middle-income countries, where to a large extent through jobs. The share of the a larger fraction of the labor force works in population of the developing world living on household enterprises or on farms. On average, less than US$1.25 a day (in purchasing power about one-fourth of adults from 35 countries in parity, or PPP) fell from 52 percent in 1981 to 22 Eastern Europe and Central Asia prefer self- percent in 2008, or from 1.94 billion to 1.29 bil- employment, but the rate varies from 10 percent lion people.47 This reduction is the result of in Azerbaijan and Hungary to 43 percent in multiple factors, but the creation of millions of Belarus and Turkey.44 In poorer countries, self- new, more productive jobs, mostly in Asia but employment is often a choice of last resort, in also in other parts of the developing world, has part because of the inability to find salaried em- been the main driving force.48 More people have ployment.45 Owning a small business is a goal to jobs now than ever before, and those jobs pro- which the poor do not always aspire.46 vide generally higher earnings. The jobs challenge    57 The BOX 1.3   temporary staffing industry is growing in developing countries A decade ago, the temporary staffing industry was seen as irrele- the modern industry of temporary staffing is only 15 years old, vant outside of high-income countries. But it is now growing rapidly and is developing rapidly. The number of temporary workers in some developing countries, even beyond large cities.a This recruited by labor brokers grew more than 10 percent in 2009 and growth is often viewed as a response to the complex regulatory 18 percent in 2010. According to some media reports, workers are framework facing employers. Temporary staffing also allows more quitting permanent jobs to move into more attractive temporary flexibility in the management of peak workloads and in adjusting roles. Some firms claim that as many as 15 percent of new recruits staffing levels up or down in line with business demands. are permanent employees switching to temporary jobs.d Competi- Depending on the context, the temporary staffing industry tion in the Indian temporary staffing industry is strong. Agencies employs different types of workers, from mainly entry-level, previ- have introduced lower recruitment fees to gain more market share ously unemployed workers seeking to gain experience to highly edu- and to drive growth. Large temporary staffing firms are entering cated, mid-level career employees looking to fast-track their careers. niche activities such as business consulting (Manpower) and train- In South Africa, temporary workers make up about 7 percent of ing (TeamLease).e the labor force; the temporary staffing industry provides employ- As temporary staffing grows, so do calls to examine the regula- ment to an average of 410,000 workers a day.b Finance—the sector in tory framework of the industry.f Some of those calls focus on the statistics that includes temporary staffing—was a close second addressing vulnerability. Workers in these jobs typically face lower to retail in employment growth from 1994 to 2009. Unskilled jobs earnings (because a portion of the pay is diverted to temporary and service-related occupations dominate the employment distri- staffing agencies). They also face a lack of benefits, coverage by bution within the labor brokering subsector. A worker employed in labor laws, and job security. Other efforts focus on professionalizing temporary staffing services is less likely to contribute to pension the industry. For instance, in 2011 TeamLease and seven other staff- funds or health insurance and is generally seen as more vulnerable. ing firms formed the Indian Staffing Federation to advocate for Temporary forms of employment have existed in India for changes in labor laws and more acceptance for the industry in a decades, partly as a way to circumvent rigid labor laws.c However, country where a vast majority of the labor force is unorganized. Source: World Development Report 2013 team. a. Dourgarian (2011) remarks that it is not the Group of 8 countries that led the pack in the growth in the staffing industry in 2011, but the BRICs (Brazil, India, the Russian ­ Federation, and China), along with Indonesia, Mexico, and Pakistan. b. The discussion on South Africa is drawn from Bhorat (2012) for the World Development Report 2013. c. World Bank 2011b. d. TeamLease 2010. e. Bajaj 2011. f. ILO 2011; Musgrave 2009. Jobs F I G U R E 1. 7   provide higher earnings and benefits as countries grow a. Average wage b. Social security coverage 100,000 100 programs, % of total employment average wage in manufacturing, contributors to social security 80 2005 PPP US$ 10,000 60 40 1,000 20 100 0 300 3,000 30,000 300 3,000 30,000 GDP per capita, 2005 PPP US$ GDP per capita, 2005 PPP US$ Source: World Development Report 2013 team. Note: GDP = gross domestic product; PPP = purchasing power parity. Each dot represents a country. 58   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Yet in a majority of countries, both indus- does not have a role to play. The quality of the trial and developing, the share of labor in total civil service is critically important for develop- income is declining.49 This trend, which has ment, whether it is teachers building skills, agri- been observed since the mid-1980s and early cultural extension agents improving agricul- 1990s, has been attributed to various forces, tural productivity, or urban planners designing from technological progress biased toward functional cities. Public works programs or em- skilled workers to global competition under- ployment programs for the demobilization of mining workers’ bargaining power. The en- combatants are also justified in some circum- trance of China and India in world trade has stances. But the private sector is the main en- doubled the size of the globalized labor force, gine of job creation and the source of roughly hence reducing the price of labor relative to that nine of 10 jobs in the world. Between 1995 and of other factors of production.50 2005, the private sector accounted for almost 90 Changes in the distribution of income in percent of jobs created in Brazil. In the Philip- recent years have actually taken place not only pines and Turkey, the fraction reached 95 between factors of production but between percent.51 workers. Two distinctions are relevant in this re- But the most remarkable example of the spect: between high- and low-skill workers, and expansion of employment through private between those who work in tradable sectors and sector growth is the case of China. In 1981, those who do not. Tradable sectors produce private sector employment accounted for 2.3 goods and services that can be exported or im- million workers while state-owned enterprises ported, such as shirts or computers. A turning (SOEs) had 80 million workers. Two decades point came in the mid-1990s, when labor earn- later, employment in private sector firms ac- ings in developing countries started to grow counted for 74.7 million workers surpassing, faster than those in industrial countries, regard- for the first time, the 74.6 million workers in less of level of skill. But the trend is more pro- SOEs (figure 1.10).52 nounced in tradable sectors, whereas low-skill In contrast to the global average, in some workers in nontradable sectors continue to be countries in the Middle East and North Africa, the most disadvantaged (figure 1.8). the state keeps a leading role as an employer—a The relatively lower wages among low-skill pattern that can be linked to the political econ- workers compared with high-skill workers, par- omy of the post-independence period and, in ticularly in nontradable sectors in developing some cases, to the abundance of oil revenues.53 countries, are consistent with an increase in re- For a long period, public sector jobs were of- turns to education. Returns to education mea- fered to young college graduates. In recent years, sure the wage premium on higher educational however, the fiscal space for continued expan- attainment for workers of the same sex with the sion in public sector employment shrank, and same age and work experience. In all regions, “queuing� for public sector jobs became more more schooling is associated with higher labor prevalent, leading to increasing transitions into earnings, but the gain is not linear. The labor informality, a devaluation of education creden- earnings of workers with tertiary education are tials, and forms of social exclusion.54 A relatively double or more than those of workers with sec- well-educated and young labor force remains ondary education only. However, workers with unemployed, or underemployed, and labor pro- only a secondary education earn little more than ductivity stagnates.55 those with elementary education. The education premium is generally higher the lower the in- come level of the country (figure 1.9). Vulnerability on a global scale Jobs are vulnerable to economic downturns— The role of the private sector and much more so in the private than in the public sector. Short-term crises may wipe out The solution to all these demographic and years of progress. They may start in a single technological challenges rests with the private country and through globalization spread over sector. That does not mean that the government entire regions or, as in the recent one, to the The jobs challenge    59 Wages F I G U R E 1. 8   in developing countries are catching up a. Relative wages in nontradable sectors b. Relative wages in nontradable sectors 50 50 non-U.S. wages to U.S. wages, % non-U.S. wages to U.S. wages, % 40 40 30 30 20 20 10 10 0 0 1985 1990 1995 2000 2005 1985 1990 1995 2000 2005 low skilled (laborer in construction) low skilled (sewing machine operator) high skilled (accountant) high skilled (chemical engineer) c. Dispersion of wages in nontradable sectors d. Dispersion of wages in tradable sectors 1.4 1.4 cross -country wage inequality, cross-country wage inequality, 1.2 1.2 coe cient of variation coe cient of variation 1.0 1.0 0.8 0.8 0.6 0.6 1985 1990 1995 2000 2005 1985 1990 1995 2000 2005 low skilled (laborer in construction) low skilled (sewing machine operator) high skilled (accountant) high skilled (chemical engineer) Source: Based on Oostendorp 2012 for the World Development Report 2013. Note: The database used to construct this figure contains wage data by occupations for an unbalanced panel of more than 150 countries from 1983 to 2008. The data are derived from the International Labour Organization (ILO) October Inquiry database by calibrating the data into a normalized wage rate for each occupation. For a description of the data, see Freeman, Oostendorp, and Chor 2011. The vertical axis in panels a and b represents the ratio between U.S. wages and non-U.S. wages. The vertical axis in panels c and d repre- sents the coefficient of variation of wages—a measure of wage inequality—across all countries in the sample. whole world. In 1995, a financial crisis in Mex- rea. In 2007, an alarming rise in food prices be- ico engulfed most of Latin America as well as got problems with food supply and inflation, other emerging countries. In 1997, a speculative increasing poverty and reducing real wages in attack on Thailand’s currency severely affected parts of the developing world.56 the economies of Indonesia, Malaysia, and Ko- 60   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Returns F I G U R E 1. 9   to education are higher in poorer countries 25 private annual returns to education, % 20 15 10 5 0 world low income lower-middle income upper-middle income high income primary secondary tertiary Source: Montenegro and Patrinos 2012 for the World Development Report 2013. Note: Reported figures are unweighted averages of country-level private returns, for the most recent year within the period 2000–10 in a sample of 69 countries. F I G U R E 1.10   In China, employment growth is led by the private sector 110 100 90 number of workers, millions 80 70 60 50 40 30 20 10 0 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 20 state-owned enterprises individual rms (fewer than 8 workers) private rms (8 workers or more) foreign-owned companies Source: Kanamori and Zhao 2004. Note: Data for foreign-owned companies in 2002 and for non-state-owned enterprises in 2003 are not available. In 2008, the bursting of asset price bubbles creating 22 million new unemployed in a single and the resulting collapse of financial institu- year. Growth in total employment, which had tions in the United States and some European been hovering around 1.8 percent a year before countries initiated a crisis of worldwide span, 2008, fell to less than 0.5 percent in 2009, and by The jobs challenge    61 In F I G U R E 1.11   developing countries, the crisis affected earnings more than employment East Asia and Paci�c Europe and Central Asia Latin America and the Carribean 0 –2 change in growth rate, % –4 –6 –8 –10 –12 GDP growth employment growth earnings growth Source: Khanna, Newhouse, and Paci 2010. Note: GDP = gross domestic product. The vertical axis measures the difference in growth rates before and after the beginning of the crisis. 2011 had not yet returned to its pre-crisis level.57 countries. The less formalized the labor market, As Europe struggles with high levels of public the more earnings shrank and the less employ- debt, vulnerabilities in its banking sector, and ment numbers gave away. In Central and East- uncertainties about the euro, and as growth de- ern Europe, where the labor market is largely celerates in China and India, it is by no means formalized, the growth rate of GDP dropped on clear that the global crisis is over. average by 12 percentage points, employment The impact of the 2008–09 crisis varied contracted by 3 percent, and earnings fell by across developing countries. Some, such as the more than 8 percent (figure 1.11).62 In East Asia, Baltic countries, the Kyrgyz Republic, Mexico, where formal employment rates are very low, Romania, Russia, South Africa, and Turkey, had the average decline in GDP growth was 5.5 per- absolute declines in employment; other coun- centage points and total employment numbers tries such as Brazil, China, and Indonesia expe- barely changed.63 rienced only a brief deceleration. Country-­ The policy response to the crisis was unprec- specific studies shed further light on employment edented in its scale, but it also involved different impacts across different population groups. For combinations of instruments, with potentially example, it is estimated that the crisis cost China different implications for jobs. Fiscal stimulus between 20 million and 36 million jobs, particu- across the world amounted to US$5.5 trillion larly among migrant workers in export-oriented in purchasing power parity, with China, Japan, sectors.58 In Mexico, it caused a decline of half a and the United States accounting for more than ­ million jobs between 2008 and the second quar- 70 percent of it.64 Of 77 countries for which ter of 2009, particularly among women, youth, data are available, 80 percent used fiscal ex- and older workers, as well as a 10 percent drop pansion. Higher-income countries favored tax in real wages.59 In Indonesia, although the ef- cuts, higher unemployment benefits, and di- fects of the crisis were mild, young, casual, and rect support for enterprises. Low- and middle- informal workers were affected.60 Across coun- income countries boosted expenditures, includ- tries, the young bore the largest brunt.61 ing spending on training and income-support Adjustment patterns, in terms of jobs lost or measures.65 Across countries, responses mainly earnings declines, also varied in developing aimed at preventing or mitigating employment 62   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 BOX 1.4  Responses to the crisis went beyond income support for the unemployed  Countries worldwide used an array of policy responses to confront initiatives reached 3.3 percent of the employed in Italy, 3.2 percent the jobs crisis. Macroeconomic stimulus and targeted sector policies in Germany, and 2.7 percent in Japan. were supplemented by policies to strengthen income support for Low- and middle-income countries spent 67 percent of their those affected; measures to boost labor demand through wage resources on direct job creation measures and public works pro- subsidies, credit policies, and public works programs; and invest- grams. Mexico, for example, extended its temporary works program ments in skills and tailor-made employment services for those most to around 250,000 workers, or 0.5 percent of the labor force. Wage affected. subsides were popular in Europe. They were most often imple- Globally, unemployment insurance played a minor role. Only mented through a reduction in social security contributions and 15.4 percent of the unemployed received benefits during the crisis, were targeted to small and medium enterprises or to disadvantaged because of the low effective coverage of unemployment insurance groups such as long-term job seekers and the young. To reach an programs. In 23 countries in the Organisation for Economic Co- effective scale in a timely manner, though, countries needed to have operation and Development, Central Europe, and Latin America, the had such programs in place before the crisis. duration of benefits was extended. Countries with large informal In comparison, efforts to ramp up training, employment ser- sectors or without unemployment schemes relied on a mix of cash vices, or specific support programs for vulnerable groups were rela- transfers and public works schemes to provide additional income tively modest in scope and scale. Building or maintaining skills took support for those in need. Colombia’s Familias en Acción—a pro- the lion’s share of the budget in this category, with Chile, Italy, Mex- gram focusing on strengthening nutrition and education for chil- ico, Turkey, and the United States spending the most. dren—increased its coverage from 1.8 million to 2.7 million families. Overall, countries with more mature social protection systems as Argentina expanded the coverage of family benefits to all informal well as established employment programs were able to respond sector workers. quickly and reduce the impact of the crisis on jobs. With few excep- Countries across all income levels took measures to boost labor tions, though, little is known about possible unintended effects of demand. Of the resources devoted to creating and protecting jobs, their programs such as whether preserving some jobs came at the high-income countries spent more than half (56 percent) on credit expense of destroying others. Careful impact analysis is only in its policies for firms. They also implemented work-sharing arrange- initial stages. ments to cushion the impact of the crisis. Take-up rates for these Sources: ILO and World Bank 2012. losses, rather than trying to offset earnings tion of the labor force.66 Youth unemployment shortfalls (box 1.4). and idleness may be high, and women may face limited job o ­ pportunities, leaving potential eco- nomic and social gains untapped.67 A repeating * * * pattern of small gains in living standards, slow productivity growth, and eroding social cohe- Demography, urbanization, globalization, tech- sion may emerge. In contrast, countries that suc- nological progress, and macroeconomic crises cessfully address these job challenges can develop bring about formidable jobs challenges. Coun- ­ virtuous circles. The results—prosperous popu- tries that fail to address them may fall into vi- lations, a growing middle class, increased produc- cious circles of slow growth in labor earnings and tivity, and improved opportunities for women job-related dis­satisfaction affecting a sizable por- and youth—may then be self-reinforcing. QUESTION 1 What is a job? The world of work is diverse, especially in devel- on what a job is and what it means almost inevi- oping countries, and it is changing rapidly. tably influence views on what policies for jobs Against this backdrop, the diversity of the words should look like. Those for whom the word job used to describe what people do to earn a living is associated with the image of a worker in an across countries and cultures should not be sur- office or a factory, with an e­ mployer and a pay- prising. Even people who speak the same lan- check, may focus on a supportive investment guage can have vastly different interpretations climate for firms. Those for whom the word also of the meaning of a job. For some, the word encompasses farming, street vending, waste conjures up an image of a worker in an office picking, and domestic employment may think or a factory, with an employer and a regular of jobs policies as including land reform, agri- paycheck. Others may think of farmers, self- cultural extension, urban policies, or the provi- employed vendors in cities, and caregivers of sion of voice to the most vulnerable workers. children or elderly relatives. According to the International Conference of Labour Statisticians (ICLS), which sets standard definitions for official use across countries, a job Work matters and words matter is “a set of tasks and duties performed, or meant The varied interpretations capture the different to be performed, by one person, including for an aspects of jobs that people value. A woman in employer or in self-employment.�71 Under this Hanoi, Vietnam, explained, “an old woman who definition, a job is not the same as employment. just sells vegetables can gain respect from others The existence of job vacancies and people with and people listen to her.�68 A man who had more than one job means that the number of worked his way out of poverty in Satgailijhara, jobs is greater than the number of people em- Bangladesh, linked the value of his job as a rice ployed. The existence of unemployment means farmer to being able to invest in his children, “I that people do not find the jobs they want. Jobs have been able to get my children educated. refer to tasks, while the wage employed, farmers, That’s the best achievement in life.�69 and the self-employed refer to the people who Ela Bhatt, a lawyer and the founder of Self do them. Employed Women’s Association (SEWA) in In- dia, described her struggle with the language of Some gray areas work, given the multiplicity of tasks that people do every day and over time: “A small farmer The ICLS definition excludes some forms of works on her own farm. In tough times, she also work from official employment statistics. The works on other farms as a laborer. When the ag- employed are part of the economically active riculture season is over, she goes to the forest to population, defined as people who contribute to collect gum and other forest produce. Year “the production of goods and services as de- round, she produces embroidered items either at fined by the United Nations systems of national a piece rate for a contractor or for sale to a trader accounts and balances.�72 The system of national who comes to her village to buy goods. Now, accounts (SNA) includes “all production actu- how should her trade be categorized? Does she ally destined for the market, whether for sale or belong to the agricultural sector, the factory sec- barter,� as well as the production of goods for tor, or the home-based work sector? Should she one’s own use, but “excludes all production of be categorized as a farmer or a farm worker? Is services for own final consumption within she self-employed or is she a piece-rate worker?�70 households.�73 This definition thus leaves out of These questions are not merely semantic. official statistics activities such as child-rearing, The words and categories that are used to de- care of the elderly, or home cooking, as well as scribe work have tangible implications. Views traveling to work. 64   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Some countries are starting to develop esti- Informality is another important gray mates of these other types of activities. The area. After nearly four decades of debates SNA offers guidelines to countries for produc- about the concept of informality, there is still ing satellite accounts reflecting forms of house- no consensus on what is meant by informal hold work that are not considered jobs. Ac- jobs. Some schools of thought link informality cording to these guidelines, “a job can refer to characteristics of firms—whether the busi- to unpaid household service and volunteer ness is registered or pays taxes. Others focus work performed by one person for a household on characteristics of workers—whether they outside the SNA production boundary but are covered by social protection or have an within the general production boundary.�74 employment contract. And yet others stress Mexico, for instance, has used this guidance to modes of production and levels of productiv- estimate the value of unremunerated activities ity to define informality. A consensus is start- in households, such as housekeeping and child ing to emerge on how to measure informality, care. These activities represented nearly a quar- but the definitions used still leave out types of ter of Mexican GDP between 2003 and 2009 work that some consider informal. Meanwhile, and were equivalent to about two-thirds of relatively few countries produce regular statis- worker wages and benefits.75 tics on informality (box 1.5). BOX 1.5  Few countries produce statistics on informality  The concept of dual economies has an old lineage.a Based on the the Delhi Group has been working with the ILO on a forthcoming recognition of dual economies and the Harris and Todaro two-sector “Manual of Surveys of Informal Employment and Informal Sector.� In model, the term informality was first coined by a U.K. anthropologist 2008, the newest version of the United Nation’s system of national in a paper about Ghana. The concept gained popularity with a accounts adopted most of the previous resolutions and recommen- widely cited report from a mission of the International Labour Orga- dations on the measurement of informality. That resulted in a broad nization (ILO) to Kenya.b Since then, studies on informality have pro- definition that includes both the informal sector and informal liferated, and the concept has become standard in development employment. However, gray areas remain in relation to activities studies, labor economics, and other disciplines. Today the causes such as farming, independent professionals, and activities among and implications of informality are the subject of an intense aca- rural workers in general. demic debate and an extended research agenda aimed at under- Few countries produce regular official statistics on informality. standing how labor markets function in developing countries.c The dearth of data is apparent in global repositories such as the A variety of approaches can be used to measure informality, ILO’s Key Indicators of the Labor Market database, which, in its most informal employment, and the informal sector. Measurement be-­ recent version, includes data on informality for only 60 countries.d A comes even more complex when informality is combined with report on informality across the world by the ILO and Women in other concepts like illegal and underground activities or nonstan- Informal Employment: Globalizing and Organizing (WIEGO) makes dard work. The lack of systematic country-level data on informality systematic use of the most thorough definition of informality, cover- has led researchers to construct their own estimates using similar ing informal employment and employment in informal firms, to but not identical criteria, which sometimes lead to diverging present data on 47 countries.e The ILO-WIEGO report shows infor- conclusions. mality rates ranging from 40 percent in Uruguay and 42 percent in Recommendations on the measurement of informality were first Thailand to 75 percent in Bolivia and 80 percent in India. drafted in 1993 by the 15th International Conference of Labour The extent of informality varies with differences in productivity ­ Statisticians (ICLS). Four years later, the United Nations created the across firms and workers, as well as with differences in the nature of Delhi Group to document and recommend methods for defining regulations and the degree to which they are enforced. Whether and collecting data on the informal sector. In 2003, the 17th ICLS, informality is the result of exit, exclusion, uneven enforcement, or through the document “Guidelines Concerning a Statistical Defini- low firm productivity is still a matter of debate. Better measurement tion of Informal Employment,� introduced a definition of informal can provide information on the magnitude of informality and pro- employment and a series of rules for its measurement. Since 2006, vide more data for the advancement of studies in this area. Source: World Development Report 2013 team. a. Boeke 1942; Lewis 1954. b. Harris and Todaro 1970; Hart 1973; ILO 1972. c. A few among the most recent are Guha-Khasnobis and Kanbur 2006; Perry and others 2007; Kanbur 2009; Blades, Ferreira, and Lugo 2011; and Vanek and others 2012. d. ILO Key Indicators of the Labour Market (database). e. Vanek and others 2012. The jobs challenge    65 What is not a job? are discriminated against, such as migrants, women, and indigenous people. Migrant work- While views about what a job means vary, a ers in sweatshops in Latin America, child sol- broad consensus exists on the types of activities diers in conflict-afflicted areas of Africa, people that should never be considered a job. Interna- trafficked for sexual exploitation in Europe and tional norms define basic human rights as the Asia, and temporary migrant workers under boundaries of what is unacceptable. The United sponsorship contracts in the Middle East are Nations Universal Declaration of Human Rights, the subjects of the most conspicuous forms of which the UN General Assembly embraced forced labor and trafficking in the world. unanimously in 1948, provides for the right to Child labor provides another striking exam- work and protection from discrimination.76 The ple (box 1.6). Although it is covered by the core Declaration on Fundamental Rights and Princi- labor standards and many countries have rati- ples at Work adopted at the ILO conference in fied the relevant ILO conventions and the UN 1998 further specifies core labor standards that Convention on the Rights of the Child, the ILO call for an end to forced and compulsory labor, estimates that 115 million children worldwide child labor, and labor discrimination, and that were involved in hazardous work in 2008. Mea- provide for freedom of association and collec- surement is complicated by legal and moral tive bargaining.77 Most countries have ratified concerns, as well as by the flawed design of sur- the conventions on forced labor, child labor, and veys that may induce parents to misreport chil- discrimination; fewer have ratified the conven- dren’s work.80 These statistics may actually rep- tions on freedom of association and collective resent only a lower boundary on the size of the bargaining (figure 1.12).78 problem. While international agreements help define Recent research shows that children work for what forms of work are unacceptable, in prac- diverse and complex reasons.81 They range from tice many people work in activities that violate household poverty to the relative accessibility their rights. Some 21 million people are esti- and affordability of schooling and from the mated to be subject to forced labor, and around preferences of families and even children re- 1 million to trafficking.79 In many cases, forced garding work and play to the influence of tech- labor is inflicted upon minorities or groups that nological change, international trade, and ur- A majority of countries have ratified the core labor standards F I G U R E 1.12   180 175 170 number of rati cations, out of 185 countries 165 160 155 150 145 140 135 Con. 87 Con. 98 Con. 29 Con. 105 Con. 100 Con. 111 Con. 138 Con. 182 freedom of association forced labor discrimination child labor Source: ILO 2012c. Note: Con. = Convention. 66   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 may affect schooling, health, fertility, and behav- ior, although establishing these links has been BOX 1.6   Not all child work is child labor challenging. Rarely is there a one-to-one trad- According to the International Labour Organization (ILO), children engaged in eoff between school and work. In many places, child labor include “all persons aged between 5 and 17 years who during a the majority of children who work are also at- specified time period were engaged in one or more of the following activities: tending school. Moreover, a child may have (a) hazardous work; (b) worst forms of child labor other than hazardous work; dropped out of school for reasons unrelated to and (c) other forms of child labor (depending on age of the child and weekly child labor. The participation of children 12 hours worked).�a The “worst forms of child labor� include any work that jeopar- dizes the health, safety, or morals of a child. Such work is determined to be haz- years and older in family farming and small ardous depending on its specific nature, the demands on children in particular household enterprises can in some cases con- industries, and the general working conditions. Apart from hazardous work, tribute to the acquisition of skills.83 the worst forms of child labor include all forms of slavery, bondage, military In sum, jobs are activities that generate actual conscription, trafficking, and using, procuring, or offering children for prostitu- income, monetary or in kind, and do not violate tion, pornography, or other illicit activities. fundamental rights and principles at work. This definition of child labor is governed by two ILO Conventions (132 and This definition includes the categories of work 182), two ILO recommendations (146 and 190), and the UN Convention on the Rights of the Child. However, international standards also provide countries covered by ICLS guidelines: wage or salary em- with some latitude in setting allowed boundaries for the involvement of chil- ployment, employers, members of cooperatives, dren in productive activities (regarding ages or the definition of hazardous family workers (including unpaid family mem- work, for example). The standards also permit limited work in the performance bers), and the self-employed. In many instances, of household chores or in light productive activities that are not considered however, these categories may fail to uniquely or harmful. clearly classify certain individuals. For instance, small farmers are sometimes wage employed or Source: World Development Report 2013 team. self-employed, but they may also be employers a. ILO 2008a. if they hire unpaid family workers. Jobs include labor activities that generate income for the household, even if no income measure can be at- banization. Children in wealthier households tributed to a person’s specific labor, as in the case may, in some settings, be engaged in child labor of household enterprises and farming. if household assets and access to finance, land, By this definition, jobs are much broader or other resources generate more demand for than just working in an office or a factory, with work from household members.82 Child labor an employer and a regular paycheck. The jobs challenge    67 © Sebastião Salgado/Amazonas—Press Images Garimpeiros (independent prospectors) at the Serra Pelada gold mine, in Brazil 68   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Notes 1970; Hart 1973; ILO 1972. For a recent study of the process of structural change in Africa, see   1. World Development Report 2013 team estimates. Losch, Freguin-Gresh, and White (2012).   2. World Development Report 2013 team estimates 23. Freeman and Schettkat 2005; Ngai and Pissarides based on data from the International Labour 2008. Organization, http://laborsta.ilo.org/applv8/data/ 24. Bardasi and Wodon 2006; Lee, McCann, and EAPEP/eapep_E.html, and World Development Messenger 2007; Maddison 2001; Ramey and Indicators, http:/data-worldbank.org/data/cata Francis 2009. log/world-development-indicators. 25. Bardasi and Wodon 2010; Gammage 2010.   3. Reid 1934, 11. 26. Charmes 2006; Gálvez-Muñoz, Rodríguez-  4. Organisation for Economic Co-operation and Modroño, and Domínguez-Serrano 2011; Hir­ Development Employment database, http:// way and Jose 2012. Burda, Hamermesh, and Weil www.oecd.org/employment/database. Tempo- (2011) argue that differences in total work by rary work refers to a mixture of seasonal jobs, gender change over the business cycle but con- fixed-term contracts, on-call workers, and verge over the long term. temporary help agency workers that varies by 27. ILO 2009a; Pagés 2010. country, depending on national definitions and 28. Autor and Dorn 2011; Gratton 2011; Holzer and available statistics. Lerman 2009.   5. Lee, McCann, and Messenger 2007. 29. Feenstra 2010.   6. ILO Department of Statistics, “Short term indi- 30. Brown, Ashton, and Lauder 2010. For U.S. parent cators of the labour market,� http://laborsta.ilo companies, according to the National Science .org/sti/sti_E.html. Foundation, the share of research performed by   7. Ghose, Majid, and Ernst 2008. Asia-located affiliates outside of Japan rose from   8. Gindling and Newhouse 2012 for the World De- 5 to 14 percent from 1997 to 2008, mainly in velopment Report 2013. China, Korea, Singapore, and India (http://nsf   9. World Bank (2011c) offers a more in-depth re- .gov/statistics/seindl12/c4/c4s4.htm). view of these issues. 31. Goswami, Mattoo, and Sáez 2011. 10. ILO Department of Statistics, “Short term indi- 32. Goswami, Mattoo, and Sáez 2011. cators of the labour market,� http://laborsta.ilo 33. A.T. Kearney 2011. .org/sti/sti_E.html. 34. United Nations Educational, Scientific and Cul- 11. Youth not in employment, education, or training tural Organization Institute of Statistics, http:// are also sometimes referred to as “NEETs.� See the stats.uis.unesco.org/unesco/TableViewer/ statistical annex for more information. Kovrova, tableView.aspx?ReportId=175. Lyon, and Rosati 2012 for the World Develop- 35. WDR team estimates of the top 20 percent of ment Report 2013; Ranzani and Rosati 2012 for PISA ratings among 12 countries in the 2009 the World Development Report 2013. Programme for International Student Assess- 12. Lyon, Rosati, and Guarcello 2012 for the World ment (PISA) of 15-year-olds. Development Report 2013. 36. This skill definition broadly follows Autor, Levy, 13. World Bank 2006. and Murnane (2003) and Acemoglu and Autor 14. ILO 2010. (2011). Other approaches distinguish, for exam- 15. United Nations Population Division, World Pop- ple, between cognitive, noncognitive, and techni- ulation Prospects 2011; ILO, Labor Force Partici- cal skills (World Bank 2010) or cognitive/problem pation Estimates and Projections. solving, learning, personal/behavioral/ethical, and 16. Rozelle and Huang 2012 and estimates from In- social and communication skills (ILO 2008b). ternational Labour Office database on labor sta- 37. Aedo and others 2012 for the World Develop- tistics, Laborsta, http://laborsta.ilo.org/, October ment Report 2013. 2011. 38. Brown, Ashton, and Lauder 2010. 17. IOM 2008; Lucas 2005; Özden and others 2011. 39. A.T. Kearney 2011. 18. IOM 2010. 40. Yoshino 2011. 19. Chenery and Syrquin 1975; Clark 1940; Kuznets 41. Examples include oDesk (https://www.odesk 1966; Rostow 1960. .com/), babajob (http://www.babajob.com/), 20. UN 2011b. Google trader (for example, http://www.google 21. Herrmann and Khan 2008. .co.ug/africa/trader/search?cat=jobs), and Souk- 22. This is the main observation that sparked the re- Tel (http://www.souktel.org/). search on the informal sector: Harris and Todaro 42. McKinsey Global Institute 2011. The jobs challenge    69 43. Blanchflower, Oswald, and Stutzer 2001. More 70. Bhatt 2006, 17. recent (2005) data of the International Social 71. Article 2, ILO 2007. Also see UN (2009). Survey Programme show similar patterns. 72. Article 9, ILO 1982, http://www.ilo.org/wcmsp5/ 44. WDR team estimates from the 2010 Life in Tran- groups/public/---dgreports/---stat/documents/ sitions Survey. normativeinstrument/wcms_087481.pdf. 45. Banerjee and Duflo 2011; Perry and others 2007. 73. UN 2009, 6–7. 46. Banerjee and Duflo 2011. Although, it is not al- 74. ILO 2009b, 42. ways the case that the self-employed always report 75. INEGI 2011. lower well-being. In their study from Ghana, Falco 76. UN 1948, http://www.un.org/en/documents/ and others 2012 for the WDR 2013 find that infor- udhr/, article 23 (1). mal firm owners who employ others are on aver- 77. ILO 1998. age substantially happier than formal workers. 78. The core international labor standards are the 47. Based on an update by the authors Chen and subject of eight conventions covering the four Ravallion 2010. areas: Convention 87 (1948), the Freedom of As- ­ 48. World Bank 2011b. sociation and Protection of the Right to Organize 49. Guscina 2006; Lübker 2007; Rodriguez and Convention; Convention 98 (1949), the Right to Jayadev 2010. Organize and Collective Bargaining Convention; 50. Bentolila and Saint-Paul 2003; Freeman 2008. Convention 29 (1930), the Forced Labour Con- 51. International Labour Office database on labor vention; Convention 105 (1957), the Abolition statistics, Laborsta, http://laborsta.ilo.org/. of Forced Labour Convention; Convention 100 52. The definition of “private sector� in China is (1951), the Equal Remuneration Convention; broad and sometimes not clearly defined in offi- Convention 111 (1958), the Discrimination (Em- cial statistics. There is differentiation between ployment and Occupation) Convention; Con- what are labeled “private firms� (a profit-making vention 138 (1973), the Minimum Age Conven- unit invested in and established by natural per- tion; and Convention 182 (1999), the Worst sons or controlled by persons hiring more than Forms of Child Labour Convention. See “Con- seven workers) and “individual firms� (those ventions,� NORMLEX Database: Information on with fewer than eight employees). Foreign- International Labour Standards, International invested firms and collectives are not part of the Labour Organization, Geneva. ILO (2012). http:// private sector in official statistics. For more de- www.ilo.org/dyn/normlex/en/. tails, see Kanamori and Zhao (2004). 79. Andrees and Belser 2009; ILO 2009c; ILO 2012b. 53. Nabli, Silva-Jaurengui, and Faruk Aysan 2008. 80. Dillon and others 2012. 54. Assaad 2012; Assaad and Barsoum 2007. 81. Basu 1999; Basu and Tzannatos 2003; Cigno and 55. Mryyan 2012. Rosati 2005; Edmonds 2008. 56. For a study of previous crises, see Fallon and 82. Del Carpio and Loayza 2012; Hazarika and Lucas (2002). For the food price crisis, see Ivanic Sarangi 2008. and Martin (2008). 83. Edmonds (2008) offers a review of the theoretical 57. ILO 2012a. and empirical evidence on child labor. 58. Giles and others 2012 for the World Develop- ment Report 2013. These losses were temporary and currently worker shortages are experienced instead. References 59. Freije-Rodriguez, Lopez-Acevedo, and Rodriguez- Oreggia 2011. The word processed describes informally reproduced 60. McCulloch, Grover, and Suryahadi 2011. works that may not be commonly available through 61. Cho and Newhouse 2010. libraries. 62. Khanna, Newhouse, and Paci 2010. 63. Khanna, Newhouse, and Paci 2010; World Bank A.T. Kearney. 2011. Offshoring Opportunities amid 2012. Economic Turbulence: A.T. Kearney Global Services 64. ILO and World Bank 2012. Location Index, 2011. Chicago: A.T. Kearney 65. ILO and World Bank 2012; Robalino, New- Global Services Location Index. house, and Rother forth­ coming. Acemoglu, Daron, and David Autor. 2011. “Skills, 66. Bell and Blanchflower 2011; Farber 2011. Tasks and Technologies: Implications for Employ- 67. World Bank 2011c. ment and Earnings.� In Handbook of Labor Eco- 68. World Bank 2011a. nomics Volume 4, ed. Orley Ashenfelter and David 69. Narayan, Pritchett, and Kapoor 2009, 19. E. Card. Amsterdam: Elsevier. 70   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Aedo, Cristian, Jesko Hentschel, Javier Luque, and Blades, Derek, Francisco H. G. Ferreira, and Maria Martin Moreno. 2012. “Skills Around the World: Ana Lugo. 2011. “The Informal Economy in De- Structure and Recent Dynamics.� Background veloping Countries: An Introduction.� Review of paper for the WDR 2013. Income and Wealth 57 (Special Issue): S1–S7. Andrees, Beate, and Patrick Belser. 2009. Forced La- Blanchflower, David G., Andrew J. Oswald, and Alois bour: Coercion and Exploitation in the Private Econ- Stutzer. 2001. “Latent Entrepreneurship across omy. Geneva: International Labour Organization. Nations.� European Economic Review 45 (4–6): Assaad, Ragui. 2012. “The MENA Paradox: Higher 680–91. Education but Lower Job Quality.� In Moving Boeke, Julius H. 1942. Economies and Economic Policy Jobs to the Center Stage. Berlin: BMZ (Bundesmin- in Dual Societies. Haarlem: Tjeenk Willnik. isterium für Wirstchaftliche Zussamenarbeit), Brown, Philip, David Ashton, and Hugh Lauder. 2010. Berlin Workshop Series. Skills Are Not Enough: The Globalization of Knowl- Assaad, Ragui, and Ghada Barsoum. 2007. “Youth Ex- edge and the Future of the UK Economy. Wath clusion in Egypt: In Search of ‘Second Chances.’ � upon Dearne, U.K.: United Kingdom Commis- Middle East Youth Initiative Working Paper Series sion for Employment and Skills. 2, Wolfensohn Center for Development, Dubai Burda, Michael C., Daniel S. Hamermesh, and School of Government, Dubai. Philippe Weil. 2011. “Total Work, Gender and So- Autor, David H., and David Dorn. 2011. “The Growth cial Norms.� Working Paper Series 13000, Na- of Low-Skill Service Jobs and the Polarization of tional Bureau of Economic Research, Cambridge, the U.S. Labor Market.� Massachusetts Institute of MA. Technology, Cambridge, MA. Processed. Charmes, Jacques. 2006. “A Review of Empirical Evi- Autor, David H., Frank Levy, and Richard J. Murnane. dence on Time Use in Africa from UN-Sponsored 2003. “The Skill Content of Recent Technological Surveys.� In Gender, Time Use, and Poverty in Sub- Change: An Empirical Exploration.� Quarterly Saharan Africa, ed. C. Mark Blackden and Quen- Journal of Economics 118 (4): 1279–333. tin Wodon, 39–72. Washington, DC: World Bank. Bajaj, Vikas. 2011. “Outsourcing Giant Finds It Must Chen, Shaohua, and Martin Ravallion. 2010. “The Be Client, Too.� New York Times, November 30. Developing World Is Poorer Than We Thought, Banerjee, Abhijit V., and Esther Duflo. 2011. Poor Eco- but No Less Successful in the Fight against Pov- nomics: A Radical Rethinking of the Way to Fight erty.� Quarterly Journal of Economics 125 (4): Global Poverty. New York: Public Affairs. 1577–625. Bardasi, Elena, and Quentin Wodon. 2006. “Poverty Chenery, Hollis Burnley, and Moises Syrquin. 1975. Reduction from Full Employment: A Time Use Approach.� Munich Personal RePEc Archive Paper Patterns of Development, 1957–1970. London: Ox- 11084, Munich. ford University Press. ———. 2010. “Working Long Hours and Having No Cho, Yoonyoung, and David Newhouse. 2010. “How Choice: Time Poverty in Guinea.� Feminist Eco- Did the Great Recession Affect Different Types nomics 16 (3): 45–78. of Workers? Evidence from 17 Middle-Income Basu, Kaushik. 1999. “Child Labor: Cause, Conse- Countries.� Discussion Paper Series 5681, Insti- quence, and Cure, with Remarks on International tute for the Study of Labor, Bonn. Labor Standards.� Journal of Economic Literature Cigno, Alessandro, and Furio C. Rosati. 2005. The 37 (3): 1083–119. Economics of Child Labour. Oxford: Oxford Uni- Basu, Kaushik, and Zafiris Tzannatos. 2003. “Child versity Press. Labor and Development: An Introduction.� World Clark, Colin. 1940. The Conditions of Economic Prog- Bank Economic Review 17 (2): 145–6. ress. London: Macmillan & Company. Bell, David N. F., and David G. Blanchflower. 2011. Del Carpio, Ximena, and Norman Loayza. 2012. “The “The Crisis, Policy Reactions and Attitudes to Impact of Wealth on the Amount and Quality Globalization and Jobs.� Discussion Paper Series of Child Labor.� Policy Research Working Paper 5680, Institute for the Study of Labor, Bonn. Series 5959, World Bank, Washington, DC. Bentolila, Samuel, and Gilles Saint-Paul. 2003. “Ex- Dillon, Andrew, Elena Bardasi, Kathleen Beegle, and plaining Movement in the Labor Share.� Contri- Pieter Serneels. 2012. “What Explains Variation in butions to Macroeconomics 3 (1). Child Labor Statistics? Evidence from a Survey Bhatt, Ela. 2006. We Are Poor but So Many: The Story Experiment in Tanzania.� Journal of Development of Self-Employed Women in India. New York: Ox- Economics 98 (1): 136–47. ford University Press. Dourgarian, Gregg. 2011. “Five Staffing Industry Bhorat, Haroon. 2012. “Temporary Employment Ser- Trends for 2011.� Staffing Talk, January 3. vices in South Africa.� Background paper for the Edmonds, Eric. 2008. “Child Labor.� In Handbook of WDR 2013. Development Economics, Vol. 4, ed. T. Paul Shultz The jobs challenge    71 and John Strauss, 3607–709. Oxford: North Hol- Guha-Khasnobis, Basudeb, and Ravi Kanbur, eds. land Elsevier. 2006. Informal Labour Markets and Development. Falco, Paolo, William Maloney, Bob Rijkers, and Mau- New York: Palgrave Macmillan. ricio Sarrias. 2012. “Subjective Well-Being, Infor- Guscina, Anastasia. 2006. Effects of Globalization on mality, and Preference Heterogeneity in Africa.� Labor’s Share in National Income. Washington, Background paper for the WDR 2013. DC: International Monetary Fund. Fallon, Peter, and Robert Lucas. 2002. “The Impact of Harris, John R., and Michael P. Todaro. 1970. “Migra- Financial Crises on Labor Markets, Household In- tion, Unemployment and Development: A Two- comes and Poverty: A Review of Evidence.� World Sector Analysis.� American Economic Review 60 Bank Research Observer 17 (1): 21–45. (1): 126–42. Farber, Henry S. 2011. “Job Loss in the Great Reces- Hart, Keith. 1973. “Informal Income Opportunities sion: Historical Perspective from the Displaced and Urban Employment in Ghana.� Journal of Workers Survey, 1984–2010.� Discussion Paper Modern African Studies 11 (1): 61–89. Series 5696, Institute for the Study of Labor, Bonn. Hazarika, Gautam, and Sudipta Sarangi. 2008. Feenstra, Robert C. 2010. Offshoring in the Global “Household Access to Microcredit and Child Economy: Microeconomic Structure and Macroeco- Work in Rural Malawi.� World Development 36 nomic Implications. Cambridge, MA: MIT Press. (5): 843–59. Freeman, Richard B. 2008. “The New Global Labor Herrmann, Michael, and Haider Khan. 2008. “Rapid Market.� Focus 26 (1): 1–6. Urbanization, Employment Crisis and Poverty Freeman, Richard, Remco H. Oostendorp, and Davin in African LDCs: A New Development Strategy Chor. 2011. “The Standardized ILO October In- and Aid Policy.� Munich Personal RePEc Archive quiry 1953–2008.� National Bureau of Economic Paper 9499, Munich. Research, Cambridge, MA. Hirway, Indira, and Sunny Jose. 2012. “Understand- Freeman, Richard, and Ronald Schettkat. 2005. ing Women’s Work Using Time-Use Statistics: The “Marketization of Household Production and the Case of India.� Feminist Economics 17 (4): 67–92. EU-US Gap in Work.� Economic Policy 20 (41): Holzer, Harry, and Robert Lerman. 2009. The Future 5–50. of Middle-Skill Jobs. Washington, DC: Center on Freije-Rodriguez, Samuel, Gladys Lopez-Acevedo, Children and Families, Brookings Institute. and Eduardo Rodriguez-Oreggia. 2011. “Effects of ILO (International Labour Organization). 1972. Em- the 2008–09 Economic Crisis on Labor Markets in ployment, Incomes and Equality: A Strategy for In- Mexico.� Policy Research Working Paper Series creasing Productive Employment in Kenya. Geneva: 5840, World Bank, Washington, DC. ILO. Gálvez-Muñoz, Lina, Paula Rodríguez-Modroño, and ———. 1982. Resolution Concerning Statistics of the Mónica Domínguez-Serrano. 2011. “Work and Economically Active Population, Employment, Time Use By Gender: A New Clustering of Euro- Unemployment and Underemployment. Adopted pean Welfare Systems.� Feminist Economics 17 (4): by the Thirteenth International Conference of La- 125–57. bour Statisticians, ILO, Geneva, October 29. Gammage, Sarah. 2010. “Time Pressed and Time ———. 1998. Declaration on Fundamental Princi- Poor: Unpaid Household Work in Guatemala.� ples and Rights at Work. Adopted by the Interna- Feminist Economics 16 (3): 79–112. tional Labour Conference at its Eighty-sixth Ses- Ghose, Ajit K., Nomaan Majid, and Christoph Ernst. sion, ILO, Geneva, June 18. 2008. The Global Employment Challenge. Geneva: ———. 2007. Resolution Concerning Updating the International Labour Organization. International Standard Classification of Occupa- Giles, John, Albert Park, Fang Cai, and Yang Du. tions. Adopted by the Tripartite Meeting of Ex- 2012. “Weathering a Storm: Survey-Based Per- perts on Labour Statistics on Updating the Inter- spectives on Employment in China in the After- national Standard Classification of Occupations, math of the Global Financial Crisis.� Policy Re- ILO, Geneva, December 6. search Working Paper Series 5984, World Bank, ———. 2008a. Resolution Concerning Statistics of Washington, DC. Child Labour. Adopted by the Eighteenth Interna- Gindling, T. H., and David Newhouse. 2012. “Self- tional Conference of Labour Statisticians, ILO, Employment in the Developing World.� Back- December 5. ground paper for the WDR 2013. ———. 2008b. Skills for Improved Productivity, Em- Goswami, Arti Grover, Aaditya Mattoo, and Sebastián ployment Growth and Development. Geneva: ILO. Sáez, eds. 2011. Exporting Services: A Developing ———. 2009a. Protecting People, Promoting Jobs: A Country Perspective. Washington, DC: World Bank. Survey of Country Employment and Social Protec- Gratton, Lynda. 2011. The Shift: The Future of Work Is tion Policy Responses to the Global Economic Crisis. Already Here. London: HarperCollins. Geneva: ILO. 72   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 ———. 2009b. Report of the Conference, 18th Interna- Kuznets, Simon. 1966. Modern Economic Growth. tional Conference of Labour Statisticians. Geneva: New Haven, CT: Yale University Press. ILO. Lee, Sangheon, Deidre McCann, and Jon C. Messen- ———. 2009c. The Cost of Coercion. Geneva: ILO. ger. 2007. Working Time around the World. New ———. 2010. Accelerating Action against Child La- York: Routledge; Geneva: International Labour bour. Geneva: ILO. Organization. ———. 2011. Private Employment Agencies, Promo- Lewis, W. Arthur. 1954. “Economic Development tion of Decent Work and Improving the Functioning with Unlimited Supplies of Labor.� The Manches- of Labour Markets in Private Services Sectors. ter School 22 (2): 139–91. Geneva: ILO. Losch, Bruno, Sandrine Freguin-Gresh, and Eric ———. 2012a. Global Employment Trends 2012: Pre- Thomas White. 2012. Structural Transformation venting a Deeper Jobs Crisis. Geneva: ILO. and Rural Change Revisited: Challenges for Late ———. 2012b. ILO Global Estimate of Forced Labour Developing Countries in a Globalizing World. 2012: Results and Methodology. Geneva: ILO. Washington, DC: World Bank. ———. 2012c. NORMLEX Database: Information Lübker, Malte. 2007. “Labour Shares.� Technical Brief System on International Labour Standards. ILO, 1, International Labour Organization, Geneva. Geneva. http://www.ilo.org/dyn/normlex/en/f?p= Lucas, Robert E. B. 2005. International Migration and NORMLEXPUB:1:0. Economic Development: Lessons from Low-Income ILO and World Bank. 2012. Inventory of Policy Re- Countries. Cheltenham, U.K.: Edward Elgar sponses to the Financial and Economic Crisis: Joint Publishing. Synthesis Report. Washington, DC: ILO and World Lyon, Scott, Furio C. Rosati, and Lorenzo Guarcello. Bank. 2012. “At the Margins: Young People neither in INEGI (Instituto Nacional de Estadística y Geografía Education nor in Employment.� Background pa- per for the WDR 2013. de México). 2011. Sistema de Cuentas Nacionales Maddison, Angus. 2001. The World Economy: A Mil- de México. Cuenta Satélite del Trabajo no Remu- lennial Perspective. Paris: Organisation for Eco- nerado de los Hogares de México, 2003–2009. Mex- nomic Co-operation and Development. ico, DF: INEGI. McCulloch, Neil, Amit Grover, and Asep Suryahadi. IOM (International Organization for Migration). 2011. The Labor Market Impact of the 2009 Finan- 2008. World Migration Report 2008: Managing cial Crisis in Indonesia. Sussex, U.K.: Institute of ­ Labor Mobility in the Evolving Global Economy. Development Studies. Geneva: IOM. McKinsey Global Institute. 2011. An Economy That ———. 2010. World Migration Report 2010—The Fu- Works: Job Creation and America’s Future. Wash- ture of Migration: Building Capacities for Change. ington, DC: McKinsey Global Institute. Geneva: IOM. Montenegro, Claudio E., and Harry Anthony Patri- Ivanic, Maros, and Will Martin. 2008. “Implications nos. 2012. “Returns to Schooling around the of Higher Global Food Prices for Poverty in Low- World.� Background paper for the WDR 2013. Income Countries.� Agricultural Economics 39 (1): Mryyan, Nader. 2012. “Demographics, Labor Force 405–16. Participation, and Unemployment in Jordan.� Kanamori, Tokishi, and Zhijun Zhao. 2004. Private Working Paper Series 670, Economic Research Sector Development in the People’s Republic of Forum, Giza, Egypt. China. Manila: Asian Development Bank Institute. Musgrave, Amy. 2009. “Labor Broking Industry Likely Kanbur, Ravi. 2009. “Conceptualizing Informality: to Face Regulation.� Business Day, August 25. Regulation and Enforcement.� Indian Journal of Nabli, Mustapha K., Carlos Silva-Jáuregui, and ­Ahmet Labour Economics 52 (1): 33–42. Faruk Aysan. 2008. “Authoritarianism, Credibility Khanna, Gaurav, David Newhouse, and Pierella Paci. of Reforms, and Private Sector Development in 2010. “Fewer Jobs or Smaller Paychecks? Labor the Middle East and North Africa.� Working Paper Market Impacts of the Recent Crisis in Middle- Series 443, Economic Research Forum, Cairo. Income Countries.� Economic Premise April Narayan, Deepa, Lant Pritchett, and Soumya Kapoor. 2010, Number 11, Poverty Reduction and Equity 2009. Success from the Bottom Up. Vol. 2 of Moving Group and the Human Development Network Out of Poverty. New York: Palgrave Macmillan; Social Protection Division, World Bank, Washing- Washington, DC: World Bank. ton, DC. Ngai, Rachel L., and Christopher A. Pissarides. 2008. Kovrova, Irina, Scott Lyon, and Furio Camillo Rosati. “Trends in Hours and Economic Growth.� Review 2012. “NEET Youth Dynamics in Indonesia and of Economic Dynamics 11 (2): 239–56. Brazil: A Cohort Analysis.� Background paper for Oostendorp, Remco. 2012. “The Occupational Wages the WDR 2013. around the World (OWW) Database: Update for The jobs challenge    73 1983–2008.� Background paper for the WDR TeamLease. 2010. Temp Salary Primer 2010. Ahmed- 2013. abad, India: TeamLease Services Pvt. Ltd. Özden, Çaglar, Christopher Parsons, Maurice Schiff, UN (United Nations). 1948. Universal Declaration of and Terrie L. Walmsley. 2011. “Where on Earth Human Rights. Adopted by the UN General As- Is Everybody? The Evolution of Global Bilateral sembly, New York, December 10. Migration 1960–2000.� World Bank Economic Re- ———. 2009. System of National Accounts. New York: view 25 (1): 12–56. UN. Pagés, Carmen, ed. 2010. The Age of Productivity: ———. 2011b. World Urbanization Prospects: The Transforming Economies from the Bottom Up. New 2011 Revision. New York: UN, Department of York: Palgrave Macmillan. Economic and Social Affairs. Perry, Guillermo E., William F. Maloney, Omar S. Vanek, Joann, Martha Chen, Ralf Hussmanns, James Arias, Pablo Fajnzylber, Andrew D. Mason, and Heintz, and Françoise Carré. 2012. Women and Jaime Saavedra-Chanduvi. 2007. Informality: Exit Men in the Informal Economy: A Statistical Picture. and Exclusion. Washington, DC: World Bank. Geneva: Women in Informal Employment: Glo- Ranzani, Marco, and Furio Camillo Rosati. 2012. balizing and Organizing and International Labour “The NEET Trap: A Dynamic Analysis for Mex- Organization. ico.� Background paper for the WDR 2013. World Bank. 2006. World Development Report 2007: Ramey, Valerie A., and Neville Francis. 2009. “A Cen- Development and the Next Generation. Washing- tury of Work and Leisure.� American Economic ton, DC: World Bank. Journal: Macroeconomics 1 (2): 189–224. ———. 2010. Stepping Up Skills for More Jobs and Reid, Margaret. 1934. Economics of Household Produc- Higher Productivity. Washington, DC: World tion. New York: John Wiley and Sons. Bank. Robalino, David, David Newhouse, and Friederike ———. 2011a. Defining Gender in the 21st Century: Rother. Forthcoming. “Labor and Social Protec- Talking with Women and Men around the World, A tion Policies during the Crisis and Recovery.� In Multi-Country Qualitative Study of Gender and Labor Markets in Developing Countries during the Economic Choice. Washington, DC: World Bank. Great Recession: Impacts and Policy Responses, ed. ———. 2011b. More and Better Jobs in South Asia. Arup Banerji, David Newhouse, David Robalino, Washington, DC: World Bank. and Pierella Paci. Washington, DC: World Bank. ———. 2011c. World Development Report 2012: Gen- Rodriguez, Francisco, and Arjun Jayadev. 2010. “The der Equality and Development. Washington, DC: Declining Labor Share of Income.� Human Devel- World Bank. opment Research Paper 2010/36, United Nations ———. 2012. Job Trends. Washington, DC: World Development Programme, New York. Bank. Rostow, Walt Whitman. 1960. The Stages of Economic Yoshino, Yutaka, ed. 2011. Industrial Clusters and Growth: A Non-Communist Manifesto. Cambridge, Micro and Small Enterprises in Africa: From Sur- ­ U.K.: Cambridge University Press. vival to Growth. World Bank Directions in Devel- Rozelle, Scott, and Jikun Huang. 2012. “China’s Labor opment Series. Washington, DC: World Bank. Transition and the Future of China’s Rural Wages and Employment.� Background paper for the WDR 2013. PART 1 Jobs are transformational Introduction to Part 1 E conomic development is about improve- ments in living standards supported by productivity growth. It also involves so- DEVELOPMENT cial change associated with urbanization, inte- gration in the world economy, and the drive toward gender equality. All of these transforma- tions are related to jobs. The development pro- cess is about some jobs becoming better and LIVING PRODUCTIVITY SOCIAL STANDARDS COHESION others disappearing, about people taking jobs and changing jobs, and about jobs migrating to other places within and across countries. Development often entails the movement of labor from rural, agricultural, and mostly sub- sistence activities to urban, nonagricultural, and mostly market-oriented activities. This move- ment transforms the lives of families and com- JOBS munities, the organization of firms, and the norms and values of societies. It can boost pro- ductivity and improve living standards and also affect the cohesiveness of society. Jobs are thus a key driver of development. haviors, they can influence trust and civic engagement. • Living standards: Jobs provide earnings op- portunities to lift people out of poverty, raise Distinguishing these three transformations pro- their consumption, and contribute to indi- vides an understanding of how jobs contribute vidual well-being more broadly. to development. People’s well-being is the ulti- mate goal, and the transformation of living • Productivity: Through job creation and de- standards captures this link directly. But sus- struction within sectors and reallocations tained improvements in living standards are im- across sectors and countries, jobs are also at possible without productivity growth or when the root of economic growth. resources are wasted through confrontation. • Social cohesion: Jobs define who people are That is why it is necessary to look at the three in many ways; by shaping values and be­ transformations jointly. 75 CHAPTER 2 Jobs and living standards Jobs are the main source of income for the majority of households and a key driver of poverty reduction. But their contribution to well-being goes beyond the earnings they provide. J obs are the most important determinant of living standards around the world. For the vast majority of people, their work is the main source of income, especially in the poorest Jobs also influence how workers see them- selves and relate to others. Most people feel that jobs should be meaningful and contribute to so- ciety. Together with other objective job charac- countries. And jobs-related events are the most teristics, the self-esteem a job provides is an im- frequent reasons for families to escape or fall portant determination of satisfaction with life. into poverty. Furthermore, as earnings increase, individual choices expand—household mem- bers can opt to stay out of the labor force or to Jobs improve material well-being work fewer hours and dedicate more time to education, retirement, or family. Opportunities Over the course of a country’s development, for gainful work, including in farming and higher productivity and labor earnings allow self-employment, offer households the means to households to allocate more time to invest- increase consumption and reduce its variability. ment and consumption activities and less to Higher crop yields, access to small off-farm en- production. Thus, schooling and retirement terprise activities, the migration of family mem- gain importance relative to work. For the past bers to cities, and transitions to wage employ- century or so, the number of hours worked by ment are milestones on the path to prosperity. youth in industrial countries has declined In addition to their fundamental and imme- steadily as access to education has increased. diate contribution to earnings, jobs affect other Similarly, the number of years in retirement dimensions of well-being, positively and nega- has increased in parallel with longer life expec- tively. Not having a job undermines mental tancy.1 Higher earnings also facilitate longer health, especially in countries where wage em- periods of job seeking, especially among ployment is the norm and the lack of employ- younger household members, often leading to ment opportunities translates into open unem- higher unemployment rates. Among men and ployment rather than underemployment. But a women of prime age (25 to 54), total working job prone to occupational accidents or work- hours (market and nonmarket) have remained related diseases can damage physical health or relatively stable, with the main change being worse. More generally, monetary, nonmonetary, the growing share of market activities among and even subjective characteristics of jobs can all women (figure 2.1). These general trends are have an impact on well-being (box 2.1). not ironclad, however. Jobs and living standards    77 BOX 2.1  There are many dimensions of living standards and many ways to measure them  Debates on how to define and measure living standards go far back of standards of living and poverty reduction worldwide. Advances in social sciences. The work by Rowntree and Booth in late 19th cen- toward the first Millennium De­ velopment Goal (Eradicate extreme tury England is usually mentioned as seminal, especially in relation poverty) have been documented using global monetary poverty to the measurement of poverty. In the 1930s, the creation of the Sys- measures. The availability of richer datasets, in turn, has supported tem of National Accounts concentrated on measuring the total mar- the emergence of newer measures of living standards, many of ket value of the goods and services produced in an economy and them multidimensional in nature. These measures combine both made gross domestic product (GDP) per capita the main indicator monetary and nonmonetary indicators of well-being, as well of living standards in general. By the 1970s and 1980s, there was a as information on their distribution across different population growing agreement that important aspects of well-being, such as groups.b health status, or exposure to crime, pollution, and urban conges- Despite this progress, important controversies remain, particu- tion, were not fully accounted for in GDP. Research also showed that larly on which indicators are more appropriate for gauging each the distribution of material amenities affected individual well- dimension of well-being and on the weights that should be attrib- being. There is now consensus that living standards depend not uted to each. Some recent proposals even suggest a revamping of only on average incomes and consumption but also on access to statistical systems to formulate better measures of production that benefits as diverse as health and education, sanitation and housing, take into consideration changes in the quality of goods, govern- and security and freedom.a ment services, and time allocated to home activities and leisure. There are ongoing systematic efforts to collect individual, There are also proposals to include among measures of living stan- household, and community data to better understand and com- dards subjective indicators of well-being and indicators on the level pare living standards in developing and developed countries. Com- and sustainability of human, physical, and environmental assets.c plete poverty profiles for different groups of the population within Other proposals emphasize subjective indicators building on a phil- a country, based on the comparison of income or consumption osophical point of view.d Aggregating indicators and comparing aggregates to international or national poverty lines, have prolifer- them over time and across space becomes more intricate in this ated. Microdata collection efforts have allowed a close monitoring case, because of differences in values and beliefs. Source: World Development Report 2013 team. a. Adelman and Morris 1973; Chenery and Syrquin 1975; Nordhaus and Tobin 1973; Sen and Hawthorn 1987; Steckel 1995; Streeten 1979. b. Among these indicators are the Human Development Index (UNDP 1990), the Human Opportunity Index (Paes de Barros and others 2009), and a large variety of multi­ dimensional poverty indexes (Alkire and Foster 2011; Bourguignon and Chakravarty 2003; Kakwani and Silber 2008). See also OECD 2011. c. Fitoussi, Sen, and Stiglitz 2010. d. This is the case, for instance, of the measures of Gross National Happiness in Bhutan by the Center for Bhutan Studies. The nature of production, consumption, nied by higher market participation among and investment activities varies across coun- women.3 Developed and developing econ- tries as well. In some, low hours of work among omies allocate a similar share of the day to youth are associated more with idleness than work. But women allocate a larger share than with schooling; in others, schooling has pro- men to activities not directly generating income ceeded at an accelerated pace. Similarly, job (figure 2.2). characteristics change with development. In Jobs do not automatically guarantee sus- rural economies where agricultural activities tained improvements in earnings and well- predominate, the purpose of household pro- being. Working people often remain mired in duction is often direct consumption. Less de- poverty. In many countries, adults in poor veloped economies tend to be characterized households are more likely to be working than by more working time dedicated to jobs with- those in nonpoor households. The poor are not out wage payments, including farming and usually characterized by lack of jobs or hours of other types of self-employment. Development work; they often have more than one job and changes the organization of work from home work long hours, but their jobs are poorly remu- to market production.2 As economies develop, nerated (box 2.2). more work is remunerated through wages and In more affluent societies, a larger share of salaries. This reallocation is usually accompa- income is derived from capital, transfers (social 78   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Working F I G U R E 2 .1   hours vary across ages age group 24–54 years 70 54–65 years Costa Rica 14–24 years 2004 United 60 States United States 2005 Ghana 1900 United States average weekly work hours, men 2006 1950 50 40 Russian Guatemala Federation 2006 1994 Mexico Latvia 2009 30 2003 20 10 100 1,000 10,000 40,000 GDP per capita Sources: Berniell and Sanchez-Paramo 2011; Ramey and Francis 2009. Note: GDP = gross domest product. The vertical axis measures weekly hours spent on production activities (market and nonmarket work), including some outside the boundaries of the system of national accounts, such as child care. The measure does not include time allocated to schooling or leisure. The horizontal axis measures real GDP per capita in 2000 US$. assistance), or savings (social insurance and qualitative studies in low-income countries, get- pensions). Still, the majority of households ting jobs and starting businesses were two of the worldwide make their living through their work, main reasons people gave to explain their rise and labor earnings represent the largest share of out of poverty.5 Conversely, a lack of job op­ total household income (figure 2.3). The main portunities reduces the ability of households to change that comes with development is the improve their well-being.6 composition of labor income.4 Jobs are not the only force that determines Job-related events are the main escape route whether a household escapes from poverty. De- from poverty in developing and developed mographic changes, such as the arrival of a new- countries alike. More than two decades of re- born, relatives moving in, or a family split be- search on poverty dynamics, spanning countries cause of death or separation, affect expenditures as different as Canada, Ecuador, Germany, and per capita, hence the household’s poverty status. South Africa, show that labor-related events The same is true of changes in nonlabor income trigger household exits from poverty (figure from assets or transfers, be they private remit- 2.4). These events range from the head of a tances, public social assistance, or pensions. household taking a new job, to family members These developments may all interact and often starting to work, to working family members occur simultaneously. For example, the migra- earning more from their labor. In a large set of tion of family members to a city for a job may Jobs and living standards    79 Women F I G U R E 2 . 2   spend more time in activities not directly generating income 100 90 80 70 time allocation, percent 60 50 40 30 20 10 0 men women men women men women men women India Guatemala Spain United States income-generating activities investment other activities Source: World Development Report 2013 team based on ISSP 2005 for Spain and the United States, 1999 Time Use Survey of India, and 2006 Guatemala Household Survey. Note: The figure refers to people aged 15 years and more. Income-generating activities is the time devoted to wage or salaried employment; farming, own-account work, self-employment with hired labor, and unpaid family labor in household enterprises; investment refers to time allocated to education, health care, and job search; other activities includes work outside the system of national accounts, for example child care, housework. Leisure and other activities associated with consumption (for example, shopping and social interactions), as well as sleep, are not included. improve the well-being not only of the migrants (education, work experience, or region of resi- but also of those who stay in the rural village. In dence) were important, but that the returns addition to receiving remittances, those who stay to these characteristics mattered more. Among behind may have access to the migrants’ land to those returns is the relative price of labor. 9 cultivate and work more as a result.7 The connection between jobs and poverty With all these changes occurring at the same reduction is not mechanistic, and not all transi- time, gauging the contribution of labor earnings tions out of poverty require a change in the type to poverty reduction is difficult. However, recent of work undertaken. Changes in the productiv- methods allowing to decompose changes in pov- ity of the same job may also be at play. In Ban- erty by sources of income confirm the funda- gladesh and Vietnam, for example, poverty tran- mental contribution of change in labor earnings sitions have been dominated not by changes in (figure 2.5). In 10 of 18 countries considered for income sources from farm to nonfarm income, the analysis, labor income explains more than but by higher income within the same sector.10 half of the change in poverty, as measured by the Richer insights on the connection between US$2.50-a-day poverty line. In another 5 coun- labor-related events and transitions out of pov- tries, it accounts for more than a third of the erty can be obtained from studies that follow reduction in poverty.8 A further decomposition ­ the same households over extended periods of of the contribution of labor income to poverty time. Studies in several countries in Asia and in reduction in Bangladesh, Peru, and Thailand Sub-Saharan Africa show that farming and off- found that changes in individual characteristics farm activities are intricately related and not 80   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 BOX 2.2  Most poor people work  It is not lack of work that defines the poor. This realization has Caution is needed in interpreting this concept, however. Outside brought to the fore the concept of the working poor, and questions the group of the working poor, there may be individuals who have about who they are, and why they remain poor even when they very low labor earnings but whose expenditures are above the pov- have jobs. First studied by researchers in some countries such as the erty line because they have other sources of income such as private United States, this concept of the working poor is now recognized transfers or earnings from social insurance or social assistance pro- globally. The International Labour Organization (ILO) has included grams. In other words, being excluded from the category of working the working poor in its statistics since the mid-1990s, and measure- poor does not mean one has high labor earnings. ments of this group have been added as a Millennium Development Another concept that indicates whether job earnings are indicator. ­ sufficient to ensure an adequate standard of living for a person The working poor are defined as employed persons in house- or a household is the living wage. This is the level of earnings holds whose members are living below one of the two international that would provide a satisfactory standard of living to workers poverty lines—either US$1.25 or US$2 a day.a Household expendi- and their families. But moving from this definition to measurement ture surveys allow for a classification of the population as poor and is difficult. With more than half of all working people engaged nonpoor, based on the level of consumption per person. These sur- in nonwage work, accurate measures of labor earnings may not veys also provide information on household members who work. be available. Moreover, there are diverse interpretations of what According to the ILO’s most recent estimate, 910 million workers— constitutes a standard family and a lack of consensus on computa- nearly 30 percent of total global employment—were living on less tion methods.d An alternative is measuring the percentage of the than US$2 a day.b The incidence is much higher among low-income population that cannot reach the poverty line with labor incomes countries. It reaches 63.7 percent in Africa and 54.2 percent in Asia.c only, as the Poverty Labor Trend Index in Mexico does.e Source: World Development Report 2013 team. Notes: For a review of the working poor in developed countries, see Blank, Danziger, and Schoeni (2006) and Brady, Fullerton, and Cross (2010); for developing countries, see Fields (2011). The content and scope of the Millennium Development Goals can be found in United Nations, “We Can End Poverty, 2015: Millennium Development Goals,� United Nations, New York. a. Indicator 18, “Poverty, income distribution and the working poor,� KILM (Key Indicators of the Labour Market) (database), 7th ed. 2011, International Labour Organization, Geneva. b. ILO 2011, 41–42. c. Estimates are for 2009 for a selection of low-income countries from the ILO KILM. d. Anker 2011. e. Poverty Labor Trend Index, National Council for the Evaluation of Social Development Policy (CONEVAL), Mexico City. FI G U R E 2. 3  Jobs are the most important source of household income 100 90 percentage of household income 80 70 60 50 40 30 20 10 0 1) ) ) ) ) 5) ) ) 1) 5) ) ) 2) 5) 7) 9) ) ) 4) 05 00 05 03 03 06 05 05 01 04 00 00 00 99 Gh 200 00 00 00 00 20 0 20 20 20 20 20 20 20 20 (2 (2 (2 (2 1 2 (2 2 (2 ( a( a( ( a( ( a( ( ( a( ( r( i( ria a sia sh Gu pal n r ia m an ia ria aw do ca ni ny m al gu an ta liv an na de ist ga ge ne em as ba Ne na kis ua al Ke ra Bo nz et jik la l Ni ag do M Bu Al Pa ca Pa Ec at Vi ng Ta Ta ad In Ni Ba M self-employment wage employment transfers other Source: Covarrubias and others 2012 for the World Development Report 2013. Jobs and living standards    81 Jobs F I G U R E 2 . 4   take households out of poverty, especially in developing countries 100 90 80 transitions out of poverty, percent 70 60 50 40 30 20 10 0 ca ile a es r ru y il a en da m n s do nd an az in ric ai do Pe at Ri Ch na ed nt Sp Br ua Af rm la St a ng Ca Sw ge er st Ec h Ge d Ki th Co ut Ar ite Ne So d Un ite Un labor events nonlabor events Source: Inchauste 2012 for the World Development Report 2013. Note: Nonlabor events include changes in nonlabor earnings (such as rents or pensions) and demographic changes. A trigger event is defined as the most important event occurring during a poverty reduction spell among a set of mutually exclusive categories of events such as changes in family structure, in sources of income, and in needs of the household. necessarily substitutes for each other. Access to vestments, also raised the odds of escaping pov- land, increases in farm yields, and access to mar- erty, particularly in Uganda. All of these factors kets are fundamental for the growth of off-farm affect the labor productivity of farmers but jobs and hence for diversification in family in- originate in land markets or food markets comes.11 Simply having work is not what mat- rather than labor markets. ters most, according to these studies, since most The largest poverty reductions documented people work in rural economies. What is impor- are associated with jobs in agriculture. The cases tant for escaping poverty is deriving greater of China and Vietnam, in the 1980s and 1990s earnings from work. respectively, testify to the importance of agricul- Other factors of production are critical for tural productivity and the forces unleashed by explaining poverty reduction through jobs, land reform, investments in rural infrastructure, particularly in rural areas. Studies from Uganda and off-farm job opportunities.13 In rural China, and Pakistan, using rural data spanning 4 and poverty reduction was associated with off-farm 10 years respectively, show that higher agricul- activities, but the workers engaged in these ac- tural productivity, the growing commercializa- tivities tended to be those who had benefited tion of agriculture, and an increase in cash crop from increased farm incomes and by obtaining production contributed substantially to pov- more education.14 Furthermore, easier access to erty reduction. The increase in the price of cash off-farm employment and opportunities for mi- crops over this period also helped.12 Improve- gration reduced the exposure of households to ments in land rights and better access to input income shocks. A similar pattern of events has and output markets, due to infrastructure in- been documented in other Asian and Sub- 82   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Jobs F I G U R E 2 . 5   account for much of the decline in extreme poverty 200 percentage of total change in extreme poverty 150 100 50 0 –50 –100 ia or ico ile va ica a r ay nd a il a as a l sh ru pa do az in an bi m an ad ur de gu Pe do Ch la aR ex nt m Ne na Br ua Gh m nd ai lv la ra ol lo ge M Pa st Ec Th Ro Sa ng M Pa Ho Co Co Ar El Ba family composition labor income nonlabor income consumption-to-income ratio Sources: Azevedo and others 2012; Inchauste and others 2012; both for the World Development Report 2013. Note: Family composition indicates the change in the share of adults (ages 18 and older) within the household. Labor income refers to the change in employment and earnings for each adult. Nonlabor income refers to changes in other sources of income such as transfers, pensions, and imputed housing rents. If a bar is located below the horizontal axis, it means that that source would have increased, instead of decreased, poverty. The changes are computed for Argentina (2000–10); Bangladesh (2000–10); Brazil (2001–09); Chile (2000–09); Colombia (2002–10); Costa Rica (2000–08); Ecuador (2003–10); El Salvador (2000-09); Ghana (1998–2005); Honduras (1999–2009); Mexico (2000–10); Moldova (2001–10); Panama (2001–09); Paraguay (1999–2010); Peru (2002–10); Nepal (1996–2003); Romania (2001–09); and Thailand (2000–09). The changes for Bangladesh, Ghana, Moldova, Nepal, Peru, Romania, and Thailand are computed using consumption-based measures of poverty, while the changes for the other countries are based on income measures. Saharan African countries. Whereas poverty re- among households with few assets. Studies from duction in rural areas in Asia is associated with Uganda and Pakistan show that the share of diversification into nonfarm activities, in Sub- household members who work also has a con- Saharan Africa, it may be more closely associ- siderable impact. Households with rising de- ated with increases in farm productivity.15 pendency ratios were more likely to remain Jobs and relapses into poverty are also con- poor or fall into poverty, while households nected. Widespread shocks such as droughts, whose share of working-age adults increased floods, and conflicts can drive households into were less likely to fall into poverty or remain in a poverty or even chronic poverty. Events specific state of poverty.18 to individuals, such as illness or poor health of the head of household, can have the same effect. In these cases it is not joblessness per se that Jobs are more than just earnings pushes families into poverty but rather the de- struction of personal and household assets.16 Jobs have consequences beyond wages and earn- And even taking these shocks into account, job ings. Other aspects such as workplace safety, loss of the head of household remains a critical stability, commuting time, learning and ad- ­ determinant of a fall into poverty.17 vancement opportunities, entitlements to pen- The poor clearly rely on their labor to make sion benefits, and other amenities are highly val- a living. The death or disability of an income ued by some workers. However, quantifying the earner ­significantly increases the odds of falling monetary value of these other aspects of a job into poverty or remaining poor, particularly is not easy. Comparable surveys in Jianyang, Jobs and living standards    83 BOX 2.3  The value of job attributes can be quantified through hedonic pricing Workers place a value on jobs that goes beyond income. At the indi- in Colombia and China to 4.2 percent in Egypt and 5.1 percent in vidual level, people assess the impact a job might have on their Sierra Leone.a This is significantly lower than the explicit valuations physical and mental well-being, as well as on their families. In addi- answered by those surveyed: 4.9 percent in China, 10 percent in tion to the earnings the job provides, they can value the stability of Colombia and Sierra Leone, and, at the highest, 25 percent in Egypt. a job, its earnings, the possibilities of advancement, or the flexibility This indicates that the revealed preference of individuals for health of working hours. Workers might also value how well a job connects insurance benefits in the job are lower than the price they express them to society, the prestige associated with it, or its contribution to they would be willing to pay. Hedonic pricing can also identify the social goals. revealed preference to pay for other less tangible job charac­ Hedonic pricing assesses how people value specific job charac- teristics. Salaried workers in Colombia, China, and Egypt would teristics through their job satisfaction or happiness more broadly. forgo up to 1.5 percent of hourly wages for jobs that are “meaning- Indicators of subjective well-being are linked through sta­ tistical ful.� In Egypt, salaried workers reveal a price tag equivalent of up analysis to various job characteristics, including earnings. Statistical to 2.1 percent of hourly wages for jobs that are non-manual or methods can be used to assess the contribution of each of these job nonroutine. characteristics to happiness or job satisfaction. This approach is especially relevant in the assessment of job The weights associated with different job characteristics in the benefits. These benefits involve a deduction from earnings in ex-­ estimated hedonic price function allow an assessment of the value change for access to a pension in old age, for instance. Jobholders workers attach to each job characteristic. The monetary value of a typically value these benefits, but they may value them less than the job characteristic can be assessed by comparing the corresponding associated deductions in earnings through social security contribu- weight in the hedonic price function with the weight of earnings. tions. If the expected value of the pension is low or uncertain, Thus, for instance, a hedonic function reveals the share of earnings they may prefer to remain in the informal sector. In contrast, a well- respondents would be willing to forgo in exchange for stability, or designed program that allows longevity risks to be pooled with for creativity at work, or for a job providing voice in the workplace.b other jobholders may be valued by the jobholder more than the Using surveys commissioned for this Report, hedonic valuation deductions associated with participation. of health insurance benefits range from 1.5 percent of hourly wages Source: World Development Report 2013 team. a. Calculations by the World Development Report 2013 team of the FAFO (Forskningsstiftelsen Fafo [Fafo Research Foundation]) 2012 Survey on Good Jobs. b. Recent examples are Hintermann, Alberini, and Markandya 2010 and Falco and others 2012. China; Risaralda, Colombia; Cairo and Fayoum, dents and diseases kill an average of 6,000 peo- the Arab Republic of Egypt; and Port Loko and ple a day, or 2.2 million a year. Most of these Free Town, Sierra Leone, showed the limited deaths (1.7 million) result from work-related ability of respondents to attach a monetary value diseases; the remainder is linked to fatal acci- to job benefits, despite expressing willingness to dents in the workplace and during commutes pay.19 Among those who do give an explicit valu- to or from work.20 Every year, more than 400 ation, the willingness to pay for pension benefits million people (nearly 15 percent of the global goes from 5 percent of monthly wages in China labor force) suffer from occupational accidents to 7 percent in Colombia and 13 percent in or illnesses involving work-related diseases. In Egypt. Lower values are given for transportation some cases, the incidence is intolerably high: allowances (2, 1, and 7 percent, respectively), but half of slate pencil workers in India and 37 per- having a permanent contract is valued more, es- cent of the miners in Latin America suffer from pecially in Egypt (3, 8, and 22 percent, respec- some stage of silicosis (an occupational lung tively) (box 2.3). disease caused by inhalation of silica dust).21 Characteristics of jobs have other less tan- Mental health can be threatened by abusive re- gible, but no less real, effects on well-being. In lations between managers and workers and particular, jobs can have a direct impact on sexual harassment. Health risks are not con- workers’ health, a key component of human fined to wage employment. Collecting and car- development and personal well-being (box rying water or cooking over open stoves, as 2.4). Ex­posure to hazardous substances causes many self-employed workers do, poses risks, an ­estimated 651,000 deaths annually, mainly affect women and these risks are more likely to ­ in developing countries. Work-related acci- than men.22 84   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 by the system of social benefits, and from dis- BOX 2.4  Work can pose risks to health and safety  crimination. In any case, a lower employment rate is one of the main channels through which Surveys of workers in garment factories in three countries underscore the disability may lead to poverty. health and safety hazards they face in their jobs. Garment workers in Indonesia, In countries where wage employment is Jordan, and Haiti have reported physical stresses linked to work, including hun- the norm, joblessness may severely affect well- ger, thirst, and severe fatigue. being. Together with income, social status has In Indonesia, more than half the workers surveyed reported that they had experienced severe thirst often or every day. Heat is a likely contributor. Asked been recognized as an important factor in the whether the factory is too hot or too cold, only about half (52 percent) reported development and maintenance of mental that temperature was not a concern. Occupational safety is an issue for many: health.26 Studies document the detrimental ef- 59 percent of workers reported concerns about dangerous equipment; 73 per- fects of unemployment and the positive effects cent were concerned about accidents; 64 percent, about dusty or polluted air; when finding a job.27 Medical research has as- and 69 percent, about chemical odors. sociated unemployment with stress, depression, In factories in Jordan, 37 percent of workers reported concerns about dan- heart disease, and alcoholism.28 Psychological gerous equipment, and 45 percent reported concerns about accidents and injuries. hardship, marital dissolutions, and suicide have In Haiti, 40 percent of workers reported that they had experienced severe also been associated with job loss.29 Depression fatigue or exhaustion occasionally, often, or every day; 41 percent reported and stress-related illnesses are becoming more frequent headaches, dizziness, backaches, or neck aches. A stunning 63 per- common with the expansion of outsourcing, cent of workers reported that they had experienced severe thirst often or labor informality, and mobility in the modern every day. workplace.30 The impact of unemployment on mental Source: IFC and ILO 2011. health appears to occur independently of the availability of social insurance or other mecha- nisms of protection.31 This is because the psy- Occupational accidents and work-related chological hardship of unemployment is also diseases have economic costs. These costs are associated with social stigma. Studies show that difficult to compute because the estimates a worker who is unemployed or who has a vul- ought to include spending on health care and nerable job faces less duress if the phenomenon sickness benefits, as well as the forgone earn- is more pervasive or if there is less inequality in ings from workdays lost. Estimating these costs the incidence of unemployment or the distribu- is particularly difficult in the case of the self- tion of vulnerable jobs. This finding demon- employed. The few studies that have tried to do strates the close interaction between a person’s so suggest that the burden on society could be job and their place in society.32 high. In Spain, in the industrial sector alone, these costs were estimated to amount to 1.72 percent of gross domestic product (GDP) in Jobs and life satisfaction 2004. In Mauritius, the cost of work-related in- juries r ­ epresented around 2.8 percent of GDP Happiness, both a personal goal and a social as- in 2003.23 Global estimates put the cost asso­ piration, is related to employment status. A large ciated with work-related sickness at around 4 body of literature shows that unemployed peo- percent of GDP.24 ple report lower happiness and life satisfaction Opportunities to participate in labor markets than their employed counterparts.33 For in- for people with disabilities vary across countries. stance, in Indonesia subjective well-being in- The employment ratio of people with disabilities creases when gaining a job and decreases when ranges from 70 percent, in Poland, to 20 percent, losing it.34 Some researchers argue that this dis- in Switzerland and Zambia, lower than the ratio content is transitory, but others point out that, for the overall population.25 Disabilities may be as long as concerns about job stability persist, preexisting conditions or the result of job-related so does un­ happiness. This “unhappiness effect� injuries or conditions. Different labor outcomes is more typically reported in men than in among persons with disabilities stem from pro- women, but evidence indicates that women are ductivity differentials, from disincentives created affected by the unemployment of their spouse. Jobs and living standards    85 The lack of employment can lower the self- esteem and undermine the s ­ ocial status of other Workers F I G U R E 2 . 6   often care more about job security family members.35 than about income When jobs are in short supply and unem- ployment becomes a problem, people change 4.9 their expectations and attitudes. Data from the World Value Surveys for a large set of countries 4.7 (both developed and developing) show that mean score, job security higher unemployment rates are associated with 4.5 lower ambitions to do meaningful work, per- haps indicating that a lack of available jobs im- 4.3 pels individuals to accept any job. It is not only one’s joblessness that may be 4.1 important to life satisfaction. In the United Kingdom, the unemployed are less unhappy in 3.9 districts in which the unemployment rate is higher, suggesting that joblessness always hurts 3.7 but that it hurts less if there are many unem- ployed people in the local area.36 The effect on 3.5 happiness of not having a job seems to be par- 3.5 3.7 3.9 4.1 4.3 4.5 4.7 4.9 tially offset by the lower social stigma when mean score, job income the lack of jobs is widespread. Joblessness also leads to a loss of contact with people through the workplace and to a contraction in related Source: ISSP 2005. Note: The analysis covers 29 countries, each represented in the figure by a dot. Respondents scored the social networks, which can erode social capital importance of job security and job income on a scale from 1 to 5, with 5 = very important, 4 = important, and undermine the sense of engagement with 3 = neither important nor unimportant, 2 = unimportant, 1 = not important at all. others.37 Simply having a job does not guarantee higher life satisfaction. Feeling insecure at work employed have higher levels of satisfaction because of earnings variability, job instability, or than the unemployed. health and safety concerns also affects a person’s Whether the self-employed express greater sense of well-being (figure 2.6).38 For wage satisfaction than wage workers depends on the workers, the type of contract and its duration context. In industrial countries and in Eastern are important; part-timers and seasonal work- Europe and Central Asia, life satisfaction is, on ers express less job satisfaction. Even workers average, similar among both groups, but in with long-term contracts may feel insecure.39 In Latin America, it is substantially lower among factories in Haiti, Jordan, and Vietnam, earn- the self-employed.42 ings from work did not influence the reported Jobs contribute to how people view them- level of life satisfaction, but working conditions selves and relate to others. Most people feel did.40 In more developed countries, jobs that strongly that their jobs should be meaningful provide more autonomy are linked to higher life and contribute to society. A 2005 survey of 29 satisfaction.41 countries asked people about the characteristics Most research on the links between jobs and that they valued in their jobs.43 Over three- life satisfaction has been conducted in settings quarters reported that it is important to have a where wage employment is the norm. A grow- job that is useful to society, and a similar share ing literature on life satisfaction in developing agreed that it is important that their jobs help regions, where a smaller share of those who other people. In nine countries, the share who work are wage earners, shows that farmers have reported that it is important for jobs to be so- the lowest levels of life satisfaction relative to cially useful was higher than the share reporting other workers and the unemployed (figure that high income is important. While most of 2.7). Meanwhile, wage workers and the self- these are high-income countries, preferences for 86   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 and social norms. Notwithstanding this, other Life satisfaction is lower among farmers and the FI G U R E 2.7  health variables used as proxies of job satisfac- unemployed tion such as absence of fear or sadness also show an association with working conditions. Research for Haiti, Jordan, and Vietnam finds working conditions such as basic hygiene and 50 health, workplace facilities, or presence of 45 unions to be associated with fewer feelings of 40 fear or sadness.44 35 30 * * * percent 25 20 Jobs have an impact on the well-being of the 15 high income person who holds them, but they can also have Latin America and the Caribbean an impact on the well-being of others. Some 10 Europe and Central Asia East Asia and Paci c jobs bring more poverty reduction and, as 5 Middle East and North Africa such, benefit those who consider eradicating 0 South Asia ed poverty to be a fundamental societal goal. ploy loye d Sub-Saharan Africa e em ers Some jobs promote higher employment rates wag l f - emp farm oye d se mpl une among women, giving more say on the way household resources are allocated, typically leading to greater spending on raising children. Sources: Gallup 2009, 2010. Gender equality, much the same as poverty re- duction, is a broadly shared societal goal. Jobs socially useful and high-income jobs did not dif- that have these additional impacts do more for fer greatly in the Dominican Republic, Mexico, development. Given such spillover effects, jobs or South Africa. play a fundamental role in the well being of Job satisfaction and other measures of non- ­ individuals and entire societies. Jobs may thus material well-being such as happiness or iden- be the center piece of a development strategy tity may be affected by cultural differences (Question 2). QUESTION 2 Growth strategies or jobs strategies? Rapid and sustained growth is generally viewed come and the possibility of redistributing re- as the main priority for developing countries, sources through the growth process itself and and as a precondition for continued increases in through government transfers.51 living standards and strengthened social cohe- Behind these sensible qualifiers, it is possible sion. Economic growth, living standards, and to point to the role of jobs. Growth is “inclusive� social cohesion can indeed move together, and when higher earnings are driven by employ- they often do—as shown, for example, by the ment opportunities for the majority of the labor remarkable experience of East Asian economies, force, particularly the poor. Recent studies show including the Republic of Korea and Singa- that the impact of economic growth on poverty pore.45 Building on the East Asian experience, reduction depends critically on the employment the conventional wisdom is to focus on growth intensity of different sectors.52 Employment and assume that increased living standards and opportunities also matter for social cohesion. ­ greater social cohesion will follow. This is the It is thus jobs that bring together the three main tenet behind “growth strategies,� “growth transformations. diagnostics,� and “binding constraints analyses,� Realizing the role jobs play implies going be- all of which aim to identify and remove obsta- yond the sequential view in which growth issues cles to economic growth and to sustain it over are addressed first and employment follows prolonged periods of time. from increased demand. Instead, jobs are seen But transformations in living standards, pro- as a medium that can make the development ductivity, and social cohesion do not necessarily transformations a reality. From a sta­ tistical happen at the same pace. Lags and gaps in rising point of view, the relationship between growth living standards can be illustrated by the differ- and employment (or unemployment) shows ent impacts growth has on poverty reduction substantial variation over time, across countries, across countries. A 2 percent annual growth rate and across sectors. In light of this diversity, a can reduce poverty rates by 1 percent in some given rate of growth does not guarantee a given countries and by 7 percent in others.46 Ethiopia, level of job creation or a given composition of Tanzania, and Zambia experienced periods of employment (box 2.5). economic growth with very little change in pov- erty incidence.47 On the other hand, important When a growth strategy may not be advances in poverty reduction have also hap- sufficient pened during periods of slow growth, as oc- curred in Brazil and Mexico during the 1990s Focusing on the aggregate relationship be- and the first half of the 2000s.48 And in some tween growth and employment downplays cases, growth is not accompanied by increased some of the most important channels through social cohesion—even though poverty may fall which jobs connect to development. The very and living standards improve for some, the ex- notion of employment as derived labor de- pectations of others remain unfulfilled. Tunisia mand does not reflect the situation of the many is a clear example in this regard: its growth rate working people in developing countries who is well above the average of the region, but it has are farmers and self-employed. The focus on nonetheless experienced serious social and po- the labor market as the transmission chain be- litical tensions.49 tween growth and employment also does not The recognition of these lags and gaps has capture the interaction of working people with led to more nuanced approaches to economic others in households, at the workplace, and in growth in which the growth being sought is society more broadly. Focusing solely on the “pro-poor,� “shared,� or “inclusive.�50 In these relationship between growth and employment versions, it is not just the rate of growth that may fail to measure how jobs can foster gender matters but also the initial distribution of in- equality, support urbanization, or contribute 88   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 BOX 2.5  The relationship between growth and employment is not mechanical  The statistical connection between economic growth and employ- ment and the percentage change in GDP. These elasticities show ment is sometimes termed Okun’s Law. In 1962 Arthur Okun found great variability over time and space, too, making it difficult to fore- that in the years immediately following World War II, a 1 percent cast net job creation over the course of development. For instance, increase in gross domestic product (GDP) in the United States in Tanzania growth elasticities of employment declined from 1.04 in brought about a 0.3 percent decline in unemployment. Since then, the period 1992–96 to 0.27 in the period 2004–08. Similar trends this empirical regularity has found support in a wide variety of coun- have been reported for Ethiopia, Ghana, and Mozambique.c In Latin tries. Recent research, however, suggests that Okun’s Law is not as America, recent estimates show that growth elasticities of employ- stable as its name implies.a ment were much lower during the global financial crisis than in pre- The debate on the stability of Okun’s Law sheds light on the char- vious crises. In other words, the Great Recession produced compara- acteristics of economic recessions and expansions. A recent study tively less net employment destruction in that region.d indicates that, in industrial countries, unemployment has become While employment and unemployment are aggregates, growth more responsive to output declines over the past 20 years. This has may also affect the composition of unemployment. Important con- been attributed to institutional reforms that have made labor mar- troversies, such as why manufacturing employment in India has kets more flexible. Interestingly, economies that suffer financial cri- stagnated despite rapid growth in the sector can be interpreted in ses and large housing price busts (such as the United States and this light.e Other studies show that, given their different labor inten- Spain in recent years) have deeper and longer increases in unem- sities, economic growth in some sectors like agriculture, construc- ployment than Okun’s Law would have predicted; whereas econo- tion, or services generates more employment than does economic mies with large short-time work schemes (like Germany, Italy, Japan, growth in manufacturing.f Investment projects in agribusiness in and the Netherlands) show less unemployment than predicted.b Ukraine, in construction in India, and in tourism in Rwanda have had While Okun’s Law relates to unemployment, other studies focus large employment impacts, not only because of the direct jobs cre- on the growth elasticity of employment. In its simplest form, this ated but also because of indirect job creation in their large network elasticity is the ratio between the percentage change in employ- of distribution channels.g Source: World Development Report 2013 team. a. Cazes, Verick, and Al Hussami 2011; Moosa 2012. b. Balakrishnan, Das, and Kannan 2009. c. Martins 2012 for the World Development Report 2013. d. World Bank 2010. e. Bhalotra 1998; Roy 2004. f. Arias-Vasquez and others 2012 for the World Development Report 2013. g. IFC, forthcoming. to peaceful collective decision making. Under- In practice, however, tradeoffs between the standing how to enhance these positive spill- three transformations can amount to more than overs from jobs might be difficult when only just lags and gaps. Depending on the nature of aggregates are considered. the jobs challenges facing a country, tensions The case of urbanizing economies such as may emerge between growth that generates jobs Bangladesh may support the idea that the three for living standards and growth that generates major transformations happen simultaneously. jobs for productivity growth or for social cohe- Taking advantage of their abundance of rela- sion. Examples abound: tively low-skilled labor, such economies can engage in world markets through light manu- ­ • In agrarian economies, increasing productiv­ facturing. Wage employment is created in large ity in smallholder farming is fundamental for numbers, providing opportunities for rural mi- poverty reduction, given the share of the pop- grants, and cushioning social tensions at a time ulation living in rural areas. But urban jobs in of rapid social change. In Bangladesh, the ex- activities that connect the economy to world pansion of the light manufacturing sector has markets and global value chains are necessary allowed for the integration of young women for growth. With limited resources to support into the labor market, at a time of falling fertility both, a tradeoff between living standards and rates. Employment opportunities for women productivity may arise. have in turn led to growing female schooling, better human development outcomes, and • In resource-rich countries, massive invest- faster poverty reduction. ments in extractive industries support accel- Jobs and living standards    89 erated rates of growth and connections with fully accounting for the negative impact of cur- international markets but generate little di- rent pollution on workers’ future health would rect (or even indirect) employment and of- make a more complete evaluation of the output ten little poverty reduction. Moreover, the potential of a growth strategy based on a given abundance of foreign exchange undermines technology. Opting for defused tensions or the competitiveness of other activities, mak- greater integration in world trade would lay the ing it difficult to create productive jobs in ground for accelerating growth in the future in a other sectors. sustainable way, which a short-term evaluation based on output growth alone would fail to con- • In countries with high youth unemployment, sider. If measures of growth captured the intan- job opportunities are not commensurate with gible social benefits from jobs, a growth strategy the expectations created by the expansion of and a jobs strategy would be equivalent. How- education systems. And the active labor mar- ever, when focusing on measured growth only, ket programs needed to defuse social tensions spillovers from jobs can easily be overlooked, in the short term may not do much for pov- and this is why a jobs strategy may be needed. By erty reduction because many of the jobless focusing on the spillovers from jobs, a jobs strat- come from middle-class families, and devot- egy highlights the different outcomes of interest ing public resources to finance them may re- in a development process. duce economic dynamism. Considering a jobs strategy is a way to call • In formalizing economies, there is an effort attention to the social value of jobs. A jobs strat- to support social cohesion by extending the egy assesses the types of jobs that do more for coverage of social protection to as many development in a particular country context. It workers as possible. Broad coverage regard- relies on qualitative and quantitative analyses to less of the type of job is often seen as part of a identify how jobs contribute to living standards, social compact. But extending coverage with- productivity, and social cohesion. And it seeks to out distorting incentives to work, save, and identify where the constraints to the creation of participate in formal systems is difficult and the jobs with the highest development payoff may have adverse impacts on productivity lie in practice. In some cases, a jobs strategy will and long-term growth. focus on increasing female labor participation, in others on creating employment opportuni- ties for youth, yet in others on creating a sup- When a jobs strategy may be appropriate portive environment for the creation of jobs in Tradeoffs between improving living standards, cities, or jobs connected to global value chains. accelerating productivity growth, and fostering This may not be too different from preparing social cohesion arguably reflect a measurement a more comprehensive growth strategy, except problem. While the contribution jobs make to that jobs would be center stage. output can be quantified, some of the spillovers Jobs strategies are not needed under all cir- from jobs cannot. Measured output does not cumstances. A jobs strategy is warranted only increase when jobs defuse social tensions, even when potentially important spillovers from though these outcomes are valued by society jobs are not realized, leading to tensions be- and may increase productivity in the future. tween living standards, productivity, and social Conversely, measured output does not decline cohesion. When improvements in living stan- when jobs in export sectors are replaced by jobs dards, productivity, and social cohesion happen producing for the domestic market, even together, as was the case in several East Asian though the oppor­ tunities to acquire technical countries, and may now be the case in urbaniz- and managerial knowledge through work tend ing economies such as Bangladesh, a growth to be higher in the export sectors. strategy may be more appropriate. Yet even re- If the spillovers from jobs could be appro­ markably successful East Asian economies such priately quantified, the tradeoffs would be fully as Korea and Singapore, which undoubtedly de- understood and an adequate evaluation of the livered inclusive growth over many decades, also output and employment potential of a given had jobs strategies at specific points in their de- growth strategy would be possible. For example, velopment histories (box 2.6). 90   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 BOX 2.6  Korea went from a growth to a jobs strategy, and Singapore the other way around  The Republic of Korea and Singapore are success stories combining The strategy identifies four pillars to achieve the 70 percent tar- long-term economic growth with rapid poverty reduction and get. The first recognizes the importance of collaboration between strong social cohesion. But at different points in time, both coun- the public and private sectors for employment creation and consists tries relied on jobs strategies. of implementing economic and industrial policies in a job-friendly Singapore was confronted with a tense social situation at inde- manner. The second aims at improving flexibility and fairness in the pendence, with both high unemployment and inter-ethnic tension. workplace and consists of a series of reforms to increase regulation Its first development strategy focused on jobs, housing, and wage in certain areas of the labor law, while decreasing regulation in oth- moderation. As unemployment subsided, the next strategy was ers. Thus the 40-hour workweek became enforceable for all compa- geared toward raising labor costs to encourage higher-value-added nies, regardless of size,a with the obligatory introduction of the work activities. This cost drive resulted in a recession, however, and since time savings system.b Simultaneously, regulations on duration of con- then Singapore has focused on growth, rather than jobs. tracts for temporary workers and fixed-term contracts were relaxed Conversely, Korea abandoned development planning in 1996, to allow for more hiring flexibility. The third pillar focuses on increas- but in 2010, it adopted a jobs strategy for the next decade as its ing labor force participation and skill development of women, youth, highest-level policy document. In October 2010, the Korean govern- and older workers. This involves developing the option of perma- ment launched the “National Employment Strategy 2020 for the Bal- nent part-time jobs, thus allowing parents to both work and care for ance of Growth, Employment and Welfare.� In the tradition of long- their children, especially in sectors suffering from labor shortages range plans, this national strategy has a clear target for 2020: an and unable to fill full-time jobs. Older workers would be retained increase in the employment rate of the working-age population longer in the active labor force by having the option to work shorter (15–64 years) to a minimum of 70 percent—the average among hours under the wage peak system.c Last but not least, the inten- industrial economies. The strategy was rooted in the mismatch tion is to facilitate welfare-to-work transitions, by encouraging able- between macroeconomic indicators that pointed to a recovering bodied welfare recipients to enroll in employment assistance pro- economy and the inability of individuals—especially youth—to find grams and by reinforcing their obligation to pursue employment. adequate employment. Sources: World Development Report 2013 team based on Huff 1994, 1995; Republic of Korea 2010. The 40-hour workweek was introduced in 2004 and applied only to companies with over 1,000 employees. a.  This system allows employees to take leave to compensate for overtime, work during holidays, or night work. b.  The wage peak system allows companies to rehire workers after they retire. c.  © Justin Guariglia / Redux Day laborer in a pineapple plantation in Pontian, Malaysia Notes 22. Al-Tuwaijri and others 2008; Brenner 1979. 23. Estimates for both Spain and Mauritius are from  1. Gershuny 2000; Krueger and others 2009; Ramey Ramessur (2009). and Francis 2009. 24. ILO 2005.   2. Ngai and Pissarides 2008. 25. WHO and World Bank 2011.   3. See Hongqin, MacPhail, and Dong (2011) for the 26. Wilkinson and Marmot 1998. case of growing female participation in China. 27. Baingana and others 2004; Murphy and Athana- On the other hand, see Gammage and Mehra sou 1999. (1999) for the case of stagnant female participa- 28. Brenner 1971; Brenner 1975; Dooley, Catalano, tion in the Middle East. and Wilson 1994; Dooley, Prause, and Ham-   4. Davis and others 2010. Rowbottom 2000.   5. Narayan, Pritchett, and Kapoor 2009. 29. Lundin and Hemmingsson 2009; Stuckler and   6. See the studies cited in Baulch 2011; Fields and others 2009a, 2009b. others 2003; and Fields and others 2007. 30. ILO 2010.   7. de Brauw and Giles 2008; Giles and Murtazash- 31. Ouweneel 2002. vili 2010. 32. Helliwell and Putnam 2004; Stutzer and Lalive   8. Azevedo and others 2012 for the World Develop- 2004. ment Report 2013. In El Salvador and ­ Romania, 33. Blanchflower and Oswald 2011; Winkelmann nonlabor incomes compensated for lower labor and Winkelmann 1998. There are valid concerns incomes as a result of the financial crisis. For about how to compare self-reported subjective Mexico, although earnings increased for the em- outcomes across countries and cultures. See King ployed, this effect was compensated for by a de- and others 2004. cline in occupied adults, resulting in a relatively 34. Gales, Mavridis, and Witoelar 2012 for the World lower contribution of labor income to poverty Development Report 2013. reduction when compared to transfers. 35. Björklund 1985.   9. Inchauste and others 2012 for the World Devel- 36. Clark and Oswald 1994. opment Report 2013. 37. Helliwell and Putnam 2004. 10. Dang and Lanjouw 2012. 38. Dooley, Prause, and Ham-Rowbottom 2000; 11. Estudillo, Sawada, and Otsuka 2008; Himanshu, Winefield 2002. Bakshi, and Dufour 2011; Lanjouw and Lanjouw 39. Bardasi and Francesconi 2004; Origo and Pagani 2001; Lanjouw and Murgai 2009; Otsuka, Estudi- 2009. llo, and Sawada 2009; Takahashi and Otsuka 2009. 40. Dehejia, Brown, and Robertson 2012 for the 12. Mansuri and others 2012a for the World Devel- World Development Report 2013. opment Report 2013. 41. Wietzke and McLeod 2012 for the World Devel- 13. Glewwe, Gragnolatti, and Zaman 2002; Ravallion opment Report 2013. and Chen 2007; Ravallion, Chen, and Sangraula 42. Graham 2008. 2009. 43. ISSP 2005. 14. Christiaensen and others 2009; de Brauw and 44. Dehejia, Brown, and Robertson 2012 for the others 2002; Giles 2006; Giles and Yoo 2007. World Development Report 2013. 15. Christiaensen and Todo 2009; Estudillo and oth- 45. Gill and Kharas 2007; Stiglitz 1996; World Bank ers 2012; Himanshu and others 2011. 1993. 16. Dercon and Porter 2011; Fields and others 2003; 46. Ravallion 2001; Ravallion 2011. Lawson, McKay, and Okidi 2006; Lohano 2011; 47. Bigsten and others 2003; Demombynes and Quisumbing 2011; Woolard and Klasen 2005. Hoogeeven 2007. 17. Fields and others 2003; Fields and others 2007. 48. Ferreira, Leite, and Ravallion 2010; Hanson 2010. 18. Mansuri and others 2012b for the World Devel- 49. The GDP per capita (in real 2000 US$) grew in opment Report 2013. Tunisia at an annual average rate of 3.4 percent 19. Bjørkhaug and others 2012 for the World Devel- between 1990 and 2008, whereas the average for opment Report 2013; Hatløy and others 2012 the Middle East and North Africa region was 2.0 for the World Development Report 2013; Kebede percent in the same period (World Development and others 2012 for the World Development Re- Indicators). port 2013; Zhang and others 2012 for the World 50. There are several measures that gauge “pro-poor� Development Report 2013. growth. See Ravallion 2004. 20. ILO 2010. 51. Ianchovichina and Lundstrom 2009; Ravallion 21. ILO 2005. 2001. Jobs and living standards    93 52. Christiaensen, Demery, and Kuhl 2011; Loayza Berniell, M. Inés, and Carolina Sanchez-Paramo. and Raddatz 2010. 2011. “Time Use Database.� World Bank, Wash- ington, DC. Processed. Bhalotra, Sonia R. 1998. “The Puzzle Of Jobless Growth in Indian Manufacturing.� Oxford Bulle- References tin of Economics and Statistics 60 (1): 5–32. Bigsten, Arne, Kebede Bereket, Abebe Shimeless, The word processed describes informally repro- and Mekonnen Taddesse. 2003. “Growth and Pov- duced works that may not be commonly avail- erty Reduction in Ethiopia: Evidence from House- able through libraries. hold Panel Surveys.� World Development 31 (1): 87–106. Adelman, Irma, and Cynthia Taft Morris. 1973. Eco- Bjørkhaug, Ingunn, Anne Hatløy, Tewodros Kebede, nomic Growth and Social Equity in Developing and Huafeng Zhang. 2012. “Perception of Good Countries. Stanford, CA: Stanford University Press. Jobs: Colombia.� Background paper for the WDR 2013. Al-Tuwaijri, Sameera, Igor Fedotov, Ilise Feitshans, Björklund, Anders. 1985. “Unemployment and Men- Malcolm Gifford, David Gold, Seiji Machida, Mi- tal Health: Some Evidence from Panel Data.� Jour- chele Nahmias, Shengli Niu, and Gabor Sandi. nal of Human Resources 20 (4): 469–83. 2008. Beyond Death and Injuries: The ILO’s Role in Blanchflower, David G., and Andrew J. Oswald. 2011. Promoting Safe and Healthy Jobs. Geneva: Interna- “International Happiness.� Working Paper Series tional Labour Organization. 16668, National Bureau of Economic Research, Alkire, Sabina, and James Foster. 2011. “Understand- Cambridge, MA. ings and Misunderstandings of Multidimensional Blank, Rebecca M., Sandra K. Danziger, and Robert F. Poverty Measurement.� Journal of Economic In- Schoeni. 2006. Working and Poor. New York: Rus- equality 9 (2): 289–314. sell Sage Foundation. Anker, Richard. 2011. Estimating a Living Wage: A Bourguignon, François, and Satya Chakravarty. 2003. Methodological Review. Geneva: International “The Measurement of Multidimensional Poverty.� ­Labour Organization. Journal of Economic Inequality 1 (1): 25–49. Arias-Vasquez, Javier, Jean N. Lee, and David New- Brady, David, Andrew Fullerton, and Jennifer Moren house. 2012. “The Role of Sectoral Growth Pat- Cross. 2010. “More Than Just Nickels and Dimes: terns in Labor Market Development.� Background A Cross-National Analysis of Working Poverty paper for the WDR 2013. in Affluent Democracies.� Social Problems 57 (4): Azevedo, João Pedro, Gabriela Inchauste, Sergio 559–85. Olivieri, Jaime Saavedra Chanduvi, and Hernan ­ Brenner, Harvey. 1971. “Economic Changes and Winkler. 2012. “Is Labor Income Responsible for Heart Disease Mortality.� American Journal of Poverty Reduction? A Decomposition Approach.� Public Health 65 (12): 606–11. Background paper for the WDR 2013. ———. 1975. “Trends in Alcohol Consumption and Baingana, Florence, Andrew Dabalen, Essimi Menye, Associated Illnesses: Some Effects of Economic Menahem Prywes, and Michael Rosholm. 2004. Changes.� American Journal of Public Health 65: “Mental Health and Socio-Economic Outcomes 1279–92. ———. 1979. “Mortality and the National Economy.� in Burundi.� Health, Nutrition and Population The Lancet 26: 568–73. Discussion Paper, World Bank, Washington, DC. Cazes, Sandrine, Sher Verick, and Fares Al Hussami. Balakrishnan, Ravi, Mitali Das, and Prakash Kannan. 2011. “Diverging Trends in Unemployment in the 2009. “Unemployment Dynamics during Reces- United States and Europe: Evidence from Okun’s sions and Recoveries: Okun’s Law and Beyond.� Law and the Global Financial Crisis.� Employ- In IMF World Economic Outlook: Rebalancing ment Working Paper Series 106, International Growth, 69–108. Washington, DC: International Labour Organization, Geneva. ­ Monetary Fund. Chenery, Hollis Burnley, and Moises Syrquin. 1975. Bardasi, Elena, and Marco Francesconi. 2004. “The “Patterns of Development: 1950–1970.� In Redis- Impact of Atypical Employment on Individual tribution with Growth: Policies to Improve Income Wellbeing: Evidence from a Panel of British Work- Distribution in Developing Countries in the Con- ers.� Social Science & Medicine 58 (9): 1671–88. text of Economic Growth, ed. Hollis Burnley Chen- Baulch, Bob, ed. 2011. Why Poverty Persists: Poverty ery, Richard Jolly, Montek S. Ahluwalia, C. L. Bell, Dynamics in Asia and Africa. Cheltenham, U.K.: and John H. Duloy. Oxford: Oxford University Edward Elgar Publishing. Press. 94   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Christiaensen, Luc, Ruchira Bhattamishra, Lei Pan, Dooley, David, Ralph Catalano, and Georjeanna and Sangui Wang. 2009. “Pathways Out of Poverty ­ Wilson. 1994. “Depression and Unemployment: in Lagging Regions: Evidence from Rural Western Panel Findings from the Epidemiologic Catch- China.� World Bank, Washington, DC. Processed. ment Area Study.� American Journal of Commu- Christiaensen, Luc, Lionel Demery, and Jesper Kuhl. nity Psychology 61 (3): 745–65. 2011. “The (Evolving) Role of Agriculture in Pov- Dooley, David, JoAnn Prause, and Kathleen Ham- erty Reduction—An Empirical Perspective.� Jour- Rowbottom. 2000. “Inadequate Employment and nal of Development Economics 96 (2): 239–54. High Depressive Symptoms: Panel Analyses.� In- Christiaensen, Luc, and Yasuyuki Todo. 2009. “Pov- ternational Journal of Psychology 35: 294. erty Reduction during the Rural-Urban Transfor- Estudillo, Jonna P., Tomoya Matsumoto, Ziauddin mation: The Role of the Missing Middle.� Paper Hayat Chowdhury, Nandika Kumanayake, and presented at the International Association of Agri- Keijiro Otsuka. 2012. “Labor Markets, Occupa- cultural Economists 2009 Conference, Beijing, tional Choice, and Rural Poverty in Selected August 16. Countries in Asia and Sub-Saharan Africa.� Back- Clark, Andrew, and Andrew J. Oswald. 1994. “Un­ ground paper for the WDR 2013. happiness and Unemployment.� Economic Journal Estudillo, Jonna P., Yasuyuki Sawada, and Keijiro 104: 648–59. Otsuka. 2008. “Poverty and Income Dynamics in Covarrubias, Katia, Benjamin Davis, Aminata Bak- Philippine Villages, 1985–2004.� Review of Devel- ouan, and Stefania Di Giuseppe. 2012. “House- opment Economics 12 (4): 877–90. hold Income Generation Strategies.� Background Falco, Paolo, William Maloney, Bob Rijkers, and paper for the WDR 2013. Mauricio Sarrias. 2012. “Subjective Well-Being, Dang, Hai-Anh, and Peter Lanjouw. 2012. “Measur- Informality, and Preference Heterogeneity in ing Poverty Dynamics and Labor Transitions Africa.� Background paper for the WDR 2013. with Synthetic Panels Based on Cross-Sections.� Ferreira, Francisco H. G., Phillippe G. Leite, and Processed. Martin Ravallion. 2010. “Poverty Reduction and ­ Davis, Benjamin, Katia Covarrubias, Kostas Stamoulis, Economic Growth? Explaining Brazil’s Poverty Paul C. Winters, Carlogero Carletto, Esteban Qui- Dynamics 1985–2004.� Journal of Development ñones, Alberto Zezza, and Stefania DiGiuseppe. Economics 93 (1): 20–36. 2010. “A Cross-Country Comparison of Rural In- Fields, Gary. 2011. Working Hard, Working Poor. New come Generating Activities.� World Development York: Oxford University Press. 38 (1): 48–63. Fields, Gary, Paul Cichello, Samuel Freije-Rodriguez, de Brauw, Alan, and John Giles. 2008. “Migrant Labor Marta Menendez, and David Newhouse. 2003. Markets and the Welfare of Rural Households in “Household Income Dynamics: A Four-Country the Developing World: Evidence from China.� Story.� Journal of Development Studies 40 (2): Policy Research Working Paper Series 4585, World 30–54. Bank, Washington, DC. Fields, Gary, Robert Duval Hernández, Samuel Freije, de Brauw, Alan, Jikun Huang, Scott Rozelle, Linxiu and María Laura Sánchez Puerta. 2007. “Intragen- Zhang, and Yigang Zhang. 2002. “The Evolution erational Income Mobility in Latin America.� of China’s Rural Labor Markets during the Re- Economía 7 (2): 101–54. forms.� Journal of Comparative Economics 30 (2): Fitoussi, Jean-Paul, Amartya Sen, and Joseph E. 329–53. Stiglitz. 2010. Mismeasuring Our Lives: Why GDP Dehejia, Rajeev, Drusilla Brown, and Raymond Doesn’t Add Up. New York: The New Press. Robertson. 2012. “Life Satisfaction, Mental Well- Gammage, Sarah, and Rekha Mehra. 1999. “Trends, Being, and Workplace Characteristics Evidence Countertrends, and Gaps in Women’s Employ- from Vietnam, Jordan, and Haiti.� Background ment.� World Development 27 (3): 533–50. paper for the WDR 2013. Gershuny, Jonathan. 2000. Changing Times: Work and Demombynes, Gabriel, and Johannes G. Hoogeeven. Leisure in Post Industrial Society. Oxford: Oxford 2007. “Growth, Inequality, and Simulated Poverty University Press. Paths for Tanzania, 1992–2002.� Journal of African Giles, John. 2006. “Is Life More Risky in the Open? Economies 16 (4): 596–628. Household Risk-Coping and the Opening of Dercon, Stefan, and Catherine Porter. 2011. “A Poor China’s Labor Markets.� Journal of Development ­ Life? Chronic Poverty and Downward Mobility in Economics 81 (1): 25–60. Rural Ethiopia, 1994 to 2004.� In Why Poverty Per- Giles, John, Dimitris Mavridis, and Firman Witoelar. sists: Poverty Dynamics in Asia and Africa, ed. Bob 2012. “Subjective Well-Being, Social Cohesion, Baulch, 65–95. Cheltenham, U.K.: Edward Elgar and Labor Market Outcomes in Indonesia.� Back- Publishing. ground paper for the WDR 2013. Jobs and living standards    95 Giles, John, and Irina Murtazashvili. 2010. “A Control Ianchovichina, Elena, and Susanna Lundstrom. 2009. Function Approach to Estimating Dynamic Pro- “What Is Inclusive Growth?� World Bank, Wash- bit Models with Endogenous Regressors, with an ington, DC. Processed. Application to the Study of Poverty Persistence in IFC (International Finance Corporation). Forthcom- China.� Policy Research Working Paper Series ing. Job Study. Washington, DC: IFC. 5400, World Bank, Washington, DC. IFC (International Finance Corporation) and ILO Giles, John, and Kyeongwon Yoo. 2007. “Precaution- (International Labour Organization). 2011. “Base- ary Behavior, Migrant Networks and Household line Data Collection: Better Work.� IFC and ILO, Consumption Decisions: An Empirical Analysis Washington, DC. Processed. Using Household Panel Data from Rural China.� ILO (International Labour Organization). 2005. Facts Review of Economics and Statistics 89 (3): 534–51. on Safety at Work. Geneva: ILO. Gill, Indermit, and Homi Kharas. 2007. An East Asian ———. 2010. Emerging Risks and New Patterns of Renaissance: Ideas for Economic Growth. Washing- Prevention in a Changing World of Work. Geneva: ton, DC: World Bank. ILO. Glewwe, Paul W., Michele Gragnolatti, and Hassan ———. 2011. Growth, Employment and Decent Work Zaman. 2002. “Who Gained from Vietnam’s in the Least Developed Countries. Geneva: ILO. Boom in the 1990s?� Economic Development and Inchauste, Gabriela. 2012. “Jobs and Transitions Out Cultural Change 50 (4): 773–92. of Poverty: A Literature Review.� Background pa- Graham, Carol. 2008. “Measuring Quality of Life in per for the WDR 2013. Latin America: What Happiness Research Can Inchauste, Gabriela, Sergio Olivieri, Jaime Saavedra- (and Cannot) Contribute.� IDB Working Paper Chanduvi, and Hernan Winkler. 2012. “Decom- 549, Inter-American Development Bank, Wash- posing Recent Declines in Poverty: Evidence from ington, DC. Bangladesh, Peru, and Thailand.� Background pa- Hanson, Gordon H. 2010. “Why Isn’t Mexico Rich?� per for the WDR 2013. Journal of Economic Literature 48 (4): 987–1004. ISSP (International Social Survey Programme). 2005. Hatløy, Anne, Tewodros Kebede, Huafeng Zhang, and “Module on Work Orientation.� ISSP, Cologne. Ingunn Bjørkhaug. 2012. “Perception of Good Kakwani, Nanak, and Jacques Silber. 2008. Quantita- Jobs: Sierra Leone.� Background paper for the tive Approaches to Multidimensional Poverty Mea- WDR 2013. surement. Basingstoke, U.K.: Palgrave Macmillan. Helliwell, John, and Robert Putnam. 2004. “The So- Kebede, Tewodros, Anne Hatløy, Huafeng Zhang, cial Context of Well-Being.� Philosophical Transac- and Ingunn Bjørkhaug. 2012. “Perception of tions of the Royal Society B 359: 1435–46. Good Jobs: Egypt.� Background paper for the Himanshu, Ishan Bakshi, and Camille Dufour. 2011. WDR 2013. “Poverty, Inequality and Mobility in Palanpur: Some Preliminary Results.� Asia Research Centre King, Gary, Christopher J. L. Murray, Joshua A. Salo- Working Paper 45, Asia Research Centre, London mon, and Ajay Tandon. 2004. “Enhancing the School of Economics and Political Science, London. ­ Validity and Cross-Cultural Comparability of Himanshu, Peter Lanjouw, Abhiroop Mukhopadhyay, Measurement in Survey Research.� American Po- and Rinku Murgai. 2011. “Non-Farm Diversifica- litical Science Review 98 (1): 191–207. tion and Rural Poverty Decline: A Perspective Krueger, Alan B., Daniel Kahneman, Claude Fischler, from Indian Sample Survey and Village Study David Schkade, Norbert Schwarz, and Arthur A. Data.� Asia Research Centre Working Paper 44, Stone. 2009. “Time Use and Subjective Well-Being Asia Research Centre, London School of Econom- in France and the U.S.� Social Indicators Research ics and Political Science, London. 93 (1): 7–18. Hintermann, Beat, Anna Alberini and Anil Markan- Lanjouw, Jean O., and Peter Lanjouw. 2001. “The Ru- dya. 2010. “Estimating the Value of Safety with La- ral Non-Farm Sector: Issues and Evidence from bour Market Data: Are the Results Trustworthy?� Developing Countries.� Agricultural Economics 26 Applied Economics 42 (9): 1085–100. (1): 1–23. Hongqin, Chang, Fiona MacPhail, and Xiao-yuan Lanjouw, Peter, and Rinku Murgai. 2009. “Poverty Dong. 2011. “The Feminization of Labor and the Decline, Agricultural Wages, and Nonfarm Em- Time-Use Gender Gap in Rural China.� Feminist ployment in Rural India: 1983–2004.� Agricultural Economics 17 (4): 93–124. Economics 40 (2): 243–63. Huff, W. G. 1994. The Economic Growth of Singapore: Lawson, David, Andy McKay, and John Okidi. 2006. Trade and Development in the Twentieth Century. “Poverty Persistence and Transitions in Uganda: A New York: Cambridge University Press. Combined Qualitative and Quantitative Analysis.� ———. 1995. “What Is the Singapore Model of Journal of Development Studies 42 (7): 1225–51. Economic Development?� Cambridge Journal of Loayza, Norman, and Claudio Raddatz. 2010. “The Economics 19 (6): 735–59. Composition of Growth Matters for Poverty Al- 96   WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 leviation.� Journal of Development Economics 93 Latin America and the Caribbean. Washington, (1): 137–51. DC: World Bank. Lohano, Hari Ram. 2011. “Poverty Dynamics in Rural Quisumbing, Agnes R. 2011. “Poverty Transitions, Sindh, Pakistan, 1987–88 to 2004–05.� In Why Shocks and Consumption in Rural Bangladesh, Poverty Persists: Poverty Dynamics in Asia and 1996–97 to 2006–07.� In Why Poverty Persists: ­Africa, ed. Bob Baulch, 145–86. Cheltenham, U.K.: Poverty Dynamics in Asia and Africa, ed. Bob Edward Elgar Publishing. Baulch, 29–64. Cheltenham, U.K.: Edward Elgar Lundin, Andreas, and Tomas Hemmingsson. 2009. Publishing. “Unemployment and Suicide.� The Lancet 374 Ramessur, Taruna Shalini. 2009. “Economic Cost of (9686): 270–1. Occupational Accidents: Evidence from a Small Mansuri, Ghazala, Slesh Shrestha, Hernan Winkler, Island Economy.� Safety Science 47 (7): 973–9. and Monica Yanez-Pagans. 2012a. “A Plot of My Ramey, Valerie A., and Neville Francis. 2009. “A Cen- Own: Land Titling and Economic Mobility in tury of Work and Leisure.� American Economic Rural Uganda.� Background paper for the WDR Journal: Macroeconomics 1 (2): 189–224. 2013. Ravallion, Martin. 2001. “Growth, Inequality and ———. 2012b. “Health or Wealth? Income Earner Poverty: Looking Beyond Averages.� World Devel- Death and Economic Mobility in Rural Pakistan.� opment 29 (11): 1803–15. Background paper for the WDR 2013. ———. 2004. “Pro-Poor Growth: A Primer.� Policy Martins, Pedro. 2012. “Growth, Employment, and Research Working Paper Series 3242, World Bank, Poverty in Africa: Tales of Lions and Cheetahs.� Washington, DC. Background paper for the WDR 2013. ———. 2011. “A Comparative Perspective on Poverty Moosa, Imad A. 2012. “A Cross-Country Comparison Reduction in Brazil, China and India.� World Bank of Okun’s Coefficient.� Journal of Comparative Research Observer 26 (1): 71–104. Economics 24 (3): 335–56. Ravallion, Martin, and Shaohua Chen. 2007. “China’s Murphy, Gregory C., and James A. Athanasou. 1999. (Uneven) Progress against Poverty.� Journal of De- “The Effect of Unemployment on Mental Health.� velopment Economics 82 (1): 1–42. Journal of Occupational and Organizational Psy- Ravallion, Martin, Shaohua Chen, and Prem San- chology 72 (1): 83–99. graula. 2009. “Dollar a Day Revisited.� World Bank Narayan, Deepa, Lant Pritchett, and Soumya Kapoor. Economic Review 23 (2): 163–84. 2009. Success from the Bottom Up. Vol. 2 of Moving Republic of Korea. 2010. National Employment Strat- Out of Poverty. New York: Palgrave Macmillan; egy 2020 for the Balance of Growth, Employment Washington, DC: World Bank. and Welfare. Seoul: Republic of Korea. Ngai, Rachel L., and Christopher A. Pissarides. 2008. Roy, Dutta Sudipta. 2004. “Employment Dynamics in “Trends in Hours and Economic Growth.� Review Indian Industry: Adjustment Lags and the Impact of Economic Dynamics 11 (2): 239–56. of Job Security Regulations.� Journal of Develop- Nordhaus, William D., and James Tobin. 1973. “Is ment Economics 73 (1): 233–53. Growth Obsolete?� In The Measurement of Eco- Sen, Amartya, and Geoffrey Hawthorn. 1987. The nomic and Social Performance, ed. Milton Moss, Standard of Living. Cambridge, U.K.: Cambridge 509–64. Cambridge, MA: National Bureau of Eco- University Press. nomic Research. Steckel, Richard H. 1995. “Stature and the Standard OECD (Organisation for Economic Co-operation of Living.� Journal of Economic Literature 33: and Development). 2011. How’s Life? Measuring 1903–40. Well-Being. Paris: OECD. Stiglitz, Joseph E. 1996. “Some Lessons from the East Origo, Federica, and Laura Pagani. 2009. “Flexicurity Asian Miracle.� World Bank Research Observer 11 and Job Satisfaction in Europe: The Importance (2): 151–77. of Perceived and Actual Job Stability for Well- Streeten, Paul. 1979. “Basic Needs: Premises and Being at Work.� Labour Economics 16 (5): 547–55. Promises.� Journal of Policy Modeling 1: 136–46. Otsuka, Keijiro, Jonna P. Estudillo, and Yasuyuki Stuckler, David, Sanjay Basu, Marc Suhrcke, and Sawada, eds. 2009. Rural Poverty and Income Dy- Adam Coutts. 2009a. “The Public Health Effect of namics in Asia and Africa. New York: Routledge. Economic Crises and Alternative Policy Responses Ouweneel, Piet. 2002. “Social Security and Well-Being in Europe: An Empirical Analysis.� The Lancet 374 of the Unemployed in 42 Nations.� Journal of Hap- (9686): 315–23. piness Studies 3 (2): 167–92. Stuckler, David, Sanjay Basu, Marc Suhrcke, and Mar- Paes de Barros, Ricardo, Francisco H. G. Ferreira, José tin McKee. 2009b. “The Health Implications of R. Molinas Vega, and Jaime Saavedra Chanduvi. Financial Crisis: A Review of the Evidence.� Ulster 2009. Measuring Inequality of Opportunities in Medical Journal 78 (3): 142–5. Jobs and living standards    97 Stutzer, Alois, and Rafael Lalive. 2004. “The Role of Winefield, Anthony H. 2002. “The Psychology of Un- Social Work Norms in Job Searching and Subjec- employment.� In Social, Developmental, and Clini- tive Well-Being.� Journal of the European Economic cal Perspectives, ed. Claes von Hofsten and Lars Association 2 (4): 696–719. Bäckman. Vol. 2 of Psychology at the Turn of the Takahashi, Kazushi, and Keijiro Otsuka. 2009. “The Millennium. New York: Psychology Press. Increasing Importance of Nonfarm Income and Winkelmann, Rainer, and Liliana Winkelmann. 1998. the Changing Use of Labor and Capital in Rice “Why Are the Unemployed So Unhappy? Evi- Farming: The Case of Central Luzon, 1979–2003.� dence from Panel Data.� Economica 65: 1–15. Agricultural Economics 40 (2): 231–42. Woolard, Ingrid, and Stephan Klasen. 2005. “Deter- UNDP (United Nations Development Programme). minants of Income Mobility and Household Pov- 1990. Human Development Report 1990. New erty Dynamics in South Africa.� Journal of Devel- York: UNDP. opment Studies 41 (5): 865–97. WHO (World Health Organization) and World Bank. World Bank. 1993. The East Asian Miracle. Washing- 2011. World Report on Disability. Washington, DC: ton, DC: World Bank. WHO and World Bank. ———. 2010. From Global Collapse to Recovery: Eco- Wietzke, Frank-Borge, and Catriona McLeod. 2012. nomic Adjustment and Growth Prospects in Latin “Jobs, Well-Being, and Social Cohesion: Evidence America and the Caribbean. Washington, DC: from Value and Perception Surveys.� Background World Bank. paper for the WDR 2013. Zhang, Huafeng, Ingunn Bjørkhaug, Anne Hatløy, Wilkinson, Richard, and Michael Marmot. 1998. and Tewodros Kebede. 2012. “Perception of Good ­ Social Determinants of Health: The Solid Facts. Jobs: China.� Background paper for the WDR Geneva: World Health Organization. 2013. CHAPTER 3 Jobs and productivity Reallocation from low- to high-productivity jobs matters more for growth in developing countries, where differences in productivity across sectors and within sectors are wide. But reallocation often amounts to little more than churning. P roductivity growth happens as jobs be- regions, however. In Sub-Saharan Africa, the come more productive, as new high- Green Revolution has not taken place on a large productivity jobs are created, and as low- scale. productivity jobs disappear. In the medium Outside of agriculture, productivity varies term, trends in employment align closely with substantially across enterprises, implying po- trends in the labor force, so there is no such a tentially large productivity gains from job real- thing as jobless growth. But the short-term rela- location. The speed at which productivity grows tionship between employment and growth is also varies. Large firms are more innovative, more complex. Large numbers of jobs are being provided that they are exposed to competition. created and destroyed simultaneously, leading At the other end, microenterprises are a diverse to structural change and spatial labor realloca- group. A vast majority of them, more prone to tion. Underneath these sectoral and spatial churning than to growth, are a means of survival changes are firm dynamics that result in a con- for the poor. Yet some are entrepreneurial, and stant restructuring and reallocation of resources, their success could boost wage employment. including labor. In developing countries, many people work in very small and not so dynamic economic Employment turbulence, units. Family farms, which often predominate not jobless growth in agriculture, average only 1.2 hectares in Asia, and 1.8 hectares in Sub-Saharan Africa. Outside Jobless growth is a popular notion, often be- of agriculture, microenterprises and household lieved to be grounded on data. However, unem- businesses account for a large share of employ- ployment rates neither explode nor vanish over ment in a majority of developing countries. time, so employment trends align closely with These businesses make a significant contribution trends in the size of the labor force. The growth to gross job creation and destruction, although of gross domestic product (GDP) certainly mat- not necessarily to net job creation and produc- ters for employment growth, but in the medium tivity growth. term it matters less than demographics and par- In agriculture, the Green Revolution has ticipation rates. Data from 97 countries over the led to higher cereal yields and to employment past decade confirm that a positive relationship growth because the new technologies are labor exists between the growth of GDP per capita and intensive. The progress has been uneven across the growth of employment per capita (figure Jobs and productivity    99 BOX 3.1  What drives economic growth?  At the risk of simplifying, four main forces lie behind increases in an measured as changes in total factor productivity. Technological economy’s per capita output. The first is the use of more capital per progress amounts to combining capital, labor, and skills more effi- unit of labor. The second is an increase in the number of people ciently, while applying new knowledge. working, relative to the total population. This happens when fertility Growth decomposition quantifies the contribution of each of declines and the share of adults in the total population increases; these four forces to economic growth. It can be done for any partic- it is also happens when women shift their work from household ular country given sufficient data on gross domestic product, capi- chores to income-generating activities. The third mechanism tal, employment, and human capital. Some of these variables may through which output can grow is by making people themselves need to be constructed or approximated; for instance, the stock of more productive. The acquisition of skills, also known as human capital in an economy is estimated based on accumulated invest- capital accumulation, allows a person to do more using the same ments, while human capital is approximated by the educational amount of capital. The fourth mechanism is technological progress, attainment of its population, corrected for the quality of education. Source: World Development Report 2013 team. 3.1a).1 The relationship is not very strong, but deed associated with a decline in employment in only in very few cases was growth truly jobless. the same year (figure 3.1b). Only in subsequent On the other hand, the short-term relation- years did this negative employment effect wane. ship between growth and employment is not Productivity growth is a turbulent pro- so straightforward. Growth happens partly cess. Analyses covering economies as different through the disappearance of low-productivity as Ethiopia and the United States in different jobs as well as through the creation of more periods over the past three decades reveal the productive jobs.2 So for the same sample of magnitude of gross job creation and gross job countries over the same decade, surges in total destruction (figure 3.2). In the manufacturing factor productivity (TFP) in one year were in- sector of developing countries, between 7 and Economic F I G U R E 3 .1   growth does not occur at the expense of jobs in the medium term a. Annual growth of GDP per capita b. Annual growth of TFP and the employment rate and the employment rate 4 4 growth of employment per capita, % correlation between TFP growth and growth of employment per capita, % 3 2 0 2 –2 1 –4 0 –6 –1 –8 –2 –10 –3 –12 –4 0 4 8 12 16 in the one year two years three years same year later later later growth of GDP per capita, % Source: World Development Report 2013 team estimates based on average growth decomposition accounting for years 1999–2009. Note: GDP = gross domestic product; TFP = total factor productivity. Data are from 97 countries. Panel a presents the relationship between annual growth of GDP per capita and the growth of the employment-to-population ratio. Each dot represents a country. Panel b depicts the correlation between annual growth in total factor productivity (TFP) and employment rate growth in the same year, and in subsequent years. 100  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 20 percent of jobs are created every year, while Simultaneous F I G U R E 3 . 2   job creation and destruction a similar proportion disappear.3 Even when characterize all economies aggregate employment was declining, as in the 1990s in Romania and Slovenia and in the manufacturing sector of República Bolivariana net job gross job gross job creation creation destruction de Venezuela, many new jobs were being cre- ECONOMY-WIDE ated. Conversely, when aggregate employment was growing by 6 percent in Mexico, jobs were Latvia disappearing at almost twice that rate. Job flows may be associated with profound Mexico transformations in the sectoral structure of the economy. Technological change often occurs for Argentina specific products and processes, causing pro- ductivity to grow at different paces in different Estonia sectors. However, the relative weight of different sectors in the economy is determined not only Hungary by technological progress, but also by market demand and nonmarket forces. When there is Slovenia an expansion of the most productive sectors, ag- gregate productivity increases. This composition Romania effect, called productivity-enhancing structural change, is well documented in the case of labor industrial economies shifts from agriculture to industry and services. (average) Analysis based on more disaggregated data sug- MANUFACTURING SECTOR ONLY gests that reallocation of labor across sectors has also been an important driver of productivity Ethiopia growth in several fast-growing East Asian coun- tries. In China, it contributed 4.1 percentage Indonesia points of the 7.3 percent annual growth in ag- gregate labor productivity over the past decade; Brazil in Vietnam, it accounted for 2.6 points out of 4.2 (figure 3.3).4 Chile Job flows are also associated with changes in the spatial distribution of employment. Taiwan, China The structural shift from agriculture in rural areas to industry and services in towns and Colombia cities may be the most visible example of spa- tial labor reallocation, but it is not the only Venezuela, RB one.5 Even within sectors, job flows often have a strong spatial dimension. New plants associ- industrial economies (average) ated with more innovative activities tend to start in large, diversified cities—incubators—with –5 0 5 10 15 20 a higher density of suppliers and labor, and share of total employment, % more fluent exchanges of information. As they mature and become more self-sufficient in in- formation, these plants move to smaller cities, Sources: World Development Report 2013 team estimates based on Bartelsman, Haltiwanger, and where land and wage costs are lower. As a result, Scarpetta 2009b and Shiferaw and Bedi 2010. Note: The figure shows annual job flows. Data are from Argentina (1996–2001); Brazil (1997–2000); many medium and small cities tend to be more Canada (1984–97); Chile (1980–98); Colombia (1983–97); Estonia (1996–2000); Ethiopia (1997–2007); specialized.6 Finland (1989–97); France (1989–97); Germany (1977–99); Hungary (1993–2000); Indonesia (1991–94); Italy The extent of spatial relocation varies across (1987–94); Latvia (1983–98); Mexico (1986–2000); the Netherlands (1993–95); Portugal (1983–98); Romania (1993–2000); Slovenia (1991–2000); Taiwan, China (1986–91); the United Kingdom (1982–98); the United sectors and countries. For instance, in the Re- States (1986–91, 1994–96); and República Bolivariana de Venezuela (1996–98). public of Korea, manufacturing dominance in Jobs and productivity    101 Labor F I G U R E 3 . 3   reallocation across sectors was a driver of productivity growth in East Asia 8 7 annual labor productivity growth, % 6 5 4 3 2 1 0 –1 M ica th no s th Eu s t a Ca ope n ca As uth ca a c a on m In olia T sia SA land an ina a . Pa e n a ilip a s Si Rep er Sou eco ie dl and ast ie ne an Asi in Ko hin di Ph aysi or Su No bea ta na fri So ri ci ne Am nd rial nom m Ch iw h In kis ap pi Af g r Pa Sa th A Ta R, C ,C ng hai et a, al nd rib do ng n Vi re M d tin al a ust eco ra M r ha E e ia d ll a Ko b- st ng Ea as nt in Ho eE id r changes within sectors reallocation across sectors Ce La Source: World Development Report 2013 team estimates based on Kucera and Roncolato 2012. Note: The figure shows the decomposition of labor productivity growth in 81 economies over 1999–2008 into productivity changes due to changes within sectors and reallocation across sectors. Seven sectors are considered: agriculture, hunting, forestry, and fishing; mining and utilities; manufacturing; construction; trade, restaurants, and hotels; transport, storage, and communication; and other services. The regional growth rates are weighted averages, with weights based on an economy’s share in regional GDP. urban centers has continued for a long period main driver of aggregate productivity growth is of time. Enterprises have been sprawling into firms becoming better at what they do (figure the suburbs of urban centers rather than leap­ 3.4).9 Entry and exit also contribute, which in- frogging to different locations as in some in­ dicates that new firms are more productive than dustrial countries (map 3.1). Similar shifts are those exiting. In general, exiting firms see their happening in Brazil, China, and Vietnam.7 In productivity decline before they close, whereas India, large manufacturing enterprises are mov- new firms tend to attain the average levels of ing away from urban centers into rural loca- productivity of their industry within five years. tions.8 Regardless of the nature of the shift, al- These complex dynamics imply that at any point most inevitably jobs are created in some places in time, firms with very different productiv- and destroyed in others. ity levels coexist, even within narrowly defined Underneath these sectoral and spatial changes industries.10 in the structure of employment are the firm The popular perception that productivity dynamics connecting job flows and productiv- grows through downsizing at the firm level is ity growth. Aggregate productivity grows when partially supported by these analyses. Indeed, existing firms become better at what they do, in many cases, employment tends to shrink in when more productive firms enter the market, firms experiencing rapid productivity growth.11 and when less productive ones exit. It also grows But downsizing is only part of the story. There when more productive firms become bigger and are also many successful upsizing firms, achiev- less productive ones become smaller. Decompo- ing both productivity growth and employment sition analyses show that in most countries the growth.12 For instance, one-fourth of manufac- 102  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 REPUBLIC OF KOREA MANUFACTURING ACTIVITIES ARE SPRAWLING INTO SUBURBS OF URBAN CORE IN KOREA EMPLOYMENT SHARES OF MANUFACTURING INDUSTRIES CITIES AND TOWNS COUNTY BOUNDARIES Manufacturing activities are sprawling out of the main urban centers in the Republic of Korea PROVINCE CAPITALS PROVINCE BOUNDARIES M A P 3 .1   40% 30% NATIONAL CAPITAL This map was produced by the Map Design Unit of The World Bank. 20% The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of The World Bank 10% Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. IBRD 39533 1960 DEMOCRATIC PEOPLE'S REPUBLIC OF KOREA Sea of 1985 DEMOCRATIC PEOPLE'S REPUBLIC OF KOREA Sea of 2005 DEMOCRATIC PEOPLE'S REPUBLIC OF KOREA SEPTEMBER 2012 Japan Japan Sea of EMPLOYMENT Japan SHARES OF SEOUL SEOUL SEOUL MANUFACTURING Incheon INDUSTRIES CITIES AND TOWNS 37°N 37°N 37°N 40% PROVINCE CAPITALS 30% NATIONAL CAPITAL Yellow Yellow Yellow Sea Sea Sea 20% Taegu Taegu COUNTY BOUNDARIES 36°N 36°N 36°N 10% PROVINCE BOUNDARIES Pusan Pusan map Thismap This was was produced by the by produced Map the Map Design Design Unit Unit Bank. of The World of The World Bank. 35° N 35° N 35° N boundaries, Theboundaries, The colors, colors, denominations denominations and any otherand any other information information shown on this map do not imply, on the part of The World shown on this map do not imply, on the part of The World Bank Bank Group, Group, any judgment any on theon judgment legal status the of any legal territory, status or any of any territory, or any endorsement or acceptance of such boundaries. Korea Strait Korea Strait Korea Strait endorsement or acceptance of such boundaries. 34°N 34°N 34°N 0 30 60 Kilometers 0 30 60 Kilometers 0 30 60 Kilometers 126°E 127°E 128°E 129°E 126°E 127°E 128°E 129°E 126°E 127°E 128°E 129°E Source: Park et al. 2011. Note: The maps show employment shares of manufacturing industries at the city or county level for 1960, 1985, and 2005. FI G U R E 3.4  Efficiency gains at the firm level are the main driver of productivity growth 14 12 annual labor productivity growth, % 10 8 6 4 2 0 –2 –4 Ge p. Po y l ia a Po e ite gal Ge es y Ki ds m nd RB il Co ile Ar bia a na ia Sl ia Ko enia ga an ni an az in c tv Re tv n do at an n Ch hi la to rtu nt to u m a, rm Br La rm la La St ov t ,C n a, ng el Fr Es lo Es r ge er Fi re zu d an th ne iw Ne d Un ite Ve Ta Un manufacturing services within between cross entry and exit Source: Based on Bartelsman, Haltiwanger, and Scarpetta 2009b. Note: The figure decomposes annual labor productivity growth. Data for industrial countries are from France (1990–95), the Netherlands (1992–2001), the United States (1992,1997), Portugal (1991–1994), the United Kingdom (2000, 2001), and Germany (2000–02). Data for developing economies are from Argentina (1995–2001); Brazil (2001); Chile (1985–99); Colombia (1987–98); Estonia (2000, 2001); the Republic of Korea (1988, 1993); Slovenia (1997–2001); Taiwan, China (1986, 1991, 1996); and República Bolivariana de Venezuela (1999). Within captures the changes at the firm level, between the changes in employment shares across firms, cross the interaction between the former two, and entry and exit the opening and closure of firms. Jobs and productivity    103 FI G U R E 3. 5  Efficiency gains and employment growth can go together 100 90 80 70 60 share, % 50 40 30 20 10 0 no. of employment value no. of employment value no. of employment value plants added plants added plants added Chile (2001–06) Ethiopia (2005–08) Romania (2000–05) successful upsizers successful downsizers unsuccessful downsizers unsuccessful upsizers Source: World Development Report 2013 team based on Amadeus Database, Bureau van Dijk, Amsterdam, and Ethiopia Large and Medium Scale Manufacturing and Electricity Industries Survey, Central Statistical Agency, Addis Ababa. Note: no. = number. The figure shows the contribution of each of the four groups to the annual growth rate of the variable of interest. The plants considered employ at least 10 workers and exist throughout the entire period. Successful upsizers are plants that increased both labor productivity and employment, successful downsizers are plants that in- creased productivity but reduced employment, unsuccessful downsizers are plants that reduced employment and productivity, and unsuccessful upsizers are plants that increased employment but reduced productivity. turing plants operating in Chile over 2001–06 be successful upsizers.13 Recent evidence based were successful downsizers, consistent with the on 26,000 manufacturers from 71 countries fur- popular perception. But another one-fourth ther shows that firms that innovated in products were successful upsizers, achieving both pro- or processes were more likely to be successful up- ductivity and employment growth (figure 3.5). sizers; they not only attained higher total factor More important, the successful upsizers con- productivity than noninnovative firms; they also tributed more to production, employment, and exhibited higher employment growth.14 aggregate productivity growth than the success- Transition economies in Europe illustrate ful downsizers. Results were similar in Romania the links between job reallocation and pro- between 2000 and 2005, and in Ethiopia between ductivity growth. Before reforms were imple- 2005 and 2009. While country experiences vary, mented, these economies suffered from large having a critical mass of successful upsizers is not distortions caused by a rigid planning system, uncommon. which prevented resources from flowing to Across countries, successful upsizers in man- more efficient uses. Liberalization led to mas- ufacturing industries tend to be younger, leaner, sive downsizing and job losses. Eventually, it also and more innovative. Among survivors in the strengthened incentives, mobility, and markets, same industry and region, younger firms were opening up space for more productive private more likely to be upsizers in Chile over 2001–06, companies. The entry of these dynamic players and successful upsizers in Romania over 2000– contributed between 20 and 50 percent of total 05. In all three countries, survivors employing labor productivity growth in the late 1990s. The fewer than 20 employees tended to upsize fewer exit of obsolete firms released resources that rather than downsize. In Romania, survivors in- could be used more effectively by new or exist- vesting more in capital per worker also tended to ing firms. Although lack of experience and small 104  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 size often made the new firms less productive census reported 1.33 million manufacturing than the average firms of more advanced coun- firms with annual sales above RMB 5 million.16 tries, these new firms were more efficient than Most of them were private. The entry of these domestic incumbents. They played a strong role new businesses and the closure of nonviable in boosting productivity in medium- and high- state-owned enterprises (SOEs) accounted for technology industries and in exerting competi- two-thirds of TFP growth in manufacturing tive pressure on existing firms.15 sectors over 1998–2006.17 China’s rapid productivity growth was also underpinned by large-scale reallocation. Beginning in 1978, economic reform efforts Most jobs are in very small farms gradually expanded the influence of markets and firms and deepened global integration. All of this created unprecedented opportunities for the Many people in developing countries work formation of private entities, including town- in very small and not very dynamic economic ship and village enterprises, and the entry of units—family farms, microenterprises, and foreign companies. During the first decade of household businesses. Although microenter- reform (1978–88), reallocation from agricul- prises are often defined as firms employing ten ture to nonagriculture activities was the source or fewer workers, many among them are actu- of almost half of all productivity growth. In the ally one-person businesses. Given their contri- following decades, however, the main drivers bution to total employment, these small eco- of productivity growth were labor reallocation nomic units cannot be ignored. Understanding out of the state sector, private sector vibrancy, their dynamics is crucial to deciphering the rela- and state sector restructuring. The scale of tionship between jobs and productivity. business entry was startling: the number of In family farms, hired labor is usually em- industrial firms rose from 377,000 in 1980 to ployed in simple tasks such as weeding and har- nearly 8 million in 1996. The 2004 economic vesting, whereas family labor usually carries out care-intensive activities such as water manage- ment, land preparation, and fertilizer applica- Smallholder F I G U R E 3 . 6   farming is dominant tion. Limited reliance on hired labor is due to the difficulty to monitor effort, and without outside Latin America machinery, farms cannot be expanded beyond Europe and the size manageable by the family’s labor, which Central Asia is typically 1 to 2 hectares.18 Family farms domi- 100 4.8 ha Asia nate even in high-income countries, and owner 1.2 ha Sub-Saharan Africa cultivation is the most common form of land 1.8 ha tenure, especially in Asia (figure 3.6). 80 Latin America Middle East and 61.5 ha At 1.2 and 1.8 hectares, respectively, average farms owned, % North Africa 3.2 ha farm size is small in both Asia and Sub-Saharan 60 Africa.19 In Asia, farmers typically own land plots, which they supplement through tenancy contracts that facilitate transfers from relatively 40 land-abundant to relatively labor-abundant households. Farms in Sub-Saharan Africa are slightly larger than in Asia, but size and the im- 20 portance of owner farming are becoming similar 0 5 10 15 20 25 30 35 40 in both regions.20 There are some exceptions to farms rented, % this pattern. Colonial governments created large farms in some developing countries, such as es- Source: FAO 2010. Note: ha = hectare. Countries in Asia include: Bangladesh, India, Indonesia, the Lao People’s Democratic tates in southern Africa, haciendas in Latin Republic, Nepal, Pakistan, and Thailand; in Europe and Central Asia: Azerbaijan, Croatia, the Czech Republic, America and the Philippines, and plantations in Georgia, the Kyrgyz Republic, Latvia, Serbia, Slovenia, and Turkey; in Latin America and the Caribbean: Brazil, the Caribbean. In Sub-Saharan Africa, large ar- Guatemala, Nicaragua, St. Lucia, St. Vincent and the Grenadines, Uruguay, República Bolivariana de Venezu- ela, and Virgin Islands (United States); in the Middle East and North Africa: the Arab Republic of Egypt, Jordan, eas are also held as customary land—owned col- Morocco, Qatar, Saudi Arabia, and Tunisia; and in Sub-Saharan Africa: Côte d’Ivoire, Ethiopia, and Madagascar. lectively by extended families, clans, or lineage.21 Jobs and productivity    105 The F I G U R E 3 . 7   employment share of microenterprises is greater in developing countries Ethiopia Egypt, Arab Rep. India Bolivia Colombia Ghana Mexico Venezuela, RB Argentina Poland Turkey Hungary South Africa Uruguay Czech Republic Slovenia Chile Romania Vietnam industrial countries (average) 0 20 40 60 80 100 share of employment, % manufacturing sector services sector Sources: World Development Report 2013 team estimates based on International Income Distribution Database (I2D2) and EUROSTAT. Note: Microenterprises are firms, formal or informal, with fewer than 10 workers. Data for developing countries are from Argentina (2006–10); Bolivia (2005, 2007); Chile (2006, 2009); Colombia (2009); the Czech Republic (2005–07); the Arab Republic of Egypt (2006); Ethiopia (1999); Ghana (1991); Hungary (2007–08); India (2004, 2009); Mexico (2004–10); Poland (2005–07); Romania (2005–07); Slovenia (2005–07); South Africa (2005–07); Turkey (2006–10); Uruguay (2009); República Bolivariana de Venezuela (2004–06); and Vietnam (2009). Data for industrial countries are from Austria, Belgium, Denmark, Finland, France, Germany, Greece, Italy, Luxembourg, the Netherlands, Norway, Portugal, Spain, Sweden, and the United Kingdom over 2005–07. Outside of agriculture, microenterprises It is often claimed that most employment, and and household businesses are dominant. More most job creation, is associated with small and than 80 percent of registered manufacturing medium enterprises, but that is generally not true establishments in Argentina, Bolivia, El Salva- in developing countries. In reality, micro- and dor, and Mexico have fewer than 10 workers.22 small enterprises account for the bulk of employ- About 90 percent of manufacturing establish- ment, even in middle-income countries (figure ments employ 5 to 49 workers in China; India; 3.7). Their share is often underestimated, because Indonesia; Korea; the Philippines; and Taiwan, economic censuses and plant-level surveys rarely China.23 The share of microenterprises is even cover the informal segment of the economy, higher outside manufacturing, reaching 94 where businesses are especially small. But data percent in the services sector of Mexico and 98 from household and labor force surveys that are percent in all modern sectors in Tunisia.24 In representative of the entire population provide several African and Latin American countries, a different picture. These small enterprises play the majority of informal enterprises consist of significant role in employment in manufactur- 1- to 3-person businesses.25 ing. They account for 97 percent of employment 106  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 BOX 3.2  Microenterprises account for most job creation and destruction  In some household and labor force surveys, employees are asked to The distribution of job flows by firm size that emerges from report the size of the firm they work for, or the size of their own busi- the manufacturing survey can be adjusted based on the distri- ness if they are self-employed. This information can be used to esti- bution of employment by plant size from Chile’s household survey, mate the distribution of employment by plant size. This distribution the National Socio­ economic Characterization Survey (Encuesta de can in turn be used to correct for the omission of informal enter- Caracterización Socioeconómica Nacional, or CASEN). Before this prises in an economic census or plant-level survey. adjustment, larger firms seem to account for most job creation and This approach was applied to Chile’s manufacturing survey, the destruction. But the adjustment shows that microenterprises con- Annual National Industrial Survey (Encuesta Nacional Industrial tribute about 80 percent of gross job flows. This estimate should not Anual), which covers more than 90 percent of employment among be taken literally, because the microenterprises for which informa- establishments with 50 workers or more, but less than half the tion on job creation and destruction is available are not necessarily employment in establishments with 10 to 49 workers. Nearly representative—the Chilean census started to include microenter- 300,000 workers in microenterprises are omitted from the survey; prises only in the late 1990s. But even with a margin of error, the 250,000 of them work in firms with fewer than 5 employees. estimate is so large that it changes the picture of job creation and job destruction. a. Contribution by rm size before correcting, 2000–06 b. Contribution by rm size after correcting, 2000–06 80 80 60 60 percent percent 40 40 20 20 0 0 0–5 6–9 10–49 50–199 > = 200 0–5 6–9 10–49 50–199 > = 200 rm employment size rm employment size shares in job creation, % shares in job destruction, % shares in job creation, % shares in job destruction, % shares in employment, % shares in employment, % Source: World Development Report 2013 team. in the manufacturing sector in Ethiopia and 39 yields per hectare tend to be higher in smaller percent in Chile. In services sectors, their role is farms, because family farms apply more labor often more important. Even in Eastern European per unit of land, even though they apply fewer countries, where private sector entry is only two purchased inputs. This inverse relationship be- decades old, microenterprises account for 10 to tween farm size and productivity was first ob- 20 percent of employment in manufacturing and served in South Asia.27 But it has also emerged for 30 to 50 percent of employment in services. in Sub-Saharan Africa as population pressure Micro- and small enterprises also play a critical on the land has led to agricultural intensifica- role in job creation and destruction (box 3.2).26 tion. In Kenya, Malawi, Tanzania, and Uganda, a 1 percent increase in farm size is associated with a 0.1 to 0.2 percent reduction in yield.28 In farms, uneven technological The use of family labor per hectare is also in- progress versely correlated with farm size.29 The relationship between crop yields and It is generally assumed that large farms are more farm size emerges, because the larger, more productive. In low-income countries, however, mechanized farms have higher productivity. But Jobs and productivity    107 Crop F I G U R E 3 . 8   yields have diverged vastly across regions 6 China, Republic of Korea, and Japan 5 4 tons per hectare Latin America and the Caribbean 3 2 Southeast Asia and South Asia 1 Sub-Saharan Africa 0 61 64 67 70 73 76 79 82 85 88 91 94 97 00 03 06 09 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 20 Source: FAOSTAT-Agriculture (database), Food and Agriculture Organization, Rome. Note: Figures are weighted averages of yields for wheat, rice, and coarse crops. constraints in land markets usually slow expan- may not be productive in Sub-Saharan Africa, sion and mechanization.30 and high-yielding varieties in irrigated areas The new technologies of the Green Revolu- may be low-yielding in rain-fed areas.35 tion contributed to job creation because they The agricultural growth associated with were labor intensive. Short-statured, fertilizer- the Green Revolution not only creates jobs in responsive, high-yielding varieties of rice and farming but also facilitates the development of wheat were developed by international agri- the nonfarm sectors.36 The adoption of mod- cultural research centers in the late 1960s. The ern technology stimulates the production and varietal improvement of other cereals such as marketing of fertilizer and other purchased maize followed. These varieties, as well as im- inputs. Increased supply of cereals stimulates proved production practices, were quickly dif- the development of food markets and keeps fused, particularly in tropical Asia. The amount the cost of living low for those who migrate to of inorganic fertilizer applied has steadily in- the cities. In addition to these backward and creased over extended periods.31 Thanks to forward links, the increase in farmers’ incomes continual technological improvements and heightens the demand for goods and services.37 sustained adoption, cereal yields have increased Cross-country analyses show that agricultural dramatically for the past several decades.32 growth has resulted in the expansion of non- Progress has been uneven across regions, farm sectors, particularly where the agricul- however (figure 3.8). In Sub-Saharan Africa, tural sector is large.38 there is no evidence to suggest that small farm- ers were slower than larger farmers in adopting the new technologies. But the Green Revolution Among firms, much churning and has not taken place on a large scale, even though few gazelles farmland has been growing scarce because of population pressure on limited c ­ultivable Outside of agriculture, productivity varies sub- areas.33 Feeding growing populations from a stantially across enterprises, implying that job shrinking amount of farm land requires Sub- reallocation could lead to large gains in aggre- Saharan African countries to increase cereal gate productivity. In India, within a narrowly yields.34 Yet improved agricultural technology defined manufacturing industry, a plant at the is location specific: improved varieties in Asia 90th percentile of the TFP distribution gener- 108  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 The F I G U R E 3 . 9   dispersion of productivity in manufacturing is greater in developing countries United States China Argentina Ecuador Chile Bolivia Uruguay El Salvador India Mexico 0 5 10 15 20 25 30 productivity ratio between 90th percentile and 10th percentile of TFP distribution Source: Pagés 2010. Note: TFP = total factor productivity. The figure shows the ratio of TFP among plants between the 90th and the 10th percentiles of the TFP distribution within narrowly defined industries. TFP is measured as physical productivity, as defined by Foster, Haltiwanger, and Syverson (2008). The data cover only the manufacturing sector. Data are from Argentina (2002), Bolivia (2001), Chile (2006), China (2005), Ecuador (2005), El Salvador (2005), India (1994), Mexico (2004), United States (1997), and Uruguay (2005). ates 22 times as much output as a plant at the efficiency tends to suffer. Large SOEs without 10th percentile. In comparison, the estimated foreign competitors are less innovative and pro- ratio is only 9 to 1 in the United States.39 The ductive than other large firms.41 dispersion of TFP is also high in a number of For a given size, young firms are also more Latin American countries (figure 3.9). Detailed likely than old firms to engage in innovative ac- data on nonmanufacturing firms are scarcer. tivities. They also have better growth prospects, But the dispersion of TFP in retail businesses in a finding consistent with evidence from indus- Mexico, and in communication and transporta- trial countries (figure 3.11).42 For example, in tion businesses in Uruguay, is also sizable.40 the 1990s, when China was in the early stages The speed at which productivity grows also of reform, human and financial resources were varies across firms. Large firms are typically concentrated in SOEs. However, the incentive more innovative than small firms. They tend to structure in these enterprises hindered innova- invest more in machinery and hire more edu- tion. In contrast, the new township and village cated workers. They are also more likely than enterprises lacked the resources to adopt new small firms to engage in activities such as devel- technology and import new equipment, but oping new product lines, introducing new tech- they were more flexible in their decision mak- nology, opening and closing plants, outsourc- ing. As a result, these younger firms were more ing, and engaging in joint ventures with foreign dynamic than large SOEs, although they were partners (figure 3.10a). Large firms produce less productive than large and medium private more with a given amount of labor, are more companies.43 likely to export, and tend to export more. They In developing countries, the dispersion of also pay substantively higher wages than micro- productivity and growth prospects across firms and small enterprises (figure 3.10b). They pay a is further widened by the large number of wage premium even controlling for age, educa- micro­ enterprises, many of which are barely tion, and other worker characteristics. Not all more than a means of subsistence for the poor. large firms are innovative, however. When size A majority of these microenterprises have lim- is supported through nonmarket mechanisms, ited capital and often even lack a fixed address. Jobs and productivity    109 F I G U R E 3 .10   Large firms tend to perform better and to pay better than small ones a. Large rms are more productive and innovative labor productivity wage b. Larger rms pay higher wages overall being exporter performance 6 export as share of sales added new product line 4 estimates, % incorporated new technology upgraded existing product line opened a new plant 2 probability to discontinued at least one product engage in innovative entered new joint venture with foreign partner activities entered new licensing agreement 0 0 20 50 80 120 closed at least one existing plant wage premium, relative to microenterprises (%) outsourced a major activity brought in-house a previously outsourced activity small large 0 10 20 30 40 50 60 70 di erence, relative to small rms (%) Source: World Development Report 2013 team based on Ayyagari, Demirgüç-Kunt, and Maksimovic 2011a; and Montenegro and Patrinos 2012 for the World Development Report 2013. Note: Panel a uses World Bank enterprise surveys covering more than 54,000 firms across 102 developing countries over 2006–10 for overall performance, and 19,000 firms across 47 developing countries over 2002–05 for innovative activities. The analysis controls for firm characteristics, industry, and country. In this panel, large firms employ 100 or more workers and small firms fewer than 20 workers. Panel b uses 138 household and labor force surveys spanning 33 countries over 1991–2010 and controls for worker characteristics. In this panel, large firms are those employing more than 50 workers and small firms 10 to 50 workers. Young F I G U R E 3 .11   firms are more likely than old ones to engage in innovative activities incorporated new technology added new product line opened a new plant upgraded existing product line brought in-house a previously outsource activity entered new joint venture with foreign partner outsourced a major activity entered new licensing agreement closed at least one existing plant discontinued at least one product –0.15 –0.1 –0.05 0 0.05 0.1 di erence, relative to rms one year younger (%) Source: World Development Report 2013 team based on Ayyagari, Demirgüç-Kunt, and Maksimovic 2011a. Note: The figure uses World Bank enterprise surveys covering 19,000 firms across 47 developing countries over 2002–05, controlling for firm characteristics, industry, and country. Statistically insignificant estimates are reported as zeroes. 110  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 BOX 3.3  Most microenterprises are in rural areas and engage in commerce  Microenterprises in urban areas, and particularly those in the infor- other manufacturing and food (16 percent) and other services (11.8 mal sector, tend to attract the attention of academics and policy percent). Similarly, the surveys of countries in Africa and in Latin makers. But microenterprises are prominent in rural areas as well. America and the Caribbean suggest that 56 to 74 percent of micro- The surveys of micro- and small enterprises in Africa and in Latin and small firms in urban areas, and 60 to 70 percent in rural areas, America and the Caribbean show that fewer than half are in cities are engaged in commerce. The surveys of household enterprises in and towns with 20,000 inhabitants or more. The urban share reaches Sub-Saharan African countries show similar patterns. These surveys 46 percent in the Dominican Republic but is below 30 percent in all identify three manufacturing activities as the most important across other countries surveyed. Even if rural towns are counted (generally, all countries: textiles and apparel, food and beverages, and wood localities with 2,000 to 20,000 inhabitants), well over half of the and forest products. These three categories account for about 75 enterprises are in strictly rural areas in most countries. percent of manufacturing enterprises in urban areas and nearly 90 The vast majority of microenterprises are engaged in commerce, percent of manufacturing enterprises in rural areas. supporting the conventional view that associates microenterprises Most of the microenterprises operate from home or on the with street vendors and petty traders. But a significant number are street. According to the surveys of household enterprises in Sub- involved in light manufacturing activities. According to the 1-2-3 Saharan Africa, 25 to 45 percent of these microfirms use home as surveys of West African countries, the most important sector in capi- primary point of operation, and 10 to 40 percent of them simply tal cities is petty trading (27.1 percent of all enterprises), followed by work on the street. Sources: Fox and Sohnesen 2012; Grimm, Kruger, and Lay 2011; Liedholm 2002. Many are ­ located in rural areas, absorbing some cumulated human and physical capital while labor slack during the low agricultural season working for a wage or a salary. Operating (box 3.3). Across 18 developing countries, 44 micro­­enterprises is a choice for them. Those percent of the people living on less than US$1 a who achieve higher productivity levels are more day in urban areas, and 24 percent of those in likely to stay in business, grow, and create job rural areas, work in a nonagricultural business. opportunities for others.46 On average, they do not earn much.44 A very small group of microenterprises actu- Nonetheless, these nonfarm activities pro- ally displays a strong performance. This group vide an important channel to diversify income bears similarities with the so-called “gazelles� of for the poor. In nine Sub-Saharan African coun- industrial countries—high-growth companies tries, most nonfarm jobs were generated by whose revenues increase by at least 20 percent households starting businesses, rather than en- annually for four years or more. In industrial tering the rapidly expanding private wage sector. countries, the term “gazelle� is used for com­ Despite being modest, earnings from household panies starting from a revenue base of at least enterprises contribute to consumption much US$1 million, which makes them very big the same as earnings from wage employment by developing-country standards. Nonetheless, do. And these small businesses offer an avenue the same dynamism can be found at a much for poor households to engage in gradually smaller scale. Data from seven Sub-Saharan more productive activities.45 African countries show that the median capital While microenterprises have a lackluster per- stock held by urban informal enterprises is less formance as a group, they are also very diverse. than US$80, whereas the capital stock for those In middle-income countries, a significant share in the top quintile averages US$5,000. The aver- of the owners of micro- and small enterprises age monthly profit of those in the top quintile are as entrepreneurial as their peers in indus- is seven times the median monthly profit. The trial countries. Their weak performance may rate of returns to capital is also relatively high in be driven more by contextual factors such as these firms.47 This heterogeneity among micro­ limited access to credit and policy-induced bar- enterprises suggests they can be an incubator riers to access technologies and markets, than for large and productive firms. by limited capacity. In several Latin Ameri- The dynamism of microenterprises matters can countries, for example, entrants into self- not only for livelihoods but also for productivity employment tend to be workers who have ac- growth. Large firms innovate more, but they are Jobs and productivity    111 Surviving F I G U R E 3 .12   firms were born larger and grew less in Ghana than in Portugal a. Initial year: All rms versus surviving rms Ghana Portugal 0.4 0.4 0.3 Surviving rms were born 0.3 Surviving rms larger than other rms resembled other rms frequency frequency 0.2 0.2 0.1 0.1 0 0 1 10 100 1,000 1 10 100 1,000 number of employees number of employees all rms, 1988 all rms, 1984 surviving rms, 1988 surviving rms, 1984 b. Surviving rms: Final year versus initial year Ghana Portugal 0.4 0.4 Surviving rms did Surviving rms grew 0.3 not grow 0.3 frequency frequency 0.2 0.2 0.1 0.1 0 0 1 10 100 1,000 1 10 100 1,000 number of employees number of employees surviving rms, 1988 surviving rms, 1984 surviving rms, 2003 surviving rms, 1991 Sources: Cabral and Mata 2003; Sandefur 2010. not all born large. In industrial countries, some a multinational conglomerate comprising 114 of the more resounding successes, from Honda companies and subsidiaries across 8 business to Microsoft, started in garages. And many suc- sectors on several continents.49 Many of China’s cessful companies in developing countries also successful clusters, such as the footwear and grew out of small household businesses. Thai- electric appliance industry in Wenzhou, also land’s Charoen Pokphand Group, founded in started from small family businesses working 1921 as a small seed shop in Bangkok by two close to each other.50 brothers, has grown into one of the world’s larg- A vibrant firm life cycle is often missing, est multinational conglomerates in agribusiness, however. Larger and older firms tend to be operating in 15 countries and encompassing stagnant, while smaller enterprises are prone to close to 100 companies.48 India’s Tata Group churning. In Ghana, for example, many firms transformed from a Mumbai-based, family- are born large and show little growth over 15 owned trading firm in the late 19th century to years (figure 3.12). In Portugal, by contrast, 112  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 F I G U R E 3 .13   The majority of firms grew little in India and Mexico a. Employment over a rm’s life cycle b. Productivity over a rm’s life cycle 11 11 10 10 employment, relative to birth productivity (TFP), relative to birth 9 9 8 8 7 7 6 6 5 5 4 4 3 3 2 2 1 1 0 0 5 9 4 9 4 9 4 35 5 9 4 9 4 9 4 35 5– –1 –1 –2 –2 –3 5– –1 –1 –2 –2 –3 < < ≥ ≥ 10 15 20 25 30 10 15 20 25 30 age o rm, years age of rm, years India Mexico United States Source: Hsieh and Klenow 2011. Note: TFP = total factor productivity. Figures show the average employment (or productivity) of firms in different age groups relative to the average employment (or productivity) of those same firms had at birth. Figures are computed using 1989–90 and 1994–95 data for India, 1998 and 2003 data for Mexico, and 1992 and 1997 data for the United States. many more firms are born as microenterprises 20 to 30 percent of household enterprises leave and grow substantially in 7 years.51 The major- the market over a two-year period, while the ity of firms are born small in India too, but they total number of household enterprises remains tend to stay small, without displaying much about the same.54 In Sub-Saharan Africa, few variation in employment over their life cycle. household enterprises expand into employment A revealing comparison involves the size of 35- beyond the household, as shown by the experi- year old firms relative to their size at birth. In ences of Ethiopia, Tanzania, and Madagascar.55 India, the size declines by a fourth. In Mexico, it In Mexico, individuals starting microenterprises doubles. In the United States, it is 10 times larger are more likely to remain the sole worker than to (figure 3.13a). Productivity growth over a firm’s increase the firm size (table 3.1).56 life cycle follows similar patterns in these coun- The wide dispersion of productivity among tries (figure 3.13b).52 businesses, the large number of unsustainable Churning—entering and exiting at a rela- microenterprises, and the stagnation of larger tively high rate—is much more common than firms all suggest that the process of market se- growth among the micro- and small enterprises lection and creative destruction that has under- of developing countries. In several Sub-Saharan pinned the rapid growth of transition economies African and Latin American countries, about 20 and East Asian countries in the past decades is percent of micro- and small enterprises enter weak in most developing countries. This weak- and leave the market in the same year. A majority ness impedes labor and other resources from of closures occur within three years of starting moving toward their most productive uses and up. Among the survivors, less than 3 percent ex- undermines both job creation and productivity pand by four employees or more.53 In Vietnam, growth. Gains from tackling the difficulties faced Jobs and productivity    113 TA B L E 3 .1   Few small firms grew in Mexico Same firms by size in 2011, % Own account 1–4 workers 5–9 workers 10 or more workers Own account 51.9 12.4  0.5  0.2 by Firms 1–4 workers 22.1 49.2  3.9  1.5 in size 5–9 workers   7.8 35.1 22.6 13.1 1987, % 10 or more workers   4.1 15.2 14.4 44.6 Source: Fajnzylber, Maloney, and Rojas 2006. Note: Rows do not add up to 100 percent because the owners of some of these firms may become salary workers or unemployed. by the start-ups and removing constraints to the are exchanged among people more effectively, growth of incumbents could be sizable, but the making everybody more productive. In clus- task is daunting (question 3).57 ters, similar firms that locate next to each other tend to benefit from a broader pool of qualified workers and common support services. Firms * * * also connect with foreign businesses through trade and investment, and, in integrating with Jobs can have an impact on the productivity of global value chains, they can acquire more ad- others, beyond the jobholder and the economic vanced knowledge, technology, and manage- unit where they belong. Jobs that have these ad- ment know-how. In all these ways, specific jobs ditional impacts do more (or less) for develop- can contribute to productivity gains of others ment. These additional impacts arise because and elsewhere in the economy. But effects can jobs differ in the way they connect with each be negative as well if jobs overuse natural re- other, and some of the connections do not oc- sources or damage the environment, thus re- cur through markets. In functional cities, ideas ducing aggregate productivity. QUESTION 3 Can entrepreneurship be fostered? Self-employment is prevalent in developing ship may be dominant, but entrepreneurship is countries, and micro- and small enterprises unlikely to be missing altogether. are a major source of livelihood for low-skilled workers. Even if only a small fraction of these Who is an entrepreneur? tiny economic units succeeded in building a viable business, with the potential to hire others, Entrepreneurship combines innovative capac- the aggregate effect on living standards would be ity to put new ideas into effect with managerial substantial. Their success would also matter for capacity to increase a firm’s efficiency within the productivity reasons. Quite a few currently large limits of known technology. Specific psychologi- enterprises in industrial countries started out as cal traits are associated with entrepreneurship, micro- and small family businesses. By contrast, ­ such as a personal need for achievement, a belief in developing countries many large enterprises in the effect of personal effort on outcomes, self- are born large, often the result of government confidence, and a positive attitude toward risk. support or privileged access to finance and in- These traits are difficult to observe or measure. formation. Breaking privileges is one more rea- But surveys comparing entrepreneurs with other son why the success of microenterprises is so workers in places as diverse as China and the Rus- important. sian Federation show that observable individual Views differ on whether there is scope to characteristics such as education, experience, help the self-employed succeed. At one time al- gender, location, and age are good predictors most every self-employed person or owner of a of entrepreneurship.61 Among microenterprises, microenterprise was seen as a potential entre- rates of return on capital tend to be higher when preneur, held back only by regulatory zeal and their owners are more educated and experienced. corruption. Substantial rates of return on capital Observable characteristics of the self-­ for micro- and small enterprises were viewed as employed can thus be used to identify individu- evidence of a potential to thrive.58 But the pen- als who have potential to become successful dulum has swung, and the conventional wisdom entrepreneurs.62 To illustrate the point, a suc- is now rather pessimistic. The large numbers of cessful entrepreneur is defined as someone who unregistered self-employed in developing coun- employs others and is not living in poverty. The tries are viewed as subsistence ­ entrepreneurs share of this group in total employment is small who are trying to make ends meet, not thriv- and relatively stable across countries at differ- ing.59 Evidence on the growth of micro- and ent levels of development.63 The share of self- small enterprises in several countries in Latin employed workers without paid employees, on America and West Africa shows that most micro- the other hand, initially increases and then de- enterprises with at least two years of operations clines with GDP per capita (figure 3.14a). At its remain at their start-up employment levels.60 peak, which corresponds to low-income coun- Embedded in the pessimism of the conventional tries, the share of self-employed workers with- wisdom is the idea that entrepreneurial ability out paid employees reaches almost three-fifths and skills cannot be easily transferred, especially of total employment. Among this group, a ma- not to adults with limited formal education. In jority are individuals with relatively low poten- this view, entrepreneurs are born, not made. If tial to succeed. Their characteristics are closer to this view is correct, attempts to convert survivor- those of wage workers than of employers.64 ship into entrepreneurship are bound to fail. The However, if each of the self-employed work- wide dispersion of productivity across firms, in- ers with high potential were to create a single ad- cluding across microenterprises, suggests, how- ditional job, total employment would increase ever, that reality is somewhere in between the substantially, somewhat more so in low-income optimistic and the pessimistic view: survivor- countries (figure 3.14b). As a share of the work- Jobs and productivity    115 F I G U R E 3 .14   Some among the self-employed have the potential to become successful entrepreneurs a. Shares of high- and low-potential b. Employment e ect of a single additional job self-employed workers in total employment created by high-potential self-employed workers 70 9 60 8 7 employment share, % % of total labor force 50 6 40 5 30 4 3 20 2 10 1 0 0 500 3,000 15,000 500 3,000 15,000 per capita GDP, US$ 2005 PPP per capita GDP, US$ 2005 PPP low-potential self-employed workers employers high-potential self-employed workers Sources: Gindling and Newhouse 2012 for the World Development Report 2013; World Development Report 2013 team estimates based on data from 36 countries. Note: GDP = gross domestic product. PPP = purchasing power parity. In panel b, each dot represents a country. ing age population, such additional job creation Removing obstacles to firm growth is thus a pre- would amount to 8 percent in Kenya, 5 percent requisite to foster entrepreneurship. in the Arab Republic of Egypt and 4 percent in Obstacles notwithstanding, entrepreneurial Costa Rica. capacity varies substantially across microen- While this calculation is hypothetical, several terprises and small firms. A distinction is often studies report that observable characteristics of made between innovative or transformative en- micro- and small informal enterprise owners, trepreneurs and replicative or subsistence entre- such as education and gender, are important preneurs.67 The former correspond to Schumpe- determinants of innovation and employment terian type of entrepreneurs, while the latter, who growth.65 In Mexico, after a business registra- generally manage micro- and small enterprises, tion reform, informal enterprise owners with are followers. Such a distinction, however, does observable traits similar to those found among not capture the broader gradation of managerial formal enterprise owners were more likely to performance that lies between the transformative register their business than those similar to wage and subsistence extremes. A study of the number workers.66 of management practices adopted by the owners of micro- and small enterprises in Sub-Saharan Africa reveals a large variation of management What constrains entrepreneurship? scores (figure 3.15). These scores are closely as- Even potentially skilled entrepreneurs would sociated with business performance.68 A broad have difficulty succeeding without access to dispersion of management scores is also found basic infrastructure and financial resources. In among relatively larger firms in India.69 their absence, managerial capacity alone may An emerging literature confirms the impor- not be enough to realize productivity gains and tance of management practices in explaining employment expansion. The investment cli- firm productivity. Although much of the focus mate matters for business performance as well. is on large firms, recent studies have turned their 116  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Management F I G U R E 3 .15   scores vary widely across small enterprises in Sub-Saharan Africa 9 8 7 6 % of �rms 5 4 3 2 1 0 1 6 11 16 21 26 management score Source: Fafchamps and Woodruff 2012. Note: The management score measures the degree to which firm owners use and master core management and business techniques. Scores are based on an evaluation of 26 techniques (26 is the highest possible score). attention to how innovation in small and me- up an offer of a wage subsidy covering 50 percent dium firms takes place. The most telling studies of the cost of hiring a worker for six months and involve management training provided for free 25 percent of the cost for another two months. to randomly selected firms whose performance Overall, these results suggest that lack of access is then compared to that of a control group of to finance is not the only constraint. firms. Evaluations of these programs find that Entrepreneurial skills, measured by the edu- the training improves the financial literacy and cation of business owners and their partici­ basic management skills of business owners. The pation in training, explain a large share of the estimated impact is also positive, but less robust, differences in productivity across firms and re- ­ when it comes to improved business outcomes gions in developing countries.71 Yet markets fail and job creation. Better outcomes are associated to nurture entrepreneurship, because knowl- with business owners who already had an ini- edge spillovers imply that some of the returns to tial understanding of the concepts and relatively acquiring or developing new managerial ideas better access to financial resources. are appropriated by others. More important On the other hand, similarly designed inter- perhaps, entrepreneurs themselves do not rec- ventions to provide financial resources to mi- ognize the relevance of management expertise.72 croenterprises, or to process their registration Only 3 percent of Brazil’s owners of micro- and with authorities, or to pay the salary of an ad- small enterprises, for instance, see management ditional employee, show mixed impacts on busi- as a binding business constraint.73 This may be ness performance.70 In Mexico and Sri Lanka, an area where information and knowledge fail- grants given to microenterprises increase the ures matter, leading to a vicious circle of low income of their owners—and then only if they productivity, low living standards, and insuffi- are male—but do not result in employment cre- cient job creation. ation. In Ghana, similar grants given to female The capacity to acquire skills and to apply business owners do not result in significant them to business seems to be one of the most growth of their microenterprises. In Sri Lanka, important characteristics of successful entrepre- only 22 percent of eligible microenterprises took neurs. Success also depends on having core skills Jobs and productivity    117 such as numeracy and literacy, as well as social from abroad can be found in the case of the gar- skills. A vast literature highlights the importance ment industry in Bangladesh (box 3.4). of entrepreneurs’ schooling as a determinant of firm growth, employment, and efficiency.74 Rus- The case for targeted management sian and Chinese business owners have more training entrepreneurs in their families and among childhood friends than otherwise similar indi- Managerial practices are linked to differences viduals, suggesting that social environment also in productivity, profitability, firm growth, and matters.75 survival.79 The development experience of the Learning can also happen through jobs. garment industry in Bangladesh suggests that Nearly half of entrepreneurs managing the 50 entrepreneurship can be fostered by exposure largest manufacturing firms in Ethiopia be- to advanced management practices and tech- gan their careers in trading companies, thereby nologies. But whether managerial capacity can learning about the market and what it takes to be improved through management training is meet demand.76 A large number of founders more debatable. Creativity, foresight, and risk and leading entrepreneurs in the light manu- taking are key elements of any innovative pro- facturing industries in Asia and Sub-Saharan cess, but the question is whether they can be dif- Africa were initially traders or employees in the fused and nurtured. marketing division of large enterprises.77 A substantial number of experiments have Integration in supply chains with larger, of- been conducted in recent years, providing evi- ten foreign, firms, is receiving much attention dence of both successes and failures of man- as a potential source of knowledge transfers.78 agement training interventions. Some patterns Indian entrepreneurs returning from Silicon emerge from a systematic review of the available Valley made Bangalore a hub of the information evidence. To be successful, management training technology industry. Perhaps the most dramatic must be kept simple, appropriate teaching ma- evidence attesting to the importance of learning terials must be available, and the training must BOX 3.4  What explains the boom in the garment industry in Bangladesh?  The garment industry in Bangladesh illustrates how important it is contributed to the proliferation of garment manufacturers by pro- to learn advanced management practices, marketing, and technol- viding a variety of valuable services including international procure- ogies from abroad. When Daewoo Corporation of Korea teamed up ment and marketing, sample making, and design reengineering. with Bangladesh’s Desh Ltd. to produce garments for export in Ban- Observing Desh’s good start in exporting, and subsequently the gladesh in 1979, the South Asian country had no modern industry. success of ex-Desh workers, highly educated people started their Little more than 20 years later, the industry was generating more own garment businesses, and wealthy families actively invested in than US$12.5 billion in export revenue. Women accounted for 80 the industry. As a result, the size of garment firms has been quite percent of its 3.6 million workers large since the beginning; their average size was 300 workers in Arguably, a wide set of factors, from financial innovation to pol- 1983–84 and 700 in 2010–11. As of 2005, owners of garment firms icy support, contributed to this development success. But it began had 15 years of schooling on average, and about 60 percent of them in 1979, when Desh sent 130 newly recruited, educated employees had completed college or university education. to Daewoo’s garment factory in Korea, where they participated in an Learning from abroad continued. Some entrepreneurs partici- eight-month intensive training course covering topics from sewing pated in training programs in Singapore, Japan, and Europe. Beyond skills to factory management, quality control, and international pro- garment enterprises in Korea, other newly industrial countries in curement and marketing—skills that they then applied in the Desh East Asia followed Daewoo into operation in Bangladesh and in- factories in Bangladesh. Within a few years, almost all the trainees vested in training Bangladeshi workers and managers. Thus, many had left Desh to start their own garment businesses. Some of the Bangladeshi traders and manufacturers had work experience in gar- ex-Desh workers joined new garment factories established by afflu- ment trading and production, including the experience of working ent businessmen, while others founded trading houses, which then at joint ventures, before starting their current businesses. Sources: Bangladesh Knitwear Manufacturers and Exporters Association 2012 ; Easterly 2002; Mottaleb and Sonobe 2011; Rhee 1990. 118  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 last for a certain minimum length of time. Com- identify those business owners with the highest plementing classroom teaching with instructors’ potential to benefit from management train- visits to trainees on the job can yield significant ing. Expert panels may be used to identify and positive effects.80 In Mexico, for example, such rank micro- and small enterprises on their po- on-site visits improved sales, profits, and pro- tential to grow, but such methods are expen- ductivity.81 But in Ghana, on-site visits and sup- sive and difficult to apply on a large scale. An port for microenterprises were not successful.82 effective alternative is a survey questionnaire It is also possible that key entrepreneurial skills designed to capture abilities, attitudes, and man- are gained more effectively through work expe- agement scores of potential trainees.87 Manage- rience in large productive firms than through ment training itself can be used as a screening training programs.83 device. Trainees with high potential often under- Entrepreneurship training for women has take new investments and expand employment had mixed results. Nurturing female entrepre- as a result of their training. Financial institutions neurship has the potential to create wider social could view such activity as an indication of po- benefits associated with female employment, tentially high investment returns. Programs that such as changes in the household allocation of combine management training with financial resources that improve family well-being, espe- support yield better firm performance in devel- cially of children. Female entrepreneurship often oping countries.88 provides employment opportunities to women Training programs can be implemented by that allow them to balance work and family private providers and financed by private inves- roles. Yet providing classroom training to female tors with a significant interest in the success of microentrepreneurs in Peru had no effect on key the entrepreneurs in whom they have invested. business outcomes such as sales and profits, even But as long as there are knowledge spillovers when some business practices improved.84 Class- and the importance of management expertise room training complemented with on-site visits, is undervalued, governments have a role to play. though, yielded positive results.85 In Pakistan Given the differing capabilities among business and Tanzania, management training improved owners, proper targeting is crucial to ensure pos- management practices and business outcomes itive returns to publicly funded programs. Ran- for male but not female entrepreneurs.86 These domized experiments in Ghana, Tanzania, and mixed results can also reflect wider constraints Vietnam indicate that the benefit of such pro- facing women in societies, including access to ef- grams generally outweighs the cost, even though fective learning in schools. the costs of implementing training programs A common finding of training evaluations is vary greatly.89 If the overall investment climate is that the potential to absorb management prac- not conducive to private sector growth, however, tices differs greatly among beneficiaries. Readily targeted training programs for better business observable individual characteristics can help skills will most likely return meager results. Jobs and productivity    119 Notes 16. Brandt and Rawski 2008; World Bank and the People’s Republic of China Development Re-  1. Labor force is used instead of employment in search Center of the State Council 2012. figure 3.1. Over a long term (10-year period), 17. Brandt, Hsieh, and Zhu 2008; Brandt and Rawski changes in unemployment rates are small, in gen- 2008; Brandt, van Biesebroeck, and Zhang 2012. eral, and employment in an economy is driven by 18. Hayami and Otsuka 1993. the size of its labor force. 19. South Africa is excluded from this figure because   2. Haltiwanger 2011; Schumpeter 1934. it is an outlier, with an average farm size of 288   3. Gross job creation is the sum of all additions to hectares. total employment. It occurs when expanding 20. In fact, tenancy markets are emerging in Sub- economic units hire workers and when new eco- Saharan Africa. See Holden, Otsuka, and Place nomic units are created. In principle, economic 2009. units can be as small as a one-person microen- 21. Otsuka and Place 2001. terprise, but most quantitative analyses refer to 22. Pagés 2010. establishments employing several workers. Gross 23. ADB 2009. The analysis is based on data from job destruction is the sum of all employment India (2004–05); Indonesia (2006); the Republic losses. It occurs when economic units close or of Korea (2004); the Philippines (2005); and Tai- contract in size. Net job creation is the difference wan, China (2006). between these two gross flows. Job creation and 24. Rijkers and others 2012 for the World Develop- destruction rates measure how many employ- ment Report 2013; Pagés 2010. ment positions emerge or disappear in a specific 25. Grimm, Kruger, and Lay 2011; Liedholm 2002. period (typically one year) relative to the number 26. See also Ayyagari, Demirgüç-Kunt, and Maksi- of existing positions. See Bartelsman, Haltiwan- movic (2011b) for analysis based on World Bank ger, and Scarpetta 2009b; Davis, Haltiwanger, and enterprise surveys; the analysis suggests that Schuh 1996. small firms contribute significantly to employ-  4. Kucera and Roncolato 2012; McMillan and ment and job creation. Rodrik 2011; Pieper 2000; Timmer and de Vries 27. The body of literature on this subject is enor- 2009. mous. See Barrett, Bellemare, and Hou 2010;   5. World Bank 2009. Carletto, Savastano, and Zezza 2011; and Larson  6. Duranton 2007, Duranton 2012 for the World and others 2012 for the World Development Re- Development Report 2013; Duranton and Puga port 2013. 2001; Henderson 2002. 28. Holden, Otsuka, and Place 2009; Larson and   7. Park et al. 2011; World Bank 2011b. others 2012 for the World Development Report ­   8. Ghani, Goswami, and Kerr 2012. 2013.   9. For a discussion on how to decompose produc- 29. For example, in Kenya in 2007, the family labor tivity growth by firm dynamics, see Foster, Halti- input per hectare in maize production was 418 wanger, and Krizan (2001) and Griliches and Re- hours for the top quartile of farms (measured by gev (1992). size), but that input reached 1,032 hours for the 10. Bartelsman and Doms 2000; Bartelsman, Halti- bottom quartile. wanger, and Scarpetta 2004; Foster, Haltiwanger, 30. The positive relationship is found not only in and Krizan 2001; Roberts 1996; Syverson 2011; such high-wage economies as Japan but also in Tybout 1996, 2000. The dispersion of productiv- India recently. See Foster and Rosenzweig 2011; ity across firms is not, by itself, sufficient to gauge Hayami and Kawagoe 1989. the efficiency of the job creation and reallocation 31. Evenson and Gollin 2003. process. Poor market structure and institutions 32. World Bank 2007. Improvement in agriculture can distort the process. See Haltiwanger 2011; technology can lead to fast growth in produc- Nelson 1981. tivity in the sector and convergence in aggre- 11. Bartelsman, Haltiwanger, and Scarpetta 2004; gate productivity. On the basis of data from 50 ­Tybout 1996. countries over 1967–92, Martin and Mitra (2001) 12. Baily, Bartelsman, and Haltiwanger 1996. found that productivity growth in agriculture 13. WDR 2013 team estimation based on Amadeus was faster than in manufacturing in these coun- Database, the Annual National Industrial Survey tries over the period. of Chile, and Ethiopia Large and Medium Scale 33. However, TFP has been increasing since the early Manufacturing and Electricity Industries Survey. 1980s, suggesting that the Green Revolution has 14. Dutz and others 2011. taken place in some areas of Sub-Saharan Africa. 15. Bartelsman, Haltiwanger, and Scarpetta 2004; See Block 2012. Rutkowski and others 2005. 34. Hayami and Ruttan 1985. 120  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 35. David and Otsuka 1994. 61. Djankov and others 2005, 2006b. See also Viva- 36. Ravallion 2005; Ravallion and Chen 2007. relli (2012) on the importance of disentangling 37. Haggblade, Hazell, and Reardon 2007. entrepreneurship drivers to craft policies target- 38. Christiaensen, Demery, and Kuhl 2011. ing high potential entrepreneurs. 39. Hsieh and Klenow 2009. 62. Methodological details of this technique to iden- 40. Pagés 2010. Figures are based on physical pro- tify high-potential entrepreneurs can be found ductivity (or TFPQ), as defined by Foster, Halti- in Gindling and Newhouse (2012) for the World wanger, and Syverson (2008). This factor is a Development Report 2013; and Grimm, Knor- measure of real output per unit of input, which ringa, and Lay (2012) for the World Develop- is computed using plant-level price deflators. ment Report 2013. TFPQ is more precise than TFPR—a revenue 63. Gindling and Newhouse 2012 for the World proxy for TFPQ that is computed using industry- Development Report 2013. level price deflators. TFPQ is a preferred measure 64. de Mel, McKenzie, and Woodruff 2008b. because TFPR combines the effects of quantities 65. de Mel, McKenzie, and Woodruff 2009; Sonobe and prices. Quantities and prices are affected and Otsuka 2006; Sonobe and Otsuka 2011. by demand factors, quality differences, mark- 66. Bruhn 2008. ups, and potential distortions. A survey of earlier 67. Baumol 2010; Schoar 2010. studies based on TFPR did not find a higher dis- 68. Fafchamps and Woodruff 2012. persion of productivity in developing countries, 69. Bloom and others 2011. but those studies are not very informative be- 70. de Mel, McKenzie, and Woodruff 2010; Faf- cause they are based on outdated methodologies. champs and others 2011; McKenzie 2010. See Tybout 2000. 71. Gennaioli and others 2011; Kelley, Bosma, and 41. Ayyagari, Demirgüç-Kunt, and Maksimovic 2011a. Amorós 2010; van der Sluis, van Praag, and 42. Ayyagari, Demirgüç-Kunt, and Maksimovic 2011a; ­Vijverberg 2005. Haltiwanger, Jarmin, and Miranda 2010. 72. Bloom and others 2011; Mano and others 2011. 43. Lin 2012; Wang and Yao 1999. 73. Estimate from Brazil’s ECINF 2003 survey. 44. Banerjee and Duflo 2011; Fox and Sohnesen 74. Fafchamps and Woodruff 2012; Gindling and 2012; Schoar 2010; Sutton and Kellow 2010. Newhouse 2012 for the World Development Re- 45. Fox and Sohnesen 2012. port 2013; Grimm, Kruger, and Lay 2011; Otsuka 46. Perry and others 2007. and Sonobe 2011; Sonobe and O ­ tsuka 2006. 47. Grimm, Kruger, and Lay 2011; McKenzie and 75. Sutton and Kellow 2010. Woodruff 2008. 76. Otsuka and Sonobe 2011; Sonobe and Otsuka 48. Mertens 2011; The Economist 2001; Charoen Pokphand Group, www.cpthailand.com. 2006. 49. Kasbekar 2007; Witze 2010. 77. Djankov and others 2006a, 2006b. 50. Sonobe, Hu, and Otsuka 2004. 78. See the initiative by the Inter-American Develop- 51. Sandefur 2010. ment Bank, “Bringing Market-Based Solutions 52. Hsieh and Klenow 2011. to Latin America and the Caribbean to Promote 53. Liedholm 2002; Mead and Liedholm 1998. Social Change,� http://browndigital.bpc.com/ 54. Results from 1-2-3 Survey. publication/?i=92819. 55. Grimm, Kruger, and Lay 2011; Kinda and Loen- 79. Bennedsen and others 2007; Bloom and others ing 2008; Loening and Imru 2009. 2011; Bloom, Schweiger, and van Reenen 2011; 56. Fajnzylber, Maloney, and Rojas 2006. Bloom and van Reenen 2007, 2010. 57. Bartelsman, Haltiwanger, and Scarpetta 2009a; 80. Drexler, Fischer, and Schoar 2011; Kairiza and Haltiwanger 2011; Hsieh and Klenow 2009; Sonobe 2012 for the World Development Report ­Syverson 2011. 2013; Mano and others 2011; Valdivia 2011. 58. Banerjee and Duflo 2004; Banerjee and others 81. Karlan, Bruhn, and Schoar 2012. 2009; de Mel, McKenzie, and Woodruff 2008a; 82. Karlan, Knight, and Udry 2012. In Ghana, the in- Göbel, Grimm, and Lay 2011; Grimm, Kruger, tervention included only a 10-hour consultancy and Lay 2011; McKenzie and Woodruff 2008. treatment, in contrast to 700 hours provided in 59. Banerjee and Duflo 2011; Schoar 2010; Sutton a successful Indian program for larger firms that and Kellow 2010; Tokman 2007. See de Soto was implemented by Bloom and others (2011). (1989) and Yunus and Jolis (1999) for a more 83. Bloom and van Reenen 2010; Bruhn, Karlan, positive view. and Schoar 2010; Bruhn and Zia 2011; de Mel, 60. Fajnzylber, Maloney, and Rojas 2006; results McKenzie, and Woodruff 2009. from World Bank’s informal enterprise surveys 84. Karlan and Valdivia 2010. conducted between 2009 and 2010. 85. Valdivia 2011. Jobs and productivity    121 86. Berge, Bjorvatn, and Tungodden 2011; Giné and nal Microdata.� Journal of Economic Literature Mansuri 2011. (38) 3: 569–94. 87. Fafchamps and Woodruff 2012. Alternatively, see Bartelsman, Eric, John Haltiwanger, and Stefano Scar- the work done by the Entrepreneurial Finance petta. 2004. “Microeconomic Evidence of Creative Lab at the Kennedy School (http://www.efinlab Destruction in Industrial and Developing Coun- .com) on psychometrics tools to uncover success- tries.� Discussion Paper Series 1374, Institute for ful entrepreneurs. the Study of Labor, Bonn. 88. Cho and Honorati 2012 for the World Develop- ———. 2009a. “Cross-Country Differences in Pro- ment Report 2013. ductivity: The Role of Allocation and Selection.� 89. Sonobe, Higuchi, and Otsuka 2012 for the World Working Paper 15490. National Bureau of Eco- Development Report 2013. nomic Research, Cambridge, MA. ———. 2009b. “Measuring and Analyzing Cross- References Country Differences in Firm Dynamics.� In Pro- ducer Dynamics: New Evidence from Micro Data, The word processed describes informally reproduced ed. Timothy Dunne, J. Bradford Jensen, and Mark works that may not be commonly available through J. Roberts, 15–82. Cambridge, MA: National Bu- libraries. reau of Economic Research. Baumol, William J. 2010. The Microtheory of Innova- ADB (Asian Development Bank). 2009. Enterprises in tive Entrepreneurship. Princeton, NJ: Princeton Asia: Fostering Dynamism in SMEs. Manila: ADB. University Press. Ayyagari, Meghana, Asli Demirgüç-Kunt, and Vojislav Bennedsen, Morten, Kasper M. Nielsen, Francisco Maksimovic. 2011a. “Firm Innovation in Emerg- Perez-Gonzalez, and Daniel Wolfenzon. 2007. “In- ing Markets: The Role of Finance, Governance, side the Family Firm: The Role of Families in Suc- and Competition.� Journal of Financial and Quan- cession Decisions and Performance.� Quarterly titative Analysis 46(6): 1545–80. Journal of Economics 122 (2): 647–91. ———. 2011b. “Small vs. Young Firms across the Berge, Lars Ivar Oppedal, Kjetil Bjorvatn, and Bertil World: Contribution to Employment, Job Cre- Tungodden. 2011. “Human and Financial Capi- ation, and Growth.� Policy Research Working Pa- tal for Microenterprise Development: Evidence per Series 5631, World Bank, Washington, DC. from a Field and Lab Experiment.� Discussion Baily, Martin Neil, Eric J. Bartelsman, and John Paper 1, Norwegian School of Economics, Ber- Haltiwanger. 1996. “Downsizing and Productiv- gen, Norway. ity Growth: Myth or Reality?� Small Business Eco- Block, Steven. 2012. “The Decline and Rise of Ag- nomics 8 (4): 259–78. ricultural Productivity in Sub-Saharan Africa.� Banerjee, Abhijit V., and Esther Duflo. 2004. “Do Working Paper Series 16841, National Bureau of Firms Want to Borrow More? Testing Credit Con- Economic Research, Cambridge, MA. straints Using a Directed Lending Program.� Dis- Bloom, Nicholas, Benn Eifert, Aprajit Mahajan, cussion Paper Series 4681, Centre for Economic David McKenzie, and John Roberts. 2011. “Does Policy Research, London. Management Matter? Evidence from India.� Pol- ———. 2011. Poor Economics: A Radical Rethinking icy Research Working Paper Series 5573, World of the Way to Fight Global Poverty. New York: Pub- Bank, Washington, DC. lic Affairs. Bloom, Nicholas, Helena Schweiger, and John Van Banerjee, Abhijit V., Esther Duflo, Rachel Glennerster, Reenen. 2011. “The Land That Lean Manufactur- and Cynthia Kinnan. 2009. “The Miracle of Mi- ing Forgot? Management Practices in Transition crofinance? Evidence from a Randomized Evalu- Countries.� Working Paper Series 17231, National ation.� Massachusetts Institute of Technology, Bureau of Economic Research, Cambridge, MA. Cambridge, MA. Processed. Bloom, Nicholas, and John van Reenen. 2007. “Mea- Bangladesh Knitwear Manufacturers and Exporters suring and Explaining Management Practices Association. 2012. “History of Development of across Firms and Countries.� Quarterly Journal of Knitwear of Bangladesh.� Bangladesh Knitwear Economics 122 (4): 1351–408. Manufacturers and Exporters Association, Ban- ———. 2010. “Why Do Management Practices Dif- gladesh. Processed. fer across Firms and Countries?� Journal of Eco- Barrett, Christopher B., Marc F. Bellemare, and Janet nomic Perspectives 24 (1): 203–24. Y. Hou. 2010. “Reconsidering Conventional Ex- Brandt, Loren, Chang-Tai Hsieh, and Xiaodong Zhu. planations of the Inverse Productivity–Size Rela- 2008. “Growth and Structural Transformation in tionship.� World Development 38 (1): 88–97. China.� In China’s Great Economic Transforma- Bartelsman, Eric J., and Mark Doms. 2000. “Under- tion, ed. Loren Brandt and Thomas G. Rawski. standing Productivity: Evidence from Longitudi- Cambridge, U.K.: Cambridge University Press. 122  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Brandt, Loren, and Thomas G Rawski. 2008. China’s de Soto, Hernando. 1989. The Other Path: The In- Great Economic Transformation. Cambridge, U.K.: visible Revolution in the Third World. New York: Cambridge University Press. Harper & Row. Brandt, Loren, Johannes van Biesebroeck, and Yifan Djankov, Simeon, Miguel Edward, Yingyi Qian, Zhang. 2012. “Creative Accounting or Creative Gerard Roland, and Ekaterina Zhuravskaya. 2005. Destruction? Firm-Level Productivity Growth in “Who Are Russia’s Entrepreneurs?� Journal of the Chinese Manufacturing.� Journal of Development European Economic Association 3 (2–3): 1–11. Economics 97 (2): 339–51. Djankov, Simeon, Yingyi Qian, Gerard Roland, and Bruhn, Miriam. 2008. “License to Sell: The Effect of Ekaterina Zhuravskaya. 2006a. “Entrepreneurship Business Registration Reform on Entrepreneur- in China and Russia Compared.� Journal of the ial Activity in Mexico.� Policy Research Working European Economic Association 4 (2–3): 352–65. Paper Series 4538, World Bank. Washington, DC. ———. 2006b. “Who Are China’s Entrepreneurs?� Bruhn, Miriam, Dean Karlan, and Antoinette Schoar. Amercian Economics Review 96 (2): 348–52. 2010. “What Capital Is Missing in Developing Drexler, Alejandro, Greg Fischer, and Antoinette Countries?� American Economic Review 100 (2): Schoar. 2011. “Keeping It Simple: Financial Lit- 629–33. eracy and Rules of Thumb.� MIT Sloan Manage- Bruhn, Miriam, and Bilal Zia. 2011. “Stimulating ment, Cambridge, MA. Processed. Managerial Capital in Emerging Markets: The Im- Duranton, Gilles. 2007. “Urban Evolutions: The Fast, pact of Business and Financial Literacy for Young the Slow, and the Still.� American Economic Review Entrepreneurs.� Policy Research Working Paper 97 (1): 197–221. Series 5642, World Bank, Washington, DC. ———. 2012. “Agglomeration and Jobs in Develop- Cabral, Luis M. B., and José Mata. 2003. “On the ing Countries.� Background paper for the WDR Evolution of the Firm Size Distribution: Facts 2013. and Theory.� American Economic Review 93 (4): Duranton, Gilles, and Diego Puga. 2001. “Nursery 1075–90. Cities: Urban Diversity, Process Innovation, and Carletto, Calogero, Sara Savastano, and Alberto Zezza. the Life Cycle of Products.� American Economic 2011. “Fact or Artefact: The Impact of Measure- Review 91 (5): 1454–77. ment Errors on the Farm Size–Productivity Rela- Dutz, Mark A., Ioannis Kessides, Stephen O’Connell, tionship.� Policy Research Working Paper Series and Robert D. Willig. 2011. “Competition and 5908, World Bank, Washington, DC. Innovation-Driven Inclusive Growth.� Policy Re- Cho, Yoonyoung, and Maddalena Honorati. 2012. “A search Working Paper Series 5852, World Bank, Meta-Analysis of Entrepreneurship Programs in Washington, DC. Developing Countries.� Background paper for the Easterly, William. 2002. The Elusive Quest for Growth WDR 2013. Economists’ Adventures and Misadventures in the Christiaensen, Luc, Lionel Demery, and Jesper Kuhl. Tropics. Cambridge, MA: MIT Press. 2011. “The (Evolving) Role of Agriculture in Pov- erty Reduction—An Empirical Perspective.� Jour- Evenson, Robert E., and Douglas Gollin. 2003. Crop nal of Development Economics 96 (2): 239–54. Variety Improvement and Its Effect on Productivity: David, Christina C., and Keijiro Otsuka, eds. 1994. The Impact of International Agricultural Research. Modern Rice Technology and Income Distribution Wallingford, U.K.: CABI Publishing. in Asia. Boulder, CO: Lynne Rienner. Fafchamps, Marcel, David J. McKenzie, Simon Quinn, Davis, Steven J., John C. Haltiwanger, and Scott and Christopher Woodruff. 2011. “When Is Capi- Schuh. 1996. Job Creation and Destruction. Cam- tal Enough to Get Female Microenterprises Grow- bridge, MA: MIT Press. ing? Evidence from a Randomized Experiment in de Mel, Suresh, David J. McKenzie, and Christopher Ghana.� Working Paper Series 17207, National Woodruff. 2008a. “Returns to Capital in Micro- Bureau of Economic Research, Cambridge, MA. enterprises: Evidence from a Field Experiment.� Fafchamps, Marcel, and Christopher Woodruff. 2012. Quarterly Journal of Economics 123 (4): 1329–72. “Identifying and Relaxing Constraints to Employ- ———. 2008b. “Who Are the Microenterprise Own- ment Generation in Small-Scale African Enter- ers? Evidence from Sri Lanka on Tokman v. de prises.� University of Oxford, Oxford. Processed. Soto.� Discussion Paper Series 3511, Institute for Fajnzylber, Pablo, William Maloney, and Gabriel the Study of Labor, Bonn. Montes Rojas. 2006. “Microenterprise Dynamics ———. 2009. “Innovative Firms or Innovative Own- in Developing Countries: How Similar Are They ers? Determinants of Innovation in Micro, Small, to Those in the Industrialized World? Evidence and Medium Enterprises.� Discussion Paper Se- from Mexico.� World Bank Economic Review 20 ries 3962, Institute for the Study of Labor, Bonn. (3): 389–419. ———. 2010. “Wage Subsidies for Microenterprises.� FAO (Food and Agriculture Organization). 2010. American Economic Review 100 (2): 614–18. 2000 World Census of Agriculture: Main Results Jobs and productivity    123 and Metadata by Country (1996–2005). Rome: Haggblade, Steven, Peter B. R. Hazell, and Thomas FAO. Anthony Reardon, eds. 2007. Transforming the Foster, Andrew D., and Mark R. Rosenzweig. 2011. Rural Nonfarm Economy. Baltimore, MD: Johns “Are Indian Farms Too Small? Mechanization, Hopkins University Press. Agency Costs, and Farm Efficiency.� Brown Uni- Haltiwanger, John. 2011. “Globalization and Eco- versity, Providence, RI. Processed. nomic Volatility.� In Making Globalization Socially Foster, Lucia Smith, John C. Haltiwanger, and Cornell Sustainable, ed. Marc Bacchetta and Marion Jan- J. Krizan. 2001. “Aggregate Productivity Growth: sen, 119–46. Geneva: International Labour Orga- Lessons from Microeconomic Evidence.� In New nization and World Trade Organization. Developments in Productivity Analysis, ed. Charles Haltiwanger, John, Ron S. Jarmin, and Javier Mi- R. Hulten, Edwin R. Dean, and Michael J. Harper, randa. 2010. “Who Creates Jobs? Small vs. Large 303–72. Chicago, IL: University of Chicago vs. Young.� Working Paper Series 16300, National Press. Bureau of Economic Research, Cambridge, MA. Foster, Lucia Smith, John C. Haltiwanger, and Chad Hayami, Yujiro, and Toshihiko Kawagoe. 1989. “Farm Syverson. 2008. “Reallocation, Firm Turnover, and Mechanization, Scale Economies and Polariza- Efficiency: Selection on Productivity or Profitabil- tion: The Japanese Experience.� Journal of Devel- ity?� American Economic Review 98 (1): 394–425. opment Economics 31 (2): 221–39. Fox, Louise, and Thomas Sohnesen. 2012. “House- Hayami, Yujiro, and Keijiro Otsuka. 1993. The Eco- hold Enterprise in Sub-Saharan Africa: Why They nomics of Contract Choice: An Agrarian Perspective. Matter for Growth, Jobs, and Poverty Reduction.� New York: Oxford University Press. Policy Research Working Paper 6184, World Bank, Hayami, Yujiro, and Vernon W. Ruttan. 1985. Agricul- Washington, DC. tural Development: An International Perspective. Gennaioli, Nicola, Rafael La Porta, Florencio Lopez- Baltimore, MD: Johns Hopkins University Press. de-Silanes, and Andrei Hleifer. 2011. “Human Henderson, Vernon. 2002. “Urbanization in Develop- Capital and Regional Development.� Working ing Countries.� World Bank Research Observer 17 Paper Series 17158, National Bureau of Economic (1): 89–112. Research, Cambridge, MA. Holden, Stein T., Keijiro Otsuka, and Frank M. Place. Ghani, Ejaz, Arti Grover Goswami, and William R. 2009. “Land Markets and Development in Africa.� Kerr. 2012. “Is India’s Manufacturing Sector Mov- In The Emergence of Land Markets in Africa: As- ing Away From Cities?� Working Paper Series sessing the Impacts on Poverty, Equity, and Effi- 12-090, Harvard Business School, Harvard Uni- ciency, ed. Stein T. Holden, Keijiro Otsuka, and versity, Cambridge, MA. Frank M. Place, 3–17. Washington, DC: Resources Gindling, T. H., and David Newhouse 2012. “Self- for the Future. Employment in the Developing World.� Back- Hsieh, Chang-Tai, and Peter J. Klenow. 2009. “Misal- ground paper for the WDR 2013. location and Manufacturing TFP in China and Giné, Xavier, and Ghazala Mansuri. 2011. “Together India.� Quarterly Journal of Economics 124 (4): We Will: Evidence from a Field Experiment on 1403–48. Female Voter Turnout in Pakistan.� World Bank, ———. 2011. “The Life Cycle of Plants in India and Washington, DC. Processed. Mexico.� Research Paper 11-33, Booth School of Göbel, Kristin, Michael Grimm, and Jann Lay. 2011. Business, University of Chicago, Chicago, IL. “Capital Returns, Productivity and Accumulation IFC (International Finance Corporation). Forthcom- in Micro and Small Enterprises: Evidence from ing. IFC Job Study: Assessing Private Sector Contri- Peruvian Panel Data.� Processed. butions to Job Creation. Washington, DC: IFC. Griliches, Zvi, and Haim Regev. 1992. “Productiv- Kairiza, Terrence, and Tetsushi Sonobe. 2012. “Are ity and Firm Turnover in Israeli Industry, 1979– Female Entrepreneurs Less Skillful in Manag- 1988.� Journal of Development Economics 65 (1): ment? A Randomized Experiment in the Garment 175–203. Industry in Tanzania.� Discussion Paper 11-24, Grimm, Michael, Peter Knorringa, and Jann Lay. National Graduate Institute for Policy Studies, 2012. “Constrained Gazelles: High Potentials in Tokyo. West Africa’s Informal Economy.� International Karlan, Dean, Miriam Bruhn, and Antoinette Schoar. Institute of Social Studies, Erasmus University– 2012. “The Impact of Consulting Services on Rotterdam, Rotterdam, Netherlands. Processed. Small and Medium Enterprises: Evidence from Grimm, Michael, Jens Kruger, and Jann Lay. 2011. a Randomized Trial in Mexico.� Working Paper “Barriers to Entry and Returns to Capital in In- 100, Economics Department, Yale University, New formal Activities: Evidence from Sub-Saharan Haven, CT. Africa.� Review of Income and Wealth 57 (S1): Karlan, Dean, Ryan Knight, and Christopher Udry. S27–S53. 2012. “Hoping to Win, Expected to Lose: Theory 124  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 and Lessons on Micro Enterprise Development.� Growth.� Working Paper Series 17143, National Innovations for Poverty Action, Cambridge, MA. Bureau of Economic Research, Cambridge, MA. Processed. Mead, Donald C., and Carl Liedholm. 1998. “The Karlan, Dean S., and Martin Valdivia. 2010. “Teaching Dynamics of Micro and Small Enterprises in De- Entrepreneurship: Impact of Business Training on veloping Countries.� World Development 26 (1): Microfinance Clients and Institutions.� Review of 61–74. Economics and Statistics 93 (2): 510–27. Mertens, Brian. 2011. “Forbes Asia’s Businessman of Kasbekar, Kiron. 2007. Only the Fittest Survive. Lon- the Year.� Forbes Asia Magazine, December 5. don: Tata Group. Montenegro, Claudio E., and Harry Anthony Pa- Kelley, Donna, Niels Bosma, and José A. Amorós. trinos. 2012. “Returns to Schooling around the 2010. GEM Global Report. London: Global Entre- World.� Background paper for the WDR 2013. preneurship Monitor. Mottaleb, Khondoker A., and Tetsushi Sonobe. 2011. Kinda, Tidiane, and Josef L. Loening. 2008. “Small “An Inquiry into the Rapid Growth of the Gar- Enterprise Growth and the Rural Investment Cli- ment Industry in Bangladesh.� Economic Develop- mate: Evidence from Tanzania.� Policy Research ment and Cultural Change 60 (1): 67–89. Working Paper Series 4675, World Bank, Wash- Nelson, Richard R. 1981. “Research on Productivity ington, DC. Growth and Productivity Differences: Dead Ends Kucera, David, and Leanne Roncolato. 2012. “Struc- and New Departures.� Journal of Economic Litera- ture Matters: Sectoral Drivers of Development ture 19 (3): 1029–64. and the Labour Productivity–Employment Re- Otsuka, Keijiro, and Frank Place. 2001. Land Tenure lationship.� International Labour Organization, and Natural Resource Management: A Compara- Geneva. Processed. tive Study of Agrarian Communities in Asia and Larson, Donald F., Keijiro Otsuka, Tomoya Matsu- Africa. Baltimore, MD: Johns Hopkins University moto, and Talip Kilic. 2012. “Can Africa’s Agri- Press. culture Depend on Smallholder Farmers?� Back- Otsuka, Keijiro, and Tetsushi Sonobe. 2011. “A ground paper for the WDR 2013. Cluster-Based Industrial Development Policy for Liedholm, Carl. 2002. “Small Firm Dynamics: Evi- Low-Income Countries.� Policy Research Working dence from Africa and Latin America.� Small Busi- Paper Series 5703, World Bank, Washington, DC. ness Economics 18 (1–3): 227–42. Pagés, Carmen, ed. 2010. The Age of Productivity: Lin, Justin Yifu. 2012. Demystifying the Chinese Econ- Transforming Economies from the Bottom Up. New omy. Cambridge, U.K.: Cambridge University York: Palgrave Macmillan. Press. Park, Jaegil, Daejong Kim, Yongseok Ko, Eunnan Loening, Josef, and Mikael Imru. 2009. “Ethiopia: Kim, Keunhyun Park, and Keuntae Kim. 2011. Diversifying the Rural Economy. An Assessment Urbanization and Urban Policies in Korea. Korea of the Investment Climate for Small and Infor- Research Institute for Human Settlements. mal Enterprises.� Paper 23278, Munich Personal Perry, Guillermo E., William F. Maloney, Omar S. RePEc Archive, Munich. Arias, Pablo Fajnzylber, Andrew D. Mason, and Mano, Yukichi, Alhassa Iddrisu, Yutaka Yoshino, and Jaime Saavedra-Chanduvi. 2007. Informality: Exit Tetsushi Sonobe. 2011. “How Can Micro and and Exclusion. Washington, DC: World Bank. Small Enterprises in Sub-Saharan Africa Become Pieper, Ute. 2000. “Deindustrialisation and the Social More Productive? The Impacts of Experimental and Economic Sustainability Nexus in Developing Basic Management Training.� Policy Research Countries: Cross-Country Evidence on Produc- Working Paper Series 5755, World Bank, Wash- tivity and Employment.� Journal of Development ington, DC. Studies 36 (4): 66–99. Martin, Will, and Devashish Mitra. 2001. “Productiv- Ravallion, Martin. 2005. “Externalities in Rural De- ity Growth and Convergence in Agriculture versus velopment: Evidence for China.� In Spatial In- Manufacturing.� Economic Development and Cul- equality and Development, ed. Ravi Kanbur and tural Change 49 (2): 403–22. Anthony J. Venables, 137–62. Oxford: Oxford McKenzie, David. 2010. Impact Assessment in Finance University Press. and Private Sector Development: What Have We Ravallion, Martin, and Shaohua Chen. 2007. “China’s Learned and What Should We Learn? Washington, (Uneven) Progress against Poverty.� Journal of DC: World Bank. ­Development Economics 82 (1): 1–42. McKenzie, David, and Christopher Woodruff. 2008. Rhee, Yung Whee. 1990. “The Catalyst Model of “Experimental Evidence on Returns to Capital Development: Lessons from Bangladesh’s Success ­ and Access to Finance in Mexico.� World Bank with Garment Exports.� World Development 18 Economic Review 22 (3): 457–482. (2): 333–46. McMillan, Margaret S., and Dani Rodrik. 2011. “Glo- Rijkers, Bob, Hassen Arouri, Caroline Freund, and balization, Structural Change, and Productivity Antonio Nucifora. 2012. “Which Firms Create Jobs and productivity    125 Jobs in Tunisia?� Background paper for the WDR Asia and Latin America: A New Sectoral Data Set.� 2013. Cliometrica 3: 165–90. Roberts, Mark J. 1996. “Employment Flows and Pro- Tokman, Victor. 2007. “Modernizing the Infor- ducer Turnover.� In Industrial Evolution in De- mal Sector.� Working Paper 42, Department of veloping Countries: Micro Patterns of Turnover, Economic and Social Affairs, New York, United Productivity, and Market Structure, ed. Mark J. Nations. Roberts and James R. Tybout. Oxford: Oxford Tybout, James R. 1996. “Heterogeneity and Productiv- University Press. ity Growth: Assessing the Evidence.� In Industrial Rutkowski, Jan, Stefano Scarpetta, Arup Banerji, Evolution in Developing Countries: Micro Patterns Philip O’Keefe, Gaelle Pierre, and Milan Vodop- of Turnover, Productivity, and Market Structure, ed. ivec. 2005. Enhancing Job Opportunities: Eastern Mark J. Roberts and James R. Tybout. Oxford: Ox- Europe and the Soviet Union. Washington, DC: ford University Press. World Bank. ———. 2000. “Manufacturing Firms in Developing Sandefur, Justin. 2010. “On the Evolution of the Firm Countries: How Well Do They Do, and Why?� Size Distribution in an African Economy.� Centre Journal of Economic Literature 38 (1): 11–44. for the Study of African Economies Working Pa- Valdivia, Martin. 2011. “Training or Technical As- per Series 2010-05, Oxford. Processed. sistance for Female Entrepreneurship? Evidence Schoar, Antoinette. 2010. “The Divide between Sub- from a Field Experiment in Peru.� GRADE (Grupo sistence and Transformational Entrepreneur- de Análisis para el Desarrollo), Lima. Processed. ship.� In Innovation Policy and the Economy, Vol- van der Sluis, Justin, Mirjam van Praag, and Wim Vi- ume 10, ed. Josh Lerner and Scott Stern, 57–81. jverberg. 2005. “Entrepreneurship Selection and Cambridge, MA: National Bureau of Economic Performance: A Meta-analysis of the Impact of Research. Education in Developing Economies.� World Bank Schumpeter, Joseph Alois. 1934. The Theory of Eco- Economic Review 19 (2): 225–61. Vivarelli, Marco. 2012. “Entrepreneurship in Ad- nomic Development: An Inquiry into Profits, Capi- vanced and Developing Countries: A Microeco- tal, Credit, Interest, and the Business Cycle. Cam- nomic Perspective,� Discussion Paper Series 6513, bridge, MA: Harvard University Press. IZA, Bonn. Shiferaw, Admasu, and Arjun S. Bedi. 2010. “The Dy- Wang, Yueping, and Yang Yao. 1999. Market Reforms, namics of Job Creation and Job Destruction: Is Technological Capabilities, and the Performance Sub-Saharan Africa Different?� Poverty, Equity of Small Enterprises in China. Washington, DC: and Growth Discussion Paper 22, Courant Re- World Bank. search Centre, Göttingen, Germany. Witze, Morgen. 2010. “Case Study: Tata.� Financial Sonobe, Tetsushi, Yuki Higuchi, and Keijiro Otsuka. Times, December 29. 2012. “Productivity Growth and Job Creation in World Bank. 2007. World Development Report 2008: the Development Process of Industrial Clusters.� Agriculture for Development. Washington, DC: Background paper for the WDR 2013. World Bank. Sonobe, Tetsushi, Dinghuan Hu, and Keijiro Otsuka. ———. 2009. World Development Report 2009: Re- 2004. “From Inferior to Superior Products: An shaping Economic Geography. Washington, DC: Inquiry into the Wenzhou Model of Industrial World Bank. Development in China.� Journal of Comparative ———. 2011a. “Brazil Urbanization Review.� World Economics 32 (3): 542–62. Bank, Washington, DC. Processed. Sonobe, Tetsushi, and Keijiro Otsuka. 2006. Cluster- ———. 2011b. Vietnam Urbanization Review: Tech- Based Industrial Development: An East Asian nical Assistance Report. Washington, DC: World Model. New York: Palgrave Macmillan. Bank. ———. 2011. Cluster-Based Industrial Development: ———. Forthcoming. Planning, Connecting, Financ- A Comparative Study of Asia and Africa. New York: ing Cities—Now. Urbanization Review Flagship Palgrave Macmillan. Report. Washington, DC: World Bank. Sutton, John, and Nebil Kellow. 2010. An Enterprise World Bank, and the People’s Republic of China De- Map of Ethiopia. London: International Growth velopment Research Center of the State Council. Centre. 2012. China 2030: Building a Modern, Harmoni- Syverson, Chad. 2011. “What Determines Productiv- ous, and Creative High-Income Society. Washing- ity?� Journal of Economic Literature 49: (2), 326–65. ton, DC: World Bank and Development Research The Economist. 2001. “Face Value: Radicalism, Center of the State Council, the People’s Republic Asian Style.� The Economist Newspaper Limited, of China. March 22. Yunus, Muhammad, and Alan Jolis. 1999. Banker to Timmer, Marcel P., and Gaaitzen J. de Vries. 2009. the Poor: Micro-Lending and the Battle against “Structural Change and Growth Accelerations in World Poverty. New York: Public Affairs. CHAPTER 4 Jobs and social cohesion Jobs can shape social interactions and the ways societies manage collective decision making. They connect people with others and can provide access to voice. J obs influence who we are and our relations with others. In most societies, jobs are a fundamental source of self-respect and social identity. Historically, family names in cohesion can also influence jobs by shaping the context in which entrepreneurs make in- vestment decisions. Empirical evidence of a connection between some cultures were associated with specific oc- jobs and social cohesion is limited by data con- cupations because people defined themselves straints, the complexity of measuring social by what they did: Miller in English, Hurudza interactions, and the multiple factors that can (master farmer) in Shona, and Suthar (carpen- contribute to social cohesion. However, cross- ters) in Hindi. country analysis of values surveys finds that job Jobs connect people with others through loss or lack of access to jobs is associated with networks. The workplace can be a place to en- lower levels of trust and civic engagement. This counter new ideas and information and to inter- is not only a rich-country phenomenon, as is of- act with people of different ethnicities. The dis- ten suggested. Unemployment can cause depres- tribution of jobs within society and perceptions sion, increase mistrust in others, and lead people about who has access to opportunities and why to drop out of community life. Migrants without can shape people’s expectations and aspirations social ties may be excluded from job opportuni- for the future, their sense of having a stake in ties that would allow them to succeed in their society, and perceptions of fairness. new environments. In extreme cases, if people, These individual influences of jobs may particularly youth, lack jobs and hope for the fu- have collective consequences. Having or not ture, they may turn to violent or criminal activ- having a job may affect key elements of social ity to compensate for the absence of self-esteem cohesion, the capacity of societies to manage and sense of belonging that a job might other- collective decision making peacefully. While wise provide. Similarly, jobs offering limited op- the frustration of unemployed youth during portunities for future growth or lacking access the Arab Spring suggests that the lack of jobs to voice can lead to alienation and frustration. can be a source of social unrest, that does not Some jobs are positively correlated with so- mean that the relationship between jobs and cial cohesion. Jobs that are empowering, build social cohesion is straightforward, immediate, agency, and provide access to voice can increase or direct. Rather, the relationship is contextual trust and people’s willingness to participate in and shaped by individuals, their values, atti- civil society. Jobs can create economic and social tudes, and behaviors, and the institutions that ties and have the potential to build incentives surround them. And it goes both ways: social to work across boundaries and resolve conflict. Jobs and social cohesion   127 And people’s trust in government and their con- intergroup relations, and the effectiveness of fidence in institutions may increase if they be- channels for resolving conflicts. Cross-country lieve that job opportunities are available to them data on political stability, the absence of vio- either now or in the future. Jobs can influence lence, and voice and accountability can be used social cohesion through their effects on social to construct an index of social cohesion at the identity, networks, and fairness. country level.6 The Nordic countries, Swit- zerland, and New Zealand, score high on this index. Although the index is a static measure, Jobs can help manage social the capacity for peaceful decision making can tensions evolve over time as societies change, through urbanization, more female employment, and News reports about the financial crisis and the the growth of a middle class. Arab Spring have broadcast a common senti- The nature of the interactions through jobs ment that unemployment, especially among affects the degree of social cohesion in commu- young people, can ignite unrest and violence.1 nities and societies. Trust and civic engagement In September 2010, a Telegraph headline re- are two measurable indicators of social cohe- ported that the “IMF Fears ‘Social Explosion’ sion at the individual level. These indicators are from World Jobs Crisis� ahead of a summit associated with the country-level index of the of the International Monetary Fund and the capacity for peaceful collective decision making International Labour Organization (ILO).2 In (figure 4.1). 2011, Le Monde linked jobs and social unrest Trust refers to the extent to which individu- in Tunisia to concerns about social justice: als have confidence in people whom they know “protesters aren’t asking the Government to personally, including family and neighbors.7 It find them a job, but denouncing the lack of can also refer to trust in people met for the first transparency and justice in the labor market.�3 time and in people of different religions and The revolution in Tunisia was sparked by the nationalities.­Civic engagement captures the protests of a fruit vendor frustrated by his in- extent to which people participate voluntarily ability to get a permit to do his job. High lev- in civil society by joining community organiza- els of youth unemployment were a significant tions, unions, political parties, or religious or- contributing factor to the riots in the United ganizations, and by engaging in civic life. These Kingdom in the summer of 2011.4 forms of involvement and activism include These events suggest that jobs can contrib- nonviolent activity, such as participating in pro- ute to social cohesion, including how societies tests, that can be constructive for social cohe- handle differences and manage tensions among sion. Civic engagement relates to social capital, different groups, and how they avoid and resolve participation, and the agency that motivates in- ­ conflicts. There are many possible ways to de- dividuals to be part of collective action. social fine social cohesion (box 4.1). But overall, ­ cohesion refers to the capacity of societies to . . . and they are influenced by jobs peacefully manage collective decision making.5 Social cohesion thus relates to the processes and Trust and civic engagement can be linked to institutions that shape how groups interact. It jobs. Having—or not having—a job may af- does not follow that collective decision making fect the way people view the world by influenc- should be imposed from above, but rather that ing their values and attitudes, including trust in channels for voice, accountability, and inclusive others and in institutions. Jobs can also provide participation of diverse groups can contribute channels for people to interact across diverse to a cohesive society. groups. Jobs with certain characteristics may contribute more to trust and civic engagement than others. Trust and civic engagement matter . . . Not having a job is associated with less self- The capacity of a country to support peaceful reported trust in high-income countries (figure collective decision making involves multiple 4.2a). The relationship is stronger with civic factors including the quality of institutions, engagement, where unemployment is linked to 128  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 BOX 4.1  What is social cohesion?  The concept can be traced as far back as the writings of Ibn Khaldun, nomic divisions within society (income, ethnicity, political party, a Muslim scholar born in Tunis in the 15th century, whose idea of caste, language).�d asabiyah is generally translated as “social cohesion.� Living during Although social cohesion has multiple definitions with differ- times of manifold conflicts, Khaldun regarded asabiyah as the soli- ences in focus and emphasis, some common threads emerge: darity of small groups (tribes) that has the power to promote • Social cohesion is generally viewed as a positive concept. It can be broader social integration.a an end in itself, as well as a means. The Organisation for Economic Four centuries later, Emile Durkheim considered cohesion in Co-operation and Development describes a cohesive society as the context of societal transformation.b He was particularly con- one that “works towards the well-being of all its members.�e The cerned with two different types of solidarity that he observed French Commissariat General du Plan defines social cohesion as “a emerging through industrialization in Europe. Primitive societies, set of social processes that help instill in individuals the sense of he found, were marked by mechanical solidarity and a strong col- belonging to the same community and the feeling that they are lective ethos based on relatively homogeneous patterns of life recognized as members of the community.�f and work. Advanced capitalist society, in contrast, with its complex division of labor, was marked by organic solidarity based on merit, •  Social cohesion relates to the interactions among individuals, respect for different roles within the labor force, and the need for groups, and societies. These interactions are seen as “the forces moral regulation. holding the individuals within the groupings in which they are� More recently, social cohesion has been related to social capital. and as linking diverse groups together.g In the 1990s, Pierre Bourdieu and others focused on the benefits •  Social cohesion contributes to sustainable social development. that accrue to individuals through their participation in groups and According to the Club de Madrid, “shared societies are stable, safe the need for individuals to invest in these relations. Robert Putnam’s and just and based on the promotion and protection of all human analysis of the conditions for creating responsive, effective, and rep- rights . . . , including disadvantaged and vulnerable groups and resentative institutions builds on this theme. Famously, northern persons.�h Italy had more of these institutions than southern Italy, and Putnam concluded that the central enabling condition was the existence •  Last, some definitions of social cohesion relate the concept to pro- of more social capital, measured through the density of local cesses and institutional characteristics. For example, social cohesion associations.c Social cohesion can be understood as a broader con- can refer to “the capacity of societies (not just groups, networks) cept than social capital in that it considers intergroup relations in to peacefully manage collective action problems.�i This definition a wider context. Easterly, Ritzen, and Woolcock define social cohe- links social cohesion to participation and civic engagement. sion (or lack thereof) as “the nature and extent of social and eco- Sources: World Development Report 2013 team based on Norton and de Haan 2012 for the World Development Report 2013; OECD 2011. a. Weiss 1995. b. Durkheim 1893. c. Putnam, Leonardi, and Nanetti 1993. d. Easterly, Ritzen, and Woolcock 2006, 105. e. OECD 2011, 17. f. Jenson 1998, 4. g. Moreno and Jennings 1937, 371. h. OECD 2011, 53. i. Woolcock 2011. lower participation in associations and dem- In developing countries, the type of job, the op- onstrations, and signing petitions (figures 4.2b, portunities the job provides, and the way jobs c, and d). With the exception of low-income connect people may be more relevant for social countries, the relationship between unemploy- cohesion. ment and active membership in an association Further indications of a connection between is significant and negative. The mixed findings jobs and social cohesion comes from looking on trust and unemployment underscore that at job characteristics. The 2005 wave of the unemployment may not always be a meaningful World Values Survey asks people whether their concept in low-income countries. Open unem- jobs involve manual or cognitive, routine or cre- ployment is frequently low in developing coun- ative tasks, and how much independence they tries and is not always concentrated among the have at work. An index of these self-­ reported worse-off, because most people work to make characteristics captures how motivating a job ends meet in the absence of social safety nets. is. The index is positively associated with trust Jobs and social cohesion    129 Trust F I G U R E 4 .1   and civic engagement go together with peaceful collective decision making a. Trust b. Civic engagement 0.8 0.9 0.7 0.8 share reporting trust in people 0.7 0.6 index of civic engagement 0.6 0.5 0.5 0.4 0.4 0.3 0.3 0.2 0.2 0.1 0.1 0 0 –2.0 –1.5 –1.0 –0.5 0 0.5 1.0 1.5 2.0 –1.5 –1.0 –0.5 0 0.5 1.0 1.5 2.0 index of peaceful collective decision making index of peaceful collective decision making Sources: World Development Report 2013 team, based on World Values Survey 2005 (database), World Values Survey Association, Stockholm; Worldwide Governance Indicators 2005. Note: The analysis includes 56 countries (panel a) and 49 countries (panel b). “Index of peaceful collective decision making� is an average of indicators of “voice and accountability� and “political stability and the absence of violence� from the Worldwide Governance Indicators. “Index of civic engagement� is the average of responses to questions from the World Values Survey on (a) active membership in associations; (b) whether the respondent participated or would participate in a demonstration; and (c) whether the respondent would sign a petition. in high- and upper-middle-income countries ciety. Moreover, trust and civic participation are (figure 4.3a). This relationship is not significant influenced by peer and social interaction effects in lower-middle- and low-income countries. (such as the trust or participation of others), Holding a job with perceived cognitive, creative, which can make it difficult to draw conclusions. and autonomous attributes is positively linked While the primary focus is on how jobs can with civic engagement indicators in all but contribute to trust and civic engagement, this re- low-income countries (figures 4.3b, 4.3c, and lationship goes in both directions. There are ways 4.3d).9 Similarly, in surveys conducted in 2012 in which social cohesion can affect jobs. Trust and in China, Colombia, and the Arab Republic of social capital (an element of civic engagement) Egypt, workers who perceived that their jobs may create an economic and political environ- involved more autonomy and greater creative ment that is conducive to economic growth.11 and cognitive content were more likely to report Trust can reduce transaction costs and overcome helping other people.10 market failures that arise because of uncertainty; it can reduce costs related to search and informa- tion, policing and enforcement, and bargaining More than correlations? and decision making; and it can be the basis for As suggestive as they are, these relationships be- the transmission and exchange of knowledge tween jobs and social cohesion do not establish and allow for innovation, coordination, and co- causality. While unemployed people may be less operation among firms.12 Meanwhile, factors likely to trust others or join associations, people such as mistrust, discrimination, fragmentation with less trust in others may also be more likely along ethnic lines, or inequality can also influ- to be unemployed or not participate in civil so- ence whether jobs are created, and what kind. 130  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 People F I G U R E 4 . 2   who are unemployed trust and participate less a. Trust and unemployment b. Active membership and unemployment 0.1 0.04 0 N marginal probability marginal probability N –0.1 N 0 N N –0.2 N –0.04 –0.3 N –0.4 –0.08 N –0.5 –0.12 –0.6 high upper lower low high upper lower low income middle middle income income middle middle income income income income income c. Demonstration and unemployment d. Petition and unemployment 0.02 0.02 0 0 marginal probability marginal probability –0.02 –0.02 N N N N N –0.04 –0.04 N N –0.06 –0.06 N –0.08 –0.08 –0.1 –0.1 high upper lower low high upper lower low income middle middle income income middle middle income income income income income Source: Wietzke and McLeod 2012 for the World Development Report 2013. Note: The analysis includes 54 countries. The vertical axis shows the marginal probability (d-probit coefficient) of individuals’ self-reported trust or civic engagement on being unemployed. The estimates control for the income, education, and demographic characteristics of respondents. Trust is based on the question, “Generally speaking, would you say that most people can be trusted or that you need to be very careful in dealing with people?� Civic engagement variables are (a) whether the respondent is an active member of one or more of nine different associations; (b) whether the respondent attended or would attend a demonstration; or (c) whether the respondent signed or would sign a petition. The line indicates the 95 percent confidence interval of each coefficient. If the line crosses the horizontal axis, the corresponding coefficient is not statistically significant. Evidence of a directional link between em- Controlling for other factors, men and ployment status and civic engagement comes women who were working in 2000 but not in from a survey in Indonesia that tracked partici- 2007 were less likely to be participating in com- pation in community meetings and volunteer munity activities than others. Conversely, men activities and interviewed the same respondents and women who were not working in 2000 but in 2000 and 2007.13 On average, participation were working in 2007 were significantly more in community activities increased 8 percent likely to be involved in community activities during the period, but it increased at different than those who were not working in 2007.15 rates among people with different work histo- Reasons not controlled for in the analysis could ries (figure 4.4).14 explain these findings; for example, people who Jobs and social cohesion    131 People F I G U R E 4 . 3   with motivating jobs trust and participate more a. Trust and motivating job b. Active membership and motivating job 0.03 –0.02 N N N 0.02 0.01 marginal probability marginal probability N 0 0.01 N N N –0.01 0 N –0.02 –0.01 high upper lower low high upper lower low income middle middle income income middle middle income income income income income c. Demonstration and motivating job d. Petition and motivating job 0.1 0.02 0.08 N marginal probability marginal probability N 0.06 N N 0.01 N N 0.04 0 N 0.02 N –0.01 0 –0.02 –0.02 high upper lower low high upper lower low income middle middle income income middle middle income income income income income Source: Wietzke and McLeod 2012 for the World Development Report 2013. Note: The analysis includes 54 countries. The vertical axis shows the marginal probability (d-probit coefficient) of individuals’ self-reported trust or civic engagement on an index assessing whether respondents think their job is cognitive, creative, or independent. The estimates control for the income, education, and demographic characteristics of the respondents. Trust is based on responses to the question “Generally speaking, would you say that most people can be trusted or that you need to be very careful in dealing with people?� Civic engagement variables are whether the respondent is an active member of one or more of nine different associations and whether the respondent attended or would attend a demonstration, or signed or would sign a petition. The line indicates the 95 percent confidence interval of each coefficient. If the line crosses the horizontal axis, the corresponding coefficient is not statistically significant. get sick lose their jobs and their ability to partici- employment can break economic and social pate in the community. New cross-country anal- ties, breed mistrust, and damage people’s sense ysis from Europe and Latin America suggests a of community and hope for the future. Not hav- casual relationship between employment status ing a job can mean losing social status as well and trust in others and institutions (box 4.2).16 as not being able to provide income for one’s The empirical results relating unemploy- family. A man laid off after 24 years of work in a ment, trust, and civic engagement imply that factory in Serbia explained, “I automatically lost losing a job means more than losing income. everything. I lost any freedom and power I had. Job loss can undermine feelings of self-worth Every­ thing was lost.�17 Ethnographies of com- and strain family and social relationships. Un- munities in Argentina, Bulgaria, and Guyana 132  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 that experienced widespread job losses in con- Having F I G U R E 4 . 4   a job means more community texts of limited new job creation are remarkably consistent in their accounts of the social impli- participation in Indonesia cations of long-term unemployment (box 4.3). employment status For communities, job loss appears to foster mistrust not only toward former employers or not working in 2000, government authorities suspected of being in- working in 2007 different or responsible for the lack of employ- ment opportunities but also among neighbors, former colleagues, and friends. This frustration working in 2000 may contribute to general dissatisfaction with and 2007 the political environment. An empirical study using the World Values Survey in 69 countries finds that joblessness can be linked with nega- not working in tive views about the effectiveness of democracy 2000 or 2007 and preferences for a rogue leader.18 Insecure jobs or jobs that people find demoralizing can have effects similar to those of unemployment. working in 2000, The lack of status, job security, or voice at work not working in 2007 can lead people to feel disempowered and hope- less about the future and to stop participating in 0 1 2 3 4 5 6 7 8 social networks.19 % change in community participation, 2000–07 In extreme cases, unemployment can con- tribute to violence or social unrest. Youth in particular may turn to gangs or other violent Source: Indonesia Family Life Survey (database), Rand Corporation, Santa Monica, CA. Note: Community participation includes joining in a community meeting; cooperative, voluntary labor; groups to compensate for the lack of ties in eco- neighborhood improvement; neighborhood watch (men); or women’s association. nomic and social life.20 A longitudinal study of youth in Ecuador found that members of gangs BOX 4.2   Do jobs cause trust? Analysis of Eurobarometer and Latinobarómetro Surveys An analysis using the Latinobarómetro and Eurobarometer values institutions over time. The estimated effects measure how a per- surveys during the 2000s makes it possible to study the evolution of centage change in, say, the unemployment rate for a cohort in a trust and jobs and links in both directions. The surveys include ques- given year predicts changes in the percentage of individuals of that tions on interpersonal trust and trust in institutions. Cohorts are same cohort reporting to trust in the subsequent year. defined and examined in the different survey years. The analysis This analysis finds that increases in unemployment are fol- looks at how social cohesion and employment conditions for the lowed by increases in trust among Europeans, but the opposite is cohorts evolve over time, controlling for certain country characteris- true among Latin Americans. At the same time, increases in self- tics that could be correlated with both trust and employment employment lead to higher trust in Europe while the opposite is status. true in Latin America. These results hold in Latin America for trust The dataset captures important features of the formation of both in government and in others. Conversely, the analysis finds social cohesion, because perceptions of trust and civic participation little evidence of a causal link from trust to jobs, except for a small are highly influenced by peer and social interaction effects. For negative impact of self-employment on trust in government in instance, an individual’s propensity to trust other people or the Latin America. These results may reflect the higher coverage of state depends on the perceived or actual trust of others belonging social protection in Europe and the lower importance of open to similar sociodemographic groups. unemployment in Latin America than in Europe. They are consis- The model simultaneously allows group level job conditions, tent with evidence that in Latin America self-employment, while a including unemployment and self-employment, to influence trust last resort for many unable to find wage employment, is valued by and vice versa. The empirics quantify how earlier changes in group- some for the independence it provides.a level employment conditions predict their trust in society and its Source: Arias and Sosa 2012 for the World Development Report 2013. a. Perry and others 2007. Jobs and social cohesion    133 BOX 4.3   Displacement and unemployment can lead to the erosion of trust and ties Downsizing of bauxite mines in Guyana state jobs entailed the loss of numerous benefits, including health The downsizing of bauxite mines in the absence of new opportuni- care and job security. In focus groups, people linked job loss to poor ties has contributed to a deterioration in family and community health, social isolation, and crime. Older men, in particular, lost face relationships in Linden, Guyana.a Between the early 1970s and the when they had to ask young relatives or employers for work. mid-1980s, bauxite mining near Linden was cut by half, and layoffs Unemployment did more than simply weaken social ties; it continued throughout the 1990s. By 1999, formal unemployment in created distrust and mutual suspicion. The restructuring created ­ Linden stood at about 40 percent, and residents complained of ris- winners as well, and they also suffered from the mutual distrust. ing crime. People who had lost their jobs began to avoid traditional social Once among the best-paid workers, miners were respected for gatherings because they were unable to afford gifts that they were their work and seen as drivers of the economy. People felt particu- expected to provide. People felt that security—once linked to larly demeaned by the downsizing process: “The people off the good health, the opportunity to pursue personal and professional job don’t get any information. They treat us like we don’t exist. fulfillment, good personal relations, respect in the community, and Yet . . . before we came off, there used to be meetings with us, social cohesion—had moved out of reach. In communities that [about] what was happening.�b were once relatively equal, people identified five or six levels of Material hardship and insecurity took a harsh toll on identity well-being. and the relations between men and women. Women directly linked men’s inability to retain their authority as breadwinners to domes- Economic reforms in Argentina tic violence. “Especially in cases of abuse, you would be surprised La Matanza is a city of 1.2 million outside Buenos Aires that was that after counseling them, the problem comes right back to the once a manufacturing center of textiles, diesel engines, household economic situation. The man can’t provide adequately for the appliances, and steel. Economic transformations in the 1990s led to home.�c Indigence was linked to shocking forms of child neglect increased reliance on technology and skilled workers. Factories in La and abuse. Some parents were said to be prostituting their chil- Matanza closed, and job opportunities became scarce. With mobil- dren. The cultivation of cannabis, the use of cocaine, and involve- ity low, people had to take up temporary or casual jobs without ment in the international transshipment of drugs were said to be unemployment or health insurance. Those who managed to find rapidly increasing among young men. jobs complained of exploitative pay, abusive treatment, and assaults Downsizing also diminished the economic resources available to their dignity. to community organizations such as churches: “The churches are in As elsewhere, job loss affected men and women’s relationships. crisis also. As individuals, we are part of the crisis, so we carry it into While some men adapted to a more egalitarian role, many re­­ church and it in turn goes into society,� one person said.d sponded to the blow to their self-esteem as breadwinners with depression or anger; women complained that violence in the Regime change and unemployment in Bulgaria household increased. Men felt joblessness undermined their roles Bulgaria massively downsized its unsustainable state enterprises in the family. following the end of the Communist regime. The disappearance of Source: Dudwick 2012 for the World Development Report 2013. a. World Bank 2004a. b. World Bank 2004a, 26. c. World Bank 2004a, 29. d. World Bank 2004a, 53. involved with drugs and guns had joined “be- placed persons, can be particularly disorient- cause they were searching for the support, trust, ing. It can influence status and identity, for and cohesion—social capital—that they main- example, for migrants who had better jobs in tained their families did not provide, as well as their places of origin. The social effects of un- because of the lack of opportunities in the lo- employment among dislocated populations cal context.�21 Similarly, analysis in the United may be especially isolating for people lacking States has found that gangs provide youth with family or other ties in their new communities. the income, respect, and social ties that they It can have implications for psychological well- were unable to find in jobs, particularly given being, as well as the ability to participate in civil the limited opportunities available in cities such society. Even migrants who find work may be as Chicago and New York that had lost stable, vulnerable if their jobs do not provide adequate unionized manufacturing jobs.22 channels to integrate within the new society or The lack of jobs among dislocated popula- if the migrants lack voice or information about tions, including migrants, refugees, and dis- their rights. 134  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Jobs (or the lack of jobs) can shape child care, education, biotechnology, and manu- social interactions facturing define job competencies and give em- ployees the chance to develop skills, participate The link between social cohesion and jobs is not in training, and increase their responsibility. necessarily direct or linear. Interactions between Results from a program implemented in nurs- jobs and societies are contextual and multidi- ing homes in Massachusetts in the United States mensional; effects can be positive as well as nega- found that having opportunities for growth im- tive. Having, or not having, a job can influence proved communication and teamwork, reduced how people view themselves in relation to others, turnover, and built self-respect and confidence with implications for values, attitudes, and be- among staff.25 haviors. Jobs can connect people with informa- Jobs can have similar effects for low-wage tion, economic activities, and other people. And workers in developing countries, and these ef- how jobs are allocated can affect whether people fects can have implications for social cohesion. think their society is fair and merit-based, believe The growth of the garment sector in Bangla- they have a stake in society, and have expecta- desh brought more than 3 million women into tions and aspirations for the future (Question 4). the workplace. Although the factory jobs were physically demanding and poorly paid, they ex- panded women’s autonomy and increased their Jobs provide social identity opportunities to participate in public life.26 “I Some jobs can contribute positively to how peo- am braver now,� a 26-year-old worker explained, ple view themselves and their relations with oth- “I understand more things which I did not be- ers. The identity conveyed by a job can influence fore.�27 Observers noted that the sight of women the social categories that individuals associate walking back and forth to work changed popu- with, their behaviors, and the norms that shape lar notions about the acceptability of women in this behavior.23 In industrial countries, jobs that the public space and their right to access public give people opportunities to learn and develop institutions.28 Coworkers travel together, share careers can be motivating and strengthen iden- information about work opportunities, and tity. In the United States, programs that provide form savings groups.29 skill development and growth opportunities to The effect of jobs on identity also holds low-wage workers aim to strengthen self-esteem for self-employed workers, including farmers and motivation.24 Public and private sector ini- (box 4.4). Jobs that provide access to voice can tiatives to establish career ladders in health care, be empowering and give workers a stake and shared interest in their work.30 Informal workers lack access to representation on the job and are BOX 4.4  Jobs, motivation, and identity in Risaralda,  similarly excluded from local government and Colombia economic associations. Associations of self-em- ployed workers and farmers help fill these gaps.31 David is the owner of a small shop in one of the rural areas of Risaralda. He was A core strategy of the Self Employed Women’s born in another part of Colombia but has lived in the region of Risaralda for Association (SEWA) in India has been to em- some years now. He loves living in this area for the safety and peace that exists. power its members and partners by increasing He has had his shop, located next to the main street of his village, for around their say in communities (box 4.5). 15 years. One of the things he loves the most about his job is the deep sense of belonging to the community that it offers. When necessary, people come to his Jobs connect people shop and ask for credit for the goods they need. Despite a few unpaid bills some of his customers have left him, he does his best to help the villagers. The Some jobs bring people into contact with others income from the shop provides only enough to subsist, and it is necessary for whom they might not otherwise encounter, in- him to engage in other businesses so that he can have an additional income. He cluding people of different ethnicities and social feels that his shop is a way of giving back to the community and that by being backgrounds (box 4.6). This connecting aspect there, he is able to provide for the needs of his neighbors. of jobs can contribute to social cohesion. Jobs can create opportunities for repeated interac- tions focused on tasks leading to interdepen- Source: Bjørkhaug and others 2012 for the World Development Report 2013. dent relationships.32 A study of political views Jobs and social cohesion    135 BOX 4.5  Voice can be extended to the self-employed: The case of SEWA  The Self Employed Women’s Association (SEWA) began in 1972 with loans, and insurance), social services, infrastructure, and training a small group of migrant women cart pullers in the wholesale cloth and capacity building. market of Ahmedabad City in Gujarat, India. These women worked Of particular concern to SEWA is the fact that the working poor, as head loaders, carrying clothes to and from the wholesale market. especially women, do not have a voice in institutions that set the They were paid on a per-trip basis, regardless of the distance they rules that affect them. The association seeks to expand the voice of traveled or the weight they carried. Often, they were not paid the its members at the local level through representation and by build- full amount they were owed because no records were kept. Ela ing capacity to participate in local councils; municipal, state, and Bhatt, head of the Women’s Wing of the Textile Labor Association, national planning bodies; tripartite boards; minimum wage and helped organize the group and negotiate with the cloth merchants other advisory boards; sector-specific business associations; and to gain fair treatment. local, state, and national labor federations. SEWA is now a member of the International Confederation of Over the past decade, SEWA has also inspired or cofounded Trade Unions and has become a model for associations of informal national and regional networks of home-workers in other parts of workers internationally. In 2011, SEWA had more than 1.3 million South and Southeast Asia, national networks of street vendors in members across India, of which over 820,000 were in Gujarat, while India and Kenya, and international networks of domestic workers the rest were in eight other states. The members are drawn from and waste pickers. While some of these networks and organizations multiple trades and occupations and from all religious and caste remain weak, most have been able to collaborate, leverage re­­ groups. sources, and influence policies. The regional and international net- SEWA stresses self-reliance and promotes organizing around the works of domestic workers, home-based workers, street vendors, central strategies of work security, income security, food security, and waste pickers have secured two international conventions (for and social security. Primarily a trade union, SEWA now engages in home-workers and domestic workers) and policies, laws, or legal a wide range of other areas, including leadership development, judgments in several countries. collective bargaining, policy advocacy, financial services (savings, Source: Chen and others 2012 for the World Development Report 2013. and the workplace in the United States finds that likely to witness ethnic violence, while riots were cross-cutting interactions at work lead to greater more frequent in cities with fewer economic ties. awareness of the rationales for views other than The existence of civil society organizations, such one’s own and for “exposing people to political as clubs, political parties, labor unions, and busi- dialogue across lines of political difference.�33 In ness associations, contributed to reducing vio- a survey of 200 managers, owners, and sales rep- lence. But economic interests provided a com- resentatives in Trinidad and Tobago, 81 percent mon motivation for community members of of the interviewees reported that their working both groups to participate in these associations.37 lives brought them into contact with people of a Jobs can also play a connecting role out- wider range of races than did their social lives.34 side of urban environments. Studies in Ghana Interactions through jobs can contribute to and Uganda illustrate how farmers connected greater trust and positive interdependence be- through networks can access information and tween groups. In the 18th century, Montesquieu increase productivity. In Ghana, pineapple farm- wrote that “the natural effect of commerce is to ers adjusted their use of fertilizer in response bring peace. Two nations that negotiate between to the successful or unsuccessful experiences of themselves become reciprocally dependent, if their neighbors. Farmers who were starting to one has an interest in buying and the other in cultivate pineapples were more likely to make selling.�35 Relations through jobs, whether built changes based on information they had received through trade or other transactions, can influ- from other farmers, showing the potential that ence social relations. on-the-job interactions and learning from oth- A 2001 study of multiethnic cities in India ers can have.38 In a qualitative survey, youth in suggests that economic interdependence, in- Ghana who were asked about the characteristics cluding through jobs, can reduce the incentives that would make a job attractive emphasized for violence between communities.36 Cities with the importance of jobs as opportunities to meet more interlinked economic relations were less new people and build social networks.39 136  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 BOX 4.6  Some jobs connect people across ethnic boundaries Surveys carried out across the world illustrate the ways jobs can connect native. It is, in the strictest sense, a medley, for they mix but do not people from different backgrounds. combine. Each group holds to its own religion, its own culture and language, its own ideas and ways. As individuals they meet . . . in the “In Sadakhlo market in Georgia, next to the borders with Armenia market place, in buying and selling.�d and Azerbaijan, one does not hear the virulent expressions of In ancient Cordoba, Spain, the marketplace represented “the mutual hatred one can hear a few miles away across the border. place of encounter over and above the gender, tribal, and faith ‘They fight, we don’t,’ says Mukhta, a trader from Azerbaijan, while divides that constituted Islamic urbanization.�e putting his arm round his Armenian colleague Ashot.�a “You don’t reconcile in a vacuum. There must be a practical pro- “According to one of the stallholders at Ergneti market, on the gramme; there must be something that brings people together. As disputed border between South Ossetia and mainland Georgia, they work together, cleaning the coffee, they talk together so they ‘There are no political questions here. The market has one language: start talking business but later they start talking family affairs. It fos- economic. That is it.’�b ters relationships and reconciliation.� f “In Guinea, members of the Malinke ethnic group are wholesal- “If I wasn’t in this job, I might have only Indian friends or African ers in the groundnut market chain, while the primary producers of friends,� said a sales manager for a processed food manufacturer in groundnuts tend to be Guerse. Malinke wholesalers and Guerse Trinidad and Tobago. “But now I have plenty, plenty friends. White farmers are willing to trade with each other. This is helping over- friends in Mayaro. Chinese friends in Port-of-Spain. And real close. come ethnic and religious tensions. . . . This willingness to trade is Closer than if you born with someone, your next-door neighbor. due to the mutually recognized possibility of profit.�c And that’s why I wouldn’t swap this job for anything else.�g “In Burma, as in Java, probably the first thing that strikes the ­ visitor is the medley of peoples—European, Chinese, Indians, and Source: Kilroy 2012 for the World Development Report 2013. a. The Economist 2000. b. Voice of America 2002. c. Spilsbury and Byrne 2007. d. Furnivall 1948, 304–12. e. Vicente-Mazariegos-Eiriz 1985, 763, cited in Briggs 2004, 326. f. Fatuma Ngangiza, Unity and Reconciliation Commission of Rwanda, quoted in BBC News 2006. g. Kilroy 2011. An experiment among farmers in rural be sufficient to build trust or change behaviors Uganda found that subsistence cotton farmers and contribute to social cohesion. The literature using social networks can change existing social on prejudice suggests that contact across groups interactions with beneficial results. The ran- can alter people’s perceptions of others.41 There domized intervention compared the impact of may also be risks. If cooperation through jobs training on agricultural productivity with the fails, tensions between groups may flare, par- impact of being paired with another farmer. ticularly if the groups have previously been in The pairs were encouraged to discuss farming conflict and blame each other.42 activities, problems, and solutions and to set a While networks connect people in positive target for increases in cultivation. The interven- ways, they can also exclude. Surveys in indus- tion encouraged exchanges of information and trial and developing countries consistently find learning by expanding farmer networks. Farm- that people obtain jobs through acquaintances. ers who participated in the project, especially Yet, reliance on networks may have negative women, significantly increased their produc- social consequences if people and groups who tivity. Connecting farmers with people outside lack such connections are left out. In Morocco, their established social circles helped spread in- after controlling for education, social status, and formation that would not otherwise have been other factors, people whose fathers did not have shared.40 formal sector jobs were significantly less likely to Jobs may not always help overcome differ- obtain formal sector jobs themselves.43 In addi- ences and tensions between groups. While in- tion to unfairness in access to jobs, family con- centives inherent in jobs can provide people nections can also influence labor earnings. For with motives to interact across gender, caste, example, in Brazil, sons’ wages are influenced by and ethnic boundaries, these incentives may not those of their parents.44 Jobs and social cohesion    137 The exclusionary nature of networks is high- Respondents at a focus group explained that, lighted by the experience of migrants moving “to get jobs, one needs someone to speak for from rural areas to cities. Migrants often choose him, particularly from Sana’a.� Young people re- destinations where they have connections. But ported that inheriting a civil service post from if they do not, they can be uprooted from fam- one’s father was not viewed as wrong under the ily and community ties that provide economic country’s civil service rules.47 and social support, including access to jobs. Jobs that are allocated based on connections They may also lack the information needed to and other circumstances beyond the control integrate into their new destinations. Migrants of an individual can influence whether people moving across borders or regions, internally dis- view society as fair. Recent work on the mea- placed persons or refugees fleeing from conflict surement of inequality of opportunities exam- or returning after a peace agreement, and sol- ines the extent to which access to basic services diers demobilized after conflict may be particu- that are essential for human development, such larly vulnerable to exclusion from job opportu- as education, health, nutrition, and sanitation, nities. This is a concern in conflict situations as is based on circumstances of birth or arises be- well as in contexts of structural transformation, cause of inequality within society (box 4.7).48 when massive numbers of people move from Application of this approach to access to jobs rural to urban areas.45 Networks also do not considers the extent to which opportunities are reach many among the self-employed, especially related to circumstances at birth, including gen- home-based workers who work in isolation and der, ethnicity, and parental educational attain- domestic workers who lack opportunities to in- ment and political affiliation, or to attributes, teract with others. including educational attainment and age.49 Re- sults from 29 countries in Europe and Central Asia indicate that inequality across groups based Jobs influence aspirations and on circumstances and attributes varies between expectations 3 and 20 percent. The share of inequality attrib- The various ways in which jobs are distributed utable to circumstances is substantial in most can affect expectations and aspirations and in- cases, contributing to more than half of the fluence whether people believe that they have a overall inequality (figure 4.5). stake in society. The jobs that other people have Circumstances at birth contribute the most can contribute to an individual’s values, atti- to inequality in Azerbaijan, followed by Uz- tudes, and behaviors. Children’s goals for the fu- bekistan, Georgia, Turkey and Albania. In these ture may be influenced by whether their parents countries, such factors contribute the most to have jobs or not, as well as by the types of jobs inequality in access to jobs. Education plays an their parents have. Frustration and even social outsized role in inequality in some countries— unrest may develop when education and effort Armenia stands out in particular, along with Al- are not rewarded or when people perceive the bania, Bulgaria, and Romania. distribution of jobs to be unfair. Similar analysis for 18 countries in Latin The Arab Spring was as much or more America using the 1990 Latinobarómetro sur- about political voice as it was about jobs. Yet vey confirms these findings. On the whole, the widespread disappointment, especially among education of the worker and the circumstances youth, about the lack of job opportunities and he or she was born into play important roles frustration with the allocation of jobs based on in explaining inequalities in access to jobs, and connections rather than merit echoed across the role of education is especially important for countries. A young person in Egypt commented, regular employment in the formal sector.50 “To work in a big company, you’ve got to have wasta [connections; literally, a middleman]. Re- * * * gardless of your qualifications, you must search for someone to secure the job for you. In some The interaction of jobs and social cohesion is not cases, you have to pay money.�46 Social assess- linear or simple to disentangle. This is an emerg- ments in the Republic of Yemen documented ing area for further research across disciplines. frustration with the allocation of jobs based on The effect of jobs on trust and civic engagement tribal, family background, or party affiliation. at the individual level suggests that exchanges 138  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 BOX 4.7  Measuring inequality of opportunities in access to jobs  The approach butes), and which circumstances contribute the most. Circumstances The concept of equality of opportunity, which can be traced back to can affect access to a job involving more than 20 hours of work a John Rawls and Robert Nozick,a stems from the idea that an individ- week through direct and indirect channels. An example of a direct ual’s chances of success in life should not be caused by circum- channel is when belonging to a minority group can affect the chance stances that are beyond the individual’s control, such as gender, of getting a job. As for indirect channels, circumstances can influ- ethnicity, location of birth, or family background. John Roemer’s ence the education of a person, which, in turn, influence the chance 1998 work formalized the principle of equality of opportunity and of getting a job. The decomposition of D is intended to measure argued that policy should seek to equalize opportunities indepen- the direct channel, which is to say the inequality attributable to the dent of circumstances.b Empirical applications of this concept use predetermined circumstances, net of the effect attributed to differ- different measures of opportunity and estimate the extent to which ences in education and experience among workers.f inequality arises because of circumstances at birth, rather than indi- vidual attributes such as effort or talent.c Caveats The Human Opportunity Index (HOI) is one approach that is A number of questions complicate the exercise and act as caveats to being used across countries and regions to analyze the opportuni- the analysis. First, how should opportunity be defined in terms of ties available to children in terms of access to basic goods and ser- access to jobs? People have different preferences about jobs, so vices such as sanitation, clean water, electricity, and basic education.d part of the measured inequality may reflect voluntary choices The HOI captures both the extent to which societies provide these rather than a lack of access. And people with certain circumstances goods and services and how equitably access to them is distributed and attributes may be more (or less) likely to be in the labor force in among groups with different circumstances in a society. the first place. Second, which circumstances should be considered? Recent work has tested the application of the HOI methodology The data only report a limited range of them, and some may simply to jobs in Europe and Central Asia and Latin America using data not be observable. Gender, minority status, and parental education from the 2006 Life in Transition Survey and the Latinobarómetro are commonly considered in the literature; and whether parents Survey.e In this case, opportunity is defined as having a job involving were affiliated with the Communist Party can be a proxy for social more than 20 hours of work a week; circumstances are the gender of status in the European and Central Asian countries, even many the individual, the educational attainment of the father, parents’ years later. However, parental education may be correlated with past affiliation in the Communist Party (in Europe and Central Asia), unobserved abilities of an individual. Controlling for the individual’s and self-reported minority status; and attributes are educational education level partially resolves this problem but does not address attainment and age. Those lacking opportunity are people working the possibility that among children who receive the same educa- fewer than 20 hours a week, the unemployed, and those who want tion, children with educated parents may acquire better skills stem- to work more. ming from unobservable inputs. The methodology makes no The HOI is the coverage rate of the opportunity, adjusted for assumptions about missing circumstances, which are likely to exist inequality between groups defined by circumstances and attributes. since information on all circumstances is not typically available Inequality is measured by a “dissimilarity index� (henceforth, D), from the same survey. The inequality or dissimilarity index has the which reflects the share of available opportunities that would have property that the index will always increase with the addition of to be reallocated to achieve the same coverage rate of opportunity more circumstances or attributes. Despite these caveats, this across all groups. A decomposition of D indicates how much circum- approach is a first step in applying the inequality of opportunity stances contribute to inequality between groups (relative to attri- analysis to access to jobs. Source: Abras and others 2012 for the World Development Report 2013. a. Nozick 1974; Rawls 1971. b. Roemer 1998. c. Roemer and others 2003. d. Paes de Barros and others 2009. e. Life in Transition Survey I (database), European Bank for Reconstruction and Development, London. f. Estimating the indirect channel—the effect of circumstances through education—would be difficult because education depends on a host of factors other than the circum- stances on which information is available. Moreover, excluding the impact of circumstances through education is justified because we are interested in measuring the extent to which inequality in access to jobs is attributable to circumstances. While circumstances may have influenced educational attainment as well, these effects would have occurred at a much earlier stage of life (primarily in childhood) and therefore do not reflect inequality of opportunities specific to jobs. and relationships established through jobs can contribute more to social cohesion than others. have broader effects on societies, including how What matters is not necessarily whether people they manage tensions between groups and col- have a job but whether the job and its charac- lective decision making. But some jobs may teristics can contribute to social cohesion. In Jobs and social cohesion    139 FI G U R E 4. 5  Inequality of job opportunities varies across countries 20 16 12 D-index 8 4 0 Se n ec Es ia pu a Be lic kh s an ov C ia Re tia Uk blic Hu ine Po ary d Sl nd eg ia on a hu ia M jiki a ac en n rg ni ro pu R Bu lic Tu ia M key Ge va Ro rgia kis a Al an er nia Ar jian ia za ru Re ni M vin Ta ani be ni Re FY tio M ont sta rb tv rz en Lit gol ar en Ky edo eg b b do st t ak roa la h to Uz a Az ba Ka la ng ra r pu ba lg La o ra o m He ov m yz a, ol de Fe n ia Cz Sl ss an Ru ia sn Bo circumstances age education Sources: Abras and others 2012 for the World Development Report 2013; based on the data from the Life in Transition Survey I (database); European Bank for Reconstruction and Development, London. Note: Opportunity is defined as having a job with 20 or more hours a week. Circumstances include gender, ethnicity, and parental education and political affiliation. The D Index is the share of available opportunities that would have to be reallocated to achieve the same coverage rate of opportunity across all groups. certain contexts, jobs can transform societ- influence identity, connect people through ies if they influence social identity and social networks, and increase a sense of fairness and norms; if they shift bargaining power within meritocracy in access to jobs have the potential households, communities, or society; or if they to contribute to social cohesion. alter power relations between groups. Jobs that 140   4 WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 QUESTION Can policies contribute to social cohesion? In Rabat, Morocco, unemployed college gradu- programs themselves are subject to corruption ates gather daily in front of government build- and governance risks. These negative experi- ings to protest the lack of jobs.51 In Juba, South ences may reveal poor program design, however, Sudan, the fledgling government faces the chal- rather than prove the impossibility for jobs poli- lenge of demobilizing 150,000 combatants and cies to contribute to social cohesion. reintegrating large numbers of internally dis- placed persons after conflict.52 For policy mak- Access to information, rights, and voice ers in countries with high youth unemployment and in countries affected by conflict, expanding Policies can take social cohesion into account by job opportunities has urgency for social and po- expanding opportunities for groups who face litical reasons, as well as for economic reasons. barriers to getting jobs and increasing access In industrial and developing countries alike, to voice and rights. People may feel frustrated the conventional wisdom is that having a job is if they perceive that jobs are allocated on the what matters for social cohesion—how societies basis of privilege and connections rather than peacefully manage collective decision making. merit and achievement. Increasing fairness and The idea that jobs can build identity, or might equality of opportunity for jobs involves in- be associated with trust or more participation forming the public about jobs and how to get in society, is often seen as relevant only for a them, and about the existence of legal mecha- narrow set of occupations in rich countries. nisms, such as antidiscrimination laws and af- Those jobs are perceived as a luxury that devel- firmative action programs to reduce discrimina- oping countries cannot afford. Even those who tion and support the inclusion of groups who concede that some jobs can do more for social lack access. But having laws on the books is cohesion in developing countries are skeptical not enough. Increasing fairness involves institu- that policies can do much beyond supporting tions for enforcement, and redress mechanisms job creation. Given that most employment is in for accountability. Although such measures can the private sector, it is unclear how or whether be motivated by multiple objectives, including the government could influence the nature of poverty reduction, they can also be considered the jobs. Some even doubt that jobs on their from a social cohesion perspective. own lead to greater social cohesion. They view Transparency and access to information jobs as only one element that can contribute to about jobs can increase fairness and equality changing values, attitudes, and behaviors within of opportunity by ensuring that vacancies are a complex web of institutional, historical, politi- widely publicized, together with information cal, and social factors. Given this multiplicity of about accessing public employment programs. influences, engineering social cohesion through Access to information about rights is similarly jobs is not an option. important for ensuring that labor practices are Negative experiences with publicly funded fair. Farmers, self-employed workers, and work- employment programs give some justification ers without formal labor contracts are often not to this skepticism. Temporary employment pro- knowledgeable about their rights in relation to grams that place people into dead-end jobs with land owners, traders, local authorities, and em- no hope for future employment may do more ployers, or about their options for appeals. Civil harm than good.53 Similarly, demobilization society organizations such as cooperatives, asso- programs in post-conflict environments risk ex- ciations of informal workers, and trade unions acerbating tensions between former opponents can disseminate information about rights and through divisive targeting.54 Social cohesion the channels to voice grievances.55 is actually undermined when jobs in publicly A related challenge is the extension of effec- funded programs are allocated to friends and tive legal protection to those who work outside relatives of government officials, or when the of legal frameworks. At the international level, Jobs and social cohesion    141 the passage of ILO conventions on domestic and home workers has extended coverage for these BOX 4.8   omestic workers: The journey to an ILO D groups (box 4.8). At the national level, countries such as Zambia and the Philippines include le- convention gal protections of informal workers in domestic Domestic work includes cleaning, cooking, gardening, child care, and elder legislation. Brazil recognizes domestic workers care. The International Labour Organization (ILO) estimates that there are 52.6 within its constitution and has extended social million domestic workers worldwide; other estimates are nearly twice as high. protection, including leave and maternity ben- Women, generally from the poorest sections of society, make up over 80 per- efits to them. The country’s National Social Se- cent of domestic workers.a Many are migrants, and child labor is common, curity Institute provides incentives for employ- especially for girls. Domestic workers, and especially migrants, are excluded from labor and social protection laws in most countries. ers who register domestic workers. Although Domestic workers have long tried to be recognized and included in the difficult to enforce in practice, Brazil, the Czech labor laws of their respective countries. In 2006, domestic worker organizations Republic, the Philippines, and South Africa began to organize internationally with the support of international trade have established minimum wages for domestic unions and nongovernmental organizations representing informal workers. workers.56 Their main demand was recognition and access to rights, including a campaign Similarly, migrant workers tend to fall out- for an ILO convention on labor rights for domestic workers. side legal frameworks. Both sending and receiv- The campaign involved extensive coordination at the country level to mobilize workers and gain support from labor ministries, trade unions, and ing countries can adopt measures to extend legal employers’ associations. As a result of this campaign, the minimum wage for protection. The government of the Philippines domestic workers was raised by 10 percent in Jamaica, and a memorandum of has a mechanism to protect its overseas work- understanding was signed to improve the conditions of Indonesian domestic ers. The government provides them with pre- workers in Malaysia. departure information and support services; it In 2011, the ILO adopted the Domestic Workers Convention and the Domes- has also signed bilateral agreements and memo- tic Workers Recommendation. The convention states that domestic workers are randa of understanding recognizing migrant to be covered under national labor laws and regulations, including those related to social protection programs. workers’ rights with receiving countries. The The process of securing an ILO convention contributed to building the government has also promoted voluntary social capacity of organizations and individual leaders and gained domestic workers security schemes for overseas workers.57 associations status with trade unions. It also created better conditions for rec- The existence and quality of institutions for ognition and enforcement of rights. In March 2012, the government of Singa- accountability can influence the extent to which pore announced that it would require employers to give one day a week off to rights are enforced in practice.58 Legal frame- the country’s 206,000 domestic workers, most of whom come from Indonesia, works rely on the ability of labor ministries, in- the Philippines, Sri Lanka, and India.b spectorates, and courts to handle disputes and hold the parties accountable. Most countries al- Source: Chen and others 2012 for the World Development Report 2013. low labor disputes to be heard in special labor a. ILO 2011a. courts or civil courts. But court proceedings b. Kennedy 2012. can be lengthy, costly, and cumbersome. In re- sponse, some countries have established alterna- tive procedures for dispute resolution, including the obligations of the state. Guarantees are of- conciliation, mediation, and arbitration before ten complemented by laws addressing job seg- court hearings.59 Cambodia introduced an Ar- regation, unequal pay, prejudice in recruitment, bitration Council in 2003 to help manage labor harassment at work, and lack of education and grievances and improve industrial relations in training.60 Affirmative action programs involve the growing garment sector (box 4.9). proactive measures for hiring women, minori- ties, and other groups subject to exclusion.61 Such programs can be mandatory or voluntary Antidiscrimination policies and apply to the public or private sectors. Legal mechanisms such as antidiscrimina- Affirmative action programs can work, but tion laws and provisions for affirmative action pitfalls are many. Evaluations yield mixed re- can facilitate access to jobs for groups who sults.62 The most extensive research is from the are excluded from opportunities or suffer from United States; it finds that programs are most stigma. Most countries have equality guarantees ­ effective when they are temporary and com- within their constitutions, generally covering bined with improvements in recruitment, train- 142  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 gram might have assisted individuals who were already higher up on the skills ladder but not the BOX 4.9  F  rom laws on the books to laws in action in average previously disadvantaged individual.64 Cambodia’s garment sector Hiring quotas for underrepresented groups The garment industry is Cambodia’s largest formal sector employer. By mid- can be enshrined in constitutions, as is the case 2008, the sector had more than 300 factories, with nearly 340,000 workers, 90 for Scheduled Castes and Tribes in India and for percent of whom were women. Labor conditions including low wages, exces- Bumiputras in Malaysia. Both countries have sive overtime, poor occupational health and safety, child labor, and antiunion shown that quotas work well but can become practices emerged as a major issue as the sector expanded. The initial response politicized.65 Quotas supported through specific was passage of a new labor law in 1997. Enforcement was poor, however. The programs have been successful. In Bangladesh, Labor Inspectorate lacked credibility; inspectors were underpaid and underre- for instance, where women’s employment rates sourced, and were seen as subject to influence. The courts were perceived as corrupt and unresponsive to the needs of workers or employers. As a result, were extremely low, the Employment Gen- strikes and demonstrations increased, and major international brands raised eration Program for the Poorest put in place concerns about the viability of operating in Cambodia. a 30 percent quota for women. This doubled In this context, Cambodia concluded a 1999 bilateral trade agreement with women’s participation in the program within a the United States. Building on a similar clause in other trade deals, the United year, with participants reporting a high level of States agreed to increase Cambodia’s import quota for garments if a semian- satisfaction.66 nual review showed that progress had been achieved in adherence to core international labor standards and standards set in Cambodian law. Following the agreement, the United States funded two International Labour Organiza- Jobs policies can shape social identity and tion (ILO) projects to support the implementation of this clause. The first, which connect people became known as Better Factories Cambodia, involved monitoring working conditions in garment factories. Access to jobs can bolster self-esteem and pro- The second program established an Arbitration Council to prevent and duce benefits for societies beyond incomes. resolve labor disputes. The council’s 30 part-time members were nominated Programs that support employment for at-risk through a process facilitated by the ILO and endorsed by unions, employers’ populations, including youth, can take into ac- organizations, and government. The council conducts mandatory but (gener- ally) nonbinding arbitration of collective labor disputes that cannot be resolved count the ways in which jobs affect peoples’ at- through mediation by the Ministry of Labor. Most disputes handled by the titudes, values, and behaviors and contribute to council involve compliance with labor law related to wages, bonuses, benefits, improved relations between groups. Arguably, and working conditions. Some cases also relate to rights, including antiunion in countries with high youth unemployment, practices, gender equality, freedom of association, and collective bargaining. targeted training programs have the potential Since its establishment in 2003, the council has heard more than 1,200 dis- to be designed to strengthen self-esteem, which putes, 70 percent of which are reported as successfully resolved. Opinion sur- can lead to greater community involvement veys indicate a high level of confidence in the council’s independence and effectiveness. In 2010, the Garment Manufacturers Association of Cambodia and reduced crime and violence. The evidence and major union federations agreed to switch to the council’s arbitration pro- remains limited and tentative, but emerging cedures for disputes over existing labor rights. The result has been an upsurge findings from some training programs targeted in the rate of awards issued by the council and a decrease in the rate at which to youth, including those in post-conflict set- parties are filing objections. Strikes per factory have fallen to their lowest level tings are somewhat encouraging. in 10 years. The Northern Uganda Social Action Fund suggests that combining vocational training, Source: Adler and Hwang 2012 for the World Development Report 2013. life skills, and counseling can increase commu- nity involvement and reduce aggression among youth in a post-conflict setting (box 4.10).67 A ing, and on-the-job training.63 Evidence from reintegration and agricultural livelihoods pro- developing countries is more limited. In South gram for high-risk Liberian youth led to a mod- Africa, affirmative action supporting blacks, est increase in social engagement and a reduc- women, and people with disabilities was com- tion in illegal activities. Participants were also plemented with incentives for firms, including less interested in recruitment into violent activi- access to licenses and contracts. An evaluation ties in neighboring Côte d’Ivoire. The program found that programs had limited impact on had no clear impact on reducing aggression reducing gaps in employment and wages but and violence, however.68 An evaluation of the narrowed differentials at the top of the wage Juventud y Empleo program in the Dominican distribution. This finding suggests that the pro- Republic found that a combination of voca- Jobs and social cohesion    143 tional and life-skills training for unemployed youth can reduce involvement in gangs and de- BOX 4.10   n post-conflict settings, well-designed programs I lay teen pregnancy.69 This is an area for further research; evidence is thin, and few evaluations of reduce social tensions employment and training programs incorporate Opportunities for youth in Northern Uganda social cohesion outcome measures such as com- Two decades of insurgency, instability, and conflict led to high rates of poverty munity participation and conflict resolution. in northern Uganda. By 2005, a measure of peace and stability had returned to Temporary employment programs can pro- the region, allowing for the demobilization and reintegration of former com- vide skills training and access to employment batants and other war-affected populations. In 2006, the government launched for youth at risk and vulnerable populations, the Youth Opportunities Program to stimulate income generation and employ- ment growth among young adults ages 16 to 35. The program provided cash particularly during crises and after conflicts.70 grants for vocational training and business materials to groups of participants These programs have a mixed record in support- with successful grant proposals. Groups had an average of 22 members, and ing employability, because they generally involve most expressed interest in tailoring, carpentry, metal works, mechanics, or jobs with low status that rarely lead to future hairdressing. earnings opportunities. But there are indications An evaluation two years after the intervention found increased investments that programs can be designed to invest in skills in skills, participation in skilled work, greater incomes, and higher savings. with benefits for social cohesion. El Salvador’s Grantees were 4 percent more likely to attend community meetings and 9 ­ percent more likely to be community mobilizers. Participants also reported Temporary Income Assistance Program targets receiving more social support from their family and the community. Further- women and youth in areas with high rates of more, men who received grants reported a 31 percent decline in aggressive violence. Early results suggest that the program behavior relative to the control group. This finding is consistent with theories has increased the self-esteem of beneficiaries and that link aggression to stress levels, low social standing, and perceived injus- reduced the recurrence of violence. 71 tice—all potentially alleviated by higher employment and incomes. Public works programs frequently rely on community participation to identify local proj- Public works in Sri Lanka’s Northern Province ects, providing forums for collective decision In Sri Lanka, a cash-for-work program initially established to re­settle 100,000 returnees following internal conflict actually assisted more than 250,000 making. Community meetings can bring to- returnees and quickly evolved into one of the largest sources of employment in gether people affected by conflict and crisis (box the Northern Province. 4.10). In Rwanda, meetings for the country’s Participants noted that in many cases the program meetings were the first public works program discussed peace building, community-level gathering that they had attended after having arrived from security, community development, and recon- camps for internally displaced populations. By many accounts, community ciliation, in addition to project-related issues. In meetings, shared meals, team work, and the involvement of elders and chil- the Republic of Yemen, fuel shortages and price dren as indirect beneficiaries of the program promoted a sense of belonging among the newly resettled families. increases in building materials stalled public Sachchithananthan Subodhini, 36 years old, from Thervipuram in the works activities in 2011. However, communi- Puthukkudiyiruppu Division of the Northern Province said that she was “very ties worked together to find creative solutions to happy. As a result of cash for work, the whole village is working as one; for our these obstacles, including using local materials own community and village.� Reflecting on her life journey since being dis- and finding alternative modes of transport.72 placed in 1995, she said that the program “had helped to bring the community Participatory aspects of programs can pro- together. . . . [T]he village seemed abandoned but the shramadana [volunteer vide a channel for voice of excluded groups. In work] helped to get the community back to its original state.� a survey of participants in Ethiopia’s Produc- tive Safety Net Program—which at 7.6 million Sources: Blattman, Fiala, and Martinez 2011 (Northern Uganda); Andrews and Kryeziu 2012 for the World beneficiaries is one of the largest public works Development Report 2013 (Sri Lanka). programs in the world—two-thirds of respon- dents said that the project had given them the first opportunity ever to be involved in a local entrepreneurial skills. Students are mentored by meeting. Many participants had not interacted professors and private sector coaches to develop with local government officials prior to the business plans. The initial results of the program program.73 show that the program motivated students and Employment programs partnering with the gave them confidence to take risks. A male par- private sector can connect people through jobs. ticipant from Tunis explained, “I have become A program in Tunisia uses the process of writing more independent. My behavior has changed. I an undergraduate thesis to teach students basic use my new skills, I am more disciplined.� Stu- 144  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 dents also explained that the program expanded ability in the labor market can improve equity. their professional networks by giving them op- They can also increase the extent to which portunities to interact with mentors. “I now people perceive that they have a stake in society. have a social network. I know whom to consult,� This perception can be especially critical when explained a female participant.74 risks of social unrest from youth unemployment While not all jobs affect social cohesion, and conflict are high. While policies with weak those that shape social identity, build networks, governance or divisive targeting can undermine and increase fairness, particularly for excluded social cohesion, well-designed programs may groups, can defuse tensions. Increasing fairness have positive effects. Jobs policies for youth at in the allocation of jobs and at work can also be risk can incorporate counseling and training in important for social cohesion. Measures that conflict resolution. Public works programs can support inclusion, extend access to voice and facilitate community participation and engage- rights, and improve transparency and account- ment between citizens and local governments. Jobs and social cohesion    145 © Ayemoba Godswill / World Bank Shopkeeper and a friend at a foodstuff shop in Mpape, Nigeria © Curt Carnemark / World Bank Rural migrants working in construction in China 146  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Notes provement; neighborhood watch (men); or wom- en’s association. These trends were likely driven   1. The International Labour Organization (2011b) by a set of factors, including political changes after estimates the impact of the unemployment rate 1998 that created more space for community dia- on social unrest to be positive and significant. logue and the decentralization of public services   2. Evans-Pritchard 2010. starting in 1999; the introduction of multistake-  3. Solletty 2011. “[L]es manifestants ne demandent holder forums for consultation on budgeting pas à l’Etat de leur trouver du travail mais dé- (the Musrenbang) in 2004, which included village noncent le fait que, sur le marché de l’emploi, les meetings; school based–management initiatives choses ne se font jamais dans la transparence et that require parental involvement; and the scaling avec justice.� up of development programs that explicitly pro-   4. Hudson and others 2011. mote community participation.   5. This definition is based on Woolcock 2011. 14. Giles, Mavridis, and Witoelar 2012 for the  6. Based on indicators from the Worldwide Gov- World Development Report 2013. Development ernance Indicators, http://info.worldbank.org/ and antipoverty programs such as the KDP governance/, and documented in Kaufmann, (Kecamatan Development Project) and its urban Kraay, and Mastruzzi 2010. counterpart, the UPP (Urban Poverty Program),   7. Fukuyama 1995; North 1990; Putnam 2000. explicitly promoted community participation   8. Delhey, Newton, and Welzel 2011. from their inception and continue to do so under   9. These aspects of jobs are also correlated with val- the new and expanded umbrella of the PNPM ues that include tolerance for others, preferences (National Program for Community Empower- for gender equality, individual choice, and voice ment). Other poverty programs (including the (though not controlling for other factors). See conditional cash transfer program, Program Welzel 2012 for the World Development Report Keluarga Harapan) usually have some commu- 2013. nity participation component. 10. Based on data from the FAFO (Forskningsstif- 15. These results are similar to findings from the Wis- telsen Fafo [Fafo Research Foundation]) Good consin Longitudinal Study, which tracked nearly Jobs Survey for the World Development Report 5,000 people who had graduated high school in 2013. Probit regressions controlling for income, 1957 over 45 years and indicated that the loss of age, education, and household characteristics of a job can lead to withdrawal from associational an index of job attributes on the question, “Not life. Getting laid off even once, especially if it oc- counting anything you do for your household, curred during a person’s prime working years, re- in your work, or within voluntary organizations, sulted in “enduring, substantively and statistically how often in the past 12 months did you actively significant lower probabilities of social involve- provide help for other people?� ment over the life course,� particularly in church, 11. Easterly, Ritzen, and Woolcock 2006; Knack and youth, and community groups. See Brand and Keefer 1997. Fukuyama (1995) notes that trust in society shapes the nature of economic trans- Burgard 2008. actions and institutions. Similarly, Arrow (1972, 16. Arias and Sosa 2012 for the World Development 357) writes that “virtually every commercial Report 2013. transaction has within itself an element of trust, 17. Petesch 2012 for the World Development Report certainly any transaction conducted over a pe- 2013. riod of time. It can plausibly be argued that much 18. Altindag and Mocan 2010. of the economic backwardness in the world can 19. Helliwell and Putnam 2004. be explained by the lack of mutual confidence.� 20. Bell and Blanchflower 2010; Cramer 2010 for the North (1990) discusses the role of trust and World Development Report 2011. informal institutions as the basis for market 21. Moser 2009, 240. economies. 22. Cramer 2010 for the World Development Report 12. Noteboom 1999; Ostrom 1990; World Bank 2004b. 2011; Padilla 1992. See Boschma (2005, 51–52) for an overview. 23. Akerlof and Kranton 2010. 13. Giles, Mavridis, and Witoelar 2012 for the World 24. Fitzgerald 2006; Holzer and Lerman 2009; Oster- Development Report 2013. Indonesia Faily Life man 2005. Survey (database), Rand Corporation, Santa Mon- 25. Wilson, Eaton, and Kamanu, 2002. ica, California, http://www.rand.org/labor/FLS/ 26. Dudwick 2012 for the World Development Re- IFLS.html. The community activities recorded port 2013; Hossain 2011 for the World Develop- covered participation in a community meeting; ment Report 2012. cooperative, voluntary labor; neighborhood im- 27. Amin and others 1998, 191. Jobs and social cohesion    147 28. Feldman 2009, cited in Dudwick 2012 for the Datta 1999; Deshpande 2008; Funston 2001; World Development Report 2013; Hossain 2011 Gudavarthy 2012; Montlake 2010; Teoh 2008; for the World Development Report 2012. Weisskopf 2004; Yadav 2010. 29. Amin and others 1998. 66. Das 2012 for the World Development Report 30. Marsden 2000. 2013. 31. Chen and others 2012 for the World Develop- 67. Blattman, Fiala, and Martinez 2011. ment Report 2013. 68. Blattman and Annan 2011. 32. Kilroy 2012 for the World Development Report 69. Ibarraran and others 2012. 2013; Pickering 2006. 70. Public works programs can be scaled up quickly 33. Mutz and Mondak 2006, 153. and reach large numbers of people. In Argen- 34. Kilroy 2011; Kilroy 2012 for the World Develop- tina, a program, Jefes y Jefas de Hogar, intro- ment Report 2013. duced after the 2002 crisis was scaled up quickly 35. Montesquieu 1951. and reached more than 2 million people every 36. Varshney 2002. month. Programs have been launched and scaled 37. Chandra 2001. up after conflicts in countries including Guinea 38. Udry and Conley 2004. and Guinea Bissau, Liberia, Nepal, Sierra Leone, 39. Anarfi, Anyidoho, and Verschoor 2008. Sudan, and the Republic of Yemen. 40. Vasilaky 2010. 71. Programs in Kenya, Sierra Leone, and South 41. Allport 1954. Pettigrew and Tropp (2011) pro- Africa also provide vocational training and sup- ­ vide a meta-analysis of intergroup studies. port for entrepreneurship for youth. 42. Austin and Worchel 1979. 72. Andrews and Kryeziu 2012 for the World Devel- 43. Gatti and others 2012. opment Report 2013. 44. Arias, Yamada, and Tejerina 2004. See also Wang 73. Andrews and Kryeziu 2012 for the World Devel- 2011. opment Report 2013. 45. Willman and Makisaka 2010 for the World De- 74. Brodmann, Grun, and Premand 2011, 3. velopment Report 2011. 46. Singerman 2007, 33. 47. Marc and others 2012, 50. References 48. World Bank 2005. 49. Abras and others 2012 for the World Develop- The word processed describes informally reproduced ment Report 2013. works that may not be commonly available through 50. Abras and others 2012 for the World Develop- libraries. ment Report 2013. 51. Amos 2012. Abras, Ana, Alejandro Hoyos, Ambar Narayan, and 52. IOM 2012. Sailesh Tiwari. 2012. “Inequality of Opportunities 53. Assessments of experience with public works in the Labor Market: Evidence from Life in Tran- programs can be found in Betcherman, Olivas, sition Surveys in Europe and Central Asia.� Back- and Dar 2004; Harvey 2011; Martin and Grubb ground paper for the WDR 2013. 2001; and McCord and Slater 2011. 54. Dudwick and Srinivasan, forthcoming; USAID Adler, Daniel, and Hans Hwang. 2012. “From Law on 2009. the Books to Law in Action: A Note on the Role 55. Chen and others 2007. of Regulation in the Production of Good Jobs in 56. Fredman 2012 for the World Development Re- Cambodia’s Garment Sector.� Background paper port 2013. for the WDR 2013. 57. Fredman 2012 for the World Development Re- Akerlof, George A., and Rachel E. Kranton. 2010. port 2013; McKenzie, Theoharides, and Yang Identity Economics: How Our Identities Shape 2012. Our Work, Wages, and Well-Being. Princeton, NJ: 58. Locke, Amengual, and Mangla 2009. Princeton University Press. 59. Purcell 2010. Allport, Gordon. 1954. The Nature of Prejudice. Read- 60. Fredman 2012 for the World Development Re- ing, MA: Addison-Wesley. port 2013. Altindag, Duha Tore, and Naci H. Mocan. 2010. “Job- 61. Holzer and Neumark 2000. lessness and Perceptions about the Effectiveness 62. World Bank 2011. of Democracy.� Working Paper Series 15994, Na- 63. Holzer and Neumark 2000. tional Bureau of Economic Research, Cambridge, 64. Burger and Jafta 2010. MA. 65. Bardhan, Mookherjee, and Parra 2012; Das Amin, Sajeda, Ian Diamond, Rushira T. Naved, and 2012 for the World Development Report 2013; Margaret Newby. 1998. “Transition to Adulthood 148  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 of Female Garment-Factory Workers in Bangla- Program in Northern Uganda.� Yale University, desh.� Studies in Family Planning 29 (2): 185–200. Innovations for Poverty Action, New Haven, CT. Amos, Deborah. 2012. “In Morocco, Unemployment Boschma, Ron. 2005. “Social Capital and Regional Can Be a Full-Time Job.� National Public Radio, Development: An Empirical Analysis of the Third January 27. Italy.� In Learning from Clusters, ed. Ron Boschma Anarfi, John Kwasi, Nana Akua Anyidoho, and Arjan and Robert Kloosterman, 139–68. Dordrecht: Verschoor. 2008. “The Economic Empowerment Springer Verlag. of Young People in Ghana.� World Bank, Wash- Brand, Jennie, and Sarah Burgard. 2008. “Job Dis- ington, DC. Processed. placement and Social Participation: Findings Andrews, Colin, and Adea Kryeziu. 2012. “Public across the Life Course of a Cohort of Joiners.� Works and the Jobs Agenda: Pathways for Social Social Forces 87 (1): 211–42. Cohesion.� Background paper for the WDR 2013. Briggs, Xavier de Souza. 2004. “Civilization in Color: Arias, Omar, and Walter Sosa. 2012. “Do Jobs Cause The Multicultural City in Three Millennia.� City & Trust? Results from Pseudo-Panel Analysis of Community 3 (4): 311–42. Euro and Latino Barometer Surveys.� Background Brodmann, Stefanie, Rebekka Grun, and Patrick paper for the WDR 2013. Premand. 2011. “Can Unemployed Youth Create Arias, Omar, Gustavo Yamada, and Luis Tejerina. Their Own Jobs? The Tunisia Business Plan The- 2004. “Education, Family Background and Racial sis Competition.� Fast Brief 83 (March), MNA Earnings Inequality in Brazil.� International Jour- Knowledge and Learning, World Bank, Washing- nal of Manpower 25 (314): 355–74. ton, DC. Arrow, Kenneth. 1972. “Gifts and Exchanges.� Phi- Burger, Rulof, and Rachel Jafta. 2010. “Affirmative Ac- losophy and Public Affairs 1 (4): 343–62. tion in South Africa: An Empirical Assessment of Austin, William G., and Stephen Worchel. 1979. The the Impact on Labour Market Outcomes.� Work- Social Psychology of Intergroup Relations. Mon- ing Paper Series 76, Centre for Research on Equal- terey, CA: Brooks/Cole Publishing Company. ity, Human Security and Ethnicity, University of Bardhan, Pranab K., Dilip Mookherjee, and Monica Oxford, Oxford. L. Parra. 2012. “The Impact of Political Reserva- Chandra, Kanchan. 2001. “Civic Life or Economic tions in West Bengal Local Governments on Anti- Interdependence?� Commonwealth and Compara- Poverty Targeting.� Journal of Globalization and tive Politics 39 (1): 110–18. Development 1 (1): 1948–837. Chen, Martha, Chris Bonner, Mahendra Chetty, Lu- BBC News. 2006. “Coffee Key to Reconciling Rwan- cia Fernandez, Karin Pape, Federico Parra, Arbind dans.� BBC World News, August 30. Singh, and Caroline Skinner. 2012. “Urban Infor- Bell, David N. F., and David Blanchflower. 2010. mal Workers: Representative Voice and Economic “Youth Unemployment: Déjà Vu?� Discussion Pa- Rights.� Background paper for the WDR 2013. per Series 4705, Institute for the Study of Labor, Chen, Martha, Renana Jhabvala, Ravi Kanbur, and Bonn. Carol Richards. 2007. Membership-Based Organi- Betcherman, Gordon, Karina Olivas, and Amit Dar. zations of the Poor. London: Routledge. 2004. “Impacts of Active Labor Market Programs: Cramer, Christopher. 2010. “Unemployment and New Evidence from Evaluations with Particular Participation in Violence.� Background paper for Attention to Developing and Transition Coun- the WDR 2011. tries.� Social Protection Discussion Paper Series Das, Maitreyi Bordia. 2012. “Stubborn Inequalities, 0402, World Bank, Washington, DC. Subtle Processes: Exclusion and Discrimination Bjørkhaug, Ingunn, Anne Hatløy, Tewodros Kebede, in the Labor Market.� Background paper for the and Huafeng Zhang. 2012. “Perception of Good WDR 2013. Jobs: Colombia.� Background paper for the WDR Datta, Nonica. 1999. “Backward Caste Movement 2013. Gains Ground.� Economic and Political Weekly 34 Blattman, Christopher, and Jeannie Annan. 2011. (39): 2630–31. “Reintegrating and Employing High Risk Youth in Delhey, Jan, Kenneth Newton, and Christian Welzel. Liberia: Lessons from a Randomized Evaluation 2011. “How General Is Trust in ‘Most People’? of a Landmine Action an Agriculturual Train- Solving the Radius of Trust Problem.� American ing Program for Ex-Combatants.� Policy report Sociological Review 76 (5): 786–807. 2011.1, Yale University, Innovations for Poverty Deshpande, Satish. 2008. “Changing Social Composi- Action, New Haven, CT. tion.� Seminar 587 (July): 23–26. Blattman, Christopher, Nathan Fiala, and Sebastian Dudwick, Nora. 2012. “The Relationship between Martinez. 2011. “Employment Generation in Jobs and Social Cohesion: Some Examples from Rural Africa, Mid-Term Results from an Experi- Ethnography.� Background paper for the WDR mental Evaluation of the Youth Opportunities 2013. Jobs and social cohesion    149 Dudwick, Nora, and Radhika Srinivasan, with Jose Holzer, Harry, and Robert Lerman. 2009. The Future Cueva and Dorsati Madani. Forthcoming. Creat- of Middle-Skill Jobs. Washington, DC: Center on ing Jobs in Africa’s Fragile States: Are Value Chains Children and Families, Brookings Institute. an Answer? Directions in Development. Washing- Holzer, Harry J., and David Neumark. 2000. “Assess- ton, DC: World Bank. ing Affirmative Action.� Journal of Economic Lit- Durkheim, Emile. 1893. The Division of Labor in Soci- erature 38 (3): 483–568. ety. New York: Free Press. Hossain, Naomi. 2011. “Exports, Equity, and Empow- Easterly, William, Jozef Ritzen, and Michael Wool- erment: The Effects of Readymade Garments cock. 2006. “Social Cohesion, Institutions, and Manufacturing Employment on Gender Equality Growth.� Economics and Politics 18 (2): 103–20. in Bangladesh.� Background paper for the WDR Eaton, Charles, and Andrew W. Shepherd. 2001. Con- 2012. tract Farming: Partnerships for Growth. Rome: Hudson, Maria, Rosemary Davidson, Lucia Durante, Food and Agriculture Organization. Jemma Grieve, and Arjumand Kazmi. 2011. Reces- Evans-Pritchard, Ambrose. 2010. “IMF Fears ‘Social sion and Cohesion in Bradford. York, U.K.: Joseph Explosion’ from World Jobs Crisis.� Telegraph, Rowntree Foundation. September 13. Ibarraran, Pablo, Laura Ripani, Bibiana Taboada, FAFO (Forskningsstiftelsen Fafo [Fafo Research Juan Miguel Villa, and Brigida Garcia. 2012. “Life Foundation]). 2012. “Good Jobs Survey.� FAFO Skills, Employability and Training for Disadvan- Institute for Applied International Studies, Oslo. taged Youth: Evidence from a Randomized Evalu- Fitzgerald, Joan. 2006. Moving Up in the New Econ- ation Design.� IZA Conference Paper, Institute for omy: Career Ladders for U.S. Workers. Ithaca, NY: the Study of Labor, Bonn, May 12. Cornell University Press. ILO (International Labour Organization). 2011a. Feldman, Shelley. 2009. “Historicising Garment Man- Global and Regional Estimates on Domestic Work- ufacturing in Bangladesh: Gender, Generation, ers. Geneva: ILO. and New Regulatory Regimes.� Journal of Interna- ———. 2011b. World of Work Report: Making Mar- tional Women’s Studies 11(1): 268–82. kets Work for Jobs. Geneva: ILO. Fredman, Sandra. 2012. “Anti-Discrimination Laws IOM (International Organization for Migration). and Work in the Developing World: A Thematic 2012. “IOM Tracking of Spontaneous Returns Overview.� Background paper for the WDR 2013. Project: Total Returns to South Sudan Post CPA to Fukuyama, Francis. 1995. Trust: The Social Virtues and June 2009.� Geneva, IOM. the Creation of Prosperity. New York: Free Press. Jenson, Jane. 1998. “Mapping Social Cohesion: The Funston, John. 2001. “Malaysia: Developmental State State of the Canadian Research.� Discussion Paper Challenged.� In Government and Politics in South- 3, Canadian Policy Research Networks, Ottawa. east Asia, ed. John Funston, 160–202. Singapore: Kaufmann, Daniel, Aart Kraay, and Massimo Mas- Institute of Southeast Asian Studies. Furnivall, John S. 1948. Colonial Policy and Practice: truzzi. 2010. “The Worldwide Governance Indica- A Comparative Study of Burma and Netherlands tors: Methodology and Analytical Issues.� Policy India. Cambridge, U.K.: Cambridge University Research Working Paper Series 5430, World Bank, Press. Washington, DC. Gatti, Roberta, Diego Angel-Urdinola, Joana Silva, Kennedy, Alex. 2012. “Singapore to Require One Day and Andras Bodor. 2012. Striving for Better Jobs: Off a Week for Maids.� Associated Press, March 5. The Challenge of Informality in the Middle East and Kilroy, Austin. 2011. “Business Bridging Ethnicity.� North Africa. Washington, DC: World Bank. PhD thesis, Massachusetts Institute of Technol- Giles, John, Dimitris Mavridis, and Firman Witoelar. ogy, Cambridge, MA. 2012. “Subjective Well-Being, Social Cohesion, ———. 2012. “Jobs to Social Cohesion: Via Interests, and Labor Market Outcomes in Indonesia.� Back- Attitudes, and Identities.� Background paper for ground paper for the WDR 2013. the WDR 2013. Gudavarthy, Ajay. 2012. “Can We De-Stigmatize Res- Knack, Stephen, and Philip Keefer. 1997. “Does ervations in India?� Economic and Political Weekly Social Capital Have an Economic Payoff? A Cross- 47 (6): 55–62. Country Investigation.� Quarterly Journal of Eco- Harvey, Philip. 2011. “A Job-Led Recovery Strategy nomics 112 (4): 1251–88. Achieving Economic Recovery through Direct Locke, Richard, Matthew Amengual, and Ashkay Public Job Creation.� Dēmos, NY. Mangla. 2009. “Virtue out of Necessity? Compli- Helliwell, John, and Robert Putnam. 2004. “The So- ance, Commitment and the Improvement of La- cial Context of Well-Being.� Philosophical Transac- bor Conditions in Global Supply Chains.� Politics tions of the Royal Society B 359: 1435–46. & Society 37 (3): 319–51. 150  WO R L D D E V E LO P M E N T R E P O RT 2 0 1 3 Marc, Alexandre, Alys Willman, Ghazia Aslam, Mi- Lucifora, and Wiemer Salverda. 237–252. London: chelle Rebosio, with Kanishka Balisuriya. 2012. Palgrave Macmillan. Societal Dynamics and Fragility: Engaging Societies Ostrom, Elinor. 1990. Governing the Commons: The in Responding to Fragile Situations. Washington, Evolution of Institutions for Collective Action. Cam- DC: World Bank. bridge, U.K.: Cambridge University Press. Marsden, David. 2000. “A Theory of Job Regulation, Padilla, Felix M. 1992. The Gang as an American En- the Employment Relationship, and the Organisa- terprise. Piscataway, NJ: Rutgers University Press. tion of Labour Institutions.� Industrielle Beziehu- Paes de Barros, Ricardo, Francisco H. G. Ferreira, Jose ngen 7 (4): 320–47. R. Molinas Vega, Jaime Saavedra Chanduvi. 2009. Martin, John P., and David Grubb. 2001. “What Measuring Inequality of Opportunities in Latin Works and for Whom? A Review of OECD Coun- America and the Caribbean. Washington, DC: tries’ Experiences with Active Labour Market Poli- World Bank. cies.� Swedish Economic Policy Review 8 (2): 9–60. Perry, Guillermo E., William F. Maloney, Omar S. McCord, Anna, and Rachel Slater. 2011. Overview Arias, Pablo Fajnzylber, Andrew D. Mason, and of Public Works in Sub-Saharan Africa. London: Jaime Saavedra-Chanduvi. 2007. Informality: Exit Overseas Development Institute. and Exclusion. Washington, DC: World Bank. McKenzie, David, Caroline Theoharides, and Dean Petesch, Patti. 2012. “The Exponential Clash of Con- Yang. 2012. “Distortions in the International Mi- flict, Good Jobs, and Changing Gender Norms grant Labor Market: Evidence from Filipino Mi- in Four Economies.� Background paper for the gration and Wage Responses to Destination Coun- WDR 2013. try Economic Shocks.� Policy Research Working Pettigrew, Thomas F., and Linda R. Tropp. 2011. Paper Series 6041, World Bank, Washington, DC. When Groups Meet: The Dynamics of Intergroup Montesquieu, Baron de. 1951. 2 Oeuvres Completes: Contact. New York: Psychology Press. De l’Esprit des Lois (2 Complete Works: Spirit of Pickering, Paula. 2006. “Generating Social Capital for the Laws). Paris: Gallimard/Pleiade. Bridging Ethnic Divisions in the Balkans: Case Montlake, Simon. 2010. “Malaysia Cautiously Chal- Studies of Two Bosniak Cities.� Ethnic and Racial lenges Longtime Affirmative Action Policies.� Studies 29 (1): 79–103. Christian Science Monitor, March 30. Purcell, Julius. 2010. Individual Disputes at the Work- Moreno, Jacob L., and Helen H. Jennings. 1937. “Sta- place: Alternative Disputes Resolution. Dublin: tistics of Social Configurations.� Sociometry 1: European Foundation for the Improvement of 342–74. Living and Working Conditions. Moser, Caroline O. N. 2009. Ordinary Families, Ex- Putnam, Robert. 2000. Bowling Alone: The Collapse traordinary Lives: Assets and Poverty Reduction in and Revival of American Community. New York: Guayaquil, 1978–2004. Washington, DC: Brook- Simon and Schuster. ings Institution. Putnam, Robert, Robert Leonardi, and Rafaella Na- Mutz, Diane, and Jeffery Mondak. 2006. “The Work- netti. 1993. Making Democracy Work: Civic Tra- place as a Context for Cross-Cutting Political Dis- ditions in Modern Italy. Princeton, NJ: Princeton course.� Journal of Politics 68 (1): 140–55. University Press. North, Douglass C. 1990. Institutions, Institutional Rawls, John. 1971. A Theory of Justice. Cambridge, Change and Economic Performance. New York: MA: Belknap Press. Cambridge University Press. Roemer, John E. 1998. Equality of Opportunity. Cam- Norton, Andrew, and Arjan de Haan. 2012. “Social bridge, MA: Harvard University Press. Cohesion: Theoretical Debates and Practical Ap- Roemer, John E., Rolf Aaberge, Ugo Colombino, plications with Respect to Jobs.� Background pa- Johan Fritzell, Stephen Jenkins, Arnaud Lefranc, per for the WDR 2013. Ive Marx, Marianne Page, Evert Pommer, and Noteboom, Bart. 1999. Inter-Firm Alliances: Analysis Javier Ruiz-Castillo. 2003. “To What Extent Do and Design. London: Routledge. Fiscal Regimes Equalize Opportunities for In- Nozick, Robert. 1974. Anarchy, State, and Utopia. New come Acquisition among Citizens.� Journal of York: Basic Books. Public Economics 87: 539–65. OECD (Organisation for Economic Co-operation Singerman, Diane. 2007. “The Economic Imperatives and Development). 2011. Perspectives on Global of Marriage: Emerging Practices and Identities Development 2012: Social Cohesion in a Shifting among Youth in the Middle East.� Middle East World. Paris: OECD. Youth Initiative Working Paper 6, Wolfensohn Osterman, Paul. 2005. “Making Bad Jobs Good: Strat- Center for Development, Dubai School of Gov- egies for the Service Section.� In Job Quality and ernment at Brookings Institution, Washington, Employer Behaviour, ed. Stephen Bazen, Claudio DC. Jobs and social cohesion    151 Solletty, Marion. 2011. “Le Chômage des Diplômés, Weisskopf, Thomas. 2004. “Impact of Reservation on Moteur de la Révolte Tunisienne.� Le Monde, Jan- Admissions to Higher Education in India.� Eco- uary 10. nomic and Political Weekly 39 (39): 4339–49. Spilsbury, John, and Karri Goeldner Byrne. 2007. Welzel, Christian. 2012. “The Contribution of ‘Good’ “Value Chain Activities for Conflict-Affected Jobs to Development and Cohesion: The Human Populations in Guinea.� Micro Report, U.S. Empowerment Perspective.� Background paper Agency for International Development, Washing- for the WDR 2013. ton, DC. Wietzke, Frank-Borge, and Catriona McLeod. 2012. Teoh, Shannon. 2008. “Poll Shows Most Malaysians “Jobs, Well-Being, and Social Cohesion: Evidence Want NEP to End.� Malaysian Insider, October 9. from Value and Perception Surveys.� Background The Economist. 2000. “An Uncommon Market in the paper for the WDR 2013. Caucasus.� The Economist, June 1. Willman, Alys, and Megumi Makisaka. 2010. “Inter- Udry, Christopher, and Timothy G. Conley. 2004. personal Violence Prevention: A Review of the “Social Networks in Ghana.� Discussion Paper Evidence and Emerging Lessons.� Background ­ 888, Economic Growth Center, Yale University, paper for the WDR 2011. New Haven, CT. Wilson, Randall, Susan Eaton, and Amara Kamanu. USAID (U.S. Agency for International Development). 2002. “Extended Care Career Ladder Initiative 2009. A Guide to Economic Growth in Post-Conflict (ECCLI) Round 2: Evaluation Report.� Faculty Countries. Washington, DC: USAID, Bureau for Research Working Paper Series RWP03-006, Har- Economic Growth, Agriculture and Trade. vard University, Cambridge, MA. Varshney, Ashutosh. 2002. Ethnic Conflict and Civic Woolcock, Michael. 2011. “What Distinctive Con- Life: Hindus and Muslims in India. New Haven, tribution Can Social Cohesion Make to Devel- CT: Yale University Press. opment Theory, Research, and Policy?� Paper Vasilaky, Kathryn. 2010. “As Good as the Networks presented at the Organisation for Economic Co- They Keep? Expanding Farmer’s Social Net- operation and Development International Con- works Using Randomized Encouragement in ference on Social Cohesion and Development, Rural Uganda.� Yale University, New Haven, CT. Paris, January 20. Processed. World Bank. 2004a. Guyana: A Poverty and Social Vicente-Mazariegos-Eiríz, José Ignacio. 1985. “La ­Impact Analysis of Bauxite Mining Reforms in Guy- ciudad hispanomusulmana: Organizacíon social aya. Washington, DC: World Bank. y formalizacíon urbana.� In Urbanismo e Historia ———. 2004b. World Development Report 2005: A en el Mundo Hispano, ed. Antonio Bonet Correa, Better Investment Climate for Everyone. New York: 2: 749–64. Madrid: Universidad Complutense de Oxford University Press. Madrid. ———. 2005. World Development Report 2006: Eq- Voice of America. 2002. “South Ossetia Market Im- uity and Development. New York: Oxford Univer- portant for Local Economy.� Voice of America, sity Press. Washington, DC. ———. 2011. World Development Report 2012: Gen- Wang, Shing-Yi. 2011. “Marriage Networks, Nepo- der Equality and Development. Washington, DC: tism and Labor Market Outcomes in China.� New World Bank. York University, New York. Processed. Yadav, Yogendra. 2010. “Why Caste Should Be Weiss, Dieter. 1995. “Ibn Khaldun on Economic Counted In.� The Hindu, May 14. Transformation.� International Journal of Middle East Studies 27 (1): 29–37. PART 2 What are good jobs for development? Introduction to Part 2 E arnings, benefits (if any), and overall job nation, and the suppression of voice among satisfaction are what matter to individual workers. These forms of work should not be jobholders. Those earnings and bene- considered jobs. fits—tangible or otherwise—are the first and How jobs contribute to living standards, most direct measure of the value a job has to so- productivity, and social ­ cohesion varies with ciety. But jobs may also affect others, positively a country’s level of development, its demogra- or negatively. Jobs for women may influence re- phy, its endowments, and its institutions. Jobs source allocations at the household level and agendas are thus inherently country specific. benefit their children. Jobs connected to world By combining the various features of an econ- markets may lead to knowledge spillovers and omy, however, it is possible to build a typology make other workers more productive. Jobs that of jobs challenges. It includes agrarian econo- provide opportunities may convey a sense of mies, conflict-affected countries, urbanizing fairness to others and help them remain en- countries, resource-rich countries, small island gaged. Spillovers like these have been the focus nations, countries with high youth unemploy- of recent thinking on development. The analysis ment, formalizing economies, and aging societ- may be organized around concepts such as gen- ies. Because the nature of the challenges varies, der, urbanization, or conflict; but much of the what makes a job good for development in one action happens through jobs. context may not be so relevant in another. And Quite often individual and social perspectives in some circumstances, tradeoffs emerge. The on jobs coincide; but not always. For instance, jobs that do the most to defuse tension in the jobs with perks and benefits may be highly short term may not do much for productivity, coveted by individuals, but they may be less or those with the highest productivity impact valuable to society if their privileges are sup- may not lead to a broad-based improvement in ported through government transfers or restric- living standards. Identifying good jobs for de- tive regulations, undermining the earnings or velopment helps visualize these tradeoffs. job opportunities of others. Because of gaps Different countries face different jobs chal- like these, jobs that look equivalent from an indi- lenges, but two forces—the migration of people vidual perspective may be different from a social and the migration of jobs—connect their jobs perspective. The contributions jobs make to so- agendas. These two flows have an impact on ciety should be assessed by taking into account countries at the sending and the receiving ends. the value they have to the people who hold them, The arrival of migrants or the outsourcing of but also the potential spillovers on others— jobs affects the living standards of both migrants positive or negative. Good jobs for development and locals. The availability of foreign workers, are those with the highest value to society. the development of migrant networks channel- Conversely, some forms of work are un- ing savings and ideas, and the arrival of multi- equivocally bad. A set of universal rights en- national firms bringing in more advanced dorsed by governments, international organi- techniques, are all bound to increase productiv- zations, and others seeks to eliminate forced ity. Family structures and community life are also labor, harmful forms of child labor, discrimi- affected by the movement of people and jobs. 153 CHAPTER 5 Valuing jobs Jobs are often assessed from an individual perspective. But they can also affect the earnings, productivity, and well-being of others—positively or negatively. These spillovers should be assessed too. T he most important impact of jobs is on services, and are obstacles to economic growth the people who hold them. Jobs provide and poverty reduction. Such jobs may look ap- earnings, can give access to benefits and pealing from an individual perspective—but insurance, and are often a source of broader life less so from a social perspective. Good jobs for satisfaction. Development, in large part, consists development are those that make the greatest of increasing these direct effects of jobs on contribution to society, taking into account the individuals. value they have to the people who hold them, Beyond the importance of jobs for those but also their potential spillovers on others— who have them, jobs matter for societies be- positive or negative. cause they can affect the earnings, employment Recognizing the multiple effects of jobs opportunities, and the productivity of others, is important for understanding the possible as well as the collective capacity to manage ten- trade­offs they entail. Some jobs greatly con- sions. Jobs can also contribute to shared social tribute to productivity growth but do not lead objectives, such as poverty reduction, environ- to poverty reduction in the short run. In other mental protection, and fairness. Often, the cases, jobs are urgently needed to avoid an un- ­ individual and social values of jobs are similar; raveling of social cohesion, but the job creation but these two perspectives may differ. For in- that can be immediately supported using pub- stance, Vietnam’s poverty rate declined with lic funds is unlikely to result in rapid produc- unprecedented speed in the 1990s when land tivity growth. An understanding of the various was redistributed to farmers and agricultural effects of jobs on aggregate well-being, both commercialization was liberalized.1 Albeit in- direct and indirect, may help identify when a creasing individual incomes, farming jobs in- virtuous circle of jobs along all three transfor- volved difficult working conditions, substantial mations may arise, and when a vicious circle variability in earnings, and no formal social looms. protection. But they made a major contribu- The value of a job for the person who holds it tion to the development of Vietnam. Bloated is a primary indication of its development pay- public utilities, on the other hand, often offer a off. But assessing the broader value the job has range of privileges to their employees even to society also requires information on the spill- when the utilities themselves provide only lim- overs the job may have on the living standards ited coverage of the population and unreliable of others, on aggregate productivity, or on social Valuing jobs    155 cohesion. Data from household, plant-level, and Rights as the foundation values surveys, as well as qualitative assessments, can be used to determine the existence of rel- While jobs can be transformational, some evant spillovers. Although quantifying all the forms of work are harmful. Those that exploit gaps between the individual and social values workers, expose them to dangerous environ- of jobs may not be possible, identifying where ments, or threaten their physical and mental these gaps lie can help make policy trade­ offs well-being are bad for individuals and societ- transparent. The analytical tools to do so can be ies. Their negative effects can be surprisingly borrowed from several disciplines in economics long-lasting. An extreme example is the impact and the social sciences. These disciplines often of the Atlantic slave trade on West Africa. A focus on spillovers from jobs, without necessar- study found that individuals whose ancestors ily calling them that. ha