Report No. 35388-MX Mexico Mexico's Competitiveness: Reaching Its Potential Programmatic AAA Phase 2 August 10, 2006 Finance, Private Sector, and Infrastructure Department Colombia and Mexico Country Managament Unit Latin America and the Caribbean Region Document of the World Bank CRE Energy Regulatory Commission (Comisidn Reguladora de Energia) CUSTA US-CanadaFree Trade Agreement DGN General Standards Directorate (Direccidn General de Nomas) DOF Official Gazette (Diario Oficial) DPL Development Policy Loan EIA Energy Information Administration (US.Department o fEnergy) EMA Mexican Accreditation Entity(Entidad Mexicana de Acreditacidn) ENESTYC National Survey o fEmployment, Wages, Technology andTraining ENIGH Mexico's National Household Income and Expenditure Survey (Encuesta Nacional de Ingresosy Gastos de 10s Hogares) FA0 Food and Agriculture Organization FCST Telecommunications Social Access Fund(Fondo de Cobertura Social de las Telecomunicaciones) FDI ForeignDirect Investment FIAS ForeignInvestment Advisory Service FM Financialmanagement FTA Free Trade Agreement FUMIAF Mexican Foundation for Agricultural and ForestalResearch (Fundacion Mexicana para la Investigacion Agropecuariay Forestal) FY Fiscal Year GATS General Agreement on Trade inServices GATT General Agreement on Tariffs and Trade GDP Gross Domestic Product GEF Global Environment Facility GNI Gross National Income GOM Government o fMexico IAF International Accreditation Forum IBRD International Bank for Reconstruction and Development ICA Investment Climate Assessment ICT Informationand communications technology IDA International Development Association IDB Inter-AmericanDevelopment Bank IEC International Electrotechnical Commission IFAI Federal Institute o f Access to Public Information (Instituto Federal deAcceso a la Informacidn Pliblica) IFC International Finance Corporation ILAC International Laboratory Accreditation Cooperation IMF International Monetary Fund IMCO Mexican Competitiveness Institute (Instituto Mexicanopara la Competitividad) INEA National Institute o fAdult Education (Instituto Nacionalpara la Educacidn de 10s Adultos) INEGI National Institute o f Statistics, Geography and Informatics (InstitutoNacional de Estadistica, Geografa e Informdtica) IICA Inter-AmericanInstitute for Cooperation of Agriculture I M S S Mexican Social Security Institute (Instituto Mexicano del Seguro Social) INIFAP National Institute o f Forestry, Agriculture, and Livestock Research (Instituto Nacional de Investigaciones Forestales, Agricolas y Pecuarias) National Nuclear Research Institute (Instituto Nacional de Investigaciones Nucleares) IPER Infrastructure Public Expenditure Review I S 0 International Standards Organization I T A M Instituto TecnoldgicoAutdnomo de Mixico ITU International Telecommunications Union JICA Japan International Cooperation Agency LAC Latin America and Caribbean Countries ii LDP Letter o f Development Policy LFC Central Light and Power (Luzy Fuerza Centro) LFPA Federal Administrative Procedures Law (Ley Federal de Procedimiento Administrativo) LIC Labor Intermediation Center LPG LiquidPetroleumGas LSPEE Electric Energy Public Service Law (Ley del Sewicio Pziblico de Energia Elictrica) MBTU MillionBritishthermal units MDGs MillenniumDevelopment Goals M&E Monitoringand Evaluation MFN Most favored nation MRA Mutual RecognitionArrangement MTEF Medium-TernExpenditure Framework NAFTA North American Free Trade Agreement NOMS Official Mexican Norms (Normas Oficiales Mexicanas) NRI Networked Readiness Index OECD Organization for Economic Cooperation and Development OTRI Overall Trade Restrictiveness Index PAC Training SupportProgram (Programa deApoyo a la Capacitacidn) PECyT Special Plan for Science and Technology (ProgramaEspecial de Cienciay Tecnologia) PEMEX Mexican Petroleum (PetroleosMexicanos) PER Public Expenditure Review PIDIREGAS Infrastructure projects with deferred expenditure recording (Proyectosde Infraestructura Dferidos en el Registro del Gasto) PISA Programme for International Student Assessment PITEX Temporary Import Program for Exporters (Programa de Importacidn Temporalpara Producir Articulos de Exportacidn) PRODESCA Program to Develop Capacity inRural Areas (Programas deDesarrollo de Capacidadesen el Medio Rural) PROFECO Federal Consumer ProtectionAgency (Procuraderia Federal del Consumidor) PROSEC Sectoral Promotion Programs (Programasde Promocidn Sectorial) PRONASOL National Solidarity Program (Programa Nacional de Solidaridad) PROSENER Energy Sector Program2001-2006 (Programa Sectorial de Energia 2001-2006) R&D Research and Development RBP Regulatory Betterment Program RIA Regulatory Impact Assessment RFTS Federal Registry o f Procedures and Services (Registro Federal de Tramitesy Sewicios) SAGARPA Secretariat o f Agriculture, Livestock, Rural Development, Fisheries, and Food (SecretariadeAgricultura, Ganaderia,Desarrollo Rural, Pescay Alimentacidn) S A R E RapidBusiness Opening System (SistemadeApertura Rbpida de Empresas) SCT Secretariat o f Communications and Transport (Secretaria de Comunicacionesy Transportes) SE Secretariat o fEconomy (Secretariade Economia) SECTUR Secretariat o f Tourism (Secretaria de Turismo) SEDESOL Secretariat o f Social Development (Secretariade Desarrollo Social) SEGOB Secretariat o f Government (Secretaria de Gobernacidn) SEMARNAT Secretariat of the Environment and Natural Resources (Secretaria deMedio Ambientey Recursos Naturales) SENASICA National Health, Innocuousness, and Food Quality Service (SewicioNacional de Sanidad,Inocuidad, y Calidad Agroalimentaria) SENER Secretariat o fEnergy (SecretariadeEnergia) SEP Secretariat o f Public Education (Secretariade Educacidn Pliblica) ... 111 SFP Secretariat of Civil Service (Secretaria de Funcidn Pziblica) SHCP Secretariat of Finance and Public Credit (Secretaria de Hacienda y Crddito Publico) SME Small andmedium enterprise SICEX Comprehensive ForeignTrade System (SistemaIntegral de ComercioExterior) S N I National Researchers System (SistemaNacional de Investigadores) SNIE National Business IncubationSystem (SistemaNacional de Incubacidn de Empresas) SOE State-ownedenterprise SPS Sanitary and Phytosanitary ss Secretariat of Health(Secretaria de Salud) SSP Secretariat of Public Security (Secretaria de Seguridad Pziblica) S&T Science and Technology STPS Secretariat of Labor and Social Security (Secretaria del Trabajoy Previsidn Social) TA Technical assistance TFP Total factor productivity TRQ Tariff rate quota UNDP United Nations Development Program USAID United States Agency for International Development USD United States Dollars USDA United States Department ofAgriculture WACC Weighted average cost o f capital WBI World Bank Institute WDI World Development Indicators (World Bank) WEF World Economic Forum WTO World Trade Organization Vice President: Pamela Cox Country Director: Isabel M.Guerrero Sector Director: Makhtar Diop Task Team Leader: Jose Luis Guasch Co-Task Team Leader: Keta Ruiz iv PREFACE In 2003, the Government of Mexico (GOM) declared increasing country competitiveness to be a key priority o f the Fox Administration. This was a response to modest growth, employment, export, and competitiveness indicators; it also signaled concerns about the country's ability to respond to the pressures induced by advancing globalization, such as competition from China and the end o f the textile quota system in 2005. Within its ongoing dialogue with the World Bank, the GOM requested assistance to develop a strategy and an implementation agenda for the overall improvement o f country competitiveness. In particular, the GOM requested assistance in identifying the agenda and producing well-focused analytical pieces on key issues to address and facilitate the implementation o f that agenda. Responding to that request, the World Bank agreed to cast its assistance in the form o f Programmatic Analytic and Advisory Activities (AAA), the first phase o f which was in the form o f non-lending technical assistance (TA). This allowed for the provision o fjust-in-time analytical and technical support, as requested by the GOM.' Pre-programmatic assistance started in fiscal year 2004 (FY04) and helped define the GOM's Competitiveness Agenda 2004-2006 (see Annex 2 for a description o f the pre- programmatic work). The AAA (FY05-08) was structured in three phases (see Annex 1). The first formal phase (FY05) supported the design and implementation o f the Mexican Competitiveness Agenda through separate pieces o f analytical work at the request of the Government. I t has also included advocacy presentations and advisory assistance through briefings and background information on relevant competitiveness themes. The first phase o f the AAA led to a $300 million Competitiveness Development Policy Loan (DPL) from the World Bank to Mexico. The first DPL was approved in March 2006 and provides tangible support to several areas o f the GOM's Competitiveness Agenda. The DPL highlightsnumerous recent actions that the GOM has taken to improve competitiveness, as well as fbture plans that will form part ofthe secondand thirdDPLs planned for FY08 and FY09. This report concludes Phase 2 of the AAA (FY06). It aims to bring the key policy messages into the public sphere and provide the incoming Mexican administration (which takes office inDecember 2006) with a sound analysis o f competitiveness constraints. This i s critical given that improving competitiveness is a long-term challenge for the next sexenio, and fbture efforts will buildon important steps already taken by the Fox administration. This report i s based on what was learned from Phase 1 o f the AAA and the first DPL, as well as additional pieces o f analytical work conducted as part o f Phase 2 and achievements o f the Fox administration to date. The background studies from Phases 1 and 2 can be found at www.wor1dbank.orrr.m. Initially, a Phase 3 o f the AAA (FY07-08) was plannedto deal indepthwith specific issues o f interest to the next administration. However, given that the World Bank will be engaged in competitiveness dialogue with the next administration vis-a-vis the second and thirdphases o f the DPL, the programmatic AAA will close withthis report. 'The AAA was proposedinthe Country PartnershipStrategyof the WorldBankGroupwith the UnitedMexican States, ReportNo. 28141-ME,April 15,2004. V MEXICO'S COMPETITIVENESS: REACHING ITS POTENTIAL PROGRAMMATIC AAA PHASE2 TABLE OFCONTENTS EXECUTIVESUMMARY ............................................................................................................................. ... ~I I MAKINGCOMPETITIVENESSAPRIORITY;BUSINESSASUSUALWILLNOTWORK . .....1 Strong Macroeconomic Management but Relatively Low Growth ..................................................... 2 Progress inReducing Poverty and Inequality. butNot Enough........................................................... 4 Analytical Underpinnings Identify Priorities....................................................................................... 6 I1 MAXIMIZINGMEXICO'S EFFICIENCYPOTENTIAL . .................................................................. 10 Improvements inMexico's CompetitionPolicy Are Needed. Particularlyto Facilitate Enforcement and Serve as an Effective Deterrent to Anticompetitive Practices.................................................... 10 Mexico's Trade Environment Has BeenTransformed But More Should Be Done to Facilitate Trade and Reduce Logistics Costs............................................................................................................... 15 Mexico Has Made Far-Reaching Regulatory Improvements but Firms Still Face Pervasive RedTape .................................................................................................................................................... 21 Mexico's Energy Sector Needs to Strengthenits Performance ........................................................... 30 Mexico Lags on ICT and Telecom and More Competition is Needed ................................................ 39 Agriculture Faces Great Challenges inBothDomestic and International Markets ............................. 58 I11 MOVINGTOWARD THEINNOVATION-DRIVENSTAGEOFDEVELOPMENT . ..................... 72 IncreasedInnovationandTechnology Development and Better Links to the Private Sector are Essential for Productivity Increases and Faster Growth ............................................................................ 72 PressingImprovements to Mexico's Quality System are Needed ....................................................... 77 Better Education Quality. Expanded Adult EducationPrograms. and Improved Training Programs are Needed ....................................................................................................................................... 89 IV ADDRESSINGSUB-NATIONALDIFFERENCES . .............................................................................. 97 V DOUBLINGMEXICO'S PER CAPITA INCOMEIN 10YEARS . ..................................................... 103 ANNEXES ANNEX 1:THE COMPETITIVENESSAAA PROGRAM ........................................................................ 105 ANNEX 2: PRE-PROGRAMMATICBANK SUPPORT TO MEXICO ................................................... 107 ANNEX 3: ANALYTICAL UNDERPINNINGS ........................................................................................... 110 ANNEX 4: GROWTHDIAGNOSTICFRAMEWORK .............................................................................. 127 ANNEX 5: THEMEXICAN ENERGY SECTOR ....................................................................................... 128 REFERENCES ................................................................................................................................................ 139 Sesernl background studizs tind exercises were undertakenfor this report. Descriptions have been compiied i r i a single volu111c andcan be accessed at t\u.w.uorldbsnk.orlr.mx. The studies cover: ! I ! Diagiiosis,monitoring, and mcastiringrcsults: IMCO Country and State le\.el analyses i 0 S M l suppcrrt programs I 0 Quality infrastructure EXECUTIVESUMMARY MEXICO'S COMPETITIVENESS:REACHINGITSPOTENTIAL 1. Can Mexico compete? This question has dogged Mexico in recent years thanks to moderate growth and productivity numbers, a declining share o f the U.S. import market, and increasing competition from lower-wage economies such as China. The objective o f this report i s to inform policy makers, the private sector, and the public at large about competitiveness priorities and shortcomings in Mexico, andto point the way forward on improving competitiveness. The report highlightsconstraints to competitiveness and identifies key high impact interventions to maximize Mexico's potential for efficiency, generate jobs, and move the country to a more innovation-driven stage o f development. It draws on a series o f focused analytical pieces that were conducted for Phases 1 and 2 o f the Programmatic Analytical and Advisory Activities (AAA), as well as extensive World Bank work on competitiveness issues in Mexico and beyond.2 The report also recognizes that the Fox administration has taken important steps that have laid a solid foundation for future efforts. Yet, improving competitiveness is a long-term challenge that the next administration will face over the coming years. 2. The key policy messages that arise from this work are the following: I. To secure strong job creation and significant poverty reduction in Mexico, increasing competitiveness shouldbe a priority and business as usual will not work. 11. Mexico needs to maximize its efficiency potential by a) encouraging competition between firms, b) facilitating trade and reducing logistic costs, c) improving business regulation and the investment climate, d) strengthening the energy sector, and e) improving the ICT (information and communications technology) and telecommunications sector. Special emphasis i s needed to improve and transform the agriculture sector. m. To advance to the innovation-driven stage o f development, Mexico needs to boost innovation and strengthen quality, education, and training (Box 1). rv. Competitiveness priorities differ at the sub-national level and active reform at this level is necessary. V. Competitiveness improvements could allow Mexico to double its GDP per capita in ten years, as a number o f countries have recently done. These messages are addressed in the following five sections o f the Executive Summary. Section VI highlights selected recent actions that the Government of Mexico has taken to improve competitiveness, as well as planned future actions. It also summarizes the objectives for moving forward covered in the other five sections. Box 1: Stages of Development: Factor-, Efficiency-, and Innovation-Driven capital, labor, and energy factors, eedto produceinnovativeproducts ~~ ~ Annex 1and Annex 2 and the backgroundstudies containdescriptions of the World Bank's assistance programto increaseMexico's competitiveness. viii I.MAKINGCOMPETITIVENESSAPRIORITY;BUSINESSASUSUALWILLNOT WORK 3. Growthand productivityimprovementsinMexico havebeenmodest inrecentyears. Over the last decade, Mexico has benefited from a relatively stable economy and has avoided the pattern o f financial crises that have plagued many o f its past political cycles, thanks in large part to prudent fiscal and monetary policies. Despite this stability, structural economic constraints have held back growth and therefore poverty reduction. Growth rates have beenparticularly low over the last decade when compared with countries o f similar income in East Asia. While real GDP per capita in Mexico only grew at an annual rate o f 1.2 percent between 1994 and 2004, it grew at 7.7 percent in China, and at 5.9 percent on average in the East Asia and Pacific region. Productivity growth rates have also been lackluster. The share o f growth attributable to total factor productivity(TFP) was negative on average between 1980 and 2003, althoughthere has been a slight increase inrecent years (Figure 1). Figure1:GDP Growth in Mexico andits Figure 2 :Competitivenessversus GNI per Capita Componentsin %:Growth Accounting, 1960-2003 Rankings (Contributionsto GDP GrowthinPercent) 0 GDP0 Capdal Labor D TFP 4 -1 ' I 1965-1979 1980-2003 1996-2003 Note: Numbers are for real GDP growth (not GDP per capita) 0 20 40 BO 80 1M) 120 I Growth CompetItIY.Mss 1nd.x Rank 20062006 Source: Faal(2005). Note: Lower-ranking numbers are better. Source: World Economic Forum (2005a) and World Bank, World Development Indicators. 4. Given the productivity numbers, it is not surprising that comprehensive competitiveness measures show Mexico to be weak-underperforming-relative to per capita income and not improving?The World Economic Forum's 2005-2006 Global Competitiveness Report gives Mexico a rank o f 55 out o f 117 countries, down from 48 in2004.4 Incomparison, Mexico's gross national income (GNI) per capita rank within that same sample of countries is 39. While this gap between income and competitiveness rankings i s not atypical for Latin America, it indicates that Mexico i s falling well short o f its potential. Incontrast, Chile, India, China, and other East Asian economies enjoy competitiveness rankings that exceed their income placing (Figure 2). The World Bank's Doing Business in 2006 reports similar findings. Mexico ranked 73 (out o f 155 countries) on the publication's "Ease o f Doing Business" scale, compared with an income per capita rank o f 38. Competitiveness i s broadly defined as the "collection o f factors, policies and institutions which determine the level o f productivity o f a country and that, therefore, determine the level o f prosperity that can be attained by an economy." World Economic Forum(2005), p.xiii. This ranking is for the Growth Competitiveness Index. Two countries (Qatar and Kuwait) that were not ranked in 2004 placed aheadof Mexico in2005, accounting for a small part o f Mexico's fall. i x 5. An often used (albeit imperfect) proxy for competitiveness is export performance. On this basis, the weak competitiveness of Mexico's economy would be reflected in the reduction in its share of U.S. imports.Mexico is highlydependentonthe U.S. economy, which absorbs 88 percent of its total exports. After experiencing an export boom inthe 1990sthanks to NAFTA, Mexico has been losing U.S. market share, particularly to China. Mexico's share o f U.S. imports fell from a higho f 11.6 percent in2002 to 10.2percent in2005. Duringthat same time period, the Chinese sharewent from 10.8percent to 14.6 percent. 6. Lagging competitiveness, and hence growth, has meant lost opportunities for poverty reduction. Nearly all the countries that have had long term success in reducing poverty have done so with higher rates o f economic growth (left panel o f Figure 3).5This association i s also true for Mexico, where poverty reduction has tracked growth in GDP per capita over the last decade (right panel o f Figure 3). Nonetheless, after experiencing robust GDP per capita growth following the 1994 Tequila Crisis, income levels have been more or less stagnant since 2000. The recent growth slowdown has meant a significant loss o f opportunities for poverty reduction, although programs such as Oportunidudes have helped offset this trends6For Mexico to breakout o f its low-growthtrajectory and cut poverty, it will have to boost growth and improve its productivityand competitiveness. Figure 3: Poverty reduction is closely associatedwith growth World +China East Aria otharlhnn China =0 .E .n ia other than India.. -c 20 Uganda India 12,0004 c80% 0 -10 -2 10 1992 1994 1996 1998 2000 2002 2004 Average per capite erawrh rats. 1981-2001 1-GDP per capita +Extreme Poverty +Moderate Poverty] Source: World Bank (2004a) Source: MexicoENIGHand World Bank, World Development Indicators;World Bank calculations 7. Aware of its predicament, the Government of Mexico (GOM) in 2003 declared increasing country competitivenessto be a key priority of the Fox Administration. This was a response to the modest indicators o f growth, employment generation, export, and competitiveness; it was also a response to concerns about the country's ability to respond to the pressures imposed by advancing globalization, such as competition from China and the end o f the textile quota system in 2005. The GOM has made progress in some fronts, but much more is needed if Mexico i s to go beyond sound macroeconomic managementand stability and achieve higher growth and productivity. 8. An impact study conducted by the Mexican Competitiveness Institute (IMCO), as well as other studies, identifiedmajor priority areas. The IMCO (2005a) study develops a model to assess Mexico's competitiveness and establishes priorities among themes in terms o f impact on investment ~ World Bank (2004a). For a completeanalysis ofpovertyreductioninMexico, see World Bank(2004~). X (gross capital f o ~ ~ ~pcr worker,7~Fromthat model, point elasticities are e ~ ~ ~ ~givingathe~ impact, ~ i o n ~ i c ~ npcrccnt ~ ~ p r o ~ ~ireithcerespect ~ n t ~ n d e ~ e ~variablcs (Table I).. d e ~ t crylcs are the ~ o ~ penv~roi~ent, i and tax r e ~ l a ~ l orchqiasory e ~ ~ taxes~ ~ ~ n s ~ atioo and ~ ~ ~ o ~trade~~ j~ oc ~~ ~~i t,tran~ i ~ nand logistics, c o r ~ p ~ i u ~ , a a , financc, arid energy, These impact e ~ ~ i provide~~u~delines priu ~ a ~ s for to be tackled rn order to increase Mexico's ~ ~ ~ ~ ~ Theyc are also~ s i ~~ p ~ i ~~ by~twoo~other rctatively e ~ m p r e h ~ ~ ~ i ~ e r ~ ~~ ~~ . d rcporls: a ~ ~ ~ ~ Focus Groups reportj and the World ~Economic ~Forum's (M'EF)~ Global ~ ~ ~ t ~ r ~ ~ $ zcss Report ~ i ~ U 5 - 2 ~Finally, thc ~ U ~ . ~ ~priority ~areast are~ ~ ~~ a ~~ byedanalyzing d i ~ ~~~ e ess c o ~ s ~ rthrough~a~growth d i a ~ ~ Q~~ ~ j rc ~ i ~ s ~~ ~~ ~~Kodrik,andVctasco,~r ~a ~~ s~ ak ~ ~ ~ Table 1: Constrrrints to C o ~ p ~ ~ ~ ~ ~ ~ t ~ n e ~ $ ' IIII Impact on investment per worker of a I Subject areas 10pcrccntiniprovcrnentin the \ariables behind each subject area I xi authorities are supporting improvements in most o f these areas," but further progress i s necessary. It i s also important to address inefficiencies in the agriculture sector. Each o f these areas i s discussed in the following subsections. A. IMPROVEMENTSINMEXICO'SCOMPETITION POLICYARENEEDED, PARTICULARLYTO FACILITATE ENFORCEMENT SERVEASAN EFFECTIVE AND DETERRENT ANTICOMPETITIVE TO PRACTICES 10. Greater competition leads to efficiency gains. Competition, and the benefits it brings with respect to lower prices, greater variety, and the development of new products, is crucial for markets to work well and for increasing the wellbeing o f a society. Policies that enhance competition are expected to bringnot only static improvements in efficiency, especially in disciplining monopolies, but also dynamic improvements in efficiency, through improvements in productivity and innovation. Recent empirical research indicates that regulation that lowers entry barriers and state control increases productivity and technological innovation, stimulates business investment, and increases long-run employment rates.".Therefore, reforms to enhance competition could have significant effects on growth. A study o f the effect of pro-competitive regulatory reform in several industries in the United States found that annual welfare gains inthat part o f GDP that was affected by reform were more than 7 percent, with 90 percent o f the benefits flowing to consumers.I2 Increasing competitive pressure can increase the probability o f firm innovationby more than 50 percent.I3 11. Studies show competition deficiencies in Mexico. For instance, according to a recent governance survey, the two largest market obstacles to business development identified by firms in Mexico are monopolies, private or p~b1ic.l~ The Global Competition Review 2003 ranking o f competition regimes in 25 countries places Mexico near the bottom with a score o f 2.25, ina rating from 0-5; this places it only above Greece (2.0) and Argentina (1.5). 12. To improve the effectiveness of competition policy in Mexico, actions are needed that will do the following: Remove the constraints on statutory authority and judicial review processes and allow for preventive measures; Enhance the collaboration by the Federal Competition Commission (CFC) with different regulatory agencies and states through harmonization agreements or/and alignjurisdiction; 0 Improve the deterrent effects o f the existing Federal Law on Competition Policy, by improving level and collection o f sanctions; and Increasepublic support for competition policy. 13. The amendments to the Federal L a w on Economic Competition have been a great advance, but the associated regulation and its application are important. The amendments-that reform 25 articles, repeal 1, and add 11more-were approved inApril 2006 and went into effect on June 29, 2006. are addressed elsewhere through a parallel AA.4 on Education; an Institutionaland Governance Review, and finance AAA, including an FSAP (Financial Sector Assessment Program). loSome of these improvements are being supported by Bank operations such as the MX Competitiveness DPL and the MX Finance and Growth DPL series. 'INicoletti and Scarpetta (2003) Nicoletti and Scarpetta (2004). I'OECD (2002). l3World Bank (2004a). l4CEESP (2005).. xii The amendments are designed to increase the effectiveness o f detecting and sanctioning monopoly practices. They classify new anticompetitive actions, increase fines and expand the capacity o f the CFC to divest the assets o f firms that engage in anti-competitive practices, in cases where they have not complied with previous orders; undertake verification visits; implement leniency programs; and order preventative measures. To make the amendments concrete, the corresponding regulation needs to be issued, which i s expected to occur by the end o f December 2006. Training and diffusion efforts will then be necessaryto support the implementation and application o f the amended law. Box 2: Competition: Selected Evidenceand Effects A studyofthe effect ofpro-competitivere that annualwelfaregains inaffectedsect consumers(OECD). Increasingcompetitivepressurecanincrea B. MEXICO'STRADEENVIRONMENT BEENTRANSFORMED MORESHOULDBEDONE HAS BUT TO FACILITATE TRADE ANDREDUCELOGISTICS COSTS 14. The trade and investment environment in Mexico has been transformed over the past two decades. This is inpart due to the 1994 launch o f the North American Free Trade Agreement (NAFTA) with the United States and Canada, and, more recently, trade agreements with the European Union, Japan, and others. Growing competitive pressures from countries such as China have also generated policy reform measures to lower trade transactions costs. The engine o f export-led expansion in the second half of the 1990s, however, has slowed down in recent years. And while exports have begun to diversify away from petroleum and relatedproducts, Mexico still remains heavily dependent on a single market-the UnitedStates. 15. Domestic reforms can have a large impact. Hanson and Robertson (2005) suggest that internal conditions in Mexico associated with its ability to expand export-supply capabilities-areas directly related to trade costs and facilitation-are a more important factor in limiting the country's export growth than China's expanded trade into the UnitedStates. Soloaga, Wilson, and Mejia (2006) find that trade facilitation reforms in Mexico could increase exports by US$29.3 billion, equivalent to 21 percent o f total Mexicanmanufacturing exports. 16. Logistic costs are particularly high in Mexico. They are estimated at over 20 percent o f product value, while the OECD average i s 9 percent." A pilot survey o f a new Logistics PerceptionIndex, aimed at measuring operators' perceptions o f countries' logistics environments, gives Mexico a score o f 3.6 (7 being the best and 1 the worst), lower than other Latin American countries like Argentina (4.5) and Brazil (3.8).16 Recent analyses by the government and the private sector, both o f whom were concerned about Mexico's potential disadvantage vis-a-vis its competitors, also highlighted considerable logistics weaknesses. 17. Key actions to facilitate trade and reduce logistics costs include the following: l5Guasch (2004). l6The LPIis a World Bankinitiative implementedwith the support of the TurkuSchoolofEconomicsandBusiness Administration,Finland.A pilot surveywas carriedout in2003. xiii Further lowering tariffs and non-tariff barriers and reducing the dispersion of tariff rates (consolidation at a few levels). Lowering trade barriers offers gains for consumers and allows firms to acquire lower cost inputs. The simplification or harmonization o f tariffs would make more productive chains viable." This is particularly important given the need to continue to diversify exports beyond oil and the major export destination, which continues to be the UnitedStates. Expanded use of information technology and reduced regulation can help lower trade costs. Making trade procedures and documentation available electronically will reduce the time and costs for firms to conduct business abroad. Key internal corridors and logistics centers could be improved, as manufacturers face high costs to deliver their products frodto the border, ports, or airports; access to those facilitates is already fairly congested. Elevated transport costs, times, and incidence o f delays have also led businesses to maintain highinventory levels. Enhancing customs efficiency. The lack o f flexibility in working hours and locations for product clearance adds delays and costs. The customs administration would also benefit from a modernization o f its infrastructure, including better use o f information technology inports, airports, and at land border crossings. Strengthening the organization of firms' supply-chains. Manufacturing companies, especially SMEs, often incur higher costs due to a lack of skills in modern supply-chain management techniques. This involves, through business development services and logistic assistance, rethinking the complete supply chain from material and services sourcing to the use o f specialized distribution companies to deliver the end product to the client. Box3: Trade FacilitationandLogistics: SelectedEvidence and Effects recent years. Ofthe top ten agriculturalproducts-gain c. MEXICO MADE HAS FAR-REACHING REGULATORYIMPROVEMENTSBUT F I R M S STILL FACE PERVASIVEREDTAPE 18. Mexico needs to reduce its regulatory burden to improve productivity. Several studies demonstrate how the regulatory burden inMexico i s still hindering business growth.'* The OECD, in its most recent reviewIg o f the effect o f product market regulations on investment and growth found that when regulation is restrictive, productivity growth is generally below the OECD average. The OECD l7 Currently, insome sectors it is more profitable to import final goods under preferential treatment than to import the inputsand raw materials inorder to produce those same goods domestically. l8 See Cotis (2005), World Bank (2006a), and IMCO (2005 a and b). l 9"Indicators o f Product Market Regulation", OECD, 2003 xiv also estimates that if those sectors that lag in terms o f productivity were to modernize their regulatory framework and align it on best practice, thenproductivity could increase by as much as 10 percent.20 19. The World Bank's Doing Business and IMCO have shown that regulation i s a drag on the Mexican economy. Compared to other countries in Latin America, Mexico clearly lags behind in: a) time and cost to open a business; b) time and cost to obtain a license; c) cost o f hiring; and d) cost o f tax compliance (World Bank, Doing Business database). The Mexican Institute for Competitiveness (IMCO) produced its own benchmarking report in 2005, which concludes that the cost of the regulatory burden currently represents 15 percent o f Mexico's GDP, while the benefits from regulatory improvement could yield a 5-10 percent increaseinGDP. 20. Regulatoryreformsto boost competitivenessincludethe following: (i) Elimination or improvement of procedures that impose unnecessary large burdens on entrepreneurs (starting a business, licensing, registering property, employment regulation, contractual enforcement, tax regulation); (ii)Streamlining of business interaction with government in order to reduce uncertainty and improve implementation efficiency. Industry sectors that are key to Mexican competitiveness and under specific competitive threat from foreign markets could benefit from a thorough review o f administrative barriers in order to identify where action can be taken to reduce cost. Such sectors and issues could include foreign trade, manufacturing, property, finance, norms and standards, transport and telecommunications, energy, and labor among others. For each sector or issue, reduction o f time, procedures, and costs o f fulfillment would help improve competitiveness. The Federal Regulatory Improvement Commission (COFEMER) can leadthese efforts, but it can exercise its full powers only ifit enjoys broad support. 21. Other importantactions to improveregulationincludethe following: Strengthening public consultation; 0 Better harmonizationof regulation betweenthe different levels ofjurisdiction; 0 Expanding the implementation o f the Rapid Business Opening System (SARE) to reduce the time and cost o f business opening regulations; creating a system to measure the SARE's effectiveness and monitor its implementation; 0 Reinforcing the Regulatory Impact Assessment (RIA) process in federal institutions, as well as creating a monitoring and evaluation framework for RIAs; 0 Increasing cooperation with sub-national jurisdictions and the private sector to keep improving regulations at all levels. Box 4: Reeulation: Selected Evidence and Effects States at a cost o f 17%of per capita income The cost of the regulatory burden inMexico *'"Economic Policy Reforms: Going for Growth and the case of Mexico", Jean-Philippe Cotis, Chief Economist, OECD, 2005 xv D. MEXICO'S ENERGY SECTORNEEDS STRENGTHENITSPERFORMANCE TO 22. An efficient energy supply at competitive prices i s a major driver of economic growth and productivity, yet Mexican prices and performance have been somewhat inadequate. Both energy and non-energy businesses' investment decisions are based on expectations about energy prices and service quality levels. And unfortunately, expectations o f Mexican businesspeople are quite low: business opinion surveys give Mexico the lowest rank o f any OECD country in terms o f energy infrastructure adequacy and efficiency (Figure 4). Natural gas, electricity, and fuel oil prices have seen dramatic increases in recent years and are among the highest in the world. High electricity costs are compounded by fluctuations in frequency and voltage, resultingin effective costs that are 10-60 percent higher than those of U.S. counterparts. When annual interruptions are compared with Latin American privatized distribution companies, the performance of Mexico's National Electric Company (CFE) i s limited. Furthermore, electricity subsidies are highlyregressive and amount to about 1.1percent o f GDP. Figure4: EnergyInfrastructureAdequacy andEfficiency(2005) Singapore Czech Republic Colombia USA Portugal Chile Venezuela Spain Turkey Russia Brazil China India Indonesia Mexico Argentina 0 2 4 6 8 10 Note: Scores (between 0 and 10) are based on business survey responses to the question: "Is the energy infrastructure adequate and efficient?" The highest value indicates the most positive perception. Source: IMD World Competitiveness Online. 23. Mexico needs to implement a number of strategic measures to ensure the long term sustainability and competitiveness of the energy sector. In the hydrocarbons sector, about US$130 billion is estimated to be required inthe period 20042012 to restore reserves rate and minimize imports through activities in exploration and production, refining, petrochemicals, and others." Likewise, the electricity sector needs an important amount o f resources to finance additional capacity and modernize interconnected infrastructure. The growing dependence on fuel imports and the gradual deterioration o f energy infrastructure calls for the implementation o f a number o f far-reaching strategic measures, which may include the following: 0 Lower fiscal dependence on oil revenue. Ways need to be found to lower the country's fiscal and budgetary dependence on oil revenues and allow for autonomous enterprise management in hydrocarbons and electricity. One way is to strengthen other sectors o f the economy so they can act as a substitute for Mexican Petroleum (PEMEX) budgetary support. The high leakage factor o f Acevedo (2004). xvi subsidies (64 percent o f value in electricity) is highly regressive and increases the fiscal burden. Private sector participation can also help minimize the fiscal burden and financing needs o f the hydrocarbons and electricity industries. Increased transparency o f sector performance i s key to inducing improvedperformance. 0 Initiate national dialogue. A structured national dialogue should include relevant stakeholders and expert analysts to review the options that could ensure the long term sustainability o f the energy sector (e.g., national expert committee). In parallel, strengthen the capacity o f political parties, senate and congressmembers, and the civil society to analyze energy sector bottlenecks. 0 Modernizethe energysector and reducedependence on fuelimports.Strategies should include a) the implementation o f a sound program o f investments and capital expenditures; b) fuel diversification measures; c) taking advantage o f private sector investment interests; and the promotion o f strategic partnerships that bring additional sources o f capital; d) design and implementation o f innovative legal and financial instruments to lower risks associated with fuel price, fuel availability, and other (e.g., innovative covenants in power purchase agreements, phasing out o f PJDIREGAS (infrastructure projects with deferred expenditure recording), fuel price hedging, fixed price coverage with reference price or cap); and e) implementation of a program to improve service quality and performance to international levels. 0 Strengthen policy and regulatory frameworks. New investments will require the consolidation o f policies and regulatory frameworks that increase longterm certainty levels for all participants in different segments o f the hydrocarbons and electricity supply chains. Box 5: Energy: Selected Evidence and Effects The quality of the electricity service ispoor comp fluctuations affect the overallproductivity levels of electronic equipment (World Bank, 20054 IMCO, E. MEXICOLAGS ICTANDTELECOM, MORECOMPETITION ISNEEDED ON AND 24. Despite impressive ICT growth during the 1990s, Mexican investment in ICT lags behind other Latin American and OECD countries. Mexico's level of ICT expenditure as a share o f the overall economy (3.1 percent) is significantly below OECD countries such as Japan (7.4 percent), U S (8.8 percent) and New Zealand (10 percent). It i s also nearly half that o f Chile and Brazil's rates o f 6.7 percent and 6.9 percent respectively. As a growing number o f studies have found,22countries with higher levels o f investment inICTs experience higher economic and social development growth. 25. The ICT sector faces a numberof issues, includingthe following: 0 Network readiness needsto be improved. 0 Fixed line growthhas not kept pace with comparable countries. 0 There is a digital divide between urban and rural areas and between states-with Southern states particularly falling behind. See OECD (2004c), "ICTs and Economic GrowthinDeveloping Countries." xvii 0 More competition is needed. Telmex, the incumbent, dominates the long distance, local, and cellular telecommunications markets. It provides service to over half o f all dial-up internet accounts and over a third o f high-speed internet access accounts. Telmex's net profit margins and earnings per share from continuing operations are more than twice that o f its closest rival. 0 Telephone charges are highinMexico compared to Latin America, especially local loop prices for businesses. 0 Mexico's ICT regulatory framework i s weak and regulators have beenreluctant to take action on controversialissues. COFETEL lacks adequate enforcement powers and independence. 26. Private sector provision of ICT infrastructure needs to be expanded. Although current government programs provide important contributions towards the development o f a knowledge economy, they do not adequately address the most important challenge: that o f dramatically increasing and expanding private sector provision o f ICT infrastructure in Mexico. To do so would require a level playing field, elimination o f entry barriers, healthy competition, and increased access to ICT services. The following are some recommendations to facilitate these tasks: 0 Create a level playing field through institutional changes. Rebuilding the credibility, effectiveness, independence, and transparency o f COFETEL is necessary. 0 Eliminate entry barriers and foster competition. The licensing regime for new entrants should be simplified and streamlined. Requirements to register contracts with COFETEL for all but the incumbent should be eliminated. Legacy voice-centric regulations should be eliminated and automatic review of regulations shouldbe required. 0 Promote equity and increase access. Broadband deployment should be fostered and local loop costs reduced through competition. The universal access program should be strengthened. The government should undertake a consensus-building campaign on the priorities for universal access and cross-subsidies issues should be addressed. Box 6: ICT and Telecommunications: SelectedEvidence and Effects countries suchas Chile (6.7%), Braz DevelopmentIndicators,2003). Business telephonecharges (factorin F. AGRICULTUREFACES GREAT CHALLENGESINBOTHDOMESTIC AND INTERNATIONAL MARKETS 27. Agricultural performance has been lagging since the 1980s, compared to the Mexican economy as a whole and to major regional competitors. This i s despite consistently highsubsides, drastic domestic reforms, and the advent o f NAFTA. Mexico i s still the largest provider o f fresh fruits and vegetables to the United States, with 56 percent market share, but is facing share erosion in most o f its 24 top export crops, losing market share to South and Central American competitors. Diversification to non-US. markets is minimal despite the signing by Mexico o f 12 free trade agreements (FTAs) other than NAFTA. xviii 28. Mexican agriculture faces great competitive challenges. As NAFTA tariff exemptions expire in 2008, Mexican producers will face renewed competition inproducts like maize, sugar, powder milk,and beans, while transitory support programs, like Procampo, will be discontinued, at least inprinciple. Also, as the United States signs FTAs with other regions and countries giving them preferential treatment, Mexico will face increasinglytougher competition inits most important export market. 29. Key to further export growth i s not so much market access as removing internal constraints. Policy recommendations to boost exports often focus on issues limiting access to foreign markets. Mexico, however, has already quite successfully tackled market access through its aggressive bilateral negotiation strategy. For Mexico, then, further export growth requires enhancing competitiveness through increased productivity inthe production and marketing chains. Efforts should include the following: Targeted promotion to diversify and expand Mexican export markets. Targeted assistance instruments could include export credit guarantee programs, export incentive programs for specific high priority products and markets, market research assistance, product and market specific export counseling, and assistancewith documentation and other export requirements. Improvement in irrigation. This i s not limited to just hydraulic infrastructure but also includes institutional strengthening to improve the technical, administrative, and financial management capacities o f the water sector to improve irrigationefficiency. Since water scarcity i s a great problem inmajor agricultural areas, and agriculture is the dominant user of water, water saving techniques need to be encouraged to raise water productivity. Increased public investments in trade-related transport infrastructure. Particular targets would include the following: additional bridges, access roads, and rail lines to cross the border; additional commercial inspection facilities; the modernization and expansion o f Mexican port facilities; updating and expansion o f intermodal transportation facilities; improved highways to handle the rapidly growing truck traffic; and speedier customs clearance procedures. Increased public investments in agricultural research and extension. Critically needed are investments in research and extension to adapt and disseminate technologies developed in other countries; also critical i s the generation o f new technologies adapted to the conditions o f Mexican resource endowments, particularly for small and medium-sized agricultural and agribusiness firms. An enhanced market intelligence system. InMexico the generation and flow of information needed by agricultural producers and agribusinesses is not keeping pace with the growth of globalization. Although much has been done to enhance market information, a large segment o f Mexican agricultural producers and agribusinesses still lack access to critically needed information to manage their risks and make sound production and marketing decisions. A better functioning land market. Despite important reforms since 1992, legal restrictions on e j i d and ~community lands still interfere with land allocation and transfers, and inlow value areas ~ ~ transactions involving outsiders are still scarce. A more efficient and effective National Health and Food Quality Service (SENASICA). Additional and stable resources to support the operations o f SENASICA would greatly improve Mexico's effectiveness inresponding to food safety and sanitary and phytosanitary (SPS) problems,24 and prevent hazards more effectively. 23An ejido is a land-owning peasant community or the land owned collectively by the members o f such a community. 24Sanitary (human and animal health) and phytosanitary (plant health) measures ensure that food is safe for consumers and prevent the spread ofpests or diseases among animals and plants. xix Box 7: Agriculture: SelectedEvidenceand Effects to $1.3 billionin 2 111. MOVINGTOWARD THE INNOVATION-DRIVENSTAGE OFDEVELOPMENT 30. If Mexico wants to advance beyond competing solely based on efficiency and move to the innovation-driven stage o f development, it needs to invest more and more effectively ina) innovation; b) quality; and c) education and training. Improvements in these areas are being supported by Bank operations,*' but further progress i s necessary. Such progress will allow Mexico to engage in higher value-added activities that are based on innovative products and processes, placing the country on a more equal footing with its OECD brethren and allowing it to achieve faster growth in order to secure strong job creation and significant poverty reduction. A. INCREASEDINNOVATIONAND TECHNOLOGYDEVELOPMENT BETTERLINKSTO THE AND PRIVATE SECTORARE ESSENTIAL PRODUCTIVITYINCREASESAND FASTER FOR GROWTH 31. Innovation i s a major driver of economic growth and productivity. Roughly half o f cross- country differences inper capita income and growth are drivenby differences intotal factor productivity (TFP), generally attributed to technological development and innovative capacity.26 These should be understood in a broad sense, referring not only to the ability to engage in research and development (R&D) activities, which eventually may or may not be transformed into new outputs, but also to the efficient use o fmodem operating technologies and to the adoption and adaptation o f new ones. 32. The Mexican innovation system performs below par compared with economies of its income level, and it has a low ability to translate investments into innovation. Private investment in R&D, both as a proportion o f GDP and as a percentage o f total national R&D expenditures, i s very low. Other indicators linked to innovation such as licensing expenditures, patent count, and so on are also low given its level o f GDP per capita. 33. Several steps need to be taken to develop a coherent innovation system with stronger and more credible institutions, a more active role for the private sector, and stronger academic- industry linkages. Efforts should focus on the following: 0 Institutional strengthening. The strengthening o f funding mechanisms and incentives for the National Innovation System across all areas and levels o f government should be continued. Even though there has been some progress through the creation o f the sectoral and mixed funds, synergies and complementarities need to be further exploited. Promoting the effective development o f national and international knowledge networks is also needed. 25The MX Competitiveness DPL I,the MX Innovationfor Competitiveness APL I,MX Basic EducationAPL 111, MXEducationQuality APL and MXTertiaryEducationAPL I,. 26 Hall and Jones (1999), Dollar and Wolf (1997). xx 0 Educated and incentive driven increases in funding. Programs should be scaled up through consolidation and funding, particularly those that emphasize technology upgrading and development; formation of the venture capital industry through programs such as those recently launched by CONACYT should be speeded up. The pace o f scaling up and speeding up should be based on performance results. Programs that support demand need to be complemented by the strengthening o fthe regional infrastructure that provides technological services. 0 Monitoring and Evaluation program. The portfolio o f programs aimed at the consolidation and reallocation o f funding should be reassessed in order to strengthen their effectiveness, coherence, alignment with objectives, and rationale. It i s key that all such reassessment should be based on proper evaluation. 0 Facilitating interactions between R&D institutions and industry. Although the Program of High Value-Added in Businesses with Knowledge and Businessmen (AVANCE) and the consortia programs can be viewed as entry points for facilitating such interactions, more efforts are needed. To make science more relevant to industry requires providingmore funding to enhance such links. It will also require a review o f the intellectual property rights system and the establishment o f a framework for public institutions to reward staff for success inproductive research. 0 Strengthening state-level sub-systems for innovation. This can be done by improving the design and implementation o f decentralized funds; emphasizing the roles played by the state-level science and technology councils; developing capacity at state level and in the economic development agencies; and expanding the collaborative R&D networks in order to establish regional innovation clusters. 0 Supporting internationalization policies. Projects to strengthen the integration o f Mexican innovations into the global system should be financed; and successful Mexicans working abroad should be engagedinan international knowledge network. Box 8: Innovation: SelectedEvidenceand Effects Roughlyhalfofcross-countrydifferences inper capitaincome and growthare drivenbydifferences in total factor productivity (TFP), generallyattributedto technological (Hall andJones, 1999; DollarandWolf, 1997). ICalculations suggest a socialreturnonR&Dof 60% inMexico (Lede Mexicanf m s invest a little above 0.1% of GDP inR&D, compared B.PRESSING IMPROVEMENTS TO MEXICO'SQUALITY SYSTEMARE NEEDED 34. Quality systems contribute to economic development in a number of ways. Standards, which are at the heart o f quality systems, lower transaction costs since they communicate reliable and consistent information regarding the characteristics and quality o f products and processes to consumers. By allowing consumers to differentiate products o f different quality, they reduce asymmetric information and promote the creation o f high-quality markets. They also enhance competition by allowing easy comparison o f products and services conforming to a given standard. Standards embody technology and act as a channel for technology transfer for the institutions and firms that adopt them. This raises national productivityand competitiveness. xxi 35. Mexico's quality system presents many deficiencies. Mexico's quality system, composed o f a standardization system, a metrology institute, a national accreditation system, certification, testing, and inspectionbodies and calibration labs, needs a major overhaul as listed below: Mexico needs to further develop its metrology and calibration services, which serve as a public good. Since the costs to develop high-level metrology services significantly outweigh the income received from providing such services in Mexico, public fhnding i s necessary. So far, the National Metrology Center (CENAM) has developed a robust set o f measurement standards, but there are some gaps. Meanwhile, low-level metrology services can be attractive to the private sector, which has developed a high geographic concentration o f testing labs and inspection bodies around Mexico City, Prices for conformity evaluation services are relatively high. Mexico needs to solve many problems with its institutional framework for defining and presenting standards. Standards that require physical or chemical measurements frequently do not include adequate consideration o f measurement characteristics needed to produce the required level o f confidence. As a result, the standards can become non-viable. Consultative committees and standardization bodies do not have strong enough ties to science and technology institutions. A mechanism to conduct costhenefit analyses o f standards has not been developed and therefore the mandatory review o f standards every five years often occurs without any information about the impact of standards. Mexico's accreditation system needs to reduce costs, possibly through increased competition. Mexico's market size and subsidies to the MexicanAccreditation Entity(EMA)have discouraged the emergence of competitors and led to high prices. Furthermore, the EMA's mix o f public and private origins has ledto a complex governance structure and slow decision-making. Other problems: To minimize technical barriers to trade, national conformity assessment procedures should be harmonized across countries. The EMA also faces challenges related to internationalrecognition, domestic promotion, and meeting domestic demand. Box 9: Quality: Selected Evidenceand Effects A studyonGermanyfoundthat standardscontributedto overallgrowthrateof 3.3% inthe period 1960-1990 (Ju Adopting 100 additionalBritishstandards raisedthe UKexport t c.BETTEREDUCATIONQUALITY, EXPANDED ADULTEDUCATION PROGRAMS, AND IMPROVED TRAININGPROGRAMS ARENEEDED 36. Quality needs to be at the top of the education agenda. Enrollment i s low in secondary and tertiary education and the system produces students with poor skills. Hence, quality and relevance become an important factor for keepingchildren inschool at the secondary and tertiary levels and linking education more closely to thejob market. . 37. School autonomy, accountability, and assessment need to be promoted. Mexico needs to continue efforts to move decision-making to the schools. Parental involvement i s required to help make schools accountable, and performance needs to be assessed and fed back into the system at the school level. xxii 38. Private financing i s needed to overcome the gap in tertiary education. Expansion through public investment is fiscally unsustainable inMexico given that taxes are only 11 percent o f GDP. Public financing needs to be redirected to provide financial aid to poor students and the broad availability o f (income-contingent) student loans would facilitate increased private investment intertiary education. 39. More resources are needed to continue the expansion of adult education. The National Institute o f Adult Education (INEA) offers secondary education certificates program^.^' INEAstudents come from poor and vulnerable groups, such as women, youth, and indigenous peoples. Inorder to expand program coverage, INEA would need additional fiscal resources, improved internal efficiency, and the introduction o f incentives to increase effective demand. The government is already pursuing some of these initiatives, but more needs to be done. 40. Small and medium enterprises (SMEs) comprise a huge part of the Mexican economy, and worker skills and training play a key role in their success. SMEs make up 99 percent o f enterprises in Mexico, employ 64 percent o f the workforce, and account for over 40 percent o f GDP. These firms face many difficulties, perform below the level o f their larger counterparts, and rarely export their goods and services. An assessment by the Secretariat o f Economy (SE) identified lack o f access to financing and skills as the main weakness o f SMEs. The mortality rate o f SMEs i s high-200,000 SME are created every year but only 70,000 (35 percent) survive two years later. The firms that fail report lack o f training and expertise intheir workforces as key reasons for their failure. 41. Impact evaluations of S M E training programs are needed. The historical proliferation o f SMEs support services inMexico and the absence o f reliable informationon their effectiveness, suggeststhat it i s necessary to take stock and formulate a new and comprehensive strategy for supporting SMEs before considering program expansion. To collect the necessary data, federal funds should be allocated to expand the ENESTYC (National Survey o f Employment, Wages, Technology, and Training) training module that covers not only manufacturing but also services and commerce sectors. Once this comprehensive database is available, rigorous cost-benefit analyses o f the larger training programs for SMEs could be carried out. The results of these evaluations would provide the basis for a national training policy for SME workers and provide incentives for private sector worker training investments. Box 10: Educationand Training: SelectedEvidenceand Effects These numbersare quite 1 In2000,24% ofMexicanad schooling. (World Bank, 2004e). Only35% ofnewSMEs survivemorethan2 years, an IV. ADDRESSING SUB-NATIONAL DIFFERENCES 42. There are substantial differences in competitiveness across states in Mexico. Although competitiveness i s usually analyzed at the national level, it is important to recognize that there is a wide 27Secondary educationcertificates are consideredthe equivalent of a secondary education. xxiii variety in the c o ~ p e t ~ ~ ~ ~Eeveise of~ , e n s es and states, Studies by the World Bank's I rities for rcfrrrms. ~ u ~ - ~analysis~is~p n~ a~ ~ crelevanta given ythat state and local ~ t ~ ~ ~ r ~ have ~ ~ r ~ s d iover ~ a ~ ji ~ i ~ ~ e ~climaten reforms, including ~ r o c e ~ to ~opens a c ~ many t ~ e t ~ e ster property, and ertfarcc contracts. 43. When ranked globally OR the cost af starting a busin ,Mexican states vary s ~ ~ ~ j ~ (Figure 5). A ~ ~ a s c a ~ ~isnthecleast costly-7.3 percent of 31 c ~ s income pcr capita--and still it ranks 33rd p ~ o b ainl ~cost tu startup. Jaliscn fol ~ costing 10.9 percent, at 52nd place, behind Gabaror-ie ~ ~ ~ ~andt Santiago~(Chlc).n Esliadode s ~ ~ a ~ s lowest at 127th place-far behind X~ma(Pent) and ~ ~ a ~ eCity.i Elowever, it ~ a ~ a IS less e x p e i ~ to~start a business in any s ~ ~ of the hfcxican states than in Jakartaor ~ a n a ~ ~ i a . Virture 5: Business Start-t'r, CostInhlexico and other selected ~ c o ~ ~ ~ i ~ ~ ~ s 44. A ~ r to ithe ~Mexican ~ o ~ i ~ e t Institute~ e ~ ~ ~ the ~ o s s ~ b for~ ~ t ~ ~ ~ ~ ~ ~ ~ i ~~~~~~, s i j n i ~ ~ oare ~J ~eb~s ~~ a~but varya across states.2sA I O perccnt ~ ~ p r o ~in?cach~of~IMCO's ~ ~~ t i ~ ~ e n ~ ~ o t i ~ ~ c ~factors~ would~result sinsan increase in ~ ~ v e per worker~of e89 ~percent~ on average. i ~ ~ ~ ~ e s ~ ~ ~ XXoweier, this average niasks co derable v a r ~ a ~across states: the state with the ~ a x i ~ ~ ~gaini ? i i o ~ u would experience an 841 percent increasc in jrives~rnc~i~. while the state with the ~ ~ j ~ ~gaini \~s'owIcIn ~ ~ u r only see a 12 percent gain. ~ ~ ~ r ~it sis not nalwaysl the worst ranked state5 that stand to pain thc ~ ~ y ~ most. can be changed by a governor or a mayor. And as the news about reforms spreads, there is increased interest to replicate success stories. 1Aguascalientes (Easiest) 8 San Luis Potosi 2 Guanajuato 9 Coahuila 3 Chihuahua 10Mexico Citv II4 Jalisco 11 Estado de Mexico 5NuevoLebn II 12 Puebla II 6 Veracmz 13 Quer6taro (Most difficult) 7 Yucatan Box 11: Sub-national Competitiveness: SelectedEvidenceand Effects . (World Bank, Doing Business database). The impact on investment per worker ofa 841% to 12%, depending on the Mexican V. DOUBLINGMEXICO'S PER CAPITA INCOMEIN10YEARS 46. Competitiveness improvements could allow Mexico to double its income in ten years, as a number of countries have recently done. As shown in Table 3, the time necessary for countries to double their income has been steadily decreasing. Particularly impressive has beenthe fast growers since World War II-for example Japan, China, Thailand, Botswana, Ireland, and Chile. Pre-industrial 350 years Britain (1780-1830) 175 years Britain in lgthcentury 65 years Fast growers since WWII (Japan, China, 10 years or less Thailand, Botswana, Ireland, Chile.. ,) 47. Inorder to double GDP per capita over a decade, annual GDP per capita growth rates o f at least 7 percent are necessary. Figure 6 shows that the East Asian and Pacific region nearly attained this rate during the 1980s and 199Os, posting average GDP per capita growth rates of close to 6 percent. Meanwhile, Latin America has lagged substantially behind the East Asian performance, and Mexico i s no exception.29Yet, the fact that other countries have recently doubled their incomes provides grounds for optimism that Mexico can do the same. However, to do so it is necessary to continue boldly addressing competitiveness constraints inMexico. 29Real average GDP per capita growth rates for Mexico were 0.5 percent, 0.6 percent, and 0.2 percent greater than LatinAmerican averages during the 1970s, 80s, and 90s respectively (World Bank, World Development Indicators). xxv Figure6: Average Growth of GDPper Capita 8 -2 East Asia Latin Middle South Asia SUb-SEhFiran EUrDpe & Paclfic Amrice East & Africa & 81 NorthAfrlca Central Caribbean Asia Note: Data for the Europe and Central Asia Region begin inthe 1980s. Source: World Bank, World Development Indicators (as reported inWorld Bank, 2004a). 48. Doubling per capita income can have a tremendous impact on poverty. Evidence shows that countries experiencing higher rates o f economic growth can be expected to reduce poverty faster (see left panel of Figure 3). For example, over the 1981-2000 period, China's poverty rate fell from more than 50 percent to about 8 percent, thanks to an impressive per capita growth rate o f almost 8.5 percent a year. Similarly, between 1993 and 2002 Vietnam cut its poverty rate inhalf, from about 58 percent to about 29 percent, by growing at almost 6 percent a years3' 49. In technical terms, cross-country analysis has demonstrated that the average total elasticity of poverty with respect to economic growth is -2.4. This implies that, without controlling for any characteristics o f the country, 1 percentage point growth in a country's mean income can be expected to reduce the incidence o f poverty in that country by about 2.4 percentage point^.^' There i s evidence that relatively high income inequality in Mexic-in 2002 the Gini index was 0.49-could cause the country's true elasticity o f poverty with respect to growth to be closer to -1.6 or -1.7.32Evenbased on the lower inequality-adjusted elasticity, doubling per capita income could theoretically eliminate poverty ~~ ~ ~~ 30Perry and others (2006). 3'World Bank (2005g) based on an updated Ravallion and Chen (1997). 32This is the author's calculation basedon Lopez and Servtn's (2005) growth elasticity ofpoverty estimates, as cited inPerryet a1(2006). TheLopez and Serven formula determinesgrowthelasticity according to a country's per capita income, poverty line, and Gini coefficient. For Mexico, the author used the following 2004 data: Rural moderate poverty line per capita (US$ per month) $90; urban moderate poverty line per capita (US$ per month) $135; GNI per capita (US$per month) $564; rural population ("ho f total) 24%; urbanpopulation (%o f total) 76%; total population 103,795,200; share o f rural population inmoderate poverty 57%; share o f urbanpopulation inmoderate poverty 42%. This data enabledus to calculate a national moderate poverty line of $121per capita per month (this number is weighted by rural andurban shares). Hence, GNI per capita / poverty line = 4.6. According to the Lopez and Servtn table, this would equate to a growth elasticity o fpoverty between -1.6 and -1.7. Data sources include World Bank, World Development Indicators (population and GNI data); World Bank (2005k) for poverty data; World Bank (2005g) for the Gini coefficient. xxvi altogether. Although this is probably unlikely, it i s clear that large growth gains from increased competitiveness would leadto plummetingpoverty rates. VI. BUILDINGONPROGRESSTO DATE 50. The Government of Mexico has recognized that competitiveness i s a top priority, and as a result, numerous positive steps have been taken to date. These actions have often been part of Mexico's Competitiveness Agenda 2004-2006 (which this AAA helped define and implement). Many prior and planned future actions are also supported by the Competitiveness Development Policy Loan (DPL), which was an offshoot o f this AAA and approved inMarch 2006. The left side o f Table 4 details some examples of prior and planned actions o f the Government o f Mexico in areas o f the DPL that coincide with the themes o f this AAA. The right side of the table contains a summary of the objectives for moving forward that are covered by this AAA report. xxvii a B . . . . . . . . X .r( I X X . . . . . . Ex . . . . . I. MAKINGCOMPETITIVENESSAPRIORITY;BUSINESSASUSUALWILLNOT WORK 1. Can Mexico compete? This question has dogged Mexico inrecent years thanks to modest growth and productivity numbers, a declining share o f the US.import market, and increasing competition from lower-wage economies such as China. The objective o f this report is to inform policy makers, the private sector, and the public at large about competitiveness priorities and shortcomings inMexico, and to point out the way forward on improvingcompetitiveness. The report highlights constraints to competitiveness and identifies key high impact areas o f action aimed at maximizing Mexico's efficiency potential, generating jobs, and moving the country to a more innovation-driven stage o f development. The report draws on a series o f focused analytical pieces conducted for Phases 1 and 2 o f the Programmatic Analytical and Advisory Activities (M); it i s also based on extensive World Bank evidence and expertise on competitiveness issues in Mexico and beyond.33The report also recognizes that the Fox administration has taken important steps that have laid a solid foundation for future efforts. Yet, improving competitiveness is a long-term challenge that the next administration will face over the coming years. 2. The key policy messages that arise from this work are as follows: I.TosecurestrongjobcreationandsignificantpovertyreductioninMexico,increasing competitiveness should be a priority and business as usual will not work. Competitiveness improvements could allow Mexico to double its GDP per capita inten years, as a number o f countries have recently done. 11. Mexico needs to maximize its efficiency potential by a) encouraging competitionbetween firms, b) facilitating trade andreducing logistic costs, c) improvingbusiness regulation and the investment climate, d) strengtheningthe energy sector, and e) improvingthe information and communications technology (ICT) and telecommunications sector. Special emphasis i s needed to improve and transform the agriculture sector. 111. To advanceto the innovation-driven stage of development, Mexico needsto boost innovation and strengthen quality, education, and training (Box 12). Iv. Competitiveness priorities differ at the sub-national level and active reform at this level is necessary. Box 12: Stages of Development: Factor-, Efficiency-, and Innovation-Driven The efficiency and inno ven stage concepts come fkom the Worl where Mexico is classified as being inthe efficiency-driven stage along d Chile, for example. At the efficiency-driven stage, finns increase productivity by adopting efficient production practices. Among the countries inthe frst stage-the "factor-driven stage"-where firms compete onprice by taking advantage o f cheap capital, labor, and energy factors, are China, India, innovation-driven stage, firms need to produce innovative products using sophisticated (e.g., Finland, Germany, and Japan). 33Annex 1, Annex 2, and the background studies contain descriptions o f the World Bank's program to increase Mexico's competitiveness. 1 STRONG MACROECONOMIC MANAGEMENT BUTRELATIVELY LOW GROWTH 3. Mexico has shown lackluster growth and productivity performancein recent years. Over the last decade, Mexico has benefited from a relatively stable economy and has avoided the pattern o f financial crises that have plagued many o f its past political cycles, thanks in large part to prudent fiscal and monetary policies. Despite this stability, structural economic constraints have heldback growth and therefore poverty reduction. Growth rates have beenparticularly low over the last decade when compared with countries at similar income levels in East Asia. While real GDP per capita inMexico only grew at an annual rate o f 1.2 percent between 1994 and 2004, it grew at 7.7 percent inChina, and at 5.9 percent on average inthe East Asia and Pacific region. Productivity growth rates have also been modest. Annual labor productivitygrowth was negative inMexico between 1990 and 2004 (Figure 7), in sharp contrast to strong positive growth rates in Chile and Argentina. The share o f growth attributable to total factor productivity was also negative on average between 1980 and 2003, although there has been a slight increase since 1996 (Figure 8). Figure7: Annual Labor Productivity Growth, Figure 8: GrowthAccountinginMexico, 1960-2003 Early 1990s to Early 2000s (Contributionsto GDP Growthin percent) 0.06 0.05 0 GW Caprtal Labor w TFP 0.04 0.03 4 0.02 3 2 0.01 1 0 0 -0.01 -1 ` I I -0'02 Colon-bia Mexico Brazil Chile Argentina 1965-1979 1980-2003 1996-2003 Note: Argentina: 1991-2001; Brazil: 1990-2002; Chile. Note: Numbers are for real GDP growth (not GDP per 1990-2003; Colombia: 1996-2000; Mexico:1990-2004. capita), Source: Multiple, compiledby World Bank (2005~). Source: Faal (2005), 4. Given the productivity numbers, it is not surprising that comprehensive competitiveness measures show Mexico to be weak-underperforming-relative to per capita income and not improving34 The World Economic Forum's 2005-2006 Global Competitiveness Report gives Mexico a rank o f 55 out o f 117 countries, down from 48 in2004.35Incomparison, Mexico's gross national income (GNI) per capita rank within that same sample o f countries is 39. Although this gap between income and competitiveness rankings is not atypical for Latin America, it indicates that Mexico i s falling well short o f its potential. In contrast, Chile, India, China, and other East Asian economies enjoy competitiveness rankings that exceed their income placing (Figure 9). The World Bank's Doing Business in 2006 reports similar findings. Mexico ranked 73 (out o f 155 countries) on the publication's "Ease o f DoingBusiness" scale, compared with an income per capita rank o f 38. 34 Competitiveness is broadly defined as the "collection o f factors, policies, and institutions which determine the level of productivity of a country and that, therefore, determine the level of prosperity that can be attained by an economy" (World Economic Forum 2005, p.xiii.). 35 This ranking is for the Growth Competitiveness Index. Two countries (Qatar and Kuwait) that were not ranked in 2004 placed aheado f Mexico in2005, accounting for a small part o f Mexico's fall. 2 Figure 9 :Competitivenessversus GNIper Capita Figure 10: Share of U.S.imports,Mexico versus Rankings China 120 16% 100 14% a13d280 12% aa -egE EO 8L 20 u) 10% 40 Ii 2 8% 20 6% Un 0 0 20 40 BO BO 1 w 12c 4% Grovth Comprt~hwwssIndex Rank, 20052006 9Qb`9 hpQb\q9` hp9%`9992000200` 200%20o=20ob200% 8 Note:Lower-rankingnumbersare better. Source: World Economic Forum (2005) and World Bank, Source: us Census Bweau, ForeignTrade World DevelopmentIndicators. Statistics. 5. An often used (albeit imperfect) proxy for competition is export performance. On this basis, the weak competitiveness of Mexico's economy would be reflected in the reduction in its share of exports to the United States. Mexico is highly dependent on the U.S. economy, which absorbs 88 percent o f its total exports. After experiencing an export boom inthe 1990s thanks to NAFTA, Mexico has been losing U.S.market share, particularly to China (Figure 10). Mexico's share o f U.S. imports fell from a higho f 11.6 percent in 2002 to 10.2 percent in 2005. Duringthat same time period, the Chinese share went from 10.8 percent to 14.6 percent. According to the governor o f the Banco de Mdxico, lost competitiveness has caused the economy to lose an additional US$30 billion in exports to the United States and has restrained economic growth. Particularly worrying i s the fact that Mexico i s losing U.S. market share in heavy goods such as fbrniture and household appliances, in which it should have a transport cost advantage due to its proximity to U.S. markets. While one concern i s that Mexico suffers from high labor costs, this i s not the whole story. A recent World Bank analysis o f the computers and peripherals market found that while labor accounted for 4 percent o f costs inChina, it only accounted for 3 percent in Mexico. In non-leather apparel, labor accounted for 15 percent o f costs in China and 20 percent inMexico. Clearly, there are other factors impeding Mexicancompetitiveness. 6. The Mexican Competitiveness Institute (IMCO) has also shown that the country i s lagging. Out o f 45 economies that IMCO defines as Mexico's competitors, the country ranks inthe bottomthird at number 31, behind Brazil and Thailand among others. IMCO considers ten separate features to be critical to a country's competitiveness, measured as the ability o f a country to attract and retain investment, local and foreign. O f those, Mexico falls particularly short on governance and legal system indicators, foundational or "precursor" sectors (transport, telecommunications, and financial services), efficient factor markets (capital, labor, and energy), human capital (education, knowledge, and innovation), and environmental sustainability. Simulations by IMCO show that if Mexico were to improve by 10 percent in all the ten features critical for competitiveness, investment could increase by as much as 60 percent. Among the ten features, five (efficient government, trade and international relations, human capital, infrastructure and governance, and legal system indicators) account for 70 percent o f this increase.36 36 IMCO (2005b). 3 7. Reforms are needed to e ~ c o ~ r investment. Mexico's savings rate, at about. I 9 pcrceiit of a ~ e CDP, IS on par with the rest of' Latin Xnxxica, but i s i ~ ~ s u f ~toc provide~the i ~ ~ e sneeded~toe ~ ~ ~ ~ e ~ i t ~ boost fong-term growth (Table S), ~ o ~ ~that~have rbeen~ables to break out uf low grotvthpaths in she ~ i early stages of ~ e v e ~ o ~such easnEast Asia and Chile, typically have done so with higher rates of ~ ~ ~ , rnvestrnenf in fixcd capital f o ~ a ~ ~ o n , Table 5: Savings, Investment,and the Current Account (as percentaf GDP. average ~ 0 ~ ~ t 1 t Gross fired Current tt Gross national capital :: account I I savings formation ' b2tlane.e i s j ~ ~ ~ for growtha and,~ henre, poverty red^^^^^^. Nearly a11 the o ~ t ~ longtenii SUCCGSSin reducingpoverty have done so with higher rates~ f ~ ~ ~ ~ per capita over thc last decade (right parici of Figure 1I).N P per capita hg-owth~ o l ~ othe ii994gTequila Crisis, income ~ ~ niore or tcss stagnant since 2000. Thc rcccnt growth slotvdow~ihas meant signific rcseductron o p ~ o ~ u ~ ~~~h~~~~~~r a ~ s 8s ~ ~ ~ ~ e sp r, o ~ such ~ hatie l ~ e l p7~offset this trends3$ d ~ For ~ ~ Mexico lo breakout of its low-growth trajectory and cut poverty, it wiII have to boost growth arid p rviry and c ~ ~~ i ~~~ ~ i~~ ~ v~ e ~ ~~ s s ~ Figure 11: Poverty reductlatiis closely associated with growth - World Mexico P f7,ffOO - T 0% 10. Mexico has made progress in many human ~ ~ ~ e indicators~and in~thce~ n~ ~d ofu ~ o ~ ~ ~ ~ o ~ extreme poverty, but poverty and inequality remain serious issues, Income per capita (GNI per capita) IS WSS6,790, among thc highest in Latin Anierica; life expectancy at birth has risen to 74 years; the under-fivc ~ ~ o r trate~droppad from 44.2 to 25 per thousand bctwecn 1990 and 2003; about 59 ~ i ~ y percent of the ~ ~ phas ~acc l to~ekean~water; and the~literacy rate is ovcr 97 percent. IIowever, ~ o ~ dcspitc ortan tan^ a ~ ~ ~ i e ~in~ health~ andnet d~u ~ ~ ~andisotfiei progress in exfrenx poverty rcduetion e ~ e ~ o ~ over the last five years, low arid i ~ e q ~ ~growth~ hase kept poverty rates high,The percentage ~ f p r ~ p l e ~ a l poverty was rcduccd from 24.2 percent in 2000 to 17.3 percent in 2005. Still, the Girti vered betwen 50 and 55, on par with Latixi ~ ~ ~ r javerages c a n to 2004 data, 47.7 percent of the ~ ~ ~p o~ ~ ~ ~ ~a poor. o~ n richest tcrt percent 1 is ~1~ The ~ s of thc population cams c)\rtr30 percent o f tatai incomc, while ehr poorest ten percent earns only 1.Ipcrcent. 5 Inipact on investmentper worker of a I Subject areas 10 percent improvementin the lariablrsbehindeach subject area 11. it should bc IIOI~~that the impact xititnbers in Tablc 5 arc for single effects or isolated ~ ~ ~ ~ r o \ ' ~in~rachecategory. A broad pac i ~ n t s of reforms across most of thc h i ~ ~ "categories ~ ~ ~ i ~ p could havc a positive b u ~ d leffect. ~ n ~ inother words, the ~ ~ t e r abetweenthe individual variables c t ~ ~ ~ should create an overall effect that i s ~ ~ ~ s largerathan the suma of~the~i ~~d ~ ~ ~ d M a ~ ~ ~ ~ i i ~ p ~ o ~ F(supereadditive effects). e ~ ~ t ~ 13. The conclusions of the impact elasticity c a ~ c ~ l ~ are os ~ s~ t ~ ~ byptwo~other relatively d ~ ~ ~ cnsise (albeit 'bwfter"} reports: a C ~ n i ~ e ~ i ~ ~FocusnGroups report and the World v e e ~ ~ World Bank report based on the written and verbal responses over 90 business pcopk in o from a range of ceorio~~jcsectors ia 3003. The R'EF report conta ~ ~ ~ f about~tkcnmcrs ~ ~ ~ n o ~ ematit factors for doing business in hiiexico. These sources are "'SO are based mostty on subjective evnfuarro of e ~ ~ r ~ p r ~ Iwhiie~ the ,IMCO aiialysis is "hard' in the seiisc that, by and ~ c r s large, it uses hard f tuai data and provides ~ ~ ~ a n t i ~estimateseof impact, 'tt is r c a s s that~ the~ ~ ~ a t j ~ ~ ~ ~ correlation between the hard and soft srudies IS quile high(over 80 percent). 14. Similar conclusions regarding obstacles to competitiveness in Mexico are derived through a "growth diagnostic" conceptual framework. This framework, developed by Hausmann, Rodrik, and Velasco (2004), aims to identifythe binding constraints to growth. Although the diagnostic has not been comprehensively applied to Mexico, it does provide guidelines. The framework is basedon the drivers o f investment and uses a decision tree to first ask whether growth i s constrained by low returns on investment, the highcost o f finance (both local and international), or both. The cost o f finance (reflected particularly in very limited access) is a key issue in Mexico, considering that the share o f credit to the private sector as a percentage o f GDP is about 17 percent, while it i s 63 percent in Chile, 98 percent in Korea, and over 100 percent inmany East Asian countries. This issue is addressed separately through a separatereport on the financial sector. 15. Turningto the returns on capital, an analysis by IMCO shows that low returns are a large part o f the story. Based on an analysis o f the 33 largest companies on the Mexican stock market (BMV), IMCO found that, on average, profits did not cover the cost o f capital (Figure 71). Specifically, only 6 o f 14 sectors had returns that exceeded their cost o f capital. Moreover even in absolute terms, the returns to capital appear low. They are inthe 5-1 2 percent range, which i s similar to the yield on government bonds in Mexico. The growth diagnosis framework provides guidance into the critical issues that are keeping those returns so low. And this AAA supports measuresthat ought to improve the returns to capital. Figure 13: Cost of Capitaland Returnson CapitalinMexico --.-- Retorno s/Capltal WACC Retorno Promedio --- WACC Promedio 15% Note: WACC stands for WeightedAverage Cost of Capital. Promedio means average. The data includes the 33 companies that compose the mainMexicanstock market (BolsaMexicana de Vafores) index (IPC). Source: IMCO (2005a) 16. The growth diagnostic framework goes on to identify a number of possible drivers to low private returns on domestic investment. They include the following: a) highmicro or macro risks; b) lack o f innovation and R&D; c) large externalities, spillovers, or coordination failures; d) hightax rates andor inefficient tax structure; and e) insufficient infrastructure, high transport, energy, telecommunications or shipping costs (see Annex 4 for a diagram o f the framework). The macro environment has been stable in Mexico for the last decade, making it unlikely to be a bindingconstraint. Micro risks, such as from corruption and crime, could potentially be binding constraints. Indeed, they came up in competitiveness focus groups held in Mexico and the World Economic Forum's executive opinion survey. However, they are not covered in this AAA since there is another Bank analysis o f this 7 subjects3'Other micro risks, such as property rights and judicial reform (contractual enforcement, are handled through two other Bank instruments. Lack o f R&D / innovation i s certainly an issue inMexico and i s addressed by this AAA. Tax rates and regulations are seen as a constraint with significant impact, but the issue has also been looked into in another Bank analy~is.~'Constraints associated with infrastructure and transport and shipping costs are addressed in the logistics, trade facilitation, and energy components o fthis AAA 17. In consequence, Mexico needs to take action on various fronts: innovation and technology development, quality, training, investment climate, trade and trade facilitation, logistics, energy, ICT and telecoms, and agriculture. These subject areas, as shown, are the result both o f extensive analytical work that the Bank has done for Mexico over the past two years and of other complementary studies done outside the Bank, in addition to the above-mentioned studies. Selected findings in each o f the priority areas are shown in Table 7. The points are discussed in more detail in the policy area descriptions in Section III. evidence complements the IMCO impact elasticities shown in Table 6. Detailed This analytical underpinnings are provided inAnnex 3. Table 7: Analytical Underpinnings-Selected Evidence Innovation Roughly half o f cross-country differences inper capita income and growth are drivenby and differences intotal factor productivity (TFP), generally attributed to technological development Technology and innovative capacity (Hall and Jones, 1999; Dollar and Wolf, 1997). Development Calculations suggest a social return on R&D o f 60% inMexico (Lederman and Maloney, 2003a). Mexican f m s invest a little above 0.1% o fGDP inR&D, compared with 1.5% inthe OECD area. Quality A study on Germany found that standards contributed to about 0.9 percentage points out ofan average overall growth rate o f 3.3% inthe period 1960-1990 (Jungmittag, Blind, and Grupp, 1999). Adopting 100 additional Britishstandards raised the U K ' s export to import trade ratio by about 14% (Swam, 1996). Education Less than 1%o f Mexican students perform at the highest level o f international student and Training assessments (Programme for International Student Assessment [PISA] Level 5) ineither reading or math, while more than halfperform at Level 1or less inboth (OECD, 2003a). These numbers are quite low given Mexico's income levels. In2000,24% ofMexican adults lackedsecondary education, eventhough they hadcompleted primary schooling. Receiving a secondary education certificate leads to higher employability and higher wages (World Bank, 2004e). Lack o f access to financing and skills are the main weaknesses o f Mexicansmall andmedium enterprises (SMEs).Only 35% o f new SMEs survive more than 2 years, andthe f m s that fail report lack o ftraining and expertise as key reasons for their failure (Secretariat o f Economy- Mexico, 2001). Investment Modernizing regulatory frameworks could increase productivity by as much as 10 % inlagging Climate: sectors (Cotis, 2005). Regulation Starting a business takes 58 days inMexico, compared with 5 days inthe UnitedStates. Dealing with licenses takes 222 days at a cost o f 159% o f income per capita, while it only takes 70 days inthe UnitedStates at a cost of 17% ofper capita income (World Bank,Doing Business database). B The cost o fthe regulatory burden inMexico is 15% o f GDP (IMCO, 2005a). Competition D A study ofthe impact ofpro-competitiveregulatory reform inseveral industries inthe United States found that annual welfare gains inaffected sectors were more than 7%, with 90% of the 39The Mexico Institutionaland Governance Review. 40The Mexico CEM: Challenges and Prospects for Tax Reform. 8 benefits flowing to consumers (OECD). Increasing competitive pressure can increase the probability o f fminnovationby more than 50% (World Bank, 2004a). According to a survey o f Mexican f m s , the two largest obstacles to business development are government monopolies and private monopolies (CEESP, 2005). Trade and Trade facilitation reforms inMexico could increase exports by $29.3 billion, equivalent to 21 Trade percent o f total Mexicanmanufacturing exports (Soloaga, Wilson, andMejia, 2006). Facilitation Mexico's global exports would have been about 25% lower without NAFTA, and FDIwould have been about 40% less. Also, the amount o f time requiredfor Mexicanmanufacturers to adopt U.S.technological innovations was cut inhalf withNAFTA (Lederman, Maloney, and Serven, 2003). Despite the gains from NAFTA, Mexico has been losing export market share within the United States inrecent years. O f the top ten types o f Mexican imports inthe United States, only two- .petroleum and total agricultural products-gained market share between 2001 and 2005 (US. Dept.ofCommerce, see Table 33). Export-supplycapabilities inMexico were more o f a constraining factor for Mexican exports than import-demand conditions inthe U.S.(Hanson and Robertson, 2005). Logistics costs inMexico are estimatedto be 20%, while the OECD average is 9%. A reduction Logistics to 12% inMexico would have a significant impact on economic activity andjob creation (Guasch, 2004; Guasch and Kogan, 2003). Mexico is not capitalizing on its enormous competitive advantage o fproximity to the United States because logistic costs are so high: as an example, Mexico spends almost double what the United States does on transport as a percentage o f GDP (IMCO, 2005a basedon INEGI)). Energy Inadequateinvestments inenergy infrastructurehave had a significant impact on TFP growth in Mexican manufacturing (Casteiieda Sabido, 2003). The prices o f fuels and electricity inMexico are among the highest o f the world. Incontrast with other OECD and developed countries, the prices o felectricity, natural gas, fuel oil and other .. fuels have increasedover the last decade. (World Gas Intelligence, EIA, EdisonElectric Institute, Eurelectric, among others). A large portion (64 percent) of electricitysubsidies are captured by the non-poor inMexico (World Bank, 2004f). The quality of the electricity service i s poor compared to other countries. Service interruptions and voltage fluctuations affect the overall productivity levels o f industries and prevent the installation of modemelectronic equipment (World Bank, 2005d; IMCO, 2005a; and others). ICT and Countries with higher levels o f investment inICTs experience higher economic and social r e1ecom .. development growth (OECD, 2004~). Mexico's level o f ICT expenditure (3.1%) as a share o f GDP is substantially lower than Latin American countries such as Chile (6.7%), Brazil's (6.7%), Argentina (5.7%), and Peru (6.9%) (World Bank, World Development Indicators, 2003). BBusiness telephone charges (factoring ininstallation costs, monthly fees, andper minute rates) inMexico are over 3 times greater than charges inArgentina and4 times greater than chargesin 4griculture BAgricultural GDP grew only 1.3% between 1984 and 2004, while total GDP grew 3%. I Agriculture as a share o f the total Mexican economy shrank from over 8% inpre-NAFTAera to 5% in 2004. Public expenditure inagriculture declined from 13.4% to 6.9% between 1990 and 2003. l Out o f the top 24 export agricultural commodities from Mexico to the United States, only 9 are increasing market share. The rest are losing market share to competitors. tNegative net trade balance (exports minus imports) for agriculture deepened from $211 million in 1991/93 to $1.3 billion in2001/03. sub-national tAguascalientes ranks 33rd globally in the costs to start a business, while the Estado de Mexico ranks 127" (World Bank, 2006b). tThe impact on investment per worker o f a 10% improvement in competitiveness variables ranges from 841% to 12%, depending on the Mexicanstate (IMCO, 2006a). 9 11. MAXIMIZINGMEXICO'S EFFICIENCYPOTENTIAL 18. For Mexico to compete in what the World Economic Forum (2005) calls the "efficiency-driven stage" o f development, Mexico needs to develop more efficient production processes and reduce costs throughout the economy. To reduce costs and improve productivity, Mexico should focus on a) encouraging competition between firms; b) facilitating trade and reducing logistic costs; c) improving business regulation and the investment climate; d) strengthening the energy sector; and e) improving the ICT and telecommunications ~ector.~'These are all high impact areas, as shown in Table 6. Mexican authorities are supporting improvements in most o f these areas,42but further progress is necessary. It i s also important to address inefficiencies in the agriculture sector (f). Each o f these areas i s discussed in the following subsections. IMPROVEMENTS INMEXICO'S COMPETITIONPOLICY ARE NEEDED, PARTICULARLY TO FACILITATEENFORCEMENT SERVEAS AN EFFECTIVEDETERRENT h'TICOMPETITIVE AND TO PRACTICES. Description and Challenges 19. Competition, and the benefits it brings with respect to lower prices, greater variety and the development of new products, has been recognized as crucial for markets to work well and for increasing the wellbeing of a society. Policies that enhance competition are expected to bringnot only static improvements in efficiency, especially in disciplining former monopolies, but also dynamic improvements in efficiency, through improvements in productivity and innovation. Recent empirical research indicates that regulation that lowers entry barriers and state control increases productivity and technological innovation, stimulates business investment, and increases long-run employment rates.43 Therefore, reforms to enhance competition could have significant effects on growth. A study o f the effect o f pro-competitive regulatory reform in several industries in the UnitedStates found that annual welfare gains in the part o f GDP affected by reform were more than 7 percent, with 90 percent o f the benefits flowing to consumers.44Increasing competitive pressure can increase the probability that firms will innovate by more than 50 percent4' 20. Competition enforcement is relatively recent in Mexico. Although Article 28 of the Mexican Constitution o f 1917 prohibits monopolies and monopolistic practices, competition policies and their implementation are relatively new in Mexico. It was only in 1993, during the important wave of reform o f the mid 1980s, that the Federal Law o f Economic Competition (Ley Federal de Competencia Econdmica) was enacted and the Federal Competition Commission (Comisidn Federal de Competencia, CFC) was created. Regulations to the law, that defined the actual procedures that the CFC would use to implement the law, were later published in 1998. 4'The topic o f taxes is not a subject analyzed inthis report since it is the subject o f the Mexico CEM: Challenges and Prospects for Tax Reform. Likewise, the topics of education, corruption and finance are addressedelsewhere through a parallel AAA on Education; an Institutionaland Governance Review, and finance AAA, including an FSAP(Financial Sector Assessment Program). 42Some o f these improvements are being supported by Bank operations such as the MX Competitiveness DPL and the MX Finance and Growth DPL series. 43Nicoletti and Scarpetta (2003) Nicoletti and Scarpetta (2004). `'World 44OECD (2002). Bank (2004a). 10 21. The Competition Law defines a range of responsibilities for the CFC. The Competition Law prohibits either absolute (price-fixing, output and distribution restrictions, market division, and concerted bidding inpublic tenders) or relative monopolistic practices if they demonstrably limit competition; the latter includes vertical market division, restrictions on resale, tied sales, exclusivity contracts, boycotts, predatory pricing, discounts, cross-subsidies, imposition o f different prices or sales conditions o f goods or services. The Competition Law provides administrative sanctions, includingcorrective conduct orders and fines, for monopolistic practice violations. The Competition Law investigates concentrations including mergers and acquisitions that combine corporations, associations, partnerships, shares, interests, assets or trusts between competitors, suppliers, customers or any other economic agents. Only large concentrations require economic agents to notify authorities prior to their completion. The CFC assesses the legality o f concentrations and could sanction them by ordering partial or full divestiture and fines. The CFC also oversees State regulations and cancels them if they constitute interstate trade barriers. 22. The CFC is an independent agency of the Federal Government with technical and operational autonomy but i s attached to the Secretariat o f Economy for budgetary purposes. The CFC does not collect any fines but rather they are remitted to a general treasury fkd. The CFC's President and four commissioners are appointed for staggered ten year terms by the President o f Mexico. This provides independence since the commissioners are protected by the duration o f their tenure and avoids the usual turnover during change in administration every six years. The CFC has a plenum, composed by five commissioners, o f which one i s the president. Resolutions o f the plenum are adopted by the majority vote o f commissioners, the president having the casting vote. 23. A number of other agencies also play a role in ensuring competition. In addition to the CFC, competition and consumer protection is also enforced through the Federal Consumer Protection Law by the Federal Prosecutor for Consumers (PROFECO). The objective o f the Consumer Protection Law i s to promote and protect consumer rights and procure equity and legal security in relationships between suppliers and consumers. Also, regulatory commissions have been established for specific economic sectors: the Federal Telecommunications Commission (COFETEL), which i s a decentralized agency (6rgano deconcentrado) o f the Secretariat of Communications and Transportation, to regulate telecommunications; the Energy Regulatory Commission (CRE), a decentralized agency of the Secretariat of Energy, to regulate electricity and gas; the National Water Commission (a decentralized agency o f the Secretariat o f the Environment and Natural Resources, SEMARNAT) to administer and preserve national waters; the National Insurance and Bonding Commission to regulate insurance and bonds; the National RetirementFund System Commission that regulates pension funds; and the Federal Regulatory Improvement Commission (COFEMER), a decentralized agency o f the Secretariat o f Economy, to guarantee transparency inthe development and application o f regulations, as well as ensure that the benefits o f regulations outweigh their costs to society. The transportation sector is regulated directly by the Secretariat o f Communications and Transportation. The Secretariat o f Finance and Public Credit, the National Banking and Securities Commission, and the Mexican central bank regulate the financial and banking sector. Although these institutions can impose price controls to limit monopolistic power, the CFC has first to determine if there i s an absence o f effective competition inthe market. Also, although these other "regulators" can award concessions, licenses, or permits in their regulated sector, they couldonly awardthemafter the CFC has given a favorable opinion. 24. OECD assessmentsshow competitiondeficienciesinMexico. An OECD assessment of Mexico's competition law and policy completed in 199846concluded that the law was well-conceived and that the 46 OECD (1999),Background Report on the Role ofCompetition Policy inRegulatory Reform, inOECD, Regulatory Reform inMexico. 11 CFC had strong e ~ ~ ~ ~ r cpowerseandt portan^ an^ a u t ~ too affect direct market regulation by other c n ~ ~ ~ ~ ~ There was, fiowcter, rio clear base o f support for c a ~ ~ p epolicy, ~arid ~the CFG's t i ~ ~ rit record could be ques~ioned.Since then, a new review by the OECDJ7~ o ~ that~ the~ ~ e d d ~ in tRc~ Erst review still perfain, butethere were now no doubts about the CFC's ~ ~ ~ ~ willingness tu engage powrful economic interests, It also asce ed that the CFC had matured into a credible and ~ ~ ? e ~ agency ~ l * r that~had c~~npiledrema ~ ~ ~ ~ a~ d c record of ac~iic~~ciiicn~ the given ~ i f ~ c ~of~ itIs~e ~ ~e s ~ ~ It ~ ~ ~g h~~ i ~howe\~er,~ the~ f, ~~ ~ ~~ lcfialiengcs~and~~ p ~ u ~ ufor ~ ~ i e s ~ ~~ ~ e d eo ~ ?~ n n . ~ ) the limited ciegee of general support for c o ~ i p ~paiicy;~ (~ii)acertain dciiciencics in ~ j ~ rjty and judicial review processes that constrain the CFC's ability to address ~ ~ ~ o r ~ and a(iv) ; some crf thc CFC's p r a e e ~ u ~and~ methods o f interface with ~ Q d e entities that reduce ~ t efficacy as a law e ~ ~ o r ~ e Iagency~ and c o ~ p e ~advocare. n s n c n ~ ~ i ~ OECB assessmc~~s discussedabove, and dcspitc Mexico having n2 policy, there are scvernl nstraints that still h i t c~ni~ct~t~un. ECB barriers-to-campctiriorl indtcalor-a nieasure o f I 1 b a ~to~~ ~~ r~ ~~ and e ~antitrusti o ~ i t ions for ~ u b enterprises--- ~ ~ c shn\vs that this index f2as d e ~ ~ r ~ o ~inaMexico and &at the country has thr: highest barriers to ~ c d c n n i ~ camong~CXCD)counrrics i ~ i g u r cI41.'' ~ ~ ~ ~ ~ Note. Lower rankingnumbersw e bcrter. above Greece (2.0) and Argentina (1.5) since Mexico's competition regime i s considered to be severely undercut by 26. Limitationsto the CFC's actions arise from two types of proceedings at the FederalJudiciary that are brought by dissatisfied parties seeking judicial relief. These two proceedings, the amparo actions ("protection appeals") against the processes and resolutions o f the CFC and the fiscal nullity actions at the Federal Administrative and Fiscal Justice can take many years to resolve and have proliferated over the years. On average it takes 401 days for these suits to be resolved, and between 1994 and 2004 the number of suits (amparos and fiscal nullity) increased from 41 to 170. (See Box 13 for a description o f how amparos have prevented the enforcement o f injunctions against Telmex.) Even if the CFC's orders survive the amparos and fiscal nullity actions, the imposed fines are not collected by the CFC but rather by the fiscal department o f the municipality o f residency o f the fined party. From 1993 to September 2005, only 3.3 percent o f the total amount fined by the CFC was collected, while 72 percent i s under reconsideration and 20 percent i s under amparo and fiscal nullity action. Box 13: Shortcomings in CompetitionEnforcement: The Case of Telmex The Federal Telecommunications Law empowers COFETEL obligations on a fm that has been declared by the CFC t Federal Competition Law. These obligations may include aggravatedby the abuse of the a COFETEL has not imposed the ob1 sanctions by repeatedly appealing th its resolution in 1998. However, Telmex still filed a number of amparos to delay the enforcement of this decision. COFETEL eventually issued a resolution establishing certain obligations for Telmex in 2000, but Telmex also appealed this resolution through judicial court annulled the origin to obtain the annulment of all acts and resolutions that derived from had to restart and redo the original proceedings ino When the CFC finally issued a new resolution dec resolution and eventually won the amparo in May which was likewise Source. Solano. del 27. Another importantlimitationof the CFC is that it lacks many powerscommonto competition commissions in other countries (Figure 16). Among these is the ability to divest monopolies, undertake searches, implement leniency programs, and order preventive measures. Due to limitations in the Competition Law, the CFC is not able to prosecute and divest or impose price controls on a dominant 5' Global CompetitionReview, Volume 6 Issue 6 June 2003. 52 An amparo i s a remedy against acts by any authority that violate any of the individual guarantees recognized by the Mexican Constitution. It is available in all legal matters and may be invoked in criminal, civil and administrative trials. If CFC's resolution imposes a fine, then the private party can request a review o f the Federal Administrative and Fiscal Justice Court, which mostly considers tax cases, but also assertsjurisdiction to review any agency. 13 firm that restricts supply and raises prices. Also, without the use of search and seizure authority, it is difficult for the CFC to detect surreptitious collusion. The lack of a leniency program for the CFC restricts its ability to provide reductions in monetary penalties to conspirators who voluntarily reveal collusive agreements. Figure 16: Powers of Selected Competition Commissions USA European Chile Australia Union Preventagreements amongcompetitors Prevent monopolistic practices Merger control Prevent predatory pricing Monopoliesdivestiture Undertake searches Implementleniencyprograms 28. The CFC has had limitedcoordinationin the past with other regulators at the federal, state, and municipal levels. Given the powers of the many sectoral regulators, of other agencies such as COFEMER and PROFECO, and o f state and municipal entities, the CFC could improve its effectiveness by working more closely with them. The CFC could benefit from monitoring the sectoral regulator's regulatory postings and comment on them. Likewise, the CFC is not taking advantage of data available from other regulators, such as PROFECO, that could be useful in detecting collusive behavior. This collaboration and taking advantage o f the possibility o f exploiting synergies could also prove useful given the declining budget o f the CFC. Although the number o f cases has increased, the CFC's budget has fallen 9 percent inreal terms between2000 and 2004. 29. Competition policy and the role of the Federal Competition Commission (CFC) are being strengthened. Amendments to the Federal Law on Economic Competition-that reform 25 articles, repeal 1, and add 11 more-were approved in April 2006 and went into effect on June 29, 2006. The amendments are designed to increase the effectiveness o f detecting and sanctioning monopoly practices. They classify new anticompetitive actions, increase fines and expand the capacity o f the CFC to divest the assets o f firms that engage in anti-competitive practices, in cases where they have not complied with previous orders; undertake verification visits; implement leniency programs; and order preventative measures. 30. Other CFC actions include strengthening of internal procedures, cooperation agreements with other institutions, and actions to build general support for competition policy. See Mexico 14 Competitiveness Development Policy Loan, Report No. 34570-MX for a full description of these activities. MovingForward 31. Mexico needs to improve its competition policy through a variety of actions. To improve the CFC's ability to address anticompetitive conditions and improve the effectiveness o f competition policy, actions are neededthat will do the following: increasethe general support for competition policy, remove the constraints in statutory authority andjudicial review processes, make more efficient use o f the CFC's declining resources, and enhance the CFC's collaboration with other government entities. These areas are based on the various reviews o f competition law and policy in Mexico, empirical evidence, and an assessment by the Bank. If appropriate actions are taken, substantial gains are possible: according to IMCO's research, a 10 percent improvement in the variables that compose the competition dimension could increase investment per worker inMexico by 7.5 percent (Table 6). 32. The amendments to the Federal L a w on Economic Competition have been a great advance, but the associated regulation and its application are important. To make the amendments concrete, the corresponding regulation needs to be issued, which is expected to occur by the end o f December 2006. Training and diffusion efforts will thenbe necessaryto support the implementation and application o f the amended law. Training and diffusion efforts will then be necessary to support the implementation and application o f the amended law. 33. Although there have been advances, more collaboration between the CFC, different regulatory agencies, and states i s needed. The CFC has established cooperation agreements with 15 states and with COFEMER, CRE, and PROFECO but more advances are needed. For instance, a planned pilot program that focuses on cooperation between states should be pursued. Regulatory agencies should also increase their interactions to improve the diffusion o f competition policy, and cooperation schemes between regulatory agencies and states regardingeconomic competitionpolicy should be encouraged. MEXICO'STRADEENVIRONMENT BEENTRANSFORMED HAS BUTMORESHOULDBEDONETO FACILITATE TRADE AND REDUCELOGISTICS COSTS Description and Challenges 34. The trade and investment environment in Mexico has been transformed over the past two decades. This is in part due to the 1994 launch o f the North American Free Trade Agreement (NAFTA) with the United States and Canada, and, more recently, trade agreements with the European Union and Japan.53Growing competitive pressures from countries such as China have also engendered policy reform measures to lower trade transactions costs to expand export opportunities for Mexicanfirms. The engine o f export-led expansion inthe secondhalfo f the 1990s, however, has turned to lower growth rates inrecent years. Exports have diversified away from petroleum and relatedproducts, yet Mexico remains heavily dependent on a single market-the United States. 35. NAFTA has increased Mexico's exports to the United States, but exports to other regions remain underdeveloped. After implementation o f the NAFTA, the growth o f manufacturing exports from Mexico intensified as maquiladora operations along the U.S. border expanded. Since the early 53 Currently, Mexicohas signed 12 Free Trade Agreements with43 countries inNorthAmerica, LatinAmerica, Europe, and Asia. 15 1990s, manufacturing has accounted for over 70 percent o f the total exports (Table 8).54 In addition, although the NAFTA has had clear and demonstrable positive impacts on growth and employment, the advantage o f preferential access to the U.S. market is being eroded by international competition. Mexico has also had limited success in establishing strong relationships with European and Asian firms and expanding exports inthese markets; this is inpart due to the lack o f incentives for maquilas to establish strategic alliances outside o f NAFTA. Continued high trade transactions costs associated with reaching these markets for Mexican firms is one important contributing factor inhibiting progress in export expansion inthis broader picture. 36. In 2003, China surpassed Mexico as the United States' second largest trading partner after Canada. Mexico i s confronting competitive challenges from China and other countries not only interms o f labor costs but also in areas directly related to trade facilitation. These include, among other factors, wider use o f information technology in trade transactions, streamlined regulatory requirements on imports, and assistance to small and medium-sized firms intrade logistics. Table 8: Mexico's Trade Indicators Imports o f goods fob (US$ bn) 174.5 168.4 168.7 170.5 I 196.8 Current-account balance (US$ bn) -18.6 II -17.6 -13.3 -8.5 I -7.4 Source: Economist Intelligence Unit (2005) 37. Trade indicators inDoing Business highlight room for improvement. Inthe World Bank's most recent Doing Business report (2006), a "trading across borders" indicator compiles procedural requirements for trading a standardized shipment o f goods across national borders. The detailed components o f this indicator suggest that Mexico's performance remains at the Latin America and Caribbean region average. Considering Mexico's geographic proximity to the United States, the time for export and import at 18and 26 days respectively indicates a beat deal of work ahead to facilitate trade. The data for the OECD countries are 12 days on average for exports and 14 days for imports to clear borders (Table 9). 54 Economist Intelligence Unit (2005). Country Profile Mexico. 16 38, Itlexico fails short of Canada's trade ~ n ~ i c a t oand bas mixed results vis-$-vis China, ~ s Cctnipared with its NAFTA alliance partner, Canada, Mexico still has ~ ~ ~room to inzprove in a ~ i ~ c ~ ~ ~ or cxa~~ptc, ~ ~ u nof~ signatures required in Mexico for cxports is 4, compared thc ~ c r and rime for export is 18 days white it is 12 days in Canada. ~ ~ ~ i i a ~required u r e s in s tcmi 11 while 011 equiscd in Canada. "lhe t Itime for ttlll~ortis 26 days a s s u ~ i ithat~ the ~US will reniain a major impo ~ i ~ xico to ~ ~ ~to approach~Canada's efficiency ~ ~ ~ ~ i u c le NAFTA alliaiicc. ~ o ~to China,~Mexico is more ~ ~ ~ r e uircd for cxport, as well as d c t c ~ ~ i ifur~ I ~ s ~ import, .vi.hile China is rn the better position in terms ctf the iiurrtber of ~ i ~ ~and days~required for ~ ~ r e s inZports. 1 'l'able 9: Doitrg Bircitiess Trade Indicators i I Indicator Mexico 1 1 1 Region LAC China Canada OECD 5 3 11 Sour 39. Tariffs impose welfare losses on citizens. Ktt, Nicita, and Olarreaga (2004) providc estimates of trade efasticies for 4,625 ~ ~ goodsp in t 17 countries, as inputs into dihc UveraI1 'I'rade ~ ~ ~ ~ ivemss fndcx (UTKD, ~norder to study the et ts of tariffs on gross domestic p r o ~ (GDP).55 u ~ ~ 'Rf factors all trade r c s t r i ~ t j ~ands ~ ~ c f e r c nand~takes them into an overall c o ~ s ~ ~ l ~ ~ a t ~ ~ ~ c ~ nt tariff' rate.5hIn contrast. thc ~ ~ d c ~ sTrade~~ I ~ ~~ ~~ cr ~Index (TRI) is based on s s o ~~ ~ ~ ~ ~ e n c kx~ce.~' The OJ'RI data suggests that 11 er elasticities, whitc increase as the l c \ d o f ~ i s a g ~ ~ ecrcascs, iat~which elas ~ a ~ ~ ~ are estimated, 15'1th t around US$?billion ~ ~ o f GDY . ~2UW) Gper year, the~ ~ Stares 1s the country inthe sample whose tar mcturc iniposcs thc largest welfare costs its citizens. Mexico has the third largest tstelf'tirc loss at 2 billion (Q.3?4 o f GDP in 2.004) tvith ir cxrsting tariff structure III the overall sample, 40. Evidence s ~ that internal~~ ~ ~ ~ ~ared~~ i ~ ~t iMexico's~export growth. Hsnson atid ~t j in~ ~ n ~ Robertson (3005, ~ n h e o suggest that in the 19913s i ~ n p o ~ - ~conditions~in~the United Stales and ) ~ ! ~ i d ctfhcr d e s ~ i ~ i a markets for h4cxrcan cxports were not tlic ~ ~ ~ i factor c~o ~ ~as t ~r a ~Fgrowth~in ~ j o ~ i ~ n ~ n hfexrcan cxports They use a gravity rnodef of trade to cScvclop a f ~ a i ~ for ~~ c~c oo m ~~ oMexico's e ~ s i ~ ~ expof? growlh into c ~ ~ ~ p o associateds with e x p o ~ " s ~ capacity, i ~ p o ~ e n ~ p ~ 1 ~ 7ows wcak growth in ~ ~ p o ~ - cs ~a pp~ ~~ y~ ~ ~ ~ i ~ ~ ~ ~ evidence suggests that interrial cond~~io~~s in hil ly capabilities --areas directly related to trade c 17 more ~ m p o r t afactor~ in limiting the c o ~ i i t ~export growth than China's cxpanded trade into the ~ ~ ~ ' s United States 4 I. Gains from trltde f ~ c ~ l j r reform ~ ~ t ~ o can be ~ ~ ~ s t a ~ Sofoaga,~Wilson, and Mejia (2006) t j a l . ~ finti that trade facilitation reforms in Mexico could inc se exports by $29.3 biiiioa, e ~ u ~to~21r ~ ~ e ~ ~ percent of total facturing exports. On the side, these figures are $14.4 bitlion and 13 3 percent, TC ir analysis assumes that t j ~ ~ ~rcfomsi would~allow Mcxico to a ~ o ~ tance between its current score" and the world average (75 cou aiong 4 variables, as foifows: i)port cffciency; 2) c u s t ~ ~ $ t; 3) r e ~ u ~ enviror ~ a ~ o as measured by the ~ e r ~ e p ~oi *f cn ~ ~ ~ tando4)nservice j ~ tmcture as measured by speed and cost of internet access and effect of the i met on business(Tabie IO). ic reforms have much more impact than the trade f ~ c ~ l j r a treforms of trading ~ o ~ partners. 'X'hr samc study ~ ~ ~ e a sther es d lated effect of domestic refomis in Mexico and ~ Mexico's trading parlners on Mexico's tr It found that domestic rcfornis a c e o ~ nfor 86 percent of ~ total export gaitts and 87 percent of total import gains. The food, tcxtiics, t r a ~ ~ F ~ ~ r t i ~ ~sectors were a~ ~ ~ ~~ and ~nmost~of the sectors, ~ ~ III a r ~~ about 50 pc ~ ~ ~ (from both foreign and domestic r e ~ ocome from i m ~ r o ~ e m e ~domestic border nieasures (port ~ $ ~ III ~ ~ s efficiency and customs envirorirne About 30 percent of the total potentialjump inexports comes from ports, j i n p r ~ v e m ~in~ Mexico's port ~ s efficiency appear tu be the most ~ ~ ~ factor,paccoun~ ~for n80 percent of thc total impact (of ~ o ~ foreip and domestic f on imports. The textiie sector si experience the largest jump in export ?ithdomestic refoms ~ ~ nan cxpected 48 per ~ r a ~ t~mcrcasc ~in tcxtile ~ ~ cxports. 'I'fic background studks (wssrw.wortdbaitk.tirg.mx) fbr this report cctntain imre detaiis 01%the tradc f ~ ~ c ~ l ~reformsi ostudy ~ a ~ n 'l'able 10: 3lcxico's Potential 'I'rude Gains from Danicstic Trade Facilitation Improvcmcnr Ii I---^ Customs Regulatory S c n k e Sector t--."-.-..----- Total , 43. Despite the benefits of increased trade, trade policy changes can entai1 risks for import- competing sectors. Krebs, Krishm and Maloney (2005) examined the r ~ ~ a ~betwecn trades policy~ p ~ o ~ ~ ~ ~ and ~ ~ ~ ~income riskdfaced~by aMexican workers, in ordcr to evaluate the welfare effcct of trade i ~ ~ ~ 1 i s reform.60Their result shows that trade policy changes have a significant short run effect on income risk for industries with highlevels o f import penetration. Moreover, the welfare results inthis analysis do not show that trade liberalization is necessarily costly to undertake. 44. Domestic policy reforms increase the benefits of trade liberalization. Nicita (2005) provides a quantitative estimate o f the effect on Mexican households from the implementation o f the Doha Development Agenda.61 The findings suggest that multilateral trade liberalization is likely not to be beneficial for most Mexican households unless complemented by domestic reforms aimed at facilitating the response of households to these new market opportunities.62However, when the implementation of the Doha development agenda i s complemented by domestic policy reforms targeted at increasing productivity and improving domestic price transmission, the overall effects become positive. The differences in effects across households are more closely associated with geographic areas o f residence rather than income level, meaning that the poor would be likely to share in the benefits o f such trade reforms as well. Measures aimed at enhancing domestic price transmission could ensure that any gains from trade reform, when accompanied by productivity-enhancing policies, would be more evenly spread throughout the country. It i s these complementary reforms to raise productivity associated with lowering trade costs that Mexico must focus on inthe decade ahead. 45. Logistics improvements are complementary to trade facilitation reforms and can lead to improved competiti~eness.~~ Logistics costs include transportation infrastructure and services (roads, ports, railways, and trucks), warehousing and inventory financial costs, and the associated firm administrative costs resulting from supply chain organization. Besides increasing productivity and competitiveness in their own right, more efficient logistics can reduce the cost o f exporting and importing and improve domestic price transmission. At the micro level, the reduction o f logistics costs enables firms to improve their financial performance, resulting in positive economic outcomes for the country. With better logistics performance, a firm will be able to reduce its inventory level and costs as well as the total costs o f goods delivered, allowing it to remain responsive to the client and consumers' preferences and ultimately increase its market share. In the context o f global sourcing o f goods, this results in increased trade and higher-value employment, leading to economic growth. Moreover, studies have shown that if a reduction in logistics costs takes place, products that were unviable to produce for export could become viable after this cost reduction. This i s likely to be the case for primary products and commodities, such as fruits and vegetables, which are likely to be produced inpoorer areas. 46. Recent trends in international trade are increasing the importance of logistics. After the decrease in tariffs following bilateral, regional, and multilateral trade agreements, countries are relying less on preferential tariffs to promote exports in the long run. Therefore, the improvement o f logistic performance (cost and quality) is progressively becoming a key factor incountries' competitiveness. It i s worth noting that increases in trade ultimately result in higher incomes and poverty reduction; Frankel 6o Krebs, Krishna, andMaloney (2005). "Nicita(2005). 62 The analysis utilizes a two step approach for which changes inprices and factors are estimatedthrough a CGE (computable general equilibrium) model and then mapped into the welfare function o fthe household using household survey data. 63 Logistics refers to the physical flow o f goods and data through the supply chain. Logistics involves the "process o f planning, implementing and controlling the efficient, cost-effective flow and storage o f raw materials, in-process inventory, finished goods and related information from point o f origin to point o f consumption for the purpose o f conforming to customer requirements" (Council o f Supply Chain Management Professionals, 2006). Logistics costs include transportation infrastructure and services (roads, ports, railways, and trucks), warehousing and inventory financial costs, and the associated firm administrative costs resultingfrom the supply chain organization. 19 and Romer (1999) found that countries closer to world markets enjoy higher levels o f trade, and that a 1 percent rise inthe ratio o f trade to GDP increasesincome per person by at least 0.5 percent. 47. Logistics costs are particularly high in Mexico. They are estimated at 20 percent, while the OECD average i s 9 percent.64A pilot survey o f a new Logistics Perception Index, aimed at measuring operators' perceptions o f countries' logistics environments, gives Mexico a score o f 3.57 (7 being the best and 1 the worst), below other Latin American countries like Argentina (4.46) and Brazil (3.83).65 Recent analyses by the government and the private sector, both o f whom were concerned about Mexico's potential disadvantage vis-A-vis its competitors, also highlighted considerable logistics weaknesses. In 2003, the government issued a Diagndstico General sobre la Plataforma Logistica de Transporte de Carga en Mixico, and the Mexican Competitiveness Institute (IMCO) performed a complete analysis o f competitiveness in 2004, with a robust chapter on logistics.66Both confirm the need to further reduce logistic costs. 48. The government has taken a number of actions over the past five years to lower trade costs. Mexico's tariff have been harmonized and reduced on a number o f products. Informationtechnology i s beingleveraged to lower trade transactions cost for firms, for instance through electronic transmission of certificates o f rules o f origin. Customs operations problems are being streamlined through electronic information exchange between agencies. And participation in international organizations-such as the WTO and APEC-has helped facilitate trade. Mexico's transport infrastructure has also been improved, including roads, railways, ports, and border crossings. See the Mexico Competitiveness Development Policy Loan, Report No. 34570-MX, for a more detaileddescription o f government actions. Moving Forward 49. Mexico would benefit from further lowering tariffs and non-tariff barriers and reducing the dispersion of tariff rates. Besides offering gains for consumers and allowing firms to acquire lower cost inputs, lowering tariffs provides a foundation for addressing non-tariff obstacles. The GOM undertook a tariff reform inDecember 2004 which reduced tariffs between 3 and 10 points on more than 9,000 tariff lines, affecting 76 percent o f all tariffs in the Mexican tariff schedule, but much more remains to be done.67The simplification or harmonization o f tariffs would make more productive chains viable.68This i s particularly important given the need to continue to diversify exports beyond oil and the major export destination, which continues to be the United States. Lower tariffs would reduce Mexico's dependence on imports from the United States as a supplier o f inputs. It would also reduce the incentive for exporters to the United States and elsewhere to choose U.S.imports over either imports from other countries or from Mexico itself. A notable feature o f China's most favored nation (MFN) trade reforms in recent years indicates a reduction in its dependence on exports. Exporters have increasingly sourced domestically, increasing the amount o f domestic value-added in exports. Continued reduction in MFN tariffrates, buildingon steps taken inDecember 2004, are clearly important. 50. Expanded use of information technology and reduced regulation can help lower trade costs. Making trade procedures and documentation available electronically will reduce the time and costs for 64 Guasch(2004) 65 The LogisticsPerceptionsIndex is a World Bank initiative implementedwith the support of the Turku Schoolof Economicsand BusinessAdministration, Finland.A pilot surveywas carriedout in2003. 66 IMCO (2005a). 67 See also the backgroundstudies for this report(www.wor1dbank.org.m). Currently, insome sectors it is moreprofitable to import final goods underpreferentialtreatment, than importing the inputs andrawmaterialsto produce those same goods domestically. 20 firms to conduct business abroad. Current GOM programs that increase the number o f instruments and agencies involved in paperless transactions are addressing this issue. Streamlining regulatory requirements for firms can also help lower trade costs. One way to do this is through further strengthening the Temporary Import Program for Exporters (PITEX) and the maquiladora export promotion programs. 5 1. Key internal corridors and logistics centers could be improved, as manufacturers face high costs to deliver their products frodto the border, ports, or airports. Elevated transport cost,s and times have also led to highinventory levels for businesses: during the 1990s,Mexico had 90 percent higherraw materials inventory levels on average than the United States.69The lack o f efficient multi-modal transportation platforms, such as logistics centers, increases the time to deliverheceive a product and the associatedcost. 52. Enhancing customs efficiency will bolster competitiveness. Average customs transactions involve many parties, documents, and data. The time for export i s currently 18 days inMexico, compares with 12 days in Canada; Mexico requires 11 signatures for imports, while only one i s required in Canada." The lack o f flexibility inworking hours and locations for product clearance adds on delays and costs. The customs administration would also benefit from a modernization o f its infrastructure, including better use o f informationtechnology, inports, airports, and land border crossings. 53. The organization of firms' supply-chains could be strengthened. Manufacturing companies, especially SMEs, often incur higher costs due to a lack o f skills in modern supply-chain management techniques. This involves rethinking the complete supply chain from material and services sourcing to the use o f specialized distribution companies to deliver the end product to the client. Another factor slowing the adoption o f modem supply-chain management techniques i s the limited availability o f efficient operators able to deliver integrated logistics services. The GOM has a relatively small Logistics Practices Innovation Program that attempts to improve supply-chain management within firms. MEXICO MADE HAS FAR-REACHING REGULATORYIMPROVEMENTSBUT RRMS STILL FACE PERVASIVEREDTAPE Description and Challenges 54. Although recent reforms have been far-reaching, entrepreneurs still face microeconomic constraints. Over the last two decades, Mexico has undergone a reform process that has redefined the economic role o f the state. Market liberalization, especially through the establishment o f NAFTA, privatization o f state-owned enterprises (SOEs), and regulatory reforms affecting 90 percent o f the legal framework, has moved the country closer to free and competitive markets. However, much remains to be done to address structural weakness that constrains productivity growth. Poverty remains high; approximately 52 percent o f the population lives in poverty, a similar level to that o f the early 9Os, and 20 percent live in extreme poverty. The informal sector represents 33 percent of GDP. At a time when Mexico is facing increased competition from China and Central American countries, Mexican entrepreneur's competitiveness i s hindered by a series o f microeconomic constraints, including the increased cost o f doing business that has beencreated by the regulatory burden. 69 GuaschandKogan(2003) 70 World Bank,Doing Business database. 21 55. Mexico needs to reduce its regulatory burden to improve productivity. Several studies demonstrate how the regulatory burden in Mexico i s still hinderingbusiness growth. In a review o f the effect o f product market regulations on investment and growth, the OECD found that when regulation i s restrictive, productivity growth is generally below the OECD average (Figure 17)." As the OECD points out, this does not imply that governments should dismantle regulations. Regulation i s necessary for well- functioning market economies. But regulation may also contribute to impede competition with detrimental consequences for resource allocation, innovation, andproductivity. The challenge for Mexico i s to design regulations that help to safeguard the public good without compromising economic performance. It is an area where learning from best performers i s crucial. The OECD estimates that if those sectors that lag interms o f productivity were to modernize their regulatory framework and align it onbest practice, thenproductivity could increaseby as muchas 10per cent. Figure17: RestrictiveRegulationof ProductMarketsis HighinOECDEmergingMarket Economies __. _____.__-_____-- ___I.. - OBanieis IO riadr and iuYtsmnr BinieiIlo tunepivntwrhlp Slate rounol 1 s I 0 5 0 i Note: 2003 values. The scale of the indicators is 0-6 from least to most restrictive. Source: OECD Indicators of Product Market Regulation (w.oecd.orgleco/pmr). 56. Benchmarking the cost of doing business in Mexico i s telling where improvements can be achieved through targeted regulatory improvement reforms. The Doing Business reports and databaseo f the World Bank have benchmarked indicators o f the cost o f doing business inMexico against more that 150 countries around the The report covers entry regulations, licensing regulation, employment regulations, registering property, court efficiency, corporate governance, creditor rights and credit information, bankruptcy, and tax issues. 57. Compared to other countries in Latin America, Mexico clearly lags behind in the following: a) time and cost to open a business; b) time and cost to obtain a license; c) cost of hiring; and d) cost of tax compliance.The Doing Business indicators contribute to a better understanding of progress in regulatory reform and demonstrate the efforts that remain to be made to improve the regulatory environment so as to allow the Mexican private sector to become more competitive. Despite its shortcomings, Mexico's overall performance in Doing Business indicators is better than the average, reachingthe top 40 percent. However, competitors such as Chile, Malaysia and Thailand are inthe top 20 percent. '*Seewww.doinebusiness.or~. 71 Cotis (2005). 22 58. 'I'irne and casts to open a ~ ~ s in~Mesicaeares high compared to China and latin Anierica. ~ ~ s data shows that it 1s cheaper and faster to open a buusincssin China (31 days, 14.5 p of incot~ieper capita) than in Mexico jSS days and 16.7 percent r e S p c c ~ ~ v(Figure~13). The picture e ~ ~ darkens c~iisIder~b1~ when C o ~ i ~ ~Mexico~ to its Latin ~ i ~ e ~ ne1a n ors. In, Chile, which r j n c with Mexico on trade with the European Union, it takes less than 27 days) the time it fexico EO create a business. Mexico also fails well short o f countries like tina (33 days), Colombia (43 days). and even Bolivia (50 days). Figure 18: Time and Costs to Start Business B Costto §tana Buscnsss(@Aof income percapes) 59. The r ~ ~ ucost ~~s ts ao e~with ~ ep ~e r ~atbusing is one of the main factors that drites ~ ~ ~~ ~ ~ conpanics into the ~ n f # r ~sector, r ~ ~ u c i nthe c ~ ~ p e t ~ ~ of ~entire~S ~sCs~ O TofSthe econumy. ~ a l g ~ e n ?bc c ~ ~sectorn IS p~a ~ ~~ c ~a~~ ~ a~r l ~~ i ~ ~shows~that is takes 222 days to icense to build a warehouse i ir in the bottom quarter o f Latin ~ 4 m c ~ ~ e ~ ~ ~ irt Colombia, I87 days in Bolvia, ;ifid f 5,I e, Costs to obtain a ticensc in Mexico (159 percent of me per capita) are not bad th inany ~ o ~in Latin~America, but there is still signi ~ r ~ ~ s room for i n i ~ r o ~ ' ~For~ ~ i i ~ . i ~ instance. it only costs 45 percent of incoiiie per capita in ~ ~ ~ c nandi125apercent in Chile. 'l'hese times ~ n and CQSES arc i ~ i ~ j ~ ~ ~the Uumght UT ~ r s that g i i l a ~ oburden can play on ~ n ~ e s audi'or~ e ~ ~ ~ ~ n corn~~ianeedecisions that i ~ ~ ~ ehave to Sacc ~ ~ u r s trying to grow their market share, 60. Other Doing ~ ~ i ~~in ~d ~~ce~nrlso ~r $ts~point to regulator? weaknesses. R/fexico IS in the bottom third of thc warldwidc ~ ~ ~ s ~ ruf~thcuCASCi of n rcing contracts, Regarding p ~ ~ p c rrt ey ~ ~ ~ t r a ~ ~ o ~ , ~ ~ Q therc are as many procedures in Mexico as in Aus and Irittand. hut it rakes at feast twice as long. ource OS bottlenecks; this procedurc m y take 1 Costs to regtster pro are high in Mexico, e ~ ~ c o u r a ~i i I in~ ~ o ~ ~ ~ and reducing thc svailabilily of collateral fbr cost o f hiringworkers. Doing Rtisir employee's arlnual salary in hkxi a11 but three Latin American cou bers fpercentagc of salary) arc 3 percent in tax r e ~ u ~ a are~~oon~~n p ~ icnough that ~ e ~ ~ e d it takes the average Mcxican e ~ ~ ~ r e ~ r536 e u r per year to pay raxes. ~ ~ n hours ~ e~ ~ ia~ r e~p rin~n ~ ~~~ ~ ~ e s ~ bothColombiaand Chile only spend41.3 hours payrngtaxes. complex regulatory framework, non-transparent business procedures and weak judicial structures remain factors inhibitinginvestment inMexico. They create indirect costs that impede Mexicanmanufacturers in their competition with other countries, and translate intojob loss and growth reduction. Figure 19: Cost of Regulationand PotentialBenefitsof Better Regulation(percent of GDP) 15% Costo de h lngresos Beneficios regulaci6n tributarbs no potencialesde la petroleros desregulacion Source: IMCO (2005a) 62. Lack of harmonization in the regulatory environment negatively affects businesses. Since regulatory responsibilities are shared among the federal, state, and municipal levels, coordination i s necessaryto eliminate the overlaps, duplication and inconsistency with federal regulations. A 2003 study commissioned by the Business Coordination Council (CCE) showed marked differences inthe quality o f the regulatory framework among states.73On the positive side, improvements in some sub-national entities have attracted reforms in others, as has been the case with the Rapid Business Opening System (SARE). Benchmarking different states and municipalities can spur this positive competition to improve regulations. The result o f two state by state studies was released in December 2005 by IMCO and the World Bank's ForeignInvestment Advisory Service (FIAS)/Doing Business. 63. COFEMER was created to improve regulation. In 2000, amendments to the Federal Administrative Procedures Law (Ley Federal de Procedimiento Administrativo, LFPA) institutionalized the Federal Regulatory Improvement Commission (COFEMER). COFEMER is a technically and functionally autonomous agency o f the Ministryo f Economy created to coordinate and to supervise the Regulatory Improvement Program o f the government. It aims to ensure the transparency o f the regulatory process and promote regulations that produce benefits greater than the costs. Mexico is one o f the few developing countries that has created a mechanism to challenge the quality o f new regulation through the establishment o f mandatory Regulatory Impact Assessment (RIA) for all new regulation that imposes costs on citizens. COFEMER plays the key role o f systematically reviewing R I A s and regulatory improvement proposals (Figure 20). 73CEESP (2003) 24 Figure 20: Institutional Design for Regulatory Review 64. The systematic review of administrative proposals is now well-grounded in law, and COFEMER i s fairly effective in improving new regulations (Figure 21). Though COFEMER has the power to present reforms to existing regulation, it is sometime difficult to achieve results without a good partnership with the responsible ministry. Inother words, COFEMER can pushfor the structural reform agenda, but it cannot carry it out on its own. Accountability and transparency o f the regulatory review process i s supported by the right to apply severe sanctions for non-compliance. Sanctions are applied by the Secretariat of Civil Service (SFP). A key enforcement power o f COFEMER is that regulations cannot be published inthe national gazette without its explicit opinion. Figure 21: RIA and Draft*-_ Regulation Review Process Proposal and RZA'; - 10 dayis,,.- I *--- =-*- RIA rent to COFEMER, "-..satisfactory? +-.em I 30 days Yrts I (publlc consultation) ,___ .__._ 4 COFEMER 65. Certain specified federal agencies must send each of their draft proposals, together with a RIA, to COFEMER, which makes them public on www.cofemer.gob.mx. COFEMER has the authority to issue a public opinion concerning each draft proposal and RIA.It also has the authority to request changes in the RIA, and, in specific cases such as high impact regulations, it can seek expert opinions on the draft proposal. For other agencies, COFEMER has "optional or not optional" authority to review and analyze all proposals with general exemptions in fiscal policy, criminal justice, and national defense. Hence, COFEMER can develop reform proposals in collaboration with these agencies. Inthe Fox administration, COFEMER has made use of its optional authority to propose new measures in at least three areas, as follows: 1) with Interior, it co-sponsored the Transparency Act (2003), 2) with Treasury, Foreign Affairs, and Labor, it co- sponsored administrative reforms to reduce federal procedures for start-ups, and 3) with states and municipalities it sponsored the federal/sub-federal fast 25 track start-up p r ~[see~beiow) ~Figure72 shows the numbcr of draft r e ~ l a ~ sento to COVE ~ ~ i ~ ~ s 1s a d e ~ ~ ~requests, bet ~ o ~ i ~ p ~ 66. ~ e ~ wplanning~activities are being ~ ~ n d ~via ~ a~~ e ~ ~~ l ~ ~ r ~ r tbe ~e Program. ~~ ~ ~ ~~ s to the LFPA ais0 iricl n ~ e c ~ ~ atoi ~organikei future j s ~ ~ Igivethepubltctheoppo~uni~yp to ire 111r e g u l a ~ o ~ p l a I ~ nAt~least, evcry two y ~ ~ ators must prepare an RHP rmthc public of their ptans for r e ~ u l a ~pr~ o ~ ~ creation of new fonnalitics. These RBPs are b e ~to build~a d~sciplirieof pertodical rcvictlf and ~ ~ a nof~thenregdatory framework and its i ~ ~ ~ ~ i ~ amend~~~en~sR13P also heips fo achieve the central ~ u ~ o othes federal reguulatosy ~ ~ i p r ~ ~ ~ ' e n ~ e n ~ The f ~ crcntc atid to modify regulations and f o r ~ ~ i aaccording ~ l ~ ~ i ~ to processes based on p ~ a ~ ~ ~ n ~ , cy, atialysis ~ oeffects,~ and public c o~n s ~ ~ ~ ain~ ~ ~o ~ ito obtain the fixgliest socia! ~ i ordcr , ~ ~ ~ beneStt" 47. '[*he First draft of the RDPs 1s ~ ~ b ~to~C~ ~~ t ~~ e d~postcd fbr~ puhiic corxirnents on and ~ ~ E ~ , ~ ~ ~ N'eb~site fw.ti.w.cot'emer.gob.~~~~as sieast twenty days. After the period of public ~ ~ ~ R for ' ER sends to each ~ ~arid ~ j s y its c~o ~ ~ ~ i~etof fthesprograms ~ well as other public e o n i ~ i ~ ereceived. Each regu ~ ~ s hen makes any c o ~ c s ~ o n d i ~ g or cxpfaiiis the reasons for reject1 fic final sersion in the ~ ~ i ~ ~ ~ a l h. Each agency must subitzit to COFEMER a periodic report on e o ~ z ~ p witha ~ ~ ~ c ~ ~ possible ~ ~ i o ~ i ~ c ato~thenprograms are asscssed, j o s 68. The last version of tfrc RBP was prcscritcd in the summer of 2005 and it ~ ~ i c ~ u da enovcfty: the d RBI' was prc'parcd with a Pr Public ~ j ~illati took place ~before the first draft o f the RBP was ~ ~ ~ e prepared. This approac~~ ma possible to utiify efforts in a new systemic a p p r o a ~to~ r~e ~ u ~ a ~ i # n , creating a c ~ ~ o ri ~~c cih ann i~between the~public,~academic. and private sectors. The idea was to set s ~~ ~ ~ ~ ~ unction with rlic private sector. The slstcmie ap~roacl~ aIso aimed to i n c o ~ ~ ~ raaholistic t c o f r ~ ~ ~system iiiparticu~areconomic sectors, 'Iogethcr, the ~ l a ~ o ~ c ~ ~ c ~ n public and pmate s efined 36 specific actloris related to regulatory I roJJernentthat could influence c ~ ~ ~ ~ p ~in a~ positive way.eThosce actions have an impact in specific areas, such as the i ~ ~ ~ / ~ s ~ f ~ l ~Transport~and~'I'e~ o~ ~: ~ ~ u n ~Labor axid~Social iSecurity, Health, Energy, Foreigntrade, o ~ c ~ ~ o ~ s ~ Finance, Techrxical Standards, Migation. and Sfates and ~ u n i c ~ ~ a ~ i ~ i ~ s . 26 and the federal district inwhich states agree to implement regulatory programs similar to the ones at the federal level. COFEMER has also signed 50 Memoranda o f Understanding with municipalities. COFEMER's support to states and municipalities has varied from providing technical assistance in establishing registries o f formalities to conducting reviews of regulatory authorities to designing enforcement schemes. As an illustration, 67 municipalities' regulatory improvement systems are currently being inventoried by the COFEMER. 70. COFEMER works with municipalities to simplify business opening procedures. The Rapid Business Opening System (Sistema de Apertura Rcipida de Empresas, SARE) was created by COFEMER and issued at the federal government level in 2002. It allowed businesses to complete two federal procedures-obtaining a fiscal identification number and name registration-in one business day. Buildingon this mechanism, some 85 municipalities (representing 35.7 percent of Mexican GDP) have also implemented the SARE, reducing the time to start up a business from over 30 days to 1-2 days (Figure 23). COFEMER aims to have implemented the SARE in a total of 100 municipalities by 2007. The SARE is also considered an important tool indeveloping compatible systems at the state and local levels. However, the SARE has beenonly partially implemented in some locations. Moreover, even with the SARE, the entrepreneur still has to register separately with labor and social security authorities as well as the statistical office. Figure23: SARE Municipalitiesto Date Source: COFEMER(2005) 71. The Federal Regulatory Improvement Council allows labor and business representatives to advise the government on regulation. The Federal Regulatory Improvement Program, designed and co- coordinated by COFEMER, is supported by the Federal Regulatory Improvement Council (Consejopara la Mejora Regulatoria Federal), whose membership was also modified by the LFPA. The president of the Council is the Minister o f Economy. Other members are the Minister of Finance, the Minister o f Labor, the Minister o f the Public Function, the General Director o f COFEMER, the Governor o f the 27 Bank o f Mexico, five business sector representatives and at least one representative from the labor, agricultural and academic sector. The President's Legal Counsel, the President o f the Federal Competition Commission, and the President o f the Federal Consumer Protection Agency (PROFECO) were added as new members after 2000. This Council mirrors the experience o f other OECD countries, involving labor and business representatives in official advisory bodies to the government for regulatory matters. 72. A regulatory moratorium was issued in 2004. In a meeting of the Council in April 2004, President Fox announced a moratorium for one year on regulations as part o f an effort to further reduce burdens on businesses. Under this moratorium, any new federal regulations must be shown to have benefits that significantly surpass the costs, or be driven by an emergency situation. This moratorium has caused a significant reduction in issuing and renewing regulations. The moratorium process has been effective in reducing the number o f regulations being issued, illustrated by the fact that 38 percent fewer regulations are now produced by the federal authorities, compared with the number before the moratorium. But it also has had a positive impact upstream, within agencies and institutions, which are now submitting34 percent fewer proposals for new regulations, compared with the volume o f proposals they used to forward for review before the m~ratorium.'~ In 2005, the Moratorium was extended for another year at the demand o f the private sector. 73. COFEMER has increased transparency in the implementation of regulations. Through open communication with the public and the private sector in particular, transparency has been reinforced. COFEMER has emphasized its importance not only with RIA'Sand commentary on draft regulations, but also with public participation in comment procedures, and with clearer legal requirements for notifications. Within the government, COFEMER disseminates knowledge about U s among institutions through training courses and its electronic portal, which has proved to be a major success. COFEMER has also developed a Federal Registry o f Procedures and Services (Registro Federal de Trdmites y Servicios, RFTS) that contains all formalities and services o f the federal administration with general exemptions infiscal formalities. 74. Transparency mechanisms have also been strengthened through the creation of the Federal Institute of Access to Public Information (Instituto Federal deAcceso a la Informacidn Phblica, IFAI). IFAI enforces the Federal Law of Transparency and Access to Public Information (Ley Federal de Transparencia y Acceso a la Informacidn Gubernamental), enacted in 2002. It also guarantees the effectiveness o f both the right to access public information and the right to privacy through data protection and promotes transparency and public sector accountability. Moving Forward 75. Mexico needs to undertake important regulatory reforms to boost its competitiveness. Specific sectors, especially in the manufacturing industry, are losing exports and even domestic market share to China and Latin American countries such as Honduras. While Mexico's labor wages cannot be reduced so as to compete with its competitors' low labor costs, the government needs to undertake actions to reduce the indirect costs of doing business. Specifically: (iii)Proceduresthat imposelargeburdens onentrepreneursneedto beeliminated, replaced, or improved (those for starting a business, licensing, employment regulation, tax regulation); (iv) Business interaction with government needs to be streamlined, so as to reduce uncertainty and implementation efficiency issues; 74COFEMER, see http://www.cofemer.gob.mx/. 28 Product market regulations: Industrysectors that are key to Mexican competitiveness and under specific competition threat from foreign markets could benefit from a thorough review o f administrative barriers in order to identify where action can be taken to reduce cost. Such sectors and issues could include foreign trade, manufacturing, property, finance, norms and standards, transport and telecommunications, energy and labor among others. For each sector or issue, reduction o f time, procedures, and costs o f fulfillment would help improve competitiveness. 76. COFEMERneeds continuingsupport. COFEMER can lead the tasks mentioned above, but it can exercise its full powers only if it enjoys broad support and acts decisively. One challenge i s sustaining the support ofthe private sector and official institutions. Earlier, ralliedby the domestic crisis and strong political support, the private sector took on an active role in pressuring agencies to improve regulation. Today many factors contribute to the more passive role o f several public and private actors; o f these two are the loss of founding members o f the Council through turnover and a review process inwhich success has become routine. Additionally, political circumstances and a fragmented Congress have shifted the private sector's lobbying efforts from the executive branch and its agencies toward Congress. This trend can only be counterbalanced by a reinforcement o f COFEMER's role v i s - h i s the institutions issuing regulations. In that sense, ensuring continuous training of public administration officials on the Regulatory Impact Assessment (RIA) framework is key to maintaining momentum while improving the upstream quality o f proposed regulations. The cornerstone of this effort lies indemonstrating the benefits o f the RIA process interms o f cost reduction to the entrepreneurs. Inthat regard, the establishment o f an ex-post evaluation framework for RIAs" would help benchmark regulatory compliance o f key administrations; it would also create strong political incentives for those administrations to streamline and simplifytheir regulatory environment. 77. Public consultation needs to be strengthened. Consultation in making, modifying, or repealing legislation and regulation in Mexico i s still weak. There i s still no legal obligation to undertake active public consultation for certain federal regulatory proposals. Nevertheless, COFEMER has included a section in the RIA questionnaire concerning public consultation. Institutional bodies are requested to specify if they established some type o f public consultation and who was consulted about the proposals. The RIA also requests a list of all the submissions that were taken into account for the draft regulation that i s beingpresented. But the input o f the private sector when it comes to regulation affecting the cost o f doing business should be reinforced. 78. Better harmonization of regulation between the different levels of jurisdiction is necessary. Today, any regulatory improvement process, including red-tape reduction at the sub-federal level, depends on the will and capacity o f states and municipalities to coordinate with each other. COFEMER, through the SARE system, can play a key role inhelping set up similar frameworks and systems between jurisdictions. It i s crucial to deepen the voluntary relationship between COFEMER and the states and municipalities. As a first step, it i s equally important to diagnose binding constraints at the state and municipal level in order to identify the different sources o f competitiveness loss o f specific industries or value chains. A benchmarking exercise between the states and municipalities could help identify best practices and drive the regulatory cost reduction agenda inthat regard. 79. Expandingthe implementationof the SARE will reducethe time and cost of working through business opening regulations. The target is for COFEMER to help implement the SARE in more than 100 municipalities, representing more than 40 percent of Mexican GDP by 2007. The output i s to be measured in the start-up time for small and medium-sized businesses (from an average o f 52 days to 3 75Ex-post evaluations would occur after regulatory changes have been implemented. Inthis way, actual impacts could be compared with the ex-ante impact assessments (RIAs). 29 days only) and a reduction in the associated cost. The outcome i s to be measured in the number o f new companies created, the value o f investment in these entrepreneurial projects and the number o f jobs created. Beyond the numerical targets, COFEMER shall also create a system to measure the SARE's effectiveness and monitor its implementation inorder to keep improvingthe processes and to facilitate the introduction o f the system to new municipalities. 80. The Regulatory Impact Assessment process infederal institutions needs to be reinforced. This component aims at generalizing the current RIA process among all agencies and institutions when applicable. This includes raising the number o f R I A s being conducted (355 in 2005) as well as the quality o f the R I A s being submitted for review. In that regard, training o f public servants in impact assessment practices remains crucial. In order to ensure compliance and accountability in the RIA process, enactment o f the by-laws o f the Federal Administrative Procedures Law (Ley Federal de Procedimiento Administrativo) is also important. 81. COFEMER needs to create a monitoring and evaluation framework for U s . To date, COFEMER does not compare the cost it estimates for regulatory proposals to actual cost, which could be measured after regulatory changes have been implemented (ex-post). Similarly, COFEMER does not track how the regulatory alternatives it suggested during the RIA process were actually applied and implemented. Finally, COFEMER does not formally benchmark the quality o f R I A s originating from different secretariats. A framework to achieve the above points should be put in place in the medium term. The impact o f such actions would be, over time, to reduce the cost o f doing business for Mexican entrepreneurs, especially compared with other Latin American countries, with an emphasis on entry, licensing, tax, and labor regulations. 82. Increasing cooperation with sub-national jurisdictions and the private sector to keep improving regulations at all levels i s important. The aim is to improve local competitiveness by the following: (i) benchmarking the cost o f doing business in at least 12 states in 2006; (ii) facilitating the creation o f competitiveness action plans at the sub-national level; and (iii)building an inventory o f existing state and municipal mechanisms for regulatory improvement. In turn, these activities shall increase the quality o f the systems used by the municipalities (percentage reduction in proceedings and procedures, improvement o f administrativeprocesses, response times, etc.) and reduce the time and costs o f business operations pertaining to entry regulation, property registry, contract enforcement, access to credit, and others. MEXICO'SENERGY SECTORNEEDS STRENGTHENITS PERFORMANCE TO Description and Challenges 83. Energy i s a major driver of economic growth, productivity and competitiveness. The energy sector i s o f highstrategic importance to the economy.76Evenslight changes in the availability and price o f hydrocarbon hels or electricity can have serious effects on economic growth. An increase in the relative price o f energy inputs increases the cost o f production o f goods and services, thereby putting pressure on profit margins. A modeling exercise by the IMF (2000) shows that a sustained US$5 per barrel increase in the price of oil results in a 0.1 percent decrease in GDP-even in oil exporting countries such as Mexico. Since January 2002, oil prices have increased from about US$20to US$55 per Another study showed that a ten percent increase in electricity infrastructure stock (generation 76See Annex 5 for a more detailed description of energy sector challenges and relevant programs and strategies. 77West Texas Intermediate Crude Oil Price 30 capacity) could lead to a general increase o f 2 percent in manufacturing growth with a higher impact on 31percent o f all manufacturingoutput." 84. Price volatility and uncertainty associated with fuel availability and the quality of service significantly affects consumers and producers in the economy. Both energy and non-energy businesses' investment decisions are based on expectations about energy prices and service quality levels. And unfortunately, expectations of Mexicanbusinesspeople are quite low. Infact, Mexico scored second to last o f 51 countries on IMD's World Competitiveness Yearbook business survey in terms o f energy infrastructure adequacy and efficiency (Figure 24). The countries that have achieved a secure supply o f fuels and electricity at competitive prices have exhibited highlevels o f economic development and consequently an increase intheir level o f competitivene~s.~~ Figure 24: Energy InfrastructureAdequacy and Efficiency (2005) Singapore Czech Republic Colombia USA Portugal Chile Venezuela Spain Turkey Russia Brazil China India Indonesia Mexico Argentina 0 2 4 6 a 10 Note: Scores (between 0 and 10) are based on business survey responses to the question: "Is the energy infrastructure adequate and efficient?" The highest value indicates the most positive perception. Source: IMDWorld Competitiveness Online. 85. The Mexican hydrocarbons industry faces important challenges. About 60 percent o f oil revenues are diverted to finance at least 35 percent o f national budgetary needs. This has a significant impact on the capacity o f the Mexican oil company (PEMEX) to finance new infrastructure for sustainable and competitive growth. Furthermore, meeting the growing demand for energy will require a significant increase in investment that the government will not be able to finance unless it severely curtails spending on other important social goals (such as health, education, and rural infrastructure). Breaking out o f this vicious circle i s one o f the most important challenges facing Mexico today. To a large extent, the technical and operational efficiency o f state-owned enterprises depend on the levels of spending approved by Treasury and the Congress every year. But beyond the spending constraints, various national and international experts agree that PEMEX needs to make additional efforts to increase operational efficiency, lower costs, and improve the quality o f service. 78Castaiieda Sabido (2003). 79Aspe (2004). 31 86, Natural gas prices have seen ~~~~~~~c increase. Despite thc fact that Mexico has thc sixtk- reserves in the ~ ~ e ~s~ ~~i i ei s ~~theccountry inipo ~ r e , 5-20 pcsccnt of Its domestic orts are mainly supplied by the US. natural gas et, one of the most volatile of the world. Natural gas prices m Mexico have rncrcassd by 236 percent over the Last SIX years and are aniongthe liiglicst inthc world (Figure 25 and Figure 26). The lack of cconornic and ~ ~ ~efficiency ~ ~ ~ c a 1 in the h ~ d r o ~ ~ sector~ hass also tra~sla~ed ~ b n into high fuel oil priccs, which have increased by 183 percent since 1999. Figure25: Reference Prim Natural Gas Mexi $14 $13 Ed 53 SENER data, 1004 2005 32 87. Electricity tariffs in Mexico have increased 77 percent since 1998, compared with a rise of only 14 percent in the United States (see Figure 27). And commercial tariffs are among the highest in the world. Indeed, average commercial tariffs in Mexico are 80 percent higher than the U.S. average (Figure 28). Transmission tariffs in Mexico are also among the highest inthe world, below only Japan and Italy. Figure27: ElectricityPrices (IJSDg / kWh) Figure28: Commercialtariffs, 2000 (USD cts./kWh) 1 5 0 1 2 5 1 0 0 7 5 5 0 2 5 0 0 3 98 99 00 01 02 03 04 Source: Consejo de Coordinadores Sources: Consejo de Coordinadores Empresarial(2004) based on U.S.Department Empresarial(2005) based on national o f Energy and Mexico Federal Electricity Commission (CFE). sources.8' When equivalent consumptionlevels (interms of capacity factor, tension, and type of service) are compared across regions, Mexicans pay higher rates across the board (See Table 11). National sources include: MBxico: www.cfe.gob.mx / informacih p6blica de la CFE / estadisticas de ventas / datos hist6ricos. Suma total de ingresospor ventas del Sector Electric0 Nacional a Mediana Industria y GranIndustria entre suma total de MW vendidos a Mediana Industriay Gran Industria. UnitedStatus: www.eia.doe.gov. Datataken from the EIAMonthly EnergyReviewNovember 2004, Average Retail Price of Electricity, Industrial. Brasil: www.aneel.gov.br. Agencia Nacional de Energia Electricade Brasil. Tarifa industrialpromedio nacional annual. Chile: www.cne.cl. Comisi6n Nacional de Energia (de Chile). Precio medio de nudo Maitencillo (zona Santiago) para abril de cada aiio. Colombia: www.creg.gov.co. Comisih Reguladora de Energia y gas (de Colombia). Tarifa industrialpromedio nacional anual. 33 industr I i I 34 250 - 200 - 150 - -eCFE -+Peru Edelnor + t-BarllElecb-oPaub loo - +Afgenbno M e w *Argentina Edesur 5 0 - .--4-I- - - -____c +- L*--,-=, -==-* 0 g,--, I 89. In Mexico, high costs are compounded by fluctuations in frequency and voltage. The result is effective costs that are 10 to 60 percent higher than those o f U.S. counterparts (Figure 30). Companies also face other barriers associated with the electricity service in Mexico: difficulty in obtaining permits for self-generation and co-generation, highconnection costs for self-generating units or new connections to the grid, and others. Figure30: Cost and Quality Comparisons,Mexicanvs. U.S. Companies Variation in Voltage Variation in Frequency Contract 2% CA = PEpcSw m */- 3 5% = w m +/.(I3% = W1fi-W -Less than +/- 0 5% - = cad'rapMBBw Lessthan +/- 0 3% DmmmUrpnaParer - I/- 5% wlmt-H-P Less than +/-3% Q'lnabvaa '* I Lessthan I/- 3% ~ ~ ~ ~ r Lessthan +/- 0 3%~ g n Cost of Electricity US cents / kWh =- 5 1 (HS-L) 10.60% extra cost due to failures A5CSW 3 6 (IndustrialAverage Tx. Ok y La) M-W- 3 9 (HS-L) CadrmRhVerB - ushl -1 386-412 (HS-L) knwmVrgnm -.II h e r 394 (HS-L) Source: Consejo de Coordinadores Empresarial (2004) based on information fiom US.electric companies and information fiom users inthe U.S. and Mexico. 35 90. Subsidies for infrastructure services absorb significant public resources in Mexico and encourage inefficient resource use. For example, subsidies (for operations and consumption) in the electricity sector amount to about 1.1 percent o f GDP (Table 13). Tariffs for commercial and industrial users are set near levels allowing full recovery o f the cost o f supplying these consumers. However, average residential and agricultural tariffs covered only 42 percent and 28 percent o f cost, respectively, during 1997-2003. Table 13: Estimate of Subsidiesfor Electricity,2003 Sectolxl (Federal FAIS (2002 As a Sham and State) estimate) Total of GDP (OA) @l.xPbillion) ConsumptionSiibsidies fiianced by CFE (off- 49 0 49 0.735.0 budget) OtherSubsidies (LFC operationandFAIS) 21 2.9 23.9 (5.35% Total 70 2.9 72.9 1.08?f~ 91. Residential electricity subsidies are highly regressive: Upper middle income households (income deciles 6, 7, and S), receive the greater part o f the consumption subsidy. The electricity subsidies also go mostly to the regions that are already more economically developed. The bulk o f the subsidy-ver 90 percent-is not a lifeline for the poor and encourages inefficiency, especially in the hot areas in the summer, which benefit from highly subsidized rates. Poverty criteria are absent in the determination o f regional electricity tariffs, unlike in water, where some municipalities set lower tariffs in poorer neighborhoods 83 92. In the manufacturing sector, the cost of energy as a percentage of total production costs increased by 2.25 percentage points to an average of 10.5 percent between 1995 and 2002. This i s comparable to labor costs, which are in the range of 4 to 20 percent. The energy-intensive industrial sub-sectorsare particularly vulnerable to volatility inprices and uncertain availability. The cost o f energy as a percentage o f the total production costs for these types of industries i s particularly high (see Table 14 below). The quality and price o f the products from energy-intensive industries are of fundamental importance, as they are used as raw materials in the manufacturing o f other diverse products. For instance, the price of cement inMexico is about twice the price inthe US, the European Union and other countries around the world (115 USD/ton, CEMEX Data 2004). Steady increases in energy prices have had serious effects on the costs o f production across different industries. 83World Bank(2005d), IPER 36 Industry Percentage I DATA FORMEXICO * When labor costs are included, the share of energy costs as a percentage of the total is 5.86 percent on average (2002). Source: Commissionfor EnvironmentalCooperation (2003) for cement inMexico; INEGI (2002) for manufacturinginMexico; EnergyInformationAdministration(2005) for United States 93. Over the last five years, there has been a steady fall in the number of energy-intensive processing facilities in Mexico. This has been largely attributed to the increasing costs o f energy and to uncertainty regarding the availability of natural gas and other fuels (Arizmendi, 2005) (Figure 31). The studies and surveys conducted by the Center for Studies on Competitiveness (CEC) at the Instituto Tecnolbgico Autonomo de MBxico (ITAM) to analyze the competitive status and prospects o f different industries-including mining, electronics, and vehicle parts-have concluded that highenergy costs and the low quality o f infrastructure services are factors contributing to lower economic efficiency and prospects for growth and competitiveness (CEC 2004a, 2004b, 2004~). 138m - Dic*o Figure31: Reduction inProcessingFacilities inMexicon4 --- __. _._ -- - - __ - ____ - - _____ __ _ _ II $ 4 h U 121Qa - - - -- - ---- _ _ __ -- - - ___-- - - -_ __--_ 1n.m - _ _ -- - - - __ - _. --- __ 2 s i m a n wpoa sapm sa42 sspm scpo, scpss Note: IMSS is the MexicanSocial SecurityInstitute(InstitutoMexican0 del SeguroSocial) Source: Consejode CoordinadoresEmpresarial(2005)basedon data from M G I . 94. Reforms to the energy sector maybe extremely difficult to achieve. Over the last decade, different groups have called for the implementation of far reaching reforms in sector policies as a necessary condition to attract sufficient private financing, improve efficiency, lower costs, and expand the sector's contribution to the federal budget and the economy. Although a national consensus has not 84Reductionestablecimientosde industriade transfomacion:Patronesregistradosen el IMSS industriade transfonnacion(INEGI, 2005). 37 been reached with regard to the nature andextent of energy sector reforms, the GOM has gradually taken conservative steps to allow limitedprivate sector participation and improve transparency and regulatory accounting. 95. Mexico has adopted a strategy to strengthen and modernize the energy sector. Specifically, the Energy Sector Program 2000-2006 (PROSENER) includes measures to strengthen the technical, economic, financial, and legal aspects o f the different energy sub-sectors. Inelectricity, it places special emphasis on: (i) fuel share diversification and security o f supply -measures (e.g., increasing the share of renewables and other sources); (ii) energy efficiency and conservation; (iii)finding ways to increase private sector participation and, (iv) increasing commercial integration with regional energy markets. In the hydrocarbons sector, PROSENER i s mostly concerned with the modernization of the oil industry includingthe transformation o f large industrial complexes, and the introduction o f incentives to increase the efficiency and competitiveness o f the sector. Inparticular, the program emphasizes the need to adopt international standards for project planning, management, and development, and for service and product quality. 96. Regulation i s being improved through an agreement between COFEMER and the Secretariat of Energy (SENER). The 2001 agreement includes measures to modify regulatory frameworks and ensure transparent regulatory accounting and public disclosure. Moving Forward 97. Mexico needs to implement a number of strategic measures to ensure the long term sustainability of the energy sector and the competitiveness of economic activities. The constraints affecting the energy sector have been repeatedly exposed over the last decadeby both expert analysts and public enterprises PEMEX and CFE. Inthe hydrocarbons sector, it has beenestimated that about US$130 billion would be needed between 2004-2012 for priority investments; with these sums available, the focus would be on restitution o f reserves and minimization o f imports allowing for activities in exploration and production, refining, petrochemicals, and others.*' Likewise, the electricity sector needs important sums to finance additional capacity and modernize interconnected infrastructure. The growing dependence on fuel imports and the gradual deterioration o f energy infrastructure calls for the implementation o f a numbero f far-reaching strategic measures, which may include the following: 0 Finding ways to lower the country's fiscal and budgetary dependence on oil revenues and allow autonomous enterprise management in hydrocarbons and electricity. This will perhaps require a programmatic approach that considers the gradual strengthening o f other sectors o f the economy to substitute PEMEX and CFE for other sources and which could sustain budgetary needs and the financing of social programs. One important move toward this end would be the continued strengthening and modernization o f the fiscal regime to guarantee sufficient, reliable and stable contributions (Le. lower risk exposure to oil revenues and eliminate possible bottlenecks to social and economic development). Also, it would be necessary to ensure high efficiency in spending on social programs with an emphasis on long term social and economic development. An increase in private sector participation is also considered essential to minimize the fiscal burden and financing needs o f the hydrocarbons and electricity industries. 0 Launching a structured national dialogue that includes relevant stakeholders and expert analysts to review the options for ensuring the long term sustainability of the energy sector ''Acevedo (2004). 38 (e.g. national expert committee). Inparallel, strengthen the capacity o f politicalparties, senate and congress members and the civil society to analyze energy sector bottlenecks. This exercise should try to integrate the political dimension and seek Congress and Senate approval only after a set o f sound strategies has been clearly identified, along with their associated economic and social costs andbenefits. 0 Establishing a strategy to modernize the energy sector and reduce dependence on fuel imports. This strategy should include the following: a) implementation of a sound program of investments and capital expenditures; b) fuel diversification measures; c) encouraging private sector investment interests and promotion o f strategic partnerships to bringinadditional sources of capital; d) implementation o f a program to improve service quality and performance to international levels; e) design and implementation o f innovative legal and financial instruments to lower risks associatedwith fuel price and fuel availability. Such measurescould include innovative covenants in power purchase agreements, phasing out o f PIDIREGAS, fuel price hedging, and fixed price coverage with reference price or cap. 0 Strengthening policy and regulatory frameworks. New investments will require the consolidation o f policies and regulatory frameworks that increase long-term certainty levels for all participants in different segments o f the hydrocarbons and electricity supply chains. MEXICO LAGS ICTANDTELECOM MORECOMPETITIONISNEEDED ON AND Description and Challenges 98. Investmentin Mexico's information and communicationstechnology (ICT) sector, especially the telecommunications sector, has grown dramatically since the 1990s. This growth is in large measure due to the gradual opening up o f the market to competition and foreign investment.According to Mexico's telecommunications regulatory agency (Comisi6n Federal de Telecomunicaciones, COFETEL), from 1999 to 2003, more than US$20 billion have been invested inthe telecommunications sector alone. In 2003, a time o f significant contraction in the global telecommunications industry, investment in the Mexican telecommunications sector amounted to US$ 2.3 billion and grew at a rate four times faster than the economy as a whole. In 2004 the GDP indicator for the telecommunications sector grew by 22.6 percent, compared to a national GDP growth rate o f 4.4 percent. 99. Despite this impressive growth in absolute levels, Mexico lags behind other Latin American and OECD countriesin terms of investmentsinthe ICT sector as the share of the overalleconomy. As Figure 32 illustrates, Mexico's level of ICT expenditure (3.1 percent) is significantly below OECD countries such as Japan (7.4 per cent), United States (8.8 percent) and New Zealand (10 per cent). It i s also nearly half that o f Chile and Brazil's rates o f 6.7 percent and 6.7 percent respectively and lower than Argentina's, Peru's, and China's rates o f 5.7 percent, 6.9 percent and 8.4 percent respectively. As a growing number o f studies have found,86countries with higher levels o f investment inICTs experience higher economic and social development growth. 86See OECD (2004c), "ICTs and Economic Growth inDevelopingCountries." 39 7 Figure 32: ICT investments as a Share of GDP,2003 y = 9E-0Sx + 5.360; Rz= 0.1935 Singapore 10.0 USA h "M 8.0 - n 5 3r 6.0 e n 0.0 4 0.0 5000.0 10000.0 15000.0 20000.0 25000.0 30000.0 35000.0 40000 GDP per capita PPP Source: World Bank, World Development Indicators, 100. Mexico falls short in terms of network readiness. Reasons for underinvestmentinMexico's ICT sector can found in a wide range of studies and global indices on the ICT sector that highlight problems with Mexico's legal and regulatory framework. For example, according to the World Economic Forum (WEF) Mexico was ranked 44th out of 102 countries in the 2003-2004 Networked Readiness Index (NRQ8' The NRIis one of the world's most widely recognized measurements of countries performance with regard to ICT readiness and usage. As can be seen inFigure 33, compared to other LatinAmerican countries, Mexico's 44th rank places it behind Brazil (30th), Chile (32th), India, China, Turkey and all OECD countries. The NRI i s a composite index that is comprised of several sub-indices, including a political and regulatory sub-index'' and on this sub-index, Mexico received the lowest ranking o f any OECD country (73) as well as a ranking lower than nine other Latin American countries, includingChile (35), Brazil (50), Jamaica (51), Uruguay (57), Costa Rica (58), Trinidad and Tobago (60), Colombia (69), and DominicanRepublic (72). 87The NetworkedReadiness Index (NRI) i s defined as "the degree o f preparation o f a nation or community to participate in and benefit from ICT developments." The NRIprovides not only a model for evaluating a country's relative development and use o f ICT, but also allows for a better understanding o f a nation's strengths and weaknesses with respect to ICT. 88Political and Regulatory Sub-index: "The priorities o f a nation are reflected inits policies and laws that inturn influence its rate o f growth and direction o f development. This component o f the NRImeasuresthe impact o f a nation's polity, laws, and regulations and their implementation on the development and use o f ICT." 40 Figure 33: World EconomicForum(WEF) NetworkReadinessIndex 2004-2005 90 - 80 - I I 70 - 60 50 I 40 30 20 10 0 Note: The y-axis is the country's rankingout of 104countries. A lowerrank is better. Source: World Economic Forum(2005b). 101. Mexican fixed line growth has not kept pace with comparable countries. Between 1998 and 2004, the number of fixed lines in Mexico almost doubled from 9.9 million lines to more than 18.1 million, with an average annual growth rate o f around 10percent and a total fixed line penetrationrate o f 17.1 telephones per 100 people in 2004. Although this growth has been significant, it i s substantially below the rates found in other similar countries, In 1991 Mexico, Brazil, and Chile had comparable fixed telephone penetration levels, however during the following decade, Mexico's growth levels were significantly lower than those o f Brazil, Chile and other OECD countries and as a result, Mexico now has significantly lower fixed line penetration levels than other Latin American countries (except Peru), as well as comparable developing countries inother regions as can be seen inFigure34. Figure34: Fixed, Mobile and InternetPenetration-Various Countries (2004) 100! ' * O 96.1 100.8 n I I 86.4 88 n r I 8 0 1 r 70.1 I 6oi r 56.5 I 40 20 0 i m main lines penetration Einternetpenetration 0 Cellular subscribers per 100 inhabitants Source: ITUdatabase; World MarketsAnalysis. 41 102. The digital divide within Mexico i s also an issue. Not only has the digital divide between Mexico and other countries, as measured by fixed telephone penetration, increased but Mexico also lags interms o f reducing the digital divide within Mexico, especially between urban and rural areas. Two indicators highlightthis trend.*' First, as is the case innearly all developing countries, telephone penetrationlevels in urban areas tends to be much higher than in rural areas. As can be seen inFigure 35, the Federal District has nearly 35.4 fixed lines per 100 people, while the Southern states o f Chiapas, Oaxaca, and Guerrero have, respectively, only 3.9,4.7 and 7.6 fixed lines per 100people. Figure 35: RegionalDistributioninMexico of MainLinesper 100 inhabitantsand GDP per Capita, (2004) OWb Federal 4 Nuevo L d n 20 16 EDBdo de Mexico Morelo Oaxacb Chiapas Guerrero San 5 bbsi I - 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 GDPper Capita Source: COFETEL 103. A second way to compare digital divide trends is to compare changes in the distribution o f telephone penetration between the region with the highest penetration level and the region with the lowest penetration level in different countries. As Figure 36 highlights, Mexico has a more unequal distribution of fixed teledensity than Chile, Brazil, and Peru. The disparity in mobile coverage between Mexico City (highest teledensity) and Chiapas (lowest teledensity) has increased tremendously inrecent years, going from less than 5 percentage points in 1997 to 33 percentage points in2002. This compares to a disparity inBrazil o f only 13 percentage points in2001. ''Theseindicators are taken from Mariscal (2005). 42 - I 1994 1995 1996 1997 1998 1999 2000 Mol Note: The FixedTeledensitypercentages measure the disparity incoveragebetweenthe regionineach countrywith the highestcoverage and the region with the lowestcoverage. 104. With respectto the internet, Mexico i s a study in contrasts.g0 the one hand it has some the On most advanced profiles o f large business Internet use in Latin America. The internet i s widely used for business-to-business, business-to-government and government-to-government transactions. As a result o f the web sites and content generatedby the e-Mexico, a comprehensive government initiative launched by the Fox administration, Mexico has received many awards for e-government initiatives. On the other hand, Mexico has one o f the least advanced profiles for consumer use o f the internet. Within Latin America, Mexico has one of the lower household internet penetration levels as can be seen inFigure 34. Within Mexico, the disparity between Internet use in different states is as great, if not greater than the telephony disparity previously described. Mexico's growth rate has also been substantially lower than Chile and Korea. As illustrated in Figure 37, Mexico increased the number o f internet users per 100 inhabitants from 1998 to 2003 around 10 times (from 1.3 users per 100 inhabitants to 12 internet users out of 100 inhabitants), while Chile's grew more than 15 times, from 1.7 to 27.2 internet users per 100 inhabitants. InKorea, the number o f internet users per 100 inhabitants increased from around 7 users per 100 people in 1998 to more than 60 percent o f the population during the same period. Finally, Mexico also lags behind other Latin American countries in ownership and use of personal computers (PCs)- Mexico has only 8.2 PCs per 100 persons, which i s lower than Chile's rate o f 11.9 and Costa Rica's rate of 19.7. See Mariscal(2005). Also, Carlson, DeSpain, Elorrieta, Flamm, Komegay, and Mahaffey, Chapter 7. "The Mexican Case." A chapter ina book beingedited for publication. 43 Figure37: InternetUsers inMexico, Chile and Korea I 70 1 1998 1999 2000 2001 2002 2003 Source; ITUdatabase 105. The mobile sector has grown tremendously but faces several challenges. From the mid-l990s, Mexico's mobile sector grew at an explosive annual rate of nearly 100 percent, though compared to other Latin American and OECD countries, Mexico's cellular penetration level i s still low. Growth has been driven mainly by the following: the speed and efficiency with which the government allocated spectrum in 1996-1 998; competition; the introduction of calling-party-pays policy by the regulator in 1997-1998; and by the widespread adoption o f pre-paid calling cards. Penetration o f wireless services was greatest in areas where fixed services had not expanded, including lower income areas. The major challenges facing the sector are as follows: the declining average revenue per user due mainly to the popularity o f pre-paid calling plans; quality o f service and consumer complaints; industry consolidation; creation o f incentives to provide internet services using wireless technologies; and extending service to rural and low income areas. 106. Fiber optic networks and cable and paid television have also seen robust growth. Mexico's high-speed fiber optic networks that carry the traffic for telecommunications and ICT service also grew dramatically since the sector opened to competition in 1996-1 997. From 1993, Mexico went from having around 15 thousand kilometers of fiber optic networks operated by Telmex to around 112 thousand kilometers of fiber optic networks operated by at least six different companies in 2003. In addition, broadband access in many countries including Mexico has been based on cable and paid television infrastructure. The number o f paid television subscribers inMexico grew from 1.81 subscribers per 100 persons in 1996 to around 4.38 subscribers per 100 persons by the endo f 2004, as shown inFigure 38 44 Figure 38: PaidTV Subscribers per 1000 Inhabitants inMexico (2004) f501 35 37 36 37 3 433 107. In 2003, several dozen cable television companies, out of 713 paid television concession holders, applied to obtain a concession to provide data and telephone service over their cable networks. Approximately twenty companies have been granted authority to carry data. However, to date, no cable television company has been granted a concession to provide voice services. COFETEL and the Secretariat o f Communications and Transport (SCT) have not granted any such licenses because, they contend, it would require that they allow Telmex to enter the cable TV sector. 108. As can be seen in Table 15, more than 60 percent of Mexican consumers who have high-speed access to the internet do so through cable networks, and cable television companies have more than 70 perce,nt share o f the total paid television market. Telmex, which dominated fixed line and mobile networks, does not compete inthe paid television market though it has successfully gained a significant and growing broadband market share though its ADSL services. Source: Select, June 2002 109. Mexico has low broadband penetration. Access to broadband internet is increasingly used as an indicator o f the development of the ICT sector. As can be seen in Figure 39, Mexico has very low broadband penetration rates not only when compared to other OECD countries, but also when compared to other Latin American countries such as Argentina, Brazil, Chile, El Salvador, Peru, and Venezuela. 45 Figure 39: Broadband Penetrationand Prices 250.0 233.3 200.0 146.7 150.0 117.0 100.0 92.9 50.0 0.0 3.0 , A 2 , 1- Broadband subscribers per 1000 inhabitants+Braodbandpricemnthly charge US$ 1 Source: ITUdatabase 110. The incumbent, Telmex, remains dominant. Telmex, as can be seen in Table 16, clearly dominates the long distance, local, and cellular telecommunications markets. It provides service to over half o f all dial-up internet accounts and over a third o f high-speed internet access accounts. Its share of high-speed internet accounts increased significantly during2002. Table 16: Telmex Market Shares Note: Internet access estimates based on number o f accounts. Source: Telegeography 2004. EMC 2005. 111. Telmex's market share in the international sector in2003 represented around 66.6 percent which i s often cited as an argument that Telmex i s too dominant. As Mexican authorities have pointed out, Telmex's market share after five years of competition is comparable to the five-year benchmark of other countries (see Figure 40). Indeed, a recent survey o f eighteen countries found that the average five-year incumbent market share benchmark i s 64.5 percent. What is remarkable about Telmex i s not that it is dominant in one market segment, but that it has retained, and in some cases increased, its market power inanumberof sectors, whichhasrippleeffects onaccess, innovation, andprices. 46 Figure 40: IncumbentMarket Share after 5 years of Competition USA 183.3 Phillipines 9 Dom Rep Korea Mexico Average Australia Malaysia Denmark Israel Chile 0 20 40 60 80 100 Note: The average is based on data from 18 countries that report market share informationfor markets opened for at least 5 years. Source: Telegeography, 2003. 112. Telmex's net profit margin and earning per share from continuing operations is more than twice that of its closest rival (Table 17). Telmex is also far ahead in operating income as a share of revenues compared to other U.S. and British companies. During the last five years, Telmex has also transformed itself into a fairly efficient company interms o f lines per employee. Table 17: FinancialPerformanceof TelecommunicationsCompanies Qecember 2004) * 2003 Source: Hoovers Detailed Annual Reports. All based on December 2004 reports. 113. Telephone charges are high in Mexico compared to Latin America. Table 18 shows the evolution o f telephone tariffs between 1995 and 2003. Although monthly subscription and usage fees have not changed much, connection charges have dropped drastically. In2003, the basic monthly tariff 47 was about $14.5 for residences and $18.3 for businesses, and the local calling charge was about 14 cents for each call. As seen in Table 19 and Table 20 prices in Mexico are much higher than other similar countries. All calls originating in the public network inMexico are subject to the basic 14 cents calling fee, unlike most Latin American countries where telephone charges are based on usage or the United States where local calls are covered by the monthly recurring charge. Businesses pay this rate for all calls, but residents pay only for calls exceeding 100per month. Telmex reports that approximately half of the residential customers make fewer than 100 calls per month. Table 18: Telephone Tariffs in Mexico, 1995-2003 ILocal call (3 minutes, USD) I0.07 I0.10 I 0.14 I 0.13 I 0.14 I0.15I 0.16 I0.15I0.14I Source: ITUdatabase, 2005. Source: ITU Business Brazil 78.80 26.70 Guatemala 16.71 28.03 Korea (Remoo 55.40 I 32.90 26.00 United States 25.70 (a) Total Basket = (installation cost/l20)*(monthly rate) * (charge p/mincall) Source: ITUdatabase with author's calculations 48 114. Local loop prices for businesses are quite high. Businesses that generate large amounts o f long distance and international data traffic, typically lease two types o f data or circuit services -one to carry the traffic from their offices to the switch o f the long distance carrier (known as a local loop), and a second for the long distance portion. The long distance price compares favorably with other countries, but not the local loop price. Table 21 shows that the price of a local loop inMexico is muchhigher than that found inmost other comparable Latin American countries. The installationor non-monthly recurring charge for a local loop with E-1 capacity i s $19,669, more than twice the price charged by Telmex's competitors and more than six times the price charged inBuenos Aires. Similarly, the monthly recurring charge for a Telmex E-1 local loop i s more than twice what a new entrant charges in Mexico City; moreover, it is more than 40 percent more expensive than the monthly recurring charges in Argentina and Brazil and more than four times as expensive as inChile. Note: Assumes a one-year lease for an E-1local access line. Prices are based on those publicly registered. Source: Company survey by World Bank. 115. People in rural areas pay relatively more for similar services. Consumers in remote rural areas not directly served by Telmex can make calls through Telecomm, a quasi-private company overseen by the (SCT). However, Telecomm is not a carrier and has to pay Telmex to transport signals from its hub near Mexico City to the final destination. Telecomm i s not considered a public operator and does not have an interconnection contract with Telmex. Telmex considers Telecomm a "large user". It also does not receive the access charges from incoming international calls. As a result, Telecomm's tariffs to users are high for the remote rural consumer (Table 22). Telmex rates, in contrast, are lower, and they do not charge for incoming calls. Telecomm has tried to establish an agreement with long distance carriers that would allow it to create a "calling-party-pays" system, but without success. Thus ultimately, people in smaller (and poorer) communities pay more for the same service than people in larger (and richer) communities. Calls to USA $1.04 $1.04 116. Mexico's ICT regulatory framework i s weak. The shortcomings and challenges o f Mexico's ICT sector-lack o f investment in rural areas, low broadband penetration, market dominance and high tariffs-are largely a reflection o f shortcomings in Mexico's legal and regulatory frameworks and institutions. Industry reports consistently rank Mexico's regulatory frameworks as one o f the weakest in Latin America." More importantly, recent reports and decisions by international organizations such as 9'See Pyramid Reports (www.pyyramidresearch.com). 49 the OECD and the World Trade Organization have also highlighted the shortcomings of Mexico's ICT policy, legal and regulatory framework, and institutions. 117. Regulatory issues were at the root of a World Trade Organization dispute with the United States. As is highlighted in B o x 14, a recent decision by the WTO involving a dispute between the United States and Mexico over telecommunications, found that Telmex is a "major supplier" and that Mexico's regulatory regime and regulator had not only failed to stop Telmex's anticompetitive practices, but that it had adopted regulations that required anticompetitive behavior. The WTO dispute resolution panel ruling was drafted to b e narrowly limited to Mexico's WTO commitments on international telecommunications. However, it is significant insofar as it carries out a detailed andbalanced analysis of Mexico's WTO commitments as well as Mexican regulations. Box 14: Telecommunicationsand the WTO: the Case of Mexico The first (and so far only) case o f WTO dispute resolution on telecommunications services (and, indeed, on services generally) was initiated in 2000 by the United States with reference to Mexico. In August 2000 the United States requested consultations on Mexico's obligations and commitments on basic and value-added services under the GATS Annex on Telecommunications and the Reference Paper. Successive rounds o f consultations did not resolve the issues raised, and in April 2002 the WTO established a dispute resolutionpanel to examine the complaint. The Panel met several times with the parties (the United States and Mexican government representatives) and with ten third-parties having an interest in the case. The Panel issued its final report to the parties in March 2004. The Panel decision was not appealed to the WTO Appellate Body, the WTO's Dispute SettlementBody unanimously approved the Panel report in June 2004, and the parties agreed on a planto redress the underlyingproblems by July 2005. Mexico had undertaken specific commitments under the GATS Articles XVI (market access), XVII (national treatment), and XVIII (additional commitments, comprising the Reference Paper). The United States claimed that, with respect to the services at issue, Mexico had: Failed to ensure that TELMEX, the largest operator, interconnects U.S. suppliers on a cross-border basis on cost-oriented, reasonable rates, terms, and conditions. Failed to prevent anticompetitive behavior, as regulations empower TELMEX to lead a cartel that fixes rates for international interconnection and restricts supply. Failed to ensure access by U.S.suppliers to public telecommunications networks inMexico, thus preventing them from providing non-facility-based services within Mexico (through commercial agencies or `comercializadoras') and international simple resale (through cross-border leased circuits). The WTO panel's 238-page final report basically ruled infavor o f the United States on the first two counts and ruled partly in favor o f Mexico and partly in favor o f the United States on the third count. On the first two counts, the panel determined that Telmex was a "major supplier" and that it had engaged in anticompetitive behavior. Furthermore, the panel found that the Mexican government had not only failed to stop Telmex's anticompetitive behavior, but that its international long distance regulations required Mexicanoperators to engage inanticompetitive behavior that was tantamount to a cartel. On the third count, the panel ruled infavor o f the United States inso far as Mexico needed to issue regulations to allow non-facility-based operators (Le., comercializadoras) to provide services, but it sided with Mexico in agreeing that Mexico had not made any WTO commitments to legalize international simple resale (ISR). Inmost OECD countries, ISR is allowed and widely usedby operators. As a result o f the panel ruling, the parties agreed in April 2004 called for the following actions which COFETEL is inthe process ofimplementing: Within two months o fadoption o f the Panel's report, Mexico will revise its international long distance rules to allow competitive negotiation o f settlement rates by eliminating uniform settlement rates, proportional returns, and the requirement that the operator with most outbound traffic negotiate the settlement rate on behalfo f all Mexicanoperators 50 Within 13 months, Mexico will have in force regulations to license `comercializadoras' allowed to resell internationalswitchedtelecommunicationsservices providedbyMexicanconcessionaires Mexico will continue to have the right to restrict international simple resale (use of leased lines to carry cross-bordercalls) 118. The government, through the SCT, i s responsible for monitoring Telmex compliance with its concession obligations; includingthe process o f gradually rebalancingTelmex's highly cross-subsidized telephone tariffs to prepare the company and Mexico's telecommunications markets for competition that was set to begin in mid-1996. Telmex's tariffs were not fully rebalanced by the end o f its monopoly, in part due to the 1994 financial crisis. The failure to rebalance Telmex's tariffs planted the seeds for subsequentchallenges. 119. Regulators have been reluctant to take action on controversial issues. Ever since Congress began debating the new telecommunications law inlate 2001, the SCT andthe regulator COFETEL have been reluctant to take action on a number of controversial issues such as unbundling, collocation, issuing resale regulation. COFETEL, however, has taken action on a number o f low-key, though important issues such as numbering, dominance rules (though these have been frozen due to legal challenges), tariffs, and local loop competition. Due to political factors, the new telecommunications law was never issued, leaving COFETEL without a strengthenedregulatory framework fostering independence and institutional credibility. 120. The 1995 Federal Telecommunication Law called for the creation o f a regulatory body as part o f the SCT but did not specify the powers and authority o f the regulator. As a result, SCT drafted a presidential decree that created COFETEL on August 9, 1996 as a regulatory body to be led by a chairman and three commissioners without set terms, who are nominated by and can be removed at the will o f the SCT Secretary. Between 1996 and 2004, COFETEL has been led by four chairmen, and has not had the stability needed to create a strong new institution charged with regulating a complex and powerful industry. 121. COFETEL lacks adequate enforcement powers and independence. One o f the main reasons for the lack o f independence i s the need for a legal figure that can give COFETEL independent decision- making power without political pressures. Also, the head o f the COFETEL is nominatedwithout a fixed term o f office. A recent OECD institutional analysis study,'* which evaluates the regulatory framework of the Water, Energy, Financial Services, and Telecommunications agencies, concluded that there is an urgent need for a national debate to discuss the creation of a regulatory figure to legitimize these regulatory entities. 122. Mexico has several fairly effective quasi-independent regulatory bodies that can be usedas models for the establishment o f a more independent and effective COFETEL. One example is the Federal Competition Commission (CFC), which was created by the Federal Competition Law. The Commission has legal underpinnings that give it independence and autonomy, and it i s headed by a board with members that enjoy ten year staggeredterms. 123. "E-Mexico'' i s a major initiative to expand ICT access throughout the country and develop on-line applications across several sectors. The objective of this initiative, as stated inPresident Fox's address to the Nation, on December 1, 2001, i s "to use the revolutiono f information and communications to project a national character and reduce the digital gap between governments, companies, homes, and individuals, with a reach to the last comer o f our country." The e-Mexico Project has three major '*See OECD (2004a) "Progress inImplementingRegulatoryReforms: Mexico." 51 initiatives: (i)the connectivity initiative that proposes to extend broadband access to internet to 10,000 locations throughout the country; (ii) the applications initiative, that aims at developing applications in education, health, SMEs, and e-Government; and (iii) implementation initiative; the last includes the buildingtelecenters, accesspoints, and localareanetworks inschools, healthcenters, community centers, and municipalities and creating associations o f producers to use the applications and to train their staff. The e-Mexico program aims to build on the existing capacity o f telecommunications infrastructure (100 percent digital), to use this infrastructure to deliver applications, and to extend ICT to provide government services directly to the citizens. 124. The e-Mexico coordinator who reports directly to the SCT Secretary officially coordinates the program. A high level official in the Office o f the Presidency is also partly responsible for monitoring implementation. e-Mexico has a small staff that is responsible for coordinating the activities o f other secretaries and ministries responsible for the e-education program, e-commerce, and e-Government. 125. Aside from the e-connection program, the e-Mexico coordinator's role i s mainly to act as a catalyst for the line-ministries to develop specific projects that have ranged from developing web pages, to implementing comprehensive and innovative programs to deliver and procure education and government services through the Internet. Mexico's information technology industry has, for example, been very supportive o f the e-Mexico initiatives developed by the Secretary o f Economy to support the development o f the Mexican software industry. The Secretary o f Economy has developed similar programs aimed at usingICT to increase the competitiveness o f Mexican companies inmore than six key industries. 126. After nearly two years of preparatory work, mostly within government, e-Mexico initiatives started operating in 2002. That year the government allocated nearly $75 million for the first o f three stages o f the e-Mexico connectivity program, which would provide a partial subsidy to a private sector company to provide broadband connectivity to 3,200 Centros Comunitarios Digitales (CCDs) or digital community centers connected via a VSAT satellite network. After extensive studies, the e-Mexico program issued on September 3, 2002 a request for proposals to establish 2,443 CCDs nationwide, especially inMexico's poorer and more rural areas. Each CCD would have at least three computers with high speed connectivity and the service provider needed to commit to managing and running these systems as well as providing a range o f services. The government subsidizes these CCDs in two ways: first, by providing free high speed connectivity via satellite transponder capacity; this the SCT had reserved for five years for social purposes from Satmex when it was privatized and from the concessions o f the satellite service providers. The winning bid would have to pay for the cost o f installing the VSAT antennas and the equipment to runthe CCDs, Second, the government subsidy would provide a one-time payment to the CCD service provider to help offset the capital costs o f buildingthe VSAT antennas and CCD equipment. The service provider was also guaranteed that each CCD would be paid $30 a monthby the school or municipality where the CCD was based. 127. On December 3, 2002 Interdirec was selected as the winner o f the first CCD program, outbidding proposals from Telmex and Avantel. Interdirec i s a relatively small company that has been using innovative wireless technology to provide distance education and internet access in several Mexican communities. By 2003, the first 3,200 CCDs were installed inall the municipalities o f the country, and in 2004, two more biddings for 2000 CCDs each were launched by the SCT. The winner o f both biddings was Telmex. By the beginningo f 2005,7,200 CCDs had been installed around the country. For 2006, the SCT expects to have more than 10,000 CCDs installed inthe country along the municipalities. 128. The Telecommunications Social Access Fund (Fondo de Cobertura Social de las Telecomunicaciones, FCST) promotes telephone access in rural and low income areas. Congress 52 authorized a US$70 million budget to develop the Fundin 2002 and a committee with representatives o f public and private sectors was created. COFETEL developed both a study to determine which localities would benefit from the fund and a feasibility study. The uniqueness of this fund is that its objective is not only to install public telephone lines, but mainly to increase telephone penetration through residential lines inrural localities with more than 500 inhabitants. Specifically, the objective i s to reduce the access gap o f localities with the least telecommunication coverage around the country. The FCST was designed to use at least 70 per cent inresidential fixed lines (with any technology) and up to 30 per cent inpublic telephone lines. 129. The FCST has some budgetary shortcomings. The FCST depends entirely on arbitrary budgetary allocations instead o f on an annual assessment on the revenues o f telephone companies o f 1 percent to 2 percent as i s the case with similar funds in Brazil, Colombia, and Peru. Inthe United States, telephone companies are collectively assessed by several universal access funds an average o f 2 percent to 5 percent of revenues. In2002, the FCST was allocated $70 million and in2003 it was allocated less than half the 2002 amount. The International Telecommunications Union (ITU) estimates that in 2003 Mexico's telecommunications sector had revenues o f around $17 billion, which means that a 1 percent tax on Mexican telecommunications companies would raise about $170 million for universal access programs on a sustainable basis. Such an assessment would provide sufficient funds to put in place a more effective, comprehensive, and sustainable universal accessprogram. 130. In2004 a first bidding o f approximately US$30 million and use o f the radio frequency spectrum (1.5 Ghz) was launched. This bidding included around 3,930 localities that will benefit around 3.2 million inhabitants. This first bidding is expected to provide around 125 thousand lines in which 30 per cent o f the cost o f each line will be covered by the Fund(capital investment) and the rest by the winning operator (operating costs). The operators will charge only the outgoing calls through a prepaid calling card scheme; the installation fee and the telephone device will be covered by the Fund.The Fund will cover the first 30 per cent of the lines demanded in each localities, the rest will be covered by the operator entirely or by incoming competitors. The incumbent, Telmex, won the bid inMarch 2005 and i s expected to start expanding the service in these localities. A new bidding i s expected to start in June or July 2005. 131. The FCST fills an important gap given that the SCT has not imposed universal access requirements on Telmex. Prior to 2002 when the FCST program was created, Mexico's only universal access programs were those contained in Telmex's concession requirements, the Telecomm payphone program, and concession build-out requirements inother carrier concessions. As a result, Mexico didnot have an explicit universal access program aimed at increasing access to telecommunication services. When Telmex was privatized in 1990, it was given a six-year exclusivity period. As part o f its concession requirements, Telmex was required to comply with certain network expansion needs by 1995, including expanding the number o f telephone lines nationwide by 12 percent each year; providing telephone service inpopulation centers greater than 5,000; and offeringrural telephony via payphones inpopulation centers greater than 500. The Telmex concession required that after 1995 it must comply with four expansion programs negotiated with the SCT. Telmex complied with its concession requirements. Since 1999 the SCT has not imposed new build-out requirements on Telmex, though it is required to continue to provide services to all areas where it currently provides services as a result o f its concession requirements. The SCT has not imposed new universal access requirements on Telmex as part o f an effort to undercut Telmex's opposition to SCT and COFETEL initiatives and regulations aimed at increasing competition. 132. Several e-Government initiatives will reduce transactions costs within government and for businesses and citizens. The government has implemented some key projects and initiatives that will 53 provide different types o f on-line services to the citizens such as the Citizen's Portal Cwww.rrob.mx), Compranet, Declaranet, and Tramitanet. The Federal Government has also enacted legislation to develop e-Government and modernize its processes (for example digital signature, E-commerce legislature package in2000, Electronic Declarations Legislation in 1998). Some keyprojects are discussedbelow: System for Government Procurement (www.commanet.gob.mx): This system allows all government purchasingdepartments to publishtheir procurement needs, includingproducts, services, leasing, and public works. Potential suppliers may access this information and bid for specific contracts usingthe same portal. There were almost 28,000 public bids conducted usingCompranet in 2003. Today, 100 percent o f public bids are registered in Compranet and 40 per cent of them were totally conducted via Internet (64% o f value). Duringthe first half o f 2004, another 80 government agenciesjoined the Compranet system. Single Registry of Authorized Persons (Registro Unico de Personas Autorizadas, RUPA): This registry will enroll citizens who work frequently with many government agencies and will issue them a universal identification number that will be used as the only identification required to work with government agencies. The decree mandating the creation o f this registry was published inMay 2004 with the expectation o f having the basic registry fully operational by the end o f November 2005. The Ministryof Finance worked together with the Ministry of Public Function (SFP) to define the rules and standards, procedures and technology for the advanced electronic signature to be used across all government agencies that would use it. RUPAwill be consistent with these regulations. Citizen's Portal of the Federal Government (www.Pob.mx): The portal is part o f the e-Mexico system and the Presidential Agenda to Promote Better Government. It represents one of the most noticeable results o f the digital government strategy. As o f today, www.g;ob.mx receives 12,000 visitors daily. Its structure represents a "world-class" e-Government portal at a national level. The portal has been acknowledged internationally and was awarded first place in the 2004 Stockholm Challenge Award for the e-government category. www.gob.mx represents a single access point to all levels o f government services and provides answers to the informational needs o f the community and a link by which to perform government-to-consumer, government-to-business, and even government- to-government transactions. Tramitanet (www.tramitanet,gob.mx): This sub-portal consists of an online catalog of federal and some state "tramites" (government procedures/papenvork) made available on the Internet. For each trumite, information i s provided about procedural steps to be followed, the supporting documentation required, location and working hours o f the government office, and cost o f fulfilling andor complying with the proceedings. The new regulatory framework mandates that electronically generated and submitted federal tramites be recognized and have the same legal effect and value as tramites conducted by traditional methods. Federal tramites can be "signed" with a high security certification procedure administered by a government entity or notary public. To obtain a certificate, an individual or corporation must first appear physically before the certifying authority to verify the user's identity. Unfortunately, this legal certainty i s not available for all state level tramites because some o f them do not have the legal frameworks inplace to support electronic forms and proceedings (digital signature, data messages, etc.). E-Government at the State and Municipal level: The states o f Puebla, Sinaloa, Aguascalientes, Baja California, and Colima are the leading states inthe performance of e-Government. By deploying interactive kiosks developed locally, Colima has become very innovative in leveraging scarce resources to offer digital services, Puebla has partnered with a private company to issue drivers' 54 licenses. It i s relevant to say that inmost o f these states, the personincharge o f IT reports directly to the governor. The great majority of the states have not leveraged the Federal Government's infrastructure to offer better services to their constituents through the use o f technology. To assist them, the Federal Government has provided a portal (www.e-1ocal.aob.m) to help municipalities and states strengthen their institutions and share good practices. Moving Forward 133. The private sector provision of ICT infrastructure needs to be expanded. Although e-Mexico, FCST, and related connectivity programs provide important contributions towards the development o f a knowledge economy, they do not adequately address the most important challenge: that o f dramatically increasing private sector provision o f ICT infrastructure in Mexico. To do so would require a level playing field, elimination of entry barriers, healthy competition, and increased access to ICT services. The following are some recommendations to bringthese goals about. Create a LevelPlayingField: Institutional Changes 134. Rebuilding the credibility, effectiveness, independence, and transparency of COFETEL i s necessary. Despite a promising beginning, COFETEL is a weak regulator that does not inspire the confidence new entrants need to make investments in this sector. This has become the single most important barrier to ICT growth in Mexico. Congress should focus on adopting amendments to the Federal Telecommunication Law that would make COFETEL as independent and transparent as the CFC, and set aside amendments over more technical issues, such as interconnection, tariff, and unbundling which could be better addressed by a more effective and independent COFETEL. The amendments should also remove the ability o f the SCT to review or oversee COFETEL and, equally important, mandate that COFETEL immediately adopts and follows transparent and accountable procedures in decision-making. COFETEL should have the power to develop and implement its own regulations and not simply be responsible for implementing those developed and approved by the SCT. The COFETEL Chairmen and commissioners, nominated by Congress, should be appointed for fixed terms, and should only be removed for gross ethical violations. To send a dramatic and clear signal that there i s an institutional break with the past, Congress could direct COFETEL to report to the Secretariat of Economy, which could have an added benefit of increasing coordination between COFETEL and the CFC-the telecommunications and competition regulators. Congress should also enable COFETEL to have adequate financial resources: the organization should be able to finance itself through fines and license fees. It should also be accountable to the Congress for its annual budget. Making COFETEL a more effective regulatory agency with increased autonomy and transparency will not only strengthen COFETEL, but it will also reduce the ability o f disaffected parties to misuse the judicial system to reverse or delay regulatory decisions through the use o f amparos. Eliminate EntryBarriers andFoster Competition 135. The licensing regime for new entrants should be simplified and streamlined. The current concessions, permits, and licensing regime in Mexico i s complex, and imposes burdensome requirements for information that cause unnecessary delays not only for new entrants, but also for current concession holders who simply wish to amend their existing concessions. Mexico would be better served by adopting a simplified licensing regime, on similar lines to those adopted by the European Union, Argentina, Chile, or Peru. A company wishing to provide domestic voice service, or a concession holder seeking to add services to its concession, needs to provide extensive financial, business, network, and technical information to SCT and COFETEL. The entire process takes at least nine to 24 months and requires the outlay o f considerable legal fees. This is in sharp contrast to Europe where class licenses are inmost cases granted automatically or within a periodof weeks. Simplifiedlicensingregimes for value- 55 added services are critical to fostering the development o f ICT service providers and applications. In Europe the process would be at most a simple one o f self-certification. InPeru, the value-added service application i s a simple two-page document, and registration is typically grantedwithin a couple o f weeks. 136. While some o f these barriers may have arguably served a purpose before Mexico's transition to a more competitive telecommunications market, it i s time Mexico adopted international best practice with a more streamlined, less burdensome, and simplified licensing regimethat is transparent and accountable and operates within a stipulated period of time. Such an approach would enable COFETEL to devote more staff and resources to revising and updating regulations and taking action against monopolistic behavior. 137. Requirements to register contracts with COFETEL for all but the incumbent should be eliminated. Currently, all concession holders must submit all of their draft contracts to COFETEL for their approval before they present them to new clients. The approvalprocess for the generic contracts can take months and COFETEL sometimes uses this process to impose additional regulatory requirementson certain services. These requirements also hinder companies from rapidly introducing and tailoring new services and contracts for clients, since the company runs a legal risk if the contract that is signed deviates significantly from the generic contract approved by COFETEL. Inaddition, all interconnection agreements between concession holders must also be approved by COFETEL before they can go into effect. Imposing such requirements on an incumbent carrier may prevent monopoly activities. However, their imposition on new entrants contributes to delays and uncertainty regarding the introduction o f new ICT services since these, because they involve new services and technologies, often do not neatly fit into the regulatory classifications o f COFETEL. International best practice allows non-dominant companies to enter into contracts freely and modify themwith their clients. 138. Legacy voice-centric regulations should be eliminated. A significant number o f COFETEL's regulations, such as the international long distance and interconnection rules, are focused on the provision o f voice services and leave much ambiguity in the rights o f companies that provide data and corporate services. COFETEL should issue resale regulations, as i s allowed under the 1995 Federal Telecommunication Law, to enable any service provider, not just companies with voice concessions, to lease and resell capacity from other companies. The refusal to issue resale regulations may have encouraged companies to invest in infrastructure earlier. Today such regulations mainly act as an entry barrier to new operators who can package services to serve small businesses and other niche markets. While voice-centric regulations need to be eliminated or revised, new regulations are needed to foster and ease the interconnectionof data networks. 139. Other voice-centric regulations are requirements imposed on the concessions granted to backbone operators. These concessions require them to certify that their customers are carrying voice traffic that terminates or originates only on the network o f a Mexican operator with a long-distance license. Such a requirement is an attempt to reduce voice-over-the-internet (VoIP) traffic by forcing backbone companies to police their customers. In most OECD countries, VoIP traffic i s classified as a data service and i s unregulated. As a result, VoIP in OECD countries has grown dramatically. More recently, COFETEL has proactively enforced its regulations by shutting down companies with only value-added licenses but that carry VoIP. 140. Automatic review of regulationsshould be required.Everytwo to three years COFETEL should review its regulations with the requirement that it amends or eliminates regulations that hinder competition, market entry, price reduction, and network expansion in order to respond to changes in technology and market conditions. 56 PromoteEquityand IncreaseAccess 141. Broadband deployment should be fostered and local loop costs reducedthrough competition. Inmarkets where there is morethan one provider of local loops, the price andquality of service improves dramatically. The government should foster increased local loop competition by promoting and easing the provision o f service to business and residential consumers by cable television service providers, by wireless local loop providers, and by data service providers. 142. The universalaccess programshould be strengthened. The e-Mexico connectivity and the FCST programs are important first steps to addressing disparities inaccess to ICT infrastructure inMexico, but they do not constitute nor provide sufficient financing for a comprehensive and sustainable universal access program. In order to reduce the digital divide within Mexico, the Congress should establish a neutral universal access tax (capped at 1.5 percent o f revenues), finish rebalancing tariffs, and transform the FCST into a sustainable universal access agency within COFETEL. As previously noted, a one percent tax on the revenue o f Mexican telecommunications operators could raise approximately $170 million for universal access, more than twice the 2002 congressional appropriation for the FCST and nearly six times the congressional appropriationfor FCST in2003. 143. FCST should be given the mandate to develop an effective and comprehensive universal access program using the FCST funds, and review and modify as appropriate all government programs that seek to foster increased access to ICT infrastructure, including the current e-Mexico connectivity initiative, Telecomm's rural telephony program, and Sepomex's ICT program. Ideally, all three programs should be part o f the FCST, allowing better coordination between them and avoiding duplication o f effort. FCST could be made a part of the new COFETEL, or it could be a part o f the new ICT Secretariat. Ifthe latter i s the case, the FCST Agency and COFETEL need to be given explicit mandates. FCST could be responsible for developing and implementing policies dealing with the allocation and administration o f FCST funds, while COFETEL could be responsible for developing regulations and enforcing them. 144. The government should undertake a consensus-building campaign on the priorities for universal access. It should consult with state and municipal authorities, identify and remove existing legal, licensing, regulatory, and other barriers to private sector participation, and ensure provision of service to underserved communities, and small businesses. The e-Mexico initiative to prioritize widespread access to broadband services over telephone access i s a unique approach that can lead to some innovative programs usingICT to foster greater economic development and improved education. 145. Cross-subsidyissues should be addressed. COFETEL should undertake a transparent and public initiative to determine whether and how Mexico's current interconnection and tariff regime contains cross-subsidies between services and companies. Where cross-subsidies exist, COFETEL should determine whether these should be phased out, as well as how to replace them with more transparent programs that will foster competition. This initiative should examine the Telmex public payphones program in towns with populations between 500 and 2,500, and Telecomm rural telephony, to determine whether the programs are financially self-sustaining, whether there are tariff or regulatory issues that would be obstacles to the independent functioning o f the programs, and whether they require a review o f their cost structure to distribute the costs equitably. The goal would be to reach the underserved with a low-priced efficient service. 57 AGFUCULTURE FACES GREAT CHALLENGESINBOTHDOMESTIC AND INTERNATIONAL MARKETS 146. Agricultural performance has been lagging since the 1980s, compared to the Mexican economy as a whole and to major regional competitors. This i s despite consistently high subsidies, drastic domestic reforms, and the advent o f NAFTA.Thus, whereas annual GDP growth in 1984-2004 averaged 3 percent for the economy as a whole it was only 1.3 percent in agriculture. In per capita terms, agricultural GDP fell at an annual rate o f 0.4 percent duringthis period. Also, while Mexico exceededthe Latin America and Caribbean (LAC) average in total factor productivity growth in 1961-1980 (1.6 percent against 1.4 percent), the opposite was true in 1981-2001 (1.5 percent against 2.3 percent). 147. Decline took place despite continuous large public spending inrural areas and in agriculture. In2005, all federal spending on rural areas (including all rural sectors) was US13.7 billionequal to 43 percent o f the administrative budget o f the federal secretarias involved inrural areas. For the average of 1996-2000, Mexico had the second highest public agricultural spending per workers inLAC, the first per hectare, and also the highest ratio between the share o f agriculture inpublic expenditure and its share in GDP. 148. NAFTA provided Mexico with new opportunities for agricultural exports but the agricultural trade balance with the United States deteriorated. Exports surged after NAFTA (71 percent increase in 1994-1 999), especially in fresh fruits and vegetables, and processed foods and beverages. Opportunities, however, were unequally distributed. Middle and large commercial farmers with irrigation facilities, mostly in the northern states, were the main beneficiaries, while small rain-fed farms remained behind. Imports also surged, more than exports, turning the agricultural trade balance with the United States more negative. The much feared collapse o f Mexican grain output did not materialize, however. The expansion in grain demand, particularly for feed grains, was taken up by imports, while domestic grain production increased slightly. 149. Market share i s being lost in the United States. Despite strong export growth, there are worrisome trends in the dominant export market, the United States, which absorbs some US$9billion or 85 percent o f Mexican food and agricultural exports. Mexico i s still the largest provider o f fresh fruits and vegetables to the United States, with 56 percent market share, but is facing share erosion inmost o f its 24 top export crops, losing market share to South and Central American competitors. Diversification to non-U.S. markets i s minimal despite the signingby Mexico o f 12 free trade agreements (FTAs) other than NAFTA with various countries and trade areas. With 4 percent o f Mexican agricultural exports, the EuropeanUnion is the secondmarket, but market share i s also being lost there. 150. Mexican agriculture faces great competitive challenges. As NAFTA tariff exemptions expire in 2008, Mexican producers will face renewed competition inproducts like maize, sugar, powder milk, and beans, while transitory support programs, like Pro~ampo~~,will be discontinued, at least in principle. Also, as the United States signs FTAs with other regions and countries, giving them preferential treatment, Mexico will face increasingly tougher competition inits most important export market. 93 Procampo is a program in Mexico that was introduced to compensate farmers for the anticipated negative effect o f NAFTA on the prices o f basic crops. 58 Declining Trendsin Agricultural Growth and Productivity 151. Declining Agricultural Growth. Mexican agriculture grew at a fast pace o f around 6 percent annually until the end o f the 196Os, fueled by the incorporation o f new land, irrigation investments, agricultural protection, and favorable commodity prices. By the 1970s, however, growth slowed down to about 2 percent despite a wide array o f government interventions. Growth continued to be weak after 1984 (1.3 percent in 1984-2004), compared to the rest o f the economy (3.0 percent), but performed somewhat better after 1994, that is, upon the signingo f NAFTA (Table 23). 1994- 2004 3.3 1.8 1.9 0.4 1984-2004 3.0 1.3 1.3 -0.4 152. Weak long-term performance can also be gauged from other indicators such as the share in GDP, employment, and exports, all o f which point to a diminishing industry. The share of agriculture (including fisheries and forestry) in GDP fell from 16 percent in 1960 to 8 percent in 1980, and less than 5 percent today. The share o f agriculture inthe labor force fell from 50 percent in 1960 to 36 percent in 1980 and some 20 percent in 2004, while the contribution to exports fell from 13 percent in 1980 to around 4 percent today. 153. Land Productivity. Land productivity, that is, average yield per hectare, grew some 30 percent in 1980-2002 (i.e., 1.2 percent annually), with growth beingmore significant since 1994. This i s the result of yield increases in individual crops plus improved cropping intensity and crop mix. While yield increases were more important inthe pre-NAFTAyears, improved cropping intensity and mix were more important inthe years after NAFTA(see FAO, 2004). 154. Other major LAC competitors have gained in land productivity more than Mexico. Thus, while Mexico's land productivity index grew only 30 percent in 1980-2002, those o f Argentina, Brazil, and Colombia show increases o f 50 to 70 percent (Figure 41). 59 Figure 41: Evolutionof ProductivityIndex in Croplands (1979-1981 = 100) 11801 I 170 160 150 140 130 120 110 100 9 0 1 ' 804 J I 1 , I , ~ I , I I I I , I I I I I I I , 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 Source: FAOSTAT 155. Labor Productivity. Labor productivity inMexican agriculture almost doubled between 1980 and 2002 (Figure 42). Performance lagged with respect to Brazil, where labor productivity more than quadrupled, and Colombia, but was above Argentina. In absolute terms, agricultural output per worker is US$2,533 (1995 prices) inMexico, US$9,988 inArgentina, US$5,737 in Brazil, US$3,769 inColombia, and US$3,766 inall LAC. Mexico, thus, is below major competitors andthe L A C average. Figure 42: Agricultural Labor Productivity Index (1979-1981 = 100) I 400 - - - Mexico 350 - - -Argentina 300 - - Brasil *Colombia 1- -- 250 - 200 -. I 4 --.....--=-- 150 -- 100 - . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 Source: FAOSTAT 156. Yields. Yield growth has been slow for most crops inMexico (Figure 44 and Figure 46), which has yields below other major L A C countries and the UnitedStates for almost all crops (Figure 43 and Figure 45). The main exception i s wheat, which Mexico produces almost entirely on irrigated lands with yields larger than those o f Chile and the United States (Figure 43). In general, most yield increases have occurred in irrigated areas. 60 Figure 43: Yields for Major Basic Grains Figure 44: Trend inYield for Mexico for Major (2001-2004, todha) Basic Grains (1983-2004. tonlha) 012~ Maize Wheal 83/86 87/90 91/93 94/97 WOO 01/04 I +-Maize +Wheat +Beam Figure 45: Yields for Major Export Crops Figure 46: Trend inYield for Mexico (2001-2004, ton/ha) (1983-2004, tonlha) ." 60 50 40 30 20 10 5' 0 O J 4 Tmafoes Awado Oranges Mangoes Melons 83/86 87/90 91/93 94/97 98/00 01/04 o Mexico IArgenlina o Brazil o Colombia IChile m USA +Tomatoes +Amado +Oranges +- Mangoes+-Melons] 157. Total Factor Productivity (TFP). TFP measures productivity growth not captured by the contributions of individual production factors. It explains the impact of technology improvements measured as the residual output growth after the effect o f growth in all factors has been considered. Annual TFP growth inMexico in 1980-2001 was 1.5 percent, smaller than the 2.3 percent o f the period 1961-80, and smaller than that o f other major L A C competitors and the L A C average (Table 24). Table 24: Annual Growth Rates of TFP inMexico and Other Countries 158. Mexican agriculture has lost its long term dynamism. During the last thirty years sectoral growth was sluggish, with agriculture falling behind the rest o f the economy as well as the agricultural 94The factors considered were land, labor, fertilizer, animal traction, machinery (tractors and harvesters) for crops and land, labor, animal capital cost, and feed for livestock. 61 160. Federal public s ~ e in~rural areas~in g ~ ~ general is also high ia Mexico. Tf'hus, 1111005, the for rural arcas ~ i ~ ~ ~ uagriculture arid othcr scctors) estitnaied under the so-csiled d j ~ I g Programa Especial Concurrente was US$13.7 billion equal to 43 percent o f the administrative budget o f the federal secretarias involvedinrural areas. A Weak but Transforming Agriculture Innovation System 161. The Mexicanagriculturalinnovation system was in crisis in the 1990s for three reasons: (1) declining performance o f the national research institute, INFAP, (lnstituto Nacional de lnvestigaciones Forestales, Agricolas y Pecuarias); (2) dismantling o f the national extension service; and (3) greater pressure to modernize technology under stiffer competition brought about by liberali~ation.'~Recent reforms have improved the overall health of the system, focusing on increased competition among research suppliers, greater stakeholder participation inpriority settings, and decentralization. 162. A model in need of overhauling. The Mexican agriculture research and extension system shares similarity with that o f other countries in the region; a state-owned, autonomous, highly centralized, nationally financed institute o f research, with a wide geographic and thematic mandate, including both research and extension, dependent on the Ministry o f Agriculture. This model produced some positive results in the past but was exhausted by the beginning o f 199Os, unable to respond adequately to producers' needs. Also, public expenditure in agricultural research decreased since the early 1980s both inabsolute terms and, especially, measuredper researcher(Figure 49). 163. The massive public extension system of the 1970s was dismantled without adequate replacement. Public agricultural extension started inthe 1940s and grew rapidly inthe 1960sand 1970s to reach 21,500 staff in 1977-1 979. It originally focused mainly on the irrigation districts. In 1981, the Extension Department at the Ministryo f Agriculture was disbanded and its functions transferred to INIA and the irrigation districts.97 Inthe second half o f the 1980s, the extension service was severely reduced and reformed to introduce decentralization, private implementation, and producers' co-payment. It concentrated mainly on commercial producers, with marginal areas being serviced by PRONASOL (Programa Nacional de Solidaridad). This parallel approach failed and public extension was almost abandoned inpractice in 1993-1 994. It was partiallyrevived later on through private service providers in the framework o f Alianza para el Campo (the leading Mexico's farm modernization program operating since 1996), particularly through PRODESCA (Programas de Desarrollo de Capacidades en el Medio Rural), a program operating in the framework o f Alianza to provide training and technical assistance to small farmers. The scope, however, is very limited. 96Roseboom(2004) 97Eckboiret al. (2003) 63 Figure 49: Developmentof researchstaff, expenditures,and expendituresper researcher in the public componentof the agriculturalresearchsystem Index (1976=100) 300.0 250.0 * - * I ' m - m - - -I (I- 200.0 150.0 100.0 50.0 I I - - - = Researchers -Expenditures Exphesearcher Source: Roseboom (2004) 164. The agriculture researchand extensionsystem has three shortcomings.They are the following: 1) lack o f response to actual demands by producers due inter alia to lack o f communication among researchers, extension, workers, and producers; 2) dependence on scarce federal funding, with very few funds coming from state governments and the private sector; and 3) dismantling of the public extension services without adequate replacement, 165. Reforms are beingintroduced to address the shortcomings in the research system. The main reforms are as follows: 1) a competitive nationwide agricultural technology fund for up- stream research, jointly financed by the National Council for Science and Technology (CONACYT) and the Ministry o f Agriculture; 2) state level research foundations, calledFunduciones Produce, which provide competitive fbding for adaptive technology research, validation and transfer; 3) the Mexican Foundation for Agricultural and Forestry Research (FUMIAF), created to help INIFAP handle ago-industry funding; and 4) competitive "mixed" science and technology funds, financed by CONACYT and state governments, that target local agricultural research issues. The first three mechanisms were introduced in the mid 1990s,andthe fourth in2OOIg8. 166. The Fundaciones Produce are a valuable mechanismbut need strengthening. Thefundaciones, created in 1996 as part o fAlianza, are administered by representatives o f producers. Their objective is to bringagricultural research closer to farmers' demands and to reflect regional priorities inthe innovation system. Activities are nationally coordinated by COFUPRO (Coordinadora Nacional de la Fundaciones Produce). Funding comes from the Federal Government mostly. Thus, 82 percent o f the Pesos 570 million 2005 budget came from the federal budget, and 18 percent from state governments. Producers may provide some limited co-financing, usually in land and labor. Research i s carried out by individual researchers or institutions through a competitive fundingprocess. Ofthe 1,005 projects financed in2005, 80 percent were for primary activities and only 5.6 percent for processing. INIFAP is the main research institution engaged inProduce projects, but its participation has declined from around 80 percent in 1996 to 57 percent in 2001 and 37 percent in 2005. Universities have gained importance, with a 22 percent '*Roseboom (2004). 64 share o f funding in 2005, while the private sector, service providers, and the Fundaciones themselves, increased their participationto 12 percent. Agro-industries participate with 1percent only. 167. The funduciones have six main challenges. The challenges are as follows: 1) bringing about closer regular interaction between all participants o f the innovation system; 2) developing financing instruments to facilitate cross-institutional collaboration inagricultural research and to enable multi-year projects; 3) increasing the participation o f the private sector; 4) strengthening technology dissemination; 5) strengthening supervision and the evaluation o f project outcomes; and 6) increasing the limited level o f funding. 168. Importance of public involvement in technology development and transfer. Without a significant increase o f public sector involvement in these areas, agricultural competitiveness may be impaired not only inthe domestic but also in the export market. Thus, a study by Malaga, Williams, and Fuller (2001) concludes that Mexican investments in yield-enhancing technologies could have a greater impact on the future level o f Mexican exports o f tomatoes and other fresh vegetables to the United States than the elimination of U.S. vegetable import tariffs under NAFTA. TheImpact of NAFTA 169. From a closed economy to NAFTA. Before the mid-l980s, Mexico had highlyprotectionist and interventionist agricultural policies. Mounting external debt, a sharp devaluation o f the Mexican Peso, the economic crisis o f the early 1980s, and international pressure, forced Mexico to abandon the import- substitution model and institute an historic process o f sweeping economic reforms. As part o f the adjustment, Mexico entered GATT in 1986, reducing its average tariff level from around 80 to 50 percent. In the agricultural sector, Mexico went further, unilaterally reducing average tariff levels to between 10 and 20 percent longbefore signingNAFTAin 1994. 170. Rapid export growth upon liberalization, before NAFTA. In the five years prior to NAFTA, Mexican exports increased 72 percent-from US$35.2 billion in 1989 to US$60.6 billion in 1994. Agricultural exports increased 50 percent inthe same period-from US$3.0 billion to US$4.5 billion. In contrast, Mexican exports o f petroleum fell 6.5 percent, and exports o f other non-petroleum products grew by 100percent (Figure 50). Figure 50: Percent Changein Agricultural and Non-Agricultural Export Value (1989-2004) 160 140 120 1 0 0 1 , 80 40 20 0 Source: COMTRADE 65 171. With NAFTA, the growth in Mexican exports accelerated. Thus, in 1994-1999, total Mexican exports grew by an annual average o f US$15.1 billion, three times as much as inthe previous five years, while agricultural exports increased 71 percent (Figure 50). Exports of petroleum and other non- petroleum goods also increased. In the most recent period of 1999 to 2004, agricultural and other non- petroleum exports continued to increase but substantially less (38 percent and 30 percent, respectively), while petroleum exports increased a great deal. 172. NAFTA was important but its impact is difficult to assess. NAFTA provided a sliding system for the elimination of tariffs and non tariff barriers." It also established norms for the application o f measures to counter threats to domestic producers and customers such as sanitary and phytosanitary requirements, anti-dumping and countervailing duties, and safeguards. A formal mechanism for resolving disputes was also introduced. Tariffs of many agricultural products were rapidly cut to zero. Only a few sensitive products were allowed tariff rate quotas (TRQs). Mexico currently has TRQ rights for the importation o f dry beans, maize, and powder milk, while the United States established seasonal tariffs and TRQs for certain fresh fruits and vegetables. TRQs on these commodities are scheduled for elimination by 2008. Mexico has not neededto resort to over-quota tariffs since import demand for TRQ products has not exceeded the quota (or the quota was unilaterally increased, as in the case o f maize). NAFTA has been an important milestone but its impact is difficult to isolate from the effects of other major events such as the Peso devaluation o f 1994-1995. Also, NAFTA follows over a decade o f drastic agricultural reform policies. The effect o f NAFTA on specific exports will be discussed further in the next section. 173. Multiplication of FTAs. Followingthe signing o f NAFTA, Mexico embarked on a global strategy to diversify its export customers through aggressive negotiation o f FTAs.Since 1994, Mexico has signed 12 FTAs: with Bolivia (1995), Costa Rica (1995), Colombia and Venezuela (G-3, 1995), Nicaragua (1998), Chile (1999), Israel (2000), the European Union (2000), Iceland, Norway, Liechtenstein, and Switzerland (EFTA, 2001), Guatemala, Honduras, and El Salvador (North Triangle, 200l), Uruguay (2003), and Japan (2005). Little research has been done on the effects o f these FTAs on Mexican agricultural trade, but there is little evidence that headway was made in developing new agricultural export markets. Nevertheless, the potential remains, eventually, for these agreements to provide some cushion against declining shares for many o f Mexico's agricultural exports to the US.market. 174. Exports surged after NAFTA but the agricultural trade balance with the United States deteriorated. NAFTA resulted in fragmented agricultural growth, which favored mostly irrigated and commercial farms growing fruit and horticulture crops mainly for exports; it also favored the export of processed foods and beverages. Overall, sectoral performance in agriculture after NAFTA has been moderate only, below the levels o f regional competitors such as Chile, Brazil, and Argentina. Agricultural imports increased substantially, more than exports, resulting in a deterioration o f the agricultural trade balance with the United States, which passedfrom US$-211 millionin 1991/93 to US$- 1.3 billionin2001/03. 175. Surprisingly, the production of basic grains did not fall but increased slightly after NAFTA. This may be due to several factors: 1) indirect price protection through Aserca's marketing subsidies, 2) the need for farmers to continue cultivating the land in order to receive the Procampo subsidy, 3) an increase in cereal yields in irrigated areas, (4) the resilience o f maize production inrain-fed areas due to tradition, risk factors, shortage o f alternatives, and subsistence consumption, and 5) the segmentation o f 99One major non-trade barrier that NAFTA eliminated was Mexico's import licensing requirements. Before NAFTA about 25 percent o f the value o f U.S. agricultural exports to Mexico (e.g. poultry, wheat, maize, dry beans) was subject to licensing or prior import permits. UnderNAFTA, licenses were converted to tariff rate quotas (TRQs). 66 Mexican cereal markets due to differences in varieties associated with tastes, location, and marketing channels. Inthis respect, local food grains have an advantage not enjoyed by feed grains where imports increased most. AgriculturalExports to the USMarket 176. Competitiveness of MexicanAgriculturalExport in the US Market. The United States i s the major market for Mexican food and agricultural products, currently accounting for about 86 percent of all Mexican agricultural exports? In2004, processedfoodproducts represented38 percent ofallMexican food and agricultural exports to the United States, above vegetables (35 percent), traditionally the largest export category to the United States. The fruits and animal products categories accounted for 15 and 10 percent, respectively. Mexican exporters to the United States enjoy several advantages: 1) proximity to consumption centers, 2) growing U.S. population o f Mexican origin and rapid expansion of Mexican foods and cuisine in the United States, 3) established marketing channels, and 4) preferential treatment underNAFTA. 177. Losing U.S. market share in vegetables. Despite continued increase in the value o f exports o f vegetables to the United States since pre-NAFTA years, Mexico's share inU.S. imported vegetables has declined almost continually from a peak o f 75 percent in 1990 to 55 percent in 2004 (Figure 51). This may be traced to two recent market phenomena. First, competition from Canadian and European exports o f vegetables grown in greenhouses (particularly tomatoes and bell peppers) has increased dramatically in the last years'". Also, the opening of the U.S. market to imports from Central and South America through bilateral agreements (Caribbean Basin and Andean initiatives) has also increased competition for Mexicanvegetables, particularly asparagus, melons, and broccoli/cauliflower. 178. Winning U.S. market share in processed foods, led by beer. The Mexican share of processed foods imported by the United States doubled between 1990 and 2004, from 9 percent to 18 percent (Figure 51). This is closely related to the strong export performance of Mexican beer inU.S. markets, and also to that o f other beverages such as tequila, processed vegetables, and bakery and confectionary products. Figure51: MexicanMarket Share of U.S.Agricultural Importsby Four Major ProductGroups (1989-2004) -- 70 60 50 40 30 20 10 n 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 I . - - - . -Fruits (FresWFrozenj-Vegetables(Fresh/Frozenj-..-AnimalsandProducts----ProcessedFoodsI Source: UNStatisticsDivision. looMost o f the analysis in this section is based on the top 24 individual commodities exported from Mexico to the UnitedStates, accounting for some 85 percent of the value o f all Mexican agricultural exports to the United States. lo'Cook and Calvin (2005) 67 179. Increase also in the share of imported fruits. Mexicanexport share inU.S. fruit imports has also grown rapidly, jumping from 14 percent in 1990 to 25 percent in 1999 before sliding back to 20 percent in2000 followed by gradual growth back to 23.7 percent in2004 (Figure 51). This growth is the result o f a surge inMexican grape and avocado exports associatedwith the lifting o f U S phytosanitary barriers. 180. Declining share in animal products. After a steady increase in 1989-1995 the Mexican share o f U.S.imports of animal and animalproducts dropped sharply duringthe height ofthe Mexicaneconomic recession in 1996 and recovered slowly in the post-recession years (Figure 51). A recent increase inthe share to about 6.5 percent in 2004 was mainly due to the U.S. ban on imports o f Canadian cattle. Nevertheless, there i s a generally downward trend inthe share; it i s related to stagnant trends inU.S. per capita beef consumption, and to increasing U.S.imports of fed cattle from Canada and o f meat products from several other countries, including New Zealand, Brazil, and Argentina. 181. Performance analysis of Mexico's top 24 agricultural exports to the United States. InFigure 52 changes in market shares since 1989 for the 24 top Mexican agricultural exports to the United States are compared with the growth o f U.S. imports o f those products inthree periods: 1) pre-NAFTA (1989- 1994), 2) early stage o f NAFTA (19941999), when most US.tariffs were eliminated and Mexico began signing free trade agreements with other countries, and 3) 1999-2004 when increased access to US. markets was given to third countries through bilateral agreements. The upper right quadrant shows products with positive changes to Mexican share and U.S. import growth during each o f the three periods-that is, products for which Mexico gained a larger share o f a growing market. The upper left quadrant includes products with negative Mexican share changes but positive U.S. import growth-that is, products for which Mexico was losing share ingrowing U.S. imports. Investments inthese products to increase productivity and eliminate remaining barriers could have the highest payoff. The bottom left quadrant shows products with negative Mexican share changes and negative US. import growth. Investments inthese products would probably have the lowest payoff. Figure 52: Market Share Changes versus US Import growth for Major Mexican Agricultural and Food Exports to the US (1989-94,1994-99,1999-04) 45 9 + 35 Other + 25 1 A Cucumbers 1 ,g 15 1 + 5 5 A Malt A Citrus .-vj iEn -5 Coffee -15 3 -25 -35 45 -25 -20 -15 -10 -5 0 5 I O 15 20 25 Mexican Sharechange(%) Source: UNStatistics Division. 68 182. Mexico i s exporting most of its top food and agricultural exports to the United States, into growth markets. Table 25 shows the details for the 24 products. United States imports from all sources decreased for only 5 products in 1999-2004; coffee, bananas, melons, sugar, and cattle. However, Mexico lost market share in 15 o f its 24 top products in 1999-2004. Note that while Mexican exports o f cucumbers, mangoes, asparagus, and cauliflowerhroccoli to the United States have all been growing in recent years, Mexico has been losing market share o f these products. Inthe case o f tomatoes, peppers, and onions, exports o f which have also been growing consistently, Mexico was able to prevent erosiono f its market share. Mexicantomato exports, for example, were losing market share to Canadian greenhouse tomatoes until 2000 when the Mexican tomato industry began adopting similar technologies and recapturing some o f its lost market'''. Inthe case of grapes and "other beverages," which experienced impressive growth in the pre-NAFTAand immediatepost-NAFTA years, the Mexican shares o f growing U.S.imports were substantially eroded inrecent years. Squash, melons, strawberries, eggplant, and cut flowers show also signs o f losing their competitive advantage. The success stories inagricultural exports to the United States are avocados and malt beverages (beer), the latter at the expense o f Canadian and European beers. Loss o f market share in asparagus, mangoes, bananas, and grapes i s due to competition from Chile, Peru, Ecuador, and Brazil; that o f melons, cauliflowerhroccoli, strawberries, and coffee i s due to competition from Central American and Caribbean exporters. Table 25: Changein MexicanShare of USImports and US Import GrowthRates for Top 24 Agricultural and Food CommoditiesExportedby Mexico to the US (1989-94,1994-99,1999-04) percent ______l__l_--_-__l_-____l__l_______ TOTALAGRICULTURE 0.3 2.2 0.5 4.3 6.9 7.5 Tomatoes -5.0 -20.7 0.1 8.4 14.9 8.9 Cattleandcalves -12.4 -1.3 70.7 11.7 -2.8 -11.5 peppers -4.4 -4.5 0.2 11.7 11.3 15.1 Cucumbers -0.6 -6.6 -6.3 4.1 5.1 19.7 Grapes,fresh 3.3 21.6 -10.9 3.8 15.3 6.2 Squash -0.2 1.8 -3.1 12.3 8.9 13.0 CauliflowerBroccoli -5.1 -6.3 -8.4 5.0 7.3 6.6 coffee -7.2 2.0 -7.9 0.4 3.1 4.8 Onions -6.7 -9.0 2.3 16.7 0.5 6.1 PeCanS -4.4 -1.3 1.9 43.5 5.3 14.8 Citrus, fresh 4.8 -15.0 10.8 30.4 28.9 4.2 Melons -26.9 15.1 -13.6 3.1 14.1 -0.6 Strawberries 17.0 1.9 -9.5 9.2 13.1 5.1 Mangoes -3.7 -14.2 -12.4 18.3 9.0 1.3 Asparagus, fresh -11.5 -4.8 -17.8 19.2 19.8 9.8 FruitJuices 0.I 0.3 -1.2 -1.9 3.5 1.0 Avocados 3.6 20.7 13.8 41.7 23.7 16.7 Redmeatandproducts 0.2 0.5 0.3 0.8 4.0 11.9 Eggplant 0.0 -3.7 -9.8 15.6 2.0 17.4 sugar -6.1 5.9 -2.9 0.4 -0.6 -1.7 cutflowers 0.5 0.9 -2.3 5.9 7.1 3.6 Bananadplantains 3.7 -2.2 -2.4 4.2 2.3 -1.3 Malt beverages 1.o 16.5 7.8 4.4 12.5 7.9 Otherbeverages 6.1 16.9 -10.9 11.4 14.8 28.1 Source: USDA. ~ IO2Cook and Calvin (2005) 69 183. Slow market diversification. Mexico has attempted to diversify export markets since the signing o f NAFTA as a means o f reducing dependency on the United States. The European Union i s the second major export market for Mexican agricultural products, but only about 4 percent o f Mexican food and agricultural exports go to the EuropeanUnion in spite o f the EU-Mexico FTA signed in2000. Substantial penetration o f E.U. markets is difficult because o f distance and fierce competition from Eastern European, African, Middle Eastern, and South American countries. The main opportunity for export growth to the European Union will likely continue to be beer. Mexico has been losing market shares in the European Union primarily to Middle Eastern and African countries with which the European Union has preferential trade arrangements. Food and agricultural exports to Canada are growing but from an extremely small base. Moving Forward 184. Key to further export growth i s not so much market access as removinginternalconstraints. Policy recommendations to boost exports often focus on issues limiting access to foreign markets. Mexico, however, has already quite successfully tackled market access through its aggressive bilateral negotiation strategy. For Mexico, then, further export growthrequires enhancing competitiveness through increased productivity in the production and marketing chains. Some specific export promotion efforts needto be included inthe policy mix. However, the key general component of any successful strategy to enhance agricultural exports must be substantial efforts in critical areas such as irrigation, transport infrastructure, and technology generation and dissemination. These areas are examined below. Targeted promotionto diversify and expandMexicanexport markets. Ingeneral, becauseof the small amounts involved, the marginal return to increased investments to promote Mexicanexports in the European Union, Canada, and Japan is likely to be low. Nevertheless, some specific Mexican items have found niches in those markets that might be exploited by targeted export promotion efforts; these should be aimed at developing strong relationships with specific foreign buyers and branding Mexican products. Such products include beer, avocadoes, mangoes, guava, peppers, dried legumes, and possibly grapes. Targeted export promotion could also be used to reverse the downward trend o f the Mexican export share o f cucumbers, asparagus, broccoli, squash, strawberries, eggplant, and cut flowers. Targeted assistance instruments could include export credit guarantee programs, export incentive programs for specific high priority products and markets, market research assistance, product and market specific export counseling, and assistance with documentation and other export requirements. Improvement in irrigation. Because of their nature, most Mexican agricultural exports require irrigation. Hence, improving this is critical to sustaining export growth. This improvement i s not limited solely to hydraulic infrastructure; to improve irrigationefficiency, institutional strengthening must also be includedto improve the technical, administrative, and financial management capacities o f the water sector. The need is particularly critical in the eastern and central regions o f Mexico characterized by a large number o f small farms with limited access to resources to expand production. Since water scarcity i s a great problem in major agricultural areas and agriculture i s the dominant user o f water, water saving techniques need to be encouraged to raise water productivity. Increasedpublic investmentsin trade-related transport infrastructure.Particular targets would include: additional bridges, access roads, and rail lines to cross the border; additional commercial inspection facilities; the modernization and expansion o f Mexican port facilities; updating and expansion o f intennodal transportation facilities; and improved highways to handle the rapidly 70 growing truck traffic. Inaddition, significant attentionmust be given to reducing the time and cost o f currently lengthy and cumbersome customs clearanceprocedures at US.border crossing points. Increasedpublic investmentsinagricultural researchand extension. Investments inresearch and extension are critically needed to adapt and disseminate technologies developed in other countries, particularly for small and medium-sized agricultural and agribusiness firms. Also needed i s the generation o f new technologies adapted to the conditions o f Mexican resource endowments. Research is also needed on a broad range o f topics related to economic, animal, and plant systems; ago-processing; business, risk, and resource management; and a host o f other topics to enhance and support decision-making all alongthe supply chains. An enhanced market intelligence system. Because globalization is forcing markets to operate in fundamentally different ways, agricultural producers and agribusinesses need to function indifferent ways too. Globalization is forcing marketing and distribution systems to be more tightly aligned with producers. Unfortunately, in Mexico the generation and flow o f information needed by agricultural producers and agribusinesses is not keeping pace with the growth o f globalization. Although much has been done to enhance market information, numerous Mexican agricultural producers and agribusinesses still lack access to information they critically need to manage risks and make sound production and marketing decisions. A better functioningland market. Important reforms have taken place inMexico's landlegislation and land administration since the constitutional reform o f 1992.Legislation now allows ejidos'03and other communities to privatize land ownership if so they decide. Also, Procede, the land titling program that has now covered more than 80 percent o f ejidos (but not many communities), has clarified land rights inthe ejidos, thus facilitating land transfers. Inejidos with high land values due to irrigation or peri-urban location, land rental and purchase markets have developed involving ejido members and outsiders. On the whole, however, legal restrictions on ejido and community lands do still interfere with land allocation and transfers, and in low value areas transactions involving outsiders are still scarce. Recent surveys also show that agribusinesses and large farmers prefer to rentrather than purchase land inorder to avoid hightransactioncosts and possible administrativeand political pr~blems.''~ This may leadto unsustainable soil and water use. A more efficient and effective SENASICA. Additional and stable resources to support the operations o f the National Service o f Health, Innocuousness, and Food Quality (SENASICA) would greatly improve Mexico's effectiveness in responding to food safety and sanitary and phytosanitary (SPS) problems, and prevent hazards more effectively. A 2004 external evaluation o f SENASICA by the Inter-American Institute for Cooperation on Agriculture (IICA) recommended changes such as the following: making available funds for emergencies or producer compensation in case of outbreaks, a closer working relation with the private sector, reducing the level o f politicization, increasing the number o f staff with hightechnical skills, and better coordination with the Ministry o f Health. IO3An ejido is a legal entity ofthe "social interests sector", as defined inMexican legislation. It is a land owning peasant community or the land owned collectively by the members o f such a community. IO4See World Bank (2005); Hollinger (2004); and Deininger and Lavadenz (2001). 71 111. MOVINGTOWARD THE INNOVATION-DRIVEN STAGE OF DEVELOPMENT 185. If Mexico wants to advance beyond competing solely based on efficiency and move to the "innovation-driven" stage of development, it needs to invest more and more effectively ina) innovation; b) quality; and c) education and training. Improvements in these areas are being supported by Bank operations,Io5 but further progress i s necessary. Such progress will allow Mexico to engage in higher value-added activities based on innovative products and processes; it will thus place the country on a more equal footing with its OECD brethren and allow the achievement o f faster growth to secure strong job creation and significant poverty reduction. INCREASEDINNOVATION AND TECHNOLOGY DEVELOPMENT BETTERLINKSTO THE AND PRIVATE SECTORARE ESSENTIAL FOR PRODUCTIVITY INCREASES AND FASTER GROWTH Description and Challenges 186. Innovation is a major driver of economic growth and productivity. Roughly half o f cross- country differences inper capita income and growth are drivenby differences intotal factor productivity (TFP), generally attributed to technological development and innovative capacity.'06These differences should be understood in a broad sense, referring not only to the ability to engage in research and development (R&D) activities, which eventually may or may not be transformed into new outputs; the differences should also extend to the efficient use o f modem operating technologies and to the adoption and adaptation of new ones. According to Prescott (1998), to understand large international income differences, it is necessary to explain differences in TFP. He argues that one o f the main candidates to explain those gaps is resistance to the adoption o f new technologies and to the efficient use o f current operating technologies, which in turn are conditioned by the institutional and policy arrangements a society employs (investment climate variables). Escribano and Guasch (2005a, 2005b), using firm level data from investment climate surveys, investigate the determinants o f productivityusinga large array of variables in four large categories: red tape, corruption, and crime; infrastructure; quality, innovation, and labor skills; and corporate governance. They find that in most countries o f Latin America and Asia covered by their study variables relatedto innovation, quality, and training are important determinants of productivity. 187. Lederman and Saenz (2003) present econometric evidence that suggeststhat innovation outcomes, namely patents per capita, are an important explanation o f the levels o f development observed around the world. On the other hand, spending on R&D, which often includes not only basic research but most expenditure on adaptation o f technologies, has typically been usedas an indicator o f innovation inputs.'" Lederman and Maloney (2003a) examine the relationship between the R&D effort and the development process and find that the share o f GDP dedicated to R&D increases with income per capita inthe average country; not only that, but they find also that several highgrowth comparator countries (such as Finland, Korea, and Israel) had dramatic take-offs relative to the benchmark, a path that China and India have 105The MX CompetitivenessDPL I, MX Innovation for CompetitivenessAPL I,MX Basic Education APL 111, the MXEducation Quality APL and MX Tertiary Education APL I, . IO6 Hall and Jones (1999), Dollar and Wolf (1997). IO7 Although it would be possiblethat firms may have innovative sales without spending anything on R&D, it has beenhypothesizedthat in generalfirms and countries needto spend on R&Dto develop absorptive or learning capacity (see Cohenand Levinthal(1990), Forbes and Wield (2000), Griffith, Redding, and Van Reenen(2003), Pavitt (2001)). 72 recently followed. Several studies have estimated economic gains to the firms investing in R&D; estimates range from 25 to 30 percent, a return far above the average return on capital estimated to be around 7 percent.lo8Furthermore, R&D spendingon the economy can increase the return to R&D several fold through knowledge spillovers that, by technology adoption and learning, improve other companies' products and productiontechniques. Lederman and Maloney (2003a) ina panel o f countries estimate that the social returns to R&D exceed the return to investments in physical capital by a factor of 6 to 10 depending upon the initial level o f income per capita. For Mexico, the calculations suggest a social return to R&D above 60 percent. 188. But R&D expenses only (or derived input measures from them) do not reveal all the intricacies at work. The newly available data from innovation surveys has opened new routes for the assessment o f the contribution of innovationto productivity by enabling an explicit estimation o f the innovation production function.log Recent examples o f this line o f research that follow a structural modeling approach']'- which owes much to Griliches (see e.g., Griliches, 1998)-are presented in Crepon, Duguet, and Mairesse (1998), Griffith, Redding, and Van Reenen (2001), Loof and Heshmati (2001), Klomp and Van Leeuwen (2001) and Correa, SQnchezGarcia and Singh (2005). These studies investigate the channels linking investment in knowledge and innovation to productivity performance at the firm level and, in general, they find a significant impact o f innovationoutput on the level o fproductivity. 189. The Mexican innovation system performs below par compared with economies of its income level, and it has a low ability to translate investments into innovation. Lederman and Maloney (2003b) analyze two key indicators o f innovation performance: the number o f patents produced and the number o f scientific publications; they conclude that Mexico is currently under-performing in both dimensions when compared with other Latin American, East Asian, and OECD countries. Dahlman and Kuznetsov (2005) construct knowledge economy indexes for Mexico and find that its relatively low ranking i s due to weaknesses in all four knowledge economy pillars, namely economic incentive regime, education, innovation, and information and communication technologies. The low performance with regard to innovation stems from a combination o f factors: 0 Insufficient private sector involvement in innovation activities. According to the innovation survey done in 2001 for the period 1999-2000, only 28 percent of the Mexican firms reported have been involved in an innovation project and 13 percent reportedperforming R&D activities. There is no information available with respect to the total amount that Mexican firms spend on innovative activities. However private investment inR&D, both as a proportion o f GDP and as a percentage o f total national R&D expenditures, is very low. Mexican firms invest a little above 0.1 percent o f GDP inR&D, which comparesto 1.5percent inthe OECD area, 0.4 percent inBrazil, 0.3 percent inChile, and 0.8 percent in China. Large companies (above 500 employees) account for the majority (69 percent) of private R&D. Some recent studies suggest that the 100 largest Mexican firms have only 12 research groups, which altogether hire less than 1,000 specialized researchers with PhDs. Those firms use a strategy o f learning by using rather than one o f learning by doing.'" Multinational corporations seem to see no advantage in shifting some o f their R&D operations across the border and hence are not providingMexicanlocal firms with role models and spillovers. lo'See Jones and Williams (1998) for a review. IO9According to Griliches (1979), this production function should describe the transformation process that runs from innovation inputsto innovation output. 'loThe bottom line of this approach is that it is not innovation investment but innovation output, measured for example by the share o f innovative sales intotal sales, that should be used to estimate the contribution of innovation to firm performance (e.g., multifactor productivity (growth). "I Resultspresented inthe CongresoNacional de Vinculacidnparala Competitividad organized by the Foro Consultivo Cient&co y Tecnoldgico. 73 74 greatly skewd towards the pubiic sector, only 32 percent of all R&D p ~ ~ s oworking~ tn ~ i e iniSustryin2004. By contrast, the OECDav e I ~ ~ linkagesebetween public~ K&D and the productive sector. Few e o ~ ~ a n i ehave d ~ ~ ~ ~ s h public R&D ~ n s ~ j ~(12 ~t ej ~r c~eofii~~ ~ i o ~ aEtmI ~s ereportedrc~~~~onships ~ s~ ons) or universities (12 p ~ r c of~~~n n o ~ afirms~ ereported r e ~ a ~ i ~ ~ swithp s e ~ i h i This implies limited pn stj to basic and applied research, reduced 13 ~ o ~ n ~ e r c i 8 l i ofaR&D,~and ~ ~ i o 8 low return to public K&D i n ~ ~ s ~ r nLopez-s .e ~ ~ Acevcdo (1002) finds that when firms haw kinks with pubiic research ~ n s their l j~k e ~~~ ~~ofo d~ ~ o ~ s ~ increased ~ ~ c a ~h s o~ ~increasts~ s~~~ ~ ~gi ~ c~ aSiniilary findings exist for the majority of ~~ i o~ n~~ l . otherOECD countri~s.Evidence o fthe low levef of ~ u ~ 1 i ei n- ~pe~r ~~ ~1x1Mex~ ~~ s ~~~i o~ by the Iow provision of t c c ~ ~ o ~ sewicesc in~the public R&D centers. Figure 53 s ~ g ~ a rough the sale of those services ~ c Q ~ ~ ~ ltesting, cR&D c o ~ ~ ~ rprojects,d ~ a ~ y , ~ c ~ e nologrcaI dcve~opnIe~~ associated with tfic National Science and nofogiff, CONAUYT). Even though the eloped \iirh finns has rcascd from 1,391 1x12000 to 3,695 in 2003, inore progress ~tccdsto be made. 75 i s about 24. With a technology coverage rate'I2 o f 8 percent in 2004, Mexico i s at the bottom o f OECD countries. 0 High concentration of the S&T activities around Mexico City. This concentration can be observed in the percentage of total researchers (50 percent), doctoral programs (34 percent), and students inthose programs (62 percent) that are located around Mexico City. Also the programs that support R&D and innovation funds and fiscal incentives still reflect a bias in favor o f Mexico City, which absorbed 51percent o f total resourcesintheir three first years o f operation. 190. The government i s aware of the crucial role that the private sector has to play to boost innovation in the country. Private sector innovation i s at the heart o f the new S&T strategy pursued by the Special Plan for Science and Technology (Programa Especial de Ciencia y Tecnologia, PECyT) hence the goal to increase the proportion o f R&D financed by the private sector from 26 to 40 percent o f total R&D expenditures inthe period 2001-2006. In2005 this proportion was already about 38 percent. This represents considerable progress for Mexico where private sector involvement in innovation activities has traditionally been very low. However, it is still far from many OECD countries where on average private participation represents 65 percent. 191. Several National Science and Technology Council (CONACYT) programs provide incentives for private sector competitiveness and business innovation. CONACYT's work plan has progressively shifted away from supporting pure research toward developing applications that will help firms become more productive. Since 2001, a series of support programs have been launched to address different needs in the innovation strategy o f firms, including sectoral and mixed funds and fiscal incentives. The sectoral funds are in collaboration with federal ministries and the mixed funds are with state or municipal governments. They bothprovide transfers to established and start-up companies on the basis o f project proposals selected through competitive mechanisms. Inthe fiscal incentives program, firms that can claim up to 30 percent of their annual investment in R&D projects are given tax credits. Other key programs include public-private innovation consortia and research networks, venture capital and seed capital initiatives, scholarships for researchers to work in firms, and firm incubators and accelerators. See Mexico Competitiveness Development Policy Loan, Report No. 34570-MX for a full descriptiono fthese programs. MovingForward 192. Mexico should seek to increase its productivity through knowledge-based integration into global value chains. Several steps need to be taken to make progress in the building of a coherent innovation system with a more active role for the private sector and stronger academic-industry linkages. Efforts should focus on the following: 0 Defining and sustaining a long run S&T strategy that i s translated into multi-year budgets which clearly identifynational priorities to increase competitiveness and productivity. 0 Continuing the strengthening o f funding mechanisms and incentives for the National Innovation Systemacross all areas and levels of government. Eventhough there has beensome progress through the creation o f the sectoral and mixed funds, synergies and complementarities need to be further exploited. There are still many enterprise support programs under different ministries and state and local authorities that have an indirect impact on technological capabilities since they are geared 'I2The coverage ratio is defined as receipts over payments o f the technology balance o f payments. 76 toward upgrading the quality o f inputs and outputs. CONACYT must continue leveraging resources bypushing different institutions' innovation agendas. Scaling up programs through the consolidation o f funding.. The focus should be on programs such as those recently launched by CONACYT that emphasize technology upgrading and development, and speed up the formation o f the venture capital industry.,. The regional infrastructure that provides technological services should be strengthenedto complement programs that support demand. Evaluating more rigorously the success of the programs. Evaluations, independently performed according to an accepted common methodology, are necessary to determine whether programs are in fact helping to improve performance o f firms involvedinthe programs. Evaluation results, translated into design and operational improvements, could also help transform the science system from supply to demand based. Facilitating interactions between R&D institutions and industry. The AVANCE (High Value-Added in Businesses with Knowledge and Businessmen) and consortia programs can be viewed as entry points for facilitating such interactions; but more efforts need to be allocated to such programs. More funding is required to make science more relevant to industry by enhancing such links (support programs focusing explicitly on this objective); moreover,, a deep reform i s also needed to enhance incentives for research organizations to cooperate with industry. This will require a review o f the intellectual property rights system and the establishment o f a framework for public institutions to reward staff for success inproductive research and links with the productive sector. Improving the design and implementation o f decentralized funds by strengthening state-level innovation sub-systems; emphasizing the roles played by the state-level science and technology councils and the economic development agencies; establishing regional innovation clusters by enlargingthe focus and sphere ofthe collaborativeR&Dnetworks . Supporting internationalization policies, by (a) financing activities that strengthen the integration o f the Mexican innovation system into the global system and (b) by engaging successful Mexicans abroad inan international knowledge network. PRESSINGIMPROVEMENTSTO MEXICO'S QUALITY SYSTEMARE NEEDED1I3 193. Mexico's quality institutional framework can be broken into the following four elements: 0 A standardization system, through which activities in the public and private sectors are regulated, including health, the environment, consumer safety, commercial information, and industrial, labor, and trade practices. InMexico, official standards are generatedby 11 secretariats and the process is overseen by the General Standards Directorate (Direccidn General de Normas, DGN) inthe Secretariat o fEconomy. A metrology institute that maintains appropriate measurement standards at the national level. In Mexico, this is done by the National Metrology Center (Centro Nacional de Metrologia, CENAM). `I3 This sectionis basedona reportby the Gmpo de Economistasy Asociados (GEA)which was funded by this AAA. 77 A national accreditation system to accredit bodies that assess conformity-testing labs, calibration labs, certification bodies, and verification or inspectionbodies. InMexico, this is done bythe MexicanAccreditationEntity(Entidad Mexicana deAcreditacidn, EMA). 0 Certification, testing, calibration, and inspection bodies--collectively known as conformity assessment bodies-that certify businesses inaspects relating to metrology and quality. 194. These four elements are closely linked. Calibration laboratories ensure that the measurements performed by testing laboratories and inspection bodies are reliable. Calibration laboratories themselves must demonstrate the accuracy and precision oftheir instruments, basedon their traceability to reference measurement standards heldat the national metrology institute(CENAM). The competence o f calibration and testing laboratories and of certification and inspection bodies must be evaluated by a national accreditation body (EMA). All aspects o f these activities rely on standards. Standards provide the basis for evaluation o f all conformity assessment bodies, and define the requirements against which conformity assessment is performed. Insum, conformity assessment provides the vital link between standards and the products, processes, and services themselves. 195. Standards contribute to economic development in a number of ways. Standards lower transaction costs since they communicate reliable and consistent information regardingthe characteristics and quality of products and processes to consumers. By allowing consumers to differentiate products o f different quality, they reduce asymmetric information and promote the creation o f high quality markets. They also enhance competition by allowing easy comparison o f products and services conforming to a given standard. Standards embody technology and act as a channel for technology transfer for the institutions and firms that adopt them. This raises national productivity and competitiveness. Standards increase the efficiency of the development, manufacturing, and supply o f products and services since they facilitate the interfacing o f processes and components, allow for flexibility through component substitution, and bring about economies o f scale. Finally, standards contribute to the safeguard o f the general public and o f the environment by establishing requirements for design and performance characteristics. 196. Nonetheless, there are cases in which standards can have a detrimental effect on economic development. This occurs when a standard imposes constraints on innovation because producers have difficulties adapting new technologies to the technical requirements o f the standard. Furthermore, standards can have anticompetitive effects if only one or a few companies are able to internalize their benefits or control their content. This occurs when the contents o f the standards cover technological areas inwhich a limitednumberof firms have property rights, exclusive knowledge. or the exclusive resources to utilize a technology. 197. Despite these opposing effects, empirical evidence suggests that under the right conditions, standards have had an important beneficial effect on growth. Time-series data for the United Kingdom showed that standards contributed to about 13 percent o f the growth inlabor productivityinthe United Kingdom from 1948 to 2002.'14Another study on the German economy found that standards contributed to about 0.9 percentage points out o f an average overall growth rate o f 3.3 percent in the period 1960-1990.1'5 198. Many of the positive effects of standards on economic growth translate directly into positive effects on trade. Compatibility and interface standards promote trade by allowing countries to specialize ]I4U.K.DepartmentofTrade andIndustry(2005) Jungmittag, Blind,andKrupp(1999). 78 in system components for which they are the most efficient. By diffusing market and technological information across borders, standards allow countries to compete in new international markets. They enhance transparency in the market and allow foreign producers to be aware o f national preferences and technical specifications and adapt their products and services accordingly. B y establishing product and process characteristics or performance, standards reduce transaction costs between business partners in distant countries and reduce information asymmetries. Standards also increase the perception o f quality and improve the competitive advantage o f firms that adopt them. Finally, when the same standards are used in different countries, they promote trade by enabling economies of scale and allowing countries to exploit their competitive advantage. 199. However, there are other circumstances in which standards will have no effect on trade or even act as a deterrent to trade. Standards cannot fulfill their role if they cannot be accessed and adopted throughout an economy. This can occur if standards are not well documented, if they are difficult to find, if they are poorly understood, or if they are protected by intellectual property rights. If each country uses its own idiosyncratic or country-specific, standards, global economies of scale will not be realized and each country may end up focusing exclusively on its domestic market. Ifproducts must be adapted to each market to conform to different national standards, additional adaptation costs can hinder trade. Standards can be purposely used as technical barriers to trade when differences between countries are not justified by legitimate health or safety objectives or technological capabilities, or when standards are not properly publicized. Harmonized standards (between countries) can also distort trade if their content strongly favors one trade partner over the other. A country may be placed at a competitive disadvantage if it does not have the technological capabilities, the industrial infrastructure or the natural resource endowments required to comply with a shared standard. 200. A review of empirical studies based on trade models finds that harmonized standards promote trade and that there i s mixed evidence on the role of country-specific (or "idiosyncratic") standards. In a time-series study o f intra-industry trade between the United Kingdom and Germany, Swarm, Temple, and Shurmer (1996) find that both country-specific and harmonized standards have a positive effect on exports and imports. Moreover, standards play a positive role in trade performance. The study finds that adopting 100 additional Britishstandards raises UK export to import trade ratio by about 14 percent. Ina similar analysis o f trade between the UnitedKingdom and Germany, and ina study o f total German foreign trade, Blind and Jungmittag (2005) finds that it i s specifically harmonized standards that enhance trade. 201. Standards in Mexico can be mandatory (Normas Opciales Mexicanas), referential, or voluntary (Normas Mexicanas). The Federal Metrology and Standardization Law was passed in 1992 and reformed in 1997 to accommodate NAFTA. Mandatory (official) standards can be issued by 11 federal government agencies and reference standards can be issued by the Federal Electricity Commission (CFE) and PEMEX. As o f June 2005, there were 789 official standards in force and 197 more inprogress (Table 26). Also as of June 2005, there were 5,65 1 voluntary norms inplace and 150 in progress. Inprinciple, anyone can propose a voluntary standard, which will then be submitted to one o f 30 technical standards committees (the committees are validated bythe DGNand supported by 8 national standardsbodies) for review. 79 Agency InForce I ~nprogress c;F 1i n m Sectur 8 0 Sagarpa 139 9 Semamat 107 15 STPS 38 3 Sedesol 0 0 ss 195 24 SCT 115 104 Segob 1 0 sSP 0 0 Total 789 197 202, Problems with the standards institutional framework include the following: e Standards that require physical or chemical measurements frequently do not include adequate consideration o f measurement characteristics-for example, traceability and uncertainty-to produce the required level o f confidence. As a result, the standards can become non-viable. This i s especially the case with voluntary standards. e Consultative committees and standardization bodies do not have strong enough ties to science and technology institutions. Even when the law requires the participation of academia, their participation frequently does not affect the relevant standard. Academia has demonstrated an increasing interest in standards, but there are currently few incentives to encourage their involvement. e There are no training programs for standards specialists. Most people learn about standards development through actually participating in the process. This lack o f training programs creates inefficiencies inthe process. e In some cases, the agency that issues the standard can be biased toward a particular sector. As a result, the overall interests of society suffer. e A mechanism to conduct costbenefit analyses of standards has not been developed. As a result, the mandatory review o f standards every five years often occurs without any information about the impact o f standards. Hence, the review becomesmerely administrative. e The DGNdoes not have sufficient resourcesto carry out all o f its mandated functions. e There is a tendency on the part o f both producers and sectoral authorities to generate standards that can serve as trade barriers. DGN's role is to prevent this. Metrology and Calibration 203. Metrology i s the study of a system of measures. Scientific metrology i s the branch o f metrology concerned with developing measurement standards and promoting their acceptance and equivalence. Legal metrology consists o f the legislative, administrative, and technical procedures established to regulate the credibility o f measurements related to official controls, trade, health, safety and the environment. 80 204. The role of a nationalmetrologyinstitute-CENAM inthe case of Mexico-is to establishthe national measurementsystem, to maintain, develop, and diffuse measurement standards for basic units, and to diffuse metrologicalexpertiseto the economy. The functions of the institute are to do the following: 0 Establish the national measurement system by adopting a national system o f units with a view to incorporating it into legislation. 0 Act as a repository for the physical realizationor reproduction o f national measurement standards. 0 Act as the primary scientific metrology laboratory that disseminates the exactitude o f its measurement standards to industryand other users via calibration services. 0 Disseminate metrological expertise by offering specialized technical consultancy and training services to industry, research institutions, and educational institutions. 205. CENAM has developed a robust set of measurement standards, but there are some gaps. Mexico's National Metrology Center (CENAM) has developed 63 national measurement standards to date, which are used as references to ensure that all measurements in the country are uniform, reliable, and in line with international standards. The National Nuclear Research Institute (Instituto NacionaZ de Investiguciones Nucleares, ININ)is incharge o f another three standards. Although these standards cover a large part o f the measurement requirements o f the country, there remain some areas that require development. One example i s the measurement o f highgas flows inview o f the increasingimportance of natural gas. Another is the production o f certified reference materials-another type of measurement standards; these have a long way to advance. Unfortunately, budgetary limitations make it difficult for CENAM to respond flexibly and rapidly to new measurement challenges. 206. Calibrationinvolvesthe determination of the relationshipbetweenan instrument's inputand the magnitude or response of its output. It serves to establish the accuracy and precision of a measuring instrument. Secondary calibration laboratories constitute the primary market for calibration from the national institute. They calibrate their own equipment to the national institute's and diffuse their measurements to the downstream market for calibration, which includes industrial producers, testing laboratories, inspection bodies, research laboratories, universities, and other final users. Many conformity assessment bodies require that equipment and measurement reference systems be calibrated, or traceable, to other widely accepted metrological references before they issue a product or system certificate. 207. Mexico's metrologyand calibration services serve as a public good, and the country needsto develop them further. Since the costs to develop high-level metrology services significantly outweigh the income received from providing such services inMexico, public funding i s necessary. However, low- level metrology services can be attractive to the private sector. In general, the higher the level in the chain o f calibration services, the greater the involvement o f the public sector because profitability decreases as one moves up the chain (Figure 55). In Mexico, this chain o f services i s not yet fully developed, since there is a shortage o f participants in different segments, and there are technical calibration deficiencies. InMarch 2005,299 calibration labs were registered with the EMA. 81 Figure 55: Metrology Services Chain ater public participation Greater privateparticipa I 208, Secondary calibration services have insufficient capacity in Mexico. As mentioned above, secondary metrologykalibration services (one step down from CENAM) are not profitable, due to the technical requirements o f labs, specialized equipment, and highly trained personnel. Secondary labs include research center labs; centers in the SEP-CONACYT network where CONACYT works with the Secretariat of Public Education; academic labs (Instrument centers of UNAM, Laboratorio de Termometria de la Escuela Superior de Fisico-Matematicas del IPN,among others); materials testing lab (LAPEM)of the CFE; and other private sector labs. The secondary service capacity is adequate in some areas, such as electricity, while more development is required in others, such as volume and firmness. Geographic coverage o f secondary services i s also lacking. About 44 percent o f calibration labs are located in Mexico City, Jalisco, and Nuevo Lebn (Figure 56). Currently, some fifth or sixth level institutions are forced to calibrate their measurements directly with CENAM because there are no secondary or tertiary labs in their area that are accredited or have the necessary level o f measurement accuracy and precision. This substantially increases the time and costs o f the calibration procedures, ultimately causing competitiveness and consumers to suffer. 82 Figure 56: Concentration of Calibration Labs in Mexico City, Jalisco, and Nuevo Le6n (Share of National Total) ReferenceMaterials Torsion norms Time and Frequency 0% 25% 50% 75% 100% 209. More inter-laboratory comparisons and proficiency tests are needed. To have a reliable Mexican metrology system, inter-laboratory comparisons and proficiency tests with calibration labs that are integrated inthe traceability chain are required. This is a common practice inall developed countries; it consists of appointing a recognized, prestigious lab or regional organization to circulate between different labs to test their measurements against known standard values. InMexico, this activity i s not sufficiently established. Inter-lab comparisons and proficiency tests occur only infrequently, meaning that in some cases, the performance o f accredited labs is not reliable. More comparisons are required in order to maintainthe confidence o f thirdparties and comply with international standards. 210. Participating in international inter-laboratory comparisons i s a requirement to maintain mutual recognition agreements. A Mutual Recognition Arrangement (MRA) inmetrology provides for the formal recognition of national measurement standards and calibration capabilities. Through measurement comparisons, an MRA establishes the degree o f equivalence o f national measurement standards maintainedby national metrology institutes and thereby guarantees the international acceptance of measurement results that are traceable to the national metrology institute, by the other MRA signatories. Mexico is a member o f international accreditation organizations, such as L A C (International Laboratory Accreditation Cooperation) and APLAC (Asia-Pacific Laboratory Accreditation Cooperation), which require the Mexican Accreditation Entity (EMA) to demand that laboratories participate innational and international inter-lab comparisons and aptitude tests inorder to be accredited. 83 211. Policies to strengthen the existing metrology system should do the following: 0 Identify deficiencies in the metrology system, including an evaluation o f the measurement instruments and other metrology equipment necessary for the future. Once detected, priorities to correct the deficiencies will have to be determined. 0 Create an inter-institutional body to unite the Mexican metrology system and make uniform the criteria that apply to legal metrology. 0 Prepare a legal metrology framework that reflects the current national and international environment. 0 Develop the national calibration system, which i s barely outlined incurrent applicable regulation. Accreditation 212. Accreditation i s defined as the procedure by which an authoritative body gives formal recognition that an organization or person i s competent to carry out specific tasks. Accreditation is sought on a voluntary basis as a proof o f competence in a given area. Accreditation provides certification and inspectionbodies, as well as testing and calibration laboratories with a means to signal that they are conducting their work to appropriate standards and that they are able to provide reliable services to support quality in firms. The accreditation body evaluates the personnel and supporting management system o f the candidates for accreditation and can request practical tests for laboratories when relevant. These tests take the form o f proficiency testing schemes, through which the measurement results o f different laboratories are compared (i.e., inter-laboratory comparison). 213. The Mexican Accreditation Entity (EMA) i s the first and, so far, the only private organization with responsibility for accrediting organizations that assess conformity. The EMA accredits testing and calibration labs and certification, verification, and inspection bodies. Before the creation o f the EMA, the DGN had responsibility for accreditation. Reforms to the Federal Metrology and Standardization Law in 1992 and 1997 opened the door for a private, thirdparty organization to take over these responsibilities. EMAbegan operatinginJanuary 1999. 214. Mexico's market size and subsidies to the EMA have discouraged the emergence of competitors. The EMA is a non-profit organization whose principal financing sources are fees for accreditation services, dues from associates, use o f the EMA brand, training courses, and donations from federal and state agencies and from regional and international organizations. While most other developed countries have a single public institution in charge o f accreditation, Mexico has adopted the US. accreditation model where a number o f private accreditationbodies fiercely compete on the basis o f price and service quality. However, subsidies from the public sector give the EMA a significant advantage over potential competitors; hence it has remained a monopoly. It i s unlikely that another accreditation body will emerge because o f Mexico's market size, sizeable investment requirements, and opposition from the EMA. 215. High accreditation costs have led to unaccredited competitors. The high costs associated with accreditation from the EMA, which are substantially higher than U.S. and international prices, have led to the emergence o f testing establishments that are not accredited but compete with EMA accredited labs. Such testingestablishments are authorized or recognized by different agencies, including SAGARPA and the Secretariat o f Health. As a result, they act as defacto accredited labs. 216. The EMA's mix of public and private origins has led to a complex governance structure and slow decision-making. The Board o f Directors comprises four sectors, each o f which has an equal 84 number (9) o f votes: the Federal Government (various secretariats); business (represented by various groupings o f private sector organizations); usersklients (labs, certification and inspection bodies); and academia and consumer organizations. Given that these four sectors often have conflicting interests, the decision-making process can be slow and there have been delays in reaching agreements that would ultimately benefit the end user: businesses that need conformity assessment services. However, the involvement o f all the different sectors leads to greater transparency and impartiality. 217. Besides the Board, the EMA has a General Assembly, Executive Commission, Executive Management, and 11 departments. The departments include certification; inspection; testing and calibration labs; technical and relationships; administration; systems; quality assurance; new projects; operations; promotion and diffusion; and human resources. The organization's size and complexity is another deterrent to the emergenceof potential competitors. 218. To minimize technical barriers to trade, national conformity assessment procedures should be harmonized across countries. Complying with a standard or a technical regulation i s only useful if compliance can be demonstrated to the buyer or the government at reasonable cost. Demonstrating compliance through conformity assessment itself i s only useful if the testing and certification requirements are similar inthe exporting country and the importing country. Iftesting laboratories are not recognized abroad, tests on products carried out in an exporting country have to be repeated by a recognized laboratory in each o f the importing countries. An adverse test report inthe importing country can result in the rejection o f an entire shipment. Likewise, if certification in one country i s not recognized abroad, domestic firms requiring quality system and environmental management certification for export purposes-for example, I S 0 9000 and I S 0 14000 registration-need to be certified by organizations in each o f the importing countries. To promote the recognition o f conformity assessment procedures, many accreditation bodies base their organizational structure and their accreditation decisions on well-recognized international guides and standards. 219. Mutual recognition of accreditation procedures promotes trade by decreasing transaction costs and eliminating technical barriers. When accreditation in one country i s recognized by other countries, the work performed by that country's certification bodies, inspection bodies, and calibration and measurement laboratories, will be accepted inother countries. 220. The EMA has been recognized by the main international accreditation bodies, IAF and ILAC. Membership in the International Accreditation Forum (IAF) and the International Laboratory Accreditation Cooperation (LAC) enhances an accreditation body's prospect o f gaining international credibility. Accreditation bodies must demonstrate that they operate at high international standards in order to join the IAF or ILAC. In this sense, IAF and ILAC membership i s a form o f assurance that accreditation bodies are competent to undertake their work and are not subject to conflicts o f interest. Furthermore, these international organizations facilitate technology transfer in areas related to quality assessment and provide a forum for learning from other experienced accreditation systems. 221. The EMA i s a signatory to various Mutual Recognition Agreements (MRA). For full recognition, accreditation bodies must establish agreements with other countries, based on mutual evaluation and acceptance o f each other's accreditation systems. Membership in an at the bilateral level (with one other country) or at the regional or international levels (with many other countries) is critical to guaranteeing the credibility of the national conformity assessment system. MRAs are based on peer evaluation processes, through which signatories evaluate each others' compliance with the agreed-upon requirements and evaluate the performance o f assessment staff. They usually cover 1I 6Also referred to as Multilateral Recognition Arrangements (MLA) inthe case o f the IAF. 85 specific accreditations types (e.g. quality management system certifiers) or products. Mexico has signed the IAF's Quality Management System (200l), Environment Management System (2004), and Product (2004) MultilateralRecognition Agreements. 222. The EMA faces the following challenges inthe short term: 0 To obtain international recognition for the accreditation o f Laboratories and Verification Units, with the aim of having tests conducted in Mexican labs (country o f origin) recognized in the destination country. 0 To fulfill the new ISO/IEC 17011 standard. Beginning in2005, the new 17011 standard i s applied to all accreditation entities inthe world and its completion will become obligatory in2006. 0 To increase domestic promotion o f the EMA. The accreditation process i s largely unknown in Mexico; hence greater promotion o f the importance o f accreditation throughout the country should be a priority. Strategic alliances with all three levels o f the government and with the private sector inall regions will be necessaryto ensurecompliance with the appropriate standards. 0 To broaden and strengthen EMA's structure so that it can provide accreditation services in areas o f highfuture demand. Certification, Testing, and Inspection 223. Third-party certification consists of the provision of assurance by an independent body that a product, service, system, process, or material conforms to one or more standards or specifications. For many standards, especially quality and safety standards, the incentives for self-enforcement are low and producers have much to gain by claiming that a non-compliant product or process adheres to a certain standard. Furthermore, the highly technical content o f some standards may render it difficult for producers to know whether they have appropriately complied with a standard. If there i s no means o f differentiatingproducts that conform to a standard from those that do not, standards are o f limiteduse. 224. Certification provides benefits for producers of goods and services, consumers, and government regulators and for international trade. Manufacturers and service providers can have their products or their management systems certified to particular standards to distinguish themselves from less reputable suppliers. Buyers benefit from certification because it allows them to compare and differentiate products and services interms o f quality, safety, or other desirable characteristics. It avoids the costs o f having each buyer independently validate the characteristics o f the products and services o f a supplier and reduces the risk o f purchasing faulty products or services. By reducing search costs for global suppliers, certification reduces technical barriers to trade. Certification by third-party organizations is increasingly included in trade contracts (Schuurman 1997). Finally, regulators benefit from certification because it provides them with a system to enforce governmental health, safety, and environmental legislation. InMarch 2005, there were 38 certification bodies accredited by the EMA. 225. Testing and inspection help demonstrate that a product or process satisfies certain technical requirements. Testing involves determining the characteristics or performance o f a product or process according to a specified procedure. Inspection i s another form o f assessment that relies on less sophisticated instruments than testing. Independent testing laboratories and inspection bodies can be contracted by a firm to obtain a test or inspection certificate as evidence that a product or process conforms to certain characteristics. In certain cases, testing and inspection are required for firms to implement a quality control system, such as I S 0 9000. The first testing labs in Mexico were created roughly 35 years ago, and in March 2005, there were 362 accredited testing labs and 791 accredited inspectionbodies. 86 226. There are internationally-accepted standards guiding the operation of certification entities, testing laboratories, and inspection bodies."' To play a credible role in the conformity assessment system, testing laboratories and inspection bodies must display many of the same characteristics as certification bodies, notably impartiality, objectivity, and confidentiality. Objectivity relies heavily on the procedures guidingthe evaluation process, the equipmentused, and the staff skills and qualifications. 227. There i s a high geographic concentration of testing labs and inspection bodies. Nearly 40 percent o f Mexico's testing labs are located in Mexico City, Jalisco, and Nuevo Lebn (Figure 57). This has translated into price differences for services in the big metropolitan zones versus the rest o f the country. Inspection bodies are also concentrated, with 50 percent of them located inMexico City, Estado de Mexico, Jalisco, and Nuevo Leon. The only exception is inspection bodies associated with energy efficiency and natural gas, which are linked to CFE and PEMEX distribution locations. The concentration o f inspection bodies creates particular problems in cases such as auto-transport, industrial parks, hydraulics, animal health, telecommunications, tourism, and others, where 100 percent of inspectors are located inthe above four regions, while demand is spreadthroughout the country. Figure 57: Concentration of Testing Labs in Mexico City, Jalisco, and Nuevo Le6n (Share of national total) Electric Food MetaVMechanic Chemistry Sanitary Pgriculture 0Yo 20% 40% 60% 228. Prices for conformity evaluation services are relatively high. Given that there are relatively few accredited firms in relation to demand, prices are essentially determined in a monopolistic fashion, with price differentials based on type o f client and service. Normally travel costs are paid by the client, which puts businesses located outside of Mexico City, Guadalajara, and Monterrey at a disadvantage. Even though accredited domestic firms consider themselves to be cheaper than international competitors within Mexico, conformity evaluationprices are three to four times greater on average than prices inthe United States and the European Union. High costs are linked to accreditation costs for certification and inspectionbodies. 229. Conformity evaluation bodies are regulated by several organizations. Certification, inspection, and testing bodies are regulated by EMA, CENAM, and the DGN, through the accreditation process. A great number o f businesses are also certified by the Standards Board (Consejo de Normalizacibn) and others such as ANCE (a domestic group) and AENOR. 230. Problems in the conformity evaluation process include the following: 'I7Guidelines for testing and calibration laboratories are documented inthe ISO/IEC 17025 standard; guidelines for inspection bodies are documented in the IOWIEC 17020 standard. 87 0 Current legislation allows people who are being evaluated to be part o f the accreditationdecision- making process and responsible for their own evaluation activities. This creates an inherent conflict o f interest. The situation was justified by the lack o f qualified people when the accreditation system was initiated, but now there are plenty o f experts from academia, research centers, and CENAM. Hence, the functional separation o f the evaluators and those evaluated i s viable and necessary. 0 Those responsible for evaluation activities often do not have sufficient technical competence in a givenspeciality, meaningthat their results lack credibility. This weakness has been identified and i s in the process of beingaddressed. 0 Federal agencies have not been very active in demonstrating the conformity assessment criteria for the standardsthat they issue. 231. Demand for conformity assessment services is relatively narrow, meaning that Mexico has a low number of I S 0 (International Organization for Standardization) certificates. Demand has been limitedto the following four sectors: 0 Mexican exporters that need a recognized certificate o f quality for their products; 0 Foreign firms that export products to Mexico; 0 Providers o f goods and services to the public sector inMexico; and Giant firms andor firms listed on the Mexican stock market. This universe encompasses a very small share of the total formal businesses inthe country. For example, only 34,000 export-oriented companies are officially registered, which represents barely2 percent o f the country's total. One manifestation o f this is the relatively low, albeit increasing number o f I S 0 9000 certificates in Mexico (Figure 58). Mexico's I S 0 9000 certificates as a share o f the Latin America and Caribbean (LAC) region's total-roughly 17 percent in 2004-has not grown appreciably over the last decade (Figure 59a). As a reference, Mexico's share o f total LAC GDP was roughly 30 percent in2004. As shown in Figure 59b, Mexico's certificates as a share of the world total increased during the 199Os, but then stagnatedduringthe early 2000s. - Figure 58: I S 0 9000 Certifications, 2004 140,000 7--- d 8 ;120,000 : a 100,000 2s .- s 80,000 5 60,000 0 0 ?! 40,000 F s8 0 20,000 Source: I S 0 (2004) 88 Figure59a: MexicanI S 0 9000 Certificatesand Share of LatinAmerican Total 4,000 1 I 25% I v) 3,500 3,000 20% 2,500 15% 2,000 0 1,500 10% n a E 1,000 500 5% 0 0% : Figure59b: MexicanI S 0 9000 Certificatesand Share of World Total 4,000 1 , 0.6% 3,500 7 0.5% 3,000 -1 5 2,500 2,000 -Mexico certificates as a share 1,500 n 5 of world total (%) 1,000 500 0 g : Em: :$ u~h : arma- maaao a ao ~ mi ; o o o o ~ m - 0 I Note: The apparent drop in the number o f I S 0 9000 certificates in Mexico in 2003 is due to the transition from the old I S 0 9000:1994 standard to the new I S 0 9000:2000 standard. In the 2003 survey, the I S 0 9000:1994 standard was officially replaced by the I S 0 9000:2000 standard and Mexico did not report I S 0 9000:1994 certificates in the survey, only I S 0 9000:2000 certificates. However, in 2003 both I S 0 9000:1994 and I S 0 9000:2000 certificates were reported in many other countries in the survey, thus giving the impression that there was a drop in the relative number o f certificates inMexico in 2003. Source: I S 0 2001,2003, and 2005. 232. The conformity evaluation culture may be less widespread due to high market concentration and informality. Relatively high concentration in many goods and services markets in Mexico means that there are few firms in each sector that have an incentive to get certified-either to compete in international markets or to deter the entrance o f new competitors. The large informal sector in Mexico means that there i s no potential demand for certification or testingfrom a large portiono f the economy. BETTER EDUCATION QUALITY, EXPANDED ADULTEDUCATION PROGRAMS, AND IMPROVED TRAININGPROGRAMSARE NEEDED Description and Challenges 233. Mexico i s not producing the human capital necessary to allow the country to live up to its potential in terms of productivity and competitiveness. The quality o f education, especially learning 89 math and science at the s c c o n c school~level, are too low for an OECD courrtrj. Also, the ~ ~ ~ system docs ~ i oproduce enou t d Iabor through its education system, especial gher e d u c ~ Ictels. ~t. ~ ~ o are needed to proclucc better quality, arid snore e; on refoms at the sch 234 E ~ ~is low in lsecondary~and~tertiary ~ ~d ~ c ~ Mexiconhas made great strides in ~ i ~ t i ~ otter the past dccadcs. However, while pn uniwrsal, enrolliimr; rates in secondary arid higher cducatxon are reta gtven national IIICO~C, and low refative to major trading partners and other OECD count example, at the secondary level, Mexican e ~ r o ~ is~ only~ 62 percent, as compared with eeo n ~ n ~ similar size and with other Latin American countries such as ~ r ~ e ~and~Chile, with net e ~ r o 1 ~ ~ ~ ~ ~ t n a t, ~ ~ d ~ ~ eat rthc~tertiary~Xcsel i~s evenestarker; Mcxico, at 2 Ipercent ~ a c e-'thrd lower than countries with milIar incnnie iekels. Ttats ~ n ~ ~ Igap ~ ~ e ~ i ~ I drop ni' high demand for ~ r ~ c ~ from higher education with rates of return ~ a ~ c s in~ ~ ~ ~cotintries.~ ~ o m e - ~ 236. Mexico i s far behind competitor countries in education quality. These results in math are compared below (Table 27) to the other countries participating in the PISA and shows the large gap in performance as regards the quality o f education. Table 27: Students by Level ofMath, selectedcountries (percent), PISA Country Below Level 1 Level 1 Level 2 Level 3 Level 4 Level 5 Level 6 Brazil 53.3 21.9 14.1 6.8 2.7 0.9 0.3 Canada 2.4 7.7 18.3 26.2 25.1 14.8 5.5 Italy 13.2 18.7 24.7 22.9 13.4 5.5 1.5 Korea 2.5 7.1 16.6 24.1 25 16.7 8.1 Mexico 38.1 27.9 20.8 10.1 2.7 0.4 0.0 Spain united States 8.1 14.9 24.7 26.7 17.7 6.5 1.4 10.2 15.5 23.9 23.8 16.6 8 2 ,Uruguay 26.3 21.8 24.2 16.8 8.2 2.3 0.5 OECD average 8.2 13.2 21.1 23.7 19.1 10.6 4 Source: OECD (2003) 237. Mexico underperforms for its income level. While Mexico's level of development is a major factor inexplaining the results, inparticular relative to other OECD members, its performance inrelation to GDP per capita suggests that it is perfonningfar under its potential (Figure 61). Figure 61: Global Performancein Mathematicsand GDP per Capita, 0 1WW zwoo 3 w w 4 w w 5ww ewoo CDPper csplta Source: OECD, PISA (2003a); EdStats. 91 238. Adult education programs are necessary to raise the skill and knowledge levels of the workforce. Given the low education quality and enrollment rates, a large proportion o f the workforce does not have sufficient education to benefit from training injob skills or to perform efficiently in the knowledge economy. Adult education programs, such as for secondary education certificates, can help. There i s evidence that receiving a secondary education certificate leads to higher employability and higher wages, suggesting higher productivity."' With complete secondary education the probability o f being employed inthe formal sector increases, on average, 13 percent for men and 10percent for women, and their hourly wage increases by almost 64 percent for men (from 1.16 to 1.90 pesos/hr), for women the effect is higher (65%) and the hourly wage rate increases from 1.89 to 3.11 pesos/hr. The effect on wages varies depending on the assumptions used, but recent simulations o f potential private benefits derived from the INEA secondary education certificates (described below) show an average internal rate o f return o f approximately 10.6 percent. This estimate i s a low threshold since it does not take into account the possibility o f INEAgraduates pursuingfurther education. 239. Mexico has an efficient adult education program funded by the National Institute o f Adult Education (Instituto Nacional para la Educacidn de 10s Adultos -INEA). INEA has modernized its out- of-school education program during its 23 years of operation and now utilizes a pedagogic approach based on modules that are flexible and competency-based. The Institute introduced information technology to expand the coverage o f its programs and has a large and decentralized operating infrastructure consisting o f 445 regional centers, 34,000 learning points, 29,000 learning cycles and 2,500 community learning centers (Plazas Cornunitarias). 240. INEA's flexible, module-based program is equivalent in quality to a lower secondary education. The education model used by INEA i s very different from those designed for children and literacy campaigns. INEA uses learning modules for basic competency that cover the secondary education curriculum (grades 7 to 9) and basic practical skills. The program is flexible in terms of schedule and number o f modules required to obtain the certification, it gives individuals credit for the knowledge or skills acquired prior to starting the program. Based on the results o f an entry test, participants may be able to skip modules that cover competencies they already have. According to assessments o f learning outcomes by the National Evaluation Center for Higher Education (Centro Nacional de Evaluacidn para la Educacidn Superior-CENEVAL,), there i s no significant difference between the quality o f the lower secondary education provided by INEA to out-of-school adults and that provided by the formal school system. 241. INEA offers a viable strategy to improve education levels of the workforce in the short term. Eachmodule is supported by a book or several books, articles, leaflets, and even a calculator and a metric tape and references to the other printedmaterials. The basic focus o f all the learning materials used by INEA is constructivist, in the sense that the adult learns new things based on what he or she already knows. Currently there are over 41 modules available for study organized in 8 subject matter groups. Adults interested in completing the twelve-module requirement to attain lower secondary certification can choose modules from the following menu o f options: (a) Mathematics; (b) Language and Communication; (c) Science; (d) Family; (e) Youth; (0 Work; (g) Citizenship and Culture; and (h) Preparation for Upper Secondary which covers a more advanced curriculum in Math, Social Sciences, Natural Sciences and Language. Some elective modules include health and human rightstopics as well as work-related skills modules. `IsThe analysis uses information from the National Employment Survey (Encuesta Nacional de Empleo, ENE) and INEA's student database. See, Annex 9: "Economic and FinancialAnalysis." In:World Bark, MX Lifelong Learning Project. Draft Report. August 18,2004. 92 242. SMEs comprise a huge part of the Mexican economy and worker skills and training play a key role in their success. SMEs make up 99 percent o f enterprises inMexico, employ 64 percent o f the workforce and account for over 40 percent o f GDP (GOM, Economic Census 1999). These firms face many difficulties, perform below the level o f their larger counterparts, and rarely export their goods and services.11gAn assessment by the Secretariat o f Economy (2001) identified lack of access to financing and skills as the mainweakness o f SMEs.The mortality rate o f SMEs i s high-200,000 SME are created every year but only 70,000 (35 percent) survive two years later. The firms that fail report lack o f training and expertise as key reasons for their failure. 243. Mexico has a proliferation of relatively small S M E support programs. Over the last 15 years, approximately 131 federal programs have been created to support small and medium enterprises (SMEs), 16 of them provide subsidized training to workers or entrepreneurs. Most o f these programs seek to improve the productivity and competitiveness of SMEs by encouraging technology upgrading and training. The 10 largest programs reported in the National Survey o f Employment, Wages, Technology and Training (ENESTYC) finance needs assessment, training and consulting services andor knowledge- sharing services."' Despite several differences among programs, they share the common goal o f improving the performance o f individual firms or groups o f firms by subsidizing support services for SMEsona cost-sharing basis. 244. The CIMO-PAC training program provided by the STPS i s the largest in Mexico.121 The program provides training services to approximately 332,000 employed workers per year on a cost- sharing basis with the employer. Due to budget cuts and lower efficiency, the number o f firms participating in the program has declined sharply, from an estimated 94,000 firms in 2001 to 7,000 in 2005 The firms benefiting from the program are preponderantly microenterprises (70%) and concentrated in the manufacturing sector (49%) and in service industries (41%). The potential positive impact o f the program on firm productivity i s minimal due to low efficiency, as indicated by overhead figures close to 60 percent, and the small scale o f the program. 245. Apart from CIMO PAC, the extent to which S M E programs reach their objectives i s unknown due to a lack of rigorous impact evaluations. The evaluations carried out by program supervisory agencies tend to be qualitative and small in scope, measuring beneficiary satisfaction or reporting coverage indicators (Le., they are monitoring studies but not impact evaluations.) Few studies measure net program benefits. Two exceptions to this norm are the CIMO-PAC program evaluations carried out by the Secretariat of Labor and Social Security (STPS) in 1995 and 1997 using quasi- experimental designs. But even these failed to account for biases from unobserved firm heterogeneity and self-selection (L6pez-Acevedo and Tan, 2004a). The lack o f reliable evaluation limits the capacity o f ' I 9 The 1999 Economic Census reportsthat only 6.7 percent o f SMEs export their goods and services. IZ0 The SMEprogramsreported in the 2001 ENESTYC survey by year of creation and supervisory agency are: CONOCER (1995) SEP and STPS; CIMO-PAC (1988) STPS; COMPITE (1997) SE; CETRO - CRECE (1996) SE; FIDECAP (2001) SE: FAMPYME (2001) SE: PAT (1994) BANCOMEXT; MEX-EX(1994) BANCOMEXT; PMT (1998) CONACyT and PAIDEC (1998) CONACyT. A description of these programsis available inL6pez- Acevedo and Tan (2004b). `'I CIMO-PAC stands for Program of ComprehensiveQuality and Modernization (CIMO), re-namedin 2001 "Programade Apovo a la Caoacitacidn-PAC." The program is administeredby the Secretariatof Labor (STPS). IZ2 Informationon the coverageof CIMO-PAC in2005 was provided by the STPS; 2001 figures are given in Lopez- Acevedo and Tan (2004a). 93 policymakers to determine program effectiveness and to assess the gains that might accrue from consolidatingprograms within or across agencies.'23 246. Dated evaluations of CIMO-PAC show mixed or positive results. The empirical evidence collected in 1995 and 1997 by the STPS suggeststhat the CIMO-PAC program o f integratedtraining and technical assistance (as originally designed) was effective in improving the performance of microenterprises and SMES.'*~ Compared with the control group, participating firms had higher rates o f capacity utilization, were more likely to adopt quality control practices, and increased investments in worker training. These improved intermediate outcomes may be associated with increased productivity, but the original analysis did not lead to this con~lusion.'~~Re-examining the evidence, Lopez-Acevedo and Tan (2004a) show that because less productive firms self-select in the CIMO-PAC program, the evaluation results were inaccurate. Usinga different methodology, the authors show that the effect o f the program is positive or mixed when production functions are estimated using first-differences, an econometric technique that controls for the self-selection bias. The government i s committed to a systematic evaluationo f the impact o f the program covering three years of operations (2002-2004) and to using the results o f this evaluation to improve program efficiency prior to expanding program coverage. These actions are supportedby an IDB loan.'26 247. Training seems more efficiently provided by the private sector. Among the several SME training programs besides PAC, possibly the most important is COMPITE, created by the SE in 1997. COMPITE i s now entirely managed, and largely funded, by the private sector. The goal o f this program i s to improve productivity and product quality in the following ways: encouraging adoption o f technology, lowering costs o f production, decreasing the size o f inventories, increasingresponse times in productive processes, and optimizing factory layout. Training is based on a methodology used successfully by General Motors to improve its productivity. The first step is an assessment o f the firm to determine which course(s) might be the most effective in improving firm performance. Once this i s known, firms select the appropriate course from those provided by COMPITE at below-market rate (there are currently 19 different courses available). The courses are organized inthree broad areas: (i) Mexican ISO-9000, which certifies the quality o f in-firmproduction; (ii) productivity improvements; iii)business ethics. Each course combines theoretical and practical aspects; upon completion, firms design strategies for improvement using the principles taught in the course and are tested on them. Firms pay nothing for the initial diagnostic, but pay training costs that vary from US$2,800-$3,900 for 4-day workshops to US$ 700-$1,600 for a 2-day workshop. Although COMPITE has not been systematically evaluated, and i s still a quite small program, there are indications that it i s more effective than federally funded and managed SME training programs. Rigorous impact evaluation requires firm-level panel data on both reasonably sized samples o f program beneficiaries and a control group o f non-participant firms with similar attributes covering pre- and post-program period. This could be done using data collected through the ENESTYC survey carried out by INEGI every two years, if a purposive sample is added to augment sample sizes o f program beneficiaries to permit more statistically robust studies o f impact evaluation. As a first step inthe direction o fcross-program evaluation studies the Secretariat o f the Econmy (SE) through the Intersecretarial Industrial Policy Commission (CIPI) completed a beneficiary database using informationreported by all SME program administrators. 124In2001 the CIMO-PACprogram was substantially modifiedand, therefore, no longer correspondsto the program evaluated inthe mid-1990s. The methodology usedinthe analysis estimatedthe program impact on sales using a Cobb-Douglas production function that relates output to different combinations o f labor and capital inputs and compared levels o f productivity between treatment and control group. Loan ME-0233 ($300 million) is the first o f a three loanprogram supporting the STPS programs. 94 Moving Forward 248. Quality needs to be at the top of the education agenda. Quality and relevance become an important factor for keeping children inschool at the secondary and tertiary levels and linking education more closely to thejob market. 249. School autonomy, accountability and assessment need to be promoted. PISA results and local experience in states such as Aguascalientes and Nuevo L e h , which have made the link between education quality and competitiveness, suggest that Mexico needs to continue efforts to move decision- making to the schools, parental involvement is required to help make schools accountable, and performance needsto be assessedand fed back into the system at the school level. 250. Private financing i s needed to overcome the gap in tertiary education. Expansion through public investment, which represented 0.7 percent o f GDP in 2001, i s fiscally unsustainable in Mexico, which only recovers 11percent o f GDP through taxes. Private institutions now account for 33 percent of students, up from 21 percent in 1992. Several o f the private institutions inMexico are world class, such as Tec de Monterrey, and some international institutions are moving in (for example, Apollo International and Laureate). However, at 0.2 percent of GDP, private financing o f tertiary education in Mexico remains far behind countries such as the United States (1.8 percent), Korea (1.9 percent) and Chile (1.7 percent). Public financing needs to be redirected to provide financial aid to poor students and the broad availability o f (income-contingent) student loans would facilitate increased private investment intertiary education. 251. More resources are needed to continue the expansion of adult education. Precise estimates o f demand for INEA's secondary education certificates are not available. But the number o f certificates grew over the last 10 years by an average rate o f 11.5 percent per year. INEA students come from poor and vulnerable groups, such as women, youth, and indigenous peoples (Schmelkes and Kaman 1996).12' This impliesthat INEAis tackling both inequity and poverty issues through its adult education programs. Moreover, INEA plans to develop new modules catering to special groups such as indigenous youth. In order to expand program coverage, INEA would need additional fiscal resources, improved internal efficiency and the introduction o f incentives to increase effective demand.'28The Government i s already pursuingsome o fthese initiatives, but more needsto be done. 252. Impact evaluations of SME training programs are needed. The historical proliferation o f SMEs support services in Mexico and the absence o f reliable information on their effectiveness, suggests that it i s necessary to take stock and formulate a new and comprehensive strategy for supporting SMEs before considering program expansion. For the PAC training program, the necessary elements neededto design such a strategy are as follows: (a) reliable information on SMEs demand for training; (b) information on the costs and benefits o f the PAC; and (c) feasibility analyses to identify more efficient ways to deliver training at the firm level. Consideration should be given to involving the private sector as a service promoter and service provider while keeping the government as co-financier and provider o f program oversight and evaluation. Once an improved strategy to deliver training services to SMEs is in place, it would be possible to plan for expansion. To achieve these goals, federal funds should be allocated to expand the ENESTYC training module covering not only manufacturing but also services and commerce sectors. Once this comprehensive database is available, rigorous cost-benefit analyses o f the larger Schmelkes, S. and Karlman, J. 1966. "La Educacibnde Adultos: Estado del Arte. Hacia Una Estrategia Alfabetizadota para Mtxico." Mexico: INEA. See also, CONEVyT-OCDE. 2003. Educaci6n de Adultos y Formaci6n para el Trabajo en Mtxico: Informe Base. 12' Program efficiency needs to be significantly increased, as only one of every four adults registered inthe INEA secondary education program receives a lower-secondary certificate. 95 training programs for SMEs could be carried out. The results o f these evaluations would provide the basis for a national training policy for SME workers and provide incentives for private sector worker training investments. 96 253, There are s ~ ~ s ~ differencesa in c o ~ ~ ~ t ~across~ states nin~ Mexico. A l t h ~ ~ ~ h ~ n ~ i ~ t j ~ ~ s ~ c ~ ~ ~ ~ ~ e ~ i ~isiusuaitly eanals v e r ~ s d at the national level, as in the rcst of this report, it is port^^^^^ tu r e ~ o ~ ~h l ti there is a wide variety 1nllte eonlpe~l~~veness ~ c icvets of d ~ ~ ~ ecitiesn and states. Studies r e ~ thcsc diffcrcnces, suggesting rcgiotial priorities for reforms. ~ u ~ - ~ a an ~ o snisap~a r ~ i c u ~ arelevant ~ r ~ y sratc and loeaf ~ o ~ e ~ ehavci~~u sr i s d ~ cover~many ~ ~ ~ ~ eclimate! rcfomis,~i n c ~ u d ~ ~ ~ ~ i ~ j n ~ t ~ e n procedurcs to opun a business, register propefly, arid enforce contracts, v irt .%fcxiccl ~2~~~~i s the first state-levcl report of the World Bank's Doing 8 t h~~~~~~~~.i It ~~ves~~gthetescopt and ifianner of regulations that enhance a s those that constrain it, Xn this project q u a ! ~ ~ ~ t ia ~icacars on business re~wlations i ~ ~ c and thcir e ~ ~ f o r c e mhate~ e ~ created for 12 cities and states that can thcn bc compared tvith R.lexico ell as 154 countries around thc world. 'I'hc E2 Mexican cities and states are: ~ ~ ~ u a s c ~ l i c n ~ c s ~ entcs; Celaya, ~ u ~ n a Chidad ~ ~ ~ u a Jtlhrex,~~ ~ ~ h u ~~huuaad ;~ l a ~Jatiseo;, MBnda, Vucatbn; ~ r a ~ ~ o Nuevo ~ ~ n; Puebla,~Puebla; ~ ~ e c Qu ~ arn, QuerBtaro: San Luis Potosi, $an Luis Potosi; ' ~ I a l r i e p ~ Bstad~ ~ ~ MCxico, Torreon, Coah ~ n ~ , and Veracrux, Veracmz. 255 In the past year, S ~ W Rof the 12 states reformed in the four areas covered in the report- ring property, getting credit, and e i ~ ~ o reontj ~ ~ ~ Changes were made io e sitless r ~ ~ l a ~ ~ ~ n -strr o n ~ f s and red~ic~n~ to taxes, d ~ o ~ ~ s o lprocedures~andneasing the burden placed on co~irts. ~ ~ ~ a i ~ 256 When rmlied g ~ o ~ on ~the1 cost of starting a b'uSines8, Mexican states vary ~ 8 ~ ~ ~ ~ j ~ ~ a n c;l:ipure 67). ~ ~ ~ ~ a s c ~ l ISi ethe~ least costly-7.3 r t e s pcrccnt of state inconie per capita-and still it ranks 33rd ~ ~ ~in cost~to startup.~fnlisco foliotvs, costing 10.9 percent, at 52nd place, b ~ ~ i jGabarone b ~ l y i ~ d ~ ~ ~ oaiad~ S s ~ ~ n t(C-hiIef. Estado de~ ~a ~~ ~~ ~~ o ICO ranks lowest at 127thplace- -far behind Lima (Peru) and C3uatcrriala City. o~ever,it is less e~ge~isive to usiness in any of the Mexi Jakarta or ~ ~ a tie number~ of ~daysato start a b ~ a ~ s across Mexican statcs, altl~ough not as widely as costs. Thc timu spent on start-up in Guan 29 days, in Snn Luis Potosi and lalisco, 51 days, in Estado dc h4Cxico44, and in ~ ~ ~52 ~ r ~ t ~ r ~ ~ 97 perty shows similar variance across es, Securing rights to property i s s to invest and f ~ ~ e trade. And~with fonnalsproperty titles, ~ ~ ~ a ~ ~ reneurs can obtain ~~~ortgaon es ~ s or land. The easiest state in Mexico in which to ion takes 29 days and casts 3.2 percent of ~ r o ~ c ~ ~ vniuc, Stili, this defay is longer than in ~ ~ a d ri30gottSi,or Bangkok. ~ r # ~ c ~ y ~ d , rakes 94 days---as long as in ~ u j ~ ~ ~ ~ and~slightly~ ~ b~ a longer than La Pax ~Boljv~a~, ~ ~ C3~ ( ~ i r-an ~n~reprene~rEstado de Mixico spcnds as many days to register as xn in ~ P a r a but~stifl~nictre~than in 520 Paulo. Howevc ~ a ~ 6.1 percent of ~ ~ value,o the enst to e ~ ~ ~ register property in Estadode MCxico is higher than the Mexicanstates. (Figurc 54 and F1 Figure 64: Time to Rcgister Property in Mexwicu and Figure65: Cost to Register Property other selected economies (?.lenScan States) 259, IMCO defines ~ ~ ~ ~ ~ t as~the abilityeta nattracts and retain ~~~~~~~~~tand finds that t ~ ~ e ~ ent Bows varv s ~ b s t a n t ~assoss~ states."2 a ~ ~ shotlin in Figure 66. the state with the tat;ccst i s hlorelos at -5200, and thc te wxth the highest 1s h.Ier;ico Ctty ( ~ ~ s ~ ~ ~ ~ ~ ~ ~ d eatrabout~$7,000, a l de fas ~ ~ t r d a d e ~ fedcrativas dc Mexico*Diagnbstico y Oh pnsas." e as a ~ ~ u n ~ ~the~growth and dct elopntent ofthc cautitsy for i a i ~ latron betweenthe Doing uu ankrngand inwttnent per vyorker. In otticr uords, srafe'i that ranked hrghfg(or poody) according io Doing R do not necessarily receive the mast (or least) ~ n v e ~ ~~ ~ workert C F i ~ n re is also a largc tariatioa in the relative p ~ ~ f ~ ~of~theadifferent states 00 1,lifCO's i n ~ $ etitiveness frtctors. Frbure 67 shows indices for each nfthe ten factors as well as the position of thc best arid worst ranked state aiong each factor. For iristan the `*Society" f a c ~ o r - ~ is h ~ ~ ~ ~ based oil 17 ~ i ~ dnf~a hc ~ y,~ inelusive, and prepared snci ~ r ~ e best ranked state scorcci a 77 while the worst raizkcd state d a 21. Figurc 67 also provide best and which uurst. In the case of Society, the stax with th ighcsf score also fieid the number nix veacss rank, tchife the state `It`ith rile Io score held the a ~ ~ t ~ Coverdl 31 r nk (see circled numbers), This result IS riot ising, but the sanie does not holdtrue fur stlicr factors, Far i ~ ~ t inathe "Pofitical" catego ~ c ~ ~ which measures five ~ ~ ctfa stable md ~ d ~ a the highest scoring state was only the best state in terms ,the aimnfJ3gure67 1s to show that there is d i ~ ~ e across~s ~ ~ c e ~ ~ different factors of cness and ~~~~~~~~~~~ r C rvcll l?IE some fact well on others This trnplies that ~~~~~~e~~~ ies arc different fbr d Figure 67: State-Level Variations in Competitiveness Factors 0 I P 3 l ! l Note: Higher Competitiveness Index scores are better. The circled numbers denote the overall competitiveness rankingof the state that received the highest (top circle) and lowest (bottom circle) score on each dimension. For example, the 5'h most competitive state inMexico received the highest International Relations ranking. Source: IMCO (2006a). 261. The possibilitiesfor improvementare substantial, but vary across states.According to IMCO, a 10 percent improvement in each o f the competitiveness factors would result in an increase in investment per worker of 89 percent on average. However, this average masks considerable variation across states: the state with the maximum gain would experience an 841 percent increase in investment, while the state with the minimumgain would only see a 12 percent gain. Surprisingly, it is not always the worst ranked states that stand to gain the most, For instance, the state with the maximum predicted gain ranks 14 in terms of the overall state competitiveness ranking, and the state with the third largest predicted gain ranks 21 (Figure 68). 100 Figure 68: PredictedImprovementsinInvestmentper Worker z\ Maximum 350%,I 8410/0 .-C 200% Average Q) m 89% C 6 -I FJ 150% Minimum 100% ------\- 12% 50% 0% 1427211931243216292820172515184302212236107 11 5 8 2 6 1 3 2 9 3 1 States Note: The predicted change in investment per worker is due to a 10 percent improvement inthe variables underlying the 10competitiveness factors. The x-axis numbersdenote the states' overall competitiveness ranking. Source: IMCO (2006a). 262. Improvements in different competitiveness factors have disproportional effects on investment.Take the state of Aguascalientes as an example. InAguascalientes, a 10percent increase in the variables underlying each of the competitiveness factors would yield a 29 percent increase in investment per worker. However, changes in only four factors-Society, Legal, Factors, and Precursors-account for 68 percent o f the overall change (Figure 69). Again, this implies different competitiveness priorities indifferent states. Figure 69: Breakdownof FactorsContributing to an ImprovementinInvestment,Aguascalientes p 4 59 pm3 57 55 53 0 51 Note: The predicted change ineach factor is due to a 10 percent improvement in the variables underlying each factor. Source: IMCO (2006b). 101 263. Benchmarking at the sub-national level can spur competition and generate needed reforms. Mexico City and the 12 states vary dramatically on Doing Business indicators, with Aguascalientes outperforming the rest (Table 28). But a big gap remains between Mexico's best and the ease o f doing business in Bangkok or Johannesburg. Similarly, IMCO showed a large variation in investment and different competitiveness factors across states. States may look for best practices within Mexico-for example, by introducing online procedures to register businesses or property-while also aiming for the pace and nature o f reform in countries like Colombia, Vietnam, or Georgia. Establishing benchmarks at the sub-national level allows reformers to focus on the main constraints to economic growth. Pressureto reform can come from contrasting how different localities implement identical national level regulations. Much o f the inefficiency i s in local administrative procedures, which can be changed by a governor or a mayor. And as the news about reforms spreads, there is increased interest to replicate success stories. 1Aguascalientes (Easiest) 8 San LuisPotosi 2 Guanajuato 9 Coahuila 3 Chihuahua 10 Mexico City 4 Jalisco 11 Estado de Mexico 5 Nuevo Lebn 12 Puebla 6 Veracruz 13 Queretaro (Most difficult) I Yucatan 102 V. DOUBLINGMEXICO'S PER CAPITA INCOMEIN10YEARS 264. Competitiveness improvements could allow Mexico to double its income in ten years, as a number of countries have recently done. As shown in Table 29, the time necessary for countries to double their income has been steadily decreasing. Particularly impressive has beenthe fast growers since World War 11-for example Japan, China, Thailand, Botswana, Ireland, and Chile. Pre-industrial 350 years Britain (1780-1830) 175 years I Britain in19* century I 65 years I Fast growers since WWII (Japan, China, 10 years or less Thailand, Botswana, Ireland, Chile., .) 265. Inorder to double GDP per capita over a decade, annual GDP per capita growth rates o f at least 7 percent are necessary. Figure 70 shows that the East Asian and Pacific region nearly attained this rate during the 1980s and 199Os, posting average GDP per capita growth rates o f close to 6 percent. Meanwhile, Latin America has lagged substantially behindthe East Asianperformance, and Mexico is no exception.'33Yet, the fact that other countries have recently doubled their incomes provides grounds for optimism that Mexico can do the same. However, to do so it i s necessary to continue boldly addressing competitiveness constraints inMexico. 266. Doubling per capita income can have a tremendous impact on poverty. Evidence shows that countries experiencing higher rates o f economic growth can be expected to reduce poverty faster (see left panel o f Figure 11). For example, over the 1981-2000 period, China's poverty rate fell from more than 50 percent to about 8 percent, thanks to an impressive per capita growthrate o f almost 8.5 percent a year. Similarly, between 1993 and 2002 Vietnam cut its poverty rate inhalf, from about 58 percent to about 29 percent, by growing at almost 6 percent a year.'34 267. In technical terms, cross-country analysis has demonstrated that the average total elasticity of poverty with respect to economic growth is -2.4. This implies that, without controlling for any characteristics o f the country, 1 percentage point growth in a country's mean income can be expected to reduce the incidence o f poverty inthat country by about 2.4 percentage point^.'^' There is evidence that relatively high income inequality in Mexice-in 2002 the Gini index was 0.49-could cause the country's true elasticity o f poverty with respect to growth to be closer to -1.6 or -1.7.136Even based on 13'Real average GDP per capita growth rates for Mexico were 0.5 percent, 0.6 percent, and 0.2 percent greater than Latin American averages during the 1970s, 80s, and 90s respectively (World Bank, World Development Indicators). Perry and others (2006). '35World Bank (2005g) based on an updated Ravallion and Chen (1997). `36This is the author's calculation based on Lopez and Servtn's (2005) growth elasticity o f poverty estimates, as cited inPerry et al(2006). The Lopez and Serven formula determines growth elasticity according to a country's per capita income, poverty line, and Gini coefficient. For Mexico, the author used the following 2004 data: Rural moderate poverty line per capita (US$ per month) $90; urban moderatepoverty line per capita (US$ per month) $135; GNI per capita (US$per month) $564; rural population (% o f total) 24%; urban population (% o f total) 76%; total population 103,795,200; share o f rural population inmoderate poverty 57%; share o f urbanpopulation in moderatepoverty 42%. This data enabled us to calculate a national moderate poverty line o f $121 per capita per month (this number is weighted by rural and urban shares). Hence, GNI per capita / poverty line = 4.6. According to 103 the lower inequality-adjusted elasticity, doubling per capita income could theoretically eliminate poverty altogether. Although this is probably unlikely, it i s clear that large growth gains from increased competitiveness would leadto plummetingpoverty rates. u a3 5 1 n the Lopez and Serven table, this would equate to a growth elasticity of poverty between-1.6 and -1.7. Data sources includeWorld Bank, World Development Indicators (population and GNIdata); World Bank (2005k) for poverty data; World Bank (2005g) for the Gini coefficient. 104 ANNEX 1:THE COMPETITIVENESSAAA PROGRAM 1. In2003, the Government of Mexico (GOM) declared increasing country competitiveness to be a key priority of the Fox Administration. This was a response to the modest growth, employment, export, and competitiveness indicators as well as concerns about the country's ability to respond to the pressures induced by advancing globalization, such as competition from China and the end o f the textile quota system in 2005. Within its ongoing dialogue with the World Bank, the GOM requested assistance to develop a strategy and an implementation agenda for the overall improvement o f country competitiveness. In particular, the GOM requested assistance in identifying the agenda and producing well-focused analytical pieces on key issues to address and facilitate the implementation o f that agenda. 2. Responding to that request, the World Bank agreed to cast its assistance in the form of a Programmatic AAA, non-lending TA, to allow for the provision o fjust-in-time analytical and technical support, and assistance to diagnose and formulate the country strategy, as requested by the GOM, inthe area o f competitivene~s.~~'Pre-programmatic assistance started in FY04 and helped define the GOM's Competitiveness Agenda 2004-2006 (see Annex 2 for a full description of the pre-programmatic work). 3. The GOM agenda, after detailed review ofanalytical work andconsultations with key stakeholders, and the pre-programmatic Bank assistance, focuses on five key areas: (i) strengthening innovation and training; (ii) improving the investment climate; (iii) reducing logistics costs; (iv) improving the energy sector; and (v) financial deepening. That selection was based on consultations and survey results with entrepreneurs, thematic and background reports and analysis, and the realities faced by the authorities. The basic principle behind the selection i s to focus on themes and initiatives to improve productivity through the factors o f production, and better usage o f them, particularly through innovation and technological development, through better fimctioning o f factor markets, reduction of logistic costs (estimated at over 20 percent o f product value), and overall improvement of the investment climate (to improve the competitive environment). 4. The Bank's assistance on Competitiveness was structured as a Programmatic Competitiveness and Trade AAA (FY05-08) in three phases: Phase I(FY05); Phase II(support through FY06); and Phase III (FY07-08). The first formal phase of the AAA has supported the design and implementation of the Mexican Competitiveness Agenda through separate pieces o f analytical work at the request o f the Government. It has also included advocacy presentations and advisory assistance through summary and briefings and background information on relevant competitiveness themes for Mexico. Since its inception, this phase was not supposedto be an encompassing analytical piece on competitiveness; rather it was designed to fill some key knowledge gaps that were necessary for the diagnosis and implementation o f the Competitiveness Agenda, and provide a non-lending TA type of assistance to move the Agenda forward. 5. Phase 2 includes additional pieces o f analytical work and consolidates what was learned from Phase 1. It also brings the key policy messages into the public sphere. The main elements o f Phases 1 and 2 are shown in Box 15. The background studies for this report (see www.wor1dbank.ora.m) contain more complete descriptions o f Phase 1 and 2 activities. Initially, a Phase 3 o f the AAA (FY07-08) was planned to deal in depth with specific issues o f interest to the next administration. However, given that the World Bank will be engaged in competitiveness dialogue with the next administration vis-&vis the second and third phaseso f the DPL, the AAA will close with this report. 137The AAA was proposedinthe Country Partnership Strategy o f the World Bank Group with the United Mexican States, Report No. 28141-ME, March 18, 2004. 105 Box 15: M a i n Components of Phase 1and 2 Assistance 1) Innovation, Qualify, and Training:Assessment andrecommendationsregarding quality infrastructureand key weakness o f the Mexican Innovation System (the latter inPhase 2); SMEprograms impact evaluation workshop (inpartnership with WBI) (Phase 1). 2) Investment Climate/Regulatory improvement: Impact o f draft labor law reform onDoingBusiness indicators (Phase 1). 3) Tradefacilitation andLogistics: Study of the logistics cost andbottlenecks of three supply chains and logistics corridors; development of logistics and transport indicators (Phase 1); analysis o fthe impact of tariffreforms on trade (Phases 1 and 2); trade facilitation impact study (Phase 2). 4) Energy: Presentationon key issues and competitivenessimpact (the latter inPhase 2). 5) ICT and TelecommunicationsSector: Challenges and Opportunities (Phase 2) 6) Diagnosis and monitoring: i)IMCO report on competitiveness diagnosis and indicators - baseline and yearly monitoring system (Phase 1); ii)IMCO report on competitivenessat the state level: indicators (Phase 2). 7) Dialogue, Advocacy and Knowledge Briefings with andfor Policymakers andMajor Stakeholders:Contribution to the competitiveness strategy o f two state development plans; presentation to various parts of the Executive, Congress and to key private sector organizations; presentation on the "Structure of the China, Mexico vs. USA trade: Winners and Losers, and Selected Cost Structure" (Phases 1and 2) 8) CompetitivenessChallengesin theAgricultural Sector: (Phase 2) 106 ANNEX 2: PRE-PROGRAMMATIC BANK SUPPORT TO MEXICO BACKGROUND 1. The World Bank worked with the Office o f the President, the main counterpart inMexico, to assist the GOM indefiningMexico's Competitiveness Agenda 2004-2006.'38 This assistance included: A report that summarized the findings from focus groups convened in June/July 2003, which identifiedkey obstacles for competitiveness. The focus groups, spreadover several states and sectors in Mexico, were a rich, highly interactive and organized way of gathering views from entrepreneurs and government officials. From the information thus gleaned, the most significant and common obstacles - such as infrastructure and logistics, regulatory burden, financing, inadequate innovation and quality - were identified for different industries and locations. Background notes and reports as requestedby counterparts, and mostly relatedto specific bottlenecks identified by the focus groups. Topics included logistics, quality and innovation, regulatory reform, human capital, regional development, SME development, financial deepening, and energy reform. The notes contributed to shaping authorities' vision o f competitiveness, identifying major competitiveness constraints and possible solutions, as well as bringing these issues to prominence (the President's Office) and suggestingthe way forward, as requested by the GOM. A report of the trends, determinants, composition and cost structure of Mexican and Chinese exports to the US.This was presented to the Office o f the President, the Ministry o f Finance, and to private sector groups. Key messages from that reports were: a). Mexico i s losing ground in several o f its 15 largest exports (including passenger cars, computer accessories, apparel and household goods); b) labor costs are on average a very small percentage of total production (and distribution) costs; and c) Mexico i s losing export volume even in products for which its proximity to the USA would be expected to give a significant advantage, due to product size or weight (e.g. furniture and household appliances.) Presentations to main stakeholders, such as senior officials from both Congress and the Executive and groups o f private sector representatives. Topics included: a) the results o f the focus group studies, b) the challenges and opportunities relatedto competitiveness; c) the structure and impact o f exports, China vs. Mexico; d) lessons and trends in energy reform; e) on the Doing Business 2005 report and competitive challenges, to the Public Credit and Financing Commission o f the House o f Representatives (Congress) and to the Association o f Private Sector Firms (complete list attached below insection 3). IMPACT 2. The GOM drew on the Bank's support and inputs to develop its Competitiveness Agenda 2004- 2006. Given the breadth o f issues identified by entrepreneurs regarding the policies and behaviors that constrain firms, the experience and impact suggestedfrom the backgroundreports, and the realities faced by the authorities, priority actions were identified by the GOM. These were based on their impact on ~ ~ 13*This was based on Mexico'sNational Development Planh2001-06,which presents three main objectives: (a) Social and Human Development (education, equity, collective initiative, harmony with the environment, confidence inpublic institutions); (b) Growth with Quality (macroeconomic management, competitiveness, inclusive development, regional development, sustainable development); and (c) Order and Respect (sovereignty, national security, democracy, federalism, population dynamics, corruption, public security, andjustice). 107 Mexico's competitiveness, the feasibility of implementation (that they only require executive and federal level approval), the horizon for implementation (within President's Fox's term) and the impact (limited) on public sector finances. The Agenda has five major areas which broadly correspond to those identified by the World Bank team. These are: i)improving the regulatory environment and reforming the investment climate; ii) improving infrastructure and reducing logistics costs; iii) strengthening innovation, education and training; iv) financial deepening; and v) securing energy supply. The agenda does not include key impacting reforms such as structural reform in the energy sector (although it does include issues o f pricing, incentives and reliability of supply) and labor reform, since these were considered too problematic and difficult to enact. Thus, although the prepared agenda is not complete, it provides an excellent starting point to address competitiveness issues inMexico. 3. The GOM set up special tasks forces or commissions by topic and selected "champions" to push forward the reforms on each topic. For example on energy, the GOM set up a high-level Grupo de Politica EnergCtica that meets monthly. On logistics, the GOM set up a Comisi6n de Corredores Logisticos that cuts acrossthe public andprivate sector, to address the reduction o f logistics costs. 4. To monitor and evaluate progress on competitiveness, the GOM decided to develop its own country competitiveness indicators, both at national and at sub-national level states and municipalities to be - - issued annually. 5. The presentation on Challenges and Opportunities on Competitiveness and the Doing Business 2005 report to the Public Credit and Financing Commission o f the House o f Representatives was well received, and as a result, the Senate created a new Competitiveness Commission that will review current legislationto propose reforms inorder to make Mexico more attractive to trade and investment. LISTOFREPORT, PRESENTATIONSANDOTHERDOCUMENTS ReportsandOther Documents TCrminos De Referencia Para la Preparacihy Facilitacibn de Grupos de Enfoque, June 2003. Constraints to Competitiveness - Results Doing Business in2004 - Mexico Mexico Limitaciones a la Competitividad Reduccion de Costos de Logistica, presentation and set o f - Country-Profile, o f Focus Groups Discussions with Entrepreneurs, July 31,2003. January 2004. documents on the topic as input for the formulation o f Mexico's Competitiveness Agenda 2004-2006, January 2004. Mexico Limitaciones a la Competitividad - Prohndizaci6n Financiera, presentation and set o f documents on the topic as input for the formulation ofMexico's Competitiveness Agenda 2004-2006, January 2004. Mexico Limitaciones a la Competitividad Energia, presentation and set o f documents on the topic as - input for the formulation ofMexico's Competitiveness Agenda 2004-2006, January 2004. Mexico Limitaciones a la Competitividad - Innovaci6n, Capacitacih y Educacibn TCcnica y Profesional, presentation and set o f documents on the topic as input for the formulation o f Mexico's Competitiveness Agenda 2004-2006, January 2004. World Bank Comments on Mexico's Competitiveness Agenda 2004-2006, March 2004. Agenda de Competitividad: 2004-2006, Government of Mexico, April 2004. Mexico Competitiveness Monitoring and Evaluation System - Terms o f Reference, May 2004. Mexico Productive Chains Logistics Costs Analysis Estudio de Cadenas Logisticas Elecci6n de las Cadenas Logisticas a Estudiar, Instituto Mexican0 del - - Terms o f Reference, May 2004. Transporte, May 2004. 108 Comments to the Instituto Mexicano para la Competitividad (IMC0)on Proposed Competitiveness Indicators, July 2004. DoingBusiness in2005 -Mexico - CountryProfile, September2004. Presentations Grupos de Anhlisis para la Identificacion de Obsthculos que Dificultan el Crecimiento de 10s Sectores Productivos, Presentation to the Focus Groups, June 2003. Competitiveness and Growth - Mexico in the Global Economy - Challenges and Opportunities. Presentationat the C A N A C I N T M conference, December 2003. Mexico - Restrictions on Competitiveness - Results from the Focus Groups with Entrepreneurs. Presentation to the Private Sector organized by the Consejo Coordinador Empresarial, December 2003. Mexico - Restrictions on Competitiveness - Results from the Focus Groups with Entrepreneurs. Presentation to the Public Sector organized bythe Office o f the President, December 2003. Mexico - Support Program for Competitiveness. Presentation to the Office o f the President and the Secretariat o f Finance, December 2003. Winners and Losers and Priorities to Improve Competitiveness. Presentation to the Office o f the President and the Secretariat o f Finance, March, 2004. Programas Bienales de Mejora Regulatoria2003-2005, COFEMER, July 2004. Competitividad y Politica Comercial, July 2004. Corredores Multimodales, Secretariat o f Transport, July 2004 Grupo de Politica EnergCtica, Secretariat of Energy, July 2004. Modernizacion del Marco Legal del SistemaFinancier0 Mexicano, Secretariat o f Finance, July 2004. ElEstimulo Fiscal a la Innovacih, CONACYT, July 2004. Electricity Issues in Mexico and Reform Experiences from Other Countries. Presentation to the Secretariat o f Finance, September 2003. Priority Energy Issues for Mexico - Some Cases from Other Countries, Presentation to the Energy Policy Group, September2003. FacilitationComercial, SAT, September2004. Reformas a la politica y legislacion de Competencia, CFC, September 2004. La Capacitacion en la Nueva Cultura Laboral, Secretariat of Labor and Social Prevention, September 2004. Methodology for Measuring Competitiveness, IMCO, September 2004. Mexico - Competitiveness and Growth - Challenges and Opportunities. Presentation to the Mexican Congress, September 2004. 109 ANNEX 3: ANALYTICAL UNDERPINNINGS 1. This annex provides a rationale for why the AA4 subject areas are important. The subject areas are the result o f extensive analytical work that the Bank has done for Mexico over the past two years and other complementary studies done outside the Bank. Selected findings in each o f the priority areas are shown inTable 30. The points are discussed in more detail inthe relevant sections o f the main report. This evidence complements the IMCO impact elasticities shown inTable 31. Table 30: Analytical Underpinnings-Selected Evidence Innovation Roughly halfof cross-countrydifferences inper capita income and growth are driven by and differences intotal factor productivity (TFP), generally attributed to technological development Technology and innovative capacity(Hall and Jones, 1999; Dollar and Wolf, 1997). Development Calculations suggest a social returnon R&D of 60% inMexico (Ledermanand Maloney, 2003a). Mexican firms invest a little above 0.1% of GDP inR&D, comparedwith 1.5% inthe OECD area. Quality A study on Germany found that standardscontributed to about 0.9 percentage points out of an average overall growth rate of 3.3% inthe period 1960-1990 (Jungmittag, Blind, and Grupp, 1999). Adopting 100 additional British standardsraisedthe UK's export to import trade ratio by about 14% (Swam, 1996). Education Less than 1% of Mexican students perform at the highest level of international student and assessments (Programmefor International StudentAssessment [PISA] Level 5) ineither Training readingor math, while more than halfperform at Level 1or less inboth (OECD, 2003a). These numbersare quite low given Mexico's income levels. In2000,24% ofMexican adults lackedsecondaryeducation, eventhoughthey hadcompleted primary schooling. Receiving a secondary educationcertificate leadsto higher employability and higher wages (World Bank, 2004e). Lack o f access to financing and skills are the mainweaknessesofMexicansmall and medium enterprises (SMEs).Only 35% of new SMEs survive more than 2 years, and the firms that fail report lack oftraining and expertise as key reasons for their failure (Secretariat of the Economy-Mexico, 2001). [nvestment Modernizing regulatory frameworks could increaseproductivity by as much as 10% in lagging Climate: sectors (Cotis, 2005). Regulation Starting a businesstakes 58 days inMexico, comparedwith 5 days in the United States. Dealing with licenses takes 222 days at a cost of 159% of income per capita, while it only takes 70 days inthe United States at a cost of 17% of per capita income (World Bank, Doing Business database). The cost of the regulatory burden inMexico is 15% of GDP (IMCO, 2005a). Competition . A study of the impact of pro-competitive regulatory reform in several industries in the United Statesfound that annualwelfare gains inaffected sectors were more than 7%, with 90% of the benefitsflowing to consumers (OECD). vIncreasingcompetitive pressure can increase the probability of firminnovation by more than 50% (World Bank, 2004a). BAccording to a survey of Mexican firms, the two largestobstaclesto business developmentare government monopolies and private monopolies (Centro de Estudios Econ6micos del Sector Privado [CEESP], 2005). hade and Trade facilitation reforms inMexico could increaseexportsby $29.3 billion, equivalent to 21 Frade percent of total Mexicanmanufacturing exports (Soloaga, Wilson, and Mejia, 2006). Facilitation 1Mexico's global exports would have beenabout 25% lower without NAFTA, and FDIwould havebeenabout 40% less. Also, the amount o f time requiredfor Mexican manufacturersto adopt U.S. technological innovations was cut inhalfwith NAFTA (Lederman, Maloney, and 110 Serven, 2003). Despite the gains from NAFTA, Mexico has beenlosing export market share within the United States inrecent years. Ofthe top ten types of Mexican imports inthe United States, only two- petroleum and total agricultural products-gained market share between2001 and 2005 (U.S. Dept. of Commerce, see Table 33). Export-supply capabilities inMexico were more of a constraining factor for Mexicanexports than import-demandconditions inthe U.S.(Hansonand Robertson, 2005). Logistics Logistics costs inMexico are estimatedto be 20%, while the OECD average is 9%. A reduction to 12% inMexico would have a significant impact on economic activity andjob creation(Guasch, 2004; Guasch and Kogan, 2003). Mexico is not capitalizing on its enormouscompetitive advantage ofproximity to the United Statesbecauselogistic costs are so high: as an example, Mexico spends almost double what the United States does on transportas a percentage of GDP (IMCO, 2005a basedon INEGI). Energy Inadequateinvestmentsinenergy infrastructure have hada significant impact on TFP growth in Mexican manufacturing(CasteiiedaSabido, 2003). The prices of fuels and electricity inMexico are amongthe highestof the world. Incontrast with other OECD and developedcountries, the prices of electricity, natural gas, fuel oil and other fuels have increasedover the last decade. (World Gas Intelligence, EIA, Edison Electric Institute, Eurelectric, among others). A largeportion (64 percent) of electricity subsidiesare capturedby the non-poor inMexico (World Bank, 20040. The quality of the electricity service is poor comparedto other countries. Serviceinterruptions and voltage fluctuations affect the overall productivity levels of industries andprevent the installation of modem electronic equipment (World Bank, 2005d; IMCO, 2005a; and others). ICT and Countrieswith higher levels of investmentinICTs experiencehigher economic and social Telecom development growth (OECD, 2004~). Mexico's level of ICT expenditure(3.1%) as a share of GDP is substantially lower than Latin American countriessuch as Chile (6.7%), Brazil's (6.7%), Argentina (5.7%), and Peru (6.9%) (World Bank, World Development Indicators, 2003). Businesstelephonecharges(factoring ininstallation costs, monthly fees, and per minuterates) inMexico are over 3 times greaterthan chargesinArgentina and 4 times greaterthan charges Agriculture - Agricultural GDP grew only I.3% between 1984 and 2004, while total GDP grew 3%. Agriculture as a share of the total Mexican economy shrank from over 8% in pre-NAFTA era to 5% in2004. Public expenditureinagriculture declined from 13.4% to 6.9% between 1990and 2003. Out of the top 24 export agricultural commodities from Mexico to the United States, only 9 are increasingmarket share. The rest are losing market share to competitors. Negative net trade balance(exports minus imports) for agriculture deepened from $211 million in 1991/93to $1.3 billion in2001103. Sub-national Aguascalientesranks 33rd globally in the costs to start a business, while the Estado de Mexico ranks 127* (World Bank, 2006b). The impact on investment per worker of a 10% improvement in competitiveness variables ranges from 841% to 12%, dependingon the Mexican state (IMCO, 2006a). 2. The AAA-supported themes are roughly in line with the GOM's Competitiveness Agenda. After the Fox administration declared increasing country competitiveness to be a key priority in2003, the Bank helped the GOM craft a Competitiveness Agenda. The agenda was completed in2004 and drew on numerous studies, surveys, and consultations in Mexico, combined with broader regional analyses. Ultimately, actions included in the Agenda were based on their impact o n Mexico's competitiveness, the feasibility o f implementation, the horizon for implementation, and the impact o n public sector finances. With all that taken into consideration, the GOM's Competitiveness Agenda converged to the following 111 subject areas: (i)improvingthe investment climate, (ii) strengthening innovation, education, andtraining, (iii)reducinglogisticscostsandimprovingtradefacilitation, (iv)improvingtheenergysector,and(v) financial deepening.While the agenda does not include key high-impact and politically sensitive reforms such as structural reform in the energy sector (although it did include issues o f pricing, incentives, and reliability o f supply), tax reform, and labor reform, it provides an excellent starting point to address competitiveness issues inMexico. It also has servedas a vehicle for dialogue with the private sector. 3. Work on the Competitiveness Agenda led to the ongoing Programmatic AAA (FY05-08). Analyses within the AAA and preparation o f the First Programmatic Competitiveness Development Policy loan (July 2005 - February 2006) have reaffirmed that the Competitiveness Agenda subject areas should indeed be priorities. As a result, the AAA-supported themes are roughly in line with the Mexico Competitiveness Agenda. The following discussion considers alternative ways o f identifyingobstacles to competitiveness, as well as a more detailed examination broken down by factor markets and total factor productivity (TFP) . Growth Diagnostic 4. One way to analyze the obstacles to competitiveness inMexico, and hence the appropriateness of the areas this AAA (and the Mexican agenda) supports, is through a "growth diagnostic" conceptual framework. This framework, developed by Hausmann, Rodrik, and Velasco (2004), aims to identify the bindingconstraints to growth. While the diagnostic has not been comprehensively applied to Mexico, it does provide guidelines. The framework is based on the drivers o f investment and uses a decision tree to first ask whether growth i s constrained by low returns on investment, the high cost o f finance (both local and international), or both. The cost o f finance (reflected particularly invery limited access) is a key issue inMexico, considering that the share o f credit to the private sector as a percentage of GDP is about 17 percent, while it is 63 percent inChile, 98 percent inKorea, and over 100percent in many East Asian countries. This issue i s addressed separately through a Development Policy Loan (DPL) on the financial sector. 5. Turningto the returns oncapital, an analysis byIMCO shows that low returns are a largepart of the story. Based on an analysis o f the 33 largest companies on the Mexican stock market (BMV), IMCO found that, on average, profits did not cover the cost o f capital (Figure 71). Specifically, only 6 o f 14 sectors had returns that exceeded their cost o f capital. Moreover even in absolute terms, the returns to capital appear low. They are inthe 5-12 percent range, which i s similar to the yield on government bonds in Mexico. The growth diagnosis framework provides guidance into the critical issues that are keeping those returns so low. And this AAA supports measuresthat ought to improve the returns to capital. 112 Figure 71: Cost of Capitaland Returnson CapitalinMexico Retorno s/Capltal 1.01.WACC 15% -I --- RetornoPromedio --- WACC Promedio Note: WACC stands for Weighted Average Cost of Capital. Promedio means average. Source: IMCO with data from Bloomberg, 12/17/2004.33 companiesthat compose the main Mexican stock market (BolsaMexicana de Valores)index (IPC). 6. The growth diagnostic framework goes on to identify a number of possible drivers to low private returns on domestic investment:a) highmicro or macro risks; b) lack of innovationand R&D; c) large externalities, spillovers, or coordination failures; d) hightax rates and/or inefficient tax structure; and e) insufficient infrastructure, high transport, energy, telecommunications or shipping costs (see Annex 4 for a diagram of the framework). The macro environment hasbeenstable inMexico for the last decade, making it unlikely to be a binding constraint. Micro risks, such as from corruption and crime, could potentially be binding constraints. Indeed, they came up in competitiveness focus groups held in Mexico and the World Economic Forum's executive opinion survey. However, they are not covered in this AAA since there is another Bank analysis ofthis subject.I3' Other micro risks, suchas property rights andjudicial reform (contractual enforcement, are handledthrough two other Bank operations-a housing DPL and a judicial reform project. Lack o f R&D / innovation is certainly an issue in Mexico and i s addressedby this AAA. Tax rates and regulations are seen as a constraint with significant impact (Figure 72), but the issue has also been looked into in another Bank analysis.'40 Constraints associated with infrastructure and transport and shipping costs are addressed in the logistics, trade facilitation, and energy components o f this AAA. ' 3 9The Mexico Institutional and GovernanceReview. I 4 OThe Mexico CEM: Challengesand Prospectsfor Tax Reform. 113 Figure 72: Tax Burden Argentina China Brazil Mexico USA Thailand Korea Chile 2 3 4 5 Note: Scores are a response to the survey question: "The overall tax burden on your enterprise, including all associated costs (tax rates plus administrative and time costs, penalties, etc.), i s estimated as (inpercent o f net revenues) (1=0-4 percent, 2=5-15 percent, 3=16-25 percent, 7=81- 100percent)." Source: World Economic Forum (2005). Impact Analyses 7. A complement to the growth diagnostic framework is an impact study conducted by the MexicanCompetitivenessInstitute (IMCO). The IMCO (2005a) study, which was part of Phase 1 of this AAA, develops a model to assess Mexico's competitiveness and establishes priorities among themes in terms of impact on investment (gross capital formation) per worker. From that model, "point elasticities" are estimated, giving the impact on investment per worker of a ten percent improvement in the respective independent variables (Table 31). As shown, the high-impact themes are the competition environment, taxes and tax regulations, regulatory and investment climate, education and innovation, trade facilitation, transport and logistics, corruption, finance and energy. These impact estimates provide guidelines for priorities to be tackled inorder to increase Mexico's competitiveness. Inconsequence, this AAA (and the MexicanCompetitiveness Agenda) analyzespolicies inmost of those themes. 114 I Subject area3 variable$ behind each subject area 8. It should hc mtcd tho1 inipact KlttKilberS in Table ?Iore for siiigte effects or isolated ~ ~ p r o ~ ~in n ~ ~cat ~ s u cach n of reform across rnost of tfte hig~-impac~a r u ~ ~ ) r ~ ~ ~ c could have a positive b ~ ~effect.n dhiotkcr~words,~the j n ~ e ~ ~betweenthe ~ ~ ~ e t i ~ ~ ~ ~ variables j v ~ ~ w ~ should create an overall impact that is s w ~ s t a n ~ ilarger than the sum of the i ~ d ~ ~ ~ ~ ~ a l ~ l l ~ ~ r n ~ ~ ~ ~ ? e ~ i e ~ ~ ~ . 9~ IMCCYs variables come from a wide variety of sources. There is an average of' ovcr 6 u ~ ~ d c variables~ per~subject area. ~ x a ~ ~ iinclude: r i ~ i ~ p ~ c s 115 10. The conclusions of the impact elasticity calculations are supported by two other relatively comprehensive reports: a Competitiveness Focus Groups report and the World Economic Forum's (WEF) Global Competitiveness Report 2005-2006. The focus group report is based on the written and verbal responses of over 90 businesspeople in Mexico from a range of economic sectors in 2003. The WEF report contains information about the most problematic factors for doing businessinMexico.These sources are "soft" in the sense that they are based mostly on subjective evaluations o f entrepreneurs, while the IMCO analysis i s "hard" in the sense that, by and large, it uses hard factual data and provides quantitative estimates o f impact. It is reassuring that the correlation between the hard and soft studies is quite high(over 80 percent). Factors Markets and TotalFactor Productivity 11. A complementary framework to the growth diagnostic to elicit key elements and priorities for competitiveness/productivityis to look at the factors of production, their markets, and how well the factors are used. National output is determined by the quality and quantity o f inputs-labor (human capital), capital, and infrastructure-as well as a residual known as total factor productivity (TFP). TFP, which can also be thought o f as know-how, is determined by some combination of innovatiodknowledge and investment climate characteristics. Well-functioning factors markets and the efficient use o f factors o f productions are key for competitiveness and increasing productivity and growth. A brief summary o f those themes i s provided in the following sections, further bolstering the analytical underpinnings for the AAA agenda. Detailed rationalizations for specific policies within each subject area discussed in the Policy Areas section. Labor 12. Labor regulations are relatively rigid. Labor regulations were identified as a constraint in the WEF executive opinion survey. Doing Business in 2006 gives Mexico a 60 out of 100 (higher numbers mean more rigid regulation) on both its "Rigidity o f hours index" and its "Difficulty o f firing index." In contrast, Chile scored a 20 on each of these indices. As part of Phase 1 o f this AAA, the World Bank estimated the impact o f a proposed labor law reform in 2004. The analysis found that the proposal to relax the regulation o f working hours alone would cause the "Rigidity o f hours index" to drop from 60 to 40. Partly as a result of strict labor regulation, Mexico has a large informal sector, and competition from the informal sector was one o f the complaints of entrepreneurs in the focus groups. IMCO has also highlighted a number o f labor market shortcomings. 13. Despite these shortcomings, labor does not emerge as one of the top constraints. This i s corroborated by the IMCO quantitative estimates (Table 31) and other studies. An IDB (2004) study shows that the Mexican labor market performs well relative to other labor markets in the region, with little adverse impact on unemployment and mobility. Figure 73 shows that Mexican's consider unemployment, low wages, and job instability to be less o f a problem than every other country in the Latin America sample. The same IDB report also shows that Mexico has the best "quality o f labor relations" and "overall efficiency in labor market performance" inthe region (Figure 74). Partly because labor is not a top constraint and partly because the subject o f labor reform is extremely politically sensitive, major structural reforms inlabor markets are not covered by this AAA. 116 Figure 73: The Most ImportantProblem,by Figure74: SummaryMeasure of OverallEfficiencyin Country (Percentageof respondents) Labor Market Performance(Index, 0-1) M&O GwbRIOJ Nlmmgm Honduln EC& Pwu d l o GurtpnrdO Venerueb Panam blffKI ArE1P"hm &aol U r W Y 00 01 0 2 03 04 0.5 06 07 0 8 Source: IDBcalculations based on householdsurvey data; Latinobarometer(1997); World Economic Forum(2001). 14. Labor productivity is impacted by the quality of human capital, which is largely determined by education and training. This AAA concentrates on adult education and training (e.g., SME support) programs, while traditional education issues (primaryhecondaryltertiary) are addressed through a series o f separate Bank projects. The WEF survey named an "inadequately educated workforce" as one o f the most problematic factors for doing business and the average education attainment of the Mexican adult population i s very low-7.6 years-compared with 11-12 years for other OECD countries. IMCO also considers the education shortfall to be a factor that stunts Mexico's competitiveness, with the prospects for the poor beingparticularly dismal: the poorest 10 percent o f the population averages only 4 years o f formal education. Ideally, SME support programs can help overcome education shortfalls. However, a 2004 World Bank study14' (associated with this AAA) highlighted several shortcomings in existing Mexican SMEprograms, includinga pressing need for impact evaluations. Capital 15. Credit and capital markets are not very effective in Mexico. A telling indicator is the level o f domestic credit to the private sector, which was a paltry 17 percent o f GDP in 2004. In comparison, values in many East Asian countries are over 100 percent. Not surprisingly,all the surveys highlight that issue. Both the focus groups and the WEF executive survey identified it as the most problematic factor. The IMCO study shows that while relatively small improvements have an impact, the big payoff comes with significant improvements in that sector. Given the importance and complexity o f this topic, it i s being handled as a separate Programmatic Finance and Growth DPL, which was presented to the Board inMarch2006. Infrastructure 16. There is ample evidence of the strong impact of infrastructure on economic growth and poverty. Yet, Mexico lags behind in that area relative to its major partners and competitors. Calderon and Serven (2004) have shown that if Mexico were to improve its infrastructure stock and quality to the levels o f Costa Rica (the region's infrastructure leader), Mexico's GDP would increase by 1.7 percentage points and inequality would be significantly reduced. Thus, not surprisingly, various components o f infrastructure have been identified as highly problematic by several sources. The focus groups considered distribution, transport costs and logistics to be an important competitiveness 14'World Bank (2004b). 117 c o ~ ~ s ~ rTheyc~3~sidered a ~ ~ ~ . rad d s to be limited and co roads to be sluw, exp and airports and ports to lack capacity for high iiiad vctlumcs. A 2005 study by the U.S. ~ i i ~ e ~ a ~ ~ ( ~ n a ~ 'I'rade ~ o ~ ~ ~ confimis s s ~ othat Mexico falls short oft a tiumber of indicators o f fogistics quality, n e s p e c ~ a ~ivhen coniparcdwith Chile and countrics in East Asia, Their evidence comes from a suwcy of ~ y 51 c n ~ ~ which ~ ~ rdemcd~ scores for cusmns, airports, seaports, r ~ e lation, burden of security p ~ o c e ~ i ~andsavailability of C ~ I ~ i p ~ e I ~resources.rWith the exception o f arporru, Mexico f'ails in r e ~ ~ e ~ ~ a y the bottom35 percent afthe countries surveyed on ail of these measures (Tabfc 32). 17 A scparaxe piece of empirical work has shown the i ~ ~ ~ ~ofr ~~n far a~s c~ ~e~inr ufranspofl costs r c , an analysis of bii 31 tradc data showed that B d e ~ e ~ ~ of r a ~ ~ ~ ~ o to rhe ?SIh percentile raises f r a ~ s costs~by 12 p e r c ~ ~ points eand ~ o ~ a ~ reduces traded voiunres by 28 percent 14' 18. Mexico's high logistics and transport costs arc hurting ~ ~ ~ ~ ~ ~ ~'i'hc iMexiconfocuss s , ~ v e ~ groups ~ ~ j ~ ~ i l the~ absence of i n ~ ~ ~ facilitiesa ("dry ports"') that would improve logistics ~ h r e d ~ n d ~ ~ ~ a ~ ~ andgreduce h~e ~cost of' t~ r ~ n s ~goods, ~ e ~ c n o ~ i~~~~p c ~exports. ~~ ~~ s iaws were~seen as i a l ~ n s very c ~ ~ ~and~~~~~~~1~I toi comply trlith. ~ ~ c ~ ~ The~data suppcsrts this assessment: logistics costs in r 20 pcrcent of product value, compared with an average of 9 p eduction nf those Iogistie costs to if Icvclof twelve perc t on economic activity and job creation, as shown elsewhere. E ~ ~ ~ r~a~taen ~s ~costs o r r so ledto high~ n v ~ levcis for ~businesses: duringthe 1990s, Mexico had 90 percent n ~ o higher raw iiiatcrials ~ n ~ e levels oon~ aver n ~ than thc U.S.id41MCO finds that Mexico i s not capitalizing on its enormous competitive advantage o f proximity to the U.S.because transport costs are so high.Indeed, Mexico spends almost double what the U.S. does on transport as apercentage of GDP.I4' 19. Improvements in the energy sector are needed, but reforms remain controversial. Over the last decade, different groups have called for the implementation o f far reaching reforms in the energy sector as a necessarycondition to attract sufficient private financing, improve efficiency, lower costs, and expand the sector's contribution to the federal budget and the economy. While a national consensus has not been reached with regard to the nature and extent o f energy sector reforms, the GOM has gradually taken conservative steps to allow limited private sector participation and improve transparency and regulatory accounting. 20. High energy costs and quality of service issues hurt productivity. Energy costs comprise 10.5 percent o f the total production costs o f manufacturing industry products (as an aggregate). This i s comparable to labor costs which are inthe range o f 4-20 percent. Highenergy prices were identified as a constraint (although not a top one) by the focus groups and unit electricity costs inMexico in2002 were higher than every other country in IMCO's list of 44 comparator countries. Moreover, while in most countries energy prices have dropped in the last five years, they have increased in Mexico. It should be noted, that while most o f the discussion in Mexico (as in the China debate) focused on labor costs, for many products inmanufacturing, energy costs as a share o f total costs are larger than labor costs. Quality o f service i s particularly poor for SMEs, which often experience variations in frequency and voltage and interruptions in service. High energy prices and poor quality then translate into reduced SME competitiveness. Large enterprises have fewer quality problems because they are generally served by more reliable, high-tensionwires. The AAA's approach inthe electricity sector is informed by the World Bank's Infrastructure Public ExpenditureReview for Mexico, which was completed in September 2005, 21. A summary of the primary issues inthe energy sector is as follows: 0 Price. The prices o f fuels and electricity in Mexico Figure 75: Electricity Prices (USDe / kWh) are among the highest o f the world. Incontrast with other OECD and developed countries, the prices o f g \ electricity, natural gas, fuel oil and other fuels have 8 increased over the last decade. (Figure 75) Steady price increases have had serious effects on the costs 7 o f production across different industries. In the manufacturing sector, the average costs o f energy as 6 a percentage o f total production costs increased USA annually by 2.25 percentage points over the period 5 1995-2002. 4 0 Subsidies. A large portion (64 percent) o f energy subsidies are captured by the non-poor inMexico.'46 hile Those funds would be better used to provide for Y urgent social needs. 98 99 00 01 02 03 04 Source:Consejo de Coordinadores 0 Quality. The quality o f the electricity service i s poor Empresarial(2005) based on national sources compared with other countries. Service interruptions and voltage fluctuations affect the overall productivity levels o f industries and prevent the installation o f modern electronic equipment. Electricity transmission and distribution also needsto be improved considerably to reduce losses. `45The original source is INEGI, BTS as reported inIMCO (2005a). 146World Bank (20040. 119 To inect projected demand, large ~n~estn~einnnew capacity are needed. ~ s ~~~~~~~~ 22. 'I'herc are also issues in tclerom sector j ~ p a c t i n ~ ~ o ~ Mexico's ~ ~ t j ~ ~ ~ c o n ~ m u ~ ~ ~ c sectorogrew conside a ~ ~ n s y as a rcsuil ofthc ~ r i v a ~ ~ ~o faTclmcx~inthe early 1990s ~ i o i ~ ~ a i thato began in 1995. Thc Fox ~ n s~rst~~on's efforts to ~ r a n s f o ~ i m y , mainly through its e-h.lextco in ,havealso resultedinnotable ~ ~ Iiowever, despite having been~tho estrly reformer in LAC faficr Chile), ~ o ~ i countries and many Latin ~ ~ ~ ecountries in tenns~ o f wide-spread r i c ~ ~ ad inccrnet ~ ~ ~ r a s ~ ~~ce~~uer ce .o ~ ~ ~ m ~ n are~nxuchihrghersin prices ~ ~ t ~ n ics. 'This Izolds true far business and r e s ~ d ~ telephom i ~ ~ ) ~ ~ ~ l ~ ~ i a ~ ~ ~ i b sfees, rre s~i d ~~~t~~~~ll e~ pco~~ ~ ~tin~charges, aridthe cost of a three minute lttcal call. The ~ e t ~ ~ e installation arid ~ i ~ ~service ciiargcs~ l y sinesscs to lease a local line (known as a local loop) III ) ~ ~ ~ blcxiccs City arc also highconiparcd with ajor cities in Latin Amenca. ~ ~ w ~ ~ ~ ~ 23 Mexico has seen ~ u L ~ ~gains~ from~ trade, ~but it i s losirtg export market share in the U.S. t ~ t ~ 1 Exports to the U.S., ~ a r t i c ~ ~ ~ NAF'A,have been an ~ ~ r l ~ i pitlar of~the Mexican economy, ~ ~ ~ a 1suggests that Mexico's global expons would hale been about 25 percent lower without and PUf would have been about 40 pcrcent less. Atso, the amount of' r i m rcyuised for hlexicaii ~nanu~ac~~~rers U,S, ~ e ~ ~ r~ i ~~ ~~ ~~was~ cut ina half ~a 1oNMTA (see Box 16 for to adopt ~ ~~ ~ ~t with ~ ~ s Lessons iiom NtWTX).Id' Despite the gains from NAETR, c) has been losing expon market share within the US. in recent years. Of the top ten types of Mexi ports inthe U.S.?only ~ ~ ~ ~ ~ c ~ gained market share belwccn1001 and 2005 (Table 33). The largest fXls e~ujp~en~.I ~ p ~ ~equipment, motor vehicles, and c o e r t gainer 111all bur one ofthe ~ a ~ e ~ owheresMexico rcgrst r ~ e 1 abir 33: \Ic\iw's Principal I: iyorts to I Mexican Share of C'S Imports (YO) I I Percentage entering maquiladoras also seem to improve their performance in the years after entry. The study also found that export experience is a significant determinant o f productivity growth: each year o f exporting experience is associated with a 4.5 percent rise inTFP.I4* 25. Trade facilitation efforts stand to be improved. Despitethe changes wrought byNAFTA and the benefits o f exporting, the U.S.market is not always easily accessible, especially for smaller firms and those far from the border, and costs to comply with export procedures can be high.Doing Business in 2006 reports that "time for export" is 18 days inMexico, double the number o f days in the U.S.Focus groups also identified the lack o f mechanisms to facilitate exports as a major constraint. Box 16: The Lessons from NAFTA A recentWorld Bankstudy, "Lessons from NAFTA for LatinAmerica and the Caribbean" (Lederman, Maloney, and Serven (2003)) highlights the following lessons: NAFTA can be credited for boosting Mexico's growth in the post 1994 period. By inducing a reciprocal reduction in trade barriers and locking-in the reforms o f recent years, it has yielded a positive impact on trade flows, foreign direct investments, and growth o f industrialproductivity. It is also responsible for the creation o f many newjobs and some reduction inpoverty rates o f recent years inMexico. Despite the positive growth effect, NAFTA has not sufficed to guarantee income convergence among North American countries. This reflects mainly pending items in Mexico's policy agenda including institutional gaps (i.e., corruption and rule o f law), deficiencies in education (both coverage and quality), a passive innovation policy, the lack o f critical infrastructure (especially in lagging regions) and some weaknesses inmacroeconomic policy. The experience has demonstrated that while positive, an FTA with a developed partner is not sufficient to increase growth on a sustainable basis unless an agenda o f complementary reforms is pursued. NAFTA benefits have been concentrated in states in the North and Center o f Mexico. Southern states have not seen much o f an impact, due to key deficiencies ininstitutions, education and infrastructure. Contrary to some predictions, NAFTA has not had a devastating effect on Mexico's agriculture. In fact, both domestic production and agricultural trade rose during the NAFTA years. The negative effects did not materialize because aggregate demand expanded rapidly in both Mexico and the U.S. in the latter half o f the 1990s, some segments o f Mexican agriculture recorded substantial increases in productivity (esp. irrigated lands), and outdated subsidies were transformed into targeted efficient programs (e.g., Procampo) which delinked transfers to farmers from current and future production levels. Mexico's experience also shows that there are some key remaining barriers to trade that have limited the growth potential o f certain sectors. Inparticular, restrictive rules o f origin affecting clothing exports to the U.S. market seem to have limited the ability o f Mexicanexporters to take full advantage o fNAFTA preferences. Source: World Bank (2005b). TotalFactor Productivity: Innovation and the Investment Climate 26. Total Factor Productivity (TFP) growth-a critical determinant of overall growth-has been limited in Mexico. Over 1960-2000 the bulk of the differences in growth between countries (45-90 percent) was accounted for by TFP-the productivity contributions above those made by physical and human ~apita1.I~' A recent study on Mexico supported these results, finding that swings ingrowth have been mostly accounted for by changes inthe contribution o f TFP, as opposed to factor inputs.'s0The bad 14*World Bank (2000). 14'World Bank (2004a). See also Lederman and Maloney (2003a), Hall and Jones (1999), Dollar and Wolf (1997). Faal (2005). 121 news is that the contribution of TFP to GDP growth was negative on average in Mexico between 1980 and 2003, although there has been an improvement since 1996 (see Figure8). A separate study found that national learning capacity, as proxied by TFP, has accounted for the vast majority o f the difference in output between Korea and Mexico since 1960 (Figure 76). Korea developed its national capacity to absorb knowledge inthe 1970sand 1980s, whereas Mexico didnot.'51 Figure 76: GDPper Capita Growth: Korea vs. Mexico Rep of Korea 1 Due to TFP gioMh or knowledge aCCUmUl8tlOnin ID 8 Korea 6 - 4 - Due to labor and 2 - capital in Korea 0 4 1 1% 197U 1975 1980 1985 19Eo l B 5 2ooo Source: Knowledge for DevelopmentProgram, WBI, as cited inWorld Bank (2005~) 27. Early growth research emphasized technological progress in explaining TFP, suggesting that differences in growth rates were driven by differences in the technologies adopted. The dramatic acceleration in income levels among the fast growing countries over the last 200 years can be understood by improvements in technology. "Technology" in this sense, however, is not limited to scientific breakthroughs o f the kind that might merit a patent. It can also include more modest advances, as well as new and better ways to organize productionprocesses, interact with consumers, or distribute goods. 28. Recent research has emphasized that TFP can also be understood to encompass more than just differences in technology. The broader environment in which firms operate-in other words, the investment climate-matters too. A better investment climate can directly improve productivity by reducing unjustifiedcosts and risks flowing from government policies and behaviors. By making it more attractive to develop and adopt new and better ways of doing things, a better investment climate will help productivitythrough its impact on technology as well.152 29. Labor productivity levels and growth are quite low by OECD standards. A separate way to measure productivity i s through labor productivity, or GDP per hour worked. Mexico has the second lowest GDP per hour worked o f all OECD countries, which equates to a -70 percentage gap with respect to US. levels (Figure 77). Growth in labor productivity has also been low. GDP per hour worked grew only 1.3 percent annually between 1995 and 2004 inMexico, making it the fourth worst performer inthe OECD (Figure 78). "* World Bank (2005~). World Bank (2004a). 122 Figure 77: Productivity Levels-Percentage Gap with Respect to US GDP per Hour Worked, 2004 20 0 - 20 - 40 - 60 - 8 0 ' 1 Figure 78: Growth in GDP per Hour Worked, 1995-2004 (percentage change at annual rate) I Source: OECD Science, Technology, and Industry Scoreboard, 2005. 30. Labor productivity growth varies dramatically by sector. A separate analysis breaks labor productivity growth down by sector and finds that the mining and utilities and transport, storage, and communication sectors were the only ones that showed annual labor productivity growth o f more than 1 percent over the period 1990-2004. Meanwhile, wholesale, retail, trade, restaurants, and hotels, construction, and other services all had negative productivity growth, yielding negative total growth duringthe period 1990-2004 (Figure 79).153 Note: these numbers may be incompatiblewith those inFigure78. 123 Figure 79: Annual Labor Productivity Growth inMexico, 1990-2004 6% 5% 4% 3% 2% 1% 0% -1% -2% -3%J I Source: World Bank calculations based on XI Censo General de Poblaci6ny Vivienda, 1990, INEGI; and EncuestaNacional de Empleo, 2004, INEGI. 31. Based on the above productivity evidence, this AAA covers measures to encourage innovation and technology adoption and improve the investment climate through better regulation and competition. Innovation and Technology Development 32. Innovation is critical and Mexico could play a more active role in fostering it. According to a recent World Bank flagship report, "technological innovation is a key source o f economic, income, and employment growth. This premise has been validated by theory and empiricism and by now has been accepted by most co~ntries."'~~Another World Bank flagship finds that what distinguished the successful resource-abundant countries (Australia, Canada, and the Scandinavian countries) from the disappointing LAC experience was precisely their ability to learn from abroad-their national innovative capacity.'s5 Current research has classified the Mexican national science and technology (S&T) learning model as one o f "passive FDI-dependent learning." This means that the country does not undertake significant efforts to facilitate local absorption and dissemination o f foreign S&T knowledge, e.g. byupgrading local workers' capabilities or by targeting the kind of FDI most relevant to the country's long-term technological learning interests. This contrasts with "active FDI-dependent learning" countries, such as Ireland and Singapore, that have a proven ability to actively facilitate and promote technological learning from FDI.Is6Finally, entrepreneurs in the focus groups noted the lack of government support for R&D. This AAA is also complemented by the Innovation for Competitiveness loan to Mexico (approved in May 2005). DeFerranti and others (2003). Closing the Gap in Education and Technology. 15'De Ferrantiand others (2002). Natural Resources to theKnowledge Economy. 156Soubbotina (2005). 124 Regulation 33. Regulation deficiencies were highlighted as a key issue in several sources. The focus groups complained about the highnumberso f norms andregulations, as well as frequent changesto them, which create uncertainty and provide opportunities for corruption. The WEF survey found that inefficient government bureaucracy was one o f most problematic factors for doing business in Mexico. Doing Business in 2006 found that starting a business takes 58 days in Mexico, compared with 5 days in the U.S. Dealing with licenses takes 222 days at a cost o f 159 percent o f income per capita, while it only takes 70 days in the US.at a cost o f 17 percent o f per capita income. IMCO also identifies regulation as a problem, ranking Mexico 36 out o f 45 countries in terms o f ease o f opening a business. According to COFEMER (Comisidn Federalpara la Mejora Regulatoria), the cost o f regulation inMexico amounts to about 15 percent o f GDP. COFEMER calculates that the benefits o f potential deregulation could reduce this cost to 5 percent o f GDP. 34. A final piece o f evidence regarding regulation comes from a study of the contribution to economic growth o f changes in different structural policies. The study analyzed the contribution to the projected change in Mexico's growth rate-during the period 2000-2010 versus the period 1991-1999-f an improvement in the government burden to the 75`h percentile o f Latin America and the Caribbean. Specifically, the study found that reducing the government burden inMexico would add 0.15 percentage points to per capita output growth during2000-2010 with respect to the 1990~.'~' Competition 35. Improvingcompetition yields a variety of benefits. Barriers to competition benefit some firms but deny opportunities and increase costs to other firms and to consumers. They also weaken incentives for protected firms to innovate and improve their productivity. Increasing competitive pressure can increase the probability o f firm innovation by more than 50 percent.15' Therefore, reforms to enhance competition could have significant effects on growth. A study o f the impact o fpro-competitive regulatory reform in several industries in the United States found that annual welfare gains in the part o f GDP affected by reform were more than 7 percent, with 90 percent o f the benefits flowing to consumers.'s9 Increasing competitive pressure can increase the probability o f firm innovation by more than 50 percent. According to IMCO's research, a 10 percent improvement in the variables that compose the "competition" dimension could increase investment per worker inMexico by 7.5 percent (Table 31). 36. Competition in Mexico is less than optimal. Deficiencies in statutory authority and judicial review processes constrain the Federal Competition Commission's (CFC) ability to address anticompetitive conditions effectively and efficiently. The CFC's decisions and proceedings are open to amparo actions ("protection appeals")160from dissatisfiedparties, which can take many years to resolve. Even if the CFC's orders survive the "amparos," the imposed fines are not collected by the CFC but rather by the fiscal department o f the municipality o f residency o f the fined party. From 1993 to September this year, only 3.3 percent o f the total amount fined by the CFC has been collected, while 72 percent i s under reconsideration and 20 percent i s under amparo and fiscal nullityaction. 37. According to IMCO's research on the impact o f 10 dimensions on investment for 45 economies, Mexico ranks 33 interms o f level o f competition and formality o f the markets. Furthermore, accordingto 157Loayza, Fajnzylber, and Calder6n (2005). Is*World Bank (2004a). 159OECD, RegulatoryPolicies in OECD Countries: From Interventionism to Regulatory Governance. I6O An amparo is a remedy against acts by any authority that violate any o f the individual guarantees recognized by the Mexican Constitution. It is available inall legal matters and may be invoked incriminal, civil and administrative trials. 125 a recent governance survey, the two largest market obstacles to business development identified by firms are government monopolies and private monopolies. The Global Competition Review 2003 ranking of competition regimes in25 countries ranks Mexico near the bottom with a score o f 2.25, ina rating from 0-5, only above Greece (2.0) and Argentina (1.5) since Mexico's competition regime is considered to be severely undercut by courts. 126 - ,- I b I 2 ea Y 5 s8 k Q b t; r 4 ANNEX 5: THE MEXICANENERGYSECTOR I.OVERVIEWANDCHALLENGESOFENERGY SECTORINMEXICO Hydrocarbons Sector 1. The energy sector in Mexico has been essential to the economic strategy o f the country since the nationalization o f foreign oil companies and the creation o f PEMEX in 1938. After the international oil crisis in 1973-1974, the government decided to increase investments in oil exploration and production. With the discovery of new oil fields duringthe early 198Os, Mexico shifted from being an oil importer to becoming a net oil exporter. In2004, Mexico ranked as the world's fifth-largest oil producer (including crude, lease condensate, natural gas liquids, and refinery gain), behind Saudi Arabia, Russia, the US and Iran.l6IToday Mexico is a major non-OPEC oil producer and has one o f the largest oil utilities o f the world (PEMEX). 2. Despitethe success and significant oil revenues, PEMEX i s the world's most indebtedoil company, with a net debt o f about 32 billion USD (2003). Historically, the Federal government has relied on PEMEX for about one third o f its budget, with PEMEX and its subsidiaries turning over an estimated 60% o f their annual revenues to Treasury (Hacienda) for the financing o f key social and infrastructure programs on health, education and other. Both PEMEX and the national budget are therefore highly vulnerable to fluctuations in international oil prices and other shocks affecting the oil market. 3. Overall, PEMEX's financial obligations to the government complicate the implementation of a sound program o f investments and capital expenditures, which i s considered essential to sustain efficient production levels and increase proven hydrocarbon reserves. Reforms to the institutional and fiscal structures that link PEMEX to Hacienda as well as measures to promote private sector participation in the industry have been proposed by several administrations. However, these proposals have found little congressional support. Natural Gas 4. Mexico has the sixth-largest natural gas reserves in the Western Hemisphere (after USA, Venezuela, Canada, Argentina and Bolivia). Yet, over the last decade the domestic demand for natural gas has outpaced the national production due mainly to the gradual installatiodconstruction o f natural gas based electricity generating plants (OCGTs, CCGTs). Today, Mexico imports about 15-20 percent o f its domestic demand for natural gas from the US and it is becoming increasingly vulnerable to price fluctuations. The increasing dependence on natural gas imports and associated high prices has led the Fox administration to prioritize the implementation o f a strategic plan aimed at increasing domestic natural gas production and lowering dependence on imports (Strategic Gas Plan). The plan includes the following measures: 0 Increase natural gas productionthrough Multiple Service Contracts 0 Diversify natural gas supply import sources and increase LNGimports 0 Reduce the flaring o f associatednatural gas 0 Expand natural gas transport, distribution and storage facilities (strengthen interconnection capacity o f pipeline grid with US) 0 Allocate more exploration funding to increase proven reserves Preliminary estimates, first three quarters 2004; EIA, DOE, 2004. Mexico Country Analysis Brief. 128 5. Mexico's downstream natural gas market has been open to private investors since the passage of the 1995 Natural Gas Law. The constitution was modified to allow private companies to become involvedinnatural gas transportation, storage, and distribution. LiquidPetroleumGas (LPG) 6. Mexico has the highest domestic and commercial consumption per capita o f LPG inthe world and it is ranked fourth inthe world interms of the volume of demand. The country's consumption has grown at 5.3% over the last decade and-due to the gradual penetration o f natural gas-it i s projected to increase at least 3.1% per annum. More than 80% o f Mexicanhouseholds rely on LPG, distributed to the final user either in portable small tanks or through tank-trucks that supply directly to stationary tanks. Today Mexico imports about a third o f LPG demand, in an international market characterized by high price volatility. PEMEX's LPGproduction reports a negative growth rate o f 0.55 percent per annum. The volume o f imports, which are subject to permits,'62i s therefore expected to increase. Since February 2003, the LPG price has been determined by a decree issued by President Fox, which will be in effect until December 2006. Underthis decree, in2005 price increases were controlled between 0.75 and 1.75 percent per month. For 2006, prices for final users will increase at a rate o f 4 percent. ElectricitySector 7. The electricity sector inMexico is characterized by a capacity stock that includes 8,250 M W of gas based independent power production (IPPs), 10,268 MW of hydroelectric capacity, 22,691 MW of thermoelectric capacity (including fuel oil and diesel), 2,600 o f coal basedpower plants, 1,365 o f nuclear capacity and 2.18 MW o f wind generation for a total installed capacity o f 46,137 MW as o f 2005. The national interconnected system has about 45,000 kmo f transmission and distribution lines. 8. Mexico has two electricity companies: Luz y Fuerza del Centro (LFyC) which serves the Mexico City metropolitan area (roughly 5 million users) and the Federal Commission o f Electricity (CFE) which serves the rest o f the country. Despite the strong technical capabilities o f the Federal Electricity Commission (CFE), one o f the largest state-owned utilities in Latin America, the sector presents a number of challenges on the technical efficiency and quality of service fronts and across segments. In addition, as in the case o f PEMEX, CFE's interdependence with Hacienda complicates the financing structure and the decision making process regarding investments and consequently the sustainability and future evolution ofthe company. 9. One o f the most important concerns o f internal and external analysts focuses on the tariff structure and the level and design of the electricity subsidy. For instance, the residential electricity subsidies are highly regressive. The upper middle income households (income deciles 6, 7, 8) receive the majority of the consumption subsidy. Despite steady tariff increases over the past 15 years, average tariff levels do not cover the costs o f providing the service. The overall degree o f cost recovery has been quite stable averaging about two thirds o f total costs in the period 1997-2003. In general, it i s recognized that the system lacks a coherent national policy framework for setting -and linking- electricity tariffs, subsidies and cost-recovery goals. Since the early 1990s, different governments have proposeddifferent models for the liberalization and reform o fthe power sector. However, these proposals have beenthe source of great political debate, public opposition and congressional rejection. LPG imports are subject to permits from the Secretaria de Economia. UntilAugust 2001, only PEMEX had a permit. However, SENER and the Secretaria de Economia issued the criteria for the expedition o f new permits. Still, today PEMEX is the only LPG importer. 129 11.KEY CHALLENGESOFTHE ENERGY SECTOR Energy sector bottlenecks will have important fiscal and macroeconomic impacts 10. The oil sector in Mexico plays a major role in economic policy. As explained before, Mexican oil export revenues account for around one-third o f government income every year. Overall, PEMEX contributes about 60 percent of its income to the federal government in royalties and taxes. Despite the highrevenues PEMEX is faced with a huge debt burden and lacks the capital resources to invest inthe modernization of the sector. One o f the major bottlenecks to the future sustainability o f the energy sector i s the systematic decline o f provendeveloped and undeveloped-hydrocarbon reserves as well as increased uncertainty regarding the size o fpotentialreserves(Figure 80). Figure 80: TotalReservesCrude Oil (Thousand Million Barrels) 55,000 50,000 45,000 9E E 40,000 35,000 30,000 1980 1985 1990 1995 2000 2005 11. In 2005, PEMEX reported proven hydrocarbon reserves o f only 14.8 mmb. The International Energy Agency (IEA) has estimated that crude oil production inMexico will decline to 4.1 mmb by 2010 and 2.7 mmb by 2030 when the expected internal demand can reach 3.5 mmb (Le. assuming an annual demand growth rate o f 2 percent). For the first time this year, the Cantarell reserves-the world's third largest oil field and the supplier o f 60% o f total oil production inMexico-will have a production fall o f about 6 percent. 12. The sustainability o f the oil stock will infact depend on new investments that allow the exploration and production o f potential national oil fields. These investments can only be triggered through either private sector participation-with mechanisms that allow a sound riskmanagement and that do not add to the country's growing contingent liability-or debt financing. 13. The possible consequences of reduced hydrocarbon reserves and the gradual deterioration o f the energy industryare two-fold. On one hand, a reduction inoil revenues will have a direct impact onpublic finances and fiscal stability thereby reducing public expenditures with the possibility o f inducing an economic recession and increasing the country indebtedness level. On the other hand, increased fuel and petrochemicals imports as well as exposure to volatile international fuel prices will have direct impacts on the competitiveness o f productive and commercial activities across the economy, including the electricity sector. 130 14. Unless Mexico implements a strategic plan that gradually substitutes the fiscal dependence on oil revenues for other sources o f income, the systematic fall in hydrocarbons reserves and an increased reliance on hydrocarbons imports will signify, without question, a bottleneck to economic and social development. InvestmentProjects under PIDIREGAS Increase the Country's ContingentLiabilitie~'~~ 15. With the intention o f promoting new mechanisms for longterm investment inenergy infrastructure, on December 1995 the Congress approved reforms to the General Law on Public Indebtedness (Ley General de Deuda Publica) and the Budget, Accounting and Federal Public Spending Law (Ley de Presupuesto, Contabilidad y Gasto Publico Federal). The reforms allowed private participation through the implementation of an instrument known as PIDIREGAS, which became operational in 1997 (PIDIREGAS means "Long-term productive infrastructure projects with deferred impact in the recording of expenditure" according to its Spanish acronym). 16. PIDIREGAS projects are investments by public sector enterprises such as CFE and PEMEX that use long-run private financing to create infrastructure capable o f generating revenue, and with a budgetary impact that i s deferred to subsequent fiscal years.164 Since 1997, a growing number o f public investment projects have been financed under the PIDIREGAS scheme. Infact, PIDIREGAS has become the main financial instrument for infrastructure development in the energy sector. In2004, for instance, PIDIREGAS represented 92% o f total PEMEX investments. 17. However, PIDIREGAS projects are regarded as public Mexicandebt and therefore private analysts and financial consultants perceive the risk associated with this instrument as equivalent to the risk on the country's sovereign debt. In other words, PIDIREGAS projects signify a contingent liability to the country. According to SHCP165, as of December 31, 2004, total liabilities o f Direct PIDIREGAS Projects amounted to US$22,155.7 million, o f which 80.1 percent corresponded to PEMEX and the rest to CFE. Table 34 below shows the participation o f total PIDIREGAS debt as a proportion o f Mexico's GDP. Note that it has increased from 1.2 percent o f GDP in 1998 to 4.9 percent o f GDP as o f end-March 2004. I PIDIREGAS 1 Domestic 2001 2.7 0 2.7 2002 3.9 0 3.9 2003 5.2 0.4 4.7 2004 (end o fMarch) 4.9 0.8 4.1 163This section is largely basedon the consultant report: Samaniego Breach, Ricardo. 2005. Mexico Infrastructure Public Expenditure Review: Targeted Issues in the Electricity Sector, The World Bank February 2005. 164See Samananiego (2005) for a detailed description o f different types o f PIDIREGAS (Le. direct and conditioned). 16'See SHCP, Informes sobre la Situacibn Econ6mica, las FinanzasPublicas y la Deuda Publica, cuarto trimestre de 2004. 131 18. At the same time total net augmented debt'66was almost constant and represented 45 percent of GDP in 1998 and 45.4 percent in 2003. In fact, the participation o f PIDIREGAS in Mexico's net augmented debt grew from 2.7 percent in 1998 to 11.5 percent in2003. 19. Table 35 also shows the importance o f PIDIREGAS liabilities in Public Sector Borrowing Requirements (PSBR). 167 In2003 they represented 1.1 percentage points o f GDP out o f total PSBR of 2.5 percent o f GDP. The approved budget for 2004 estimated an increase o f PIDIREGAS liabilities to 1.5 percent o f GDP out o f a total PSBR of 2.7 percent o f GDP. According to SHCP, fiscal pressure from investment projects financed by the private sector (PIDIREGAS-type schemes) will more than double from 1.1 percent o f GDP in 2003 to 2.4 percent in 2008. In 2005 alone, energy infrastructure financed through PIDIREGAS reached 144 billion pesos. Table35: Public Sector BorrowingRequirements,2003-2004 (% of GDP I 2003 I 2004 (approved) I.PUBLICDEFICITEXCLUDING 0.4 0.3 A. Budgetary revenues 23.7 22.7 Oil rplnted 7 9 7 1 *** VRP is the Voluntary RetirementProgramof the Public Sector. FARAC is the trust fund createdfor the toll road restructuring program Source: SHCP The traditional definition of public debt covers debt under direct budgetary control and is used for the purpose of budgetarylegislation. Specifically, it coversthe domestic debt of the federal government and the external debt of the public sector, including public, developmentbanks, and a portion o f the PIDIREGAS debt (an amount equivalent to amortization obligations for the next two year is recorded in the traditional debt upon completion of individual PIDIREGAS projects). The Public Sector Augmented Debt encompasses all government and public-sector guaranteed debt, thus covering the entire non-financial public sector as well as the public development banks. The componentsof the augmenteddebt are reported on a quarterly basis, for information purposes.Budget laws set legal approved ceilings for most of them on an annual basis. (Source: IMF. Mexico: Selected Issues, IMPF Country Report No. 04/418, December 2004). In2001, Hacienda(SHCP) beganofficially using aPublic SectorBorrowingRequirementsor PSBR broader measure of the deficit. Budget documents include amortization and interest on the debt incurred in connectionwith direct PIDIREGAS projects that have beencompleted.The PSBR includes all direct PIDIREGAS investment expenditure(net of amortization) plusthe accruedinterest on direct liabilities (net of interest expenditure included in the budget).With respectto conditioned PIDIREGAS projects, investmentand debt service are not included inthe PSBR, nor accountedfor as debt inpublic debt statistics. Outlays for these projects are recordedas operating expenses on a cashbasis once services are provided. 132 20. Although there i s no explicit limit or ceiling on PIDIREGAS debt ex ante (the only mention in Mexican laws i s that PIDJREGAS deserve special treatment and are subject to Congressional approval'68), budget laws set legally-approved ceilings for most items o f public debt-including PIDIREGAS--on an annual basis. According to SHCP, the scheme o f infrastructure investment financing through PIDIREGAS has already reached a limit and it will soon be necessary to find alternative means o f finan~ing.'~~ Indeed, the solution to the financing o f required investments in the energy sector will require strategic measures and instruments that do not increase the stock o f country's contingent liabilities. FuelPricingPoliciesHave anImpacton Competitiveness 21. Fuel pricing in Mexico i s established by the government and results in the over and under-pricing o f specific fuels and fuel products, which create market distortions with important economic and environmental consequences. Low levels o f productive efficiency, due in part to the fact that PEMEX lacks the resources to modernize its refining capacity, have a direct impact on fuel costs and the volume o f fuel imports. 22. Relative to true demand, there i s a shortage in Mexico o f light products (e.g., diesel oil) and a surplus ofheavy products (mainly fuel oil). Subsidized fuel oil promotes the maintenance and dispatch of old generating capacity in the electricity sector, leading to lower efficiency and reliability with direct effects on competitiveness across the economy. The over-pricing o f diesel and gasoline on the other hand increases the operating costs o f the industrial and commercial sectors, also with effects on competitiveness. 111. DESCRIPTION OFRELEVANT PROGRAMSAND STRATEGIES EnergySector Program 23. The Energy Sector Program 2001-2006 (PROSENER) defines the general policies and strategic lines o f the Fox's administration for the electricity and hydrocarbons sectors. The program rests on seven principles which include 1) national ownership, 2) security o f supply, 3) social contract, 4) sector modernization, 5) private sector participation, 6) sustainable development, and 7) commitments to future generation. The program outlines the following nine strategic objectives: 1. Ensure appropriate security-of-supply levels in a system with international quality standards, competitive prices and world class public and private companies. 2. Offer a solid judicial framework to develop the energy sector, providing economic agents with judicial security and certainty, ensuringat the same time national ownership and State leadership. 3. Foster the participation o f Mexicanenterprises inenergy infrastructureprojects. 4. Increase the use o f renewable energy sources and promote energy efficiency and conservation or savings. 5. Use nuclear energy - and pacific applications - in a safe way maintaining the highest international standards. 6. Become a sector leader inriskprevention for productive operations. 168Mexican laws concerning PIDIREGAS may be obtained at www.diputados.gob.md1eyinfo. 16'See "Muestran Pidiregas agotamiento: SHCP", Reforma, September 15,2004. 133 7. Become a sector leader inenvironmental protection. 8. Become leaders in the generation, development, assimilation and application of scientific and technological knowledge, as well as on the development o f highly qualified humanresources to promote the sustainable development of the sector. 9. Increase and strengthen the international energy cooperation and participate in the supply and demand sides o f global energy markets. 24. The PROSENER describes indetail what actions would be followed duringthe period 2000-20006 to achieve the ten objectives. Specific strategies are listed in B o x 17below: Box 17: Prosener 2000-2006 General Strategies PETROLEUM INDUSTRY 1. Transform PEMEX's industrial complexes and productive chains into a world class petroleumindustry 2. Increase hydrocarbons reserves and establish a program o f exploration-exploitation-extraction while protecting the environment 3. Strengthen PEMEX's refining capacity 4. Ensure the necessaryvolume o f natural gas supply at competitive prices 5. Promote structural changes inthe LPG market to ensure sufficient supply 6. Design and implement policies to strengthen PEMEX's petrochemical industry 7. Evaluate the international oil policy o f the country and, design and implement negotiation strategies to maximizerevenues from oil export operations 8. Develop a culture o f excellence and continuous improvement within State owned enterprises 9. Design a new fiscal regime that allows PEMEX to generate profits, conduct necessary investments, and guarantee an annual income to supply the fiscal sector 10. Improve instruments o f regulation to enhance and complement the existing regulatory framework with the intention o f adjusting to the changes occurring in the natural gas and LPG industries and the experience o f permit holders 11. Promote the participation of the social and private sectors in the refining industry in order to complement the supply needs o f oil products inthe mediumto long term 12. Promote the implementation o f legal and regulatory frameworks to increase the certainty o f public and private investmentsinorder to guarantee the expansion o f the electricity sectors as needed ELECTRICITY INDUSTRY 1. Promote the liberalizationof the sector 2. Promote a State policy that a) redefines and distributes subsidies targeting the poor, b) supports generators that protect the environment, c) supports rural electrificationprojects 3. Adopt labor best practices to achieve better coordination between workers, managers, and users 4. Extendthe electricity service to areas with highpoverty levels 5. Promote a change inthe fiscal regime o felectric enterprises 6. Provide legal certainty to participants in the sector and contribute to ensure State ownership and management 7. Promote the development electric and thermal self-supply projects in both the social and private sectors, particularly inthe industrial sector and the government 8. Develop programs, projects and actions to exploit renewable energy sources 9. Develop programs, projects and actions to promote energy savings 10. Maintain and improve the technical capacity o f human resources and materials to ensure security o f nuclear and radioactive installations 11. Include legal and feasible technological solutions to the problem o fradioactive waste 12. Strenghten thejudicial framework o f the nuclear in