Report No. 35353-TH Thailand Northeast Economic Development Report November 2005 Poverty Reduction and Economic Management Sector Unit East Asia and Pacific Region Document of the World Bank i Table of Contents Foreword........................................................................................................................... ix Preface........................................................................................................................... ix Acknowledgements ....................................................................................................... x Executive Summary.......................................................................................................xiii Record.........................................................................................................................xiii The Need for Change.................................................................................................. xv The Constraints to More Rapid Development ........................................................ xvi The Agenda: General Principles and Priority Measures...................................... xxv Introduction....................................................................................................................... 1 Convergence .................................................................................................................. 1 Agglomeration............................................................................................................... 2 Scope............................................................................................................................... 3 Approach and Content ................................................................................................. 4 I. Record ............................................................................................................................ 6 Growth ........................................................................................................................... 6 Growth Gap................................................................................................................. 6 GDP Shares................................................................................................................. 8 International Comparison.......................................................................................... 10 Growth Divergence................................................................................................... 11 Provincial Growth Divergence ................................................................................. 14 Structural Change - the Long Haul........................................................................... 15 Structural Change - the Short Haul........................................................................... 17 Poverty ......................................................................................................................... 19 Incidence and Numbers............................................................................................. 19 Self-assessment......................................................................................................... 21 Spatial View - Provinces........................................................................................... 22 Spatial View - Tambons............................................................................................ 25 Eradicating Poverty................................................................................................... 27 Poverty Registration.................................................................................................. 28 Shared Growth ­ Regions......................................................................................... 29 Shared Growth ­ Provinces ...................................................................................... 31 Durable Ownership................................................................................................... 32 II. Constraints ................................................................................................................. 34 Cities............................................................................................................................. 34 Urbanization and Development ................................................................................ 34 ii Primate City .............................................................................................................. 36 Drivers, Spillovers and Congestion .......................................................................... 38 Extended Bangkok Area and Beyond....................................................................... 39 Urbanization and Industrialization............................................................................ 41 Enterprises................................................................................................................... 42 Manufacturing Value Added..................................................................................... 42 Employment Dynamics............................................................................................. 44 Sector Composition................................................................................................... 46 Concentration............................................................................................................ 47 Industry and Regional Groups .................................................................................. 50 Products, Size and Exports ....................................................................................... 51 Firm Productivity...................................................................................................... 53 Decomposing Firm Productivity............................................................................... 54 Technological Capability.......................................................................................... 56 Northeast Exporters .................................................................................................. 57 Board of Investment Promotions .............................................................................. 58 BOI Zones................................................................................................................. 59 Industrial Estates....................................................................................................... 62 Construction Permits................................................................................................. 63 Business Regulations ................................................................................................ 64 Banks......................................................................................................................... 65 Workers ....................................................................................................................... 67 Jobs ........................................................................................................................... 67 Demographics ........................................................................................................... 68 Working-Age Population.......................................................................................... 69 Job Entry................................................................................................................... 71 Wage Employment and Skills................................................................................... 72 Occupation................................................................................................................ 74 Wages........................................................................................................................ 75 Returns...................................................................................................................... 76 Supply and Demand.................................................................................................. 78 Unions and Minimum Wage..................................................................................... 80 Labor Protection Legislation..................................................................................... 82 Migration................................................................................................................... 83 Remittances............................................................................................................... 86 Students........................................................................................................................ 88 Access ....................................................................................................................... 88 Private and Public Spending..................................................................................... 90 Efficiency, Targeting, and Quality............................................................................ 92 Infrastructure.............................................................................................................. 94 Roads and Phones ..................................................................................................... 94 Household Access..................................................................................................... 97 ICT............................................................................................................................ 99 Public Spending ...................................................................................................... 100 Mega-Infrastructure Projects .................................................................................. 101 iii Mekong Region.......................................................................................................... 104 Global Trade Integration......................................................................................... 104 Growth .................................................................................................................... 106 Trade ....................................................................................................................... 108 Agenda.................................................................................................................... 110 East-West Corridor ................................................................................................. 111 Villages....................................................................................................................... 113 Agricultural economy ............................................................................................. 113 Labor Productivity .................................................................................................. 115 Land Quality ........................................................................................................... 117 Rural poverty .......................................................................................................... 118 Farming, Enterprises, and Poverty.......................................................................... 119 Subsistence.............................................................................................................. 120 Public Spending ...................................................................................................... 121 Rural Programs ....................................................................................................... 122 Irrigation ................................................................................................................. 124 Research and Extension.......................................................................................... 126 Weather Risk........................................................................................................... 127 New Weather Insurance.......................................................................................... 129 Value Chain ............................................................................................................ 130 Rice............................................................................................................................. 132 Global Commodity ................................................................................................. 132 Yields and Varieties................................................................................................ 133 Producers, Collectors and Millers........................................................................... 135 Wholesalers, Retailers and Exporters ..................................................................... 137 Marketing Chain ..................................................................................................... 138 Costs and Margins................................................................................................... 140 Cost Reduction........................................................................................................ 142 Competitiveness and Comparative Advantage....................................................... 143 Rice Pledging Scheme ............................................................................................ 145 Rice Issues .............................................................................................................. 146 Rice Investment Options......................................................................................... 147 Silk.............................................................................................................................. 149 Global Commodity ................................................................................................. 149 Trade Liberalization................................................................................................ 150 Silk in the Northeast................................................................................................ 151 Silk Yarn................................................................................................................. 152 Sericulturist Farmers............................................................................................... 154 Silk Yarn Reelers And Traders............................................................................... 155 Weavers and Garment Producers............................................................................ 156 Marketing Chain ..................................................................................................... 157 Competitiveness and Comparative Advantage....................................................... 159 Government Policies............................................................................................... 160 Value Added Potential............................................................................................ 161 Silk Issues ............................................................................................................... 162 iv Public Expenditures.................................................................................................. 163 Unitary Government ............................................................................................... 163 Spending Gap.......................................................................................................... 165 Sectoral Spending ................................................................................................... 167 Provinces................................................................................................................. 168 Wages...................................................................................................................... 170 Capacity .................................................................................................................. 171 Past and Present Reforms........................................................................................ 172 Agenda and Function.............................................................................................. 173 Area......................................................................................................................... 174 Reform Components............................................................................................... 175 III. Strategies................................................................................................................. 176 Taking Stock.............................................................................................................. 176 Hypotheses.............................................................................................................. 176 Need for Change ..................................................................................................... 177 Agenda ....................................................................................................................... 178 Growth .................................................................................................................... 178 Poverty Reduction................................................................................................... 180 Public Expenditures ................................................................................................ 182 Value Chain ............................................................................................................ 184 Value Chain ­ Policy Simulations.......................................................................... 184 Summary of the Findings........................................................................................ 187 References...................................................................................................................... 195 v Table of Figures Figure 1: Regional Per Capita GDP, 1970 to 2004, 1988 Prices....................................xiv Figure 2: Number of Poor, 1988 to 2002 .........................................................................xiv Figure 3: Real Per Capita GDP Growth in Greater Mekong Region, 1993 to 2003.......xvi Figure 4: Primary City Indicies, 1983 to 2000................................................................xvii Figure 5: Spatial Distribution of Manufacturing Employment, 1996/7 and 2001/2..... xviii Figure 6: Regional Board of Investment Promotion Certificates, 1997 to 2005..............xix Figure 7: Returns to Education of Monthly Wage Earners Relative to Less than Primary in the Northeast, 1991 to 2004........................................................................xxi Figure 8: Population Pyramids in the Northeast and Bangkok, 2002 ............................xxii Figure 9: GMS GDP (Current Dollar Billion), 1995 to 2003....................................... xxiii Figure 10: Agricultural Value Added by Agricultural Worker, 1991 to 2004 .............. xxiii Figure 11: Government Spending, FY 1999 to FY 2003 (Baht Per Capital, 1999 Prices).....xxiv Figure 12: Report Content.................................................................................................. 4 Figure 13: Regional Growth in Thailand........................................................................... 7 Figure 14: Regional population and GDP shares, 1970 to 2004....................................... 9 Figure 15: Growth in Northeast and East Asia................................................................ 10 Figure 16: Regional Economic Growth Divergence, 1970 to 1986 and 1986 to 2004.... 12 Figure 17: Growth Convergence among Provinces, 1975 to 1986 and 1986 to 2003..... 14 Figure 18: Regional GDP Composition, 1970 to 2004.................................................... 16 Figure 19: Regional GDP Composition, 1996 to 2004.................................................... 18 Figure 20: Regional Poverty Trends, 1988 to 2002 ......................................................... 20 Figure 21: Self-Assessed Poverty of Communities in 1991 and 2001.............................. 21 Figure 22: Poverty Maps, 1988 to 1994 and 2002........................................................... 23 Figure 23: Tambon-Level Poverty Map of Thailand, 2000.............................................. 26 Figure 24: Share of Population Registered as Poor and 2002 Poverty Headcount in Northeast Provinces (%)................................................................................. 28 Figure 25: Poverty and Growth, 1988 to 2002................................................................. 30 Figure 26: Growth and Poverty in Provinces, 1988 to 2002 ........................................... 31 Figure 27: Durable Goods Ownership (Percent), 1988 to 2002...................................... 33 Figure 28: Urbanization Indicators.................................................................................. 35 Figure 29: Primacy Indices .............................................................................................. 37 Figure 30: Urbanization and Regional Cities .................................................................. 40 Figure 31: Manufacturing GDP and Exports, 1991 to 2004 ........................................... 43 Figure 32: Spatial Distribution of Manufacturing Employment, 1996/7 and 2001/2...... 45 Figure 33: Manufacturing Employment 1996/7 and 2001/2 by Technological Characteristics................................................................................................ 46 Figure 34: Spatial Distribution of Employment of PICS Industries................................. 47 Figure 35: Regional Employment Concentration, 1996/7 and 2001/2............................. 49 Figure 36: 2-Digit PICS Industries, Real Value Added (1988 Prices) and Total Value Added (%), 1996 to 2004 ................................................................................ 50 Figure 37: 4-Digit Products by Regions........................................................................... 52 Figure 38: Regional Distribution of Total Factor Productivity....................................... 53 vi Figure 39: Simulated Productivity Outside of Bangkok Using Returns of Bangkok and Vicinity............................................................................................................ 55 Figure 40: Kernel Density Plots of TCI by Region........................................................... 56 Figure 41: BOI Applications, Approvals, Certificates and Start-Ups (% of GDP), 1994 - 2004................................................................................................................. 59 Figure 42: BOI by Regions and Zones (%) ...................................................................... 60 Figure 43: BOI Investment Zones..................................................................................... 61 Figure 44: Industrial Estates............................................................................................ 62 Figure 45: Construction Area Permits ............................................................................. 63 Figure 46: 2005 Doing Business Survey ­ Khonkaen and Bangkok ................................ 64 Figure 47: Commercial Bank by Regions, 1999 and 2005............................................... 66 Figure 48: Labor Productivity, 1991 to 2004................................................................... 67 Figure 49: Demographic Indicators................................................................................. 68 Figure 50: Labor Force Composition .............................................................................. 70 Figure 51: Regional Unemployment Rates and Educational Attainment by Age (%)...... 71 Figure 52: Employment Composition............................................................................... 73 Figure 53: Occupational Structure by Region of the Employed (E), Wage Workers (W) and Monthly Wage Workers (M) (%), February 1991, 1996 and 2004 ......... 74 Figure 54: Wages in February 1991, February 1996, and February 2004 ..................... 75 Figure 55: Returns to Education of Monthly Wage Earners Relative to Less than Primary ..... 77 Figure 56: Population Aged 15 to 60 years-old with at least Lower Secondary Education (%)................................................................................................................... 78 Figure 57: Relative Labor and Relative Labor Demand of Monthly Wage Workers (Substitution Elasticity of 2), 1991 to 2004 .................................................... 79 Figure 58: Unionization and Minimum Wages ................................................................ 81 Figure 59: Covered, Public and Uncovered Sectors according to the 1998 Labor Protection Act, Northeast and Rest of Thailand, 1991 to 2004...................... 82 Figure 60: Migration........................................................................................................ 84 Figure 61: Household Remittances .................................................................................. 87 Figure 62: School Participation Rate, 1988 to 2002........................................................ 89 Figure 63: Private and Public Spending on Education, 2002.......................................... 91 Figure 64: Efficiency and Quality of Public Education Spending ................................... 93 Figure 65: Rural Infrastructure, 1977 to 2000................................................................. 95 Figure 66: Regional Indicators of Household Infrastructure (Percent), 1988 to 2002 ... 98 Figure 67: ICT in Thailand's Regions (%)....................................................................... 99 Figure 68: Public Per Capita Spending on Transportation and Infrastructure, FY 2002 and FY 2003.................................................................................................. 100 Figure 69: Investment Indicators and Planned Infrastructure Spending....................... 102 Figure 70: Thailand Trade by Country, 1980 to 2003 ................................................... 105 Figure 71: Growth in the Greater Mekong Subregion................................................... 107 Figure 72: GMS Trade.................................................................................................... 109 Figure 73: Agricultural GDP, 1970 to 2004 .................................................................. 114 Figure 74: Agricultural Labor Productivity................................................................... 116 Figure 75: Fertilizer and Pesticides Use and Poverty among Farming Households, 2002..... 117 Figure 76: Poverty in Urban and Rural Areas, 2002..................................................... 118 Figure 77: Farming, Enterprises, and Poverty, 2002 .................................................... 119 vii Figure 78: Households Type and Land Size, Percent, 2002........................................... 120 Figure 79: Public Spending on Agriculture, FY 2002.................................................... 121 Figure 80: Farmers Debt Moratorium and Village Fund, 2002 .................................... 123 Figure 81: Irrigation ...................................................................................................... 125 Figure 82: Public Expenditure on Agricultural Research and Extension, FY 2003 (%)......... 126 Figure 83: Agricultural Growth and Weather Risk........................................................ 128 Figure 84: Agricultural Value Added and Household Production of Key Commodities, Percent.......................................................................................................... 131 Figure 85: Glutinous and Non-Glutinous Rice, 2002.................................................... 134 Figure 86: Rice Subsistence Farmers and Poverty. 2002 .............................................. 136 Figure 87: Marketing Channels for Rice in the Northeast, 2004................................... 139 Figure 88: Distribution of Benefits and Margins along the Marketing Chain............... 141 Figure 89: Indicative Costs for Hom Mali and Glutinous Rice in the Northeast, Baht/KG ....... 142 Figure 90: Nominal Price Protection and Domestic Resource Cost.............................. 144 Figure 91: Paddy Pledging Scheme Procedure.............................................................. 145 Figure 92: Share of Sericultural Farmers by Poverty Status (Percent of All Households)........ 151 Figure 93: Silk Marketing Chain for the Northeast........................................................ 158 Figure 94: Competitiveness and Comparative Advantage of Thai Silk.......................... 159 Figure 95: Public Expenditures...................................................................................... 164 Figure 96: Regional Public Expenditures ...................................................................... 166 Figure 97: Northeast Public Spending as Percent of Non-Bangkok Average, Percent and Baht Per Capita, FY 2003............................................................................. 167 Figure 98: Per Capita Public Expenditure and Poverty ................................................ 168 Figure 99: Overall Public and Capital Spending versus Gross Provincial Product, 1999 to 2003 .......................................................................................................... 169 Figure 100: Northeast Spending Composition and Regional Public Sector Workers and Wages............................................................................................................ 170 Figure 101: A. Monthly Income of Civil Servants' Family by Position (2001). B. Civil Service Composition by Education Level (2004).......................................... 171 Figure 102: Proposed Budget Breakdown ..................................................................... 172 Table of Tables Table 1: Members of the NEED Steering Committee........................................................ xi Table 2: NEED Background Studies................................................................................. xii Table 3: Companies and Respondents of Khonkaen Doing Business Survey 2005.......... xii Table 4: Indicative Cost Estimates for Investment Options............................................ 148 Table 5: Advantages and Disadvantages of Silk Races in Thailand .............................. 153 Table 6: Summary of THAISEM Rice Simulations ......................................................... 186 Table 7: Summary of THAISEM Rice Simulations by Regions (% Change).................. 186 Table 8: Summary of THAISEM Silk Simulations .......................................................... 186 Table 9: Summary of THAISEM Silk Simulations by Regions (% Change) ................... 186 viii Table of Boxes Box 1: Thailand's Regions.................................................................................................. 5 Box 2: International Experience on Lagging Regions...................................................... 13 Box 3: The Regional City Project..................................................................................... 41 Box 4: Business Case Studies of Northeast Exporters...................................................... 57 Box 5: Eastern Seaboard Program................................................................................... 61 Box 6: Return Migration and the Asian Crisis ................................................................. 85 Box 7: Infrastructure-Led Development - The Case of the Former East Germany........ 103 Box 8: Rainfall-indexed Insurance for Indian Farmers ................................................. 129 Box 9: One-Tambon-One-Product (OTOP).................................................................... 160 Box 10: Summary of Growth Recommendations ............................................................ 179 Box 11: Summary of Poverty Reduction Recommendations........................................... 181 Box 12: The Case for Northeast Public Resources......................................................... 182 Box 13: Summary of Public Expenditure Recommendations ......................................... 183 Box 14: Growth............................................................................................................... 187 Box 15: Poverty............................................................................................................... 188 Box 16: Cities.................................................................................................................. 188 Box 17: Enterprise.......................................................................................................... 189 Box 18: Workers ............................................................................................................. 190 Box 19: Students ............................................................................................................. 191 Box 20: Infrastructure..................................................................................................... 191 Box 21: Mekong Region.................................................................................................. 192 Box 22: Villages.............................................................................................................. 192 Box 23: Rice.................................................................................................................... 193 Box 24: Silk..................................................................................................................... 193 Box 25: Public Expenditures .......................................................................................... 194 ix Foreword Preface Regional development is about promoting economic growth, to share the benefits of growth across regions and communities, and to connect regions within the country and with the rest of the world. This agenda has for a long time played a central role in Thailand's development model. There is little doubt that commitment to regional development by policymakers, firms and civil society has contributed to the enviable record of Thailand's regions on growth and poverty reduction. But there is also great interest in improving the effectiveness of economic policies in achieving balanced regional development, promoting backward communities and supporting private sector growth in poor areas. Exploring these issues for Isan, the Northeast of Thailand, the country's most populous and poorest region, is the motivation for this joint study by the Northeastern Region Economic and Social Development Office (NEESO) and the World Bank. The report is organized around three basic topics: the economic record, the constraints to economic development and strategies to promote economic development. It is written for senior policymakers, development practitioners both within Thailand and the international community at large, and the Northeast population. It offers a systematic, policy-oriented analysis of Northeast economic development and draws lessons for balanced regional growth in Thailand. The consultations in Khonkaen and Bangkok have shown that Thai policymakers, civil servants, researchers, and civil society do not have to be convinced about the importance of balanced economic development. But they are keen to learn from specific experiences on how to do it better. We hope the reader will find that this report provides a refreshing and sometimes provocative look at familiar issues ­ and also sheds new light on the way forward. This report is dedicated to the Isan people who for centuries carved out a modest living in difficult conditions from an austere, acidic soil that suffers from frequent floods and droughts. By learning to make do with what they have, they have developed an admirable resilience, and made much progress in improving their incomes and quality of life, while looking forward to a better future. x Acknowledgements This report was prepared in partnership between the Northeastern Region Economic and Social Development Office (NEESO) and the World Bank. The principal author of the report is Kaspar Richter, World Bank task manager. The core team included K. Montree Boonpanit, NEED Project Manager and NEESO Deputy Director, Paavo Eliste, Aphichoke Kotikula, Tim Purcell, Jutamas Thongcharoen and Khuankaew Varakornkarn. The NEED team undertook a wide range of consultations at various stages of the preparation of the report, including during three NEED steering committee meetings and eight workshops on background studies as well as the synthesis draft report in Khonkaen and Bangkok. We gratefully acknowledge the discussions with and guidance from representative of the NEED steering committee. Overall guidance for the work from the National Economic and Social Development Board (NESDB) was provided by K. Arkhom Termpittayapaisith (Deputy Secretary General, NESDB), K. Kitisak Sinthuvanich (Deputy Secretary General, NESDB), Dr. Priyanut Piboolsravut (CDP-PAM project manager, NESDB), K. Komol Chobchuenchom, Former Deputy Secretary General, NESDB and K. Dacha Wanichwarot, NEED Project Director, NEESO. Overall guidance from the World Bank was offered by Homi Kharas (Sector Director and Chief Economist, East Asia and Pacific Poverty Reduction and Economic Management (EASPR)), Indermit S. Gill (Sector Manager and Chief Economic Advisor, EASPR) and Kazi Mahbub-Al Matin (Lead Economist, EASPR). Indermit Gill helped to shape the executive summary. World Bank peer reviewers (Hans P. Binswanger, Ijaz Nabi and Mark W. Sundberg) as well as Zafar Ahmed, Zhi Liu, Stephen D. Mink and Ian Porter contributed insightful comments. Cheanchom Thongjen and Jeep Neravan assisted in production of the report. The report draws on NEED background studies provided by Alessandro Alasia, Naoya Azegami, Nareenot Damrongchai, Björn Dressel, Ulrich C. Hess, Sumeth Kaenmanee, Wiroj Kaewrueng, Punthumadee Katawandee, Suthiphan Kompinyoparp, Aphichoke Kotikula, Dusita Krawanchit, Jintanar Lamlaor, Phumsith Mahasuweerachai, Ornsaran Manuamorn, Chairat Monaiyapong, Erica Noda, Hector Ibarra Pando, Tim Purcell, Sunan Samrianram, Kanokwan Senamontri, Nongluck Suphanchaimat, Patcharee Suriya, Anongnuch Thienthong, Frank Walsh and Patrick Paul Walsh (Table 2). NESDB and the National Statistical Office readily provided easy access to macroeconomic and survey data. The report takes inspiration from the 2002 World Bank Growth Report on Brazil by Indermit S. Gill and Mark R. Thomas, work by Deon Filmer, Dilip Parajuli, Ana Revenga and Khuankaew Varakornkarn on education in Thailand, and research on binding constraints to growth by Ricardo Hausmann, Dani Rodrik and Andrés Velasco. Finally, we also are indebted to the staff at NEESO, NESDB and the World Bank Resident Mission, and the many key informants and farmers who generously gave their time to teach us about their lives. xi Table 1: Members of the NEED Steering Committee 1. Mr Komol Chobchuenchom, Former Deputy Secretary General, NESDB 2. Mr Somchet Taeracoop, Deputy Secretary General, NESDB 3. Mr Kitisak Sinthuvanich, Deputy Secretary General, NESDB 4. Mr Arkhom Termpittayapaisith, Deputy Secretary General, NESDB 5. Mr Dacha Vanichvarood, Director, NEESO 6. Mr Porametee Vimonsiri, Senior Advisor in Policy and Plan, NESDB 7. Ms Ladawan Kumpa, Representative of the Competitiveness Development Office, NESDB 8. Director of the Community Economic Development and Income Distribution Office 9. Ms Suwanee Khamman, Director of the Quality of Life and Social Development Office 10. Ms Anothai Nutasarin, Representative of the Office of the Agricultural Extension and Development (Region IV, Khon Kaen) 11. Mr Amorn Wongsurawat, Representative of the Thai Chamber of Commerce (Northeastern Region Office) 12. Mr Pakorn Leesirikul, Representative of the Federation of Thai Industries (Northeastern Region Office) 13. Mr Supachai Saiwirat, Representative of the Bank of Thailand (Northeastern Region Office) 14. Ms Piyanut Piboolsravut, Project Director, CDP-PAM 15. Mr Montree Boonpanit, Project Manager, NEED Project 16. Mr Teera Worapan, Policy and Plan Analyst 7, NEESO 17. Ms Tassanee Bounoun, Policy and Plan Analyst 6, NEESO xii Table 2: NEED Background Studies Background Study Authors Northeast Rice Value Chain Alessandro Alasia, Tim Purcell, Nongluck Suphanchaimat, Patcharee Suriya, Jarinya Saiyut and Mungkorn Promsang Northeast Silk Value Chain Alessandro Alasia, Tim Purcell, Chairat Monaiyapong, Sunan Samrianram, Kanokwan Senamontri, Jintanar Lamlaor, Suthiphan Kompinyoparp, Wiroj Kaewrueng, and Eing-on Chaiwongsa Thailand Spatial Equilibrium Alessandro Alasia, Jason Skotheim and Tim Purcell Model Northeast Business Case Studies Sumeth Kaenmanee, Anongnuch Thienthong and Phumsith Mahasuweerachai Northeast Development Strategy Sumeth Kaenmanee, Anongnuch Thienthong and for Poverty Reduction Phumsith Mahasuweerachai Regional Labor Markets Patrick Paul Walsh and Frank Walsh Regional Investment Climate Kaspar Richter, Patrick Paul Walsh and Aphichoke Kotikula Khonkaen Doing Business Survey Punthumadee Katawandee, Naoya Azegami, Nareenot Damrongchai, Dusita Krawanchit Regional Public Expenditures Kaspar Richter and Aphichoke Kotikula Public Expenditure Reforms Björn Dressel Infrastructure in East Germany Björn Dressel Primate City ­ Literature Review Erica Noda Weather-indexed Insurance for Ulrich C. Hess, Ornsaran Manuamorn, Hector Ibarra Agriculture in Thailand Pando Table 3: Companies and Respondents of Khonkaen Doing Business Survey 2005 Company Respondents Chandler and Thong-ek Law Office Limited. Ratana Poonsombudlert Law Society of Thailand Udom Suphasindhu, Narisorn Kedroj, Wirakarn Suthasetkul Tilleke & Gibbins International Cynthia Pornavalai, Pimvimol Vipamaneerut, John Limited Fotiadis, Charunun Sathitsuksomboon, Kasamesunt Teerasitsathaporn Taveesin Chanruchai Law Office Taveesin Chanruchai, Kasemsant Pradapkarn Khonkaen Land Office, Itipol Kal-on-sil, Pitsanu Tultham, Krisada Department of Land Hanchai, Wirasak Ukahad, Wanakorn Pupewkok, Tawatchai Darachalermkul, Pichai Pae-koa xiii Executive Summary This report is about balanced economic development in the Northeast of Thailand. It is about growth and poverty reduction, cities and villages, enterprises and workers, skills and education, infrastructure and trade, and rice and silk. Northeast economic development is only part of Thailand's development challenge, but it is among the most important. We look back at how the Northeast has fared in terms of growth, poverty reduction and social capital over the last decades relative to other regions in Thailand. We examine today's challenges, analyze the constraints facing those who seek to meet these challenges, and then propose an agenda that both outlines a general strategy and the specific priority actions. Record The Northeast's image has seen little changes over the last decades. Isan has a reputation of being a tranquil and backward region, far distant from Thailand's economic hubs, for a life burdened by the toils of the field rather than the stresses of modernity. But the image is misleading; its economic record suggests a rather different reality. Aided by a dynamic and rapidly changing economy, the region has had three major accomplishments: it has grown quickly, it has noticeably reduced poverty, and it has still preserved its strong communities. Stellar Growth During the last 35 years, the Northeast was one of the fastest growing economies in the world. The Northeast's average per capita growth rate of 3.3 percent since 1970 has rivaled that of Latin America, South Asia or the group of high-income countries. Its economy is three times as large now than in 1970: GDP per capita in 2004, measured in 1988 prices, amounted to Bt34,000, compared to only Bt11,000 in 1970 (Figure 1). The rise is even more impressive in US Dollar terms due to the appreciation of the Baht vis-ā- vis the US Dollar. The Northeast's GNI per capita reached US$740 in 2004, compared to US$94 in 1970. With economic growth came change in the composition of output. Agriculture accounts for just under one fifth of GDP, compared to close to two fifths in 1970. Industry increased from the early 1990s onwards and contributes now as much to GDP as agriculture. And the service sector recorded the largest gains: it provides today over three fifths of GDP, compared to over two fifths three and a half decades ago. xiv Figure 1: Regional Per Capita GDP, 1970 to 2004, 1988 Prices 160,000 140,000 120,000 100,000 80,000 60,000 40,000 20,000 0 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Bangkok Center Northeast North South Thailand Rapid Poverty Reduction With value-added per person growing three-fold in the Northeast, household living standards improved dramatically. The poverty headcount fell from 48 percent in 1988 to 17 percent in 2002, and in spite of population growth, the number of poor dropped from 9 million to 4 million people (Figure 2). Thailand has made such impressive inroads into poverty that in the own assessment of poor people, lack of food is no longer an attribute of poverty. Rising living standards are visible in higher income and consumption as well as more durable goods. For example, over two thirds of Northeast households had refrigerators in 2002, compared to only one seventh in 1988. Almost all families own a television today, relative to only just over one in three in the late 1980s. Bicycle ownership went down over time, as households switched from bicycles to motorcycles or even cars. Figure 2: Number of Poor, 1988 to 2002 20,000,000 15,000,000 10,000,000 5,000,000 0 1988 1990 1992 1994 1996 1998 2000 2002 Northeast North Central South Bangkok xv Vibrant Communities While economic forces have increased incomes and assets, the strength of community life remains one of the Northeast's hallmarks. Kinship networks of the extended family, widespread land ownership, low mobility (with the notable exception of the adolescents), and Buddhist values have created communities with strong social capital. Over nine in 10 Northeast households participate in local groups and the provision of social services. The Need for Change In spite of these admirable achievements, the Northeast is not known as an example to emulate. Perhaps one reason is that more has been expected of the Northeast, especially by the Isan people themselves. As they look to the North, South, Center and Bangkok, or across Thailand's borders, the Isan may rightly feel that Northeast development has not been rapid enough. Being home to a growing fraction of Thailand's poor provides harder evidence of this shortcoming. And not participating as much as other regions in Thailand's economic growth over the last decade makes more rapid economic development a pressing policy matter. Low Growth and Productivity Relative to Other Thai Regions Economic growth, while decent by international standards, lacks behind Thailand's other regions. Since 1970, annual economic growth fell short by one percentage point compared to the national average, and the Northeast's contribution to Thailand's GDP fell from 16 percent to only 9 percent even though the population share remained constant at around one third. The main factor behind lower economic growth is weak productivity gains. Much of the Northeast's human, physical and natural resources are absorbed in low-yielding activities. In 2004, the Northeast worker generated only one sixth of the value added of the average worker in Bangkok, Central, East and Vicinity, and just over two-thirds of the output of a worker in the North. And the gap to other regions is rising. Since 1990, labor productivity growth in the Northeast fell short by 0.4 percent compared to the North, by 0.5 compared to Thailand, and by a remarkable 7.7 percent compared to the East. Low Growth Relative to its International Neighbors Missed growth opportunities are also apparent from a look across the border. While the Northeast is poorer than other regions in Thailand, it traditionally has been richer than its GMS neighbors. Yet, differences in growth performance begin to unravel the basic income geography that had broadly remained unchanged for many decades. Excluding the rest of Thailand, the Northeast's share in GMS GDP dropped from close to one third in 1995 to just over one seventh in 2003, and the GNI per capita of China's Yunnan Province exceeded the one of the Northeast for the first time in 2004 (Figure 3). xvi Figure 3: Real Per Capita GDP Growth in Greater Mekong Region, 1993 to 2003 210 200 190 180 170 160 150 140 130 120 110 100 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Lao PDR Vietnam Northeast Yunnan Province Cambodia Myanmar Growing Concentration of Thailand's Poor The slower pace of change has held back poverty reduction. With poverty falling faster in other regions, poverty has become more concentrated in the Northeast. While one in two poor persons lived in the Northeast in 1988, three in five poor persons resided there in 2002. Poverty is about 60 percent higher in rural areas, where livelihood depends mostly on agriculture, than in urban areas, which offer jobs in industry and services. Northeast rice farmers alone account for over two fourths of Thailand's poor. Growing Strain on Isan Communities Faced with low agricultural yields and absence of off-farm jobs, about one in two Northeast families rely on migration and remittances to boost incomes and support the local economy. At the same time, almost one in two Northeast villages report many problems with migration. Migration separates family members and exposes migrants to great risks. While some get help from friends or family members who previously migrated, others who do not have such a social network end up in inhumane working conditions. The Constraints to More Rapid Development Economic growth and trade integration are proceeding at a rapid pace all around the Northeast. Unless the Northeast responds to and shapes the changes taking place in Thailand and East Asia, its growth will continue to fall behind, poverty will remain high, and migration will continue to trouble communities. While much is at stake, the opportunities are even greater. Thailand can transform the Northeast's economy and spread prosperity to neighboring regions and countries alike. Overcoming growth differences presents an enormous challenge, since the Northeast lacks certain key drivers for change. The Center, and to a lesser degree the North, relies on manufacturing, the South on agriculture, and Bangkok on transport, communication and business services for economic transformation. By contrast, the Northeast's dominant activity is retail trade, a sector without cross-regional importance and sustained by household remittances from other regions. What prevents the Northeast from making the most of its opportunities? xvii A Dominant Primate City Around the world, densely-populated urban areas have been a force behind economic concentration. They provide markets and allow firms to benefit from economies of scale, specialization and the rapid diffusion of knowledge and innovation. Bangkok, one of the world's most cosmopolitan cities, is the nation's big throbbing heart. It dominates Thailand's urban development like few cities in other countries. It had around 6.3 million inhabitants in 2000, which was about 17 times the number of residents of Thailand's second largest city. While the degree of Bangkok's primacy is unusual, the factors of primacy conform to experience elsewhere. Bangkok is the country's capital for a highly centralized government; has access to a major port; is a conduit for inter-regional traffic; is located above most of Thailand's groundwater and surrounded by fertile lands. Bangkok's primacy provides strong advantages for enterprises compared to outlying regions. They include easy access to export channels, lower transport costs, better utilities, higher labor productivity due to a skilled labor force, a more developed financial sector, close proximity to the public administration and policy makers, and strong forward and backward linkages to input and output markets. Changes in the growth dynamics of secondary cities have widened the gulf between the extended Bangkok area and other regions even further. Mirroring trends elsewhere in East Asia, the largest population growth over the last two decades has taken place in Bangkok's peripheries. In the early 1980s, the second and third largest cities were Nakhon Ratchasima in the Northeast and Chiang Mai in the North. By 2000, they had dropped back to 5th and 8th place, respectively. They have been replaced by Samut Prakan and Nonthanburi, both cities in Bangkok's vicinity, which used to be ranked only as 12th and 25th largest city (Figure 4). Clearly, Bangkok's strong pull factor has undermined urban development in outlying regions. The Northeast's urbanization rate remains the lowest in the country at 18 percent, barely changed compared to the early 1990s. Figure 4: Primary City Indicies, 1983 to 2000 3 0 2 5 2 0 1 5 1 0 5 0 1 9 8 3 1 9 9 0 2 0 0 0 L a r g e s t C it y to 2 n d L a r g e s t C ity L a r g e s t C it y t o 2 n d t o 4 th L a rg e s t C ity xviii The Inability to Attract Manufacturing The trends in the pattern of urbanization are directly related to the dynamics of the manufacturing sector, the principal driver of Thailand's recovery from the Asian crisis. Manufacturing exports increased from less than 40 percent of GDP before the crisis to close to 60 percent of GDP as of today. As the importance of manufacturing has grown, the East and Central regions have taken off: their contributions to manufacturing GDP have exceeded Bangkok's since 1996 and 2003, respectively. Firms in need of a large plant site are attracted to the Bangkok fringe, as it shares some of the agglomeration advantages, such as proximity to export facilities and input supplies, but avoids some of the disadvantages, such as high land cost. However, congestion in Bangkok and Vicinity has not benefited the Northeast, North and South, even though they offer cheap land and labor (Figure 5). The Northeast contributes only 4 percent to the manufacturing sector's value added, the same fraction as in the early 1980s. Figure 5: Spatial Distribution of Manufacturing Employment, 1996/7 and 2001/2 xix The differences in growth dynamics of manufacturing are linked to differences in the sectoral composition. Based on its comparative advantage, the Northeast specializes in labor-intensive and resource-intensive sectors, such as wearing apparel, textile and food processing, which have either contracted or grown modestly since the mid-1990s. The fast- expanding sectors are differentiated industries, such as electronic parts, machinery, and auto parts. These sectors rely on strong enterprises linkages and hence locate primarily around the Bangkok area. Northeast firms also lag behind in technological capabilities that are important for diffusion of knowledge and innovation. Low agglomeration and weak technological capabilities ultimately lead to low efficiency in combining capital and labor to generate output. According to the 2004/5 Productivity and Investment Climate Survey, the Northeast's total factor productivity is almost 30 percent less than Bangkok's. The pull of the agglomeration in the extended Bangkok area is so strong that government policies to promote the regional spread of investment have had little success. Since 1987, the Board of Investment (BOI) has divided the country in three zones based on proximity to Bangkok, and offered higher incentives to outlying zones. Nevertheless, the Northeast's share in BOI promotions averaged only 4 percent since 1997 (Figure 6). Even those firms investing in Zone 3 locate typically as close as possible to Zone 1 in order to limit transport costs while maximizing investment incentives. For example, about one half to three quarters of Northeast investment promotions go to Nakhon Ratchasima, which is in the Southwest corner of Zone 3, and close to Bangkok. Similarly, supply driven infrastructure projects are unlikely to succeed without a clear market demand. Only four industrial estates or parks are located in the Northeast, compared to 26 in the East. In addition, industrial estates dilute the impact of the BOI zoning policy as they offer similar incentives as those presented by Zone 2 or Zone 3. Figure 6: Regional Board of Investment Promotion Certificates, 1997 to 2005 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1997 1998 1999 2000 2001 2002 2003 2004 Apr 2005 Northeast North South Other Regions The unitary structure of Thailand's rules and regulations reinforces the importance of agglomeration effects. The same business, bankruptcy and labor laws apply in Bangkok as in Khonkaen. On paper, the costs of doing business are the same across the country, as outlying regions have no latitude to attract enterprises through adopting business friendly xx procedures. At the same time, firms in and around Bangkok benefit from lower shoe leather costs for obtaining investment promotions, permits and licenses through proximity to public institutions in Bangkok. The concentration of enterprises in and around Thailand's capital leads to a similar clustering in finance. Bangkok alone accounts for two thirds of commercial domestic deposits and three quarters of credits, and Bangkok and Center are home to about two thirds of all branches of domestic banks. By contrast, the Northeast accounts for only 5 percent of both deposits and credits. This gap reflects low household income and lack of business demand rather than low access (13 percent of all branches are located in the Northeast) or high cost of borrowing (formal interest rates are similar to those charged in Bangkok). In addition, about one quarter of all credits are provided through the three main public financial institutions (Government Saving Bank, Government Housing Bank and Bank for Agriculture and Agricultural Cooperatives), which are adequately presented throughout the country. An Unfriendly Environment for Skilled Workers Finding employment in the Northeast is not a problem. Even during the Asian crisis, the employment rate dropped by no more than 4 percent. This is due to two factors. First, wages are downward-flexible as labor unions are weak and minimum wage legislation is not effectively enforced. For example, wage earnings fell by 9 percent during the Asian crisis, and more than half of the daily wage workers received wages below the minimum wage in 2004. Second, when labor demand in industry and services falls short, workers become farm laborers either on the family farm or elsewhere. Although agricultural employment in the Northeast has declined by about 10 percent since the early 1990s, still almost one in two workers work in agriculture even during the slack season. While getting a job in the Northeast is easy, receiving a good wage is harder. Less than two fifths of Northeast workers earned a wage at the age of 35, and just over one fifth earned a monthly wage. This compares to two thirds and one half in Bangkok, respectively. These differences in wage employment rates link back to occupation and education, as wage employment is more common outside of agriculture and among skilled workers. Among Northeast workers, about one third is employed in services and one fifth in industry. By contrast, for Northeast workers earning a monthly wage, these shares are four fifths for services and one seventh for industry. In the Northeast, almost one in two workers on a monthly payroll have vocational or university education, compared to only one in five among wage workers and no more than one in ten among all workers. Wage employment is not only harder to come by, but it is also less well enumerated. Northeast wages, whether paid daily or monthly, are around 50 percent less than those in Bangkok. Since more Northeast than Bangkok workers receive daily wages and daily wages are only about 40 percent the level of monthly wages, the average wage received per month in Bangkok is twice as large as in the Northeast. Education is a strong determinant of wages, especially in the Northeast. For example, workers with university degrees earn about four times as much, and those with vocational training about three and a half times as xxi much, than those without completed primary education (Figure 7). This compares to three times as much for university education, and one and a half times as much for vocational training in Bangkok. These wage premia have changed little over time, even though the workforce has become more educated. For example, the share of Northeast workers with at least primary education has increased by almost 20 percent since the early 1990s. Figure 7: Returns to Education of Monthly Wage Earners Relative to Less than Primary in the Northeast, 1991 to 2004 600 550 500 450 400 350 300 250 200 150 100 50 0 F91 A91 F92 A92 F93 A93 F94 A94 F95 A95 F96 A96 F97 A97 F98 A98 F99 A99 F00 A00 F01 A01 F02 A02 F03 A03 F04 A04 Primary Lower Secondary Upper Secondary Vocational University Jobs have a lot to do with education and training. The better educated are more likely to find wage employment and receive higher wages. Much of the improvement in labor market outcomes in the Northeast over the last 15 years is linked to the dramatic progress in education. School participation rates for 14-year old children in the Northeast increased from less than 40 percent in 1988 to over 80 percent in 2002. The enrollment gap to other regions has been eliminated for all age groups up to upper-secondary education. Among the 15 to 21 year-old, the share of Northeast students who attended upper secondary, vocation, or university education increased from 5 percent in 1988 to over one quarter in 2002. Since less than one in twenty Northeast students attend a private school, the bulk of this progress was achieved through the public education system. However, the transition to a modern, child-centered and participatory education system in the Northeast has only begun. Northeast students fall behind in test scores at grade 9 and 12, partly because teachers have lower qualifications. In addition, school fees are a strong deterrent for students to attend private vocational and university education. Northeast household with students in private vocational institutions spend on average 14 percent of their income on fees. Yet, it is those degrees that fetch the highest premium in the labor market. Given the lack of jobs and lower wages, workers turn to migration, especially among the young. This leads to a twin-peak population structure in the Northeast, with many children and adults of 30 years or older, and a single-peak structure in Bangkok, with a high concentration of 20 to 35 year-old (Figure 8). But perhaps the most important effect of migration is remittances. More than one in two Northeast households benefited from such payments in 2002, compared to around 45 percent in 1996. Among receiving households, xxii these remittances amounted to around one third of household income, and they lowered poverty from 17 percent to 12 percent. Figure 8: Population Pyramids in the Northeast and Bangkok, 2002 Age Pyramid, NE Region 2002 Age Pyramid, Bangkok 2002 90 90 80 80 70 70 oup oup grega 60 50 arey 40 gregaraey 60 50 40 eviF 30 eviF 30 20 20 10 10 0 0 7.5 5 2.5 0 2.5 5 7.5 7.5 5 2.5 0 2.5 5 7.5 Female Male Female Male Percent of Population Percent of Population Adequate Infrastructure for Internal, Not Regional, Integration When assessed relative to the rest of Thailand, the Northeast appears adequately endowed with transport infrastructure. Road length, unpaved and asphalt, was expanded continuously in the Northeast, as in the rest of the country. The Northeast has two railway lines and eight airports, which connect the region to Bangkok. Such infrastructure integrates the Northeast with other regions, but it also facilitates the supply of goods and services from the major urban growth centers to the Northeast. This is especially attractive if other infrastructure continues to lag behind. While progress was achieved in power - almost all Northeast households had electricity connections by 2002, Northeast firms in the 2004/5 PICS complain more often than other firms about power outages. The Northeast also compares poorly in terms of water, sanitation and communication investment. For example, only one in seven Northeast business establishments have access to computers, and just over one third of them have internet connections. While the rural fixed telephone net remains patchy, mobile phones make up to some degree for the low coverage. Strong economic growth in the GMS region raises the benefits from close ties among its member countries. The economy of the Northeast is only half the size of China's Yunnan Province and two fifths the size of Vietnam (Figure 9). While informal cross-border trade has always taken place, but subregional integration in formal trade is a more recent development. Since 1980, Thailand's exports increased in real terms annually by 1 percent to Lao PDR, by 13 percent to Cambodia, by 24 percent to Vietnam, and by 23 percent to Myanmar. These are encouraging developments but the direct benefits from trade integration to the Northeast remain small. Less than one percent of the around 13,500 Thai export companies are located in the Northeast. While the bulk of exports and imports with Lao PDR go through customs in the Northeast, trade with Vietnam, which accounts for the largest part of exports to the Mekong region, takes place mostly through the sea-route, by- passing the Northeast. The Northeast will only capture a greater share of the expanding trade among GMS countries if trade through the land route becomes less cumbersome. xxiii This requires improvements of infrastructure and custom regulations not just in the Northeast but also in other GMS countries. Figure 9: GMS GDP (Current Dollar Billion), 1995 to 2003 240 220 200 180 160 140 120 100 80 60 40 20 0 1 2 3 4 5 6 7 8 9 Other Thailand Northeast Vietnam Yunnan Province Myanmar Cambodia Lao PDR Low Productivity in Agriculture, the Principal Activity of the Poor 1. While cities are important growth drivers, the bulk of the Northeast population resides in villages. Over four in five families live in rural areas. The need for higher income is perhaps nowhere greater than in agriculture. The Northeast generates just over one fifth of Thailand's agricultural GDP, even though the region accounts for one half of the farms and two fifths of the agricultural land (Figure 10). Low agricultural productivity is linked to factors like small farm size, low market power of farmers, limited irrigation and lack of fertilizers and pesticides use. But perhaps the most important reasons are weak natural resources and the focus on rice production, a water-intensive crop. The Northeast has a long dry season as well as porous and highly saline soils which retain water poorly. Figure 10: Agricultural Value Added by Agricultural Worker, 1991 to 2004 0.12 0.10 0.08 0.06 0.04 0.02 0.00 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Vicinity East Central West North Northeast South Over nine in ten Northeast farmers produce rice, but lack of water limits most of them to wet-season cropping. A large group of subsistence rice farmers exist alongside a small group of commercial farmers. The first group tends to produce glutinous rice for own consumption through rain fed main season cropping, while the second group applies irrigation to produce non-glutinous varieties destined for urban and export markets. Similarly, there is a duality in the production of silk with coexistence of subsistence oriented farmers with more commercial farmers, differentiated by the type of sericulture xxiv practiced (polyvoltine, poly-bivoltine and bivoltine). The immediate challenge is to raise productivity of subsistence farmers and integrate them into value chains. While efficiency gains through reforming production, processing and trade systems are still possible, the focus should be on increasing revenues by identifying highly processed and transformed rice and silk products that fetch a high value. In the medium term, improvements in rural living standards hinge on on-farm and off-farm diversification. A Noticeable Shortfall in Public Spending Regional economic development depends, among other factors, on how key sectors are funded with public resources. Channeling public resources to disadvantaged regions, if done well, can be a powerful way of promoting convergence in living standards. Yet, the Northeast receives fewer public resources than any other region and the expenditure gap with other region has remained fairly constant last five years (Figure 11). The Northeast obtained in FY 2003 Bt6,400 per capita (1999 Prices; US$160), which was one third less than the Center and 27 percent less than the North and the South. The spending shortfall compared to these three regions was close to around 30 percent in FY 1999 and FY 2003. Figure 11: Government Spending, FY 1999 to FY 2003 (Baht Per Capital, 1999 Prices) 11,000 10,000 9,000 8,000 7,000 6,000 5,000 4,000 3,000 2,000 1,000 0 FY99 FY00 FY01 FY02 FY03 Northeast Central North South The Northeast is disadvantaged especially with regard to capital spending. Relative to the non-Bangkok average in FY 2003, capital expenditures were 48 percent lower and recurrent expenditures 17 percent lower. Lower spending overall translates into lower spending by function. Among the 11 main functions, the gap in FY 2003 was 8 percent for education, 33 percent for social security, 39 percent for transport and housing, and 43 percent for agriculture. The low per capita shortfall for education is related to the importance of spending on teachers' wages which are normalized across the country. This evidence is consistent with a systematic bias arising from a rigid budget allocation system that preserves existing differences in staffing and facilities by applying fixed budget norms with little regard towards performance and new spending priorities. Such norms have a larger bearing on recurrent spending, especially salaries, whereas capital spending is by its very nature more discretionary and affected by political bargaining. The concentration of the economic sector in and around Bangkok is reinforced by the organization of Thailand's public sector. The highly centralized fiscal system grants only limited autonomy to lower government levels in terms of functions, area, staffing, funding xxv and decision making. The central government appoints the chief local officials, determines local salaries, and approves local budgets. Even local utilization of the restricted funding is to a large part centrally mandated. For example, staffing levels and staff appointments of local governments are now centrally controlled. Local authorities are required to hire personnel and pay salaries, wages, and benefits in accordance with central regulations that often result in overstaffing and overspending. A second issue is local administrative capacity. The Northeast has the lowest number of government staff per person. Northeast public wages are from 5 percent to 56 percent lower than in other regions, as government staff is less qualified than in other regions. The Agenda: General Principles and Priority Measures Economic affluence is associated with prosperous enterprises, and enterprises locate where they expect the highest profitability. Firms will only invest in the Northeast if resources, business climate and markets are as or more favorable for their products than in other regions: economic development of the Northeast is connected to economic development in the country as a whole. Thailand has also thus far connected with its neighbors through Bangkok. While this has worked well for the country in general, this strategy may now have become a constraint for Northeast growth. The time may be right to augment the strong regional links through Bangkok and proximate areas with equally strong subregional international links through other parts of Thailand, especially the Northeast. But the success of these shifts in strategy in helping Northeast living standards converge with those of other Thai regions will depend on how well-prepared the Isan population is to compete. Fostering such a conducive climate will require government actions to upgrade services and institutions centered on three pillars: Thailand, the poor and the Greater Mekong Subregion. What is Good for Thailand is Good for the Northeast In the foreseeable future, the Northeast economy is dependent on the dynamism of the national economy. In the absence of weather shocks, growth in the Northeast tracks growth in Thailand closely, and jobs in the extended Bangkok area provide employment to Northeast workers whose remittances support a large service sector in the Northeast. To sustain Thailand's economic expansion requires a focus on the new growth locations. Peri- urban areas in Bangkok's neighborhood have attracted population inflows and have become the core of the manufacturing sector, Thailand's most important growth driver. The specific priorities are: ˇ First, improving the business environment in the manufacturing in Central and East will be essential to ensure that these companies improve their productivity and continue sustaining Thailand's export boom. This entails addressing deficits in infrastructure and business services, such as improvements in the logistics system and the provision of one-stop government centers. ˇ Second, prosperity in Bangkok depends not so much on large scale industries but more on high quality business and producer services as well as high amenity and xxvi sophisticated cultural products. These activities require dense, high transaction business environments with easy accessibility and flourish through the low cost of doing business, the ­ by international standards ­ low costs of living, and the cosmopolitan flair. This implies a focus on urban mass transit infrastructure and communication. What is Good for its Poor is Good for the Northeast In the longer term, economic convergence will depend on how well Thai policymakers meet three challenges--improved skills, service delivery to rural areas, and general government. ˇ First, the skills of Northeast workers have to improve to allow them to compete on par with workers from other regions for decent jobs: what is good for Northeast workers is good for the Northeast. While the Northeast has succeeded in closing the enrollment gap up to upper-secondary education, access to vocational education is still lagging and test scores are lower at higher grades. Employers reward vocational and university education, but not upper secondary education. ˇ Second, the prospects for rural livelihoods have to be made better. Growth has helped to lower poverty, but the duality of commercial and subsistence farmers has reduced the impact. Grass-root initiatives support rural areas but they do not differentiate between poor and non-poor villages. While this makes these programs politically more sustainable, it dilutes the effect on poverty. By exploiting the overlap of high poverty incidence and large number of poor in the Northeast, geographical targeting of public programs can help to eliminate poverty by the end of the decade. Such programs should allow for community level monitoring and bottom-up inputs through community plans. These changes should be integrated in public expenditure reforms that emphasize results-orientation, area allocations and monitoring and evaluation. However, the main channel for poverty eradication will not be government programs but improvements of rural incomes. Reducing costs, improving productivity and on- and off-farm diversification will go some way. These efforts should be complemented by measures to raise the value-added of products and reduce the vulnerability to weather shocks. ˇ Third, strengthening and empowering public administrations from villages to provinces will enable them to provide effectively the services demanded by enterprises. Zoning policies and industrial estates have failed to promote investment in outlying regions due to off-setting incentives as well as institutional weaknesses. Government officials should have the mandate and funds to improve the local business climate to attract investment. Increased responsibilities of province-level officials should facilitate the coordination between tambon administrations and municipalities, while more power for revenue collection and decision making in municipalities should support the development of secondary cities. xxvii What is Good for the Greater Mekong Subregion is Good for the Northeast Finally, promoting and integrating with a prosperous Greater Mekong Subregion can turn the Northeast from a land-locked into a land-linked region. This will require reducing structural and institutional impediments to the movements of goods, people, and capital. Lowering transaction costs should also help GMS to follow the example of ASEAN, which was transformed through stable macroeconomic environments, reliable and transparent investment rules, and foreign investment in internationally integrated production systems. The priority measures are: ˇ First, overcoming inadequate transport and communication linkages and promote common networks in transport, power distribution, trade and commerce will help to boost competitiveness by integrating markets and exploiting scale economies. ˇ Second, to allow the integration of markets for products and services as well as for inputs such as finance, labor and energy, physical investments should be accompanied by investments in easing processes and building capacity. This includes the harmonization of legal and regulatory frameworks and the facilitation of cross-border flows. Introduction Convergence During the last 35 years, the Northeast was one of the fastest growing economies in the world. The Northeast's long-run growth rate has rivaled that of Latin America, South Asia or the group of high-income countries. Strong growth also led to a steep fall in poverty, dropping from close to one in two people in the late 1980s to less than one in five today. Nevertheless, the Northeast is not widely known as an example to emulate. Perhaps one reason is that more has been expected of the Northeast, especially by the people of Isan themselves. After all, the country is blessed with an energetic populace. Perhaps the population compares itself with Malaysia, Singapore, Taiwan and Korea, and other countries in the region that are universally praised as stellar performer. Or it may be that this economic growth performance has been accompanied by high inequality and social costs, thus diminishing the quality of growth in the eyes of some observers. But the most likely reason is the Northeast's performance compared to Thailand's other regions. During the last four decades, growth in the Northeast has not kept track with Thailand's growth pole in the Central region and Bangkok. And the gap is increasing: the growth divergence doubled over the last 17 years compared to the previous 17 years. This study is about the Northeast of Thailand, and it is about balanced regional development. It is about growth and poverty, cities and villages, enterprises and workers, skills and education, infrastructure and trade, and rice and silk. Regional economic convergence is only one part of the development challenge, but it is in Thailand among the most important. This study shows why. We look back at how the Northeast has fared in terms of growth and poverty reduction over the last 35 years relative to other regions in Thailand. We look at the resources and institutions available today for Northeast economic development. And we look ahead at what the challenges are for the future, and how to think of approaching them. Much of Thailand continues to grow rapidly, driven to a considerable extent by the growth pole of the extended Bangkok area. Urbanization is proceeding, but from a lower level and slower than in other countries in the region. Urban centers look for improved competitiveness, peri-urban areas to upgrading of services, and rural areas for off-farm diversification and farm productivity growth. Economic policies have to strike a balance between supporting lagging regions to reduce poverty and achieve economic integration one the one hand, and tackling growth constraints in prosperous regions that provide the underpinning for the country's economic growth on the other hand. The complexity of responding to these demands is greater than ever, and the cost of getting things wrong very high. Failing to improve the competitiveness of the urban centers today would have a huge economic and social impact ­ and highly expensive to fix later. Neglecting the needs of people remaining in the poor areas and villages would be costly in human, economic and political terms. 2 Agglomeration Many developing countries have well-recognized areas where poverty has been persistently high and economic growth has not kept up with other regions. The western provinces of China, the Northeast of India, the Southern States of Mexico, the West of Argentina, and the Northeast of Brazil are just few examples of such "lagging regions". Similarly, developed nations, such as the Canada, Italy, or the US also have regions with chronically low incomes compared to national averages. The Northeast of Thailand, the country's most populous region, is also an example for a lagging region. The challenge of such regions is to grow and converge with the other regions in the country. Economic geography, a branch in economics developed in the 1990s, is all about where economic activity takes place. It offers two concepts that are important in understanding lagging regions (Krugman 1998). One part of the literature argues that differences in economic development across locations can emerge from underlying, inherent differences in those locations, such as climate, sea access and geography. Another part of the literature explores how such initial disadvantages embedded in geography, climate, policy biases or cumulative outcomes of historic accidents, can lead to regions failing to develop a self- enforcing economic dynamism. Thin markets with little backward and forward linkages, low purchasing power, weak skills of the labor force and of local administrations combine to make them unattractive for business. In many ways, Thailand is a prime example for the relevance of these ideas. Perhaps due to inherent geographical weaknesses and population pressures, the Northeast was historically a poor region. The region was populated predominantly by Lao families who resettled there in various waves from the 14th century up until the mid 19th century. The region formed for centuries a buffer zone between the Lao and Siamese kingdoms. Towards the end of the 19th century, Thailand started administrative reforms to transform a loosely integrated kingdom into a nation state governed from Bangkok, a location with sea access surrounded by fertile lands. While Bangkok became quickly the undisputed political, economic, and financial center of Thailand and well integrated through trade with other countries and continents, the Northeast economy continued to be dominated by subsistence farming with little trade relations to surrounding areas up until the mid 20th century (Gustafson 1994). The dramatic modernization process, which transformed the country from a poor rural nation into a fast growing economy, has not benefited all regions equally. A high-wage, high- income economy in Bangkok and surrounding areas, driven by dynamic industrial and service sectors, coexists with a much less developed and more rural economy in the rest of the country. 3 Scope The main message of this report is that in order to promote growth convergence of the Northeast with the rest of the country, Thailand needs to create an environment conducive to productivity growth in outlying regions. Thailand's regional economic convergence agenda entails overcoming location disadvantages, integrating markets within the Mekong region, improving local institutions, aligning investment incentives, and increasing social stability by improving skills and lowing poverty. The report is about regional economic convergence. We look at human, physical, social and natural resources and ask whether they contributed or hindered economic convergence. Of course, gains from, say, improvements in education and health or natural resource management should not be seen only in this light. Income is only one factor in public welfare and social outcomes contribute to human capabilities that are ends in themselves. There are many drivers for regional economic convergence. Only some of them are dealt with comprehensively, and some are not covered at all. This restriction reflects the emphasis recommended by NESDB and peer reviewers at the concept stage, the desire to avoid duplication with existing literature, data constraints, and, last but not least, the need to keep the task manageable. One such decision was to emphasize selectivity over comprehensiveness. This report does not propose a comprehensive Northeast development strategy but elaborates a set of issues relevant to Northeast development. Cursory treatment is given to topics such as natural resource management; on-the-job training, the service sector, small-and-median enterprises and social outcomes. Some issues are not covered at all, such as health, gender equality, and inequality; as well as topics that are important to Thailand's economic success but do not explain differences in regional performances, such as national fiscal, monetary and exchange rate policies. The report also does not deal with the issue of province clusters, as Thailand's public sector reform agenda quickly evolved beyond this ultimately stalled initiative. The report varies greatly in terms of the level of detail, and this limitation alone would make this report unfit to be a fully-fledged strategy. Many sections focus on the big picture to capture how the Northeast as a region stands out compared to the North, South, Center and Bangkok. Only some parts look also at province level differences, and even fewer sections talk about variations across districts (amphoe), sub-districts (tambons), or villages. Rice and silk are the two sectors analyzed most carefully, although only a fraction of this work could be integrated into this synthesis report. These sections build on original field work involving a participatory value chain survey. In addition to training two government staff during the survey period, NESDB funded a two-week value-chain training course, held by Agrifood Consulting International which conducted these surveys. The detailed findings from this project, in addition to the simulations run with the especially developed Thailand Spatial Equalibrium Model (THAISEM), are provided in electronic format together with this report. 4 Approach and Content Based on the work commissioned for this report and the contributions of other researchers in Thailand and elsewhere, this report reviews the Northeast's economic convergence with the rest of Thailand and proposes selective policies to support balanced regional development in Thailand. It is organized in three sections (Figure 12). The first part lays out the Northeast's record on growth and poverty reduction. While the Northeast has a good performance on both scores, it is nevertheless worse than in the rest of Thailand. The second part tries to account for the slower pace of growth and poverty reduction. In the spirit of the methodology proposed in Rodrik (2004) and Hausmann et al (2004), it attempts to identify the main binding constraints for economic convergence. This approach emphasizes looking at trends in quantities as well as prices in search of idiosyncrasies that set the Northeast apart from Thailand's other regions and hence could account for differences in economic performance.1 It investigates in detail how the Bangkok growth pole has interplayed with the economic convergence of the Northeast across different domains. It starts with Thailand's urbanization, moves on to labor markets, both from the demand side (enterprises, with a focus on manufacturing) and the supply side (current and future workforce); turns to infrastructure within the Northeast and across the border in the Greater Mekong Subregion (GMS); then looks at villages in general and at rice and silk value chains specifically; and finally considers public expenditure allocations and reforms. The final part pulls together the findings and attempts to prioritize policies. Figure 12: Report Content Outcomes Growth Poverty Reduction Localities Cities Villages Mekong Region Labor Market Enterprises Workers Rice Silk Value Chain Students Resources Intrastructure Public Expenditures 1At the most aggregate level, Hausmann et al (2004) distinguish three types of binding constraints: high cost of finance, low private appropriability of social returns and low social returns. The second part of this report can be viewed as an exploration of why social returns to investment are lower in the Northeast than in other regions. The other two candidate constraints are considered only briefly, as they are unlikely to vary across Thailand's regions. 5 Box 1: Thailand's Regions Since the Royal Thai Government North (RTG) does not have any regional level representation, there are Northeast different ways of grouping Thailand's 76 provinces into regions. The 8-way breakdown consists of Bangkok municipality, Central Vicinity (5 provinces), Central (6 provinces), East (8 provinces), West (6 provinces), North (17 provinces), Northeast (19 provinces) and the South (14 East West provinces). For most of the report, we will use a 5-way breakdown, where Vicinity, Central, East and Bangkok & Vicinity West are grouped into one region called "Center". Occasionally, South such as for the analysis of the 2004/5 Thailand Productivity and Investment Survey, we will lump Bangkok and Vicinity together, in which case Center will refer to the collection of Central, East and West only. 6 I. Record Growth Growth Gap The Northeast's economy expanded greatly over the last four decades. GDP per capita in 2004, measured in 1988 prices, amounted to Bt34,000, compared to only Bt11,000 in 1970. The rise is even more impressive in US Dollar terms due to the appreciation of the Baht vis- ā-vis the US Dollar. The Northeast's GNI per capita reached US$740 in 2004, compared to US$94 in 1970.2 In spite of this strong performance, the Northeast's progress is barely visible in Figure 13.A when plotted relative to Thailand's other regions. The Northeast was the poorest region in 1970, and has remained the poorest region until today. Indeed, Northeast's 2004 GNI per capita is no more than 30 percent of Thailand's income level. As a country, the Northeast would be the only region Thailand's to be classified as a low income country. While the North and South expanded at a comparable speed, although from higher levels, growth in Bangkok and especially the Center was clearly faster. Figure 13.B takes the data from Figure 13.A, but displays it differently, as regional GDP per capita relative to the Northeast. The Northeast's income gap is constant relative to the North; increases moderately since the mid 1980s compared to the South; rises continuously, and at a higher rate since the mid 1980s, relative to the Center; and increases between the mid 1980s and 1993, before declining to about the same level as in 1990, compared to Bangkok. Average annual real per capital GDP growth of the Northeast over the entire period equaled 3.3 percent, which is slightly above the North (3.1 percent) and somewhat below the South (3.7 percent). It lagged compared to Bangkok, which expanded by around 4.1 percent, and to the Center, which grew by 5.3 percent. Overall, between 1970 and 2004, the Northeast growth gap compared to Thailand amounts to one percentage point. The Northeast's growth gap increased over the last 35 years. From the early 1970s to mid 1980s, the period which marked the transition from import substitution to export orientation, Northeast income dropped from 45 percent to around 40 percent of the Thai average (Figure 13.C). From the mid-1980 to mid-1990s, a twin export and investment boom increased national growth more than Northeast growth, and the relative income level dropped to one third. The growth miracle was unsustainable and sowed the seeds of the subsequent collapse. While the Asian crisis hit the Northeast slightly less than the Thai economy, the recovery since 1999 has also been weaker. Overall, the growth gap since 1987 amounts to 1.5 percent, about double the number from the previous 17 years. 2The World Bank uses 2003 GNI per capita, adjusted by smoothing the exchange rate with the Atlas conversion factor, to group countries into four categories: low income, $765 or less; lower middle income, $766 - $3,035; upper middle income, $3,036 - $9,385; and high income, $9,386 or more. 7 Figure 13: Regional Growth in Thailand A. Regional Per Capita GDP, 1970 to 2004, 1988 Prices 160,000 140,000 120,000 100,000 80,000 60,000 40,000 20,000 0 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Bangkok Center Northeast North South Thailand B. Regional GDP Per Capita Relative to Northeast, 1970 to 2004, Northeast=100 1,000 900 800 700 600 500 400 300 200 100 0 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Bangkok Center Northeast North South Thailand C. GDP Per Capita Growth of the Northeast (Blue) and Thailand (Green), 1970 to 2004, Percentage Points 15 10 5 0 -5 -10 -15 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 8 GDP Shares GDP per capita measures the output of goods and services produced in a country, divided by the population. The Isan population totaled 12 million in 1970 and 21 million in 2004, implying an annual population growth rate of 1.7 percent. By contrast, the Thai population increased from 34 million to 64 million over the same period, equivalent to an annual population growth rate of 1.8 percent (Figure 14.A). Overall, the economic and population growth rates imply that the Northeast GDP share fell from 16 percent in 1970 to 10 percent in 2004 (Figure 14.B). The North's share contracted from 16 percent to 9 percent and the South's share from 11 percent to 9 percent. The joint share of these three regions in Thailand's value added fell from 43 percent in 1970 to no more than 28 percent in 2004. In other words, two thirds of the population contributes less than one third of the nation's value added. The Thai economy is more and more concentrated around Bangkok and the Center. Their GDP shares rose more or less continuously during the 1970s and 1980s. Bangkok's share peaked in 1993, even prior to the Asian Crisis, and declined to around 27 percent in 2004, the same level as of 1970. By contrast, the Center's share continued to rise in the 1990s up to now. It accounted for 45 percent of Thai GDP in 2004, compared to 35 percent in 1990 and 30 percent in 1970. Clearly, in the last decade or so the dynamism of the Thai economy spread from Bangkok to surrounding regions.3 A simple way of visualizing the differences in regional contributions to the population and value added per capita is through circle maps. Figure 14.C displays province-level population and GDP per capita numbers, where the radius of the circle is proportional to size. Northeast provinces are depicted in yellow and other provinces in green ­ with the exception of outliers shown in red: these are provinces whose values exceed the 75th percentile value by three times the range between the 25th and the 75th percentile. In terms of population size, the only outlier is Bangkok, an issue that we will return to later. In terms of GDP per capita, there are six outliers, all located in and around Bangkok. Moving from the left to the right figure, the Northeast region shrinks in size, indicating that its economic potential does not match its population potential. 3Bangkok posted strong growth in 2004, the first year since 1993 in which its GDP share rose. It is too early to tell whether this marks a turnaround in the trends. 9 Figure 14: Regional population and GDP shares, 1970 to 2004 A. Regional Population Shares, 1970 to 2004 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Northeast Bangkok Center North South B. Regional GDP Shares, 1970 to 2004, 1988 Prices 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Northeast Bangkok Center North South C. Population Circle Map, 2002 D. GDP Per Capita Circle Map, 2002 10 International Comparison In spite of the strong record, the Northeast's progress seems modest when plotted against the trend of the main middle income countries in East Asia. The region as a whole grew by 5.5 in real per capita terms from 1970 to 2003, pulled up by an annual average per capita growth rate of 7 percent in China (Figure 15). Thailand's annual growth average is 4.3 percent, and Malaysia's and Indonesia's 4.0 percent. Only the Philippines (1.1 percent) grew slower than the Northeast after losing the growth momentum from the early 1980s onwards. Overall, the Northeast grew solidly since 1970 by world standards, but the record looks less impressive when contrasted with its dynamic neighbors and Thailand's regional competitors. In the global context, the Northeast growth looks very healthy by almost any measure. The Northeast grew on average in real per capita terms by 3.3 percent every year since 1970, compared to 2.2 percent for OECD countries, 1.3 percent for upper middle income countries, 2.6 percent for lower middle income countries, and 1.6 percent for low income countries. These differences matter hugely. Growing at the pace of the Northeast implies a tripling of per capita income within 35 years. Growth at OECD speed would have led to only a doubling, and expansion at low income country pace to only a 75 percent rise of per capita income. Figure 15: Growth in Northeast and East Asia Northeast, Thailand and East Asian countries: Real Per Capita GDP Growth, 1970 to 2003, 1970=100 1,000 900 800 700 600 500 400 300 200 100 0 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Northeast Thailand Malaysia China Indonesia East Asia Philippines 11 Growth Divergence Time to take stock: The Northeast is not a stagnating but a lagging region. Its economy is three times the size now than it was in 1970, a respectable performance by almost any benchmark. But Thai standards are taxing: The Northeast grew slightly faster than the North, but somewhat slower than the South, significantly slower than Bangkok, and much slower than the Center. Thailand's per capita GDP more than quadrupled, so that the Northeast's per capita income level dropped from 45 percent to 30 percent of the national average. The Northeast people, representing one third of the population, contribute no more than one tenth to the national value added, and the recent performance is worse than past performance, in spite of the Asian crisis. The average annual growth shortfall relative to Thailand overall increased to 1.5 percent since 1986, compared to 0.7 percent from 1970 to 1986. Figure 16 displays the convergence challenge. The figure at the top plots the real per capita regional GDP average growth rate for the five regions from 1970 to 1986 against the log of real per capita regional GDP in 1970. The Northeast ranks lowest relative to the horizontal axis, as it was the poorest region in terms of per capita income in 1970, followed by the North, South, the Center and Bangkok. By contrast, the North ranks lower than the Northeast relative to the vertical axis, as the Northeast grew faster (3.3 percent compared to 2.7 percent) over this period. Remarkably, growth in Bangkok was only moderately faster (3.3 percent). Overall, the period was marked by regional divergence as indicated by the upward-sloping regression line. The average growth rate increased with the initial income level of a region.4 The figure at the bottom shows the same plot for the period from 1986 to 2004. Relative to the figure above, the data points move towards the northeast, as the regions have higher initial incomes and higher growth rates post-1986 than pre-1986. The ranking vis-ā-vis the horizontal axis remains unchanged, as the differences in growth rates between 1970 and 1986 were not large enough for any region to overtake another region. The increase in the growth rates of the less well-off regions is small compared to rise for Bangkok and the Center. In turn, the slope of the regression line remains positive and higher than before.5 Regional income levels diverged even more post-1986 than pre-1986. The neoclassical growth model predicts that low income regions grow faster than high income regions. Yet, the Northeast, Thailand's poorest region, grew slower than Thailand since 1970. And the growth gap increased since 1986. However, there is one important caveat: the Northeast growth rate was higher (3.3 percent) in the second than in the first phase (3.1 percent). 4Formally, the regression slope is equivalent to a test of unconditional convergence. Assuming each region has the same steady-state output, the slope of the regression line is equal to , which captures the speed of convergence to the steady-state. A positive signifies divergence, a negative convergence (Barro and Sala-I- Martin 1991). 5The speed of divergence increases from 0.36 percent during 1970 to 1986 to 0.76 percent during 1986 to 2004. This calculation is only illustrative, as the number of regions is too low for a statistical analysis. 12 Figure 16: Regional Economic Growth Divergence, 1970 to 1986 and 1986 to 2004 6891-079119 98 5 5. 70 PRG CENTRAL NT 5 CPfoof 4. teaRt e thwort h 5 SOUTH 3. GlaunnAlaeR BANGKOK KO NORTHEAST RTHE NORTH 5 2. 8.5 9.5 10.5 11.5 Log of 1970 Real Per Capita GRP GR Fitted al es tted v ues Outsi e ortheast ts d N Northeast 4 002- CENTRAL 86 5 19 5. PRG BANGKOK CP 5 of 4. etaR htwor SOUTH 5 GlaunnAlaeR 3. NORTHEAST NORTH 5 2. 8.5 9.5 10.5 11.5 Log of 1986 Real Per Capita GRP Fitted values Outside Northeast Northeast 13 Box 2: International Experience on Lagging Regions Mexico, Brazil, and China all have regions whose per capita GDP levels are lower and poverty incidence higher than the national averages. The southern part of Mexico, traditionally the poorest region in the nation, grew during the 1990s at less than half of the national per capita growth rate of 3.5 percent. The economy of the Northeast of Brazil almost stagnated during the second half of the 1980s, leading to a 1.5 percent per capita growth gap to Brazil as a whole. Yunnan, the poorest province in the Southwestern region of China, reached only just over half the national growth rate of 12 percent during 1999 to 2003. Different factors have hindered economic activity in these regions. In spite of abundant natural resources, growth in southern Mexico is held back by a complicated tax and property right system as well as latent ethnic conflicts. In the Northeast of Brazil, difficult weather conditions reduce agricultural productivity, leading to large migration to metropolitan areas. Economic reforms in China focused on urban development, while rural-urban mobility is discouraged through regulations. In the late 1980s, Mexico launched the Oportunidades program to provide a social safety net for the poor, and increased federal and state resources to resolve land conflicts, improved highway linkages, and promote manufacturing. Nevertheless, poverty in the rural South fell only by 2 percent compared a nationwide reduction of 18 percent. In the Northeast of Brazil, the government focused on providing education and training for the poor. Perhaps aided by such programs, Northeast per capita growth reached 2.7 percent during 1990 to 1998, about 0.2 percent in excess of the national growth rate. China launched a rural development initiative in early 2001. It includes educational reform, infrastructure improvement, and agricultural development, while the restrictions on labor mobility remain in place. The initiative focuses on Yunnan, Sichuan, Guangxi, and Guizhou, which are four poor states in the South and Southwest. 14 Provincial Growth Divergence The convergence discussion focused on regional differences. But regions are broad aggregates, comprising poor and rich areas. We can bring out some of this variation by taking provinces as unit of analysis. That way, the number of observations increases from 5 regions to 76 provinces, 19 out of which are located in the Northeast. Figure 17 modifies Figure 16 in two ways. It looks at growth convergence among provinces rather than regions, and refers to the period 1975 to 2003 rather than 1970 to 2004, reflecting data availability at the province level.6 Starting with the 1975 to 1986 period, Northeast provinces, marked in blue, are clustered to the left on the horizontal axis. Moving to the right, we come across Northern provinces (in red), then Southern provinces (in yellow) and finally Central provinces (in green) and Bangkok. This reflects the 1975 income distribution. Looking at each region separately, two features stand out. First, differences in 1975 income levels within the Northeast and within the North are much smaller than among the South and especially the Center. Second, provinces within the Northeast and the North show strong convergence, with less well-off provinces have higher per capita growth rates, while the picture for the South and Center is more blurred. At the national level, the period from 1975 to 1986 was marked by province convergence, as shown by the downward-sloping regression line. The convergence among Northeast provinces implies that they are stacked more closely to each other in 1986. In contrast to 1975 to 1986, the regression line is upward-sloping for 1986 to 2003, and there is little evidence for convergence either within regions or nationwide.7 Figure 17: Growth Convergence among Provinces, 1975 to 1986 and 1986 to 2003 30 9861 6 20- 15 75- C_RAYON 86 19 N_KAMPH BK_BANGK BK_PATHU 19 C_SARAB C_PHRA PPG 4 N_TAK N_PHICH N_CHRAI N_UTTAR NE_SINE_CHAIY SA NE_NAKHO N_UTHAIC_CHAIN C_CHACH C_RAYON S_CHUMP S_PHUKE PPG N_LAMPH CP S_PHANG C_CHONB 10 NE_UBONR NE_KHONK N_LAMPA 2 CP NE_MAHAS NE_SURIN S_PATTA C_SUPHA N_PHITS C_CHANT N_CHMAI NE_ROIET NE_NONGK BK_SAMUT of NE_LOEI N_PHETCN_NAKHO C_SINGB C_PRACH of e NE_UDONT N_SUKHO N_MAEHOC_ANGTH C_NAKHOC_RATCH C_NAKH S_NARAT S_SURAT C_PHETC S_SONGK BK_NONTH ATUN atR C_SAMSA NE_YASOT NE_NAKHO S_NAKHO NE_KALAS S_PHATTS_YALA C_PHACH S_TRANG C_KANCH S_RANON NE_KHONK C_TRAT S_KRABI C_LOPBU etaR C_SARAB C_CHACH C_CHONB C_SAMSA ht 0 N_LAM PH NE_SAKON C_SAMSO C_PHACH C_PHETC C_LOPBU C_RATCH wor N_NAN NE_BURIR N_PHRAE BK_BANGK BK_SAMUT Gla NE_MAHAS NE_KALAS N_NAKHO NE_YASOT NE_NAKHO NE_CHAIY -2 htworGla N_KAMPH S_PHUKE 5 NE_NAKHO N_PHICH S_PAT TAC_ANGTH C_SUPHA S_NAKHO S_SURAT S_KRABI N_PHIT SS_TRANG C_NAKHO C_SINGBS_SONGK C_PRACH C_NAKH NE_ROIET NE_UBONR S_PHATT NE_UDONT N_LAM PA C_SAMSO NE_SI SA NE_BURIRN_NAN NE_SAKON N_PHRAEC_CHANT NE_SURIN NE_LOEIS_NARAT N_PHETC N_UTHAIC_CHAIN N_SUKHOS_YALAC_TRAT N_UTT AR N_CHMAI S_CHUMBK_NONTH P S_SATUN BK_PATHU C_PHRA N_CHRAI S_PHANG nnuAlaeR -4 8.5 9.5 10.5 11.5 nunAlaeR N_TAK NE_NONGK N_MAEHO 0 C_KANCH S_RANON 8.5 9.5 10.5 11.5 Log of 1975 Real Per Capita GPP Log of 1986 Real Per Capita GPP Fitted values Center Northeast Fitted values Center Northeast North South Bangkok North South Bangkok 6Southichack, (1998) uses province-level data to study regional convergence in Thailand between 1975-1995 period. He finds evidence for inter-provincial conditional divergence. The level of human capital, measured in a number of years of formal education, had a positive and significant impact on regional growth rates, while physical infrastructure, defined as per capita expenditure on public infrastructure, was insignificant. Provinces with a higher share of agriculture in GPP grew less fast. 7During 1975 to 1986, provinces converge at a rate of 0.39 percent per year. During 1986 to 2003, provinces diverge at a rate of 0.85 percent. 15 Structural Change - the Long Haul Economic growth typically brings about structural change in the sectoral compositions of output. One of the stylized facts of development, postulated as far back as 1939 (Fisher 1939 and Clark 1940), is that it comes with shifts in output from the primary (agriculture) to the secondary (manufacturing, mining, and construction) and the tertiary sectors (services). On the basis of comparative advantage, Thailand's leading sectors should be agriculture and related processing industries. Yet, during the 1960s and 1970s, protection of capital- intensive manufactures and export taxes on rice and other commodities suppressed the size of the agricultural sector. While these policies were abandoned during the 1980s, economic development brought about a shift away from agriculture. This sector fell as a share of GDP from one quarter in 1970 to less than 10 percent in 2004, whereas industry increased from less than a quarter to 46 percent, and services increased from just over 50 percent to 45 percent as of today. Agriculture underwent a larger contraction in the Northeast than in other regions, although from a higher level. The share dropped from close to two fifths to just under one fifth of GDP (Figure 18). Industry increased only from the early 1990s onwards, rising from around 15 percent to around one fifth. The service sector expanded by more than 15 percent since 1970 and accounted in 2004 for over three fifths of GDP. While Northeast development was focused more on services than on industry, the North expanded more the industrial than the service sector. The North had the same three-sector breakdown of value added in 1970 as the Northeast. By 2004, agriculture had declined by 20 percent, which matches the contraction in the Northeast. Yet, while the share of the Northeast industry increased just 4 percent, the share of the North industry increased by three times as much. Furthermore, the growth acceleration of the North since the mid-1980s coincided with a faster pace of industrialization. The South grew faster from a higher income level than both the North and the Northeast. While the sectoral trends look similar between the South and the Northeast up to the mid-1980s, the South has managed to reverse the decline in the agricultural share since then. The sector contributed in 2004 more than one third to regional value added, the highest share of any region. The Center stands for manufacturing. Industry contributed more than two third of regional GDP in 2004, compared to less than one third in 1970. The industrialization accelerated since the mid-1980s. Finally, Bangkok, at this aggregated level, experienced the smallest change. Services accounted in 2004 three quarter of value added, as they did in 1970. The residual is contributed by industry, which has declined since the mid- 1980s. Thailand's regions took different development routes. The Northeast stands out for a growing service sector; the North for a growing industrial sector; the South for the resilience of agriculture; the Center represents the classic case of economic growth through industrialization; and Bangkok generates income foremost through services. The most important changes since the mid-1980s are the accelerated pace of industrialization outside of Bangkok and the revival of agriculture in the South. 16 Figure 18: Regional GDP Composition, 1970 to 2004 Bangkok Center 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Agriculture Industry Services Agriculture Industry Services North South 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Agriculture Industry Services Agriculture Industry Services Northeast 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Agriculture Industry Services 17 Structural Change - the Short Haul The last section looked at long-term changes in sectoral composition of value added. But the growth dynamics in the next few years will reflect foremost recent structural changes. How did the Asian crisis affect GDP composition? Which sectors are driving the recovery in the last years? The 1997/98 Asian crisis hit both the Thai and Northeast economies hard, but within two years the recovery was under way helped by supportive macroeconomic policies and increases in external demand. In the last decade, real per capita growth rates of the Northeast tracked Thai growth rates closely (Figure 13.C). This includes the periods of negative growth during the Asian crisis and the recovery over the last five years. However, the growth of the Northeast economy is more volatile and occasionally derailed by weather- related shocks on agriculture. Most recently, the Northeast fell behind Thai growth in 2000 and 2004 by 3 and 5 percent, respectively. The Asian crisis led to important changes in the Northeast. Construction, finance, and retail trade contracted as a share of regional GDP by 6 percent, 3 percent and 2 percent, respectively (Figure 19). The same holds in other regions: the shares of construction, finance and retail fell across the country. The differences emerge in what sectors led the recovery. In the Northeast, the upturn focused on manufacturing and transport and communication, which expanded by 4 percent and 2 percent. In the North and the Center, it centered almost entirely on manufacturing (rising by 7 percent and 6 percent, respectively); in the South on agriculture (rising by 4 percent); and in Bangkok on transport and communication and on other services (rising by 5 percent and 3 percent, respectively). The largest sectors of the Northeast economy in 2004 were retail (26 percent), agriculture (19 percent), manufacturing (15 percent), transport (8 percent) and education (7 percent). This is similar to the North apart from a shift from retail to manufacturing, both of which account for 20 percent. The South is more centered on crops and forestry and fishing (26 percent and 9 percent, respectively), at the expense of retail (15 percent) and manufacturing (14 percent), while transport is of similar importance (7 percent). The Center stands out for its high concentration in one subsector. Manufacturing alone accounts for 62 percent of the value added, followed by retail with only 8 percent. Finally, Bangkok's manufacturing sector adds only 23 percent, while transport and communication contribute almost as much (21 percent), followed by retail (16 percent) and finance (8 percent). These regional differences point to an important development challenge for the Northeast. In contrast to other regions, the Northeast lacks a key driver for economic change. The North and the Center are ahead of the Northeast in terms of manufacturing, the South in terms of agriculture, and Bangkok in terms of transport, communication and other services. The Northeast is dominated by the retail sector, which lacks strategic or cross-regional importance and is largely supported through household remittances from other regions. This pattern is a sign of lack of specialization in and deepening of alternative sectors. 18 Figure 19: Regional GDP Composition, 1996 to 2004 Bangkok Center 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 1996 1997 1998 1999 2000 2001 2002 2003 2004 1996 1997 1998 1999 2000 2001 2002 2003 2004 South North 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 1996 1997 1998 1999 2000 2001 2002 2003 2004 1996 1997 1998 1999 2000 2001 2002 2003 2004 Northeast 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1996 1997 1998 1999 2000 2001 2002 2003 2004 Agriculture, Hunting and Forestry Fishing Mining and Quarrying Manufacturing Electricity, Gas and Water Supply Construction Wholesale and Retail Trade; Repair of Motor Vehicles, Motorcycles and HH Goods Hotels and Restaurants Transport, Storage and Communications Financial Intermediation Real Estate, Renting and Business Activities Public Administration and Defence; Compulsory Social Security Education Community, Social and Other Services 19 Poverty Incidence and Numbers With value-added per person growing four-fold in Thailand and three-fold in the Northeast, we would expect dramatic improvements in household living standards. Using the series of cross-sectional Socio-Economic Surveys (SES), we can trace poverty from 1988 to 2002 at the provincial, regional, and national levels. The national poverty headcount, defined as the share of people living in households with income below the poverty line, fell from 32.6 percent in 1988 to 9.8 percent in 2002 (Figure 20.A). In spite of population growth and Asian crisis, the number of poor dropped from 17.7 million to 6.2 million over this period. Thailand has already reached its MDG poverty target of halving the poverty headcount between 1990 and 2015. In addition, the Royal Thai Government (RTG)'s 9th National Economic and Social Development Plan target of poverty incidence under 12 percent has been met. Both targets were achieved ahead of time. Poverty reduction was not limited to Bangkok and surrounding areas but extended to all regions in the country. If we take reductions in percentage points, progress in the Northeast was fastest. Between 1988 and 2002, the poverty headcount fell 31 percentage points in the Northeast, compared to 24 percentage points in the South, 22 percentage points in the North, 18 percentage points in Center and 3 percentage points in Bangkok. However, poverty reduction becomes more difficult the lower the level of poverty, so percentage changes relative to the initial level is a more accurate performance indicator. Compared to the 1988 levels, the proportional reduction in poverty was largest in Bangkok, followed by the Center, South, and North, and slowest in the Northeast. With poverty falling faster in other regions, poverty becomes more and more concentrated in the Northeast. One in two poor persons lived in the Northeast in 1988, compared to one in three of the total population (Figure 20.B). The Northeast still accounted for roughly one third of the total population in 2002, but the share of poor had increased to 60 percent. This translates into 3.8 million poor living in the Northeast, and 2.3 million in the rest of the country. 20 Figure 20: Regional Poverty Trends, 1988 to 2002 A. Poverty Headcount 50 45 40 35 30 25 20 15 10 5 0 1988 1990 1992 1994 1996 1998 2000 2002 Bangkok Central North Northeast South Thailand B. Number of Poor 20,000,000 15,000,000 10,000,000 5,000,000 0 1988 1990 1992 1994 1996 1998 2000 2002 Northeast North Central South Bangkok 21 Self-assessment Poverty means different things to different people. The poverty statistics shown in the previous section are based on objective measures of household income and the consumption basket. The advantage of this approach is comparability across time and space. But the concept has also important weaknesses. The most important is perhaps that it ignores the people's own perception on poverty. While statistics may tell us that economic growth has made the population better-off, the people may actually feel worse due to the many changes that come along with economic development. One important source on community well- being is the NRD2C data set of the Ministry of Interior. It collects information on a range of economic, social, institutional and happiness indicators. Since the evaluations are provided by village committees, it can be thought of as a hybrid between objective and subjective data. NRD2C confirms the concentration of poor villages in the Northeast. Out of the almost 65,000 villages in 2001, almost 25 percent are esteemed to be poor according to the information provided by the Community Development Department of the Ministry of Interior. Two thirds of these villages are located in the Northeast. A 2001 conference, organized by the Community Organization Development Institute and the Thailand Development Research Institute, provided an alternative evaluation of regional poverty. Representatives of the poor assessed the shares of poor, middle-income and rich populations in 1991 and in 2001 (Jitsuchon 2001). The Northeast is estimated to have 70 percent poor in both 1991 and 2001. This matches with SES in that it is the highest number across regions, but the share of poor is much higher. This could reflect that the subjective notion of poverty at the community level is more encompassing than just economic well-being. The assessment provided also causes of poverty, which included lack of land, jobs, and credit access, degraded natural resources, poor health, old age, bad luck, and bad behavior (gambling, greed, laziness, and alcohol). Even more interesting were the responses on the issues not linked to poverty, such as lack of food, access to utility and education; insecure tenure; and indebtedness. Figure 21: Self-Assessed Poverty of Communities in 1991 and 2001 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1991 2001 1991 2001 1991 2001 1991 2001 Northeast | North | Center | South Poor Middle Rich 22 Spatial View - Provinces Provincial variations in living standards are even more pronounced than regional differences. Provinces with low poverty exist alongside provinces with high poverty. While the Northeast and the South include the very poorest provinces, these regions also comprise provinces in the lowest poverty bracket where the poverty incidence is less than 7.5 percent (Figure 22.A). Provinces with high poverty headcounts also tend to have large populations and hence a large number of poor people. A high concentration of poor people is found in the Northeast and the very South, the provinces with the highest incidence of poverty. For example, the five provinces who contribute most to poverty in Thailand are all located in the Northeast and have poverty headcounts more than twice the national average. They alone account for three-tenths of all the poor in Thailand. Similarly, out of the poorest 12 provinces that accounted for close to three-fifths of all the poor in Thailand in 2002, 11 come from the Northeast (Figure 22.B). While the Northeast includes provinces with little poverty such as Ubon Ratchathani, Nong Khai or Chaiyaphum, it also comprises the poorest provinces with the largest number of the poor, such as Surin, Sisaket and Buriram. This is an importance difference with the North, the second poorest region in Thailand. In the North, the provinces with the highest poverty incidence, such as Tak, Uthai Thani or Mae Hong Son, tend to be remote and sparsely populated. Hence, they contribute only moderately to national poverty due to low population density. In comparison to other regions in East Asia, the overlap of high poverty incidence with large number of poor, as in the Northeast, is rather unusual (Figure 22.C and Figure 22.D). For example, areas with high poverty incidence and low population density include the western provinces of China (Xinjiang and Tibet), the Northern Mountains areas of Vietnam, the upland areas of Laos and the eastern provinces of Indonesia and PNG. Low-incidence and high-density areas include the Mekong River and Red River delta areas in Vietnam, Vientiane plain and Mekong River Corridor in Lao PDR, and the Luzon island in the Philippines. Only a few areas are characterized by high poverty incidence and a large number of poor. Poverty incidence and number of poor overlap in the eastern provinces of the Philippines, on Java Island in Indonesia, and in the Yunnan province of China. Poverty transcends national borders. Localities with high poverty on one side of the border tend to have high poverty also on the other side of the border. The area with the most significant cross-border spillovers of poverty incidence is the Greater Mekong sub-region. This suggests an important role of geography in determining poverty, which goes beyond the influence of national history, policies and institutions. 23 Figure 22: Poverty Maps, 1988 to 1994 and 2002 A. Poverty Headcount and Number of Poor People 1988 B. Contribution to Total Poverty by Province (Northeast Provinces in red), 2002 Contribution to total poverty 2002 (NE provinces shown in red) 10.0 100 9.0 90 ) 8.0 80 %( ) cen 7.0 70 %( viorp oni 6.0 60 butir eachfo 5.0 50 ontc noitubirtnoC 4.0 40 evita ul 3.0 30 umC 2.0 20 1.0 10 - - NakhonRatchsimaSiSaKetBU inmani inaosm onhoo un an anw i i i ri n atok bupghS rionao aThaBuri ngani TrnogNThgP atnggaanihet uri uri ngom amn ratu ket Baatc itsaSong ngha kokBurnulokKhlaPacmh i soonamPa mpan thonKaemartta n ajni g laai agrkaireuo nei a nan NKkah aneawm ng uri ao hitng ditoh Sri T urRah asivn p l akh raeuanni aBbuiraka Nay araumn Sakyut honthayaaBuriaburi Raanumen nggTPa gkhS Bhoth Phuho -nBuri SuSam phaKutP nr udon PhatthaeabHoKh rgiriK hhapth NakhNoSn KanPkonNLu arathha m Yngam n L abaiThSa h BieutriphumdahanKhahothRayLoopBPhayPhicaluUttanaS agbua YaKhiTh LaChia ChiaTadChaPhetcUthkhon Saara hanRaoaiyaMukNongSuk R Ph hh non on C Na Kanc Ch Nakh CohnaiN SCCoemut.S.AChanthNonth raTtretch SuPh PhathphaPrachinBSiAh tS Ka Pm Prac Mhu No NakhonS Amn MahaS UbonRaLt Cha SaP.N Nak Samu Contribution to total income poverty Cumulative contribution to income poverty 24 C. Provincial Poverty Headcount Map of East Asia and Pacific, PPP$2/day, 2002 D. Provincial Number-of-Poor Map of East Asia and Pacific, PPP$2/day, 2002 25 Spatial View - Tambons Even within provinces, there are large differences between poor and non-poor communities. Combining household survey data with census data, we can obtain poverty estimates at the tambon (sub-district) level across the whole of Thailand. While there are other data sources for local information on living standards such as NRD2C, this approach has one key advantage: it gives the same poverty measure at the district level as is used at the national level. This consistency makes this method appealing to policymakers. It provides confidence that we are measuring the same thing, whether we are talking about districts, provinces, regions, or the nation. The 2000 poverty mapping results show that deprivation is concentrated in a relatively small number of tambons (Healy and Jitsuchon 2002). For example, the poorest third (34 percent) of all tambons, and the poorest sixth (16 percent) of all villages and urban blocks, accounted for more than two thirds (70 percent) of all poor in Thailand (Figure 23). Over two thirds (71 percent) of the tambons, and over half of all villages and urban blocks (53 percent), in the Northeast had a poverty incidence at least 50 percent in excess of the national average. This analysis shows that poverty in Thailand varies among regions, cities, and villages. There are large differences in the abilities of families to cover basic needs not just among regions, say the Northeast relative to the Center, and provinces, like Nongbua Lumphu compared to Ubon Ratchathani, but also within provinces and within districts. Why is this insight important? Thailand is aspiring to eradicate poverty nationwide. The poverty map can aid to reach this objective. It helps to make visible those poor who are otherwise hidden behind the averages of large regional aggregations. First, the knowledge of poverty incidence at a detailed spatial scale can improve the geographical targeting of interventions to improve people's lives. Policymakers can draw on this information when planning public investments in education, health, sanitation, water, transport, and other sectors. Further, poverty maps can be combined with other available geographically disaggregated data ­ e.g. geographic databases of transport infrastructure, public service centers, access to input and output markets, information on natural resources quality and natural disasters ­ to yield a rich array of information relevant for poverty analysis and policy making. Finally, it can assist communities in the development of local poverty reduction strategies. It provides local stakeholders with the facts that are required for local decision making and for negotiation with government agencies. Poverty maps thus become an important instrument for local empowerment. 26 Figure 23: Tambon-Level Poverty Map of Thailand, 2000 27 Eradicating Poverty The RTG has embraced the objective of eradicating mass poverty by the end of this decade. In order to reach this goal, the RTG has adopted a number of grassroot policies, such as Village Fund, People's Bank, Asset capitalization and 30 baht health care scheme. The extent of the RTG's poverty effort can be gauged by the volume of resources spent on all anti-poverty programs. Total expenditure on all anti-poverty programs was approximately Bt35 billion in FY1999, which constituted 4.2 percent of total public expenditure and 0.74 percent of GDP. It increased substantially to 10.4 percent of public expenditure and 2.3 percent of GDP in FY2002. However, many of these programs have low levels of coverage and high leakages of benefits to the non-poor as they cover large populations. Improved targeting by defining better criteria for allocating resources will be essential to reduce the population of poor people. For example, the village fund is a revolving fund of Bt1 million (about US$23,000), distributed nationwide to the about 70,000 villages over a three-year horizon launched in 2001. A key characteristic of this program is that it covers every single village in the country, regardless of whether the village is poor or non-poor. In addition, the bulk of the beneficiaries of the program are non-poor households. For the same resources, the poverty impact of the village fund could be increased by allocating more funds to poor villages or providing loans at more favorable terms to low-income households. One way to illustrate the implication of lack of targeting is to investigate the reduction in the poverty gap under the assumption of perfect targeting.8 The poverty gap measures the average consumption shortfall of the poor relative to the poverty line. Considering the rural population only, it equaled 3.1 percent in 2002. Assuming that the transfer is both perfectly targeted and fully consumed, the sum of all poverty gaps across rural individuals is the minimum income transfer needed to bring all rural poor just up to the poverty line. Under these assumptions, an income transfer of Bt25.2 (= 0.031 x rural poverty line of Bt813) per person per month would be required to eliminate poverty. The total annual volume of income transfers for rural poverty eradication would then be Bt13.1 billions (= Bt25.2 x 12 months x 43,300,000 persons). This is equivalent to no more than 1.3 percent of central government spending in fiscal year 2002, or just one sixth of the estimated budget spent on the Village Fund Program. By contrast, maintaining perfect targeting within regions but assuming that this amount is allocated according to current public expenditure patterns, the Northeast would receive only 30 percent rather than over 60 percent (its share in national rural poverty), and Northeast poverty would fall by only half. If the sum was evenly spread across the rural population (as in the Village Fund Program), the Northeast would obtain 40 percent, and Northeast poverty would fall by under two thirds rather than being eliminated. 8Perfect targeting implies that each individual below the poverty line would receive a transfer equal to the shortfall of consumption below the poverty line. Assuming that all this income transfer is consumed, all previously poor individuals would then have a consumption level just equal to the poverty line. No individual above the poverty line would receive any transfer. The following numbers are hypothetical and few developing countries would choose to continue making income transfers to the poor in perpetuity. Perfect targeting is impossible in practice, not all income is consumed, and transfers based on the shortfall of consumption (or income) to the poverty line to the poor have significant disincentive effects. 28 Poverty Registration Another important grass-root policy is the poverty registration initiative. The RTG launched a nationwide poverty registration program lasting from January 5, 2004 until March 31, 2004. Each poor person was invited to register herself at the District (Amphoe) branch of the Ministry of Interior and to fill out a form stating the major reasons of poverty. The District office passes on the roster of the poor to the village committee for assessment of their validity. The revised rosters are then used by the RTG for assistance in alleviating the identified problems. It is too early to evaluate the success of this program, as the government is still in the process of providing help to registered poor. Most importantly, we cannot tell whether the aided families escape poverty at least in the short term, let alone the long term. Nevertheless, information on the number of registered poor is already available. The evidence at the national level suggests that, unless the village committee screens carefully, there may be a large leakage to non-poor. Already after two thirds of the registration period, the number of persons declaring themselves as poor exceeded the number of income-poor according to the national poverty line. In the Northeast, the share of registered poor and poverty headcount according to the 2002 SES coincides remarkably close at the regional level (Figure 24). However, these measures differ greatly at the province level. In particular, the share of registered poor tends to be too high in provinces with low poverty headcounts, such as Ubon Ratchathani. In general, the share of registered poor varies much less from one province to another than the poverty headcount ­ the respective standard deviations are 3.3 compared to 11.2. Furthermore, while direct and practical assistance is clearly useful, families are poor often due to a multitude of inter-related factors. In many cases, there is no `silver bullet' for solving poverty and one-off assistance will not lift households permanently out of poverty. This suggests that incorporating geographical targeting into the design of a range of anti-poverty program could greatly aid its effectiveness. Figure 24: Share of Population Registered as Poor and 2002 Poverty Headcount in Northeast Provinces (%) 45 40 35 30 25 20 15 10 5 0 m a ai ni m t i am n i Ka tc lasinhas im phu Phu Et en Kh sothontchat ha BurirumCh areon Su hano rin Ke Ka Loe ya Lam Roi Nakho t P Chai Bua Nong Ya Ra Mukdahan SiSa onThanSarakh Na hon Ud on Ra Sak Khon hon Nong Ubon Am Nak Maha Nak Share of Registered Poor 2002 Poverty Headcount 29 Shared Growth ­ Regions What accounts for the sharp improvement in poverty at the national, regional, and province levels? Many factors could play a role, including government policies, the international economic environment, or the ingenuity of the Thais in bettering their living standards. In view of Thailand's growth performance, the primary candidate has to be that poverty reduction is related to income growth. While international evidence supports this hypothesis (Chen and Ravallion 2004, Ravallion 2001 and Field 2001), it is by no means commonly accepted. High inequality in Thailand leads some to posit that growth is essentially irrelevant to the problem of poverty, which should be regarded solely as an issue of "cutting the pie" rather than "the size of the pie". Even allowing that Thailand's growth has been unequally shared across regions, there is no doubt that economic growth led to poverty reduction. Figure 25.A shows that regions with higher growth have more poverty reduction. Between 1988 and 2002, the economy in the Center grew in per capita terms annually by 5.3 percent and poverty declined annually by 11.8 percent. The Northeast grew only by 3.4 percent and poverty decline only by 6.9 percent. Combining GDP growth rates and poverty reduction rates leads to the concept of the poverty elasticity of growth. It gives the percentage change in poverty for a one percent increase in real per capita GDP. For every one percent growth of regional GDP between 1988 and 2002, poverty fell by 2 percent in the Northeast, 2.2 percent in the Center, 2.6 percent in the South, 3 percent in the North, and 3.7 percent in Bangkok. This seems to suggests that growth was least effective in reducing poverty, or least "pro-poor", in the Northeast. However, the issue is not that clear-cut. First, the elasticity depends on the poverty measure. Figure 25.B shows the poverty elasticities for the poverty gap, which equals the average income shortfall of the poor relative to the poverty line. The greater this distance, the higher is the poverty gap. For this poverty indicator, the Center's elasticities is lowest at 2.5, compared to 2.6 for the Northeast, 3.2 for the South, 3.3 for the North and 3.7 for Bangkok.9 Second, the elasticity tends to be higher the greater the initial income (Chen and Ravallion 2004, Heltberg 2002). A reduction in poverty of one percentage point represents a smaller proportional change if initial poverty is 48 percent, as for the Northeast in 1988, rather than 4 percent, as for Bangkok in 1988. Finally, a given rise in income reduces poverty more the lower initial inequality. Since inequality in the Northeast is higher than in the South, Center, or Bangkok, economic growth would have resulted in less poverty reduction (Figure 25.C). In this context, the performance of the North stands out. In spite of lower initial income levels and higher inequality, the North achieved greater poverty reduction per one percentage growth than the South and the Center. Overall, the Northeast's poverty elasticity can be accounted for by its high initial poverty and inequality, although its performance looks weak relative to the North and, to a lesser extent, the South. 9The poverty gap overcomes one criticism of the poverty headcount, namely that it is indifferent to the distribution amongst the poor. For example, the poverty headcount could fall if resources were reallocated from the very poor to the poor. For the PPP$ per day poverty line, the elasticity of growth for the poverty gap between 1981 to 2001 was -3.3 for East Asia, -3.7 for Eastern Europe and Central Asia, -1.8 for Latin America, -4.3 for the Middle East and North Africa (Chen and Ravallion 2004). 30 Figure 25: Poverty and Growth, 1988 to 2002 A. Rate of decline in poverty and rate of real per capita growth in regional GDP 15 Bangkok t oun dcaeH Center 12 ytre ovP in South e 9 linceDfo North e atR Northeast 6 2.5 3.0 3.5 4.0 4.5 5.0 5.5 Real Growth Per Capita in GDP B. Elasticity of Growth: Percentage decline in poverty headcount and poverty gap for one percent increase in real per capita growth of regional GDP Northeast Center South North Bangkok 0 -1 -2 -3 -4 Elasticity of Growth: Poverty Headcount Elasticity of Growth: Poverty Gap C. Regional GDP Per Capita and Regional Inequality in 1988 90,000 0.480 80,000 0.460 70,000 0.440 60,000 50,000 0.420 40,000 0.400 30,000 0.380 20,000 0.360 10,000 0 0.340 Northeast Center South North Bangkok Regional GDP Per Capita (1988 Prices) Regional Inequality (Gini Index) 31 Shared Growth ­ Provinces We just found that regions with higher growth had larger declines in poverty. Figure 26 looks at the same issue, this time for provinces. This analysis will tell us whether the large positive impact of economic growth on poverty reduction, which is evident at the regional level, is replicated across most provinces. The scatter plots on the right show how the rate of decline in the poverty headcount relates to real per capita GDP growth, separately for pre- crisis years (1988 to 1996) and post-crisis years (1996 to 2002). The contrast between these two periods is stark. Pre-crisis, more growth is associated with more poverty reduction. All 73 provinces had positive growth, and only four had higher poverty, and in three of these four provinces poverty was no higher than 5 percent in 1988. Overall, a higher rate of growth led to more poverty reduction, as indicated by the upward-sloping fitted line. By 2002, 25 provinces increased income above 1996 levels, and 41 provinces reduced poverty below 1996 levels. More growth was not related to more poverty reduction. The panel on the left combines rates of growth and poverty reduction for the whole period from 1988 to 2002.10 Clearly, some provinces have done much better at translating growth into poverty reduction also within the Northeast. One average, the Northeast has done worse than the North and the South. Figure 26: Growth and Poverty in Provinces, 1988 to 2002 96 BKKCEN NORTH 19 60 88- 50 C_ANGTH C_SAMSA 19t 40 C_RAYON N_LAMPH BK_SAMUT BK_PATHU C_TRAT N_UTHAI 30 BK_BANGK C_SAMSO C_PHACH N_LAMPA uno C_NAKHO S_SURATNE_KHONK C_RAYON C_PHRA N_LAMPH C_PHRA 20 N_PHICHN_NAN C_CHAINNE_CHAIY N_SUKHOS_TRANG C_SINGBC_NAKH C_PHETC S_SONGKS_SATUN BK_NONTHC_SARABC_PHACH C_RATCH N_UTTAR S_PHANGN_NAKHO C_CHACHNE_MUKDA NE_LOEIC_PRACHC_CHANT N_UTHAINE_ROIET C_KANCH N_PHETC N_CHRAI N_KAMPHNE_SURIN N_UTTAR S_PHATTN_LAMPA 10 NE_SIN_CHMAI N_PHRAEC_RATCHNE_NAKHO S_KRABI NE_YASOT NE_MAHAS NE_NAKHP C_LOPBU N_PHITS S_NAKHOC_SAMSA C_CHONB C_SARAB N_PHICH N_TAK N_MAEHONE_BURIR NE_NONGK S_NARAT N_PHAYA C_SUPHA NE_KALAS S_CHUMPNE_UDONT SANE_UBONR C_PHETC NE_SAKON S_YALA C_CHACH N_PHAYA 0 S_RANON S_PHUKE S_PATTA C_LOPBU N_CHMAI adceHytre 0 C_TRAT C_PRACH N_NAKHO BK_SAMUT -10 C_NAKHO N_NAN ovPni C_ANGTH N_PHITS -20 C_SUPHA neil BK_BANGK N_SUKHO -30 C_SINGB N_PHRAE C_NAKH ecD -40 C_CHANT N_KAMPH C_CHAIN N_PHETC of -50 BK_NONTH N_TAK -6 C_CHONB eatR C_SAMSO N_CHRAI -10 -5 0 5 10 15 20 BK_PATHU N_MAEHO Real Growth of PC GDP 1988-1996 Fitted values Center Northeast NORTHEAST SOUTH North South Bangkok NE_UDONT S_PATTA NE_NAKHO NE_KHONK S_YALA 60 NE_NAKHP S_PHUKE 0022- C_CHONB NE_KALAS 50 NE_SURIN S_NAKHO 9961 40 NE_UBONR S_KRABI nt 30 N_KAMPH NE_BURIR C_SUPHA S_NARAT ouc S_PHANGS_PHUKE NE_YASOT 20 S_RANON C_CHANTS_TRANG NE_NONGK C_SINGB NE_SAKON S_CHUMP 10 C_TRATN_MAEHO C_NAKH BK_NONTH S_SATUN NE_MUKDA S_CHUMP C_LOPBU N_CHRAI C_CHACH C_SAMSONE_MAHASC_RATCH NE_SAKON N_PHITS S_PATTA N_PHRAE NE_UBONR NE_MUKDA S_SONGK eadHyt 0 N_CHMAIN_NAKHO N_PHAYASANE_KALASC_PHETC S_NARAT S_YALNE_LOEI NE_SIURAT S_PHATTNE_YASOT C_SARAB S_NE_ROIET S_SONGK C_PRACH N_SUKHOC_CHAIN N_PHETC NE_BURIR C_KANCH NE_SURIN N_TAKS_KRABI S_NAKHO NE_SI SA er 0 N_NANNE_CHAIY NE_NAKHPN_PHICH C_PHRA N_LAMPH NE_MAHAS S_SURAT -10 NE_UDONT BK_PATHU N_UTTAR BK_BANGK N_LAMPA NE_NAKHOC_PHACH C_SAMSA NE_KHONK NE_CHAIY S_SATUN N_UTHAIC_NAKHO C_RAYON NE_ROIET ovPni -20 BK_SAMUT NE_LOEI S_PHATT neil -30 NE_NONGK S_TRANG ecD -40 of -50 0 5 10 15 0 5 10 15 Poverty Elasticity of Growth 1988-2002 eatR -6 C_ANGTH -10 -5 0 5 10 15 20 Graphs by Region Real Growth of PCGDP 1996-2002 Fitted values Center Northeast North South Bangkok 10Ranong in the South and Kanchana Buri in the Center, the only two provinces with negative growth rates over this period, are omitted from the picture. 32 Durable Ownership Rising living standards express themselves not just in higher income and consumption, but also in ownership of durable goods. They can improve the quality of life in multiple ways. Refrigerators and washing machines cut down the time required to get through daily household chores; television provides access to information; and motorcycles help to get around places. Most of these goods require electricity connection. The expansion of power networks lays the ground for making household work less strenuous and information access easier. Durable goods also have a sales value and are as such an indication of wealth, a form of savings. They provide insurance against economic uncertainty and a way of preparing for future expenses. Ownership of small durable goods varies in response to changes in income. Looking at trends in durable good ownership as a whole can help to confirm the findings on poverty. The distribution of durable goods ownership both across time and space is in line with expectations. Comparing regions, the Northeast comes out lowest for refrigerators, washing machines, and radios, reflecting the highest levels of income poverty. However, economic growth has helped to reduce the ownership gap to other regions considerably. For example, more than two thirds of the Northeast population lived in households with refrigerators in 2002, compared to only one seventh in 1988. Clearly, televisions are among the first items purchased once some spare income is available: almost all households owned a television in 2002, compared to only just over one third in 1988. Patterns of durable ownership that serve similar functions are also telling about income progression. Bicycle ownership went down over time, as households substituted from low to high quality durables with higher income. Households switch from bicycles to motorcycles or even to cars. Compared to other regions, the Northeast has the highest ownership of bicycles but the lowest ownership of cars, indicative of its income ranking. In spite of the increases, most Northeast households have little material possessions. More than 30 percent of the Northeast population has no refrigerator, more than 80 percent has no washing machine and about 95 percent has no car. This suggests that for most families little or nothing is left over from the daily labor after the most urgent food and consumption needs are covered. 33 Figure 27: Durable Goods Ownership (Percent), 1988 to 2002 A. Refrigerator and Washing Machine 100 90 80 70 60 50 40 30 20 10 0 Refrigerator Refrigerator Refrigerator Refrigerator Refrigerator Washing Washing Washing Washing Washing Bangkok Center North Northeast South Machine Machine Machine Machine Machine Bangkok Center North Northeast South 1988 1990 1992 1994 1996 1998 2000 2002 B. Radio and Television 100 90 80 70 60 50 40 30 20 10 0 Radio Bangkok Radio Radio Radio Radio Television Television Television Television Television Center North Northeast South Bangkok Central North Northeast South 1988 1990 1992 1994 1996 1998 2000 2002 C. Bicycles, Motorcycles, and Automobiles 100 90 80 70 60 50 40 30 20 10 0 Bicycle Bicycle Bicycle Bicycle Bicycle Motorcycle Motorcycle Motorcycle Motorcycle Motorcycle Automobile Automobile Automobile Automobile Automobile Bangkok Central North Northeast South Bangkok Central North Northeast South Bangkok Central North Northeast South 1988 1990 1992 1994 1996 1998 2000 2002 34 II. Constraints Cities Urbanization and Development Urbanization refers to the process of growth in the population share living in cities, towns, and sub-urban areas. It is a territorial response to a structural shift away from agricultural in the economy, associated with features like division of labor, advanced production technology, variety in goods and services traded and population density and diversity. Worldwide, urbanization is indicative of a country's per capita income level: the correlation coefficient is around 0.75 to 0.80. It is no wonder then that urbanization is often taken to be synonymous to economic development. By international standards, East Asia's urbanization levels are low. In 2000, 36 percent of the population lived in cities and towns, less than half of Latin America's level. Upgrading of infrastructure, spread of urban areas to envelope rural areas, and rural-urban migration are projected to increase urbanization levels (ADB, WB and JICA 2005). Over the period of 2000 to 2015, the population living in cities with more than 1 million residents is expected to increase by about half (to 500 million), and the population living in mega-cities of more than 10 million residents will rise by a similar proportion to 120 millions. Even by East Asian standards, Thailand's urbanization level as well as its urbanization rate (change in urbanization levels) are low (Figure 28.A). Thailand is less urbanized than Korea, Malaysia, the Philippines, Indonesia and China. Only Vietnam, Lao PDR and Cambodia, whose income levels are about one fourth to one seventh of Thailand's, are less urbanized. Since 1970, Thailand's urbanization rate is second lowest among these countries, higher only than Vietnam's. The urbanization rate slowed from the early 1980s onwards, coinciding with the policy shift from import-substitution to export orientation, and fell even behind Vietnam's. Thailand urbanization is low for its income level also from a global perspective. Compared to other countries, Thailand's urbanization was on par with its income level in 1963 (Figure 28.B). Yet, urbanization lacked income growth, and by 2003, Thailand's urbanization level of 33 percent was about 20 percent below compared to the average urbanization degree of countries of its income level. Even allowing for an under-recording of in-migrants to urban areas in the registration of the Ministry of Interior, lack of widespread urbanization remains a salient feature of Thailand's development path. Across the world, densely-populated urban areas have been a force behind development. They provide markets for outputs, inputs, labor and other services and allow firms to profit from economies of scale and scope, specialization and the rapid diffusion of knowledge and innovation. For example, the US experience over the last century shows that while major production centers move across sectors and regions over time, economic activity in manufacturing and service sectors has always concentrated in and around cities (Gordon et al 2003). The slow pace of urbanization is linked to lower fertility in urban areas, out- migration from Bangkok to surrounding areas (and outlying regions during the Asian crisis), as well as a cautious policy approach toward urbanization. However, arguably the most important factor is the metropolis Bangkok which dominates urban development in Thailand. 35 Figure 28: Urbanization Indicators A. Urbanization in East Asian Countries, 1970 to 2003 90 80 70 60 50 40 30 20 10 0 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Cambodia Lao PDR Vietnam Indonesia Philippines China Thailand Malaysia Korea B. Urbanization and Per Capita GNI, 1963 to 2003 Urbanization vs Log GNI Per Capita (Altas ), 1963-2003 seulav 1963 0 10 SGP BEL detiF/ HKG 80 URY MLT ISRBHS FRA ISLAUS SWEUSA 60 COLBLZ ESPTTO CHL ITAVEN AUT BRAPER IRQ ZAF GRC JPNIRL PRI NLDLUX NOR CHE % 40 EGY NIC PRYHLDZAGTM CRI JAM SYR MEXPAN BRB FIN MARGUYSYC TUNFJI SLV P 20 SOMTGOSDN IND ZAR CHN PRTLBY AFGCMR TH ALKAGHA COGHNDDOMMYS CIVPECU VCT ZMB Z WE GAB 0 RWAMWINPLBDIB HTICAFMRTNGA FAKEN OMNPNG NER BEMDG TCD BPORU 4 6 8 10 Log PC GNI seulav 1973 0 10 SGP BEL detiF/ HKG 80 URY ISR MLTCHL ARG ITABHS KWT GBRDEUAUS NZLISL VIRNOR USA LUXCAN DNKSWE AUT FRA 60 COL IRQBRA PER MEX TTO VENESPPRI SAU IRL GRCLB Y PYFFINNLD CHE JPN NCL % 40 SEN COG EGYMAR SLV DOMBLZ SURJAM ZAF BRB P 20 MLIHTIPAKBENMRT THAZAR CAF CIVGUY BOLTUNDZATFJICRIPAN ECUZMB PRYGTM NICSYRMY SYC SUR GAB SOMGNB CMR TGOLKAPHLVCT LBRHND PRT AFGK EN OMNPNG ZWE SWZ 0 RWANPLBFA B DI MWIBGDGMB IDNINDCHNSLENGAGHABWA LSOSDNLB TCDSMDG NER BPORU 4 6 8 10 Log PC GNI seulav 1983 0 ISL KWT 80 VEN HKGNZLGBR SGP BEL detiF/ 10 CHLURY ARGMLT ISR BHR AUS DEUDNKWE S ARE DMACOL PER TTO SAU 60 BRABGR ESP VIRFRACANUSA LUXNOR JOR MEX IRQ PRI ANTITA BHSLBYAUT PYFJPN NLD FIN CHE CRIBLZECU TUN DOM HUNSUR GRC NCL % 40 CA FGHA S ENZMB HND LBR BOLEGYSLVNIC CYP GABIRL SYC NGA CIVVCT PHL MAR COGJAMPRYYRPANA OMN IDNTON 20 THAZ BW CMR KIR MUSM BEN GTKNAFJI TURSMYSDZZAF IRNBRB ATGPRT P SOM TCD GNB INDCOMZAR WELCA NAM 0 ETHMWIBGD MLICHNTGOKEN SLB PNGVUTSWZ NPL BFA BDIRWA BTNNERB HT IGMPAKMRTGUY SLEMDG LKA S DNLSO BPORU 4 6 8 10 Log PC GNI seulav 1993 0 10 MAC SGP BEL HKG detiF/ VECHL N URY ARG BHRBHSR MLT NZISL GBR AUS DEU ISL DNK LUX 80 PRISAU SWE UKR BLRLTUE BGRPER COLLVA JORLBN BRA RUS ME XB GA ESPARE FRA CANUSA ARM TTO OMN CYP ITAAUTNORCHE 60 STCZEHUN JPN NIC GEO BOL GRC IRLNCLFIN NLD MRTMDACOG SYR MAR DO ROMKAZTUN POLDMAUR CRI PYF SENCMRCPV PHL ECUMKPRY MHLSVKPANZAF HRVTMYS BLZBWA SYC BRB PRT % 40 CIV GTM DDZAVCT MUS SLE NGA ALBSTPCAF UZB TKM BEN ZMBGUY GHA GNQ SLV JAM FJI COMKIR HNDEGY IDN KNA ATG P 20 ETHUGA NERMLI TZA GNBSDNGMB CHN THA VNM ZARMDGHTIAGOPAKGINZWEO KENTJKYEM INDTGOKGZ A VUT SWZTONNAMFSMGRD LCA 0 BDI NPLMWIFA TCD BLAO BGD RWA BTN LKLSSLB PNG BPORU 4 6 8 10 Log PC GNI seulav 2003 0 10 SGPHKG AUSDEUISL BEL detiF/ LUX BRA GAB VENCHL ARG URY LBN NZL CANSWE USA GBR 80 DJI BLRJOR DNK COLPER MEX CZE TTO KOR E SP NOR ARM UKR ITAFRA NLD AUT JPN CHE COG LBR GH MDA DZATUNJAMVCT IRN TURDMALVA EST 60 MRT MNG MKDLVMHL MYSLTUPOLHUNPLW GRC FIN IRL CMRAZE NICBOLPHLMAR KAZDOM ZAFBWACRISVK GEO SLE K HNDSYR PRY CPVECUBGRRUS YUGROM S FJI PANHRV SYCBRB SVN P RT % 40 GYALB BIHGTM GRD MUS ZAR GNB MOZCATZA ZMBSDN P MLIKGZUZB IND FTGOKENPAK STPHTIGIN COM AGOGUYCHN THA LCA KNA A TG 20 GMB NGA BENSENCIV IDNIR TKMETONNAMMDV VNMYEM SWZ ETH LSO LKA VUT WSM FSM BDI MWI TJK TCDBFA BGD ERIRWA MDG NERNPL LAO UGA KHM SLB 0 TMPPNG BTN BPORU 4 6 8 10 Log PC GNI 36 Primate City Thailand's urbanization is not just low but municipal areas are concentrated in and around one city: Bangkok. Thailand is dominated by the primate city of Bangkok. It is the political (capital, unitary government sector), economic (location for enterprise headquarters and export hub), financial (location for banks and stock markets), transport (focal point for the country's transport and communication infrastructure) and cultural (important temples) center. Bangkok stands near perfectly at the geographical midpoint of Thailand, almost equidistant from its northern and southern borders as well as centered between its eastern and western boundaries. The sheer size and activity of primacy cities becomes a strong pull factor, bringing additional residents into or close to the city and causing the primate city to become even larger and more disproportional to smaller cities in the country. While not all countries have a primate city, for those that do, the primate city dominates cultural and economic life in that country. Standard examples of primate cities are London and Paris in Europe, Santiago and Buenos Aires in Latin America. Examples of countries without primate cities are the United States, (largest city and financial center New York, capital Washington, D.C.); Brazil (Sao Paulo and Brasilia); and Australia (Sydney and Canberra). Measures for the primacy of a city involve comparisons of the size of the largest city in a country compared to the next largest cities in terms of population.11 Figure 29.A compares Thailand with other countries that are often given as examples for countries with primate cities for two indicators. Thailand stands out as the most extreme case of primacy. Cities such as Beijing, Jakarta, Kuala Lumpur, or Manila have all become global centers for international trade, communications, employment of migrants, and foreign direct investment, and experience unprecedented prosperity as a result of their advantages. But Bangkok dominates urban development in Thailand like perhaps no other city in other countries. In 2000, Bangkok had around 6.3 million people, which was about 17 times the number of inhabitants of Samut Prakan, the second largest city, where 380,000 people lived. In fact, Samut Prakan lies in the vicinity of Bangkok, as does the third largest city, Nonthanburi. About twenty years ago, Samut Prakan was only the 12th largest city, and Nonthanburi only the 25th largest city. Only Udon Thani, the fourth largest city in Thailand with 290,000 inhabitants, is far away from Bangkok's gravity at 564 kilometers distance. The rise of these urban agglomerations in Bangkok's neighborhood has led to a fall in the primacy index for Thailand, while at the same time widening the gap between the extended Bangkok area and the rest of Thailand (Figure 29.B). According to some estimates, the extended Bangkok area could include as many as 17 million people (Webster 2005). Thailand's changes in city rankings reflect a more general regional trend. The most rapid population growth in East Asia is taking place in peri-urban peripheries (Webster 2002). Neighboring cities are connecting with each other and form into larger urban clusters. These include large parts of China's coastal zone, the Philippines' National Capital Region, the cross-border cluster of Singapore-Riau-Johore, and Bangkok, Vicinity and Eastern Seaboard in Thailand. 11Mark Jefferson introduced the concept of the primate city in 1939. According to his definition, primacy is present when the largest city's population is several times larger than the population of the second largest city (Jefferson 1939). 37 Figure 29: Primacy Indices A. Thailand and Asian and Non-Asian Countries 20 15 10 5 0 Thailand ypt UK h KoreaIndo nesia laysia zil nam ina Ch India tina Bra xico Ma Viet Me Eg France PR Argen Sout Largest City to 2nd Largest City Largest City to 2nd to 4th Largest City B. Thailand, 1983 to 2000 30 25 20 15 10 5 0 1983 1990 2000 Largest City to 2nd Largest City Largest City to 2nd to 4th Largest City 38 Drivers, Spillovers and Congestion Bangkok is one of the world's most cosmopolitan cities, the center of an extended area covering one quarter of the Thai population and to more than half a million expatriates, which attracts millions of tourists every year. While the degree of Bangkok's primacy is unusual, the factors of primacy conform to experience elsewhere. Bangkok is the country's capital for a highly centralized government; has access to a major port; is a conduit for inter- regional traffic; and is located above most of Thailand's groundwater. Historically, Ayutthaya, Thailand's capital from 1350 to 1767, was unusual among South Asian economies for its strong role in international trading. The port of Ayutthaya was an entrepot, an international market place where goods from the Far East could be bought or bartered in exchange for merchandise from the Malay/Indonesian Archipelago, India, or Persia, as well as local wares or produce from Ayutthaya's vast hinterland. After the fall of Ayutthaya, the new kingdom that emerged first at Thonburi, at the western bank of the Chaophraya river, and later at Bangkok, at the eastern bank, continued to rely heavily on trading for its economic base. The nearby fertile areas of the Central Plain provided the rice for exports by the Bangkok government. Hence, the central government had little interest in developing and integrating with outlying provinces, and began strengthening its administrative hold on these regions only in the 1860s. While international trade helped Bangkok to move from traditional to more processed goods, other regions continued subsistence farming. Overall, drivers such as comparative advantage, economies of scales, transport facilities and centralized administration combined to build Bangkok's primacy into the structure of the Thai economy (Biggs et al 1990). Bangkok's primacy provides strong advantages for enterprises compared to other regions. They include easy access to export channels, lower transport costs, better utilities, higher labor productivity due to a skilled labor force, a more developed financial sector, close proximity to the public administration and policy makers, and, most importantly, strong powerful agglomeration economies through strong forward and backward linkages to input and output markets, which increase profitability for all firms. Nevertheless, not all firms locate in Bangkok. Immobile factors such as land and natural resources (processing makes them easier to transport), lower land and labor costs, and local markets for small enterprises, such as handicrafts or manufacture of farm tools, lead to spatial dispersion. While primate cities are vital economic growth poles for their countries, they can also be a structural hindrance for the development of lagging regions. They typically become the center for both economic and social services, reinforcing disparities between the primate city and other cities. Their strong pull factor can undermine the development of thick markets and institutional capacity in outlying regions (Henderson 1988 and Krugman 1995). Problems associated with primate cities are not only issues balanced regional development but also within the primate city itself. Urban growth produces challenges. Vibrant enterprises pose rising demands for business services that meet modern standards. For example, Japanese vehicle manufacturers operating in Thailand report that Bangkok traffic congestion has increased the level of stocks that they need to hold (JBIC 2004). In addition, the poor, living frequently in peri-urban, informal settlements, require basic infrastructure, sanitation and housing. While urban incomes are higher than rural incomes, the cost of living is more expensive and social capital often weaker. 39 Extended Bangkok Area and Beyond While low urbanization has not impeded economic growth in Thailand, it has contributed to the sharp contrast between the extended Bangkok area and the rest of the country. As we have already discussed, the most dynamic urban areas are located in close proximity to Bangkok (Figure 29.B). We have also already seen that Bangkok's economic structure is very different that the Center's (Figure 19), which includes localities such as the Ayutthaya and Eastern Seaboard industries areas. While Bangkok is dominated by the service sector, the Center is the home of much of Thailand's manufacturing base. Bangkok is a leading hub for Asian media and marketing services, high quality business and producer services (law firms, accounting firms, management consultants), high amenity and sophisticated cultural products (design and fashion, jewelry, cuisine, health and spa services), as well as over 60 international organizations and a large diplomatic community (Webster 2005). These activities flourish through the low cost of doing business, the cosmopolitan flair, and the, by international standards, lost cost of living. By contrast, firms in need of a large plant site are attracted to the Bangkok fringe. These localities share some of the agglomeration advantages of Bangkok, yet they avoid some of the disadvantages, especially the high cost of land. Within a predominantly rural country, the Northeast stands out as the most rural region. In 2002, only every sixth person lived in an urban or suburban area, compared to one in five in the North, one in four in the South, and one in three in the Center (Figure 30.A). And within the Northeast, there are again large differences. Two fifths of the Northeast's population live in the provinces of Nakhon Rathchasima, Ubon Ratchathani, Udon Thani and Khonkaen, which comprise the four major cities of the same names. Figure 30.B shows the ranking of the nine largest cities after Bangkok in 1983 for 1983, 1990 and 2000. With the sole exception of Udon Thani, all cities fell in their rankings, and only four (Udon Thani, Nakhon Rathchasima, Chiang Mai and Hat Yai) remain in the top ten. Udon Thani is a former US military base, is a major center of resident expatriates and located nearby an important archeological site as well as the Friendship Bridge to Lao PDR. Similarly, their economic performance has been disappointing. Only Khonkaen and, modestly, Nakhon Ratchasima increased between 1981 and 2003 their provincial-level per capita GDP compared to the national average, and none of the provinces had income levels in excess of the national mean (Figure 30.C).12 Unfortunately, the concerns about the sustainability of an innovative regional cities project launched in the second half of the 1980s have been confirmed (Box 3). 12To what extent are changes in income levels linked to changes in the size and density of cities? While it is generally accepted that geographical concentration of economic activity can raise productivity, there is no consensus about the magnitude of the effects, whether population size or density matters more, and whether there is a minimum threshold of city size or density before these effects are triggered. 40 Figure 30: Urbanization and Regional Cities A. Regional Urbanization, 1988 to 2002 1 0 0 9 0 8 0 7 0 6 0 5 0 4 0 3 0 2 0 1 0 0 B a n g k o k C e n t r a l N o r t h N o r t h e a s t S o u t h 1 9 8 8 1 9 9 0 1 9 9 2 1 9 9 4 1 9 9 6 1 9 9 8 2 0 0 0 2 0 0 2 B. Ranking by City Population of Largest Nine Cities outside Bangkok, 1983, 1990 and 2000 25 20 15 10 5 0 Nakhon Chiang Mai Khon Kaen Hat Yai South Ubon Nakhon Udon Thani Songkhla Phitsanulok Ratchasima North Northeast Ratchathani Sawan North Northeast South North Northeast Northeast 1983 1990 2000 C. Per Capita GDP of Selected Provinces relative to National Average, 1981 and 2003 130 120 110 100 90 80 70 60 50 40 30 20 10 0 Nakhon Chiang Mai Khon Kaen Hat Yai Ubon Nakhon Udon Thani Songkhla Ratchasima North Northeast South Ratchathani Sawan Northeast South Northeast Northeast North 1981 2003 41 Urbanization and Industrialization The lack of successful development of regional cities in outlying regions is one reason for the small manufacturing sector. But why is low industrialization related to low urbanization? Industrialization typically moves people away from immobile factors, such as land and natural resources, on which agriculture draws on. In addition, industrialization creates incentives for production and market activities to cluster. Manufacturing firms tend to locate close to large urban centers to minimize transport costs and realize scale economies. Industrialization increases the need for both physical infrastructure and human capital, which are factors that large cities are better at providing. In this way, urbanization leads to concentration at the same time (Puga 1998). In addition, in spite of high rents, small firms also cluster in large cities, where they benefit from diversified market niches, mobile labor supply, good infrastructure and services and proximity to input suppliers and large firms. This underscores the importance of supporting development of infrastructure and services in regional growth centers. The next section will take a closer look at the experience of the Thai manufacturing sector. Box 3: The Regional City Project In 1985, the Ministry of Interior, Department of Local Administration, and the World Bank signed a US$27.5 million loan for the Thailand: Regional Cities Development Project. Its objectives were to stimulate economic development by strengthening four municipalities as productive urban centers, increase their financial autonomy and help reduce migration to the central region. The locations were Khonkaen and Nakhon Ratchasima in the Northeast, Chiang Mai in the North, and Songkhla in the South. Specifically, the project components included improvements in infrastructure, capacity building at the local as well as central levels, in addition to establishing a fishport and industrial area in Songkhla. The project was closed in 1994 and rated satisfactory based on good progress on infrastructure, administration, and fishport operations. The project led to follow-up project including eight other cities using only Thai resources. However, the project's performance audit report noted that the sustainability of the project remains uncertain due to a number of reasons. First, the centralized government system is the main constraint to the development of regional cities. Majors as well as provincial governors require increased power for revenue collection and decision making. Second, the project preparation (6 years) and implementation (8 years) took a long time partly due to the highly centralized system. Devolving more project responsibilities to the local government should lessen the need for inter-agency coordination at the national level. The Thai Cabinet deliberated on matters concerning project preparation and implementation on no less than 28 occasions. Third, as the population and economic activity spread to outlying areas, infrastructure and development programs are demanded outside the municipal boundary. There is need for a mechanism of implementing projects jointly by the tambon administrations and municipalities. 42 Enterprises Manufacturing Value Added Manufacturing has grown in importance over the last 25 years. In spite of investment slump and Asian crisis, the manufacturing sector expanded continuously its share in GDP. The sector is now approaching two fifths of GDP, compared to one third of GDP before the Asian crisis and just under one fifths of GDP in the early 1980s (Figure 31.A). The expansion is closely linked to the boom in exports, which increased from around one fifth of GDP in the early 1980s to around 45 percent before the Asian Crisis and now contribute close to two thirds of GDP. Manufacturing accounted for 87 percent of all exports in 2004, compared to 80 percent in 1993, 45 percent in 1986 and just over 30 percent in the early 1980s.13 Within manufacturing exports, the share of high-tech products increased from 58 percent in 1993 to 78 percent in 2003 (Figure 31.B). While exports of textile and garments products have declined, exports of electrical machinery and parts, non-electrical machinery and parts, and vehicles and parts have surged. While the importance of manufacturing has grown, the role of Bangkok and Vicinity as Thailand's factory hub has declined. During the 1980s, their combined share was between 65 to 70 percent (Figure 31.C). By the time of the Asian crisis, it had fallen to just over 50 percent, and now stands at 46 percent. Most of the decline was due to Bangkok, which took the brunt of the adjustment triggered by the Asian crisis. Since the early 1980s, Bangkok's contribution to Thailand's manufacturing value added fell by 45 percent, and the Vicinity's contribution contracted by 15 percent. By the same token, the manufacturing share of the Central's 6 provinces almost tripled, and of the East's 8 provinces almost doubled. These two subregions contributed just under one fifth of manufacturing GDP in 1981, and now account for twice as much. The East's contribution has exceeded Bangkok's since 1996, and the Central's contribution has topped Bangkok's share since 2003. By contrast, the Northeast, North and South have not benefited from the expansion of the manufacturing sector outside of Bangkok and Vicinity. Both the North and the Northeast contribute only 4 percent to the sector's value added, which is the same fraction as in the early 1980s. The South did even worse, and its share contracted from 5 percent to 3 percent over this period. The trends for the Northeast and especially the North look more encouraging since the late 1980s, with an average annual GDP growth rate in manufacturing of 10 percent, about 2 above the national average. Between 1999 and 2004, the annual growth rate dropped to 6.4 percent in the North and 4.4 percent in the Northeast compared to a national average of 6.5 percent. In any case, whatever expansion has taken place over the last decade and a half occurred from very low levels of value added. The evidence from labor productivity shows the same pattern. While the East and Central have overtaken Vicinity and Bangkok in terms of value added per manufacturing worker, the North remains the least productive region, reaching no more than 60 percent of the output per worker in the North and 10 percent of the output per worker in the East, Vicinity and Central. 13This does not imply that manufacturing equals to 57 percent of GDP (close to ninety percent of two thirds of GDP) as manufacturing value takes into account the use of intermediary inputs. 43 Figure 31: Manufacturing GDP and Exports, 1991 to 2004 A. Manufacturing as Percent of GDP 40 35 30 25 20 15 10 5 0 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Bangkok Vicinity Central East West North Northeast South B. Manufacturing Exports by Product Type as C. Regional Breakdown of Manufacturing GDP Percent of Total Manufacturing Exports 100% 100% 90% 80% 80% 70% 60% 60% 50% 40% 40% 30% 20% 20% 10% 0% 0% 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Bangkok Vicinity Central East West North Northeast South Labor intensive products High-tech products Resource based products D. Labor Productivity in Manufacturing, 1991 to 2004 1.10 1.00 0.90 0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Bangkok Vicinity East Central West North Northeast South 44 Employment Dynamics Out of Thailand's 7400 tambons, some 2700 had manufacturing establishments in 1996/7. Yet, manufacturing jobs were available in four fifths of Bangkok's tambons and over half of Center's tambons, but only in just over one third of the North's and the South's tambons and only one fifth of the Northeast's tambons ( Figure 32.A). Five years later, the clustering of companies in the extended Bangkok area increased even further. Almost all of Bangkok's 154 tambons and more than three fifths of the Center's 1932 tambons offered manufacturing employment. Manufacturing employment also increased in the Northeast, where the share rose to just under one third, and the North, where it rose to two fifths. Only the South did not experience any increase. As manufacturing value added shifted from Bangkok and Vicinity to other areas, the geographical distribution of companies and their work force have changed as well. Combining the 1996/7 and 2001/2 manufacturing censuses, we get a detailed picture of the dynamics in the spatial spread of employment among business establishments from just prior to the Asian crisis to well into the recovery. Figure 32.B plots circles of employment for enterprises with 10 workers or more, where the size of the circles corresponds to employment levels.14 This data set covers just over half of all manufacturing employment. There is a strong concentration of employment in and around the Bangkok area. Other parts of the countries are dominated by gaps, representing areas without manufacturing employment. The employment numbers tell a similar story. Between 1996/7 and 2001/2, Bangkok's employment share declined from 28 percent to 23 percent and the Center's share increased from 56 percent to 60 percent ( Figure 32.C). The Northeast's employment share increased by 2 percent while the North's and South's contributions remained unchanged. Median employment of enterprises is highest in Bangkok and Center, and lowest in the Northeast, where it is no more than 17 workers among enterprises with at least 10 workers. 14The 1996/7 data is taken from the 1997 Industrial Census collected during 1996 and 1997. The 2001/2 data draws on the listing for the 2002 Business and Trade Census, which covered the entire manufacturing sector. The restriction of 10 workers or more is imposed for two reasons: the 1997 census did not cover all enterprises with less than 10 workers; and the 2004/5 Thailand Productivity and Investment Survey, which we will investigate later, also focused on companies with at least 10 workers. Each dot is randomly placed within a tambon (subdistrict). Red circles identify outliers where employment figures are greater than the 75th percentile by three times the difference between the 25th and 75th percentiles. 45 Figure 32: Spatial Distribution of Manufacturing Employment, 1996/7 and 2001/2 A. Share of Tambons with Enterprises (%) 100 90 80 70 60 50 40 30 20 10 0 1996/7 2002/3 Bangkok Central North Northeast South B. Cartogram Maps C. Employment Shares (%) and Median Number of Workers per Enterprise 100% 50 90% 80% 40 70% 60% 30 50% 40% 20 30% 20% 10 10% 0% 0 1996/7 2002/3 1996/7 2002/3 Bangkok Central North Northeast South Bangkok Central North Northeast South 46 Sector Composition Much of Thailand's employment is concentrated in a few sectors. Food products and beverage, wearing apparel, textile and furniture were the largest employers in 1996/7, accounting for over two fifth of all manufacturing jobs, and remained among the five most important sectors in 2001/2. The spatial pattern varies widely by industry. Figure 34 shows the employment maps for the eight industries covered in the 2004/5 Thailand Productivity and Investment Climate Survey (PICS), which includes Thailand's biggest employers. The concentration of employment in individual industries exceeds that of employment overall. The Northeast is most represented in wearing apparel, textiles, food processing and furniture, which account for about half of total employment. The two sectors that experienced the largest increases at the national level were electronic parts, which almost tripled its share from 2.9 percent to 8.2 percent, and wearing apparel, which rose from 7.5 percent to 9.7 percent. All regions benefited from their expansion, with the exception of the South in electronic parts. The employment increase of electronic parts came together with a higher GDP share, while wearing apparel employment rose in spite of negative GDP growth. Together with printing and starch products, these were also the sectors that saw the largest gains in the Northeast. The only large scale industry that experienced a decline between 1996/7 and 2001/2 was textile. The employment structure changed as a result of these adjustments at the industry level. Figure 33 shows a breakdown of the manufacturing sector by technological characteristics, as developed by the OECD.15 The Northeast and the South stand out as the regions with the lowest shares of differentiated and science-based employment. While the South is dominated by resource-intensive industries, the Northeast largest sectors are labor-intensive industries. Apart from Bangkok and Vicinity and the South, the contribution of these sectors increased between 1996/7 and 2001/2. Figure 33: Manufacturing Employment 1996/7 and 2001/2 by Technological Characteristics 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% North Central BKKVIC East Northeast South Thailand North Central BKKVIC East Northeast South Thailand Resource-intensive Labor-intensive Scale-intensive Differentiated Science-based Resource-intensive Labor-intensive Scale-intensive Differentiated Science-based 15This classification is applied in Lall (2000). Resource-based industries (such as food processing and wood products) draw on natural endowments, labor-intensive products require typically low technical skills; scale- intensive products use complex technologies but are not cutting edge; differentiated products use advanced design; and science-based products apply modern technologies. 47 Figure 34: Spatial Distribution of Employment of PICS Industries A. Food Processing B. Textile C. Wearing Apparel D. Auto Parts E. Electronic Parts and Appliances F. Rubber and Plastics G. Wooden Furniture and Product H. Machinery and Equipment Concentration The employment maps suggest important differences in diversification across Thailand. We can measure the employment correlation among localities for a given distance with the help of a correlogram. In principle, one might expect employment shares to be closely positively correlated in nearby tambons. Figure 35.A shows that outside of the Northeast, tambon 48 employment up to a distance of 22 kilometers is indeed positively correlated. By contrast, tambon employment is negatively correlated with nearby tambons in the Northeast. In other words, while Northeast firms tend to locate far from each others, firms in other regions tend to locate close to each other. This suggests that the Northeast is not benefiting from agglomeration. Figure 35.B shows the Herfindahl index by region before and after the Asian crisis. It equals the sum over the squared employment shares of all 4-digit industries. A higher index implies a larger concentration. In general, the South and the Northeast lag behind other regions in terms of diversification. Among the 125 4-digit manufacturing industries in Thailand, 29 expanded and 40 declined, while the rest changed by less than 0.1 employment share. As a result, the manufacturing sector became more concentrated, as indicated by the rise in the Herfindahl index. Is diversification linked to the expansion of manufacturing employment? Focusing on the changes between 1996/7 and 2001/2, the rise of Central's employment share would indicate it does, while the contraction in Bangkok's share and the rise in the Northeast's share would indicate it does not. Indeed, relating growth rates in manufacturing employment over this period to the Herfindahl index in 1996/7 across Thailand's 845 amphoes (districts) gives no clear relationship. At the sectoral level, the same finding holds for wearing apparel, while the correlation is positive for electronic parts in districts that did not have any electronic parts industry before the crisis. This supports the notion that knowledge spillovers and innovation are less relevant for a labor-intensive industry like wearing apparel, which relies foremost on a low-cost labor force with basic skills but essential for a differentiated industry like electronic parts, which relies more on linkages to other companies. 49 Figure 35: Regional Employment Concentration, 1996/7 and 2001/2 A. Correlogram 2 1 0 B an g k ok a n d V ic . N orth -1 N orth e a s t C en te r S ou th -2 -3 -4 0 1 2 3 4 5 6 7 8 9 1 0 1 1 D is ta n c e (1 1 k m p e r la g ) B. Herfindahl Index 2 5 0 0 2 0 0 0 1 5 0 0 1 0 0 0 5 0 0 0 1 9 9 6 / 7 2 0 0 1 / 2 B K K V I C C e n t r a l E a s t N o r t h N o r t h e a s t S o u t h C. Amphoe Growth Rates in Manufacturing Employment from 1996/7 to 2001/2 relative to 1996/7 Herfindahl index 00 10 0 80 0 krw_t 60 gr 0 40 0 20 0 0 2000 4000 6000 8000 10000 hhi97 50 Industry and Regional Groups The 2004/5 PICS survey provides us with a rich data set to investigate the two trends that dominated the manufacturing sector since the late 1980s: the relocation of manufacturing outside of Bangkok and Vicinity to the East and Central; and the expansion of high-tech products relative to labor-intensive products. The 1,385 manufacturing firms, surveyed from March 2004 to February 2005, were selected purposefully to emphasize export-orientation and high value-added. The survey presents a rich source for exploring location and performance issues of sizable manufacturing companies. We classify the eight industries covered in the survey into two categories. The low-tech group includes food processing, textiles, wearing apparel, wooden furniture and product; and the high-tech group contains auto parts, electronic parts/electrical appliances, rubber and plastic, machinery and equipment. The group labels should not be taken literally. Some textile industries may employ high-tech equipments, while some equipment factories could be rather labor- intensive. Rather, they provide a convenient way of distinguishing the two groupings, which are nothing more than aggregates over the eight industries. While the PICS selected industries at the 4-digit level, it is nevertheless instructive to compare the growth performance for the 2-digit industries from which they were drawn. They amounted close to 50 percent of Thailand's manufacturing value-added both in 1996 and 2003 and slightly less in-between (Figure 36). At this aggregated 2-digit level, these eight industries declined during the Asian crisis but have recovered since and are now in real terms one quarter above the 1996 level. However, the low-tech group grew annually in real terms by between -1.5 percent to 3.6 percent, while the high-tech group expanded by 3.6 percent to 8.9 percent. As a result, the share of the high-tech group rose from one fifth in 1998 to over one third in 2003. The reason for aggregating industries into two groups is that we need sufficient observations in each category to explore also differences across regions. The Survey covers separately the North, Northeast, Central, Bangkok and Vicinity, East and South. For parts of the discussion, we will aggregate regions into two groups: Bangkok and Vicinity ("Bangkok") versus the rest of Thailand. This is required due to the limited number of observations, but also motivated by the trend identified above: the importance of Bangkok and Vicinity in terms of manufacturing output has declined relative to the rest of Thailand since the late 1980s. Figure 36: 2-Digit PICS Industries, Real Value Added (1988 Prices) and Total Value Added (%), 1996 to 2004 800,000 100% 700,000 90% 80% 600,000 70% 500,000 60% 400,000 50% 300,000 40% 30% 200,000 20% 100,000 10% 0 0% 1996 1997 1998 1999 2000 2001 2002 2003 1996 1997 1998 1999 2000 2001 2002 2003 Food Products and Beverages Textiles Food Products and Beverages Textiles Wearing Apparel Funiture; Manufacturing n.e.c. Wearing Apparel Funiture; Manufacturing n.e.c. Rubber and Plastic Products Machinery and Equipment Rubber and Plastic Products Machinery and Equipment Electrical Machinery and Apparatus Motor Vehicles Electrical Machinery and Apparatus Motor Vehicles 51 Products, Size and Exports Having organized the firms into regional and technology categories, we can now explore the differences between these groups across three dimensions: products, company characteristics, and the investment climate. The investment climate indicators come in two types: the firms' perceptions of both general constraints to operations and obstacles to doing business.16 Firms focus on similar 4-digit products across the regions, especially in the high-tech sector, although their shares are different. Among the 11 products represented in the sampling frame among low-tech industry, only six are among the top three products in any of the regions (Figure 37). And among the 26 high-tech products, only seven are among the three most important products. The Northeast is engaged in products that are produced also in at least another two regions (furniture, wearing apparel and textile preparation for low-tech; and electronic parts, plastic products and auto parts for high-tech). By contrast, among low-tech industries, Central stands out for sugar products and the South for wooden container production; and among high-tech industries, the South stands out for rubber products and the North for agricultural machinery. However, the weights of these products differ widely from region to region. In the Northeast, North and South, low-tech industries are less concentrated than high-tech industries, while there are little differences in the more advanced regions of Bangkok and Vicinity, Central and the East. This is consistent with the idea that economic development comes with diversification, especially of high-tech industries.17 We now turn to company characteristics. Relative to high-tech firms in the other regions, Bangkok's firms tend to be older, more domestically owned and less export- or import- oriented. They have a smaller workforce, although more often hired from other regions, and draw more on raw materials from other regions. Bangkok's low-tech industries have broadly similar features as Bangkok's High-tech industries, with the exception of a higher export- and import-orientation. The Northeast's low-tech industries stand out for low foreign- ownership and low export-orientation. Finally, we compare investment climate indicators between Bangkok and other regions. Bangkok high-tech companies report less problems with infrastructure, and more difficulties with business support services, skilled labor shortages, corruption, competition from imports and utility prices. Bangkok's low-tech firms also report obstacles in macro stability, anti- competitive practices and high taxes. 16The survey identified constraints through the firms' rating of 18 issues according to the severity of each given constraint ("closed" question), and obstacles through firms' selection of the three biggest problems out of a list of 22 issues ("open" question). 17Imbs and Wacziarg (2003) show that as poor countries get richer, sectoral production become less concentrated and more diversified. Sectoral specialization applies only to high-income countries. This holds not just for the shift from agriculture to manufacturing and services, but also for the manufacturing sector alone. 52 Figure 37: 4-Digit Products by Regions A. High-Tech Industries 90 80 70 60 50 40 30 20 10 0 Plastic toPartsApplian ce ic arts e ic tic arts ber tic toP lianc toParts Plastic Plasticac hinery tron Plas toP Rub Plas Au Electron Au M Elec Au Dom mApp Au Electronic Electronic Do Ag << BKKVIC | << Central >> | << East >> | << North >> | << Northeast >> | South >> B. Low-Tech Industries 90 80 70 60 50 40 30 20 10 0 el WearApparPrep ile roc gar el FurnituPrep re xtile rod iture rniturerApparel Textile FishProdFurnituodC re tain Text FurnitureFruitVegP Su Furniture Te FishP FruitVegProcFurnWearAppar on FuWea Prep Wo << BKKVIC | << Central >> | << East >> | << North >> | << Northeast >> | South >> 53 Firm Productivity The next sections will relate these constraints to a measure of firm-level performance: total factor productivity. It is a multi-factor productivity measure that represents the efficiency of the firm in transforming inputs, including skilled and unskilled labor and capital, into outputs, defined as value added or the differences between sales and intermediary goods. The estimation follows the Olley and Pakes (1996) routine which allows for simultaneity and selection biases.18 Subtracting from value added labor and capital inputs weighted by the estimated production elastiticies gives us total factor productivity. Figure 38 shows the regional graphs of the distribution of firm level productivity relative to the national average, which is normalized by a mean of unity and a variance of 0.1. Mean productivity is higher in Bangkok than outside of Bangkok, and high-tech industries have higher mean productivity both within and outside Bangkok than low-tech industries. Figure 38: Regional Distribution of Total Factor Productivity N o rth C e n tra l-2 pro v B a n g ko k+ 2 c e n tra l p rov. .3 .2 .1 0 E a s t N o rth e a s t S o u th onit .3 acrF .2 .1 0 .2 .4 .5 .6 .7 .8 .9 11 .1 .21 .31 .41 .51 .6 .2 .4 .5 .6 .7 .8 .9 11 .1 .21 .31 .41 .51 .6 .2 .4 .5 .6 .7 .8 .9 11 .1 .21 .31 .41 .51 .6 A c tu r a l TFP L ow -T ec h T F P D is trib utio ns N o rth C e n tra l-2 p ro v B a n g ko k+ 2 c e n tra l p ro v. .3 .2 .1 0 E a s t N o rth e a s t S o u th noitca .3 .2 .1 Fr 0 .2 .4 .5 .6 .7 .8 .9 11 .1 .21 .31 .41 .51 .6 .2 .4 .5 .6 .7 .8 .9 11 .1 .21 .31 .41 .51 .6 .2 .4 .5 .6 .7 .8 .9 11 .1 .21 .31 .41 .51 .6 A c tu r a l TFP H ig h-T e c h T F P D is trib u tio n s 18Simultaneity bias occurs if input demand is in part determined by the manager's knowledge of productivity levels. As inputs and outputs are determined at the same time, inputs are not exogenous to output levels. Selection bias stems from productivity being affected the location choice. 54 Decomposing Firm Productivity What are the correlates of firm productivity? We decompose the contributions of 4-digit product characteristics, company characteristics and investment climate variables on firm level productivity.19 We model productivity to depend on company characteristics (employment size at birth, age, share of skilled workers, share of local workers, share of raw materials from local sources, presence of imports, presence of exports and presence of foreign ownership), product characteristics (4-digit industry dummies) and our sets of company closed and open responses to the investment climate. As expected, total factor productivity is positively related with high skill worker share, export-orientation and foreign- ownership, although these factor matter more among Bangkok's high-tech industries. Regarding investment climate indicators, the findings are similar to the descriptive statistics quoted previously. The most striking differences are as follows. Skilled labor shortages reduce productivity for Bangkok's high-tech industries but not otherwise. Utility prices, political instability, lack of business support corruption and crime lower the performance of Bangkok companies, while infrastructure, telecommunication and taxes hold back productivity outside of Bangkok. What is the overall impact of investment climate differences on productivity? Predicting firm performance from company characteristics and product dummies suggests that they account for 79 percent of firm productivity. However, this is likely to be a lower bound of the impact of investment climate on performance. The investment climate affects location choice, so that company and product features themselves reflect investment climate influences. A key difference between Bangkok and other regions lies precisely in the returns to such company and industry characteristics. Figure 39 shows the regional distributions of simulated firm productivity relative to the national mean. We allow the companies outside of Bangkok to have Bangkok returns to firm characteristics and the same product structure as in Bangkok. Firms outside of Bangkok now have higher returns than Bangkok, as the values of firm characteristics that increase productivity, such as age, size and ownership are in fact better than in Bangkok. A key question is why are the product portfolios so different? For companies outside of Bangkok products, this may reflect the presence of export-oriented multinational, or the dependence on local raw materials for product development. While their productivity levels are lower in these activities when compared to more traditional products of the Bangkok and Vicinity, their contribution to the export boom and economic recovery over the last years was crucial. The analysis suggests that addressing deficits in institutions, infrastructure, and business services outside of Bangkok and Vicinity is essential for sustaining the export boom. 19This methodology is similar to Dollar et al (2003), Escribano and Guasch (2004) and Haltiwanger and Schweiger (2005). 55 Figure 39: Simulated Productivity Outside of Bangkok Using Returns of Bangkok and Vicinity N o rth C e n tra l -2 p ro v B a n g ko k+ 2 c e n t ra l p ro v . .3 .2 .1 0 E a s t N o rth e a s t S o u th n oitc a .3 .2 Fr .1 0 .2 .4 .5 .6 .7 .8 .9 1 1 .1 .21 .31 .41 .51 .6 .2 .4 .5 .6 .7 .8 .9 11 .1 .21 .31 .41 .51 .6 .2 .4 .5 .6 .7 .8 .9 11 .1 .21 .31 .41 .51 .6 S im u la t e d T F P L o w - T e c h S im u la te d T F P u s in g B a n g k o k R e tu rn s N o rth C e n tra l -2 p ro v B a n g ko k+ 2 c e n tra l p ro v. .3 .2 .1 0 E a s t N o rth e a s t S o u th n oitc a .3 .2 Fr .1 0 .2 .4 .5 .6 .7 .8 .9 11 .1 .21 .31 .41 .51 .6 .2 .4 .5 .6 .7 .8 .9 11 .1 .21 .31 .41 .51 .6 .2 .4 .5 .6 .7 .8 .9 11 .1 .21 .31 .41 .51 .6 S im u la te d T F P H ig h - T e c h S im u la te d T F P u s in g B a n g ko k re tu rn s 56 Technological Capability The findings on total factor productivity are supported by evidence on regional variation in technological capability. The 2004/5 PICS provides the information needed to construct an index of technological capabilities across manufacturing establishments in Thailand (World Bank 2005). TCI draws on the taxonomy developed by Lall (1992), which identifies and categorizes firm-level technological capabilities into investment, production, and linkages activities. TCI permits useful comparison of the technological capabilities across firms and enables econometric analysis of the influences on the acquisition of firm-level technological capabilities. TCI provides a composite measure of technological capabilities composed of information about firm-level technological behavior that is provided by the rich data collected in the PICS. TCI is composed of 27 separate technical activities. Investment technological capabilities is represented by 6 separate technical activities; production technological capabilities by 14 separate technical activities; and linkages technological capabilities by 7 separate technical activities. A single point is given for each technical activity the firm has performed; for higher levels of IT-related investments and computer- controlled machinery, an additional point is scored. Therefore, each firm is ranked out of a total technological capability score of 29, and the result is normalized to give a value between 0 and 1. Establishments located in the East and Central score highest and those in the North, Northeast and the South regions score lowest on the TCI scale (Figure 40), while establishments located in Bangkok and Vicinity fall somewhere in between. The businesses located in the East and Central are consistently ranked first and second, respectively, also in each of the three areas of activity: Investment (0.529 and 0.471), Production (0.428 and 0.396), and Linkages (0.370 and 0.333). Establishments in Bangkok follow with overall average TCI score of 0.415, and also in Investment (0.411), Production (0.370), and Linkages (0.316). Establishments located in the Northeast region have the poorest average Investment and Linkages TCI scores of 0.303 and 0.202, respectively. Figure 40: Kernel Density Plots of TCI by Region South Bangkok Northeast East Central ytis North enD North Central Bangkok East Northeast South 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Technological Capabilities Index 57 Northeast Exporters The PICS analysis indicates that enterprise in the Northeast have lower productivity than Bangkok-based firms. Yet, while the Northeast manufacturing base is small, it includes many foreign-owned and export-oriented companies. What businesses locate in the Northeast? What products do they produce? Box 4 gives eight examples of export-oriented firms, both local and foreign-owned. These firms produce agricultural (fruit and vegetable cans, flour and sweetener, Jasmine rice, chicken products) and industrial (iron roofs, cargo services, trucks, and carpets) goods and services. Three features stand out. First, the Northeast attracts companies through access to raw materials, workers, and land as well as BOI incentives. Second, the products either build on traditional activities (rice, livestock and fabrics), service the storage and transportation needs resulting from the distance to consumer markets (cargo and trucks), or supply housing materials for the over 20 millions regional population. Third, while all companies consider lack of skilled workers a serious constraint, daily workers constitute a large part of the work force and only some firms use wages to attract qualified employees. Box 4: Business Case Studies of Northeast Exporters Thaisun Food Product Co. is a Japanese and Taiwanese company producing fruit and vegetable cans. It located in Nong Khai province in 1988, attracted by the availability of labour and the special BOI investment incentives. While originally exclusively export-oriented, it now also sells to the Thai market. The company still buys its raw materials from the same five local suppliers as at the beginning of its operations. The main problem is the lack of qualified workers. Corn Products Amardass is an agricultural joint venture, founded by the 2001 merger of a Thai firm and a US company, which holds 80 percent of the stock. The two main products are flour and sweetener, about 70 percent of which are exported. The company chose the Northeast for raw materials (mainly cassava) and BOI investment promotions. Among the 300 employees, 20 percent are daily workers, receiving wages slightly above the minimum wage. The company is sometimes confronted with labor shortage. Deimos Holding Co.,Ltd was founded in 2000 in Nakhon Ratchaseema to provide cargo services to US mother company Effen Food Co. Ltd, which provided half of the founding capital. Services include storage of goods, packaging and truck forklift logistics. The company has expanded from 10 permanent employees to 32 as of 2004. Monthly wages are on average Bt7,600 in 2003. Bluescope Lysagh, Ltd. is a subsidiary of an Australian multinational steel company. Founded in 1988 as a coated steel factory in Prathumthani, the company has since expanded to Khonkaen, Rayong, Chiangmai and Songkhla. Specializing in iron roof construction and instalment, about 70 percent of it goods are sold domestically. The main obstacle to expansion of this medium-size company is lack of qualified workers, even though the company pays wages above the levels of civil servants and other private employers. The company provides basic in-house training to workers. Chiameng Rice Mill Corporation Co., Ltd, founded in 1937, originated from the first rice mill in Bangsue, Bangkok. By 1955, the company started exporting Jasmine rice under the brand name "Golden Phoenix", known as "Hong Thong" rice in Thailand and widely recognized for its swan logo. This family company, with its headquarter in Bangkok, has four branches and produces the full circle of jasmine rice products, starting from seeds, processing, quality-enhancement to packaging and distribution. The company has invested continuously in modernizing its technology, which it considers superior to its competitors. The main foreign competition includes basmati-rice producing companies from India and Indonesia. There is little competition in the input market as the company purchases directly from farmers, selecting high quality unhusked rice or which it pays a price premium. The location of the Sri Sa Ket branch was chosen 58 for raw material supplies and land area. This branch employs around 150 workers with a salary of around Bt6,000 to Bt6,500. The main problem is lack of skilled labor. Kawna Kaisod Co.Ltd., a Thai company established in 1981 with BOI support, began chicken export in 1993. It belongs to a group of five companies, whose activities range from breeding and broiler farms, a frozen chicken factory and a chicken feather factory. The company invests regularly in upgrading of its technology and exports about 80 percent to Europe and Asia. Raw materials are sourced mostly from local suppliers. Since its foundation, employment has increased almost every year to reach 1,400 employees in 2004. Only 15 percent of the workers are on monthly contracts. The main obstacle in employment is the lack of qualifications of the employees. A part of the profits is reinvested to avoid the need for bank loans. Khon Kaen Cho Thawee Co., Ltd is a truck and trailer company was founded by a former rice mill owner. The owner relied on the know-how of his foreign-trained sons to expand over the last 30 years its business. In the mid-1990s, the company entered a joint venture with Emil Doll Gmbh to upgrade its technology. The foreign company selects raw materials and is in charge of boosting production efficiency. The most important products today are trailers and half-trailers that can handle heavy freight on every type of road. Some trailers have anti-vibration systems as well as axle systems that allow every wheel to turn freely enabling them carry manufactured concrete weighing up to 160 ton per block. The company is careful to select only high-quality inputs, most of which are imported from abroad. Four fifths of the customers are foreign, and much of the domestic customers are state-owned companies. The company expanded from around 200 permanent employees in 1994 to 300 employees after the merger, before it declined to 150 workers after the Asian crisis. Today, it has recovered to 400 employees. Salaries, which were cut by about 20 percent in 1997, are on average Bt8,300. The main problem is lack of qualified labor. The company reinvests most of its profits, but also takes out Bank loans to fund capital investment. Carpet Maker Co.,Ltd. is a joint venture of Thai business people, established in 1985 in Khonkaen. Originally producing silk, it started in 1987 to produce carpets for domestic market, and began in 1998 to focus on exports. Today, exports account for 80 percent of total sales. In contrast to its technologically more advanced two main competitors, the company specializes in high-quality handmade production. Cloth and wool raw materials are imported from Europe, while the glue is ordered from a domestic supplier. The company had 335 employees in 2004, compared to 270 employees in 2001. About 60 percent are monthly employees, receiving a salary of about Bt6,000. One important obstacle is to find qualified employees for executive level responsibilities. The company draws on Bank credits to respond to new funding needs, in addition to credits from raw material suppliers. Board of Investment Promotions The Board of Investment (BOI) is a powerful instrument of industrial policy in Thailand. Established in 1960, its objective is to promote new investment through providing tax privileges and import duty exemptions. The 1977 Investment Promotion Act empowers the BOI to grant various fiscal and non-fiscal incentives for foreign and domestic investment that meet national economic development. They include exemptions from import duties, corporation income tax and dividend tax; and permissions to bring in expatriate worker, own land and operate in otherwise prohibited industries for foreign firms. In the first two decade of its existence, the basic thrust was to support capital-intensive industrialization. In 1983, the emphasis shifted towards promoting export orientation. Explicit criteria for the granting of investment promotions were introduced for the first time. They included the generation of foreign exchange through export; employment creation; utilization of local raw material; and industrial decentralization. However, the BOI system still kept many small and medium firms out of exporting or of supplying exporters, as it was mostly large firms that received 59 BOI promotional benefits. During the last years, about 20 to 30 percent of promoted investment was from Thai companies, 40 to 50 percent from foreign companies, with the rest accruing to joint ventures. In general, the BOI has been a follower, not a leader. BOI promotions tend to lag behind the business cycle. Figure 41 shows the amounts of investments from 1994 to 2004 along the four steps in the application procedure: from valid ("net") applications to approvals, issue of promotion certificate and the start-up of factory operations. While the amount of net applications declined in line with GDP growth during the Asian crisis, the value of promoted investment as percentage of GDP started declining in 1999 and increased only in 2004, about four years after the recovery process began. The investment of promoted firms that started operations is still no more than one percent of GDP in 2003 and 2004. The same trends are evident at the sectoral level. For example, not until the mid-1980s, after a large devaluation, took the electronics sector really off, and the BOI started issuing more electronics promotions in 1987 than in its entire existence to date (Christensen et al 1993). In any case, to the extent that tariffs and taxes have become less important over time, the importance of BOI incentives has declined. Figure 41: BOI Applications, Approvals, Certificates and Start-Ups (% of GDP), 1994 - 2004 25 10.0 20 5.0 PDGfotnecr 15 0.0 tnecr Pe 10 -5.0 Pe 5 -10.0 0 -15.0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Net applications received Net applications approved Promotion certificates issues Promoted firms starting operation LHS - Real PC GDP Growth BOI Zones Since 1987, BOI investment promotions pursue the objective of regional development. The country was divided in three zones based on proximity to Bangkok, and the incentives offered were differentiated according to the zone a business is located in. The zones evolved over time, and as of today, Zone 1 comprises Bangkok and Vicinity and Zone 2 provinces from the East, Central, West and Phuket in the South (Figure 43). How effective has the zoning policy been? The ultimate criteria of success would be regional convergence, which we have already investigated. A weaker yardstick is to investigate the distribution of BOI allocations of investment promotion certificates by regions and zones. Even here, the influence of BOI is limited, as private investment decisions are influenced by a range of factors. In spite of the BOI zoning policy, the investment promotions are concentrated on Bangkok and Center (Figure 42). From 1997 to April 2005, the Northeast's share averaged 3.8 percent, compared to a GDP of 11 percent. The North and the South, with GDP contributions of about 9 percent, received 3.2 percent and 6.4 percent on average, respectively. 60 However, the BOI zoning has succeeded to the extent that investment has shifted from Bangkok to Zone 2 areas. Even this development is likely to have resulted also from congestion and high land rents in Bangkok. The overlap of investment promotions and manufacturing employment confirms this picture (Figure 43.B). In the North, Northeast and South, some manufacturing activity has emerged in large provinces with adequate infrastructure and low wages. Even those firms investing in Zone 3 locate typically as close as possible to Zone 1 in order to limit transport costs while maximizing investment incentives. For example, about one half to three quarters of Northeast investment promotions go to Nakhon Ratchasima, which is in the Southwest corner of Zone 3, and close to Bangkok. Overall, the BOI zoning policy has influenced the spatial pattern of industrialization, but it has failed to induce widespread industrialization beyond Bangkok and the Center. Figure 42: BOI by Regions and Zones (%) A. Regional BOI Promotion Certificates, 1997 - 2005 A. BOI Promotion Certificates by Zones, 2001 - 2005 100% 100% 90% 80% 80% 70% 60% 60% 50% 40% 40% 30% 20% 20% 10% 0% 0% 1997 1998 1999 2000 2001 2002 2003 2004 Apr 2005 2001 2002 2003 2004 Apr 2005 Northeast North South Other Regions Zone 1 Zone 2 Zone 3 Other Zone 3 Northeast North South 61 Figure 43: BOI Investment Zones A. BOI Zones 2005 B. Manufacturing Employment, 1996/7 and 2001/2 C. BOI Zones, 1987 to 2004 1 9 8 7 1 9 9 3 U n til 2 0 0 3 2 0 0 4 Box 5: Eastern Seaboard Program As part of the general policy switch from import-substitution to export-promotion, Thailand launched the Eastern Seaboard Program in the 1980s with the support from the Japanese government's Overseas Economic Cooperation Fund. This is the most ambitious attempt to promote infrastructure-led development of an area outside Bangkok. The project was organized around the newly discovered natural gas supply in the Gulf of Thailand. The plans included initially large scale investment in heavy industry, ranging from steel mills to gas and oil processing. In the end, these most ambitious projects were scaled back for financial reasons, but two sea ports (Map Ta Phut and Laem Chabang), a sizeable industrial estate and an export promoting zone were established. This area of three million people is in Rayong province up to 190 km distant from core Bangkok. The export-oriented factories were a success and contributed to the export boom in the late 1980s and early 1990s as well as the economic recovery from the Asian Crisis. By 1997, 200 companies had relocated from Bangkok to the Eastern Seaboard. 62 Industrial Estates Industrial estates operated by the industrial estate authority of Thailand (IEAT) are a second instrument of the RTG for promoting investment. The 1979 IEAT act provides special incentives for investors locating in industrial estates situated in regional areas. These industrial estates comprise both general industrial zones and export-processing zones. The incentives are similar to those offered by the BOI, with the principal difference that IEAT grants investment privileges only to investment projects located in the industrial estates. Apart from industrial estates, there are industrial zones created by Ministry of Interior, privately-run industrial parks as well as industrial estates managed jointly by the public and private sector. Most industrial estates were created during 1986 to 1996, before demand for additional industrial estates was cut short by the Asian crisis (Figure 44). Industrial estates have failed to promote investment diversification to remote areas. The principal reason is that supply driven infrastructure projects are unlikely to succeed without a clear market demand for the services provided. Industrial estates themselves are located primarily in Zone 1 and Zone 2. Since industrial estates offer similar incentives as those presented by Zone 2 or Zone 3, firms have little reason to relocate to Zone 3. There are only five industrial estates or parks in the Northeast, four in the North and two in the South. The bulk of industrial estates are located in the Eastern Seaboard area (Box 5) or in central provinces just north of Bangkok. These industrial estates have been successful in forming clusters of companies, especially in Bangkok, the East and the Central region. Industrial estates have attracted foremost foreign firms and large Thai firms, but much less small-and- median enterprises. Nevertheless, firms from industrial estates near regional urban centers typically form various linkages to the local supplier network of smaller firms in the urban area. Figure 44: Industrial Estates 63 Construction Permits Construction area permits are typically considered leading economic indicators because they tend to increase or decrease before the economy/business cycle changes as a whole. For example, if building permits increase and residential construction is growing, one can expect an increase in demand for durable goods to follow, such as major appliances that will be placed in new homes. The increase in building permits foreshadows the increase in the demand for durable goods. Figure 45.A shows that residence, commerce, industry and other construction area permits declined already in 1995 or 1996, one to two years before the Asian crisis. However, with the exception of residence permits, the recovery in the construction indicators is sluggish. In the same way as this indicator may tell us something about the status of the business cycle, it can also inform us about the regional distribution of economic activity. Figure 45.B displays the shares in total construction area permits by region from 1995 to 2003. Only the South and Center increased their shares during this period. The Northeast's share declined since 2000 and accounted for no more than 5 percent of all construction permit in 2003, the lowest share since 1992. Figure 45: Construction Area Permits A. Construction Area Permits and Real Per Capita GDP, 1989 to 2003 30,000 15.0 )s er et M 25,000 10.0 e aruqS 00 20,000 5.0 ,01( t sit 15,000 0.0 m erP cenreP a reA 10,000 -5.0 noi ctur 5,000 -10.0 stnoC 0 -15.0 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Residence Commerce Industry and Other LHS - Real PC GDP Growth B. Regional Share in Total Construction Area Permits, 1995 to 2003 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1995 1996 1997 1998 1999 2000 2001 2002 2003 Northeast North South Center Bangkok 64 Business Regulations Business regulations affect directly the productivity of investment and economic activity of micro to large enterprises. It covers a range of issues, from the costs to starting a business to the costs of closing a business. Measuring quantitatively costs of doing business has long been a challenge. Recently, the 2004 and 2005 World Bank Doing Business Surveys proposed a new methodology based on quantitative indicators of business regulations and their enforcement. These surveys of 145 countries referred to a business or workers of precisely specified characteristics which operate in the country's most populous city. During spring 2005, we returned to the same respondent as for the 2005 Thailand survey with the same questionnaires apart from one key modification. The location of the business was assumed to be Khonkaen, the administrative center of the Northeast, rather than Bangkok. Five modules of the original survey were amenable to this adjustment (Starting a business; hiring and firing workers; enforcing contracts; closing a business, and registering property). The responses for Bangkok and Khonkaen are shown in Figure 46. The differences are generally minor. Compared to Bangkok, the costs of doing business in Khonkaen are no higher, and perhaps somewhat lower. If enterprises do not locate to Khonkaen, government red tape does not appear to be the reason. Figure 46: 2005 Doing Business Survey ­ Khonkaen and Bangkok 50 40 30 20 10 0 Khonkaen Bangkok Starting a Business - Days Starting a Business - Cost ($100) Hiring and Firing (100 = Worst) Enforcing Contracts - Months Enforcing Contracts - Cost (% of Debt) Closing a Business - Years Closing a Business - Cost (% of Estate) 65 Banks Thailand's economic growth came along with a rapid development of the financial sector. Broad money (M2) increased from about two-fifths of GDP in the early 1980s to 90 percent in the early 2000s. This is a ratio far higher than the norm for a country at Thailand's income level. Over the last 30 years, the financial sector expanded also in the Northeast. The amount of deposits mobilized by commercial banks increased from 10 percent of Northeast GDP in 1975 to 22 percent of Northeast GDP in 2004. This is only moderately lower than the numbers for the North, South or the Center. Only Bangkok, Thailand's financial and commercial hub, towers above the rest, with a deposit-GDP ratio of 110 percent in 2004. This points to the size of the Bangkok market and the ties of commercial banks to large enterprises with headquarters in the capital. Most branches of domestic commercial banks are concentrated in Bangkok and Center, reflecting the economic potential of these regions. About two thirds of all branches were located their in 2005, slightly more than in 1999. This compares to 13 percent in 2005 for the Northeast, down from 15 percent in 1999 (Figure 47.A). While this is lower than the regional population share, it exceeds the Northeast's GDP share of 10 percent. Furthermore, the number of commercial branches increased almost fourfold in 30 years, from 102 in 1975 to 468 in 2005. Finally, branches are also more equitably spread than deposits and credits. Bangkok accounted in February 2005 for two thirds of all deposits, the Center for one fifth, and the Northeast, North and South each for five percent (Figure 47.B). Credits are even more concentrated on Bangkok, which accounts alone for three quarters of all credits. The Northeast's share was five percent in 2005, almost the same as in 1999. The Asian crisis was linked to overexposure to foreign debt and risky loans. While banks have benefited from the acceleration in economic activity since the late 1990s, they have remained more cautious in managing funds and providing credits. Broad money, which had increased to over 100 percent in the early 2000, declined again to 90 percent in 2004. While the amount of deposits increased by 14 percent in real terms between 1999 and 2005, the value of credits declined by 11 percent over the same period, even though lending rates have declined (Figure 47.C). Credits exceeded deposits by 13 percent in 1999, whereas deposits exceed credits in 2005 by the about the same share. Commercial banks in the Northeast have also become more conservative in lending. Credits exceeded deposits by 20 percent in 1996 and 8 percent in 1999, but deposits surpass credits by 17 percent in 2005. In fact, only commercial banks in Bangkok released loans in excess of deposits in 2005. While credit supply to enterprises is sluggish, household and consumer credit has increased by annually 15 percent over the last three years. Much of this increase was due to influential public financial institutions. In 2004, about one quarter of all credits was extended by public financial institutions. The three main public financial institutions are the Government Saving Bank, the Government Housing Bank and the Bank for Agriculture and Agricultural Cooperatives (BAAC). They accounted in 2004 each for between 7 to 8 percent of the total credit extended. These institutions are well presented through branches in provinces and districts throughout the country. 66 Figure 47: Commercial Bank by Regions, 1999 and 2005 A. Number of Branches by Region, March 1999 and February 2005 4000 3500 3000 2500 2000 1500 1000 500 0 Mar-99 Feb-05 Bangkok Center Northeast North South B. Deposits and Credits by Region (Million Baht, 1988 Prices), March 1999 and February 2005 6000000 5400000 4800000 4200000 3600000 3000000 2400000 1800000 1200000 600000 0 Deposits 1999 Credits 1999 Deposits 2005 Credits 2005 Bangkok North Northeast Center South C. Commercial Bank Lending Rates, April 1997 to April 2005 20 15 10 5 0 Apr-97 Apr-98 Apr-99 Apr-00 01-Apr Apr-02 Apr-03 Apr-04 Apr-05 Nominal Interest Rates Real Interest Rates 67 Workers Jobs Low living standards in the Northeast are to a large extent a reflection of the lack of well- paying jobs. Unsurprisingly, one of the most common complains from Northeast's villagers is the absence of work that pays a decent wage. Nevertheless, there are different views on the performance of Thailand's labor market. Some characterize it as flexible and efficient, pointing to low unemployment rates, low labor organization, and high worker mobility. Others argue that labor protection laws reduce the competitive edge of firms. And yet others point out that it is inequitable and fails to deliver on job security. The following sections attempt to characterize the contribution of the labor market to regional convergence in economic well-being. After all, what is central is employment growth such that Northeast workers find gainful employment, have adequate productivity levels that are fairly compensated, and achieve reasonable income security for workers. We will first look at labor supply in terms of quantity, quality, and prices, and then explore the impact of two labor market institutions (minimum wage and labor protection act). Finally, no discussion of the Northeast labor market would be complete without investigating worker migration to other regions. Perhaps even more than for the rest of this report, it will be important to look at the Northeast relative to the other regions. The discussion will focus on the period from the early 1990s to 2004, which allows us to see the workings of the labor market through the experience of the Asian crisis. But first of all we set out the basic challenge: raising labor productivity. Figure 48 plots regional GDP divided by employment from 1991 to 2004. Throughout the entire period, Northeast workers are the least productive. In 2004, they produced only a sixth of what the average worker in Bangkok, Central, East and Vicinity generated, and reached no more than 70 percent of the output of a Northern worker. While labor productivity grew by annually 2.3 percent, this was 0.4 percent less than in the North, 0.5 percent less than in Thailand overall and a remarkable 7.7 percent less than in the East. Ultimately, lifting living standards in the Northeast will only be possible if this productivity gap is closed. Figure 48: Labor Productivity, 1991 to 2004 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Bangkok Vicinity East Central West North Northeast South 68 Demographics Population size and population growth are intimately related to the labor market. Steady increases in the labor force and employment are one important driver for higher national income. Thailand's population has grown rapidly. It more than doubled from 26 million in 1960 to 64 million in 2004. The population increased in all regions, but especially in Bangkok (Figure 49.A). Its population tripled between 1960 and 2000, compared to a 2.5 times rise in the South, a 2.3 times rise in the Northeast and Center, and a doubling in the North. One factor behind these differences is fertility rates. Due to rising incomes and government-sponsored family planning programs, the total fertility rate declined from over 6 live births of women in childbearing age in 1964/65 to around 2 in the mid-1990s. In turn, the national population growth rate fell from 3.1 percent in 1960 to less than 1 percent today, or from 2.9 percent to below 1 percent in the Northeast (Figure 49.B). In spite of this decline, the Northeast population is still younger than in other regions, apart from the South. Around 30 percent of the population are younger than 15 years of age, compared to no more than 18 percent in Bangkok or 23 to 24 percent in the Center and the North (Figure 49.C). This suggests that large increases in population are likely to persist for some time. Even with lower fertility rates, the large increase in the number of young adolescents implies strong increases in population numbers. This poses major challenges for policy-makers in providing accessible education and employment opportunities as they reach working age. But there is also an upside. Schooling of the current and future generations will be more egalitarian than of past generations. This will be an important force for regional convergences over the long-term. Furthermore, a very high dependency ratio makes saving difficult for most working individuals and families. This problem should decrease as today's young adolescents reach working age, presenting the opportunity for reducing poverty ­ as long as there are jobs for them. Figure 49: Demographic Indicators A. Regional Population, 1960 to 2000 21,000,000 18,000,000 15,000,000 12,000,000 9,000,000 6,000,000 3,000,000 0 Bangkok Center North Northeast South 1960 1970 1980 1990 2000 B. Regional Population Growth Rates, 1960 to 2000 C. Household Members Aged Less than 15 Years, 1988 to 2002 5 36 33 4 30 3 27 2 24 21 1 18 0 15 Bangkok Center North Northeast South Bangkok Central North Northeast South 1960 1970 1980 1990 2000 1988 1990 1992 1994 1996 1998 2000 2002 69 Working-Age Population Job creation is important from both an economic and a social point of view. Being without a job can undermine the self-esteem of workers and the well-being of their families. Between 1994 and 2004, the labor force has grown from around 32.5 million to around 36 million in Thailand, and from 10.5 million to around 12 million in the Northeast. Figure 50 looks at the changes in the composition of the population aged 15 years or older ("adults") over the last decade. Figure 50.A to Figure 50.E refer to August of the year, Thailand's harvest season. A number of points stand out. First, unemployment, including seasonal unemployment, is low. It affected no more than 3 percent of the adult population in August. Even during the Asian crisis, where GDP declined by 11 percent in 1998, did not increase unemployment beyond that level. This suggests that almost all adults looking for jobs were able to find some kind of job. The high adaptability of the Thai labor market to changes in labor demand is likely to be linked to two factors: downward-flexibility of wages and the safety net of farm employment. Second, the Northeast employment share has declined from a high level. Over four fifths of the adult population were part of the labor force in August 1991. The share dropped sharply during the Asian crisis and was in August 2004 still some 7 percent below the 1994 level. The North and the Center experienced similar trends, while the employment shares remained constant in Bangkok and the South. The declining importance of agriculture explains some of these changes. Bangkok is unaffected, as agricultural employment is negligible, as is the South, where the agricultural sector remains viable. The South's employment share is now as high as the Northeast's, equal to close to three-quarters of the adult population. Third, the Northeast has the lowest employment shares during the agricultural slack season in February. Non-seasonal and seasonal unemployment in the Northeast rose to as much as 8 percent around 2000 and remained still at 4 percent in February 2004, around 2 to 3 percent higher than in other regions (Figure 50.F). However, the gap in employment shares from August to February declined from 12 percent in 1994 to 8 percent in 2004, again related to the declining agricultural work force. Fourth, other factors behind lower employment shares are education and household work. The shares of Northeast people staying out of the labor force in order to attend schools or to stay home for household work increased over the last decade by over 50 percent and over 40 percent, respectively. These increases are the highest in any region. Overall, unemployment is low but employment shares have declined substantially. Some of this drop should not be a cause for worry. People postpone their entry to the labor market to invest more in education and become more productive workers later. Education also changed expectations so that fewer workers are willing to take on physically demanding farm labor jobs. But labor force participation also declined for bad reasons: labor demand declined during the Asian crisis and decent jobs have remained scarce since then. If half of the decline in participation rates was due to a "discouraged worker" effect, Northeast unemployment in August would be 5 percent rather than 1 percent. 70 Figure 50: Labor Force Composition A. Bangkok - August B. Central - August 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Employment Unemployment NILF-Seasonal NILF-Education NILF-Household Employment Unemployment NILF-Seasonal NILF-Education NILF-Household NILF-Too Old NILF-Other NILF-Too Old NILF-Other C. North - August D. Northeast - August 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Employment Unemployment NILF-Seasonal NILF-Education NILF-Household Employment Unemployment NILF-Seasonal NILF-Education NILF-Household NILF-Too Old NILF-Other NILF-Too Old NILF-Other E. South - August F. Northeast - February 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Employment Unemployment NILF-Seasonal NILF-Education NILF-Household Employment Unemployment NILF-Seasonal NILF-Education NILF-Household NILF-Too Old NILF-Other NILF-Too Old NILF-Other 71 Job Entry While the Asian crisis in now almost ten years ago, the decline in labor force participation rates suggests that its impact is still being felt. The young entrants to the labor market are perhaps the group that has been affected most. Figure 51 looks at the trends in the regional shares of unemployment of the labor force by age. It shows four years: the early 1990s (1991), the last year before the crisis (1996), the peak of the crisis (1999), and the latest figure (2004). Three general features are worth emphasizing, although they apply especially to the Northeast. First, unemployment is higher at young age than at old age across all regions. Second, young workers are subject to the largest movements in unemployment rates. Employers adjust to changes in the economic outlook foremost by reducing intakes and increasing dismissals of young workers. Northeast unemployment rates for the age group of 15 to 35 year-old are the highest in the country. Third, young workers have not yet fully recovered from the Asian crisis. Across all regions, entry to wage jobs has become more difficult for the 15 to 20 year-olds. Unemployment for the 15 year-old labor market entrants in the Northeast is still around 50 percent higher in 2004 than in 1996. Figure 51: Regional Unemployment Rates and Educational Attainment by Age (%) A. Unemployment, February 1991 B. Unemployment, February 1996 20 20 ) %( 15 ) e %( 15 atRt et Ra en m 10 nte oy pl myolp 10 emn U 5 em Un 5 0 0 15 20 25 30 35 40 45 50 55 60 Years of Age 15 20 25 30 35 40 45 50 55 60 Years of Age Center Northeast North South Bangk ok Center Northeast Nor th South Bangkok C. Unemployment, February 1999 D. Unemployment, February 2004 20 20 ) ) %( et 15 %( 15 et Rat en Ratne m 10 m 10 oylp oylp em em Un 5 Un 5 0 0 15 20 25 30 35 40 45 50 55 60 15 20 25 30 35 40 45 50 55 60 Years of Age Years of Age Center Northeast North South Bangkok Center Northeast North South Bangkok 72 Wage Employment and Skills To understand the changes in the labor market since the early 1990s, this section takes a closer look at the characteristics of the employed ("workers"). With economic development, more people leave family work, primarily in agriculture, to take up wage employment. In fact, if people talk about jobs, they mean jobs that pay a wage. The share of Northeast workers who receive wages at the age of 35, increased from around 30 percent in 1991 to 45 percent in 1996 before dropping off to just under 40 percent in 2004 (Figure 52.A). Among workers aged 15 years or older, still two-thirds are in non-wage employment in Northeast. Up to the mid-1990s, the Northeast managed to close the gap to other regions but it widened again relative to Bangkok and Central after the Asian crisis. The share of workers earning monthly salaries is yet smaller, and increased by less, than the share of wage workers. It rose from around 9 percent in 1991 to 13 percent in 1996 and to 15 percent in 2004 in the Northeast (Figure 52.B). This is the lowest percentage in the country, and compares to over one in two workers in Bangkok. Similar to wage employment, the gradients at young age became steeper between 1996 and 2004. Overall, between 1991 and 2004, the average age increased from 31 to 34 for a wage worker and from 33 to 35 for a monthly salaried worker. While increases in wage employment slowed after the Asian crisis, the education profile of the workers improved throughout the period. Figure 52.C compares education attainment for workers, wage workers and monthly wage workers between the Northeast and the rest of the country. Four points stand out. First, clearly, rising school enrollment has led to a better skilled labor force across the country. For example, the share of workers with primary education or less dropped from 90 percent in 1991 to 70 percent in 2004 in the Northeast, and from 77 percent in 1991 to 68 percent in the rest of the country. Second, wage workers have better education than workers, and monthly wage workers have higher degrees than wage workers, again both in and outside the Northeast. For example, almost one in two worker on a monthly payroll had vocational or university education in the Northeast, compared to one in five wage workers and one in ten workers. Third, the educational attainment has increased faster after the Asian crisis for all employment categories. For example, the share of monthly wage workers with vocational or university degrees increased by 10 percent between 1996 and 2004, after having declined by 3 percent between 1991 and 1996. This suggests that employers have become more demanding in terms of skill requirements. Finally, while the education profile of the Northeast is worse for workers and wage workers than the rest of the country, this is not the case for wage workers paid monthly. The only difference is that there are more workers with vocational education rather than university degrees. Workers with vocational education account for one third of all monthly salaried workers in the Northeast, compared to a quarter in the rest of the country. This finding squares with a generally worse education profile in the Northeast, as monthly salaried jobs are much harder to come by in the Northeast than in the Center or Bangkok. 73 Figure 52: Employment Composition A. Wage Employment as Percentage of Overall Employment by Region; February 1991, February 1996 and February 2004 80 80 80 ) ) %(t 60 %( )%( 60 en nte m oypl myo 60 entm mE 40 pl oypl m Ed 40 mE 40 aged gea geda W 20 W W 20 20 0 0 0 15 20 25 30 35 40 45 50 55 60 15 20 25 30 35 40 45 50 55 60 15 20 25 30 35 40 45 50 55 60 Years of Age Years of A ge Years of Age Center Northeast North South Bangkok Center Northeast North South Bangkok Center Northeast North South Bangkok B. Monthly Wage Employment as Percentage of Overall Employment by Region; February 1991, February 1996 and February 2004 80 80 80 ) ) %( %( )%( ent 60 ent 60 nte 60 m m oypl oypl myolp mEd 40 mEd 40 mEd 40 age age gea Wyhl 20 Wyhl Wy 20 ont ont M M hlntoM 20 0 0 0 15 20 25 30 35 40 45 50 55 60 15 20 25 30 35 40 45 50 55 60 15 20 25 30 35 40 45 50 55 60 Years of Age Years of Age Years of Age Center Northeast North South Bangkok Center Northeast North South Bangkok Center Northeast North South Bangkok C. Educational Attainment of Workers (E), Wage Workers (W), and Monthly Wage Workers (M) of the Northeast and the Rest of Thailand (%); February 1991, February 1996 and February 2004 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% E E E W W W M M M E E E W W W M M M F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F96 F04 << Northeast | Rest of Thailand >> None Less than Primary Primary Lower Secondary Upper Secondary Vocational University 74 Occupation The increases in education attainment, (modest) rise in wage employment shares and the reduction in seasonality at the microeconomic level, and the reduction in agricultural value- added at the macroeconomic level all point to changes in the occupational structure. Figure 53 shows the employment shares by sector of occupation, separately for each region, as before for 1991, 1996 and 2004. Thailand has experienced a substantial adjustment across sectors. First, the importance of agriculture as job provider has declined across the country. The Northeast has seen the largest reduction of 20 percent, followed by the North, with around 15 percent, and the South, with around 12 percent. Remarkably, the Northeast had in February 2004 a smaller employment share in agriculture than the South. Nevertheless, agriculture remains the dominant employer, even during the off-season, still providing jobs to more than 45 percent of Northeast workers. Second, as expected, the importance of agriculture declines as we restrict attention to wage employment. For the Northeast, just under one quarter of wage workers, and less than five percent of monthly wage workers, are in agriculture. However, a revival of agriculture has taken place since the Asian crisis in terms of wage employment. In all regions outside Bangkok, wage employment shares in agriculture increased between 1996 and 2004. The South increased during this period agricultural wage jobs by a remarkable 20 percent. This suggests that commercial farming has done well over this period. Third, the key sector that provides monthly wage jobs in the Northeast, North or South is services rather than industry. About four-fifths of all monthly salaried position in the Northeast are provided by wholesale, retail, social and other services. Overall, the labor market analysis shows that the employment and skill structures are different across regions. The Northeast typically fares worse than other regions. However, while these trends are persistent until today, employment has become older, more educated and somewhat more often wage paying across all regions since the Asian crisis. In addition, the Northeast narrowed the gap to other regions. Figure 53: Occupational Structure by Region of the Employed (E), Wage Workers (W) and Monthly Wage Workers (M) (%), February 1991, 1996 and 2004 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% E E E W W W M M M E E E W W W M M M E E F96 E F04 W W W M M M E E F96 E F04 W W W M M M E E F96 E F04 W W W M M M F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F91 F96 F04 F91 F96 F04 F91 F91 F96 F04 F91 F96 F04 F91 F91 F96 F04 F91 F96 F04 << NORTHEAST | << NORTH >> | SOUTH >> << CENTRAL | BANGKOK >> Agriculture Manufacturing Other Industry Commerce Transport & Comm'tn Other Services Agriculture Manufacturing Other_Industry Commerce TranspComm Other 75 Wages Perhaps the most important aspect of a job is the wage it pays. Figure 54.A shows average real wages received over the last month, including bonuses and overtimes pay, separately for workers on a monthly payroll, daily payroll and overall average. Northeast wages increased around 40 percent in real terms from 1991 to 1996, boosted by strong economic growth, and changed little from 1996 to 2004, reflecting the combined impact of Asian crisis and recovery. Monthly wages are about 140 to 160 percent higher in the Northeast than daily wages. Northeast wages were between 60 to 90 percent lower in 1991 than in Bangkok but the gap dropped to around 50 percent by 2004. The Northeast wage gap declined also relative to the other regions. Figure 54.B shows monthly wages separately for those paid monthly and those paid daily. The spatial and time trends are similar those of monthly wages. Differences in daily and monthly wages reflect foremost differences in skills and sector. Figure 54.C and Figure 54.D break down wages of monthly wage workers by education and occupation. Wages increase with higher education. A Northeast university graduate earned in the early 1990s almost 250 percent more than a worker with primary education or less. However, wages of university graduates in 2004 were still below the 1996 levels, and the wage premium fell to less than 200 percent. Upper secondary education earned lower wages than lower secondary education. The highest wages are earned in the non-commerce service sector in the Northeast. Agricultural wages have risen above those earned in commerce since 1996. Figure 54: Wages in February 1991, February 1996, and February 2004 A. Wages during the last 30 Days B. Monthly and Daily Wage 9000 8000 160 8000 7000 140 7000 6000 120 6000 5000 100 5000 4000 4000 80 3000 3000 60 2000 2000 40 1000 0 1000 20 F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F96 F04 << NORTHEAST | << NORTH >> | << SOUTH >> | << CENTRAL >> | BANGKOK > 0 0 All Monthly Daily F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F96 F04 << NORTHEAST | << NORTH >> | << SOUTH >> | << CENTRAL >> | BANGKOK >> Monthly (LHS) Daily (RHS) C. Monthly Wage Worker Wages by Education D. Monthly Wage Worker Wages by Occupation 18000 14000 16000 12000 14000 10000 12000 8000 10000 8000 6000 6000 4000 4000 2000 2000 0 0 F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F96 F04 F91 F96 F04 << NORTHEAST | << NORTH >> | << SOUTH >> | << CENTRAL >> | BANGKOK >> << NORTHEAST | << NORTH >> | << SOUTH >> | << CENTRAL >> | BANGKOK > Agriculture Manufacturing Other_Industry Commerce TranspComm Other Services Primary or Less Lower Secondary Upper Secondary Vocational University 76 Returns This section takes the analysis of wages and education one step further. It estimates returns to education over time using Mincer regressions. In its most basic form, log hourly wages are regressed on education, controlling for age, aged squared, gender, province and urbanization.20 Figure 55.A to Figure 55.E show the coefficients on education levels from a series of such cross-sectional regressions for monthly wage workers. Education is divided into six attainment levels (less than primary, primary, lower secondary, upper secondary, vocational and university) and the returns are estimated relative to workers with less than primary education, the omitted category. The results confirm the wage analysis. Across all regions, education, and especially vocational and university education, yields a wage premium. The returns to education increased up to the mid-1990s and tended downwards since then. Furthermore, wage differentials tended to increase up to the crisis and have fallen since then. The Northeast stands out in two ways. First, the differences in the returns by education level are larger than in other regions. Second, the premium for vocational education is 200 percent larger than upper secondary education and only 50 percent smaller than university education. The corresponding gaps in Bangkok are 60 percent and 180 percent, respectively. We can verify these findings using years of schooling an alternative measure of education attainment, drawing on the Socio-Economic Survey. Among monthly wage workers in the Northeast, another year of education increased monthly wages by 15 percent in 2002, compared to 18 percent in 1988 (Figure 55.F). The wage premium of another year of education is highest in the Northeast, followed by the North, Center, South and Bangkok.21 The ranking is constant over time, apart from switching between the Center and South. The labor market signals the high value of education. This is not surprising. More remarkable is the lack of increases in the value of education over the last decade in spite of the export boom, and not just in the Northeast, but also Bangkok and the Center, Thailand's manufacturing centers. Traditionally, the low educational attainment of the majority of Thai workers appeared to be a serious constraint for Thailand to become a producer of technology-intensive products, as Thailand may loose its comparative advantage in labor- intensive industries compared to countries such as Lao PDR, Vietnam, and China that still have low labor costs. Yet, the lack of rising returns to education casts doubts over this hypothesis. The next sections will look at different explanations for this finding. 20The calculation yields, under certain conditions, private returns. Social returns would require incorporating the public costs of providing education net of any external effects (through endogenous growth other channels) of education (Psacharopoulos 1995). The analysis makes no attempt at measuring these effects, implicitly assuming that they have not changed differentially across skill levels over time. 21The returns to education levels are calculated using Kennedy (1981). Blunch (2004) reports similar estimates using SES data from 1994 to 2002. He notes that returns are overall lower with household fixed-effects, although the gradient remains similar. This comes at a cost of dropping household weights and households with one wage worker. Similarly, Hawley (2004) finds with LFS data from 1985, 1995 and 1998 that for the group of 24 to 35 year-old wage workers an additional year of schooling provides has a return of 11 percent. The return remained stable over the period. 77 Figure 55: Returns to Education of Monthly Wage Earners Relative to Less than Primary A. Northeast, 1991 to 2004, LFS B. North, 1991 to 2004, LFS 600 600 550 550 500 500 450 450 400 400 350 350 300 300 250 250 200 200 150 150 100 100 50 50 0 0 F91 A91 F92 A92 F93 A93 F94 A94 F95 A95 F96 A96 F97 A97 F98 A98 F99 A99 F00 A00 F01 A01 F02 A02 F03 A03 F04 A04 F91 A91 F92 A92 F93 A93 F94 A94 F95 A95 F96 A96 F97 A97 F98 A98 F99 A99 F00 A00 F01 A01 F02 A02 F03 A03 F04 A04 Primary Lower Secondary Upper Secondary Vocational University Primary Lower Secondary Upper Secondary Vocational University C. South, 1991 to 2004, LFS D. Center, 1991 to 2004, LFS 600 600 550 550 500 500 450 450 400 400 350 350 300 300 250 250 200 200 150 150 100 100 50 50 0 0 F91 A91 F92 A92 F93 A93 F94 A94 F95 A95 F96 A96 F97 A97 F98 A98 F99 A99 F00 A00 F01 A01 F02 A02 F03 A03 F04 A04 F91 A91 F92 A92 F93 A93 F94 A94 F95 A95 F96 A96 F97 A97 F98 A98 F99 A99 F00 A00 F01 A01 F02 A02 F03 A03 F04 A04 Primary Lower Secondary Upper Secondary Vocational University Primary Lower Secondary Upper Secondary Vocational University E. Bangkok, 1991 to 2004, LFS F. Return to another Year of Education, 1988 to 2002, SES 600 20 550 18 500 16 450 14 400 12 350 10 300 8 250 200 6 150 4 100 2 50 0 1988 1990 1992 1994 1996 1998 2000 2002 0 F91 A91 F92 A92 F93 A93 F94 A94 F95 A95 F96 A96 F97 A97 F98 A98 F99 A99 F00 A00 F01 A01 F02 A02 F03 A03 F04 A04 Northeast North South Central Bangkok Primary Lower Secondary Upper Secondary Vocational University 78 Supply and Demand One explanation is that the supply of educated workers has increased so quickly that it has outpaced rising demand of education, and hence suppressed returns to education. Clearly, education levels increased strongly across all regions and age groups since the early 1990s (Figure 56). The left column of Figure 57 shows relative labor supply for four adjacent skill groups (university to vocational; vocational to upper secondary; upper secondary to lower secondary; and lower secondary to primary). As schooling levels of young new entrants into the labor market increased, relative labor supply improved. Relative labor supply increased by between 30 to 100 percent over the entire period. The only exception is vocational education relative to upper secondary education, which decreased by 40 to 50 percent. Katz and Murphy (1992) developed a framework to derive relative labor demand from wage and labor supply trends. Following the methodology laid out in Sanchez-Paramo and Schady (2003), the right column of Figure 57 shows relative labor demand for a constant elasticity of substitution of 2 across skill levels.22 Compared to the early 1990s relative labor demand declined for university, upper secondary and lower secondary education. In other words, only relative demand for vocational degrees increased. The same hold relative to 1996, with the exception that labor demand for university relative to vocational degrees remained constant in the Northeast.23 These findings contrast with evidence from Malaysia (Tan and Gill 1998) and trade-oriented economies in Latin America (Sanchez-Paramo and Schady 2003). The lack of rising labor demand for higher skill levels suggests that Thailand does not engender the kind of rapid technological progress that would fuel wage growth in the formal sector. Figure 56: Population Aged 15 to 60 years-old with at least Lower Secondary Education (%) A. February 1991 B. February 2004 80 ) ) 80 %( %( ero ero M M 60 60 or or noita noita ucdE 40 ucd y ardnoceSr 20 EyradnoceSr 40 20 we we Lo Lo 0 0 1 5 2 0 2 5 3 0 35 40 45 50 55 60 15 2 0 2 5 30 3 5 4 0 45 50 5 5 60 Y ea rs of A g e Y ea rs of A g e Ce nte r N orthe as t No rth S ou th B a ng ko k Ce nter N orth ea st N o rth S ou th B an gk ok 22Following Katz and Murphy (1992) and Sanchez-Paramo and Schady (2003), relative labor demand is backed out from relative wages and relative labor supply from a simple supply-demand model under the assumption of a constant elasticity of substitution of 2. The economy is assumed to be operating on the demand curve, and labor supply is taken to be inelastic in the short run. In order to net out compositional changes in the labor force, each cross-section is reweighed to replicate the average gender-age (in five year brackets) structure over the entire period. In practice, the reweighing makes little difference to the findings. Assuming elasticities of unity or 3, as common in the literature, produces similar results. 23Using LFS data from 1989 to 1995, Moenjak and Worswick (2003) find that at the upper secondary level, vocational education has a higher return than general education. 79 Figure 57: Relative Labor and Relative Labor Demand of Monthly Wage Workers (Substitution Elasticity of 2), 1991 to 2004 Northeast Supply Northeast Demand 7.0 3.0 6.5 6.0 2.5 5.5 2.0 5.0 4.5 1.5 4.0 3.5 1.0 3.0 2.5 0.5 2.0 1.5 0.0 1.0 0.5 -0.5 0.0 F91 A91 F92 A92 F93 A93 F94 A94 F95 A95 F96 A96 F97 A97 F98 A98 F99 A99 F00 A00 F01 A01 F02 A02 F03 A03 F04 A04 -1.0 F91 A91 F92 A92 F93 A93 F94 A94 F95 A95 F96 A96 F97 A97 F98 A98 F99 A99 F00 A00 F01 A01 F02 A02 F03 A03 F04 A04 North Supply North Demand 7.0 3.0 6.5 2.5 6.0 5.5 2.0 5.0 4.5 1.5 4.0 3.5 1.0 3.0 0.5 2.5 2.0 0.0 1.5 1.0 -0.5 0.5 -1.0 0.0 F91 A91 F92 A92 F93 A93 F94 A94 F95 A95 F96 A96 F97 A97 F98 A98 F99 A99 F00 A00 F01 A01 F02 A02 F03 A03 F04 A04 F91 A91 F92 A92 F93 A93 F94 A94 F95 A95 F96 A96 F97 A97 F98 A98 F99 A99 F00 A00 F01 A01 F02 A02 F03 A03 F04 A04 South Supply S o u th D e m a n d 7.0 3 .0 6.5 6.0 2 .5 5.5 2 .0 5.0 4.5 1 .5 4.0 3.5 1 .0 3.0 2.5 0 .5 2.0 0 .0 1.5 1.0 -0 .5 0.5 0.0 -1 .0 F91 A91 F92 A92 F93 A93 F94 A94 F95 A95 F96 A96 F97 A97 F98 A98 F99 A99 F00 A00 F01 A01 F02 A02 F03 A03 F04 A04 F 9 1 A 9 1 F 9 2 A 9 2 F 9 3 A 9 3 F 9 4 A 9 4 F 9 5 A 9 5 F 9 6 A 9 6 F 9 7 A 9 7 F 9 8 A 9 8 F 9 9 A 9 9 F 0 0 A 0 0 F 0 1 A 0 1 F 0 2 A 0 2 F 0 3 A 0 3 F 0 4 A 0 4 Center Supply Center Demand 7.0 3.0 6.5 6.0 2.5 5.5 2.0 5.0 4.5 1.5 4.0 3.5 1.0 3.0 2.5 0.5 2.0 0.0 1.5 1.0 -0.5 0.5 0.0 -1.0 F91 A91 F92 A92 F93 A93 F94 A94 F95 A95 F96 A96 F97 A97 F98 A98 F99 A99 F00 A00 F01 A01 F02 A02 F03 A03 F04 A04 F91 A91 F92 A92 F93 A93 F94 A94 F95 A95 F96 A96 F97 A97 F98 A98 F99 A99 F00 A00 F01 A01 F02 A02 F03 A03 F04 A04 Bangkok Supply Bangkok Demand 7.0 3.0 6.5 6.0 2.5 5.5 2.0 5.0 4.5 1.5 4.0 3.5 1.0 3.0 2.5 0.5 2.0 0.0 1.5 1.0 -0.5 0.5 0.0 -1.0 F91 A91 F92 A92 F93 A93 F94 A94 F95 A95 F96 A96 F97 A97 F98 A98 F99 A99 F00 A00 F01 A01 F02 A02 F03 A03 F04 A04 F91 A91 F92 A92 F93 A93 F94 A94 F95 A95 F96 A96 F97 A97 F98 A98 F99 A99 F00 A00 F01 A01 F02 A02 F03 A03 F04 A04 Lower Secondary to Primary Upper Secondary to Lower Secondary Lower Secondary to Primary Upper Secondary to Lower Secondary Vocational to Upper Secondary University to Vocational Vocational to Upper Secondary University to Vocational 80 Unions and Minimum Wage Another potential reason for the modest wage increases could be weak labor market institutions that protect workers' interests. The next two sections look briefly at three dimensions, namely unionization, minimum wages and labor protection legislation. The conclusion is that these institutions are weak and have little bearing on the functioning of the labor market. However, these weaknesses are a long standing feature of Thai labor markets and cannot account for the change in wage trends after the Asian crisis. For the labor market to promote prosperity, it should assure that demand for labor is met by supply at a price that is acceptable for both parties. This coincidence of wants is what creates wage employment. Most labor protection aims to enhance job security by making dismissal costly to the employer. By the same token, it also can have the unintended effect of making hiring more costly. While a fully fledged discussion of these institutions and their impact on wages and employment is beyond the scope of this report, the focus is on coverage and enforcement. First, labor unions play a negligible role in Thailand. The 1975 Labor Relations Act 2518 provided private and state enterprise employees the right to form labor unions. In 1991, the right was withdrawn from state sector employees, before it was granted again in the 2000 State Enterprise Employees Relations Act (Chandoevwit 2004). According to the latest available survey data, only 1 percent of private sector workers were members of a labor union in the Northeast, and only 1.5 percent worked in a firm with a labor union (Figure 58.A). Even in Bangkok, the corresponding numbers (5.7 percent and 8.2 percent, respectively) are still low in international perspective.24 Second, the minimum wage has fallen in real terms over the last decade and is not enforced. The minimum wage was introduced in Bangkok in 1972 and nationwide in 1974. There were 14 minimum wage levels in 2004, varying at the province level and ranging from a daily rate of 133 to 170 baht. While nominal minimum wage rates are increasing, in real terms the minimum wage has fallen by around 2 percent per year since 1996 (Figure 58.B). Nevertheless, minimum wages are relatively high compared to wages paid in some regions. Assuming that the median monthly wage is not affected by minimum wage levels, the toughness index is defined as the ratio of the minimum wage over the median monthly wage. Minimum wages are much "tougher" in the Northeast and North, where the index is over 90 percent, compared to Bangkok, where the index is no higher than 65 percent (Figure 58.C). At the same time, compliance with minimum wages among daily wage workers, who have lower wages than monthly wage workers, is low. Defining non-compliance as a daily wage level less than 95 percent of the province-specific minimum wage, then non-compliance is as high as 55 percent in the Northeast and declining, compared to no more than 15 percent in Bangkok (Figure 58.D). Most likely, this is not a reflection of differential enforcement but simply of differences in the wage scale. 24Unionization rates are 9 percent in Malaysia, 11 percent in the Philippines and in South Korea, and 17 percent in Brazil. 81 Figure 58: Unionization and Minimum Wages A. Unionization of Workers and Firms with Labor B. Real Minimum Wage, February 1994 to 2004 Unions, 1998 (Percent) 10 4,200 4,000 8 3,800 6 3,600 4 3,400 3,200 2 3,000 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 0 Union Member Firm with Union Bangkok Center North Northeast South Bangkok Center North Northeast South C. Ratio of Minimum Wage over Median Monthly D. Non-Compliance with Minimum Wage Levels Wage, February 1994 to 2004 for Daily Wages, February 1994 to 2004 110 80 70 95 60 50 40 80 30 20 65 10 0 50 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Bangkok Center North Northeast South Bangkok Center North Northeast South 82 Labor Protection Legislation Labor protection laws cover only a part of the labor market and enforcement is weak. Key legislation includes the 1990 Social Security Act and the 1994 Workmen's Compensation Act. The enforcement of severance pay during the Asian crisis as stipulated the Labor Protection Act was very low. In 1998, only one in 20 laid-off workers in firms with less than 10 workers, one in five laid-off workers in firms with 10 to 99 workers, and one in two laid- off workers in firms with more than 100 workers received severance pay. More recently, the 1975 labor protection law was amended in 1998.25 It protects workers in terms of the general right of employees, working hours, wages and other payments, and other aspects. The 1998 Labor Protection Act applies only to private employees and employers in the industrial and service sector. Including government and state-enterprise employees, the covered sector comprises around 14 million people.26 Excluded are the self-employed, agricultural workers and unpaid family workers, which amounted to 19.5 million or about 60 percent of all workers in 2004. While almost 99 percent of the monthly wage workers are either covered by the 1998 Labor Protection Act or employed in the public sector, the share drops to 70 percent for other wage workers and 6 percent for non-wage workers. The Northeast accounts one fifth of the covered and public sector workers, but two fifths of the uncovered workers. Figure 59: Covered, Public and Uncovered Sectors according to the 1998 Labor Protection Act, Northeast and Rest of Thailand, 1991 to 2004 36,000,000 30,000,000 24,000,000 18,000,000 12,000,000 6,000,000 0 F91 A91 F92 A92 F93 A93 F94 A94 F95 A95 F96 A96 F97 A97 F98 A98 F99 A99 F00 A00 F01 A01 F02 A02 F03 A03 F04 A04 NE_Covered NE_Uncovered NE_Public ROT_Covered ROT_Uncovered ROT_Public 25In 2004, unemployment insurance was introduced, covering about 10 million people. 26Garen and Jeraputtiruk (2005) and Jeraputtiruk (2004) discuss in detail the provisions of the 1998 Labor Protection Act. They estimate that it increased labor costs by 5.8 percent in large covered firms and by 0,8 percent in small covered firms. Using a two-sector model of covered and uncovered sectors, they analyze the impact of the legislation on wages and employment comparing the pre-act period of 1994 to 1997 to the post- act period of 1999 to 2000. As predicted by theory under elastic labor demand, the imposition of labor protection leads to mobility out of the covered sector and a rising wage gap between covered and uncovered sectors. However, as they acknowledge themselves, the results are likely to be contaminated by the Asian crisis, which lead to a movement of workers into the informal sector. 83 Migration Workers migrate in search of high wages. Within-year variation in agricultural employment and better wage job opportunities have for a long time resulted in large inter-regional migration. Generations of Isan people have left their villages to seek employment in the service sector in Bangkok, in manufacturing in the Eastern Seabord or in the tourist industry along the coast. The demand for low-skilled migrants in these regions is fuelled due to economic expansion and the diminished appeal of physical labor among the well-education local workforce. Villagers embrace migration reluctantly for economic reasons. According to the 2001 NRD2C, almost one in two Northeast villages report many problems with migration. This is the most pressing concern, aside from dry season farming, and compares to ratios of one in four to one in six in other regions. Figure 60.A decomposes the wages differences between the Northeast and the rest of the country as of February 2004 into two parts. The first component tells us about the average wage premium due to higher returns for worker characteristics. The second component gives us how much of the difference is due to better characteristics for a given return. If the first component dominants the difference, then higher wages outside the Northeast are due to better returns to worker characteristics rather than better characteristics as such. This is indeed the case: the bulk of the monthly and daily wage differences are accounted by higher returns outside the Northeast. Among monthly wage workers, older and better educated do better outside the Northeast, while among daily wage workers, younger worker do better and education makes little difference. The Northeast and the North move a larger share of the workforce (turnover between 10 to 16 percent compared to a national average of 7 to 8 percent) across sectors and regions than other areas.27 According to the 2000 Population Census, 18 percent of women and 15 percent of men born in the Northeast no longer reside there. This is about 4 percent higher than the shares in the North, and 10 percent to 12 higher than the shares in the South. Migration affects mostly young prime aged adults. This leads to a twin-peak population structure in the Northeast, with many children and adults of 30 years or older, and a single-peak structure in Bangkok, with a high concentration of 20 to 35 year-old (Figure 60.B). Aside from the young, mobility among other population groups in the Northeast is not particularly high. High shares of land ownership and agricultural rice farming lead to a sedentary lifestyle. Among the population that stayed in the Northeast, about two thirds stayed always in the same amphoe (district), and less than one in ten moved in the last decade. Other regions show higher between-amphoe mobility, with the highest incidence observed in the Center and Bangkok. Figure 60.C shows an increase in mobility during 2000. This is likely to be linked to return migration during the Asian crisis. As firms laid off especially the young and low-skilled in Bangkok, they returned to their home villages, triggering an increase in Northeast unemployment above Bangkok levels (Box 6). 27A 1995 survey by the National Statistical Office revealed that there were about 1 million migrant from the Northeast working in Bangkok. This accounted for about 40 percent of the whole country's migration of about 2.5 million. 84 Figure 60: Migration A. Oaxaca Decomposition February 2004: Rest of Thailand vs. Northeast 120 110 100 90 80 70 60 50 40 30 20 10 0 -10 Returns Characteristics Returns Characteristics -20 -30 -40 Monthly Wages | Daily Wages Lower Secondary. Upper Secondary. Vocational. University Age B. Population Pyramids in the Northeast and Bangkok, 2002 Age Pyramid, NE Region 2002 Age Pyramid, Bangkok 2002 90 90 80 80 70 70 oup oup grega 60 gr 60 50 gea 50 eary 40 eary 40 eviF 30 eviF 30 20 20 10 10 0 0 7.5 5 2.5 0 2.5 5 7.5 7.5 5 2.5 0 2.5 5 7.5 Female Male Female Male Percent of Population Percent of Population C. Amphoe Mobility in the Northeast and in Bangkok, Percent of Population, 1988 to 2002 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 1988 1990 1992 1994 1996 1998 2000 2002 1988 1990 1992 1994 1996 1998 2000 2002 Always in the same More than 10 years Less than 10 years Always in the same More than 10 years Less than 10 years 85 Box 6: Return Migration and the Asian Crisis Movement of workers across regions and sectors is an important adjustment mechanism to changes in the macroeconomic situation. This clearly happened during the Asian crisis. Even though Bangkok-based enterprises and financial institutions were at the core of the crisis, the Northeast was the hardest hit regions in terms of employment (World Bank 2000). Employment dropped about 4.2 percent from 1996/7 to 1998/9, compared to 2.6 percent in the North and increases of 0.6 percent in Central, 2.2 in the South and 2.4 percent in Bangkok. Aggregate wage earnings fell by 8 to 9 percent in the Northeast and Central, compared to 4 percent in the South, no change in the North, and an increase of 3 percent in Bangkok. During the first quarter of 1997, just before the crisis, 531,000 workers in agriculture moved to construction and 130,000 moved to the services sector. After the onslaught of the crisis in the first quarter of 1998, the net flow of workers out of agriculture declined by almost three fifths of the 1997 level in construction, and 100 percent in services. Clearly, reduction in labor flows out of agriculture was an important adjustment during the crisis. Return migration to rural areas is a reflection of the same phenomenon. In the first quarter of 1997, about 623,000 individuals reported having returned from Bangkok during the last year. This could be temporary migrants who move to Bangkok for contract work and return home upon completion of their contracts. This number increased in the first quarter of 1998 to 852,000 workers. 86 Remittances Migration affects the Northeast not just through its impact on the demography. Perhaps the most important effect is through remittances.28 Figure 61.A shows the differences in household income sources across regions. The Northeast has the lowest level and share of wages and salaries; the lowest level and share of non-farm profits; the highest share of non- monetary income; and, outside of Bangkok, the highest level and share of private transfers. The last feature points to remittances provided by migrant family members. But how large are these remittances? Across all regions, the share of households receiving remittances increased between 1996 and 2002 (Figure 61.B). The proportion is highest in the Northeast. More than one in two households benefited from such payments in 2002, compared to around 45 percent in 1996. Among receiving households, these remittances amount to around one third of household income in the Northeast, Center and North (Figure 61.C). With the exception of Bangkok, these income shares increased between 1996 and 2002, even though more households became recipients. Remittances lower poverty across all regions, and the reduction is largest in the Northeast. The poverty headcount in households without remittances is about 5 percent higher than in households without remittances (Figure 61.D). For 2002, this amounts to a drop in poverty by almost one third. Can we be sure that the causality runs from remittances to lower poverty rather than the other way around? We can get an indication by comparing characteristics of households by remittance status. Figure 61.E looks at the education of the household head. Clearly, non- receiving household heads are better educated than household heads who receive remittances. Similarly, non-receiving households are more likely to live in urban areas than receiving households. Those characteristics would suggest higher poverty among receiving households, yet we find the opposite. This suggests that remittances help receiving households to escape poverty. Overall, labor migration and remittances make the Northeast labor market as well as household living standards dependent on the rest of Thailand. Prosperous Bangkok and Center regions boost labor demand for Northeast workers who with their improved skill levels are in a better position to access well-paid jobs than in the past. 28Another important effect is the experience and assets that migrants bring to the Northeast if and when they return to their communities. 87 Figure 61: Household Remittances A. Household Income Sources, 2002 1 0 0 % 9 0 % 8 0 % 7 0 % 6 0 % 5 0 % 4 0 % 3 0 % 2 0 % 1 0 % 0 % B a n g k o k / V i c i n i t y C e n t e r N o r t h N o r t h e a s t S o u t h W a g e s a n d s a l a r i e s N o n - f a r m p r o f i t s F a r m I n c o m e T r a n s f e r s P r o p e r t y I n c o m e N o n - m o n e t a r y i n c o m e B. Share of Households Receiving Remittances (%), C. Remittances as Percent of Household Income 1996 and 2002 among Receiving Households, 1996 and 2002 60 40 50 30 40 30 20 20 10 10 0 0 Bankgkok Vicinity Center Northeast North South Bankgkok Vicinity Center Northeast North South 1996 2002 1996 2002 D. Poverty Headcount by Remittance Status (%), E. Educational Status of Household Head by 1996 and 2002 Remittance Status (%), 2002 20 100% 90% 15 80% 70% 10 60% 50% 5 40% 30% 0 20% nkgkok ity rth 10% Vicin Center uth No South Vicinity nter Ce North So Ba Northeast Bankgkok Northeast 0% Without Remittance With Remittance 1996 | 2002 No Education Primary Lower Secondary Without Remittance With Remittance Upper Secondary Vocational Tertiary 88 Students Access Education is an end in itself and a vital part of individuals' capacity to lead lives they value. More education is associated with better family health and participation in society, higher productivity of farmers, workers and small-business owners alike, and thus lower poverty. The society as a whole benefits from an educated population. A skilled workforce contributes to higher economic growth, as East Asia's experience over the last decades demonstrates. If education is broadly shared across the population, chances are that growth will be as well. School participation rates have increased impressively over the last 15 years.29 Thailand's primary school enrollments were already high in the early 1990s and have reached near-universal levels today. Secondary education outcomes were relatively low in early nineties but have shown noticeable improvements over the years. This trend is in line with the requirement of 9 years of mandatory schooling from 2004 onwards, as stipulated by the 1999 National Education Act. School participation rates for 14-year old children increased by almost 30 percentage points from less than 50 percent in 1988 to over 80 percent in 2002 (Figure 62.A). Remarkably, progress in the Northeast was even faster than in other regions, increasing from less than 40 percent in 1988 to over 80 percent in 2002. The enrollment gap has been eliminated for all age groups up to upper-secondary education (Figure 62.B). Even for students aged 18 to 21 years, the gap in the school participation rate dropped from 14 percent in 1992 to 3 percent in 2002. Access improved for children from poor and non-poor households alike Figure 62.C). Over the medium to long-run, the equalization of educational opportunities will be an important driver for regional convergence. Higher school participations have substantially improved the skill levels of the young entrants of the job market. Among the 15 to 21 year-old, the share of Northeast students with at least lower secondary education increased from just over one half to more than 70 percent. In 1988, less than 5 percent attended upper secondary, vocation, or university education. By 2002, this share increased to over one quarter (Figure 62.D). Nevertheless, the Northeast continues to lag behind in access to vocational and university education. The share of vocational and university students are only 1 and 7 percent in the Northeast, respectively, compared to 5 and 11 percent in other regions. 29The education system is organized into 6 years of primary education, plus 3 years in lower secondary, 3 years in upper secondary and 4 years of higher education. Thailand's education/age groups are grades 1-6 and 6-11 years old for primary school , grades 7-9 and 12-14 years old for lower secondary, grades 10-12 and 15-17 years old for upper secondary level. The statistics are based on the SES. The School Participation Rate (SPR) is defined as the proportion of a given age-group children that are enrolled in school regardless of their schooling level. Due to a modification in the SES questionnaire, net and gross enrollment rates from 2002 are not comparable to prior years. Vocational education covers lower, upper, and higher education. University education includes teacher training. 89 Figure 62: School Participation Rate, 1988 to 2002 A. School Participation Rate: Northeast and Rest of Thailand 1 1 .8 .8 .6 .6 .4 .4 .2 .2 0 0 6 10 15 20 25 6 10 15 20 25 Years of Age Years of Age 88 90 92 94 88 90 92 94 96 98 00 02 96 98 00 02 B. School Participation Rate: Gap of the Northeast to C. School Participation in the Northeast among 6 to 21 Year- the Rest of Thailand by Age Group Old: Poor versus Non-poor Students 8 100 Age 6 to11 Age 12 to 14 Age 15 to17 Age 18 to 21 4 90 0 80 -4 70 -8 -12 60 -16 50 1988 1990 1992 1994 1996 1998 2000 2002 -20 Poor Non-Poor -24 1988 1990 1992 1994 1996 1998 2000 2002 D. School Attendance by Level among the 15 to 21 Year-Old: Northeast and Rest of Thailand 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 1988 1990 1992 1994 1996 1998 2000 2002 1988 1990 1992 1994 1996 1998 2000 2002 Lower Secondary Upper Secondary Vocational University Outside School Lower Secondary Upper Secondary Vocational University Outside School 90 Private and Public Spending The 1999 National Education Act lays out an ambitious agenda of improving learning outcomes. The transition to a modern, child-centered and participatory education system is a difficult challenge, especially in the Northeast, which is struggling with the twin constraints of low private and public resources. Private schools play a limited role in Thailand's education system. In 2002, only 4 percent of Northeast pupils attended a private school, compared to between 12 percent and 28 percent in other regions (Figure 63.A). Similar to other regions, private school enrolment matters least at the primary and secondary level, and most at the tertiary and vocational level.30 Costs are one of the reasons why private education is rare in the Northeast. While public education is provided free of charge, a 1999 survey on the demand for education identified financial constraints as the main reason why children aged 12 to 17 did not enroll in lower or upper secondary school. Northeast households spent less on education than other regions. Out-of-pocket expenses of households are substantially higher in private schools than in public schools. Low household income may prevent parents to send children to private schools. In 2002, a Thai student paid on average 6 times as much for attending a private instead of a public lower secondary school, and 5 times as much for private instead of public vocational education (Figure 63.B). Northeast households spent 14 percent of household income on private vocation education. Another reason for the low private sector share might be that public schools provide better education. The RTG has traditionally been committed to fund adequately education. Education is the largest single sector in the budget. It expanded from 21 percent of total central government spending in FY1993 to 22.4 in FY1997, was effectively protected from expenditure cuts during the Asian crisis, and further increased to 23.8 percent in FY2003. This share is comparable or higher to allocations of many high-income countries, such as Korea, Japan and the United States. Education receives an even higher share of Northeast spending of the central government. More than two fifths of government spending go towards education. However, in terms of spending per pupil, the Northeast clearly lacks behind other regions outside of Bangkok. Combining data on public expenditure with data on household access to education from the household survey, the benefit incidence of public spending can be explored. Relative to the non-Bangkok average, the funding gap per pupil is 15 percent in primary, 29 percent in secondary, and 22 percent in tertiary education (Figure 63.C). Furthermore, the bulk of what is spent is allocated by salaries, leaving little to other, quality enhancing inputs. Total salary spending absorbs about two thirds of the recurrent education spending. Salary spending of the Northeast in FY 2002 accounted for about 81 percent of recurrent education spending expenditure on kindergarten/primary, 73 percent on secondary education; 52 percent on tertiary education, and 62 percent on tertiary education. 30The Northeast has 19 public and private institutions for tertiary education, in addition to extension campuses of a number of central and regional state universities. 91 Figure 63: Private and Public Spending on Education, 2002 A. Private School Enrolment by Level (% of Total Enrolment) 70 60 50 40 30 20 10 0 Primary Secondary Tertiary Vocational Northeast BKK/Vic. Central North South B. Household Education on Private and Public Vocational Education: Baht per Month and Share of Income 1600 15 1400 12 1200 1000 9 800 6 600 400 3 200 0 0 Private Public Private Public Bangkok Central North Northeast South Bangkok Central North Northeast South C. Central Government Spending on Education (Baht per Pupil) 80000 70000 60000 50000 40000 30000 20000 10000 0 Primary Secondary Vocational Tertiary Northeast Central North South 92 Efficiency, Targeting, and Quality There is another way of looking at the data: the combination of high enrollment rates and low expenditures suggest a high efficiency of education spending the Northeast, at least for access. Figure 64.A shows such frontiers for per student secondary education spending versus net secondary enrolment.31 Northeast provinces are placed close to the efficiency frontier. A vital part of the RTG's opportunity programs is access to education, supported by several targeted programs. The three most important RTG programs in education are the education loan; school lunch; and government scholarships. The school lunch program supports undernourished children in primary education. The Northeast enjoys better access than other regions, but poor and non-poor children benefit equally. Only students in primary and lower secondary school are eligible for the government scholarship program. The government scholarship program is insignificant, reaching less than 1 percent of all pupils. The Education Loans Fund was set up in 1996 to support destitute students from low-income families in pursuing higher level-education. The student loan scheme is open only to students at the upper secondary, vocational, and tertiary levels. Spending increased from Bt14 bn in FY00 to Bt27 bn in FY01 and to Bt28 bn in FY03. Levels of coverage are very low. About two fifths of student loan recipients are from non-poor households, indicating high levels of leakage. The Northeast does not fare worse in terms of pupil-teacher ratios or pupil per classroom ratios than other regions. They are around 20, low by international standards. However, teachers have generally a lower qualification than in other regions. While children in the Northeast perform similarly to other children at young age, their test scores fall behind as they go through the school system; the gap is widest at Grade 12, where the Northeast scores 1 to 1.5 points less for verbal skills and 4 to 5 points less overall than the other regions (Figure 64.B) (Punyasavatsut and Revenga 2003). Thailand's reform agenda includes guaranteeing 12 years of schooling free of charge; providing 9 years of mandatory schooling from 2004 onwards, to be raise to 12 years from 2015 onwards; operating 175 Local Education Authorities; upgrading teacher training, modernizing curricula to meet international standards and promoting school accountability towards parents. The implementation of the ambitious education reform program will only be possible with a strong commitment by the RTG to education. Reaching universal schooling of 12 years will involve improving the targeting of key education programs, such as the education loan program. Providing quality education will entail attracting motivated and well-qualified teachers, as well as giving teachers adequate tools and technology in the class room. One source for funding such reforms would be to allow for increases in the student-teacher ratios (World Bank 2000). 31Efficiency is measured by calculating the distance between observed input-output bundles and an efficient frontier, defined as the maximum attainable output for a given level of inputs. The frontier is constructed with the commonly used Free Disposable Hull technique. Input inefficiency refers to the excess input consumption to achieve a level of output; and output inefficiency to the output shortfall for a given level of inputs (Herrera and Prang 2005). Out of 11 regional indicators on gross and net primary and secondary enrollment rates averaged, the Northeast has the highest efficiency for 10 indicators. 93 Figure 64: Efficiency and Quality of Public Education Spending A. Efficiency Frontier of Net Secondary Enrollment Rate and Public Secondary Spending Per Pupil in 2000 and 2002 Secondary Enrollment vs Education Expenditure Secondary Enrollment vs Education Expenditure 90 C-RAYON N-UTTAR 80 N-NAN N-SUKHO S-PHUKE N-NAN NE-NONGB C-CHAIN NE-UDONT NE-KALAS NE-NONGB C-SAKAE NE-NAKHO NE-MUKDAN-PHITS NE-ROIET BK-PAT HU N-PHICH S-SATUN NE-ROIETNE-MAHAS N-PHRAE N-UT HAI NE-SAKON N-UTTAR N-KAMNE-YASOT NE-LOEI PH N-PHRAEC-SINGB NE-MAHAS ) NE-SURIN NE- IS-PHAT T NE-NONGKNE-KHONK SA N-PHAYAHAIS-YALA C-PRACHN-LAM PH %( 80 NE-NAKHO 70 NE-SINE-NAKHON-UT S-NAKHO C-PHRAC-NAKHOPA C-ANG THBK-BANGK NE-NAKHO SAC-RATCH N-PHICH N-PHITS BK-NONTH ) C-SAKAE NE-CHAIY S-CHUM PN-LAM C-NAKHO N-PHETC S-PHATT C-SARAB N-PHAYA S-CHUM P NE-BURIR C-SUPHA S-SONGK BK-NONTH NE-BURIR NE-KHONK NE-KALAS C-PHACH N-LAMPH C-TRAT nte NE-SURIN S-SURAT S-TRANG C-PHACH C-LOPBU NE-UBONR NE-NONGK NE-YASOTC-SARAB C-ANGTHC-SINGB S-PHANGN-PHETC NE-AMNAT C-KANCH C-RATCH N-CHRAI C-SAMSO NE-AMNAT C-CHONB S-PHUKE %(tne ml S-NAKHO C-LOPBU olrnEy N-SUKHOC-PRACH C-SUPHA NE-MUKDA C-PHETC C-SAMSA C-RAYON mllor S-KRABIN-NAKHO 60 N-CHMAI C-CHANT C-PHET C N-KAMPHN-CHMAI NE-LOEI S-SURAT BK-BANG K S-SONGK S-RANON NE-UBONR N-TAK BK-SAMUT 70 NE-SAKON BK-SAMUT C-NAKHON-NAKHO C-KANCH C-CHONB S-YALA S-KRABI ardnoceS S-SATUN S-PHANG C-TRAT S-NARAT C-CHACH S-PATTA N-LAMPA C-SAMSO NE-CHAIY 50 S-TRANG C-CHACH C-PHRA Enyradnoc S-NARAT S-PATTA NE-UDONT N-CHRAI etN 60 C-CHAIN SeteN C-SAM SA S-RANON 40 C-CHANT BK-PATHU N-TAK N-MAEHO N-MAEHO 50 30 .005 .01 .015 .02 .025 5000 10000 15000 20000 Public Expditure on Secondary Education Per Student (Baht) Public Expditure on Secondary Education Per Student (Baht) Other Regions Efficiency Frontier Other Regions Efficiency Frontier Northeast Northeast DataSource: CGD 2000, SES 2000 Data Source: CGD 2002, SES 2002 B. 2001/2 Test Score Results by Level (Office of Education Standards and Evaluation, MOE) 50 40 30 20 10 0 Grade 3 Thai Grade 6 Thai Grade 9 Thai Grade 12 Verbal Grade 12 Total Northeast BKK/Vic Central North South 94 Infrastructure Roads and Phones One important ingredient to Thailand's as well as East Asia's economic success story is infrastructure. But does it also explain also the Northeast's lack of economic convergence? Public investment focused on locations with high economic potential, but progress extended beyond Bangkok and Center to rural areas. The Northeast, an area of the size of the Netherlands (105 million r