Costa Rica’s Development From Good to Better Systematic Country Diagnostic Ana Maria Oviedo Susana M. Sanchez Kathy A. Lindert J. Humberto Lopez Costa Rica’s Development From Good to Better Ana Maria Oviedo, Susana M. Sanchez, Kathy A. Lindert, J. Humberto Lopez © 2015 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington, DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved 1 2 3 4 18 17 16 15 This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http:// creativecommons.org/licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the follow- ing conditions: Attribution—Please cite the work as follows: Oviedo, Ana Maria, Susana M. Sanchez, Kathy A. Lindert, and J. Humberto Lopez. 2015. Costa Rica’s Development: From Good to Better. Systematic Country Diag- nostic. Washington, DC: World Bank. License: Creative Commons Attribution CC BY 3.0 IGO Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official content or error in this translation. World Bank translation. The World Bank shall not be liable for any ­ Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third-party- owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to the Publishing and Knowledge Division, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Cover photo: © Andy Morffew, 2015; http://www.andymorffew.com/Galleries/Costa-Rica-Trip-2015/. Used with permission. Further permission required for reuse. Cover design: Karol A. Keane/Keane Design, Inc. Contents Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix About the Authors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi Abbreviations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Costa Rica’s Development Model: Enduring Democracy, the Social Compact, Open Economy, and the Green Trademark. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Inclusion: Inequality, Jobs, and Skills. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Growth and Competitiveness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Sustainability: Fiscal, Social, and Environmental Pressures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Governance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Priority Areas, Links, and Complementarities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Note. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 1. Introduction and Country Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 2. Poverty and Shared Prosperity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Poverty: Relatively Low Levels, but Recent Stagnation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Profile of the Poor: Increasingly Urban, Larger Households, and Lower Educational Attainment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Shared Prosperity and the Rise of the Middle Class. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Rising Inequality, in Contrast to Regional Trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Limited Poverty Responsiveness to Growth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 3. Inclusive Growth? Inequality, Jobs, and Skills . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Labor Earnings and Inequality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Fewer Jobs for Poor and Unskilled Workers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Education: Not Making the Grade, Especially for the Poor . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 Widening Earnings Gaps for Poor and Unskilled Workers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Redistribution Has Not Offset Rising Inequality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 4. The Nature of Costa Rica’s Growth and Its Constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Growth Analysis and Trends. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Policies Explaining Growth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 iii Competitiveness Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 5. Sustainability of Costa Rica’s Development Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Fiscal Sustainability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Sustainability of the Social Compact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 Sustaining the Green Trademark: Managing Natural Resources and the Environment. . . . . 119 Governance Challenges Hamper Government Effectiveness and Regulatory Growth. . . . . . 125 Notes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132 6. Priority Areas, Linkages, and Complementarities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 Appendix A: Team Engagement and Consultation Process . . . . . . . . . . . . . . . . . . . . . . . . . . 143 Appendix B: Country Comparators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 Boxes 1.1 Historical Underpinnings of Costa Rica’s Social Compact. . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 1.2 Costa Rica’s Celebrated Green Trademark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 2.1 How Have Methodological Changes Affected Poverty Measures in Costa Rica?. . . . . . . . . 25 2.2 Indigenous People in Costa Rica: A Very Small and Disadvantaged Minority. . . . . . . . . . . 29 3.1 Female Household Heads and Immigrants Face a Tough Labor Market. . . . . . . . . . . . . . . . 46 3.2 Nicaraguan Workers in Costa Rica Do Not Appear to Drive Down Native Workers’ Wages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 4.1 Intel and Development of the High-Tech Sector in Costa Rica. . . . . . . . . . . . . . . . . . . . . . . . 73 4.2 Corporate Governance of State-Owned Electric Utility ICE. . . . . . . . . . . . . . . . . . . . . . . . . . 91 5.1 The Rice Sector: Failure of a Productive Development Policy. . . . . . . . . . . . . . . . . . . . . . . . 128 5.2 The CCSS Lacks an Integrated Management System Over Resources . . . . . . . . . . . . . . . . . 129 Figures 2.1 Poverty in Costa Rica is Low by LAC Standards, but it Has Stagnated Recently. . . . . . . 24 B2.1.1 Poverty Rates Adjusting the Non-Food Component of the Poverty Line with the Overall CPI, 2001–2009. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 2.2 Poverty Varies Widely by Region, but Most of the Poor Are in the Central Region. . . . . . 27 2.3 The Highest Poverty Incidence Is in Border Areas. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 B2.2.1 Indigenous People Are Located Mostly in the Southern Region. . . . . . . . . . . . . . . . . . . . 29 B2.2.2 Access to Basic Services Is Lower for Indigenous Communities. . . . . . . . . . . . . . . . . . . . 30 B2.2.3 Indigenous People Are More Likely to Work in Self-Employment, or to Be Inactive. . . 31 2.4 The Middle Class in Costa Rica Is One of the Largest in LAC. . . . . . . . . . . . . . . . . . . . . . 33 iv Contents More People Moved Out of Poverty than Into Poverty. . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 2.5 Shared Prosperity by Country in LAC Mean Income Growth (Overall Population). . . . . . 35 2.6 Before the Crisis, Shared Prosperity Was High in Poorer Regions . . . . . . . . . . . . . . . . . . 36 2.7 Evolution of Inequality and Income Distribution in Costa Rica and in LAC . . . . . . . . . 37 2.8 Despite Regional Variation, Inequality in the Central Region Explains Most of 2.9 Overall Inequality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 2.10 Growth Benefited the Poor Relatively More Only Until 2007 . . . . . . . . . . . . . . . . . . . . . . 39 2.11 Contributions of Growth, Prices, and Income Inequality to Poverty Reduction, 2001–2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 3.1 Labor Income Contributed Significantly to Poverty and Inequality Changes, 2001–2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 3.2 Sectors that Employ Mostly Unskilled Labor Have Created Few Jobs. . . . . . . . . . . . . . . . 43 3.3 Unemployment Has Been Increasing Steadily Since 2007 . . . . . . . . . . . . . . . . . . . . . . . . . 44 3.4 Unemployment Has Hit the Poor Particularly Hard. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 3.5 Fewer than 40 Percent of Poor and Extreme Poor Adults Generate Any Labor Income. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 B3.1.1 Female Heads-of-Household and Single Mothers Have Higher Participation, and (for the Poor) Higher Unemployment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 B3.1.2 Immigrant Women Have a Much Higher Unemployment than All Other Groups. . . . . . 47 3.6 Costa Rica Achieved Substantial Gains in Primary Education Attainment, but Not in Secondary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 3.7 Relatively Few Costa Ricans Complete Secondary School. . . . . . . . . . . . . . . . . . . . . . . . . 49 3.8 Education Outcomes Are Linked to Income. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3.9 Parental Education Plays a Determinant Role in Children’s Education Investment. . . . . . 50 3.10 Regional Disparities in Education Are Large, but They Are Decreasing. . . . . . . . . . . . . . 51 K3.1.1 Main Reasons for Being Out of School . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 3.11 Costa Rica’s PISA Performance Is Below Expectations Given Its Secondary Education Spending. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 3.12 Large Disparities in Learning Outcomes by Socioeconomic Status. . . . . . . . . . . . . . . . . . 54 3.13 Real Minimum Wage Changes Closely Follow Changes in Labor Productivity . . . . . . . 56 3.14 Earnings Gaps Across Education and Income Levels Are Widening . . . . . . . . . . . . . . . . 57 3.15 Incomes Evolved Very Differently at the Top and the Bottom Before and After the Crisis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 3.16 Inequality for Different Income Concepts, Circa 2010. . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 4.1 Costa Rica GDP Growth Averaged 4.7 Percent During 1990–2014 . . . . . . . . . . . . . . . . . 63 4.2 Growth Compares Well with Respect to LAC. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 4.3 Lagging Behind in Converging Towards U.S. GDP per Capita . . . . . . . . . . . . . . . . . . . . . 64 4.4 Growth Volatility Is Low by International Standards. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 4.5 Declining Contribution of Labor to GDP Growth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 4.6 Labor Productivity Increased in All Sectors, and Workers Moved to Services . . . . . . . . 66 4.7 Investment and Exports Explain Recent Slowdown of GDP Growth. . . . . . . . . . . . . . . . 67 4.8 Services Remain the Primary Driver of GDP Growth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Contents v Agriculture Employs the Largest Number of Low-Skill Workers. . . . . . . . . . . . . . . . . . . . 69 4.9 Creation of New Firms Remains Low, Contributing to Low Job Creation. . . . . . . . . . . . 69 4.10 Costa Rica Has Been Very Successful in Attracting Foreign Direct Investment. . . . . . . 72 4.11 4.12 Costa Rica is a High-Tech Hotspot. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 Exports of Medical Devices Have Grown and Diversified . . . . . . . . . . . . . . . . . . . . . . . . . 74 4.13 Merchandise and Service Exports Are Well Diversified Into High 4.14 Value-Added Areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 4.15 Inflation Has Dropped to Single Digit and Real Exchange Rate Has Appreciated . . . . . . . 79 4.16 Labor Costs Are High in Low-Skill Sectors, Compared to Central American Neighbors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 4.17 Most of the Jobs Lost Are Mostly Unskilled Labor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 4.18 Returns to Education Reflect Growing Demand for Skilled Labor. . . . . . . . . . . . . . . . . . 82 4.19 The Higher Education System Produces Mostly Social Science Graduates . . . . . . . . . . . 83 4.20 Infrastructure Access and Coverage Is Good in Costa Rica. . . . . . . . . . . . . . . . . . . . . . . . 85 4.21 Quality of Infrastructure Is Low, Compared to Structural Peers. . . . . . . . . . . . . . . . . . . . 86 4.22 Access to ICT Is Unequal Across Regions and by Type of Activity (percent of households). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 4.23 Connectivity to Global Shipping Networks Is Low . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 4.24 Logistics Environment Is Weak. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 4.25 Electricity Tariffs Have Increased Consistently Since 2007 in Costa Rica . . . . . . . . . . . . 89 4.26 Benchmarking ICE’s Performance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 B4.2.1 Aggregated Index of Corporate Governance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 4.27 Costa Rica Investment in Infrastructure, 1994–2012 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 4.28 Costa Rica Has Burdesome Procedures (Country Ranking) . . . . . . . . . . . . . . . . . . . . . . . 93 4.29 Public Banks Dominate the Financial System. Dollarization Is Widespread, December 2014. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 4.30 Financial Depth Has Improved Since Banking Sector Liberalization. . . . . . . . . . . . . . . . 95 4.31 Costa Rica Lags Behind Comparators in Terms of Financial Depth. . . . . . . . . . . . . . . . . 95 4.32 Credit Growth to Private Sector Slowdown after Global Crisis. . . . . . . . . . . . . . . . . . . . . 96 4.33 Share of Credit to Agriculture and Industry Contracted. . . . . . . . . . . . . . . . . . . . . . . . . . . 97 4.34 Use of Bank Loans to Finance Investments and Working Capital by Firms. . . . . . . . . . . 97 4.35 Poorer and Female Costa Ricans Have Lower Levels of Financial Access . . . . . . . . . . . . 98 5.1 Central Government Fiscal Balance and Revenues. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 5.2 Primary Fiscal Balance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 5.3 Overall Fiscal Balance Deteriorated Sharply After the Crisis. . . . . . . . . . . . . . . . . . . . . . 103 5.4 Compensation of Central Government Employees. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 5.5 Public-Private Income Premium by Occupation, 2012. . . . . . . . . . . . . . . . . . . . . . . . . . . 105 5.6 Social Investment Increased Steadily Over the 20th and 21st Centuries . . . . . . . . . . . . 108 5.7 Costa Rica’s Social Spending Is the Highest in Central America. . . . . . . . . . . . . . . . . . . 108 5.8 Impressive Achievements of Costa Rica’s Social Compact. . . . . . . . . . . . . . . . . . . . . . . . 109 5.9 Costa Rica’s Public Spending on Health Care Is on Par with OECD Countries . . . . . . 110 5.10 Satisfaction with Public Health System Is Slowly Declining. . . . . . . . . . . . . . . . . . . . . . . 112 vi Contents 5.11 Costa Rica Has Heavily Invested in Education for Decades. . . . . . . . . . . . . . . . . . . . . . . 112 5.12 Costa Rica Spends More on Education Than Other MICs, but Less Than Most HICs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 5.13 More than 70 percent of Education Spending Goes to Primary and Tertiary. . . . . . . . 114 5.14 Spending on Secondary Education Is Relatively Low, Given its Level of Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 5.15 At Higher Levels, Public Education Spending Becomes Regressive . . . . . . . . . . . . . . . . 116 5.16 Most Social Protection Spending Is on Social Security and Cash Transfers. . . . . . . . . . 116 5.17 Pension Coverage Among the Elderly Is High, Thanks to the Non-Contributory Pension. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 5.18 Coverage of Major Programs Reaches Half of the Poor, at Best. . . . . . . . . . . . . . . . . . . . 118 5.19 Most Major Programs Do Not Successfully Target the Poor . . . . . . . . . . . . . . . . . . . . . . 118 5.20 Poverty Would Marginally Increase Without Avancemos . . . . . . . . . . . . . . . . . . . . . . . . 119 5.21 Costa Rica’s Success in Reforestation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 5.22 Use of Insecticides, Herbicides, and Fungicides/Bactericides (kg/ha), Circa 2011. . . . . 122 5.23 Access to Water, Sanitation, Sewage, and Connection to Sewage Treatment, Circa 2012 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 5.24 Costa Rica Has Better Governance Indicators than LAC but Lags Behind OECD Countries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 5.25 From a Two-Party to a Multi-Party System in the Last 25 Years. . . . . . . . . . . . . . . . . . . 126 5.26 The Public Sector in Costa Rica Is One of the Largest in LAC in Terms of Workers. . . . . . 131 5.27 High Socioeconomic Groups Are More Satisfied with the Government . . . . . . . . . . . . 131 6.1 Priority Areas, Linkages, and Complementarities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 Tables 2.1 Poor and Non-Poor Households Have Different Characteristics . . . . . . . . . . . . . . . . . . . . . . 31 3.1 Changing Sources of Inequality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 5.1 There Is a Large Number of Public Sector Institutions in Costa Rica, 2011. . . . . . . . . . . . . 127 Contents vii Acknowledgments We would like to thank the members of Fleischhaker, David Yuravlivker, Mateo the Costa Rica Country Team from all Clavijo, Sonia Plaza (GMFDR); Klas Sander, Global Practices and the International Carter Brandon, Christian Albert Peter, Sarah Finance Corporation (IFC), as well as the Guel (GENDR); Svetlana Edmeades, Erick many stakeholders in Costa Rica (govern- Fernandes (GFADR); Fernando Montenegro, ment authorities, think tanks, academia, and Christine Lao Peña (GHNDR); Marcelo civil society organizations), who have con- Becerra, Javier Botero, Melissa Adelman, Ciro tributed to the preparation of this document Avitabile (GEDDR); Thomas Kenyon, Melissa in a strong collaborative process (see appen- Rekas, Maria Pluvia Zuñiga, Armando dix A for a description of this process). We Heilbron (GTCDR); Cecilia M. Briceño- are very grateful for their inputs, knowledge, Garmendia, Harry Moroz, Xijie Lv, Doyle and advice. Gallegos, Pau Puig Gabarro, Rocio Sanchez The following people provided guidance Vigueras, Elena Gasol, Elene Allende (GTIDR); and inputs: Fabrizio Zarcone (Country Man­ Gabriela Elizondo Azuela, Bartley Higgins, ager for Costa Rica), Frank Sader (Principal Juan Carlos Cardenas Valero (GEEDR); Strategy Officer, CGECF), Louise J. Cord Rekha Reddy, Patricia Caraballo, Will Price, (Practice Manager, GPVDR), Auguste Tano Miquel Dijkman (GFMDR); Davide Kouame (Practice Manager, GMFDR), Zucchini, Marcelo Buitron, Lourdes Linares, Fernando Leonardo Hernandez (Special Advisor Irina Luca, José Rezk (GGODR); Armando to the Costa Rica SCD/Lead Economist, Guzmán, German Freire, Dianna Pizarro, GMFDR), Maryanne Sharp (Country Opera­ Alicia Maguid, Claudia Soto (GSURR); Luiz tions Adviser, LCC2C), Oscar Calvo-Gonzalez T. A. Maurer (GCCCS); Luis E. Gutierrez (Program Leader, LCC2C), and Ayat Soliman (LCCCR); Jovana Stojanovic (LCC2C); Jorge (Program Leader, LCC2C). Augusto de la Araujo (LCRVP); Laura Chioda, Daniel Torre (Chief Economist, LCR) chaired the Lederman, Julian Messina, Jaime Rafael de Quality Enhancement Review meeting that Pinies Bianchi (LCRCE); Alma Kanani provided guidance to the team during the (LCROS); Jorge Rodriguez-Meza (DECEA); preparation phase. Angela Fonseca (CLAPA); and Mi Hyun Bae The following members of the SCD team (CRKI4). contributed substantial inputs and/or partici- The team would like to thank the peer re- pated in several workshops: Leonardo viewers Ambar Narayan (GPDVR), David Lucchetti, Mateo Salazar, German Caruso, Rosenblatt (DECOS), and Ariel Fiszbein Liliana Sousa, Laura Moreno (GPVDR); Emma (Inter-America Dialogue) for their com- Monsalve, Pablo Acosta, Melissa Zumaeta ments. We thank other teams preparing SCDs Aurazo (GSPDR); Friederike (Fritzi) Koehler- for useful exchanges, in particular Friederike Geib, Jennifer Keller, Chris Dychala, Cornelius (Fritzi) Koehler-Geib, Kinnon Scott ix (Panama), Oscar Calvo-Gonzalez (El also like to thank Desiree Gonzalez, Cathryn Salvador), Marco Hernandez and Liliana Summers, Michelle McCue (LCC2C), Diana Sousa (Honduras), Zeinab Partow (Uruguay), Galeano, and Diana Lachy (GMFDR) for Emmanuel Skoufias (Brazil), and Barbara their support to the task and the finalizing of Cunha (Colombia). Finally, the team would the report. x Acknowledgments About the Authors Ana Maria Oviedo is an economist working Free Trade Agreements. She has conducted at the Social Protection and Labor Global research on growth determinants, financial Practice, where she leads technical and pol- markets and poverty, labor markets, and icy dialogue on social protection systems small and medium enterprise development. and labor markets with the governments of She has a PhD in development economics Colombia, Costa Rica, and El Salvador. She and a master’s degree in economics, both has also led technical and advisory work on from the Ohio State University. skills, employability and informality in Bolivia and Ecuador, and has participated Kathy A. Lindert is the Program Leader in various studies at the regional and sub-­ for Social Protection & Labor, Poverty, regional level. She was a core team member Education, and Health for the Central of the World Development Report 2014, Risk America Region. She has worked in the and Opportunity, where she led the chapter World Bank for nearly 22 years covering on household risk management. In addition, the issues of social protection, conditional she has worked extensively on investment cash transfers, social spending, redistribu- climate issues, providing analysis and techni- tion and poverty. She joined the World cal assistance to client countries in Africa, Bank in 1993, first working on food subsidy Latin America, and Central Asia to imple- reforms in the Middle East & North Africa ment and monitor business regulation re- Region and then in the Latin America forms. She has conducted research on a Region. Kathy spent five years working at broad range of topics, including skills, edu- the World Bank’s Country Office in Brazil cation quality, employment, informality, (from 2003-08), where she managed the labor and product market regulation, and World Bank’s partnership with the Bolsa productivity. She holds a Ph.D. in Economics Familia program. From 2008-12, she was from the University of Maryland, and an the Sector Manager for Social Protection Economics BA from the University of for the countries in Europe and Central Lausanne, Switzerland. Asia. She has ­authored or co-authored pa- pers on Redis​ tribution and Transfers in Susana M. Sanchez is a senior economist in Latin America, poverty in Panama and the Macroeconomics and Fiscal Management Guatemala, Designing & Implementing Global Practice group of the World Bank, Household Targeting Systems, the US wel- where she carries out economic analysis in fare system, the Nuts and Bolts of Brazil’s the Latin American Region. She has also led Bolsa Familia Program, political economy technical and advisory work on access to fi- and the media’s treatment of CCTs in nance issues in Mexico, Brazil, El Salvador, Brazil, local governance and decentralized Guatemala, Nicaragua, Turkey, and Romania. implementation of CCTs, and the self-tar- In addition, she has worked on assessing in- geting of food subsidies in North Africa. vestment climate constraints and impacts of She has a Ph.D. and M.A. in Economics xi from the University of California at Davis, for Economic Policy and Poverty Reduction and a B.A. in International Economics East Africa. Humberto has an extensive from the American University of Paris. publication record in diverse areas such as fiscal policy, optimal currency areas and real J. Humberto López is the Director of the exchange rate misalignment, armed conflict Central America Department of the World and development, and pro-poor growth. He Bank’s Latin American Region with respon- has also been the editor of three books on sibility for the Bank’s portfolio, lending, Free Trade Agreements, Remittances and strategy, and dialogue for Guatemala, El Development, and the Latin American Salvador, Honduras, Nicaragua, Costa Rica, Investment Climate, and was the lead author and Panama. During his years in the World of the World Bank 2006 Latin American Bank, he has occupied positions of signifi- Flagship on growth and poverty reduction. cant responsibility including Director for Before joining the Bank permanently, Economic Policy and Poverty Reduction of Humberto was a Professor of Economics at the Latin American region; Deputy Chief of the University of Salamanca (Spain) and a staff of the World Bank Group during the Visiting Professor at Louisiana State first year of President Jim Kim, and Manager University, Baton Rouge (US). xii About the Authors Abbreviations ARESEP  Regulatory Authority of Public Services (Autoridad Reguladora de los Servicios Públicos) BCCR Central Bank of Costa Rica (Banco Central de Costa Rica) CACM Central American Common Market CAFTA-DR Dominican Republic–Central America–United States Free Trade Agreement CANATUR National Chamber of Tourism CCH Costa Rican Chamber of Hotels CCSS  Costa Rican Social Security Administration (Caja Costarricense de Seguro Social) CEDLAS Center for Distributional Labor and Social Studies CEQ Commitment to Equity Project CGR Controller General of the Republic (Contraloría General de la República) CINDE  Costa Rica Investment Promotion Agency (Coalición Costarricense de Iniciativas de Desarrollo) COMEX Ministry of Foreign Trade (Ministerio de Comercio Exterior) CONARE National Council of Rectors (Consejo Nacional de Rectores) CONARROZ Corporación Arrocera Nacional CPI Consumer Price Index CR-ICT Costa Rican Institute for Tourism DALY Disability-Adjusted Life Years EHPM Encuesta de Hogares de Propósitos Múltiples ENAHO Encuesta Nacional de Hogares FA Frente Amplio FDI Foreign Direct Investment FODESAF  National Development and Family Allocations Fund (Fondo de Desarrollo Social y Asignaciones Familiares) FONABE National Scholarship Fund (Fondo Nacional de Becas) FSI Financial Soundness Indicators FTZ Free Trade Zone GAM Greater Metropolitan Area of the Central Valley of Costa Rica GDP Gross Domestic Product GHG Greenhouse Gas GMM Generalized Method of Moments HIC High-Income Country ICE Costa Rican Electricity Institute (Instituto Costarricense de Electricidad) ICT Information and Communication Technology IDB Inter-American Development Bank IMF International Monetary Fund xiii INA National Learning Institute (Instituto Nacional de Aprendizaje) INEC  National Institute of Statistics and Census (Instituto Nacional de Estadísticas y Censos) IPCC Intergovernmental Panel on Climate Change LAC Latin America and the Caribbean LAPOP Latin American Public Opinion Project LPI Logistics Performance Index MINAE  Ministry of Environment, Energy, and Telecommunications (Ministerio de Ambiente, Energía y Telecomunicaciones) MNC Multinational Company NPL Non-Performing Loan OECD Organisation for Economic Co-operation and Development ONF National Forestry Office PAC Partido Acción Ciudadana PDP Productive Development Policy PEFA Public Expenditure and Financial Accountability PES Payment for Environmental Services PLN Partido Liberacion Nacional PPP Purchasing Power Parity PROCOMER Center for Promotion of Foreign Trade (Promotora del Comercio Exterior) PUSC Partido Unidad Social Cristiana REER Real Effective Exchange Rate ROC Regional Operations Committee SCD Systematic Country Diagnostic SEDLAC Socio-Economic Database for Latin America and the Caribbean SERCE Second Regional Comparative and Explanatory Study SME Small and Medium Enterprise SOE State-Owned Enterprise TERCE Third Regional Comparative and Explanatory Study TFP Total Factor Productivity UNESCO United Nations Educational, Scientific, and Cultural Organization WBES World Bank Enterprise Surveys WDI World Development Indicators xiv Abbreviations Executive Summary Costa Rica’s Development island of peace at a time when armed conflict was predominant in Central America but it Model: Enduring also played a key role in the Esquipulas Agree­ ments that set the basis for the pacification of Democracy, the Social Central America. Compact, Open Economy, Likewise, the country’s Social Compact traces its roots back to the middle of the 20th and the Green Trademark century. Many of the central tenets of the So­ Sometimes referred to as the “Switzerland cial Compact were instituted during the of Latin America,” Costa Rica stands out for 1940s–50s, including: (i) enacting the Labor being among the most politically stable, pro­ Code and establishing a professional civil ser­ gressive, prosperous, and environmentally con­ vice isolating civil servants from changes in scious nations in the region. This model has the Executive; (ii) guaranteeing basic social brought important economic, social, and envi­ rights in the Constitution and adopting basic ronmental dividends, with sustained growth, welfare legislation; (iii) establishing a universal upward mobility for a large share of the popu­ health care system under the Costa Rican So­ lation, important gains in social indicators, and cial Security Agency (Caja Costarricense de significant achievements in reforestation and Seguro Social, [CCSS]); and (iv) guaranteeing conservation. Perhaps most important, the public education for all. This ambitious Social country experienced shared prosperity, with Compact has had the backing of high and ris­ strong income growth of the bottom 40 percent ing social spending. Spending on ­ education, for many years, and poverty rates in Costa Rica health, and social protection started to rise in are among the lowest in Latin America and the the 1950s, reaching over three percent of GDP Caribbean. by 1958, nearly ten percent by the end of the Indeed, Costa Rica has had uninterrupted 1990s, and then doubling during the first de­ democratic regimes since 1953, making it the cade of the 2000s. At 20.8 percent of GDP in oldest working democracy in Latin America. 2012, Costa Rica’s social spending is one of the Following a two-month civil war in 1948, the highest in Latin America and on par with Or­ Legislative Assembly was democratically ganisation for Economic Co-operation and elected to draft a new Constitution. The post–­ Development (OECD) countries. civil war administration also abolished the And the Social Compact has achieved army, and ever since Costa Rica has instead many successes, particularly in the delivery relied on a police force to maintain order. In of universal services. Costa Rica’s trademark 1953, the first presidential elections under the universal and integrated health care system, new Constitution were held, and since then managed by the CCSS has provided access to the country has held 15 additional elections health care to its entire population, including (the last in 2014), widely considered peaceful the poor and bottom 40 percent. It is widely and transparent. The country was not only an recognized that this health care system, Executive Summary 1 combined with the expansion of safe water was joining the Central America Free Trade and sanitation, are the key reasons behind Agreement (CAFTA-­ DR), which provided a Costa Rica’s impressive health outcomes: more stable and reliable framework for Costa life expectancy at birth improved from 61.6 Rica’s trade with the United States as its main years in 1960 to 72.5 years in 1980 and to trading partner, introduced changes to the 79.7 years in 2012, similar to the OECD aver­ legal framework to promote transparency, en­ age of 80.7 years. Almost 100 percent of all sured a secure and predictable environment births are attended by skilled medical person­ for investors, and led to the breakup of gov­ nel, and the infant mortality rate has fallen ernment monopolies in the telecom and in­ from 90 to 10.6 deaths per 1,000 live births. surance sectors. Following CAFTA-DR, Costa The social security and pension system has Rica has entered into further trade agree­ also been quite inclusive, and today most of ments with Canada, China, The European the population is covered by at least one of Union, Mexico, Peru, and Singapore, thus the programs, which have been extended consolidating its open economy. Another key throughout the country and include contrib­ building block of the economic model was the utory pensions, social pensions, workers’ passage of the Free Trade Zones (FTZ) law in compensation, and numerous social assis­ 1981, which started a wave of Foreign Direct tance benefits. Finally, the country has also Investment (FDI) into the country: first with shown a strong commitment to investing in low-tech sectors (textiles), and gradually at­ education, nearly reaching the spending tar­ tracting companies in high-tech sectors (elec­ gets of eight percent of GDP that were ap­ tronics, advanced manufacturing, medical proved by the Assembly in 2010. The literacy devices and life sciences, and services for rate for adults 15 years old and above is al­ export). most universal (97 percent), and the share of This outward-oriented development adults 25 years and above who had no formal model has been successful in attracting FDI education has declined from 21.2 percent in and moving Costa Rica up the global value 1950 to 3.4 percent today. Younger cohorts chains over the past two decades. The devel­ are also attaining more years of education as opment model has transformed the country time goes by; for example, the cohort 25–29 from a rural ­ based economy in agriculture-­ years old had only about four years of educa­ the mid-1950s to one with high value-added tion on average in 1950; in 1980 it had on av­ industries that contribute to several global erage a complete primary education (above value chains. By 2014, FTZs accounted for 53 six years), and in 2010 it had a complete basic percent of exports, and Costa Rica’s economy education (nine years on average). is very intensive in FDI (4.2 ­percent of GDP). Costa Rica’s economy has been trans­ In this regard, the decision by Intel to set up a formed by its outward-oriented policies. semi-conductor assembly and test plant in Trade openness has been a critical building Costa Rica in 1996 was clearly a turning block of this model, dating back to 1963, point in attracting foreign investors. Indeed, when Costa Rica joined the Central American after Intel, other well-known multinationals Common Market and to Costa Rica’s mem­ such as Abbott Laboratories (now Hospira), bership in the Caribbean Basin Initiative in Baxter Healthcare, and Procter and Gamble 1983. But perhaps the most significant step have invested in the country; and by the late 2 Executive Summary 1990s manufacturing and services had over­ Indeed, the country is one of the most taken agriculture in their contribution to visited nations in Latin America. In 2013, GDP. Costa Rica’s exports are well diversified close to 2.5 million tourists visited the coun­ with high-tech manufacturing and services try. This implies that the number of visitors playing an increasingly important role. Costa per capita (at around 0.5) is above popular Rica contributes to at least five major high- destinations in the Caribbean basin such as tech global value chains: electronics, medical Mexico (0.2), or the Dominican Republic devices, automotive, aeronautic/aerospace, (0.38). While ­ sun-and-beach tourism clearly and film/broadcasting devices. Electronics, is part of Costa Rica’s attractions, eco-tour­ medical devices, and other manufactured ism (an area where it has been a pioneer) is goods now make up about 60 percent of goods also very popular with many travelers visit­ exports. The increasingly diversified agricul­ ing the national parks and protected areas. tural portfolio (bananas, pineapples, coffee, Moreover, the country typically ranks at the etc.) accounts for most other goods exports, top of the Latin American Region Travel and while low value-added products such as tex­ Tourism Competitive Index compiled annu­ tiles are losing their relevance. Services ex­ ally by the World Economic Forum to mea­ ports (tourism, business services) also play sure the factors that make a destination an important role in the economy and have attractive for the travel and tourism industry. also diversified. Business services (mainly These hallmarks of Costa Rica’s develop­ Informa­ tion and Communication Technol­ ment model have produced clear economic ogy, ICT, related) account for almost half of and social dividends—and poverty rates all services exports. are among the lowest in Latin America and Finally, Costa Rica has also built a the Caribbean. GDP per capita has tripled world-­ renowned “Green Trademark” cen­ since 1960 and now stands at US$13,876 tered on conservation, reforestation, and (current PPP). Growth has averaged 4.5 national parks. It is the only tropical coun­ percent between 2000 and 2013, above the try in the world that has reversed deforesta­ Latin American average of 3.8 percent. tion, increasing the area covered by forests Using a US$4 poverty line, only 12 percent from 26 percent in 1983 to 52 percent today. of the country’s population is considered to Reforestation is attributed, in part, to Costa be poor (less than half the Latin American Rica’s Payments for Environmental Services average). Taking a US$2.5 poverty line, (PES) program, which provides incentives only 4.7 percent of the population is poor for forest conservation and rehabilitation (or about one-third of the Latin American of an average of 310,000 hectares per year of average). Moreover, only 1.4 percent of the privately owned lands over the past decade. population lives under the US$1.25 poverty National parks and protected areas com­ line, well below the World Bank Group’s prise 26 percent of land area in Costa Rica, Twin Goals poverty objective for 2030. Not and they have become an important desti­ surprisingly, the middle class has become nation for tourists. Costa Rica also stands the largest socioeconomic group of the coun­ out as the only country in Latin America try, comprising about 47 percent of the that has adopted the goal of achieving car­ population. bon neutrality by 2021. Executive Summary 3 However, despite these impressive achieve­ Second, Costa Rica has not converged to ments, there are a number of emerging chal­ countries of higher income, such as the US. lenges that will need to be addressed to Unlike the top regional performers (Chile, main­ tain the country’s successful develop­ Panama, and Uruguay) Costa Rica’s per ca­ ment path. pita GDP has not shown any signs of conver­ First, despite reasonable growth and a gence towards the US level in the last 25 strong commitment to the Social Compact, years. For example, in 1990, Costa Rica’s per poverty reduction has stagnated and in­ capita income was 13 percent of the US level, equality is rising. The long-term trend sug­ approximately as it is today. At the same gests rising or stagnating inequality across time, the per capita GDPs of Chile, Panama, most of the past 25 years, in stark contrast and Uruguay have increased from 13, 12 and to the significant decline in inequality in 9 percent of the US per capita GDP in 1990 the broader LAC region. More recently, the to 21, 17 and 17 percent today, respectively. gap between the rich and poor has wid­ Production costs that weaken the country’s ened significantly since the global crisis. competitiveness prevent Costa Rica from Although growth recovered promptly after joining the top growth performers. These the global crisis, above the regional aver­ production costs are affected by relatively age, job creation for low-skilled workers high wages that limit the country’s competi­ has been feeble, contributing to rising un­ tiveness in low value-­added sectors, as well employment and pushing returns to higher as by a number of investment-­ climate related education upwards. Growth has been un­ factors such as rising electricity costs, weak even, with lower growth and job creation infrastructure, and burdensome regulations. in sectors that are more likely to employ And third, fiscal pressures threaten to unskilled workers (e.g., construction and undermine the sustainability of the coun­ agriculture). Not surprisingly, inequality try’s Social Compact and Green Trademark, has increased, with the widening gap be­ and prevent it from undertaking much tween the earnings of rich and poor work­ needed investments in public infrastruc­ ers mirroring large disparities in human ture. Without fiscal consolidation, the defi­ capital and educational attainment. More­ cit could push public debt to unsustainable over, despite high spending on social pro­ levels and threaten the country’s economic, tection benefits and the broader Social social, and environmental gains. The fiscal Compact, taxes and transfers have not situation has deteriorated substantially proven to be effective in redistributing in­ since the global crisis, with the overall defi­ come to compensate for these disparities. cit of the Central Government growing to Consequently, rising inequality offset the 5.6 percent of GDP in 2013, and is projec­ poverty-reducing impact of growth in the ted to have surpassed six percent in 2014 late 2000s, and reversed what should have and to reach 6.6 in 2015. In tandem, public been a decline in poverty between 2010 debt increased from 25 percent of GDP in and 2014, with poverty increasing by 0.4 2008 to 37 percent in 2013, with projections percentage points instead of falling by a ­ of 63 percent by 2019 unless corrective projected three percentage points during measures are implemented. This recent de­ the post-crisis “recovery” period. terioration stems from counter-cyclical 4 Executive Summary measures under-taken during the crisis and risen in recent years, averaging 0.52 during structural forces that will require actions on the period from 2010 to 2014, which corre­ multiple fronts. sponds both to a revised household survey Moreover, the current political landscape and the post-crisis recovery. In contrast, the and institutional framework add an addi­ rest of Latin America witnessed a marked tional layer of complexity for approving and decline in inequality, with the regional aver­ implementing key reforms needed to address age falling by five points, from 0.57 in 2000 these emerging challenges. The shift from a to 0.52 in 2012. As a result, Costa Rica has two-party to a multi-party system in the last gone from being the least unequal country fifteen years has resulted in more complex in LAC after Uruguay in 2000, to being and lengthy reform processes. The delays and around the median country out of 17 coun­ difficulties in passing comprehensive and tries for which internationally comparable meaningful reforms, particularly on sensitive data exist for 2012. issues such as tax reform, has resulted in a Moreover, the incomes of the bottom mismatch between the demands of the popu­ 40 ­percent have fallen behind in recent years. lation and the “political delivery”. This is an The recent changes in the income distribution increasing concern given the growing need are also apparent when looking at the evolu­ to react and adapt quickly to changing global tion of the incomes of the bottom 40 percent developments. Likewise, capacity constraints, of the population, which grew about two per­ related to weak sectoral planning and bureau­ centage points less than the mean income cratic inefficiencies, also affect the ability of growth between 2010 and 2012. The stub­ the public sector to implement policies and bornly high inequality level not only offset the execute public investment projects. poverty-­ reducing impacts of growth in the late 2000s, but also reversed what should have been a decline in poverty of about three percentage Inclusion: Inequality, points between 2010 and 2014, into a poverty increase of 0.4 percentage points. Jobs, and Skills The widening income gap between the Despite its progressive social top and bottom quintiles of the population ­policies and in contrast to the trends in the reflects the mismatch between the pattern of Latin American region, inequality is rising growth and employment versus the skills in Costa Rica. Long-term trends over differ­ profile of the workforce. As discussed above, ent periods indicate that inequality has ei­ Costa Rica’s outward-oriented economic ther increased or remained flat since the late model has favored the development of high 1980s. Changes in household survey meth­ value-added sectors, such as electronics, odologies create problems of comparability medical devices, IT business services, and so over time.1 During the period from 1989 to forth. In contrast, lower value-added sectors, 2000, the Gini measure of inequality aver­ such as construction, domestic services, and aged 0.45, rising from 0.44 in 1989 to 0.48 agriculture, have grown more slowly—or by 2000. During the period from 2001 to even contracted. Costa Rica’s push toward 2009, the Gini fell and then rose again, aver­ high-end sectors has been quite successful. It aging 0.50 across the period. Inequality has has also created excess demand for skilled Executive Summary 5 workers, increasing their market value. And six years are those mainly employing low- yet, Costa Rica has not sufficiently increased skilled workers. Thus, it is no surprise that its stock of high-skilled workers. Surpris­ unemployment increased the most for this ingly, despite the long-standing commitment segment of workers. At the other end of the to invest in public education, less than half spectrum, sectors that employ mostly high- of the cohort 25–29 years old had achieved a skilled workers, such as financial services, secondary or higher education by 2010, fur­ real estate, personal services, and others, are ther adding to the stock of low-skilled adult growing fast, pushing returns to higher levels workers, a legacy of the 1980s crisis. Costa of education upward. Rica’s educational attainment is lower than As such, the earnings gaps between rich in peer countries in LAC (such as Chile and and poor and between the skilled and un­ Panama), and significantly lower than peer skilled have widened since the “inflection countries in Europe (such as Croatia and point” of the global crisis. Specifically, be­ Lithuania), and far lower than graduation tween 2007 and 2013, the earnings of those rates in the OECD. Moreover, indicators of in the top quintile increased relative to the the quality of education, such as OECD’s median earner (from 2 to 2.19), whereas PISA or UNESCO’s TERCE tests, place Costa the earnings of workers in the bottom quin­ Rican students behind most countries with tile fell in relation to the median (from 0.52 spending at comparable levels. Even more to 0.29). A similar picture is seen in dispar­ worrisome, scores on ­ international tests have ities by skill levels: for workers with tertiary worsened in recent years. Thus, the shift education, the ratio to the median income from low to high value-­ added sectors with­ increased (from 1.89 to 2.01), whereas the out a corresponding increase in the stock of ratio for workers with secondary complete skilled workers has produced the same result or incomplete remained flat, and for work­ on the distribution of income as Skill-Biased ers with primary education or less the ratio Technological Change—where low-skilled fell (from 0.79 to 0.73 of the median income). labor is gradually replaced by capital. And Indeed, the returns to education have in­ yet, with low graduation rates, the stock of creased alongside the rise in inequality in unskilled workers ­ continues to grow. Costa Rica, in contrast to trends in most The labor-market effects of this structural other LAC countries. mismatch of skills and jobs have become Furthermore, despite Costa Rica’s ambi­ even more pronounced since the global cri­ tious Social Compact, taxes and transfers sis. Job creation in sectors that employ mostly have not proven to be effective in redistribut­ low-skilled labor has been low or negative ing income to compensate for these dispari­ since 2007. Unemployment started to rise ties. This is explained by two main factors. during the global crisis and has continued in­ First, although revenue collection in Costa creasing thereafter, reaching nine percent in Rica is on par with upper-­ middle- and high-­ 2014. The poor, women, and youth have been income countries (at around 22 ­ percent of disproportionately affected in their inability GDP for general taxes plus social security to find jobs, as the sectors where job creation contributions and other revenues), general has been minimal or even negative in the last tax revenues are low (13.5 ­ percent of GDP), 6 Executive Summary and the taxation structure is not very progre­ income levels in contrast to the top perfor­ ssive. Since 1953, the Assembly has approved mers in the Latin American region. 1,259 tax exemptions (nearly half of them Importantly, the country faces several with unclear definitions of the taxes ex­ challenges to its competitiveness, and the empted), amounting to 5.6 percent of GDP. economy is showing signs of built-up vul­ Hence, the country has a limited ability to nerabilities. These constraints reflect two counterbalance income inequality through forces: the mismatch of skills and jobs, and tax policies. Second, social assistance pro­ investment-climate related factors. ­ grams in Costa Rica are fragmented, with First, relatively high wages have made fairly low coverage among the poor and weak Costa Rica less viable in low value-added sec­ targeting accuracy. Even the most widespread tors, while insufficient supply of skills ham­ program (school meals) reaches only half of pers competitiveness in high value-­ added the population in the lowest income quintile. sectors. On the one hand, high income levels, A large share of the non-poor benefit from coupled with generous social benefits, have social programs, and the targeting accuracy resulted in high reservation wages across the of programs is low compared to other coun­ board. Relatively high wages have made the tries. For example, only 44 percent of Costa country less competitive in low value-added Rica’s conditional cash transfer program sectors. On the other hand, the push toward (CCT) “Avancemos” go to those in the poor­ high value-­ added sectors such as electronics, est quintile as compared with 74 ­ percent of medical devices, and IT services has been CCT benefits under Panama’s “Red de Opor­ quite successful; but insufficient supply of tunidades” Program. high-skilled workers has created excess de­ mand for high skills. This, in turn, has put upward pressure on the salaries of high- Growth and skilled workers. And second, numerous investment-cli­ Competitiveness mate-related factors hinder Costa Rica’s com­ Costa Rica’s average growth perfor- petitiveness. Rising electricity costs, high mance has been positive over the past de­ logistics costs, and burdensome regulations, cades. As discussed above, GDP per capita plus the appreciation of the real exchange has tripled since 1960 against a 260 percent rate over the past eight years, are eroding the increase in the region. Growth has averaged country’s external competitiveness. Com­ 4.5 percent between 2000 and 2013, above bined with high wage costs, these costs have the Latin American average of 3.8 percent. made the country less competitive in low val­ Looking at the post global-­crisis period, the ue-added sectors. For example, textile ex­ country’s economy recovered similarly to a ports have dropped dramatically over the last set of comparator countries (Chile, Croatia, decade, falling from 13.3 percent of exports Dominican Republic, Lithuania, Panama, in 2000 to 1.6 percent in 2014, and this trend and Uruguay). Yet, as discussed above, since has been mainly attributed to high produc­ 1990 the country has not shown any sign tion costs in Costa Rica compared to neigh­ of convergence towards the US per capita boring countries. Although Costa Rica has Executive Summary 7 excelled in its push towards high value-ad­ created structural pressures on the country’s ded sectors, weak infrastructure outside the fiscal accounts: unlike other countries which Free Trade Zones and low backward links implemented temporary fiscal policies, Costa between export-oriented multi-national cor­ Rica chose to implement steep, permanent in­ porations and local firms have limited poten­ creases in public salaries in 2008–2010. As a tial spillover effects to the broader economy. result, the wage bill of the central government The quality of transport services and road increased from 5.5 percent of GDP in 2008 to and port infrastructure in particular is rela­ 7.4 percent in 2014. At the same time, how­ tively weak. The appreciating real exchange ever, government revenues, which had risen rate has also had negative ­ consequences for steeply prior to the crisis, dropped back to the exports and the tourism sector, as well as for average level for 2000–05. Thus, the over­ all FDI coming into or exiting the country. For deficit of the central government grew, reach­ example, in 2014, Intel’s announcement to ing 5.6 percent of GDP in 2013, and it is pro­ relocate its microchip manufacturing plant to jected to surpass six percent of GDP in 2014, Vietnam, and Bank of America’s subsequent and reach 6.6 percent of GDP in 2015. As a announcement of the closing of its service result, public debt increased by 12 percentage center, have sparked renewed dialogue about points in five years, reaching 37 percent of the importance of diversifying exports and GDP in 2013, and is projected to surpass the improving competitiveness. Although Intel somewhat worrisome level of 63 percent of has maintained its engineering and design GDP by 2019, unless corrective measures to services in Costa Rica, exports of electronics control expenditures and/or increase revenues contracted by almost 20 percent in 2014. are implemented. However, such measures are Still, export diversification has helped, and constrained by a number of rigid­ ities in the strong growth in medical device production system, including numerous expenditure has helped offset this contraction in mandates, earmarked revenues, and tax ex­ electronics. emptions. These rigidities leave only five per­ cent of the annual budget for strategic allocative decisions in any given year. Further­ Sustainability: more, the budget process is fragmented, with only the central government budget requiring Fiscal, Social, and approval by the Legislative Assembly. Without Environmental fiscal consolidation, the deficit could push public debt to unsustainable levels and Pressures threaten the country’s Social Compact and Sustainability is threatened by the dete­ Green Trademark. riorating fiscal situation. As in many other Indeed, there are tensions between the high countries, Costa Rica faced the global crisis by level of social spending and Costa Rica’s ability implementing countercyclical fiscal policies. to deliver on its ambitious Social Compact. In This was welcomed given the magnitude of health, despite the country’s long-standing the crisis and the fiscal space the country had commitment to universal coverage, important at the time (public debt was below 25 percent “cracks in the system” reduce the efficiency of of GDP in 2008). However, these policies this cornerstone of the Social Compact. The 8 Executive Summary current organization of the system has led to commitment to carbon neutrality entails not increasing wait times and patient frustration. only maintaining the current levels of forest Out-of-pocket spend­ ing has increased along­ cover but also reducing the current levels of side rising public spending on health, and greenhouse gas (GHG) emissions from all budget allocations do not take into account sectors. While a large share of GHG emissions demographic and demand changes, leading to can be achieved from better management of inefficiencies and inequities of health care. forested and agricultural land, the country Furthermore, an out­ dated infrastructure and also needs to reduce carbon emissions from information management system lowers qual­ fossil fuels coming from transport, from con­ ity and decreases transparency and efficiency struction, and increasingly from general elec­ of resource management. In education, out­ tricity consumption. This creates significant comes have been particularly disappointing, trade-offs: the PES model is itself dependent despite high public spending. Drop­ out rates on revenues generated from taxes on gasoline; are high, test scores are low, and the system is and increasing the production of clean energy not producing the skills needed to meet the necessitates in some cases intervening in pro­ demands implied by Costa Rica’s pattern of tected areas. Second, urbanization has in­ growth. Finally, despite high coverage of the creased air and water pollution, and there is social protection system, different pension re­ no long-term plan to protect the environment gimes exacerbate inequities and social assis­ from these threats. The country needs to im­ tance programs are fragmented and weakly prove its infrastructure quality, from public targeted, with limited impacts on poverty and transportation to waste treatment capabilities. inequality. In the face of rising inequality com­ Finally, the intensive use of agricultural land bined with the deterioration of the fiscal situa­ has generated a worrisome level of agro-­ tion, these inefficiencies erode support for the chemical use, and more generally the lack of a Social Compact. There are growing signs that proactive approach to territorial planning the middle and upper classes are slowly start­ jeop­ardizes the gains in environmental con­ ing to “opt out” of public services and pay for servation. Put differently, the status quo is at private ones (as is widespread in other Latin odds with the long-term goals of the country American countries). To the extent that this in terms of environmental sustainability. process continues, it will seriously undermine the commitment of these groups to finance universal services that they no longer use, fur­ Governance ther eroding the quality and legitimacy of the On top of the challenges affecting the system. three basic pillars of development (inclusion, Similarly, Costa Rica’s global leadership in growth, and sustainability), there is an addi­ environmental conservation faces the chal­ tional one that cuts across the pillars: a public lenge of growing pressures on the use of re­ sector administration that has not moder­ sources and of urbanization. First, the nized in tandem with the economy, therefore Pay­ment for Environmental Service (PES) limiting the State’s ability to deliver. Despite model competes with rising land use opportu­ Costa Rica’s good standing on governance nity costs, which makes it fiscally unsustain­ relative to the LAC region, both perceptions able in the long run. Furthermore, the and evidence suggest that its institutions and Executive Summary 9 procedures have not been able to adapt to the Business (118th). It’s startling that a country challenges of a new economic and social en­ that depends so heavily on FDI fares so vironment. Governance challenges hamper poorly in those categories. Costa Rica’s effectiveness in several ways. First, in a governance structure where power is spread across many actors, political grid­ Priority Areas, lock has hampered the adoption of reforms Links, and in many instances over the past decade(s). As a result, a consequence of this gridlock in Complementarities passing reforms has been the proliferation of Several threads weave across this “tril­ an increasing number of public (and often ogy” of challenges for inclusion, growth, and autonomous) institutions created to address sustainability—and point to priority areas for ­ specific problems. Second, the budget process action. One thread involves the interactions and numerous earmarked expenditures re­ between inequality and growth, which hinge duce the margin for the executive to control on the mismatch of skills and jobs. Another public investment and current expenses – or strand is the dual challenge of maintaining to make strategic allocative decisions. Third, competitiveness of high value-added sectors, public investment – ­ particularly in infra­ while enhancing the viability of traditional structure – has been at a standstill for de­ low value-added sectors. Mounting fiscal cades. Various factors are named as obstacles pres­sures threaten the sustainability of the So­ to the implementation of public infrastruc­ cial Compact and Green Trademark. Finally, ture and social projects in the various line the challenges of governance also weave ministries, such as the cumbersome pro­ across the development agenda, limiting the cesses associated with the Public Procure­ capacity of the public sector to adopt reforms, ment Law, deficiencies in the environmental deliver services, and execute infrastructure regulatory framework, cumbersome checks projects. These inter-connected challenges and reviews throughout the process, and so highlight a number of priority areas that forth. Fourth, perception and evidence sug­ Costa Rica needs to address to continue on a gest inefficiencies in public service delivery sustainable and inclusive growth path. (as discussed below for basic social services). And finally, the institutional complexity of Education and skills. Costa Rica needs the public sector makes for convoluted pro­ to build a skilled workforce to supports its cedures that increase transactions costs of in­ trajectory towards a high value-added teracting with public institutions and service economy and to reduce the skills-income agencies. The relatively poor ranking of gap. With fewer than half of young adults Costa Rica in the Doing Business Indicators graduating from secondary school, and (ranking 83rd out of 189 countries in 2015) is with performance on test scores falling, partly attributed to burdensome procedures Costa Rica’s labor supply does not appear within and between public institutions. Costa to be well adapted to generate the skills Rica also ranks poorly on indices of Protect­ needed for the labor market. Thus, building ing Investors (181st), Enforcing Contracts a more skilled workforce will ensure that (129th), Paying Taxes (121st) and Starting a the country remains competitive in high 10 Executive Summary value-added sectors, and that more workers standardized learning assessments, and a can access these better paying jobs, includ­ more effective governance and incentive ing those in the bottom 40 percent of the framework. population. As these changes are structural, Given the high-skilled profile of job op­ for the most part they are also long term in portunities in Costa Rica, systemic efforts nature. Workers cannot just “acquire an ed­ are also needed to enhance the quality and ucation overnight.” As such, tackling these relevance of tertiary education. Currently, challenges will require a three-pronged ap­ the tertiary education system is heavily bi­ proach: (i) strengthening the quality, reten­ ased towards social science and humanities, tion, and relevance of the education system and produces few STEM (Science, Technol­ (from pre-school through secondary ogy, Engineering and Mathematics) gradu­ school)—which will help build the skills of ates, further contributing to the skills “tomorrow’s workers”; (ii) improving the shortage in high value-added sectors. More­ quality and relevance of tertiary education; over, outdated and bureaucratic procedures and (iii) strengthening the technical train­ for recognizing foreign degrees create obsta­ ing system for the workforce. cles for Costa Rica to “import” the skills Bold actions are needed to overhaul Costa needed to sustain its high value-­ added Rica’s education system. Given the country’s growth model. Quality and accountability level of development and high education mechanisms are also needed, with perfor­ spending, the education system seriously un­ mance agreements with public universities derperforms in quality (as demonstrated by and quality accreditation standards for both test results), retention (low completion rates), university and non-university tertiary and relevance (as indicated by low returns to education training and lower levels of education). Al­ Finally, the country needs to expand the though high rates of secondary school drop­ offering and relevance of technical training, out are a symptom of the broader challenges which is the most direct way to build the in the system, imbalances in the allocation of skills of the current workforce. Again, stron­ public spending favor primary (41 percent) ger quality standards, certification of techni­ and tertiary education (32 percent) with rela­ cal programs, and accountability of training tively little allocated to the secondary level institutes could help. The National Learning (27 ­percent). Indeed, both the share of public Institute (INA) could also improve coordina­ spending and the allocation per student in tion with private-sector employers to design secondary education are low by international market-relevant curricula and course offer­ standards and given Costa Rica’s level of de­ ings so as to better respond to the needs of velopment. Moreover, inequities in learning the growing sectors of the economy. outcomes start early in life—and affect moti­ vation and abilities to learn throughout the Competitiveness and the business school years. In addition to rebalancing ­climate. Boosting growth and inclusion in spending towards secondary school and early the labor market requires confronting the childhood development, Costa Rica needs to dual challenge of maintaining competitive­ strengthen teacher quality and improve ac­ ness of high value-added sectors, while im­ countability through regular monitoring with proving the viability of low value-added Executive Summary 11 sectors. For instance, improving the integra­ jobs, and this is true in particular for firms tion of export-­oriented and domestic firms that operate outside Free Trade Zones through backward linkages could sustain the (FTZs). In addition, the country needs to growth among small and medium enter­ improve its waste management and clean prises (SMEs), generating jobs in mid- and energy production capabilities to be able to low-skilled occupations. This can be done reduce GHG emissions, and water and soil by lowering operation costs to improve the pollution. Infrastructure improvement productivity of labor and counterbalance the poses a number of tradeoffs, including the high labor costs in Costa Rica compared to need to intervene in protected areas (in the its neighbors, for example by lowering the case of clean energy production), as well as costs of doing business through regulatory the need to control the current fiscal defi­ simplification. cit. Given the necessity of continuing to in­ vest in infrastructure, and the reality of Infrastructure. Reducing the infrastruc­ fiscal constraints, Costa Rica must look for ture deficit would increase competitiveness, options for private sector participation in growth, and environmental sustainability. the maintenance and upgrading of its Costa Rica’s historical efforts to build an infrastructure. extensive network of infrastructure in Both growth and inclusion would benefit nearly all productive service areas (water, from the many complementarities involved sanitation, transport, electricity, and tele­ in improving education, competitiveness, communications) are clear from the infra­ and infrastructure. A well-educated work­ structure stock: the country has two times force with relevant skills is fundamental for the road and three times the rail density of sustaining economic growth and increasing the average middle-income country; access productivity. In parallel, closing the educa­ to electricity is nearly universal; and mo­ tion gap between the poor and non-poor is bile penetration is higher than the OECD also highly relevant for inclusion by provid­ average. Yet, the near freeze in public infra­ ing opportunities those in the bottom 40 structure investment until the 1990s, as percent. Lowering the costs of doing business well as recent fiscal constraints, have taken will boost competitiveness across various a toll on the country’s ability to upgrade sectors. Furthermore, increasing infrastruc­ and maintain its infrastructure. Further, ture spending would stimulate construction, the government has faced significant chal­ thereby creating more jobs for the large stock lenges in executing infrastructure invest­ of low-skilled workers. ment in a timely manner. As a result, today At the same time, actions are needed to roads and ports have among the lowest ensure the sustainability of Costa Rica’s de­ quality marks in the LAC region. Electric­ velopment model: ity prices have doubled since 2007 due to weather related variable hydroelectric out­ • Fiscal sustainability: Improving the put, causing increased use of thermal units fiscal stance to restore sustainability re­ and high operating costs, among other fac­ quires ­reforms to manage expenditures tors. This infra­structure deficit reduces the and increase revenues. On the expen­ potential of local firms to grow and create diture front, these include containing 12 Executive Summary the wage bill of the consolidated public the health system for accountability, ef­ sector, as public sector wages, both in ficiency, and performance. The ­ country government and more so in state-owned also needs to increase the effectiveness enterprises and other public institutions, of social protection programs by (i) har­ are well above the private sector at all monizing eligibility criteria and social employment categories; and reviewing information systems; (ii) improving per­ the fiscal sustainability of the pension formance monitoring and evaluation; and system, particularly of special pension (iii) reducing institutional and program regimes in the public sector. In addition, fragmentation. a comprehensive reform of the budge­ • Environmental sustainability: To sus­ tary process is needed to ­ increase effi­ tain its celebrated “Green Trademark,” ciency and transparency in all public Costa Rica needs to balance environmen­ sector entities. Finally, curtailing the tal and natural resource management earmarking of revenues, which cover goals, by: (i) reviewing the sustainability more than half of primary spending, of the PES mechanism for conservation; would make the budgetary process more (ii) modernizing water and solid waste flexible. Comprehensive reforms are also treatment (infrastructure, ­ service deliv­ needed to increase revenues. For exam­ ery, regulatory framework, capacity); ple, the 1,259 tax exemptions approved (iii) increasing the supply of renewable since 1953 that comprise almost six energy by making regulations on the use percent of GDP need to be thoroughly of protected areas more flexible; (iv) im­ reviewed. Curtailing those exemptions plementing a comprehensive transport would make the tax system more ratio­ policy, including measures to reduce nal and progressive, as well as produce growth of demand for energy associated higher revenues. This reform is essential with transport; (v) reinforcing regulation for restoring sustainability to the fiscal and oversight of agro-chemical use and accounts, which is a necessary condition incentivizing the expansion of “green” for achieving Costa Rica’s economic and (organic) agriculture; and (vi) improving social objectives. territorial planning, land management, • Social sustainability: In addition to and management of natural and man- strengthening education, priority areas made disasters. for sustaining the Social Compact include health and social protection. Costa Rica Governance. Progress in all the priority needs to modernize its universal health areas discussed above hinges on improving system to improve quality by: (i) strength­ the capacity of the public sector to plan and ening the health care model to enhance implement policies, execute public invest­ capacity to adapt to demographic and ment projects, deliver services, and increase epidemiological change while ensuring accountability. Despite Costa Rica’s good quality and timeliness of service ­delivery; governance levels compared to the rest of the (ii) improving the financial model of LAC region, there is a growing perception of budget and resource allocation; and (iii) low effectiveness of government institutions. improving the management model for Cumbersome regulations, in many cases Executive Summary 13 resulting from lack of coordination among filled to inform better policy decisions. Al­ institutions, make the process of starting and though there is a large and productive re­ running a business—­ particularly a non-FTZ search and policy analysis community small or medium enterprise—more chal­ studying Costa Rica, there are a few issues lenging. Low levels of transparency and ac­ where having further research and data countability lower the efficiency of public collection would provide more information spending. The current political landscape, to help design more concrete policy re­ where political minorities have the power to forms to tackle the issues presented above. delay votes, further reduces the margin for For example, the specific factors driving approving and implementing needed re­ secondary dropout are still not well under­ forms. The need to improve governance is stood. In the labor market, it is not clear apparent in all priority areas, for instance by whether the contraction of employment in increasing accountability in the education agriculture, manufacturing and construc­ sector (e.g., by tracking student achievement tion is of a cyclical or structural nature, to reward teacher and school performance). and this has important implications for Employment creation would benefit from low-skilled unemployment. Likewise, al­ streamlining business regulations as well as though there are strong signs that reserva­ the public procurement and investment pro­ tion wages are high, there could be more cesses to improve infrastructure. A more studies to quantify this ­ phenomenon bet­ consolidated budget, fewer tax exemptions, ter. To improve the efficiency of the public and more control over spending by autono­ sector, it is crucial to identify the specific mous institutions could greatly help to re­ governance bottlenecks in executing infra­ duce the current fiscal deficit, and would structure projects, as well as their cost im­ improve the capacity to monitor results of plications. Also, the articulation public spending. In turn, results-based man­ mech­ anisms to improve the effectiveness agement would help to boost the efficiency of social programs need to be based on a of public spending, for example, enforcing thorough institutional mapping of social the use of common information systems and programs. Finally, environmental conser­ modernizing the M&E frameworks in the vation needs to be better linked with eco­ social sectors. nomic activity, and a key knowledge gap in this regard is how to link sustainable pro­ An agenda for knowledge. Finally, a duction and rural landscapes number of knowledge gaps need to be to conservation. Note 1. The long-term trends in poverty and inequal­ data series occur in 2001 and 2010. As such, ity are subject to caveats regarding income measurement of welfare across the three time measurement due to several changes in sur­ periods of 1989–2000, 2001–09, and 2010–14 vey methodology across that time period. is not strictly comparable. See chapter 2 and The main “breaks” in comparability of the box 2.1. 14 Executive Summary 1. Introduction and Country Context Costa Rica is a small country, with am- Latin America. There is a clear checks-and- bitious economic, social, and environmental balances system between the Executive and goals. The country covers a land area of the Legislative Assembly. Costa Rica’s politi- 19,700 square miles (51,100 square kilome- cal norms have long placed a high premium ters), with a population of 4.9 million, of on achievement of wide socio-political con- which about three-quarters live in urban sensus on major policies. Participation in areas. An upper-­ middle income-country, elections remains high, with 70 percent vot- with GDP per capita of US$13,876 (current ing in 2010 and 68 percent in 2014, although PPP), it is under consideration for member- it has declined from levels around 80 percent ship to the Organisation for Economic Co- during elections in the 1990s.1 This political operation and Development (OECD), with stability, which has earned Costa Rica the discussions scheduled for this year. Costa label of “Switzerland of Latin America,” has Rica’s development model centers on four been attributed to several factors, including key pillars: its long-standing democracy, an the abolition of the army after a two-month ambitious Social Compact, an outward-ori- civil war in 1948, the limited presence and ented economy, and its celebrated Green power of land-based oligarchies, and the Trademark. This model has achieved many 1949 Constitution, which created one of successes, including sustained healthy Latin America’s first welfare states. growth rates, improvements in social indi- The country has also forged a far-reaching cators, environmental gains, and one of the Social Compact, with universal access to basic lowest poverty rates in the Latin America services and a long tradition of social rights. and Caribbean region. Costa Rica’s Social Compact traces its roots Anchored in a region rife with political back to the middle of the 20th century, and is and social turmoil until the late 20th century, founded on constitutional guarantees of basic Costa Rica stands out as the longest working social rights, an extensive Labor Code, public democracy in Latin America. The country— education, and a universal health care system one of the smallest in Latin America and the managed by the Social Security Agency (box Caribbean (LAC)—started a process of de­ 1.1). As a result, most of the population has mo­ cratic nation building back in the 19th access to improved water sources (96 per- century, with the development of the coffee cent), sanitation (94 percent), and electricity industry. Stability and consensus were nur- (99 percent).2 It is widely recognized that tured from the country’s early years and con- Costa Rica’s universal health care system, tinued during the transition to the modern along with the expansion of safe water and democracy that started in 1949 (box 1.1). sanitation, are the key reasons behind its im- Since then, elections have been held every pressive health outcomes.3 Life expectancy at four years, with peaceful transitions of power birth increased from 61.6 years in 1960 to between political parties, making it the oldest 79.7 years in 2012, just one year less than the working democracy in OECD average of 80.7 years. Almost all births Introduction and Country Context 15 Box 1.1  Historical Underpinnings of Costa Rica’s Social Compact Costa Rica’s social contract has its roots in the pattern of its colonial settlement and the structure of coffee production. Largely isolated, sparsely populated, and without precious metals to export, Costa Rica was a uniformly poor subsistence economy during the colo- nial period (1600–1800). The farmers of the Central Valley worked their own land on homesteads. Given the dominance of the farm household, Costa Rica has been described as a “rural democracy” during this period, with minimal social divisions and few class dis- tinctions, in contrast to neighboring countries. Following independence in 1821, coffee production grew rapidly and became the en- gine of Costa Rica’s economic growth. Coffee became a driver for the development of the country’s infrastructure, institutional organization, and productive structure. Roads were built to export coffee to Europe. Coffee production accelerated moving from communal land rights to the privatization of land, and the predominance of small landholdings and the notion of individual property rights contributed to low in- equality and shared wealth. Abundant land supply and labor scarcity contributed to high wages, contributing to a more equitable distribution of the income from coffee exports. As international demand increased, large exporters were forced to buy coffee from the smallholders, while in turn smallholders depended on large exporters for the beneficio—the hullery where coffee is washed, dried, and packaged for export—thereby creating a strong mutual interdependence. By the middle of the 19th century, this facili- tated the consolidation of the Coffee Pact—the implicit contract developed between the large exporters and smaller producers. In part, this pact was feasible because Costa Rica had a small territory with a small population concentrated in the Central Valley, where all faced the same adversities and risks. The implicit contract was a social mech- anism for dealing with these risks and overcoming the shortcomings of small size. The social relationships implicit in the Coffee Pact gradually led to the emergence of a na- tional egalitarian ideology that validated the participation of small producers and the landless workers not only in the distribution of the economy’s product but also in the po- litical arena. This led to Costa Rica’s version of the liberal state, in which the interests of different sectors could be democratically expressed, rules were universally applicable, and institutions and legal codes promoted the growth of economic activity and a democratic society. Liberal governments created a legal framework that protected individual and pro­ perty rights, guaranteed market freedom, and placed multiple limits and counterbalances on power. This egalitarian ideology provided the underpinnings for Costa Rica’s notable Welfare State. In 1941, President Rafael Angel Calderon Guardia created the social security sys- tem (Caja Costarricense de Seguro Social, CCSS) to provide public health services, which 16 Introduction and Country Context Box 1.1  continued have become universal since then. In 1943, a Labor Code was enacted and an article on social rights (garantías sociales) was written into the Constitution of 1949. The social rights included health, housing, social assistance, public services, taxes, and minimum wages, among others. Source: Adapted from Gonzalez-Vega and Cespedes (1993). are attended by skilled medical personnel, Salvador, Guatemala, Honduras, and and the infant mortality rate stands at 10.6 ­ Nicaragua), which spearheaded trade inte- deaths per 1,000 live births, a vast improve- gration in Central America, eventually ment since 1960 which recorded 90 deaths leading to a customs union. A second per 1,000. Moreover, by 2011 the literacy rate milestone took place 20 years later (1983), for adults 15 years old and above had reached when Costa Rica joined the Caribbean 97 percent, and the share of adults 25 years Basin Initiative, strengthening its trade re- and above who had no formal education was lations with the United States. But perhaps 3.4 percent in 2010. Younger cohorts are at- the most significant step was joining taining more years of education as time goes CAFTA-­ DR (the Dominican Republic- by. For example, the cohort of young adults Central America-­ United States Free Trade aged 25–29 years had only about four years of Agreement), which provided a more stable education on average in 1950; in 1980 it had and reliable framework for Costa Rica’s on average a complete primary education (six trade with its main trading partner, the years); and by 2010 it had a complete basic United States, introduced changes to the education (nine years on average). This is in legal framework to promote transparency, part the result of a strong commitment of the ensured a secure and predictable environ- society to invest in education, as reflected in a ment for investors, and led to the breakup recently approved constitutional mandate of government monopolies in the telecom (June 2010) to raise education spending from and insurance sectors. Following CAFTA-­ six to eight percent of GDP. DR, Costa Rica has entered into further Costa Rica’s economy has been trans- trade agreements with Canada, China, the formed by its outward-oriented policies, European Union, Mexico, Peru, and Singa- centered on trade openness, export diver­ pore, thus consolidating its open economy. sification, Free Trade Zones (FTZs), and Another key building block of the econo­ foreign direct investment (FDI). Trade mic model was the passage of the FTZ law open­ ness has been a critical building block in 1981 (amended in 1990 and 2010), of this model, dating back to 1963, when which started a wave of FDI into the coun- Costa Rica joined the Central American try: first with low-tech sectors (textiles), Common Market (CACM, made up of El and gradually attracting companies in Introduction and Country Context 17 high-tech sectors (electronics, advanced the Payments for Environmental Services manufacturing, medical devices and life (PES) program, which provides incentives sciences, and services for export). to forest conservation and rehabilitation, This outward-oriented development reaching an average of 310,000 hectares per model has been successful in attracting FDI year of privately owned lands over the past and moving Costa Rica up the global value decade. Costa Rica has also set aside 26 chains over the past two decades. The devel- percent of its land area for protected areas, opment model has transformed the country including national parks, which have be- from a rural ­ agriculture-based economy in come an important destination for tourists the mid-1950s to one with high value-added (53 ­percent of tourists visited those parks industries that contribute to several global in 2012).4 Tourism has become a dynamic value chains. By 2014, FTZs accounted for sector, contributing 4.6 percent of GDP and 53 percent of exports, and Costa Rica’s econ- 14.2 percent of total exports in 2013. Costa omy is very intensive in FDI (4.2 percent of Rica also stands out as the first country to GDP). Exports are well diversified with adopt the goal of achieving carbon neutral- high-­tech manufacturing and services play- ity by 2021 (box 1.2). ing an increasingly important role. Costa This development model has produced Rica contributes to at least five major high- clear economic and social dividends—and tech global value chains: electronics, medical poverty rates are among the lowest in Latin devices, ­automotive, aeronautic/aerospace, America and the Caribbean. GDP per capita and film/broadcasting devices. Electronics, has tripled since 1960 and now stands at medical devices, and other man­ ufactured US$13,876 (current PPP). Growth has aver- goods now make up about 60 percent of aged 4.5 percent between 2000 and 2013, goods exports. The increasingly diversified above the Latin American average of 3.8 per- agricultural port­ folio (bananas, pineapples, cent. Using a US$4 (2005 PPP) poverty line, coffee, etc.) accounts for most other goods only 12 percent of the country’s population exports, while low value-added products is considered to be poor (less than half of the such as textiles are losing their relevance. Latin American average). Taking a US$2.5 Services exports (tourism, business services) poverty line, only 4.7 ­ percent of the popula- also play an important role in the economy tion is poor (or about one-third of the Latin and have also diver­ sified. Business services American average). Moreover, only 1.4 per- (mainly Information and Communication cent of the population lives under the Technology, ICT, related) account for almost US$1.25 poverty line, well below the World half of all services exports. Bank Group’s Twin Goals poverty objective Finally, Costa Rica has built a world-­ for 2030.5 Not surprisingly, the middle class renowned “Green Trademark,” centered on has become the largest socioeconomic group conservation, reforestation, and protected of the country, comprising about 47 percent areas. It is the only tropical country in the of the population. world that has reversed deforestation, in- However, despite these impressive achieve- creasing the area covered by forests from 26 ments, underlying vulnerabilities challenge percent in 1983 to 52 percent today (box Costa Rica’s development model. There is a 1.2). Reforestation is attributed, in part, to sense that, although Costa Rica has achieved 18 Introduction and Country Context Box 1.2  Costa Rica’s Celebrated Green Trademark Costa Rica takes up only 0.03 percent of the earth’s surface but is considered to be one of the 20 countries with the greatest biodiversity in the world, with 52 percent covered by forests and 26 percent in protected areas. This was not always the case. Although more than half of the country was covered by forest in 1950, deforestation had stripped this to just 26 percent by 1983 under the pressures of a growing population and massive conver- sion of forest to pasture, underpinned by ­policies that supported cattle ranching and agri- cultural development. Green Framework and Payments for Environmental Services. Recognizing the potential economic and global value of the forests and the risks to environmental degradation of rapid deforestation, several framework laws established the basic principles of Costa Rica’s “Green Trademark” aimed at rehabilitating wooded lands and setting aside vast areas of land as protected areas. It took years of policy debate and consensus building to develop a national approach of paying for environmental services—which now takes the form of the world-renowned Payments for Environmental Services (PES) program. This program builds on a long chain of earlier legislation that made the costs of reforestation tax de- ductible (Forest Law No. 4475 of 1969), established reforestation as a legal imperative (Forest Law No. 6184 of 1977), and established fiscal incentives through the creation of certificates that rewarded landowners for reforestation and allowed for significant forest authority intervention in the use of forest resources (Forest Laws No. 7032 of 1986 and 7174 of 1990). The PES was formally established in 1995 when the government adopted Forest Law No. 7575. This law established a framework for payments to landowners for these eco- system services, and established the National Fund for Forestry Financing (Fondo Nacional de Financiamiento Forestal, FONAFIFO), to manage the PES and established a new certificate for forest conservation to reward landholders for their ecosystem ser- vices. Besides the 1995 Forest Law, the government enacted other environmentally re- lated laws: (i) the 1995 Environment Law No. 7554, which mandates a “balanced and ecologically driven environment for all”; (ii) the 1996 Forest Law No. 7575, which man- dates “rational use” of all natural resources and prohibits land cover change in forests; and (iii) the 1998 Biodiversity Law, which promotes the conservation and “rational use” of biodiversity resources and includes measures to conserve, protect, and sustainably exploit biological resources to ensure quality of life for future generations and the sur- vival of natural heritage. Subsequently, the Fiscal Simplification and Efficiency Law No. 8114 (2001) fixed FONAFIFO’s share of fuel tax revenues to 3.5 ­ percent, guaranteed through the Ordinary National Budget. In addition, the Decree of the President of the box continues next page Introduction and Country Context 19 Box 1.2  continued Republic No. 32868 in 2005 (“Inaugurating and Regulating a Water Charging Scheme”) introduced a mandatory payment for ecosystem services via instituting a water tariff structure through which 25 percent of the proceeds from water charges are allocated to the PES Program. National Parks and Protected Areas. The National System of Protected areas was introduced in the late 1980s, followed by the National System of Conservation Areas (Sistema Nacional de Áreas de Conservación, SINAC) was created in 1994 to organize the country into 11 large conservation areas, most of which are based around a major national park, to avoid a situation in which the protected areas would serve as isolated “green islands” in an other- wise improperly managed landscape. SINAC oversees over 160 protected areas, of which 26 are designated national parks. Other areas are designated wildlife refuges, biological re- serves, national monuments, forest reserves, national wetlands, and protected zones. Carbon Neutrality. In May 2007, the Costa Rican government announced its intention to become 100 percent carbon neutral by 2021. Efforts are underway to begin offsetting all of the country’s carbon dioxide emissions using a wide range of budgetary, legal, and finan- cial incentives including measures to promote biofuels, hybrid vehicles, and clean energy. Another key component of the national strategy will be a “C-Neutral” label to certify that tourism and certain industrial practices mitigate all of the carbon dioxide they emit. Under the new certification system, tourists and businesses would be charged a voluntary “tax” to offset their carbon emissions, with one ton of carbon valued at US$10. The funds would be used for conservation, reforestation, and research in protected areas. Sources: Bennet & Henniger (2009), Chomitz et al. (1998), Rodriguez Zunega (2003), WorldWatch Institute (2015), and Brown and Bird (2010). impressive “levels” of performance for many dissatisfaction in the quality of the health care indicators (relatively low poverty rates, long system; rising electricity costs; a near halting life expectancy, near-universal literacy, envi- of infrastructure investment; increasing air ronmental conservation), it is increasingly be- and water pollution; and the deteriorating fis- coming constrained on the “deltas” of cal balance. In turn, these risks are symptoms ado­ pting the changes needed to make the of deeper structural problems in the economy transition to a modern, higher-­ income econ- and the public sector. The implications of omy that is aspiring to join the OECD. These these challenges are worrisome: although the “cracks in the system” are apparent in many economy is expected to continue to grow at dimensions: rising inequality; growing unem- healthy rates, poverty reduction has stagnated, ployment, especially among the poor; consis- the incomes of the bottom 40 percent are fall- tently high school dropout rates; increasing ing behind, and inequality is rising. 20 Introduction and Country Context These challenges affect the basic pillars un- to threaten Costa Rica’s development derlying poverty reduction and shared pros- model. The fiscal situation stands out as perity: inclusion, growth, and sustainability. one of the most pressing development challenges facing Costa Rica. The recent •• On the inclusion front, inequality is deterioration stems from a combination rising, despite the country’s progressive of countercyclical measures undertaken policies and ambitious Social Com- during the crisis and structural forces. pact—and in contrast to the marked Without fiscal consolidation, the deficit decline in inequality in much of the could push public debt to ­ unsustainable LAC region. Consequently, inequality levels and threaten the country’s eco- offset the poverty-reducing impacts of nomic, social, and economic gains. growth in the late 2000s, and reversed Indeed, fiscal pressures, as well as rising what should have been a decline in inequality, and increasing dissatisfaction poverty, with poverty increasing by 0.4 with social services all combine to create percentage points instead of falling by ­ tensions in the social compact. Finally, a projected three percentage points Costa Rica’s leadership in ­environmental during the post-crisis recovery period. conservation faces the challenge of The widening income gap between rich growing pressures on the use of resources and poor reflects changes in the labor and of urbanization. Changing economic market, weak educational outcomes, incentives threaten the gains in reforesta- and a mismatch between the pattern tion, and the use of agro-­chemicals is of growth and the skills profile of the worrisome. Urbanization has increased workforce. energy use, as well as air and water pollu- •• On the growth front, although Costa tion; and the country has not ­developed Rica’s growth performance over the a long-term plan to ­protect the environ- past decades has been positive, it has ment from these threats. not made much headway in conver- gence with the United States, lagging This Systematic Country Diagnostic (SCD) behind structural peer ­ countries, such takes stock of Costa Rica’s path to success, and as Chile, Panama, and Uruguay. More­ reflects on the development constraints and over, the country faces several chal­ len­ opportunities that the country faces going for- ges to its competitiveness, including ward. It does so through diagnostics of the the mismatch of skills and jobs, invest­ trends, drivers, and challenges to provide ele- ment-climate related factors, production ments to answer three basic questions: costs, and weaknesses in public-­ sector 1. To what extent has the Costa Rican administration. As a result, the econ- development model been inclusive? omy is showing signs of built-up 2. What has driven growth in Costa Rica vulner­abilities. in recent years, and what are the bottle- •• And finally, on the sustainability front, necks that need to be addressed? the deterioration of fiscal balances, ten- 3. How sustainable is the development sions in the Social Compact, and various model of Costa Rica economically, risks to the Green Trademark all combine socially, and environmentally? Introduction and Country Context 21 The SCD is organized as follows. Chap- delivery of public services, and the green ter 2 examines trends in poverty and trademark of the country. Finally, based on shared prosperity. Chapter 3 assesses the the previous analyses and a participatory inclusiveness of growth, with a focus on process with partners and stakeholders in the challenge of rising inequality and the Costa Rica as well as within the WBG,6 mismatch of labor opportunities and the Chapter 6 identifies priority areas for pol- skills of the workforce, with implications icy to ensure that Costa Rica stays on a for the poor and the bottom 40 percent sustained path of growth, inclusiveness, of the population. Chapter 4 analyzes the and shared prosperity. In several parts of drivers of growth, as well as key con- the analysis, this SCD relies on compari- straints to growth looking forward. Chap- sons with other countries, including regio­ ter 5 identifies risks to the sustainability nal peers, structural peers, and aspi­ rational of development, including growth, the peers (OECD countries). 7 Notes 1. Estado de la Nación (2012) and Estado de la 4. DeShazo and Monestel Vega (1999) found that Nación (2014). tourists spend a lot of time in those national 2. Averages for 2010–13 from World Bank World parks. Development Indicators using the Find-­ My- 5. The WBG’s Twin Goals are (i) to reduce the per- Friends Tool. Moreover, following the liberal- centage of people living under US$1.25 per day to ization of the telecommunications market due three percent globally by 2030, and (ii) to increase to CAFTA-DR, mobile-­ cellular penetration the income growth rate for the bottom 40 percent levels have quickly caught up with other coun- relative to average growth, within each country. tries in the region, reaching 135  percent per 6. See appendix A for a full description of the 100 inhabitants and 94 ­ ­ ouseholds percent of h process. in 2014. 7. See appendix B for a description of selection of 3. Montenegro (2013). comparator countries used in the analysis. 22 Introduction and Country Context 2. Poverty and Shared Prosperity Poverty rates in Costa Rica are among the lowest in Latin America and the Caribbean, and the middle class is the predominant group. Nonetheless, poverty increased and then stagnated during and after the global crisis. The patterns of poverty show an uneven level of develop­ ment in the country, with the Central region being more prosperous relative to the peripheral regions. Moreover, income inequality is rising, in contrast with the historical decline observed in most countries in Latin America. Consequently, inequality offset the poverty-reducing im­ pact of growth in the late 2000s, with poverty rising instead of declining despite economic growth. Poverty: Relatively Low (11 percent), Chile (10 percent), and Uruguay (8 percent) and less than half of ­ Levels, but Recent the LAC average in 2012. Official poverty figures show a modest Stagnation decline since the 1990s, and a stagnation of One of the great outcomes of Costa poverty since 2010. Using the official pov- Rica’s sustained GDP growth and Social erty line and the per capita income aggre- Compact is the reduction of poverty that gate, overall poverty in Costa Rica declined took place over the second half of the 20th from around 31 percent to 19 percent be- century. The poverty headcount—measured tween 1989 and 2007 (figure 2.1b). Poverty with an internationally comparable poverty reduction followed a cyclical pattern: after a line of US$4 per day—was reduced from period of sharp decline in 1989–1994, pov- around 40 percent in 1989 to 27 percent by erty stagnated until 2000; then during the the turn of the century. Poverty decreased economic expansion of 2001–2007, it de- further to 18 percent in 2007, falling at a clined modestly at about 0.6 percentage rate of about 1.5 percentage points per year points per year. The global crisis reversed (figure 2.1a). However, poverty reduction some of these gains, causing poverty to in- stagnated during and after the global crisis. crease to 21.6 percent in 2009; and it has re- Trends for extreme poverty—defined inter- mained stable during the 2010–2014 period nationally in Latin America and the Carib- at around 24 percent. Extreme poverty fol- bean (LAC) as per capita income under lowed a similar pattern, declining from 9.9 US$2.50 a day—are similar, with a sustained percent to 3.7 percent between 1989 and fall from 14 percent in 2001 to seven per- 2007, and then increasing to 5.2 in 2009. cent in 2007, and a slower decline after Extreme poverty has averaged 7.2 percent 2010 to an average of five percent.1 Costa in 2010–2014. By 2014, the number of poor Rica’s poverty rate of 12 percent was the amounted to 1.17 million, including 0.45 lowest in Latin American after Argentina million of extreme poor (box 2.1).2 Poverty and Shared Prosperity 23 Figure 2.1  Poverty in Costa Rica is Low by LAC Standards, but it Has Stagnated Recently a. Poverty reduction similar to LAC until 2008 (International poverty line, 1989–2013) 50 40 Poverty rates (%) 30 20 10 0 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 EHPM ENAHO LAC (US$4/day) Costa Rica (US$4/day) LAC (US$2.5/day) Costa Rica (US$2.5/day) b. Poverty has stagnated in recent years in Costa Rica (National poverty line, 1989–2014) 50 8 6 40 GDP per capita growth (%) 4 Poverty rate (%) 30 2 20 0 10 –2 0 –4 2007 2013 2014 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2008 2009 2010 2011 2012 EHPM ENAHO Overall poverty GDP per capita growth Extreme poverty Source: Elaboration using SEDLAC (CEDLAS and The World Bank); and INEC data and World Development Indicators. 24 Poverty and Shared Prosperity Box 2.1  How Have Methodological Changes Affected Poverty Measures in Costa Rica? The analysis of trends in poverty and inequality over time must be treated with some caution due to changes in methodologies used both for the collection of household survey data and for the calculation of poverty lines in Costa Rica. Two household surveys are used to measure poverty from 1989 to 2014, but they do not allow for comparisons on poverty rates. The surveys are Encuesta de Hogares de Propósitos Múltiples (EHPM) up to 2009 and the Encuesta Nacional de Hogares (ENAHO) from 2010 to 2014. Several improvements in the methodology to estimate household incomes were introduced in the ENAHO. For example, the new questionnaire includes more and better questions on labor and non-labor income; the consumption basket was revised using the consumption patterns found in the 2004 National Income and Expenditure Survey; and there were methodological changes in how to adjust for under-declaration of incomes and the imputation of labor income. Figure B2.1.1  Poverty Rates Adjusting the Non-Food Component of the Poverty Line with the Overall CPI, 2001–2009 35 30 Poverty rate (%) 25 21.6 20 18.7 15 10 2001 2002 2003 2004 2005 2006 2007 2008 2009 Poverty rate (official) Poverty rate (simulation) Source: Calculation based on INEC. box continues next page Poverty and Shared Prosperity 25 Box 2.1  continued Costa Rica uses two income-based poverty lines, one for extreme poverty and another for overall poverty. The extreme, or “food,” poverty line is the value of the basic food basket required for an individual to meet the minimum caloric intake. The overall poverty line considers other basic resources beyond food. A methodological change in calculating the poverty line was also introduced with the ENAHO. Before 2010, both components of the poverty line were adjusted using the food consumer price index (CPI). This was likely to have resulted in overestimation of poverty when food prices were rising faster than overall prices, as happened in 2007–2009. In 2009, for example, the poverty rate would have been 18.7 percent, about three percentage points lower, if the non-food component of the poverty line had been updated using the general CPI (Figure B2.1.1). Since 2010, the official poverty line is adjusted using the food CPI for the food component of the poverty line and the non-food CPI for the non-food part. Comparing Costa Rica’s poverty rates with those of other countries presents a series of challenges, but harmonized survey data are available for that purpose from the Socio- Economic Database for Latin America and the Caribbean (SEDLAC). The harmoniza- tion was developed by the Center for Distributional Labor and Social Studies (CEDLAS) at the University of La Plata, Argentina, and the World Bank. The harmonized survey data allow for cross-country comparability of poverty and inequality measures. However, per capita income and poverty lines from SEDLAC differ from official estimates. Extreme poor are individuals with a harmonized per capita income lower than US$2.50 a day, while poor are individuals with a harmonized per capita income lower than US$4 a day (2005 PPP). International comparable data from SEDLAC are available only up to 2012. Source: Adapted from Cadena et al. (2013) and INEC (www.inec.go.cr). Profile of the Poor: 35 percent in the Brunca and Chorotega regions, as compared with 19 percent in Increasingly Urban, Larger the Central region. In terms of their numbers, however, the Households, and Lower poor are concentrated in the Central Region. Educational Attainment About half of the 1.7 million poor Costa Ricans Spatially, poverty rates are the lowest live in the Central region, with the other half in the Central region, as compared to the distributed almost evenly across the rest of the other regions. Poverty rates vary signifi- country (­figure 2.2b). The relative high concent­ cantly across the country (figure 2.2a). In ration of poor in the Central region mirrors particular, poverty rates are as high as population density as this region concentrates 26 Poverty and Shared Prosperity Figure 2.2  Poverty Varies Widely by Region, but Most of the Poor Are in the Central Region a. Poverty headcount, 2001 and 2014 b. Concentration of the poor (%), 2014 50 45 39 7.4 40 11.6 36 9.5 35 31 32 31 Poverty rate (%) 29 30 28 29 11.1 12.1 25 25 25 20 19 15 13 10 48.3 5 0 Atlantica Brunca Central 2001 2014 Chorotega Norte Pacifico Central Central Atlantica Brunca Chorotega Norte Pacifico Central Source: Calculations using 2001 EHPM and 2014 ENAHO survey. Note: The sample of the household survey allows estimating poverty in the six regions of the country: Central; Chorotega; Pacifico Central; Brunca; Huetar Atlantica; and Huetar Norte. Poverty rates in 2001 and 2014 are not strictly comparable (see box 2.1). Poor are individuals with a per capita income below the official poverty line. 62 percent of the country’s now mostly urban con­ centrated in only five counties in Costa population. Rica in 2011 (figure 2.3b). Regional disparities are significant, and Poor and non-poor households have sig- poverty rates are much higher in the border nificantly different characteristics. Table 2.1 areas. The 2011 county-level map of income compares the profile of the poor and the (monetary) poverty produced by INEC re- non-­ poor, as well as those in the bottom 40 veals a large heterogeneity of poverty rates and top 60 percent of the income distribution within and across regions.3 In particular, in 2014. Non-poor households tend to have poverty rates range from over 50 percent higher levels of human capital: heads of house­ along the border with Nicaragua and Pan- hold had nine years of education on average ama to less than 15 percent in some coun- among the non-poor, compared with about ties located in the Central region (figure five years among poor and extreme poor 2.3a). The cantons with the highest poverty households. The poor and extreme poor be- rates are also those where most of the in- longed to larger households compared to the digenous population lives (box 2.2). Inter- non-poor. They were also more likely to live estingly, the poverty map also shows a high in households with larger dependency ratios concentration of poor households, with al- of younger children compared to the non-­ most a quarter of poor households poor; the proportion of individuals 12 years Poverty and Shared Prosperity 27 Figure 2.3  The Highest Poverty Incidence Is in Border Areas Source: Elaborated with data from the 2011 Census reported in INEC’s website (www.inec.go.cr). Note: The figure shows the proportion and concentration of poor households in 2011. Poor are households with a per capita income lower than the official poverty line. Since numbers refer to households instead of individuals, they differ from those shown in figure 2.1b. old or younger in poor households was twice Despite the modest reduction in income as large as that observed in non-poor house- (monetary) poverty reduction that the holds in 2014. country experienced during the first decade Income-generation capacity also varies. For of the 2000s, the 2011 Census revealed a example, the average per capita income of the significant reduction in the percentage of non-poor is 17 times higher than the income households considered poor from a multi- of the extreme poor and seven times higher dimensional measure, related mostly to im- than the income of the overall poor. The poor proved access to basic infrastructure and are more likely to work in the agricultural sec- services.4 Thus, the percentage of house- tor and to be self-employed, while the non-­ holds in multidimensional poverty—that is, poor tend to work more in services and to be who did not meet one or more of the four formally employed. The poor are also more basic conditions—dropped from 36.1 likely to live in female-headed households, but percent in 2000 to 24.6 percent in 2011.5 ­ non-poor women are twice as likely as poor The share of households who did not meet women to participate in the labor force. the “education” component of the multi-­ Indicators of “multidimensional pov- dimensional indicator fell from 15.2 to erty” improved between 2000 and 2011. 8.3 percent, as school attendance of 28 Poverty and Shared Prosperity Box 2.2  Indigenous People in Costa Rica: A Very Small and Disadvantaged Minority Costa Rica’s indigenous population is small, but occupies a vast share of the land. In the Latin American region as a whole, indigenous people make up about seven percent of the total population (over 36.6 million people). In comparison, in Costa Rica they represent 2.4 percent of the population (104,143 indigenous people). As such, Costa Rica, along with El Salvador, is the country with the smallest proportion of indigenous people in the entire region. In terms of land, legally established indigenous territories cover about seven percent of the Costa Rican territory (about 350 thousand hectares), but 91 percent of the indigenous population is concentrated in three cantons—Talamanca, Buenos Aires, and Hojancha (figure B2.2.1). In the first decade of the 2000s, the indigenous population experienced a similar demo- graphic transition as the rest of the Costa Rican population, with a falling share of chil- dren and a growing share of elderly adults. The dependency rate fell from 108 percent in Figure B2.2.1  Indigenous People Are Located Mostly in the Southern Region Source: Elaboration based on Census 2011. box continues next page Poverty and Shared Prosperity 29 Box 2.2  continued 2000 to 79.8 percent in 2011, showing the sharp decline in the share of children that re- sulted from a decline in fertility rates, from 9.6 to two between 2000 and 2011, together with an increase in the share of elderly adults, from four percent to 8.8 percent. Even if in- digenous households are still larger than non-­ indigenous (3.6 versus 3.4 people per house- hold) these differences have been falling over time. Historically, indigenous populations have had lower access to basic services than the non-­ indigenous. Despite a broad coverage of basic services to most of the population of Costa Rica, indigenous people still lag behind in access to electricity, water supply, and sewage. According to the 2011 Census, 80 percent of indigenous households had an electricity connection, against 99 percent of non-indigenous. Still, compared to other indigenous populations in LAC, access to electricity is higher than in Colombia (58 percent), El Salvador (62 percent), Nicaragua (50 ­ percent). As for access to clean water, 31 percent of the percent), or Panama (40 ­ indigenous population had no access to sewage (public or private), against four percent of non-​indigenous. And only 67 percent had access to piped water, against 94 percent of non-­ indigenous (­ figure B2.2.2). However, these indicators have improved significantly since 2000, as the 2000 Census reported that only 53 percent of households had access to clean water, 50 percent had access to some form of sewage, and 67 percent had electricity. In addition the share of households living in overcrowded conditions fell from 30 percent to 11 percent. Human capital accumulation and income generating opportunities are also lower for in- digenous populations. Indigenous people report lower educational attainment than non-­ indigenous: average years of education among the indigenous is 6.1, against 7.7 for non-indigenous (although this shows progress from the 3.9 years on average in 2000). Figure B2.2.2  Access to Basic Services Is Lower for Indigenous Communities Access to basic services, 2011 % of population in each group 120 93.5 99.4 100 75.9 80.5 80 67.2 59.1 60 40 30.5 20 23.6 20 10.4 8.2 3.9 4.1 2 0 Septic tank Sewage system None Well Natural source Piped water Sewage Water supply Electricity Indigenous Not indigenous Source: Elaboration based on Census 2011. 30 Poverty and Shared Prosperity Box 2.2  continued Illiteracy is at 7.7 among indigenous and 2.2 among non-indigenous, and attendance to formal education among people 5–24 years old is 64.9 against 71.7 for non-indigenous. This translates into lower employment and income opportunities among the indigenous, as they have a higher probability of working in self-employment or unpaid work, and to be doing household chores rather than studying, for those who do not have an income-​ generating activity (figure B2.2.3). Figure B2.2.3  Indigenous People Are More Likely to Work in Self-Employment, or to Be Inactive Indigenous people are more likely to work in self-employment or be inactive 80 72 70 % of population in each group 59.8 60 50 40 37.3 30 26.9 24.3 19.5 20 17 14.3 10.7 10 7.7 6 5.3 2.9 1 0 Wage/ Self-employed Unpaid worker House chores Studying Other inactive Retired salary worker Employed Inactive (55.5% indigenous; 49.8% non-indigenous) Indigenous Not indigenous Source: Elaboration based on 2011 census and Programa Estado de la Nación (2014). Table 2.1  Poor and Non-Poor Households Have Different Characteristics Bottom Top Characteristics Extreme Poor Overall Poor Non-Poor 40 percent 60 percent Household Characteristics Age of the head 48.8 50.9 50.1 50.1 50.4 Male-headed household (percent) 54.6 56.7 63.9 58.7 64.4 Proportion age 0–12 (percent) 23.7 21.2 11.5 19.9 10.2 Proportion age 13–18 (percent) 12.1 11.2 7.9 11.5 7.0 Proportion age 19–70 (percent) 57.9 56.8 72.7 58.6 75.2 Proportion age 70+ (percent) 6.4 10.7 7.9 10.0 7.7 Household size (number) 3.6 3.7 3.2 3.7 3.1 Education of household head (years) 5.4 5.6 9.2 5.9 9.9 Monthly per capita income (CRC) 24,845 59,174 431,636 82,124 500,020 table continues next page Poverty and Shared Prosperity 31 Table 2.1  continued Bottom Top Characteristics Extreme Poor Overall Poor Non-Poor 40 percent 60 percent Employment sector (percent) Primary sector 22.1 18.5 8.9 18.1 7.5 Manufacturing 10.6 12.2 11.4 12.8 11.0 Construction 7.9 8.3 5.7 8.7 5.1 Utilities 8.6 5.7 6.8 6.1 6.8 Retail 16.4 18.6 18.3 17.9 18.4 Services 20.8 25.4 42.9 25.2 45.8 Domestic services 13.1 10.9 5.8 10.9 5.0 Labor force (percent) Employee 29.9 47.3 75.1 56.6 76.3 Employer 0.8 1.1 4.3 1.5 4.7 Self-employed 28.6 25.4 14.2 21.8 13.6 Unpaid worker 3.3 1.9 0.7 1.4 0.7 Unemployed 37.4 24.2 5.8 18.7 4.7 Female labor force participation 26.7 32.1 55.6 34.5 59.8 Source: Calculations based on ENAHO. Note: Poor and extreme poor are individuals with a per capita income lower than the official poverty and extreme poverty lines, respectively (the national extreme and overall poverty lines were CRC 45,116 and CRC 96,565 in June 2014, respectively). children ages 7–17 increased and repetition population in 2001 to almost 47 percent by fell. At the same time, in 2011, 9.3 percent 2012, mirroring trends in the LAC region of households did not satisfy the “health” (figure 2.4a). As figure 2.4b shows, Costa condition, while 7.8 percent did not satisfy Rica’s middle class is among the largest in the “consumption” condition, and 6.2 per- LAC in percentage terms, surpassed only by cent did not satisfy the “housing” condition Argentina (54 percent) and Uruguay (57 per- (still, all three measures were lower than in cent). Still, about one-third of the population 2000). Increased access to piped water and (those with incomes between US$4 and sewage, increased access to electricity, and US$10 a day) remains vulnerable to falling a lower dependency ratio contributed to re- back into poverty if hit by a shock. ducing the share of households who did not Upward mobility has contributed to the meet these conditions. rise of the middle class. Trends in the share of poor and the middle class conceal movements Shared Prosperity and the in and out of poverty. Between 2003 and 2009 (a period for which comparable data exist for Rise of the Middle Class Costa Rica), increasing incomes resulted in Nearly half of Costa Ricans are classified important upward mobility. About 17 percent as “middle class.” The middle class, interna- of Costa Rica’s population (57 percent of the tion tionally defined as the share of the popula­ originally poor) escaped poverty during those that lives on US$10 to US$50 per day, has ex- years (figure 2.5).6 At the same time, however, panded significantly, from 32 ­percent of the about eight percent of the total population fell 32 Poverty and Shared Prosperity Figure 2.4  The Middle Class in Costa Rica Is One of the Largest in LAC a. Evolution of middle class in Costa Rica and LAC, 2001–2012 50 45 40 35 Middle class (%) 30 25 20 15 10 5 0 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 EHPM ENAHO Costa Rica LAC b. Class composition in LAC (percent of population), 2012 100 9 13 18 22 27 27 29 30 80 33 34 35 28 39 40 30 44 47 54 57 60 47 44 38 43 40 43 41 38 40 40 36 36 62 41 36 56 32 31 20 35 33 33 28 28 29 25 24 23 21 21 10 12 11 8 0 Guatemala Honduras El Salvador DR Colombia Mexico Ecuador Bolivia Paraguay LAC Peru Brazil Panama Chile Costa Rica Argentina Uruguay Overall Poverty Vulnerable Middle Class Rich Source: Calculations using SEDLAC data (CEDLAS and the World Bank). Note: Poor are individuals with a per capita income lower than $4 a day (2005 US$ Purchasing Parity Power per day). Vulnerable are those living with a per capita income between US$4 and US$10 a day. Middle class is defined as the proportion of individuals with an income between US$10 and US$50 a day. Rich are those with an income higher than US$50 a day. Regional estimates are population-weighted averages of country-specific estimates for Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, and Uruguay. SEDLAC international comparable data is available only up to 2012. Numbers are not strictly comparable before and after 2010 (see box 2.1). Poverty and Shared Prosperity 33 Figure 2.5  More People Moved Out of Poverty than Into Poverty 70 62.9 60 51.0 50 % of population 40 30 21.6 23.4 20 16.9 12.5 10 7.7 4.2 0 Always poor Poor to non-poor Non-poor to poor Never poor Costa Rica (2004–2012) LAC (2003–2009) Source: Vakis et al. (2014), who apply the parametric approach of Dang and Lanjouw (2014) to all countries in LAC using SEDLAC data (CEDLAS and the World Bank). Note: Poor are individuals with a per capita income lower than $4 a day. Estimates of poverty at the regional level are population-weighted averages of country-specific estimates for Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, and Uruguay. Country-specific figures in regional estimates come from 2004 or 2012 surveys, or the nearest year in cases in which 2004 and/or 2012 data are unavailable. Due to comparability issues, Costa Rica estimates come from 2003 and 2009 surveys. International comparable SEDLAC data is available only up to 2012. into poverty, including 11 percent of those who incomes of the poorest fell behind. Income were previously non-poor. Downward mobility growth for the bottom 40 percent fell to 1.6 was higher in Costa Rica than in the LAC re- percent over the period from 2007 to 2012, gion, while upward mobility was lower. About significantly lower than average income 23 percent of the total population of LAC (half growth in Costa Rica (3.4 percent) and of the originally poor) escaped poverty between among the lowest in LAC (figure 2.6b). 2004 and 2012, while only four percent (about This lower income growth for those in the two percent of originally non-poor) moved into bottom 40 percent of the population did poverty during those years. not allow many individuals to escape pov- Costa Rica experienced many years of erty after 2010, and poverty reduction shared prosperity, with the incomes of the stagnated. poorest 40 percent growing faster than the Before the crisis, even the poorest regions average. Income growth of the bottom 40 benefitted from shared prosperity. Between percent was six percent between 2003 and ­ 2003 and 2007 there was some income con- 2007, higher than the mean income growth vergence across regions, with the ­ incomes of of 3.9 percent. Moreover, the incomes of the bottom 40 percent growing at more than those in the bottom 40 percent also grew 10 percent in Brunca and Chorotega (the faster in Costa Rica as compared with their poorest regions in the country), against 3.4 counterparts in other countries in the LAC percent in Central region (figure 2.7a). Even region (figure 2.6a). in the Central region, incomes at the bottom However, after the global financial cri- 40 and at the average grew at similar rates. sis, shared prosperity declined and the figure 2.7b) However, after the global crisis (­ 34 Poverty and Shared Prosperity Figure 2.6  Shared Prosperity by Country in LAC Mean Income Growth (Overall Population) . Circ 2003–2007 14 12 10 Annu li d rowth r t (%) 8 6 4 2 0 –2 2004–2007 2003–2007 2003–2007 2004–2007 2003–2007 2003–2007 2003–2006 2005–2009 2002–2006 2004–2007 2003–2007 2003–2007 2003–2007 2003–2008 2003–2006 2000–2006 2004–2007 –4 –6 AR HD EC ES BR CR CH NI MX PE DR UR PY CO PA GU BO b. Circ 2007–2012 14 12 10 Annu li d rowth r t (%) 8 6 4 2 0 –2 2007–2012 2007–2012 2007–2012 2007–2012 2007–2012 2007–2012 2007–2012 2008–2012 2007–2012 2005–2009 2006–2011 2007–2012 2010–2012 2007–2012 2006–2012 2007–2011 2006–2011 –4 –6 BO PA PE UR PY BR AR CO EC NI CH DR CR ES MX HD GU Mean income growth (bottom 40%) Overall population Source: Calculations using SEDLAC data (CEDLAS and the World Bank). Note: Country-specific figures come from 2003, 2007, and 2012 surveys, or the nearest year in cases in which data for those years are unavailable. Due to comparability issues, Costa Rica estimates come from 2010–2012 surveys in panel b. International comparable SEDLAC data is available only up to 2012. Poverty and Shared Prosperity 35 Figure 2.7  Before the Crisis, Shared Prosperity Was High in Poorer Regions a. Before global crisis (2003–2007) b. After global crisis (2010–2014) 12 12 Annualized growth rate (%) Annualized growth rate (%) 10 10 8 8 6 6 4 4 2 2 0 0 –2 –2 –4 –4 ca ica a e a ca e l ica l l ra ra l rt eg ra rt eg ra un un nt nt nt No nt No nt ot nt ot Br la Ce Ce Br la Ce or Ce or At At Ch Ch co co cifi cifi Pa Pa Mean income growth (Bottom 40%) Mean income growth (overall population) Source: Calculation based on EHPM for the 2003–07 period and on ENAHO for the 2010–12 period. Note: The sample of the household survey allows estimating growth of incomes in the six regions of the country: Central; Chorotega; Pacifico Central; Brunca; Huetar Atlantica; and Huetar Norte. incomes of the bottom 40 percent grew more rose again, averaging 0.50 across the period. slowly. In particular, in the Central and Norte Inequality has risen in recent years, averag- regions, average incomes grew faster than in- ing 0.52 from 2010 to 2014, a time period comes at the bottom, reversing the pre-crisis that corresponds both to the post-crisis re- trend in shared prosperity. covery and to the period covered by the new ENAHO household survey (box 2.1).8 In contrast, the rest of Latin America Rising Inequality, in witnessed a marked decline in inequality, Contrast to Regional with the regional average falling by five points, from 0.57 in 2000 to 0.52 in 2012 Trends figure 2.8a).9 As a result, Costa Rica has (­ Income inequality has risen or stag- gone from being the least unequal country nated since the 1990s. Changes in house- in LAC after Uruguay in 2000, to being hold survey methodologies create problems around the median country by 2012.10 In for the comparability of measure of inequal- non-LAC OECD countries, income in- ity over time (box 2.1).7 Nonetheless, long- equality has been consistently much lower, term trends over different periods indicate with a Gini coefficient 0.30 in 1995–2012. that inequality has either risen or remained Rising inequality in Costa Rica is under- flat since the late 1980s. During the period pinned by the widening gap between the from 1989 to 2000, the Gini measure of in- earned incomes and skillset of rich and equality averaged 0.45, rising from 0.44 in poor workers in Costa Rica (see chapter 3). 1989 to 0.48 by 2000. During the period Similar to most countries in LAC, in- from 2001 to 2009, the Gini fell and then come is highly concentrated among the 36 Poverty and Shared Prosperity Figure 2.8  Evolution of Inequality and Income Distribution in Costa Rica and in LAC a. Gini coefficient, 1989–2012 b. Income distribution, 2012 0.60 70 0.55 61 Percent of total income (%) 60 0.50 51 Gini coefficient 50 0.45 0.40 40 0.35 30 22 22 22 0.30 20 15 0.25 10 5 0.20 2 0 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 Poor Vulnerable Middle Rich Class LAC Costa Rica (Official) Non-LAC OECD LAC Costa Rica Source: Calculations using SEDLAC data (CEDLAS and the World Bank) for LAC countries and OECD Stats for non-LAC OECD countries; and EHPM/ENAHO for Costa Rica. Note: Numbers for LAC are calculated using pooled data in Panel A and population-weighted country-specific averages for Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, and Uruguay in Panel B. Numbers for non-LAC OECD countries are un-weighted average of country-specific averages for non-OECD countries: Australia, Austria, Belgium, Canada, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, Korea, Luxembourg, Netherlands, New Zealand, Norway, Poland, Portugal, Slovak Republic, Slovenia, Spain, Sweden, Switzerland, Turkey, united Kingdom, USA. Poor are individuals with a per capita income lower than US$4 a day (2005 US$ Purchasing Parity Power per day). Vulnerable are those living with a per capita income between US$4-US$10 a day. Middle class is defined as the proportion of individuals with an income between US$10-US$50 a day. Rich are those with an income higher than US$50 a day. SEDLAC international comparable data is only available up to 2012. Inequality numbers are not strictly comparable before and after 2001 and before and after 2010 (box 2.1). rich in Costa Rica. About 22 percent of in- inequality fell since 2010 in all but the Central come is held by the rich in Costa Rica, a region, where the capital city of San Jose is lo- similar share to the average in LAC (figure cated. Thus, the Central region seems to have 2.8b). The top 20 percent of the population driven the overall increment of the Gini by one earns 54 percent of the income, while the point. The weight of the Central region in ex- share of income going to the middle class— plaining the overall Gini is confirmed by fi­ gure spanning the 48th to 96th centiles of the pop­ 2.9b, which shows that inequality in the Cen- ulation in 2012—amounts to 61 percent. tral region accounts for three-quarters of total The average income of the rich was 10.3 inequality in Costa Rica. times higher than that of the poor in Costa Rica, slightly higher than the LAC ratio Limited Poverty (9.5 times). Within the country, inequality varies across Responsiveness to Growth regions. Brunca, Norte, and Chorotega are the During the 2000s, poverty was not very regions with the highest income inequality, responsive to growth in Costa Rica. Indeed, while Pacifica and Atlantica are the least un- the elasticity of poverty to growth has been equal (­ figure 2.9a). It is interesting to note that low: between 2001 and 2007, a one percent Poverty and Shared Prosperity 37 Figure 2.9  Despite Regional Variation, Inequality in the Central Region Explains Most of Overall Inequality a. Regional differences in income inequality, b. Contribution of each region to total inequality 2010 and 2014 (%), 2014 0.56 0.60 0.55 6.2 0.54 4.4 0.52 0.51 0.51 0.55 0.50 0.51 0.51 0.51 0.49 Gini coefficient 0.48 0.48 6.1 0.50 0.45 3.6 0.45 5.3 0.40 0.35 74.3 0.30 Nacional Central Chorotega Pacifico Brunca Atlantica Norte Central Chorotega Pacífico Central 2010 2014 Brunca Huetar Caribe Huetar Norte Source: Calculations using 2010 and 2014 ENAHO survey. Note: The sample of the household survey allows estimating inequality in the six regions of the country: Central; Chorotega; Pacifico Central; Brunca; Huetar Atlantica; and Huetar Norte. increase in GDP per capita was associated fell, further widening the gap between the with only a 0.60 percent reduction in the of- poor and the rest of the population. ficial poverty rate.11 In comparison, the pov- Over time, rising inequality and price erty-growth elasticity in LAC over the same effects have offset the contribution of ­ period was much higher, reaching 1.43. After growth to poverty reduction. Changes in the global financial crisis, the growth-poverty income poverty stem from growth, inequal- elasticity decreased even further in Costa ity, and inflation. Inflation could affect pov- Rica, to 0.44 over 2010–13, whereas in LAC erty through its effects on real wages, cost it increased further to 2.0 in 2010–2012.12 of goods and ­ services, and the regressive This low ­poverty-growth elasticity in Costa nature of the inflation tax. Costa Rica’s suc- Rica was also observed for periods before cess in controlling inflation could suggest 2000, particularly during the 1990s.13 progress in poverty reduction. A decompo- The global financial crisis marks a clear sition of changes in poverty suggests shift- inflection point, after which the incomes in ing effects of these factors (figure 2.11). the wealthiest deciles grew significantly faster During the pre-crisis period, poverty fell by and the poor fell behind. Growth incidence 3.8 percentage points, helped by strong curves (figure 2.10) reveal that before 2007, growth and changes in inequality that posi- growth was largely pro-poor. Indeed, the in- tively affected the bottom end of the income comes of the bottom deciles of the income distribution. However, these gains were distribution grew faster than those in the top partially offset by rising prices. During the deciles. During and after the crisis, however, global crisis, growth slowed, food prices the incomes of those in the poorest deciles rose, and inequality worsened—so that the 38 Poverty and Shared Prosperity Figure 2.10  Growth Benefited the Poor Relatively More Only Until 2007 Annualized growth in mean income 10 8 Annualized growth rate (%) 6 4 2 0 –2 –4 –6 0 10 20 30 40 50 60 70 80 90 100 Deciles of income distribution 2001–2007 2010–2014 Source: Elaboration based on EHPM and ENAHO. Figure 2.11  Contributions of Growth, Prices, and Income Inequality to Poverty Reduction, 2001–2014 Overall poverty change 8 6 2.4 4 1.8 2.3 Percentage points 2 3.7 2.3 1.4 0 –2 –3.3 –3.0 –5.8 –0.3 –4 –6 –2.1 –8 –10 2001–2007 2007–2009 2010–2014 Growth Food Prices Price adjustment of poverty line Inequality Source: Calculations based on EHPM for official poverty measures from 2001 to 2009 and on ENAHO for official poverty measures from 2010 onwards. Note: The poor are individuals with a per capita income lower than the official poverty line (the overall poverty lines was CRC 96,565 in June, 2014). Poverty numbers are not strictly comparable before and after 2010 (see box 2.1). Poverty and Shared Prosperity 39 net effect was an increase in poverty of 2.6 during that period (2010–14). If income in­ percentage points during these years (2007– equality had not increased, growth alone 09). Finally, in the post-crisis recovery pe- would have decreased poverty by three per­ riod, rising inequality more than offset the centage points during the recovery of 2010– positive contribution of growth to poverty 14.14 Instead, poverty rose by 0.4 percentage reduction, such that poverty rose slightly points during that period. Notes 1. The rate of global extreme poverty, measured 7. The long-term trends in poverty and inequal- as the share of the population living with under ity are subject to caveats regarding income US$1.5 per day, is 1.4 percent (2012). See measurement due to several changes in sur- World Bank Poverty and Inequality Database. vey methodology across that time period. 2. Please note that poverty rates after 2010 are The main “breaks” in comparability of the not comparable to previous years due to meth- data series occur in 2001 and 2010. As such, odological changes. measurement of welfare across the three time 3. The small area estimation technique follows periods of 1989–2000, 2001–09, and 2010–14 Elbers, Lanjouw, and Lanjouw (2003) and is not strictly comparable. See box 2.1. combines information from surveys and cen- 8. Inequality based on official per capita income suses to produce estimates of monetary pov- presents a similar trend, decreasing from erty in small geographic areas. By applying 0.52  to 0.48 in 2000–2005 and increasing to this technique, INEC combined information 0.51 in 2009. After the global crisis, inequal- from the 2011 Census together with the 2011 ity increased marginally from 0.51 to 0.52 in ENAHO survey to produce poverty rates at 2010–2014. The ENIG (Encuesto Nacional de the county level. For more information on the Ingresos y Gastos) finds that income inequal- 2011poverty map visit INEC’s website  (www​ ity remained constant with a Gini of 0.534 .inec​.go.cr). after adjusting for methodological differences 4. The multidimensional poverty measure from in the definition of  income (email exchanges INEC captures four basic conditions that are with INEC’s staff http://www.inec​ .go.cr/). strongly related to monetary poverty: housing, 9. See World Bank (2014e). health, education, and access to other goods 10. LAC averages and comparisons cover 17 coun- and services (purchasing power). Each condi- tries for which internationally comparable tion has one or more indicators. A house­ hold data exist. See World Bank (2007) and World is considered poor if  it cannot satisfy one or Bank (2014e). more condition. 11. This elasticity is somehow larger when using 5. Morales (2013). the international poverty rate. 6. The analysis on the movement in and out of 12. Measured using a US$4 a day poverty line. poverty relies on a synthetic panel estimated 13. See World Bank (2007). using the methodology of Dang and Lanjouw 14. The decomposition follows Kolenikov and (2014). The synthetic panel results for poverty Shorrocks (2005), who extended the standard do not completely align with the national or Datt and Ravallion (1992) decomposition to in- international figures presented earlier, be- clude price changes. The effect of inflation dur- cause the sample is restricted to households ing 2007–2009 on the poverty rate is due mainly whose head is between 25 and 65 years of age. to measurement issues (see box 2.1). 40 Poverty and Shared Prosperity Inclusive Growth? Inequality, Jobs, 3.  and Skills Costa Rica’s inequality trends over the last two decades present a dilemma. The widening income gap between the top and bottom quintiles of the population reflects changes in the labor market and a mismatch between the patterns of growth versus the skills profile of the workforce. With weak educational outcomes, the Costa Rican labor force is not well adapted to a labor market that increasingly demands high skills, and unemployment has increased among the poor and low- skilled workers, particularly since the global crisis. The earnings gaps between rich and poor, and between the skilled and unskilled, have widened. Furthermore, despite Costa Rica’s ambitious So­ cial Compact, taxes and transfers have not proven to be effective in redistributing income to com­ pensate for these disparities. Labor Earnings and labor incomes.1 And this has also been the case for Costa Rica, where changes in labor Inequality incomes have been the driving force behind Rising inequality stands out as a pressing reductions and increases in poverty and in- challenge for Costa Rica, particularly in light equality over the past 15 years. Specifically, of its enduring commitment to a far-­ reaching increases in labor incomes contributed to a Social Compact. As discussed in chapter 2, 4.2 percentage point reduction of poverty the gap between the rich and poor has wid- and a 1.0 point reduction in the Gini coeffi- ened significantly since the global crisis. This cient during the pre-crisis period of 2000– follows two decades of rising or stagnating in- 07, but this trend was reversed during the equality—in stark contrast to the significant global crisis and changes in labor incomes decline in inequality in the broader Latin resulted in a 2.2 percentage point increase in America and Caribbean (LAC) region. These poverty and a one percentage point increase trends are particularly puzzling given Costa in the Gini coefficient during the period Rica’s long-standing investment in an ambi- from 2007 to 2009 (­ figure 3.1). Over the tious Social Compact and high social spend- post-crisis recovery period, changes in labor ing, as discussed in more detail in chapter 5. incomes have contributed to a slight increase Most inequality stems from differences in in poverty and an additional one-point in- earned labor incomes. In general, inequality crease in the Gini coefficient (2010–14). can come from many sources: differences in Moreover, most inequality derives from labor earnings, pension income and other differences in labor earnings in the private public sector transfers, and in the returns to sector. A decomposition of the Gini coefficient capital. Throughout Latin America and the in 2001, 2007, and 2012 reveals that labor in- Caribbean, changes in poverty and inequal- come explained over 80 percent of overall in- ity have been mainly driven by changes in come inequality (table 3.1).2 However, while Inclusive Growth? Inequality, Jobs, and Skills 41 Figure 3.1  Labor Income Contributed Significantly to Poverty and Inequality Changes, 2001–2014 a. Overall poverty change b. Income inequality change 3 2.6 0.015 0.012 0.010 Change in Gini coefficient 2 0.010 Percentage points 1 0.4 0.005 0 –1 0.000 –2 –0.005 –3 –0.010 –4 –3.8 –0.010 –5 –0.015 2001–2007 2007–2009 2010–2014 2001–2007 2007–2009 2010–2014 Labor Non-labor Total Labor Non-labor Total Source: Calculations based on the EHPM for the 2001-2009 period and on the ENAHO for 2010 onwards. Note: Poor are individuals with a per capita income lower than the official poverty line (the poverty line is C 96,565 in June, 2014). The figure presents the Shapley Decomposition of poverty and inequality changes (see Barros et al. 2006; and Azevedo, Sanfelice and Cong Nguyen 2012). Poverty and inequality numbers are not strictly comparable before and after 2010 (see box 2.1). Table 3.1  Changing Sources of transfers (social assistance) reduced inequality Inequality only marginally and only until 2007: they be-   2001 2007 2012 came regressive, contributing to increased in- equality by 2012.3 This confirms that larger Capital 2.97 3.83 6.47 increases in the top portions of the income Pensions 5.81 7.47 10.91 distribution (and in the public sector in par- Transfers −0.30 −0.61 1.23 ticular) in the second half of the 2000s played Labor Income (Public) 27.24 24.64 27.11 an important role in the increase of inequality, Labor Income (Private) 54.55 55.00 44.89 which social assistance programs were not Housing + Others 9.74 9.67 9.39 Source: Elaboration using data from SEDLAC (CEDLAS and The able to counterbalance. World Bank). Fewer Jobs for Poor the labor income share decreased over the 2000s to 72 percent, this decline was entirely and Unskilled Workers explained by changes in labor incomes in the Poor and unskilled workers are finding private sector, while the share of inequality limited job opportunities due to the long- stemming from differences in incomes from term shift in Costa Rica’s pattern of growth public sector jobs remained constant over that and the recent contraction following the period. Although the contributions of non-la- oriented global crisis. Costa Rica’s outward-­ bor income remain relatively small, the share economic development model has brought of inequality deriving from differences in re- about a significant shift in the demand for turns to capital and pension income doubled, skilled and unskilled labor. As discussed in from about nine percent to nearly 18 percent chapter 4, it has favored the development of over the period (table 3.1). Public sector high value-added sectors, such as 42 Inclusive Growth? Inequality, Jobs, and Skills electronics, medical devices, IT business Skill-­Biased Technological Change, have in- services, and so forth. In contrast, lower creased local demand for high-skilled labor value-­added sectors, such as construction, at a time when Costa Rica has a larger rela- domestic services, and agriculture have tive supply of low-skilled labor.4 High and grown more slowly—or even contracted. increasing labor costs, relative to other The growth of low-end manufacturing (such Central American countries, could explain ­ as textiles), construction, and agriculture the decrease in low-skilled job creation in has particularly slowed since the onset of lower value-added sectors, as discussed in the global crisis. Job creation has been mini- chapter 4. mal or even negative in sectors like con- This structural mismatch of skills and jobs struction, domestic services, and agri­ culture, has translated into rising unemployment and which employ over 50 percent of low-skilled falling demand for low-skilled labor, particu- workers (figure 3.2). At the other end of the larly since the crisis. Overall unemployment spectrum, sectors that employ mostly skilled climbed from 4.6 percent just before the crisis workers, such as financial services, real es- to 7.3 percent during the crisis. Even with tate, personal services, and others, are grow- economic recovery post-crisis, unemployment ing fast (though from a low base in the case has steadily risen to nine percent of the work- of sectors such as utilities or financial ser- force in 2014 (­figure 3.3). These trends reflect vices). Changing patterns in foreign direct a growing mismatch between the demand for investment (FDI) and trade, along with labor and the skills profile of the workforce. Figure 3.2  Sectors that Employ Mostly Unskilled Labor Have Created Few Jobs Agriculture Domestic service Construction Manufacturing Transport, storage and communications Retail, restaurants and hotels Electricity, gas and water Financial services and real estate Community, social and personal services –40 –20 0 20 40 60 80 Job creation, 2007–2013 (percentage change) Workers with primary complete or less, 2013 (percent) Source: Elaboration based on EHPM and ENAHO. Inclusive Growth? Inequality, Jobs, and Skills 43 As such, unemployment has hit the poor in the top two quintiles is less than five per­ particularly hard. The unemployment rate for cent. The differential in the unemployment the bottom quintile in Costa Rica exceeds 20 rate between the poor and non-poor is signifi- percent, far higher than for other income groups cant and increasing, reaching more than 15 (figure 3.4a). By comparison, unemployment percent in 2013 (figure 3.4b). In contrast, Figure 3.3  Unemployment Has Been Increasing Steadily Since 2007 10 9.0 9 8 Percentage of the workforce 7.3 7 6 6.0 5 4.6 4 3 2 1 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Figure 3.4  Unemployment Has Hit the Poor Particularly Hard a. Unemployment rate by quintile b. Unemployment rate and poverty Percent of labor force in each group 25 25 20 20 Percent of labor force 15 15 10 10 5 5 0 0 Q1 Q2 Q3 Q4 Q5 2007 2008 2009 2010 2011 2012 2013 Costa Rica (2013) El Salvador (2012) Poor Non poor Honduras (2013) Panama (2012) Source: Elaboration based on data from ENAHO (2013). 44 Inclusive Growth? Inequality, Jobs, and Skills ­ nemployment rates in the bottom quintile u the extreme poor, the opposite is true: 60 among other Central American countries (El percent or more of working-age adults in Salvador, Honduras, and Panama) are signifi- these groups are either inactive or unem- cantly lower—hovering around five percent—­ ployed. Put differently, less than 40 percent likely reflecting higher rates of informality in of these adults generate any form of labor their labor markets. Interestingly, higher un- income. Moreover, this situation has employment among the poor in Costa Rica is slightly worsened since 2007, when the not associated with an increase in the share of share of inactive plus unemployed was self-employed workers (a proxy for informal below 60 percent for the poor and slightly employment). above for the extreme poor. Furthermore, over 60 percent of poor Moreover, the labor market presents a and extreme poor adults do not generate bleaker picture for women than for men. Fe- any labor income. There is a stark contrast male labor force participation has been his- in terms of labor market activity between torically low in Costa Rica and is among the the poor and the non-poor in Costa Rica. lowest in the LAC region. By 2013, slightly As figure 3.5 illustrates, over 60 percent of over half (50.8 percent) of women of work- the working-age adults who are non-poor ing age were active in the labor market, and are employed, while about 30 percent are participation has fluctuated around 50 inactive and about four percent are unem- percent during the 2000s, increasing only ployed. In contrast, among the poor and marginally. Moreover, unemployment rates Figure 3.5  Fewer than 40 Percent of Poor and Extreme Poor Adults Generate Any Labor Income Labor market status and poverty 100 Share of working age population (%) 90 80 70 60 50 40 30 20 10 0 2007 2008 2009 2010 2011 2012 2013 2007 2008 2009 2010 2011 2012 2013 2007 2008 2009 2010 2011 2012 2013 Extreme Poor Poor Non-poor Unemployed Inactive Employed non-low earners Employed low earners Source: Elaboration based on EHPM and ENAHO. Inclusive Growth? Inequality, Jobs, and Skills 45 have been consistently above the average, Education: Not Making going from 6.9 percent in 2007 to 10.9 percent in 2013. As a result, the share of working-age the Grade, Especially for women who are actually generating labor in- come is below 40 percent. Poor women who the Poor are heads of household and immigrant Despite significant investment, the edu- women have particularly high unemploy- cation system has not adapted to provide the ment rates (box 3.1). economy—or skills needed for the changing ­ Box 3.1  Female Household Heads and Immigrants Face a Tough Labor Market In the last two decades, the Costa Rican population has seen a significant change in the composition of households, with an important increase in the share of households headed by women. Between 1990 and 2013, the share of households headed by women doubled from 18 percent to 36 percent. According to some studies, this increase implied a higher female participation in the labor market, especially among women who are low skilled and more likely to live in poverty. Indeed, the participation of female household heads in the labor force is much higher than the average female participation, although it is espe- cially high among non-poor household heads (figure B3.1.1a). Figure B3.1.1  Female Heads-of-Household and Single Mothers Have Higher Participation, and (for the Poor) Higher Unemployment a. Female participation rate, 2013 b. Female unemployment rate, 2013 100 25 22.4 22.0 Percent of WAP in category 83.6 Percent of labor force 80 74.8 20 69.4 in category 58.4 15 60 52.7 50.8 10.7 40 10 6.2 20 5 2.7 1.7 0 0 Non- Poor Non- Poor Total Non- Poor Non- Poor Total Poor Poor Poor Poor Single Female heads All Single Female heads All mothers female mothers female Note: WAP is working age population, 15–64 years old. 46 Inclusive Growth? Inequality, Jobs, and Skills Box 3.1  continued Poor women who are heads-of-household experience particularly high unemployment rates. Female heads-of-household who are poor, and in particular single mothers, have an unemployment rate of over twice the average for women (figure B3.1.1b). Given that these women are more likely to have low skills, their employment opportunities are fewer, and as a result they experience higher unemployment. Immigrant women have even higher unemployment rates. Immigrants, particularly from Nicaragua, represent a large share of the labor force in Costa Rica, in particular in low- skilled employment. As shown in figure B3.1.2, women born in Nicaragua and other Central American countries have an unemployment rate around twice the average for women, whereas for men the rate is below average. Figure B3.1.2  Immigrant Women Have a Much Higher Unemployment than All Other Groups Unemployment rates, by gender and country of birth (2013) 25 Percent of workforce in group 20.9 20 17.7 15 10.2 10 7.2 6.5 5.4 5.3 5 4 0 Men Women Men Women Men Women Men Women Native Nicaragua Other Central America Other immigrants Source: Gindling and Oviedo (2008); elaboration also based on ENAHO. to ensure opportunities for the poor. Educa- and bottom 40 percent of the population. tional outcomes are surprisingly weak, given Given the country’s level of development and the level of spending and development. Like high spending, its education system seriously many middle-income countries, Costa Rica underperforms in terms of retention (as evi- is finding that achieving basic levels of edu- denced by high school dropout rates) and cation, such as near universal literacy and quality (as demonstrated by weak test results). completion of primary school, is not enough Moreover, disparities in education out­comes to generate the skills needed to sustain the are large, putting the poor and bottom 40 demands of its evolving economy—or to percent at a significant disadvantage for promote inclusive opportunities for the poor learning and earning. Inclusive Growth? Inequality, Jobs, and Skills 47 Low educational attainment: Less than half of the cohort 25–29 years old had achieved a secondary or higher educa- overall and for the poor tion by 2010, further adding to the stock of Despite large public investment and impor- low-skilled adult workers, a legacy of the tant achievements, average school attain- 1980s crisis. Costa Rica’s educational attain- ment is still low in Costa Rica. The growing ment is lower than in peer countries in LAC investment in education in the country has (such as Chile and Panama), and significantly brought important gains in terms of literacy lower than in peer countries in Europe (such and primary school completion. By 2011, as Croatia and Lithuania), and far lower than the literacy rate for adults 15 years old and graduation rates in the Organisation for Eco- above had reached 97 percent, and the share nomic Co-operation and Development of adults 25 years and above who had no for- (OECD) (figure 3.7). mal education went from 21 percent in 1950 Inequities in educational attainment are to three percent in 2010. By 2010, 83 percent closely related to poverty. Figure 3.8a shows of adults had achieved at least a complete gross attendance rates by quintile for primary primary education (­ figure 3.6a). However, and secondary. Attendance in primary is the average level of education today is below higher at lower levels of income, which is what would be expected given Costa Rica’s positive in the sense that it is very high for GDP per capita and education spending. In- all children, even the poorest. However, deed, today, fewer than half of young adults ­ figure 3.8b shows that the share of children 18–29 years old—and less than 40 percent of who are over-age (that is, who are a year or the overall workforce—have completed sec- more older than the correct age for their ondary education (figure 3.6b). grade) is higher for children in poorer quin- Indeed, compared with other countries, tiles. Taken together, these statistics suggest dropout rates are quite high in Costa Rica. Figure 3.6  Costa Rica Achieved Substantial Gains in Primary Education Attainment, but Not in Secondary a. Educational attainment of adults 25+ years old b. Secondary completion by age groups, 2013 100 100 % with secondary complete or more % of adults (25+ years old) 80 75 60 50 46.5 38.1 38.7 40 32.8 33.6 25 20 0 0 [18–29] [30–39] [40–49] [50–65] [18–65] 50 55 70 75 90 00 10 60 65 80 85 95 05 20 19 19 19 19 19 19 19 20 19 19 19 20 Age groups Tertiary (any) Secondary (any) Primary (any) No education Source: Elaboration based on data from Barro and Lee (2010); and ENAHO 2013. 48 Inclusive Growth? Inequality, Jobs, and Skills Figure 3.7  Relatively Few Costa Ricans Complete Secondary School Share of the population aged 25–29 that has completed secondary school or higher (the residual dropped out before completion) Completion of secondary school or higher, ages 25–29 100 94 91 % of population ages 25–29 81 81 80 63 60 51 49 49 39 40 20 0 Si U a Ni er r rm nd ra lv ia Gu du a Sw Ja ia Fr ny Gr ce OE ce ua a Ch D ile ly lo ru Sp ia Ar pin a ge es rt a Br al il in a C M ep. Ce El don ico Ho rag a Cr ore n Ph na n n ica at ras a ap A m o re az n u th ti ilip m Po ntin ca ic al itz pa Pa ai m ost LA C Ita nt Sa es n b ug ng S l A ad Co Pe an ee a Ge erla Li oa ica R In ex em R m Ko of ic Do C bl pu Re Source: Elaboration based on Barro and Lee (2010). Figure 3.8  Education Outcomes Are Linked to Income a. Gross attendance, 2013 b. Age disparities in attendance, 2013 140 131 40 36 123 119 34 120 % of children in primary 108 106 109 106 104 99 29 100 93 30 28 25 80 21 20 20 60 13 40 9 10 8 20 0 0 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Primary education Secondary education Under-age children in primary education Over-age children in primary education Source: Elaboration based on ENAHO. that even though enrollment is high among enrollment rates in secondary are signifi- poor children, repetition rates are also high, cantly higher for children in richer quintiles. ing and therefore the quality of their learn­ There are large disparities in enroll- might be lower and their incentives to drop ment in day care and preschool, but these out after completing primary may be higher. even out by the time children enter pri- Consistent with that story, we observe that mary school. As ­ f igure 3.9a shows, by age Inclusive Growth? Inequality, Jobs, and Skills 49 four about 60 percent of children from At around age 12, some children begin to households with educated parents are at- drop out, and those from households with a tending a day-care center, as compared low “educational environment” (that is, chil- with 40 percent for children in households dren whose parents have low education) with medium education attainment, and have a much faster dropout rate than chil- about 20 percent for children from house- dren from more educated households (figure holds with low education.5 Attendance in 3.9b). In fact, a recent study of the factors preschool is also higher for children from behind human capital accumulation finds wealthier families: virtually all five to six that, although context variables, such as year olds from the top quintile of the pop- place of residence, explained a significant ulation attend preschool, as compared with share of the educational attainment inequal- 79 percent of those in families in the poor- ity in the 1990s, at present only household est quintile.6 Enrollment equalizes by the and individual level variables, such as paren- time children start first grade, and it re- tal education, explain such outcomes.7 In- mains close to 100 percent for all children come status also affects enrollment, with throughout primary, reflecting Costa Rica’s only 77 percent of teens aged 13 to 17 from long-standing guarantee of basic public households in the poorest quintile attending education. school as compared with 93 percent of those The transition from primary to second- from households in the richest quintile.8 ary school is particularly prone to high drop­ Dropouts are highest in seventh grade, out rates, especially among children from just after the completion of primary educa- households with weaker parental education. tion. For many children, this is linked to Figure 3.9  Parental Education Plays a Determinant Role in Children’s Education Investment a. Attendance in daycare centers and preschools, b. Attendance in formal schooling by by age and parental education* children’s age and parental education* 100 100 % of children in age group % of children in age group 80 80 60 60 40 40 20 20 0 0 0 1 2 3 4 5 6 6 7 8 9 10 11 12 13 14 15 16 17 Age (years) Age (years) High education Medium education Low education Source: Programa Programa Estado de la Nación (2013). * Parental education is defined as the average education level of all adults 18 years and older in the household. A “low” level corresponds to an average of less than primary; “medium” corresponds to complete primary and incomplete secondary; and “high” corresponds to complete secondary and above. 50 Inclusive Growth? Inequality, Jobs, and Skills their poor performance during primary transportation), costs (of school supplies school (repetition and low content learn- and others), and household chores (for ing), which affects their motivation to con- certain groups of girls). tinue during secondary. However, some Regional gaps in educational attainment children who drop out of the system also also exist, but they are slowly decreasing. As face barriers such as distance (lack of figure 3.10a shows, the median educational ­ Figure 3.10  Regional Disparities in Education Are Large, but They Are Decreasing b. Net enrollment in upper secondary, by district groupings according to secondary completion rates (Q1 = lowest 20%; Q5 = highest 20%) 50 45 40 35 30 25 20 15 10 5 0 District Q1 District Q2 District Q3 District Q4 District Q5 Net enrollment rate in upper secondary 2000 Net enrollment rate in upper secondary 2011 Source: Elaboration based on Census 2011. Source: Elaboration based on Trejos and Saenz (2012). Knowledge gap 3.1  Why Are So Many Kids Dropping Out of School in Costa Rica? The skill-biased development that Costa Rica has experienced has raised returns to education, particularly secondary and tertiary. Given a large secondary school network in the country and universal primary completion, it is therefore surprising that more than half of the children who finish primary fail to complete secondary education. Although some general factors are known (see figure K3.1.1), such as lack of motivation, distance (in rural areas), and gender bias, the specific factors that push children out of school are still unknown, as are the critical ages at which early signs of dropout can be detected. continues next page Inclusive Growth? Inequality, Jobs, and Skills 51 Knowledge gap 3.1  continued Figure K3.1.1  Main Reasons for Being Out of School 100 10 12 14 22 % of secondary -age out of 9 10 80 38 17 13 8 7 school children 10 60 16 22 32 13 10 37 40 36 16 20 46 38 23 23 17 0 Guatemala Honduras Nicaragua Costa Rica El Salvador 2011 2013 2009 2013 2013 No money Not interested Has to work Family Other Source: World Bank (forthcoming 2015). attainment of adults 15 years and older is up enrollment in upper secondary is below to twice as high in the Central Region (dark 50 percent. green in the map) as in the peripheral dis- tricts around it (dark red districts). A rank- Poor quality of education ing of all districts in Costa Rica according to The quality of education in Costa Rica is also the share of adults 17–21 who have com- very low, given its level of spending. At the pleted secondary education shows that there primary school level, the latest results from is a 42-percentage point gap between the the TERCE examination of Latin American completion rate for the bottom and the top students by UNESCO show that Costa Rica is 20 percent of districts in terms of completion the only country in the sample where stu- (26.3 percent against 68.2 percent). A similar dents in both third and sixth grades per- gap is seen in terms of net upper secondary formed worse in TERCE (2013) than they enrollment (­ figure 3.10b) across districts, al- did in SERCE (2006) in reading and in math- though the gap has been closing in recent ematics.10 Similarly, the performance of Costa years, as low-­performing districts have in- Rican students at the secondary school level creased their enrollment rates faster than is also weak. Speci­fically, the performance of high-performing ones.9 Notice, however, that 15-year-old students on the international even in the best-­ performing districts net PISA mathematics exam is far lower than 52 Inclusive Growth? Inequality, Jobs, and Skills Figure 3.11  Costa Rica’s PISA Performance Is Below Expectations Given Its Secondary Education Spending 600 Singapore Hong Kong SAR, China Korea, Rep. 550 Switzerland Netherlands Estonia Poland Finland Belgium PISA 2012 Math score Australia Austria Ireland Czech Republic 500 New Zealand France Denmark Iceland United Kingdom Latvia Norway Slovak Republic Italy Spain Portugal United States Lithuania Sweden Hungary Israel Serbia 450 Romania Bulgaria Thailand Chile Malaysia Mexico Costa Rica 400 Brazil Colombia Indonesia 350 0 10 20 30 40 50 Secondary education spending per student (% GDP per capita), circa 2009 Source: Elaboration based on data from Barro and Lee (2010) and UNESCO. would be predicted by Costa Rica’s spending Widening Earnings Gaps on secondary education (figure 3.11).11 More- over, since these tests are taken only by those for Poor and Unskilled who remain in school by age 15, and since dropout rates are so high in Costa Rica, the Workers average skill levels and learning outcomes of Wages are relatively high across the all 15 year olds (including those in and out- board—even in low value-added sectors. side school) are likely even lower. As discussed in chapter 4, Costa Rica has rel- Moreover, there are large disparities in atively high wages compared to Central learning outcomes by socioeconomic sta- American neighbors, in both agriculture and tus. UNESCO’s SERCE study also shows low-­tech industry (light manufacturing)— that ­children from higher socioeconomic two sectors that traditionally employ low- backgrounds obtain better scores than skilled workers. Surprisingly, neither rising their peers from lower socioeconomic rates of unemployment nor the continuous backgrounds (­ figure 3.12).12 influx of low-skilled migrant workers from Inclusive Growth? Inequality, Jobs, and Skills 53 Figure 3.12  Large Disparities in Learning Outcomes by Socioeconomic Status 0.3 0.2 Standardized scores 0.1 0 –0.1 –0.2 –0.3 –0.4 –1 –0.5 0 0.5 1 1.5 2 Index of socioeconomic conditions 6th grade 3rd grade Source: Elaboration based on UNESCO (2006). Knowledge gap 3.2  Are Reservation Wages High in Costa Rica? The increase in unemployment rates, especially among poor workers, and the marginal changes in self-employment, a proxy for informal employment, is a puzzling finding. Most LAC countries respond to a decrease of jobs in the formal sector with a surge in informal employment, in the absence of a safety net that enables workers to look for other formal jobs. In Costa Rica, however, this did not happen, even if there is no clear evidence that the social safety net provides sufficient income support to those who are unemployed. Thus, the ques- tion arises of how high reservation wages are in Costa Rica, in a context of high average wages and a high cost of living relative to other countries in the region.a a. For evidence on the cost of living in Costa Rica by international standards, see Programa Estado de la Nacion (2014). Nicaragua (box 3.2) have put down­ ward workers in formal salaried jobs, especially in pressure on low-skilled wages. Moreover, the low-skilled jobs.13 And while changes in min- minimum wage does not appear to be driv- imum wages over the 1980s–1990s did result ing artificially high low-skilled wages. Mini- in a small reduction in employment, this ef- mum wages are binding for most Costa Rican fect was minor, and there were no effects on 54 Inclusive Growth? Inequality, Jobs, and Skills Box 3.2  Nicaraguan Workers in Costa Rica Do Not Appear to Drive Down Native Workers’ Wages Immigrants, particularly from Nicaragua, represent a large share of the labor force in Costa Rica, in particular in low-skilled employment. In 2013, over 10 percent of those employed in Costa Rica were foreign-born, of which eight percent were Nicaraguan. Nicaraguan immigrants in Costa Rica are largely unskilled: less than 20 percent com- pleted secondary school while over a third (35 percent) did not complete primary school.a However, Costa Ricans also have low schooling on average—in fact, 46 percent of both Nicaraguan and native-born adults finished primary school but did not complete secondary. This significant overlap in educational attainment suggests the potential for competition between these two groups for low- or unskilled jobs. This can be particularly true in specific sectors and specific parts of the country due to the non-random disper- sion patterns of immigrants. Four sectors of employment—agriculture, domestic service, commerce and hotels/restaurants, and construction—account for 75 percent of low- skilled immigrant employment in 2013. For context, these same four sectors accounted for only 59 percent of low-skilled native-born employment. In some parts of the country, foreign-born workers account for significant shares of workers in these four sectors, including over 20 percent of all workers in domestic services and construction in Huetar Norte and the Central regions. The presence of immigrant workers has little bearing on the wages of low-skilled Costa Rican workers—and is actually correlated with higher wages among higher-skilled Costa Ricans. Exploiting the sectoral and geographical differences in immigrant employment in Costa Rica, log wage regressions reveal positive correlations between native wages and the proportion of ­foreign-born workers in the same employment category as natives, where the employment category is defined at the region and sector level. This suggests positive complementarities between immigrant labor and skilled native labor. This is a sizeable relationship, with native workers who completed secondary receiving a wage premium of 22 percent at the average share of immigrants for this group (8.3 percent).b At the same time, no statistically significant relationship is evident between unskilled native workers and immigrants—suggesting that wages of native workers are not being lowered by immigrant labor. Similar analysis found no evidence of increased unemployment among native labor in regions with more immigrants. Source: Elaboration based on Census 2011 and ENAHO (2013). a. These education rates are calculated for the adult population 20 and above using the 2011 Costa Rica Census. b. Note that this is not necessarily a causal relationship—for example, it could be that immigrant workers are more likely to migrate to higher-paying regions and/or work in higher-wage sectors resulting in a higher share of immigrant workers in these higher-paying sectors. Inclusive Growth? Inequality, Jobs, and Skills 55 wages or employment in the uncovered sec- Even though earnings have grown in real tors (self-­employment). More recently, terms for low-skilled workers, they have changes in the minimum wage have been fallen in relative terms to high-skilled work- very modest, mostly correcting for inflation, ers. As figure 3.14 shows, relative to the me- and the evolution of the real minimum wage dian earnings level, between 2007 and 2013 has closely tracked the evolution of labor workers in the top quintile increased the productivity (figure 3.13). Moreover, a study ratio of their earnings to the median earner of the increase in the enforcement of mini- by 0.19 (from 2 to 2.19). At the same time, mum wage policy shows that it had a small workers in the bottom quintile decreased positive effect on wages, especially where the the ratio of earnings to the median by 0.23 minimum wage is most binding (at lower lev- (from 0.52 to 0.29), while workers in the els), but it did not affect employment in any second to fourth quintile kept the ratios to segment of the ­ distribution.14 Rather, it ap- the median almost flat. A similar picture is pears that high income levels, coupled with seen when income groups are divided ac- generous social benefits, have resulted in cording to workers’ education. For workers high reservation wages across the board in with tertiary education, the ratio to the me- Costa Rica. dian income increased from 1.89 to 2.01 be- Despite relatively high wages even for tween 2007 and 2013, for workers with low-skilled workers, the earnings gap be- secondary complete and secondary incom- tween rich and poor workers is widening. plete it remained flat, and for workers with Figure 3.13  Real Minimum Wage Changes Closely Follow Changes in Labor Productivity Labor productivity and mimimum wage indices (1991–2013) 150 140 130 120 110 100 90 00 08 06 09 03 05 04 02 07 01 10 98 96 99 93 95 94 13 92 12 97 91 11 20 20 20 20 20 20 20 20 20 20 20 20 20 20 19 19 19 19 19 19 19 19 19 Real minimum wage index (1991=100) Labor productivity index (1991=100) Source: Elaboration based on BCCR and IMF. 56 Inclusive Growth? Inequality, Jobs, and Skills Figure 3.14  Earnings Gaps Across Education and Income Levels Are Widening a. Relative earnings by quintile (ratio to median) 2.5 2 1.5 Ratio 1 0.5 0 2007 2008 2009 2010 2011 2012 2013 Q5 Q4 Q3 Q2 Q1 b. Relative earnings by education (ratio to median) 2.5 2.0 1.5 Ratio 1.0 0.5 0 2007 2008 2009 2010 2011 2012 2013 Any tertiary Secondary complete Secondary incomplete Primary complete or less Source: Elaboration based on data from EHPM and ENAHO. Inclusive Growth? Inequality, Jobs, and Skills 57 primary or less the ratio fell from 0.79 to of incomes for individuals in the poorest 0.73 of the median income. This resulted income decile (­ figure 3.15a) and the richest in a widening of the gap between the high- figure 3.15b), divided according to decile (­ est-paid and the lowest-paid workers, the educational attainment (skills level) of which is at the root of the increase in in- the house­ hold head. figure 3.15a shows equality that occurred in the second half of clearly that before 2007, the incomes of in- the 2000s. dividuals in the poorest decile, whose head Indeed, the incomes of the poorest and of household had fewer than eight years of those with the lowest skills grew faster be- education, grew faster than the incomes of fore the crisis—and fell the most since the those individuals (also in the poorest figure 3.15 plots the evolution crisis. ­ decile) whose household head had more Figure 3.15  Incomes Evolved Very Differently at the Top and the Bottom Before and After the Crisis a. Per capita income index for 1st decile (2001=1 and 2010=1) 1.8 1.6 1.4 Income index 1.2 1 0.8 0.6 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 b. Per capita income index for 10th decile (2001=1 and 2010=1) 1.3 1.2 1.1 Income index 1 0.9 0.8 0.7 0.6 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Education [0,8] Education [9,13] Education [14+] Source: Elaboration based on data from EHPM and ENAHO. 58 Inclusive Growth? Inequality, Jobs, and Skills education. Moreover the incomes in the conditional and unconditional cash trans- poorest decile also grew much faster than fers); and indirect taxes and transfers (such the incomes of individuals in the richest as value added taxes, and fuel subsidies).18 decile, which actually fell between 2001 and In Costa Rica, direct and indirect transfers 2007. In contrast, starting in 2008 low in- and taxes decrease income inequality by comes begin to decline, and high incomes three Gini points, from 0.51 to 0.49, a re- begin to rise, which becomes more appar- duction comparable to Brazil, Mexico, and ent after 2010.15 Interestingly, educational Uruguay, and larger than the reduction in gaps appear to matter more at the bottom Bolivia and Peru. The evidence available of the distribution before 2007, suggesting for these six countries suggests that fiscal that income increases were the highest policy in LAC is not able to counterbalance among the ­ lowest-skilled and poorest work- the inequality pressures from labor ers in the pre-crisis period. income. On the other hand, the experience from Redistribution Has Not OECD countries shows that redistributive policy has the potential to reduce inequal- Offset Rising Inequality ity. As figure 3.16b shows, non-LAC OECD Finally, redistributive policies have and select LAC countries, including Costa not been effective in compensating for in- Rica, have similar levels of market income equities associated with labor earnings and inequality before any fiscal intervention. skills. Redistribution policies have had a The Gini coefficient hovers around 0.4 and modest effect on overall inequality. The 0.5 for all these countries. However, there challenges facing Costa Rica and other are significant differences in the impact of LAC countries for reducing poverty in the transfers (direct taxes on income, payroll future highlight the importance of ensur- taxes, and direct transfers) between the ing that fiscal policy has a strong equity LAC group and the OECD. The Gini that focus and redistributes the gains from results after taking into account direct taxes growth among those who are less well off.16 and transfers in the OECD group is some- The Commitment to Equity Project (CEQ), where around 0.3, while it remains almost a joint initiative of Tulane University and unchanged at around 0.5 in Costa Rica and the Inter-American Dialogue, has devel- the rest of the LAC group. According to oped a harmonized methodology to assess World Bank (2014e), the low tax revenue the distributional impact of taxation and and high reliance on indirect taxation in public expenditure throughout Latin LAC are important reasons for these differ- America.17 Figure 3.16a shows the gradual ences. This is also the case of Costa Rica. If impact of fiscal policy in six LAC countries compared with non-LAC OECD countries, including Costa Rica. Fiscal policy is de- fiscal revenues are low and rely heavily on composed into direct taxes (such as in- indirect taxes, which are neutral in relative come and payroll taxes); direct transfers terms and therefore tend to reduce the pro- (such as non-­ contributory pensions, and gressivity of the overall tax system.19 Inclusive Growth? Inequality, Jobs, and Skills 59 Figure 3.16  Inequality for Different Income Concepts, Circa 2010 a. Gini coefficient in Costa Rica and five selected LAC countries 0.58 0.60 0.56 0.55 0.54 0.55 0.51 0.51 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.49 0.49 0.49 0.49 0.49 0.49 0.49 0.48 0.48 0.50 0.46 0.46 0.45 Gini Coefficient 0.40 0.35 0.30 0.25 0.20 Bolivia Brazil Costa Rica Mexico Peru Uruguay Market income Net market income Disposable income Post-fiscal income b. Comparison with non-LAC OECD countries 0.60 0.55 0.50 Gini coefficient 0.45 0.40 0.35 0.30 0.25 0.20 ug s h m y ak F i n b l i c ite Is gal ov d p u rk p d Au ium x e ed a N e e r urg Sw er ny er s w Ko d a a to d a Ca Italy Au e l a a st nd Ja lia Gr pan n g in rt m Co Me ay Pe a Br ile Fr nd Po ce il d S ce ru lg ic a o n St el Bo ica Cz D or ia az Ur ate itz land ni e c en wa L u Sw s t r i I r ad i Ze re st xic Es lan Sl lan Re l a n n m e Be ubl d ra liv N en K i pa Po d o Re a Ch ra an ee u t h ma la la R u G bo n e Ic Ne ite Un ov Un Sl Gini for Disposable Income (Non-LAC OECD) Gini for Market Income Gini for Disposable Income (LAC) Source: Sauma and Trejos (2014) for Costa Rica and World Bank (2014e) for the rest of the countries. World Bank (2014) results are based on Lustig and Pessino (2014), Paz Arauco et al. (2014), Higgins and Pereira (2014), Scott (2014), Jaramillo (2014), Bucheli et al. (2014). Note: All Gini are computed using 2009, except for Mexico and Costa Rica, which is 2010. The figure shows the Gini coefficient for different income definitions. Market Income is the income received before any fiscal intervention. Net Market Income subtracts direct income and payroll taxes to Market Income. Disposable Income is Net Market Income plus direct transfers. Post-Fiscal Income adds indirect subsidies and subtracts taxes from disposable income. Final Income is Post-Fiscal Income plus in-kind public transfers on health and education (World Bank 2014e). 60 Inclusive Growth? Inequality, Jobs, and Skills Notes 1. World Bank (2014e). 15-year-olds globally. See www​ .oecd.org​ 2. This follows Trejos and Oviedo (2012), who /­pisa. find that labor income for skilled workers, 12. World Bank (forthcoming 2015). followed by income for employers, capital 13. Salaried employment (which is mostly for- returns, and contributory pension income mal in Costa Rica) represented 62.4 percent explained 85 percent of inequality in 2009. of employment among workers living in 3. According to Trejos and Oviedo (2012), the poverty, and 78.7  percent among ­ non-poor share of the Gini explained by social assistance workers in 2013, whereas 33  percent of went from −0.3 in 2001 to −0.9 in 2009. poor workers were self-­ employed, against 4. See World Bank (2006a) and Lücke (2013). 16  percent of ­ non-poor. Hence, the mini- 5. Education attainment of the household is mum wage policy applies to the large major- defined as the average education level of all ity of workers. In the late 1990s, almost adults 18 years and older in the household. A three-quarters of all workers’ earnings were “low” level corresponds to an average of less concentrated around the five main minimum than primary; “medium” corresponds to com- wage categories (out of the total 19 different plete primary and incomplete secondary; and wages). See Gindling and Terrell (2004). “high” corresponds to complete secondary and 14. See Gindling, Mossaad, and Trejos (2013). above. See Estado de la Nación (2013). 15. Due to the methodological change between 6. Estado de la Nación (2013). EHPM and ENAHO it is not possible to con- 7. Trejos and Murillo (2012). struct a single income index for the entire 8. Estado de la Nación (2013). period. 9. See Trejos and Saenz (2012), and Trejos and 16. World Bank (2014e). Murillo (2012). 17. Important caveats and assumptions need to 10. Bilagher (2014). TERCE (the Third Regional be considered when analyzing results from Comparative and Explanatory Study) is a large-­ the CEQ project. The approach is static and scale study that assesses the performance of does not take behavioral responses into pupils in third and sixth grades of primary school account. On the expenditure side the method in mathematics, reading, and writing (language), focuses only on transfers and expenditures plus natural sciences in the case of sixth grade. in health and education and does not take Fifteen LAC countries took part in 2013, plus the into account other infrastructure ­ spending, Mexican state of Nuevo León (Mexico). SERCE or the quality of the services delivered. second study with the same ­ is the ­ characteristics Finally, the approach does not consider the that took place in 2006. See http://www.unesco​ externalities of public expenditures (World .org/new​ /­e n/santiago/­e ducation/education​ Bank 2014e). -assessment-llece​/­third-regional​-­comparative 18. This report presents results for Costa Rica -­and-explanatory-study-terce. (Sauma and Trejos 2014) together with the 11. The Programme for International Student first round of CEQ results from six countries Assessment (PISA), a triennial international (Argentina, Bolivia, Brazil, Mexico, Peru, and survey, tests the skills and knowledge of Uruguay) published in World Bank (2014e) and 15-year-old students. Around 510,000 stu- based on Lustig and Pessino (2014), Paz Arauco dents in 65 economies took part in the PISA et al. (2014), Higgins and Pereira (2014), Scott 2012 assessment of reading, mathematics (2014), Jaramillo (2014), Bucheli et al. (2014). and science representing about 28  ­ million 19. Sauma and Trejos (2014), World Bank (2014e). Inclusive Growth? Inequality, Jobs, and Skills 61 The Nature of Costa Rica’s Growth and 4.  Its Constraints Costa Rica’s small and open economy has achieved solid growth with an outward-oriented model combined with macroeconomic and political stability. This model has encouraged export diversifi­ cation towards higher value-added sectors and increasing labor productivity. However, the economy is showing signs of built-up vulnerabilities, resulting in weak job creation and loss of competitive­ ness. How can Costa Rica achieve higher levels of growth? Evidence points to three priority areas: education & skills, infrastructure, and government effectiveness and regulatory quality. Costa Rica experienced solid economic More broadly, GDP per capita did not growth in the past 25 years, as compared to make much headway in convergence with Latin America and the Caribbean but not the United States (figure 4.3). GDP per with other countries. Its real growth averaged ­ capita grew by 83 percent to US$ 5,839 4.7 percent since 1990, above the Latin (constant, 2005 prices)—higher than the America average of 3.1 percent (figure 4.1). ­ average of countries in LAC (48 percent) During 2003–2007, Costa Rica’s GDP growth and the world (35.5 percent). In terms of was particularly strong at 6.7 percent on av- Latin American per capita income ranking, erage, well above the regional average for Latin Costa Rica is the seventh richest country America of 4.6 percent. During the crisis, of the region (in 2005 constant US$), just growth slowed to 1.2 percent in 2008 and contracted by one percent in 2009. Following Figure 4.1  Costa Rica GDP Growth the global crisis, Costa Rica recovered simi- Averaged 4.7 Percent During larly to a set of peer comparator countries 1990–2014 (Chile, Croatia, Dominican Republic, Lithua- nia, Panama, and Uruguay) (figure 4.2). How- Annual GDP growth ever, this recovery has been slower than that 10 of upper-middle­ -income countries or com- 8 parator countries in Latin America and the 6 Caribbean (LAC). Growth decelerated in Percent 2013 to 3.5 percent and remained stable in 4 2014. The outlook is that growth will not ac- 2 celerate in 2015, amid ­ deteriorating fiscal bal­ ances and a rising public debt-GDP ratio, but 0 it is expected to pick up afterwards. The recent –2 closure of Intel’s microchip factory is expected 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016f to dampen growth by about one-half to three-­ quarters percentage points in 2014–15.1 Source: World Development Indicators and GDP forecast. The Nature of Costa Rica’s Growth and Its Constraints 63 Figure 4.2  Growth Compares Well with Respect to LAC 7 6 5.5 4.9 Average GDP growth (%) 5 4.5 4 3 2 1 0.9 0 1990–1999 2000–07 2008–09 2010–13 Costa Rica LAC Upper MICs Structural peers LAC peers Source: Calculations based on World Development Indicators. Figure 4.3  Lagging Behind in Converging Towards U.S. GDP per Capita GDP per capita (constant 2005 US$, % of U.S.) 22 20 18 % of U.S. GDP per capita 16 14 12 10 8 2000 2008 2006 2004 2002 1990 2010 1998 1996 1994 1992 2012 Chile LAC Panama Costa Rica Uruguay Source: Calculations based on World Development Indicators. 64 The Nature of Costa Rica’s Growth and Its Constraints Figure 4.4  Growth Volatility Is Low by International Standards 2.5 Coefficient of variation of GDP Growth, 2000–13 CRO 2.0 1.5 MEX LIT PAR OECD URU 1.0 SAL BRZ NIC HON CRI DOR 0.5 LAC CHL Middle income PAN Central America ECU PER GUA COL World 0 0 1 2 3 4 5 6 7 8 Average GDP Growth (%), 2000–13 Source: World Development Indicators. as it was back in 1960 and 1990. Growth points a year to 1.9 percentage points (figure has displayed low volatility in international 4.5). Over the 2010–2014, the contribution comparisons during 2000–2013 (figure 4.4). of labor further declined to 1.1 percentage points. Although growth accounting is a de- Growth Analysis and scriptive methodology and, by definition, does not provide specific insights into the Trends factors that underlie TFP growth, economists When examining growth by factor of have consistently relied on measures of the production, the contribution of capital has growth “residual” TFP as a gauge of a coun- been relatively stable in the past two decades. try’s technological change and innovation. However, the mid-2000s saw a sharp decline In this light, TFP growth reflects a realloca- in the contribution of Total Factor Produc- tion of inputs to the sector of the economy tivity (TFP). Between 2000 and 2004, TFP with higher productivity gains. Moreover, declined strongly and even turned negative, this is consistent with observed increases in dragging overall growth by about 1.7 per- labor productivity, measured as output per centage points a year (down from the positive worker (figure 3.13 in chapter 3 and figure 4.6). contribution to growth of about 0.9 percent- Weakened investment and net exports age points a year over the 1995–1999 period). have dragged growth in recent years. In TFP recovered over the 2005–2009 period, the mid-2000s, the external sector had a pos- but employment growth declines resulted in itive contribution to growth from an a sizeable reduction in labor’s contribution ­ aggregate demand perspective, as exports to output growth from about 3.2 percentage outweighed imports (figure 4.7). With the The Nature of Costa Rica’s Growth and Its Constraints 65 Figure 4.5  Declining Contribution Figure 4.6  Labor Productivity of Labor to GDP Growth Increased in All Sectors, and Workers Moved to Services Average percentage points contribution to GDP growth 6 Average percentage points change, 1991–2014 2.0 0.9 0.9 0.9 0.2 1.5 0.2 4 3.2 1.3 2.6 2.5 1.0 1.0 1.9 1.1 2 0.5 0.9 0.41 1.8 1.9 1.8 1.9 1.8 0.0 –0.69 0 –0.5 –1.7 –0.4 –1.0 –2 –1.5 Output per Employment worker change: 1.84 rate change: –0.22 Intersectoral reallocation Activities not adequately defined –4 effect 1990–94 1995–99 2000–04 2005–09 2010–14 Services Industry Capital Labor TFP Agriculture Source: Calculations with data from BCCR. Labor adjusted for Source: Results from a Shapley decomposition of growth using education levels. data from World Development Indicators and BCCR. global crisis, however, exports declined renewed dynamism to the economy due to sharply without an accompanying strong de- new investments and expansion in the provi- cline in imports. Softer economic growth in sion of services.3 The tourism sector also the United States (Costa Rica’s main trading showed healthy growth of eight percent a partner) as well as in China have weakened year in both 2013 and 2014, though its con- demand, and the recent closure of Intel’s mi- tribution to overall GDP remains small (ac- crochip factory in 2014 has further reduced counting for about 4.6 ­ percent of GDP). exports. Intel’s electronics exports alone ac- However, manufacturing, agriculture, counted for 20 percent of merchan­ dise ex- and construction have reduced their contri- ports. The lost dynamism in exports has been 2 bution to GDP growth. The manufacturing exacerbated by rising imports outside free sector has experienced a loss of competi- trade zones (FTZs). The contribution of in- tiveness due to exchange rate appreciation, vestment has also declined, lowering overall growing wages for low-skilled workers, and GDP growth. rising electricity prices. The construction The service sector continues to sustain sector, which experienced a boom in the growth. It has accounted for more than 60 2000s, has not returned to pre-­ crisis levels. percent of GDP growth (about 2.7 percent- The data on new construction permits (ex- age points of growth per year) before and cluding infrastructure and all public con- after the global crisis (figure 4.8). Economic struction) shows that during the boom years growth has been particularly strong in the the construction sector doubled the num- telecommunications and tourism sectors. ber of permits from 2.2 to 4 million of The telecom liberalization has brought a square meters per year; but, after adjusting 66 The Nature of Costa Rica’s Growth and Its Constraints Figure 4.7  Investment and Exports Explain Recent Slowdown of GDP Growth Contributions to real GDP growth 15 10 Percentage points 5 0 –5 –10 –15 04 14 02 08 09 10 12 00 01 03 05 06 07 11 13 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Consumption Investment Exports Imports Source: Calculations based on data from BCCR. Figure 4.8  Services Remain the Primary Driver of GDP Growth Contributions to real GDP growth 10 7 Percentage points 4 1 –2 –5 03 04 05 09 10 12 13 14 00 01 02 06 07 08 11 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Services Manufacturing Agriculture Construction Other The Nature of Costa Rica’s Growth and Its Constraints 67 Knowledge gap 4.1  Is the Contraction of Employment in the Agriculture, Manufacturing, and Construction Sectors Cyclical or Structural? Costa Rica has seen a significant contraction in employment in the construction, manu- facturing, and agricultural sectors. Cyclical patterns could be the culprits as the pattern emerged during the global crisis for all three sectors, compounded by the Roya coffee rust fungus and weather factors in agriculture. However, the contraction in jobs has persisted, which is particularly worrisome given that these sectors employ a higher share of the poor and lower-skilled workers—and given the sustained rise in unemployment. What are the structural factors in these sectors that are causing this sustained contraction? abruptly in 2009, the number of new per- past three years, the economy has been add- mits has contracted in the last two years. ing around 10,000 jobs per year, which is Amounting to about five percent of GDP, lower than the usual average annual increase the construction sector is relatively under- in the labor force of about 38,000. The un- developed in Costa Rica when compared to derperformance of the construction and ag- Panama (10 percent) or Uruguay (eight riculture sectors is particularly worrisome percent)). This underdevelopment is partly considering that they employ a higher share attributed to weakened public infrastruc- of low-skilled workers (­ figure 4.9). Another ture spending, which has been constrained explanation for weaker job creation is that by the limited fiscal space and anemic im- new firms, which tend to create more jobs, plementation of projects, and lower credit are not appearing in sufficiently high num- growth. Agriculture output also stagnated bers following the global crisis (figure 4.10). due to weather factors and the spread of the Roya fungus (“coffee rust”).4 As a share of GDP, agriculture accounts for five percent Policies Explaining of GDP, though it generates 12 percent of jobs. Agriculture is an important source of Growth jobs for the bottom 40 percent of the popu- Macroeconomic stability lation (18 percent) and the extreme poor One of the reasons behind growth perfor- (22 percent). Agricultural products, mainly mance is that the country has made efforts coffee, pineapples, and bananas, account for to maintain reasonable macroeconomic sta- 25 percent of exports. bility. Costa Rica learned the importance of The slowdown in economic activity after macroeconomic stability the hard way: dur- the global crisis has translated into weaker ing the Latin American crisis of the early job creation. As discussed in chapter 3, the 1980s it suffered a cumulative per-capita in- unemployment rate stood at nine percent come loss of nearly 20 percent. Following in 2014, which was relatively high by Costa that crisis, a new government started a se- Rican historical standards (figure 3.3). In the ries of structural reforms and committed to 68 The Nature of Costa Rica’s Growth and Its Constraints Figure 4.9  Agriculture Employs the Largest Number of Low-Skill Workers Number of workers by sector, 2013 Retail, restaurants, and hotels Community, social, and personal services Agriculture, forestry, and fishing Financial services and real estate Manufacturing Transport, storage, and communications Domestic service Construction Electricity, gas, and water 0 100 200 300 400 500 600 Thousands of workers Any primary Secondary incomplete Secondary complete Any tertiary Source: Calculations based on data from ENAHO. Figure 4.10  Creation of New Firms Remains Low, Contributing to Low Job Creation New firms created per 1,000 working-age people 18 16 14.6 14.5 14 13.9 12 11.8 10.0 9.9 10 9.4 8 7.4 6 4 3.5 2 0 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: Entrepreneurship dataset, Doing Business. The Nature of Costa Rica’s Growth and Its Constraints 69 pursuing macroeconomic stabilization.5 and falling further since then. But perhaps the Those policies proved successful in achiev- most significant step was joining CAFTA-DR, ing a high and sustainable rate of growth av- after an intense national debate that required eraging nearly five percent per annum in a referendum for its ratification. 7 CAFTA-DR 1983–2006. The fiscal deficit was virtually provided a more stable and reliable frame- eliminated at the end of that period, and work for Costa Rica’s trade with its main trad- public debt was reduced by two thirds, of ing partner, the United States, introduced which less than 20 percent was from exter- changes to the legal framework to promote nal sources. The decline of the debt burden transparency, ensured a secure and predict- was due in part to the Ministry of Finance able environment for investors, and led to the initiatives relating to the creation of a con- breakdown of government monopolies in the solidated cash management office and the telecom and insurance sectors.8 Following implementation of a debt management strat- CAFTA-DR, Costa Rica has entered into fur- egy. Monetary and exchange rate policies ther trade agreements with Canada, China, have evolved from a crawling peg policy to the European Union, Mexico, Peru, and Sin- exchange rate band (2006) to abandonment gapore, consolidating its open-economy of the pre-defined bands (2015). Inflation agenda. Exports now represent around 55 has dropped to single digits since 2009. percent of GDP, compared to around 27 per- Until the global crisis, Costa Rica’s GDP cent in 1980. The openness of the economy had not ­contracted in 25 years. (measured by the ratio of exports plus im- ports to GDP) increased to 100 percent of GDP from 65 percent during the same period. Trade liberalization, FDI, and An explicit policy to attract foreign di- structural transformation rect investment (FDI) complemented trade Moreover, outward-oriented policies have liberalization efforts. Trade agreements, played a key role in fostering trade. Trade lib- through provisions governing investment, eralization has been nurtured through various reduce the risk of expropriation and ensure actions and dates back to 1963, when Costa against the discrimination of foreign firms. Rica joined the Central American Common The passage of the Free Trade Zones law in Market (CACM, made up of El Salvador, 1981 (amended in 1990 and 2010), which Guatemala, Honduras, and Nicaragua). Join- was adopted to promote the export of non-​ ing the CACM required the elimination of traditional products, has contributed to trade barriers among member countries and attract FDI into Costa Rica.9 Free Trade ­ fueled an increase of manufacturing exports Zones accounted for 51 percent of mer- to the CACM. A second milestone took place chandise exports in 2014, compared to 21 20 years later (1983), when Costa Rica joined percent in 1997. Whereas Costa Rica’s in- ­ the Caribbean Basin Initiative, strengthening vestment attraction efforts in the 1980s fo- its trade relations with the United States.6 Like cused on lower value-­ added sectors such as other Central American countries, Costa Rica textiles, since the mid-1990s it has shifted to began to reduce tariffs unilaterally in the mid- attracting FDI with high value-added sec- 1980s, with trade tariffs declining from 53 tors, especially electronics and advanced percent in 1985 to about 5.2 percent in 2004 manufacturing, medical devices and life 70 The Nature of Costa Rica’s Growth and Its Constraints sciences, and ICT-related services. The were able to contribute to the development of coun­ try received US$407 million (three high-tech manufacturing and the country’s percent of GDP) in FDI inflows in 1997, status as an attractive FDI location. and this grew to US$2.7 billion by 2013 (5.4 Trade liberalization and FDI have made percent of GDP). FDI decreased by about Costa Rica stand out for its vibrant high-tech 1.2 percentage point of GDP, reaching 4.2 and ICT sectors. Agricultural production percent of GDP in 2014, partly affected by now accounts for a little over five percent of the exit of Intel and Bank of America (figure GDP, compared to 13 percent in the early 4.11a). Nevertheless, Costa Rica is one of 1980s. Moreover, the service sector’s share of the world’s most FDI-intensive economies, GDP has increased, becoming the main sec- with its FDI-to-GDP ratio exceeding inter- tor of the economy. The share of the manu- national levels from 1985 through 2010.10 facturing sector in GDP has stagnated at The composition of FDI has shifted towards about 22 ­ percent in the past 25 years. At the services, which accounted for 8.3 percent of same time, the employment share of manu- total FDI inflows in 2002–05, but increased facturing decreased from 25 percent to 12 to 28 percent in 2010–14, partly attributed percent.12 As a result of attracting FDI in to the liberalization of the telecom and in- high-technology sectors, Costa Rica’s share surance sectors (figure 4.11b). In 2014, of high-tech manufacturing and ICT exports Costa Rica attracted US$474.4 million of is among the highest among upper-mid- new investments across 39 projects that cre- dle-income countries and Latin America ated 10,281 jobs, offsetting the 4,343 jobs (figure 4.12). These trends suggest a fast lost due to the exit or downsizing of some transformation of the economy into sectors foreign firms.11 in which Costa Rica had a competitive A successful institutional cluster has been advantage. effective in supporting trade and FDI. In the The economy contributes to a number context of Productive Development Policies of global value chains and has been able to (PDPs) that Costa Rica has put in place since upgrade its product mix over time. Some the 1970s, one success story is the PDP to at- examples are electronics, medical devices, tract FDI and promote exports. In this task auto­motive, aeronautic/aerospace, film/ there are three main actors: the Costa Rican broad­ casting devices, and offshore ser- Investment Promotion Agency (CINDE), vices. During the 1990s the profile of new 13 PROCOMER (Promotora del Comercio Exte­ companies shifted from a focus on electri- rior de Costa Rica), and the Ministerio de cal assembly to electronics. The product Comercio Exterior (COMEX). All three work mix for medical devices is complex and so- in tandem to bring foreign direct investment phisticated, including products in the areas to Costa Rica and promote exports. Collabo- of optics, dental, cardiovascular, and breast ration between CINDE and high levels of implants. Over time, many firms operating government, including the president, success- in the medical device cluster have vertically fully brought Intel to Costa Rica in 1996 (box integrated, received raw materials, and 4.1). By streamlining processes and facilitat- ­ performed different activities in the value ing the removal of bottlenecks for multina- chain (sustaining engineering, process de- tional companies in the FTZs, these agencies velopment) (figure 4.13). In the case of The Nature of Costa Rica’s Growth and Its Constraints 71 Figure 4.11  Costa Rica Has Been Very Successful in Attracting Foreign Direct Investment a. Foreign direct investment (inflows), 1997–2014 4.0 8 3.5 7 3.0 6 2.5 5 Billions of US$ % of GDP 2.0 4 1.5 3 1.0 2 0.5 1 0 0 97 98 99 00 01 02 04 05 06 07 08 09 10 11 12 03 13 14 20 20 20 20 20 20 19 20 20 20 19 20 19 20 20 20 20 20 FDI (billions of US$) FDI (% of GDP) (RHS) b. Composition of FDI inflows, 2000–2014 100 80 6 29 60 8 % of total FDI inflows 39 40 66 35 48 34 31 20 27 11 20 0 –20 2000–09 2010 2011 2012 2013 2014 Agriculture Commerce Manufacturing Services Real estate Other Source: Calculations based on data from BCCR. 72 The Nature of Costa Rica’s Growth and Its Constraints Box 4.1  Intel and Development of the High-Tech Sector in Costa Rica In 1996, Intel, the world’s largest computer chip maker, announced plans for the establishment of its US$300 million semiconductor assembly and test plant in Costa Rica. The country’s economic and political stability, its proximity to the United States, and its pro-business environment were all important. Some bottlenecks in Costa Rica’s investment climate raised some concerns, however, such as inadequate supply of skilled labor, infrastructure and logistics (airport upgrades), tax incentives, and improvements in permits and construction processes.a The country’s tertiary educational system was strengthened with the introduction of a series of programs to increase both the number of graduates with engineering and technical degrees and the proficiency of the graduates.b The stock of graduates from engineering programs in Costa Rica reportedly increased by almost 40 percent between 2002 and 2011.c CINDE also worked with existing investors to “back link” potential suppliers—both local, through the Costa Rica PROVEE program, and foreign—of products and services that transcended sectors, identifying commonalities between medical devices and electronics in their procurement of clean room services, plastics, and metal mechanics.d Intel contributed to the development of the high-tech sector through three channels: (i) it had a direct impact on employment, investment, trade, output, and the development of the technology cluster; (ii) it served as a catalyst for repositioning Costa Rica as an attrac- tive investment location, through its impact on the country’s technical education, incen- tives laws and regulations, and infrastructure;e and (iii) it increased the confidence of foreign investors through the demonstration effect. Intel’s decision to invest in Costa Rica gave a “signaling effect” that helped solidify the country’s emergence as an attractive desti- nation for FDI in high-tech manufacturing. Companies that followed Intel in establishing manufacturing and global service operations in Costa Rica included Abbott Laboratories (now Hospira), Procter & Gamble (P&G), and Hewlett Packard. In April 2014, Intel announced its decision to close its Costa Rican microchip assembly fac- tory and move its assembly operations to China, Malaysia, and Vietnam. While all micro- chip assembly operations will move to Asia, Intel will keep its engineering and design departments and its global service center in Costa Rica, a move that showcases Costa Rica’s competitiveness in high value-­ added segments. It is unclear what the repercussions will be of Intel’s decision to leave because Costa Rica’s high-tech exports have become increasingly diversified. As it has in the past, Costa Rica will need to strategically position itself and capi- talize on its strengths to continue to attract FDI and increase job creation. a. MIGA (2006). b. Lederman et al. (2014). c. Lederman et al. (2014) with data from UNESCO). d. MIGA (2006). e. MIGA (2006). The Nature of Costa Rica’s Growth and Its Constraints 73 Figure 4.12  Costa Rica is a High-Tech Hotspot 50 High-tech and ICT exports, 2012 40 40 32 30 Percent 19 20 10 0 High-technology exports ICT goods exports ICT service exports (% of manufactured exports) (% of total goods exports) (% of service exports, BoP) Costa Rica LAC OECD Middle income Structural peers Source: World Development Indicators Figure 4.13  Exports of Medical Devices Have Grown and Diversified Export of medical devices 1,500 1,200 Millions of US$ 900 600 300 0 01 8 09 0 1 12 3 4 02 03 04 05 06 07 1 1 1 0 1 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Disposables Medical and surgical instruments Therapeutics devices Diagnostic equipments Source: CINDE (2015). aeronautics, firms are focusing on higher-­ exposure to weather or other shocks. With value service activities—such as software its integration into the global value chain of and design and engineering—and their medical devices, the country has been able product mix has upgraded as well.14 to survive the phasing out of Intel’s manu- Not only have exports tilted towards facturing plant. Exports of medical devices higher value-added sectors but they have have grown at an average annual rate of 12 also become more diversified, minimizing percent in the last decade (from US$0.55 74 The Nature of Costa Rica’s Growth and Its Constraints billion in 2004 to US$1.9 billion in 2014), services, shared service centers (for exam- while electronics grew at four percent per ple, back-office services for one company, year on average. The product mix in the including accounting, finance, procurement, medical device cluster is complex and so- and human resources), business ­ process phisticated, including products in the areas outsourcing (to a lesser extent), architec- of optics, dental, cardiovascular, and breast ture, and digital animation (figure 4.14b). implants. With its increas­ ingly diversified However, backward linkages from mul- agricultural portfolio (bananas, pineapples, tinational corporations (MNCs) to local coffee, and agroindustry), agricultural ex- companies remain weak. On average, local ports have been shielded from wide swings suppliers provide only 24 percent of inputs due to weather and diseases (figure 4.14a). to MNCs in Costa Rica.15 Other studies cite Business services (mainly ICT related) ac- that this percentage is even lower, at one count for almost half of all service exports. percent of MNCs’ inputs. Less than 20 Leading activities in this sector also in- percent of inputs sourced locally were in- ­ clude design and engineering, professional corporated in MNCs’ exports, indicating Figure 4.14  Merchandise and Service Exports Are Well Diversified Into High Value-Added Areas a. Medical devices have expanded their share 50 % of merchandise exports 40 30 23 20 20 17 13 10 8 8 7 3 2 0 Electronics Other Medical Agroindustry Pineapple Other agr Bananas Coffee Textiles manuf. devices b. Exports of ICT services have increased 80 % of service exports 60 45 40 32 20 17 5 1 0 Tourism ICT services Other business Transport Other services 1999 2014 Source: Calculations based on data from Procomer and BCCR. The Nature of Costa Rica’s Growth and Its Constraints 75 that the inputs are likely of low value and Capitalizing on protected areas not central to the value chain.16 Locally Costa Rica has successfully capitalized sourced products are largely low valued-­ on its environmental policies and Green added products and services such as pack- Trademark, by embracing and developing aging, printing materials, and services (for a sustainable tourism industry.21 In 2013, instance, cleaning and security).17 However, close to 2.5 million tourists visited the there is tangible demand for increased country. This implies that the number of local sourcing by MNCs. Of the MNCs in- visitors per capita (at around 0.5) is above terviewed by the World Bank in 2014, 93 popular destinations in the Caribbean percent declared an interest in local pro- basin such as Mexico (0.2) or the Domini- curement, and 40 percent claimed to have can Republic (0.38). While sun-and-beach a supplier development program already in tourism clearly is part of Costa Rica’s at- place. Local firms face difficulties in com- tractions, eco-­tourism (an area where it has plying with MNCs’ requirements, such as been a ­pioneer) is also very popular, with quality standards, delivery times, or up- many travelers visiting the national parks grading of technical capabilities. Moreover, and protected areas. Moreover, the country local small and medium enterprises may be ranked sixth in the region and 47th overall deterred from seeking business with MNCs according to the 2013 Travel and Tourism due to the uncertainty around the process Competitive Index compiled annually by of upgrading their operations.18 The gov- the World Economic Forum to measure the ernment’s programs to support backward factors that make a destination attractive linkages have had some tangible results, al- for the travel and tourism industry. Costa though these are quite small. The programs Rica’s large and mature ecotourism indus- have insufficient scale and scope to signifi- try has earned it a reputation as a top cantly influence the development of pro- ecotourism destination in the world. The ­ ductive linkages between domestic SMEs tourism industry is one of Costa Rica’s and MNCs.19 main sources of foreign exchange (14.2 Although backward linkages may be percent of total exports in 2013); its direct small, the presence of MNCs has had posi- share in GDP was 4.6 percent in 2013 (the tive spillovers through knowledge transfers. indirect share was 12.1 percent), and it One venue for these knowledge spillovers is contributed directly 4.6 percent of total through labor movement from MNCs to employment (indirectly, it contributed 11.5 local companies. A recent study for Costa percent).22 The tourism sector comprises a Rica finds a positive spillover generated by diverse set of operators (hotels, airlines, the movement of workers from MNCs to transportation companies, restaurants, etc.) local firms. Former workers of MNCs create that work together. Costa Rica’s main at- new firms with lower mortality rates or tributes for tourists are its beautiful na- boost firm performance for their new local tional parks, pleasant climate, and social employer.20 By inducing skill upgrades and and economic achievements.23 Although its changes in education programs, the pres- infrastructure has deficiencies, it has been ence of MNCs have also benefit local firms able to sustain the accelerated growth of (box 4.1). the industry since the 1980s. Currently, a 76 The Nature of Costa Rica’s Growth and Its Constraints sector-wide traveler tax is used for interna- large markets, educated labor force, and po- tional promotion of the Costa Rican “brand,” litical stability.25 According to the latest which prevents free riding and ensures that Global Com­ pet­itiveness report, the country marketing for the country is well funded. scored well for health and primary educa- The tourism sector is well organized, with tion, higher education and training, tech- a strong public-private collaboration where nology readiness, and innovation potential, incentives are aligned. The sector is orga- and this contributed to Costa Rica’s being nized around two umbrella organizations: ranked as the third most competitive econ- the Costa Rican Chamber of Hotels (CCH), omy in Latin America, behind Panama and a very influential organization regrouping Chile. However, other countries in the re- the largest hotel chains (and several smaller gion and beyond have caught up in terms of hotels), and the National Chamber of Tour- stability and education of their labor force. ism (CANATUR), where tourism-related In contrast, Costa Rica’s education system businesses are represented (travel agencies, has not adapted to provide the skills needed hotels, restaurants, etc.). The Costa Rican for the changing economy. Private firms Institute for Tourism (CR-ICT) is the auton- ­ perceived that rising costs are affecting their omous public institution in charge of setting operations (41 ­ percent), with ones in the ag- tourism policy. The board of directors of the ricultural (62 percent) and manufacturing CR-ICT has a good balance of political ap- (50 percent) being the most affected.26 The pointees and representatives of the industry, cited factors for rising production costs are and the CEO of CR-ICT often comes from electricity (24 percent), fuel prices (17 per- the tourism private sector. Over time CR- cent), and salaries (13 percent). ICT has adopted the role of promoter of pri- Competitiveness pressures are not new: vate sector development, coordinator of private firms have reported these invest- sectoral dialogue, and administrator of tour- ment climate obstacles for many years. The ism development incentives.24 Key elements World Economic Forum, which carries out in the successful relationship between these yearly surveys to construct its index of three bodies are their strong complementa- global competitiveness, has asked business rity, the consultative approach to developing executives to identify the most problematic policy (there is a continuous dialogue be- factors for doing business. For Costa Rica tween the private and public sector), and the in 2014, the top five areas were inefficient alignment of incentives (to maintain the government bureaucracy (27 percent), in- Green Trademark). adequate supply of infrastructure (24 per- cent), access to financing (11 ­ percent), Competitiveness corruption (eight percent), and restrictive labor regulations (eight percent). These Challenges areas are very similar to the ones identified Increasing production costs seven years ago.27 The World Bank Enter- weaken Costa Rica’s competitiveness. For prise Surveys (WBES) also provides data many years, the attraction of strategic for- on the perception of the private sector on eign investment relied on the country’s growth. In 2010, the latest survey, the top unique combination of location close to five problem areas were access to finance The Nature of Costa Rica’s Growth and Its Constraints 77 (26 percent), practices of informal sector (2011) examine the role of structural reforms (23 percent), inadequately educated work- in raising economic growth in Central Amer- force (13 percent), business licenses and ica for 1960–2009.30 Benchmarking Costa Rica permits (nine percent), and labor regula- against top performers in LAC, Araujo et al. tions (five percent). (2014) found the largest effects on per capita These investment climate constraints do GDP growth from lowering inflation, increas- have a harmful effect on firms’ productiv- ing financial depth, increasing trade openness, ity. Using objective indicators of invest- improving infrastructure, and improving the ment climate constraints, Fajnzylber et al. education level of the workforce.31 When Costa (2009) find that improvements in regula- Rica is benchmarked against OECD countries, tory compliance would have the most ben- the largest effects on growth will come from a impact on the productivity of Costa eficial ­ better road ­ infrastructure, greater financial Rican firms.28 With an expected contribu- depth, lowering inflation, improving the edu- tion ranging from 61 to 72 percent depend- cation level of the workforce, and increasing ing on the benchmark selected, the regu­​ trade openness. The magnitude of the estimated latory compliance captures the effect of effects suggests that per capita GDP growth regulation and institutional quality and is could have been two to 14 ­ percent higher, pro- proxied by sales declared for tax purposes viding an indication of forgone growth and its and external audits. Governance indicators, potential impact on poverty reduction. Swis- proxied by bribes, crime losses, and secu- ton and Barrot (2011) conclude that structural rity costs, rank second, contributing be- reforms in the financial sector (particularly tween 19 to 22 ­ percent of the productivity improvement in bank supervision and devel- increase. Infrastructure quality, which is opment of domestic security markets) and in proxied in this analysis by firms’ losses product markets (agriculture liberalization, caused by electricity outages, explains be- flexibility in electricity [independent regulator tween one and 10 percent of the productiv- and wholesale market], and telecommunica- ity increase. tions with private participation) could raise Moreover, these constraints affect overall growth by about two percentage points. In fact, growth rates. Cross-country growth regres- Costa Rica has already ­ undertaken a major re- sions have been used to identify statistically form in telecommunications, which has con- and economically significant determinants of tributed positively to economic activity. growth rates. While these regressions are not Furthermore, competitiveness challenges without their limitations,29 they provide useful could affect Costa Rica’s strong track record information on counterfactual estimates on for attracting FDI. The exit of Intel from its the potential impact on growth due to improve­ manufacturing operations, foreign investors’ ment of different investment climate areas. growing shift towards service functions, and Two recent studies are of interest to bench- increasing competition from other countries mark Costa Rica’s ­economic growth: (i) Araujo for FDI mean that Costa Rica cannot lag be- et al (2014) studies economic growth in Latin hind in its capacity to upgrade the areas that American and the Caribbean for 1970–2010, are important to continue to attract FDI and updating the work of Loayza, Fajnzylber, & maximize their spillover into the local econ- Calderón (2005); and (ii) Swiston and Barrot omy. A study of five global value chains 78 The Nature of Costa Rica’s Growth and Its Constraints highlights areas that could strengthen their persistently above that of the United States growth, such as increasing the supply of (its major trading partner) and large capital skilled labor. inflows put appreciating pressure on the ex- The consequences of these competitiveness change rate. The real effective exchange rate challenges have become more evident after the (REER) index appreciated by 12.3 ­ percent in global crisis. These challenges are the result of 2011–13, negatively affecting Costa Rica’s in- several forces: monetary and exchange rate ternational competitiveness. During that pe- policies that have resulted in an appreciation riod, the Central Bank of Costa Rica (BCCR) of the exchange rate, an education system that actively intervened in the foreign exchange is not generating enough graduates demanded market, buying foreign currency to keep the by the high-technology sectors, inadequate exchange rate above the lower end of the tar- supply and quality of infrastructure, and pub- get range. As a result, net international re- lic sector administration that has not evolved serves rose from US$4.8 ­billion at end-2011 quickly enough to respond to the change of to US$7.3 billion at end-2013. The BCCR the economy. Addressing them will be impor- sterilized the resulting monetary expansion tant to preserve Costa Rica’s competitive edge. with domestic bonds, which pay a substan- tially higher interest than the returns it gets on international reserves. Consequently, in Exchange rate appreciation 2013, the BCCR quasi-fiscal deficit rose to Although inflation has dropped to single 0.9 percent of GDP, and its debt stood at digits since the global crisis, the real effective about 10 ­percent of GDP. exchange rate has appreciated, negatively af- Greater exchange rate flexibility could fecting Costa Rica’s international competi- have several advantages for the country. tiveness (figure 4.15).32 Domestic inflation In early 2015, the BCCR abandoned the Figure 4.15  Inflation Has Dropped to Single Digit and Real Exchange Rate Has Appreciated CPI inflation, year-on-year, 2001–2014 Real and Nominal Exchange Rates 20 300 350 15 400 Percent 10 450 500 5 550 0 600 M 08 Au 07 Ju 9 M 07 Ja 1 Ja 2 Ja 3 Ja 4 Ja 5 06 Ja 7 Ja 8 Ja 9 Ja 0 Ja 1 Ja 2 13 14 Fe 0 Oc 08 De 09 14 Ja 13 No 12 Ju 12 Ap 11 Se 11 -0 1 0 0 0 1 0 0 0 0 0 1 l-1 n- n- n- n- n- n- n- n- t- p- r- g- n- c- b- v- n- n- n- n- n- n- n- - - n ay ar Ja Ja Ja Ja Real Exchange Rate (indexed to Jan-2007) Nominal Exchange Rate (Col/USD) Source: Calculations based on data from BCCR The Nature of Costa Rica’s Growth and Its Constraints 79 exchange rate band, allowing more flexibility income levels, coupled with generous social to the exchange rate, as part of its transition benefits, have resulted in high reservation to inflation targeting. Allowing pressures wages across the board. Figure 4.16 plots on the external position to be absorbed by hourly wages (in constant 2001 US Dollars) the exchange rate would facilitate facing ex- between 2000 and 2012 in two sectors that ternal shocks and could help prevent a sud- traditionally employ low-skilled workers: den drop in reserves. Changes in world agriculture (primary sector) and low-tech prices would be transmitted more directly industry (light manufacturing) for Central to the local economy, thereby aligning in- American countries and Chile (one of Costa centives to changing international condi- Rica’s structural peers). Except for Chile, tions. Further, as both creditors and debtors Costa Rica displayed the highest wages in perceive higher risk regarding the direction these sectors over the period 2000–2012.34 and magnitude of exchange rate fluctua- After 2006, the gap in real wages between tions, the new exchange rate regime could Costa Rican workers and the others started lead to a gradual de-dollarization of the to widen, but it did even more so after 2010, economy.33 partly attributed to the appreciation of the real effective exchange rate. Textile exports have dropped dramatically from 13.3 per- High wages and mismatch of cent of exports in 2000 to 1.6 percent in skills and jobs 2014, mainly attributed to high production Relatively high wages have made Costa Rica costs in Cost Rica in relation to neighboring less viable in low value-added sectors. High countries. Figure 4.16  Labor Costs Are High in Low-Skill Sectors, Compared to Central American Neighbors a. Primary sector: Labor cost b. Low tech: Labor cost per hour (constant 2001 $US) per hour (constant 2001 $US) 5 5 4 4 Constant 2001 US$ Constant 2001 US$ 3 3 2 2 1 1 0 0 01 12 01 12 02 03 04 05 06 20 7 08 09 10 11 02 03 04 05 06 07 08 09 10 11 0 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Chile Costa Rica El Salvador Honduras Nicaragua Panama Source: Calculations based on harmonized household survey data from SEDLAC. 80 The Nature of Costa Rica’s Growth and Its Constraints Changing patterns in FDI and trade levels of education. This decline in the returns have created excess demand for high skills, to education contributed in part to the his- putting upward pressure on the salaries of toric fall in inequality in many LAC countries high-skilled workers.35 Most of the reduc- over the past 15 years.36 Costa Rica’s path has tion in jobs comes from sectors that tend been different from the rest of the region, for to employ low-skilled workers (figure 3.2), trends both in inequality and in the returns such as agriculture and construction. It is to education. Specifically, in Costa Rica, the precisely in these sectors where job creation returns to education fell during the 1980s has been minimal or even negative in the (over a decade before similar patterns emerged last six years (figure 4.17). At the other end in the region), thus contributing to a reduc- of the spectrum, sectors that employ mostly tion in inequality during that period.37 How- skilled workers, such as financial services, ever, contrary to LAC, the returns to education real estate, personal services, and others, remained roughly constant during the 1990s are growing fast (though from a low base in and early 2000s in Costa Rica, and they the case of sectors such as utilities or finan- started to rise again after 2007. Thus, the pre- cial services). mium for having completed secondary edu- As such, returns to education have in- cation rose by 11 percentage points between creased along with inequality in Costa Rica in 2007 and 2013 and by six percentage points the 2000s, in contrast to trends in most other for those with any tertiary education (figure LAC countries. In much of the region, the re- 4.18). This is consistent with a pattern of turns to higher levels of education began to skill-biased technological change, which gen- fall in the 1990s—and continued during the erates relatively more skilled occupations, a 2000s—as more students achieved higher phenomenon also documented in a study of Figure 4.17  Most of the Jobs Lost Are Mostly Unskilled Labor Change in employment by sector and education level (thousands), 2007–2013 Agriculture Mining Construction Manufacturing Retail, restaurants and hotels Transport, storage and communications Domestic service Electricity, gas and water Community, social and personal services Fin. services and real estate –60 –40 –20 0 20 40 60 80 100 Some primary Secondary incomplete Secondary complete Any tertiary Source: Authors’ elaboration based on EHPM and ENAHO. The Nature of Costa Rica’s Growth and Its Constraints 81 Figure 4.18  Returns to Education Reflect Growing Demand for Skilled Labor 1.00 0.95 0.89 Wage premium over primary complete 0.80 0.60 0.41 0.40 0.30 0.20 0.17 0.14 0.00 Secondary incomplete Secondary complete Any tertiary 2007 2013 Source: Elaboration based on EHPM and ENAHO. the evolution of occupations over the last two have high demand (and growing wages) decades, and which shows that Costa Rica today.40 The technology sectors, in particu- has followed a pattern of changes in occupa- lar, are affected by the relatively small income countries.38 tions similar to high-­ skilled labor force. For example, a study of And yet, the workforce is not well skills needs in network technology esti- adapted to the current needs of the labor mates that the gap between demand and market, creating a deficit of workers with supply of skilled labor in Costa Rica in the skills demanded by the fastest growing 2011 was 36 percent, and for 2015 this gap sectors. Indeed, less than half of the cohort was projected to expand to 48 percent; the 25–29 years old (42.3 percent) had achieved worst among the eight Latin American a secondary or higher education by 2010, countries included in the study.41 Another which is worrisome for a country that aims study by the Costa Rican Chamber of Tech- for a structural shift into “new economy” nology of Information and Communications sectors. In turn, the higher education sys- (Cámara de Tecnologías de Infor­ mación y tem is biased towards social sciences and Comunicaciones) showed that, in 2008, the does not sufficiently generate graduates country had lost an estimated US$72 mil- with the skills ­ demanded, such as basic sci- lion, as a consequence of the lack of techni- ence and engineering (­ figure 4.19).39 In cians and professionals in this area. The other words, students still prefer to study technology sector is not the only one af- careers that had growing demand during fected by this trend: CANATUR (Costa Rican the 1980s and 1990s, rather than those that Chamber of Tourism), called attention to a 82 The Nature of Costa Rica’s Growth and Its Constraints Figure 4.19  The Higher Education System Produces Mostly Social Science Graduates University graduates by university University graduates by areas of type (thousands), 2013 studies (thousands), 2000 & 2013 Natural resources Natural resources 0.9 Art & Literature Art & Literature 1.4 Health Engineering 2.9 Engineering Basic science 3.0 Basic science Health 6.7 Education Education 11.0 Social sciences Social sciences 20.5 0 5 10 15 20 25 0 5 10 15 20 25 Public Private 2013 2000 Source: CONARE. shortage of human capital in the tourism needed for recognition, requirements for ob- industry. taining recognition and equivalency may Moreover, regulatory procedures do vary depending on the institution carrying not facilitate the recognition of foreign de- out the evaluation. There is no unified proce- grees, which could help to fill the skills gaps. dure for presenting appeals against adverse The norms that govern the recognition of decisions. qualifications, particularly the Regulations The difficulties created by this fragmented Governing Article 30 of the Agreement for and outdated system can result in denial of Higher Education Coordination of State work authorizations for otherwise qualified Universities, were defined in 1986 and have workers with foreign degrees. Thus, Costa not been updated. Institutional arrange- Rica has obstacles to either “produce” or “im- ments are also fragmented: academic qualifi- port” the high skills needed to sustain its cations and university degrees are evaluated high value-added growth model. Regulations by the various higher education institutions and procedures should be updated and uni- that form CONARE (Consejo Nacional de fied, in line with international criteria; and Rectores, National Council of Rectors), while measures should be taken to facilitate ac- those granted by other kinds of higher edu- creditation, adoption of a list of institutions cation institutions (para-­university institu- with established quality, and efficient pro- tions) are evaluated by the Council for cessing of “similar cases” whereby applicants Higher Education. Although CONARE has have similar titles, degrees, and qualifications defined common criteria for the documents as previous candidates. The Nature of Costa Rica’s Growth and Its Constraints 83 Infrastructure: the percentage of households with Internet access increased from 15 percent to 47 telecommunications, transport, percent; however, it is still highly concen- and electricity trated in the Central region, with much Infrastructure coverage and access are rela- lower access in areas with higher poverty tively good. Over several decades of invest- rates or with indigenous populations (top ment, the country built an extensive network row of figure 4.22). Interestingly, cantons of highway and feeder roads, electric power that have Free Trade Zones (FTZs) have on grids, and telecommunication system in the average better cellphone and Internet pen- context of the “Social Compact” that also put etration (bottom row of figure 4.22) than an emphasis on providing universal access to cantons with no FTZ. Internet access in infrastructure services that would improve the cantons with FTZs is 15 percentage points wellbeing of the population.42 Costa Rica has higher than in cantons without them (42 two times the road and three times the rail percent versus 27 percent, respectively). density of the average middle-income country Similarly, cell phone use is five percentage and is behind only Lithuania and the OECD points higher in cantons with FTZs (ICT on these indicators (figure 4.20). Access to and quality broadband are required in the electricity in rural areas and mobile services is high-tech sectors which predominate in nearly universal. The liberalization of the tele- the FTZs).45 At the other extreme, cantons communication sector allowed for fast growth with Protected Areas, which attract tour- in access to mobile and broadband services.43 ists, have lower access than the rest. The vast coverage of infrastructure has histori- cally supported Costa Rica’s trade and growth Transport. Despite good access, the quality and standards of living. of transport services is deficient. On the two The overall quality of infrastructure is most widely used international rankings of low compared to structural peers. Figure transport service provision—the World 4.21a shows the World Economic Forum Economic Forum’s Global Competitiveness Index on the overall quality of infrastruc- Index and the World Bank’s Logistics Perfor­ ture, showing that Costa Rica has the worst mance Index (LPI)—Costa Rica barely reaches ranking. This low ranking is mainly attrib- the middle point among the 160–170 coun- uted to the poor quality of ports, roads, and tries ranked, scoring below peer middle-­ railways (figure 4.21b). income countries and well below the average OECD country. Costa Rica ranked 87th out of Telecommunications. Despite progress 160 countries on the 2014 LPI, with a score of in the coverage of telecommunication ser- 2.7 (figure 4.24). Costa Rica’s score is compa- vices, areas outside the Free Trade Zones rable to the middle-income average but well (FTZs) and in protected areas have lower below the OECD average of 3.7. The country’s access, creating regional disparities in com­ mediocre ranking can be traced to poor petitiveness for local firms. Between 2008 perfor­ mance in customs and border clear- and 2012, the percentage of households ance, but also to poor quality of transport in- 46 with a cellphone increased from 69 percent frastructure, particularly in infrastructure that to 97 percent.44 During the same period, concerns port services. Costa Rica’s score and 84 The Nature of Costa Rica’s Growth and Its Constraints Figure 4.20  Infrastructure Access and Coverage Is Good in Costa Rica Road density (2011, unless Access to electricity, 2010 indicated otherwise) Lithuania 132 Lithuania 100 OECD 132 Croatia 100 Costa Rica 83 OECD 99 Chile 98 Croatia 53 Costa Rica 98 Uruguay (2004) 44 Dom. Rep. 94 Middle income 41 Uruguay 93 Dom. Rep. (2000) 26 LAC 84 Panama 21 Middle income 79 Chile 10 Panama 74 0 50 100 150 70 80 90 100 110 KM roads por 100 sq km of land % of rural population Access to cellular services, 2013 Access to fixed broadband, 2013 Panama 163 OECD 29 Uruguay 155 Lithuania 22 Lithuania 151 Croatia 22 Costa Rica 146 Uruguay 21 Chile 134 Chile 12 OECD 122 Costa Rica 10 Croatia 115 Panama 8 Middle income 103 Middle income 6 Dom. Rep. 88 Dom. Rep. 5 50 100 150 200 0 20 40 Subscribers per 100 people Subscribers per 100 people Source: Road density was calculated with data from the International Road Federation and Food and Agricultural Organization. Access to mobile and broadband: International Telecommunication Union, World Telecommunication/ICT Development Report and database, and World Bank estimates for mobile cellular subscriptions and fixed broadband subscribers per 100 people. Access to electricity: World Bank Development Indicators rank on the quality of its port infrastructure is dex’s user assessment of road and port quality the lowest among middle-income peers, scor- tively). This (with scores of 2.8 and 3.0, respec­ ing a 3.0 on port quality and ranking 115th out score puts the country behind the perfor- of 160 countries (figure 4.21b). A bit more mance of similar middle-income countries promising is the air transport quality in which (which average 3.5 to 3.7) and very far from Costa Rica scores 4.6 and ranks 61st (out of the infrastructure services delivered in the av- 160). The results for transport quality are sim- erage OECD country (5.2 to 5.3). In contrast ilarly poor on the WEF’s competi­ tiveness to other countries, Costa Rica has not made index. Costa Rica scores about 40 percent of progress in its connectivity with global ship- the maximum score of seven points on the in- ping networks. The Nature of Costa Rica’s Growth and Its Constraints 85 Figure 4.21  Quality of Infrastructure Is Low, Compared to Structural Peers a. Quality of Overall b. Costa Rica´s infrastructure quality: Infrastructure: 1 to 7 Value (1 to 7) and Country Ranking 6 6 5.6 5.2 5 4.7 4.9 4.9 5.0 5 4.6 4.0 4 3.6 3.6 3.7 4 3.6 2.8 3.0 3 3 2 2 1.9 1 1 0 0 Rails Roads Ports Overall Air Electricity Ur ep. y Cr e Li atia Pa nia a CD in ica m e l Do om ua m i (91th) (119th) (115th) (103th) transport supply Ch dl R .R ua OE na ug o c id a th (61th) (42th) M ost e C Source: World Economic Forum (2014). Weak road and port infrastructure cause Air transport becomes attractive and cost-­ major losses and delays in land and sea effective as a modal alternative when goods transport.47 An extreme example of poor are highly perishable or, more important, road quality comes from rural routes serving when value added is over a certain threshold. the pineapple industry. On the trip from And although ports in Costa Rica have the farm to distribution center, product losses worst rating in the region, the air transport can be 50 percent larger than for the much market appears to be competitive for LAC longer trip from Costa Rica to Rotterdam.48 standards.50 In fact, air transport quality in Costa Rica also performs at the bottom of Costa Rica scores 4.6 and ranks 61st out of the middle-income group in quality of its 160 in the LPI (figure 4.21b). port infrastructure, with a score of 3.0, and ranking 115th out of 160. A salient illustra- Electricity. Electricity tariffs have in- tion is the Port of Limón. Responsible for creased significantly in recent years, affect- most of Costa Rica’s maritime traffic, the ing production costs. Between 2007 and port has very low productivity and long de- 2013, the real industrial electric tariff in- lays: the median time between a ship’s port creased from eight to 20 US cents per kWh arrival and its dock arrival is more than 15 (figure 4.25). Industrial and commercial hours, compared to a normally acceptable electricity tariffs are competitive relative to range of five hours.49 the rest of Central America but high rela- Not surprisingly, firms in FTZs tend to tive to other competitors with larger mar- rely more intensively on air transport—which kets, such as China, Mexico, and the is more efficient than ports in Costa Rica. In United States. value terms, about two-thirds of FTZs’ ex- The rise in electricity tariffs originates ports are shipped via air transport, which is a from a combination of factors due to the en- more expensive but faster transport mode. ergy mix and governance issues of the sector. 86 The Nature of Costa Rica’s Growth and Its Constraints Figure 4.22  Access to ICT Is Unequal Across Regions and by Type of Activity (percent of households) a. Households that have used a computer in b. Households that have used the Internet in the last 3 months (percent) the last 3 months (percent) c. Cantons with and without FTZs: Access to d. Cantons with and without Protected Areas (PA): infrastructure services Access to infrastructure services Electricity Electricity Cellphone 85 Cellphone 90 89 85 Fixed phone line 54 Fixed phone line 54 64 67 Desktop 27 Desktop 42 38 28 Internet 27 Internet 47 42 27 0 50 100 0 50 100 % of households % of households Canton without FTZs Canton with FTZs Canton without PA Canton with PA Source: Calculations based on the 2011 Population Census First, due to its high reliance on hydropower and the country has struggled to meet de- generation, the power sector is particularly mand with renewable energy sources.51 Be- vulnerable to rainfall variability. Hydropower thermal tween 2000 and 2013, the share of ­ output has been increasingly variable over generation increased from one percent to 12 the past decade due to successive dry years, percent, raising the short-run marginal cost The Nature of Costa Rica’s Growth and Its Constraints 87 Figure 4.23  Connectivity to although technical losses in transmission and Global Shipping Networks Is Low distributions are low (10 ­ percent) compared to other countries, ICE’s operating perfor- Liner shipping connectivity index, 2004 and 2014 mance could be improved (figure 4.26). With 60 197 ­customers per employee, ICE (Instituto 50 48 Costarricense de Electricidad, Costa Rican 44 40 35 Electricity Institute) is the worst performer 33 30 26 23 among regional peers (Colombia, the Do- 21 20 minican Republic, Honduras, Panama, and 12 10 6 Uruguay). Although there is not data avail- 0 able on the level of salaries and benefits pro- vided to ICE’s employees, using data from the . Ur a y Do hile in a e th a a Pa D ep m ua m i c ni C t Ri .R id roa ENAHO the average central government to ua OE co na C ug a m C st Li e Co private sector wage premium was ­ estimated dl M 2004 2014 at 23 percent in 2012—and this premium was 49 percent for other public sector entities (for example, state-owned enterprises).53 Third, Figure 4.24  Logistics high depreciation costs—due to an acceler- Environment Is Weak ated depreciation schedule—have also in- creased operational costs that result in higher Logistic performance index, 2014 electricity tariffs. For example, depreciation 5 4.1 accounted for 17 percent of generation cost 4 3.7 in 2013.54 Finally, ICE expanded its genera- 3.1 3.2 3.2 3.3 3 2.7 2.7 2.7 2.9 tion capacity through expensive leasing 2 arrangements that have led to higher opera- ­ tional costs and electricity tariffs. Although 1 these arrangements allowed ICE to finance 0 its generation expansion, they have carried higher interest costs and shorter maturity y ca e Cr . Li tia Pa ia a ile rf D er ep m ua m Pe C n m Ch Ri oa .R ua st OE co na ug or than other financing alternatives. These leas- a th in m Ur st Do le Co d ing arrangements (alquileres) accounted for id M Be Source: World Development Indicators 15.4 percent of operating costs in 2010–12 and will increase to 19 percent in of generation due to high fuel cost. Thus, the 2013–2015.55 transition path toward a carbon neutral econ- Indicators of service quality, such as wait- omy has been a major challenge: in the last ing periods for new connections and fre- two years, existing thermal generation capac- quency and duration of outages, are below ity has been dispatched during the entire those of its structural peers. This is particu- year. The cost of thermal generation is influ- larly troublesome because of Costa Rica’s in- enced by oil prices and by expensive thermal terest in developing high-tech industries for generation leasing agreements. Fuel expendi- which highly reliable electricity is essential. tures are now reflected in the tariff.52 Second, According to the World Bank Enterprise 88 The Nature of Costa Rica’s Growth and Its Constraints Figure 4.25  Electricity Tariffs Have Increased Consistently Since 2007 in Costa Rica Electricity tariff (constant 2013 US cents per KWh) 30 Constant 2013 US cents per KWh 25 23.9 20.2 20.7 20 18.2 15 10 5 0 Residential Commercial Industrial Weighted Average Tariff 2007 2008 2009 2010 2011 2012 2013 Source: Calculations based on data from CEPAL (2014b). Figure 4.26  Benchmarking ICE’s Performance a. Customers per staff, 2013 b. System Losses, 2013 1,400 35 33.1 1,283 Sample 29.9 1,200 30 Avg. 22% 1,000 25 Sample 813 18.5 Number Percent 800 Avg. 575 20 15.7 600 15 10.4 10.3 400 297 10 273 197 210 200 5 0 0 . . ca a as ep ca a as ep ay ay a* a* m m ur Ri ur Ri .R .R u gu bi bi na na ug nd nd m m ta a m m u Pa Pa st Ur Ur lo lo Ho Ho s Do Do Co Co Co Co Source: Authors’ calculations. Grupo ICE (Costa Rica), UTE (Uruguay), CDEEE (Dominican Republic), EDEMET (Panama), ENEE (Honduras) and Colombia (EPM Energy Colombia). The data for Colombia is for 2011. Surveys, private companies in Costa Rica re- in terms of frequency of outages Costa Rican ported an average waiting period of 39 days firms experience 1.3 outages in a typical for a new electricity connection, which is the month, this is higher than OECD countries highest in Central America and 17 days (0.4), Panama (0.6), or Uruguay (0.2). A higher than the regional average. Although similar pattern is observed for the duration The Nature of Costa Rica’s Growth and Its Constraints 89 of outages. To address quality problems, ICE of electricity with renewable sources only. offers agreements to businesses in which the Costa Rica will need to double its installed ca- quantity and quality of electricity supply are pacity in the next 20 years. A large hydro proj- clearly indicated, and those businesses can ect is under construction (Reventazon, 305 finance ICE’s investments in exchange for MW) and will start operations in 2016–17, lower tariffs.56 significantly contributing to increase the sys- Sector governance is weak, resulting in tem’s reserve margin. The project cost has bottlenecks that affect the overall operational been estimated at around US$1.4 billion, and economic efficiency of the sector. The which at US$4.6 million per MW is on the roles and responsibilities of ICE, ARESEP high end of the range for new hydropower (Autoridad Reguladora de los Servicios Públi­ projects in Central America and clearly above cos, Regulatory Authority of Public Services), global benchmarks. This much-needed capac- and MINAE (Ministerio de Ambiente, En­ ity is necessary to ensure the security of the ergía y Telecomunicaciones, Ministry of Envi- country’s supply; however, the high capital ronment, Energy, and Telecommunications) cost will have an impact on the capital compo- overlap in some areas, causing inefficient nents of the tariff (depreciation and return on planning and policy formation outcomes. assets). Indeed, investment costs could be fur- MINAE is currently developing the VII na- ther reduced in future capacity additions if a tional energy plan with public consultations competitive mechanism for granting hydro (“mesas de dialogo”). The key focus areas for concessions and resulting electricity contracts the consultations are: energy efficiency, dis- was in place. Another large hydro project tributed generation, optimizing the electric- (Diquis, 700 MW) is confronting complex en- ity matrix, and social and environmental vironmental and social issues.57 Expanding issues. Costa Rica’s institutional framework generation capacity with geo-thermal energy, may not be sufficiently flexible to respond to which is not affected by rainfall variability and the strategic and expansion needs of the sec- can be dispatched in base-load, will require tor. ICE faces a significant financing gap handling environmental regulations with pro- (Costa Rica needs to invest over $8 billion in tected areas as most volcanos are in national the energy sector over the next twenty years), parks. The second challenge is how to finance but there is no clear plan on how to increase the large costs associated with any project. private sector participation to attract needed ICE’s capacity to issue new debt is limited to investment, especially considering the exist- certain thresholds. The legal framework limits ing legal constraints. The corporate gover- private sector participation through indepen- nance of ICE compares somewhat poorly dent power producers (IPP) and build-oper- with peer countries in Latin America (see ate-transfer (BOT) schemes to no more than box 4.2). 30 percent of installed capacity. Expansion of the electricity supply is con- strained by social-environmental restrictions, Addressing the infrastructure gap. the capacity of ICE to finance new generation The deficiencies in productive infrastruc- assets, and legal caps on private sector partici- ture raise concerns about the adequacy of pation. The first challenge is how to address an the levels of spending and implementation increasing gap between the demand and supply capacity for infrastructure projects. Costa 90 The Nature of Costa Rica’s Growth and Its Constraints Box 4.2  Corporate Governance of State-Owned Electric Utility ICE The corporate governance of state-owned electric utility ICE was considered poor when evaluated in 2009 against five key components of corporate governance (legal soundness, board and CEO competitiveness, professional management, performance orientation, and transparency and disclosure). ICE ranked 10th in the aggregated “corporate governance index” when compared to a peer group of 13 state-owned enterprises (SOEs) operating in the Latin American region. ICE ranked 11th in the group, in terms of performance orienta- tion (process of setting objectives, the instruments used to set objectives and their en- forcement, and the authority that conducts the assessments). This is indeed one of the areas where ICE seems to be weak. Figure B4.2.1  Aggregated Index of Corporate Governance 1.0 0.8 Index (0–1) 0.6 0.4 0.2 0 ca a ic ru a ia ca ico go r 2 y ay 1 il do bi in ua az bl liv er Pe er ai Ri ba gu ex m nt pu ua Br ug Bo m Ti Ti To a lo ra M ge Re Ja Ec st Ur Co Pa Ar d Co n an ica ad in id m in Do Tr Source: Adapted from Andres et al. (2013). Note: Tier 1 encompasses SOEs that have desirable conditions for developing good corporate governance. Tier 2 encompasses SOEs that meet only the minimum conditions considered necessary to implement a corporate governance program. Rica has made significant efforts to in- has recommitted to an investment pro- crease infrastructure investment since the gram. In fact, in 2012 Costa Rica invested structural freeze of investment imposed in the most in infrastructure as a percentage the nineties but has not yet been able to of GDP (5.47 percent) of any Latin American ­ leverage significant private sector funds. country.58 However, the vast majority of After investing no public funds in infra- this investment is public sector spending: structure throughout the 1990s, Costa Rica private participation represents only 10 The Nature of Costa Rica’s Growth and Its Constraints 91 percent of total spending (­ figure 4.27). The sanitation has been very low, but there has necessity of continuing to invest in infra- been a recent increase. Poor implementation structure and the reality of fiscal con- of infrastructure projects also contributes to straints mean that Costa Rica must the large infrastructure gap.60 For example, encourage more private sector participa- the Legislative Assembly takes between 1.5 tion in the construction and maintenance to 2.2 years to approve external loans by the of its infrastructure. The lack of investment ­ Inter-American Development Bank (IDB). in maintenance is particularly acute in the Also, of the $2,000 million in loans from road sector, where a cycle of underinvest- multilateral institutions only 28 percent has ment leading to deterioration that requires been executed.61 Limited execution is also still more resources has been created.59 reflected in the contraction of 2.6 percent, Expenditure on public infrastructure is while it grew in most sectors. A failure of in- uneven across sectors, implementation is stitutions is also apparent in the roads sector deficient, and expenditure has not been able where resources made available by multilat- to keep up with needed investments. Public eral organizations have not been put to use investment has been dominated by the because of inadequate management. transport and energy sectors with increasing Infrastructure limitations and growing investment in telecommunications in recent environmental challenges could affect fur- years. Historically, investment in water and ther expansion of the tourism industry. As Figure 4.27  Costa Rica Investment in Infrastructure, 1994–2012 Total infrastructure investment 5 4 3 % of GDP 2 1 0 00 08 06 09 03 05 04 02 07 01 98 10 96 99 93 95 94 92 12 97 11 20 20 20 20 20 20 20 20 20 20 20 20 20 19 19 19 19 19 19 19 19 Total Costa Rica - Private Costa Rica - Public Source: Comisión Económica para América Latina y el Caribe (CEPAL 2014). 92 The Nature of Costa Rica’s Growth and Its Constraints discussed earlier, the stock of infrastructure many indexes of competitiveness (figure covers most of the country, protected areas 4.28): Doing Business 2015 (83rd out of 189 still have coverage deficiencies, and in par- countries), Protecting Investors (181st), En- ticular quality of infrastructure is low. forcing Contracts (129th), Paying Taxes Moreover, the growing urban population (121st), Starting a Business (118th), Getting creates pressure on public services like Credit (89th), and Resolving Insolvency waste management (solid and water), which (84th). These rankings suggest issues to be do not comply with international standards addressed in order to improve firm produc- for environmental sustainability. To make tivity and unleash faster rates of economic sure that green tourism can continue to growth. The complex organization of the grow in Costa Rica, it is necessary to coor- government and lack of coordination among dinate the efforts of a large number of pub- agencies produce lack of clarity regarding lic actors, from infrastructure, health, steps and agencies and time to process regu- training, and other institutions. latory requirements. For example, when the telecommunication sector was liberalized, Business regulations the new private entrants reported delays in Business regulations and red tape also affect installing their systems due to the slow ap- competitiveness. Costa Rica ranks poorly in proval of tower building permits by the Figure 4.28  Costa Rica Has Burdesome Procedures (Country Ranking) Enforcing Contracts Protecting Investors Lithuania 14 Chile 56 Croatia 54 Croatia 62 Chile 64 Panama 76 Dominican Republic 73 Lithuania 78 Panama 84 Dominican Republic 83 Uruguay 106 Uruguay 110 Costa Rica 129 Costa Rica 181 Higher ranking Lower ranking Higher ranking Lower ranking Starting a Business Paying Taxes Lithuania 11 Chile 29 Panama 38 Croatia 36 Chile 59 Lithuania 44 Uruguay 60 Dominican Republic 80 Croatia 88 Costa Rica 121 Dominican Republic 113 Uruguay 140 Costa Rica 118 Panama 166 Higher ranking Lower ranking Higher ranking Lower ranking Source: Doing Business 2014–2015. The Nature of Costa Rica’s Growth and Its Constraints 93 municipalities.62 Environmental, health, and system and undermines the effectiveness of affected by construction are the areas mostly ­ monetary policy since it operates on a smaller these factors. monetary base.64 Public banks account for 50 percent of the system’s lending to the private sector, private banks’ share is 31.4 percent, and Access to finance non-bank financial intermediaries (for exam- Costa Rica’s financial system is highly dollar- ple, cooperatives, finance companies, and mu- ized and dominated mainly by public banks, tualistas) represent about 19 percent of the although the system has become more diversi- total system’s credit to the private sector. fied in recent years. Dollarization is wide- Credit to the private sector has grown spread in Costa Rica. Forty-one percent of slowly, and its level is below that for struc- deposits and loans to the private sector were tural peers. Liberalization of the banking denominated in dollars at the end of 2014 (fig- sector in the mid-1990s reignited credit, ure 4.29).63 Higher interest rates for loans in which grew from a low 18.5 percent in 1998 local currency and a stable exchange rate ex- to 32 percent in 2004 and 50 percent in 2013 plain the attractiveness of borrowing in dol- (figure 4.30). This level of financing is still lars. The widespread dollarization of the low when compared to structural peers and economy adds vulnerability to the financial OECD countries (figure 4.31). Figure 4.29  Public Banks Dominate the Financial System. Dollarization Is Widespread, December 2014 60 60 50 50 Percent of total private sector credit Percent of deposits with banks 16 22 40 40 30 30 20 23 20 13 2 7 10 10 0 0 Public banks Private banks Other financial Demand Savings Term intermediaries Foreign currency Local currency Foreign currency Local currency Source: Calculations based on data from the BCCR. 94 The Nature of Costa Rica’s Growth and Its Constraints Figure 4.30  Financial Depth Has Improved Since Banking Sector Liberalization Domestic credit to private sector, 1990–2013 60 50.4 50 40 % of GDP 30 20 10 0 90 92 94 96 98 00 02 04 06 08 10 12 20 20 19 19 20 20 19 19 19 20 20 20 Credit growth slowdown after the global Figure 4.31  Costa Rica Lags crisis and its growth has been uneven Behind Comparators in Terms of across sectors. Credit growth accelerated in Financial Depth 2014, reaching 17 percent as a result of the government having removed in August Domestic credit to private sector 2013 the limits it had imposed at the begin- OECD ning of the year. Credit in foreign currency Chile grew faster than in local currency, but that Panama Croatia trend has so far been reversed in 2014 Costa Rica when credit in local currency grew faster Lithuania (figure 4.32). Growth in credit is uneven Dom. Rep. across different sectors of the economy. The Uruguay area of highest growth within banks’ port- LAC folio is retail lending (consumer and hous- ing), which averaged 52.6 percent of private 0 50 100 150 credit in 2010–14, double the average in the % of GDP 1990s (25.7 percent). Lending for housing, 2013 2004 services, and consumers have increased Source: World Development Indicators. The Nature of Costa Rica’s Growth and Its Constraints 95 Figure 4.32  Credit Growth to Private Sector Slowdown after Global Crisis Annual growth rate of credit to private sector, 2000–2014 80 60 40 Percent 20 0 –20 11 12 14 13 01 10 07 02 04 03 05 06 08 09 00 n– n– n– n– n– n– n– n– n– n– n– n– n– n– n– Ja Ja Ja Ja Ja Ja Ja Ja Ja Ja Ja Ja Ja Ja Ja Foreign currency Local currency figure 4.33). In their share of total credit (­ over the age of 15 has an account at a formal contrast, the share of lending to agriculture financial institution, well above the regional and industries decreased in the past two average of 51 percent. The gap in access be- decades to 4.4 percent in 2010–2014. tween those in the bottom 40 percent of in- The penetration of financial services come earners and the top 60 percent is only among firms and the population, especially 5 percentage points, about half than for the the poor, is low. Costa Rican firms do not regional average. compare favorably in their use of banks to The banking system has adequate levels finance investments or working capital (22 of capitalization and asset quality. Accord- and 30 percent, respectively) when com- ing to the Financial Soundness Indicators pared to the Latin American average or (FSIs), banks had a capital adequacy ratio structural peers (see figure 4.34). In the case of 16.2 percent in July 2014, well above in- of individuals, access to finance indicators ternational standards. Non-performing are above regional peers; however, poorer loans (NPLs) stood at 1.7 percent, with segments (for example, the bottom 40 high level of provisions. Profitability, mea- percent of income earners) of the Costa ­ sured as return on equity, was stable in Rican population have much more limited 2013/2014 at around eight percent, but access. According to the World Bank Finan- lower than other countries in the region, cial Inclusion database from 2014 (Findex), such as Chile (21 percent) or Panama (15.6 65 percent of the Costa Rican population percent). 96 The Nature of Costa Rica’s Growth and Its Constraints Figure 4.33  Share of Credit to Agriculture and Industry Contracted Composition of credit to private sector 35 30 25 20 Percent 15 10 5 0 n s r r rc r in tio tu um st ic us rv uc du ul m ns Ho ric m tr In S Co ns Co A Co 1990s 2000s 2010–2014 Source: Calculations based on data from BCCR. Figure 4.34  Use of Bank Loans to Finance Investments and Working Capital by Firms 80 60 Percent of firms 40 20 0 p. C il m c ti ni Ch LA .R Ri u o u n u m Cr th Ur st Do P Li Co Investments Working capital Source: Calculations based on data from World Bank Enterprise Surveys of 2013 for Lithuania and Croatia and 2010 for LAC countries. The Nature of Costa Rica’s Growth and Its Constraints 97 Figure 4.35  Poorer and Female Costa Ricans Have Lower Levels of Financial Access 80 70 % of population in each group 60 50 40 30 20 10 0 Total (% age 15+) Female (% age 15+) Bottom 40% (% age 15+) Top 60% (% age 15+) Costa Rica LAC Source: 2014 Findex surveys. Notes 1. IMF (2015). increased and the tax base broadened. On the 2. Estado de la Nación (2014, page 140). expenditure side, most agricultural subsidies 3. The telecommunication sector attracted large were eliminated, public sector salaries were FDI flows, produced a significant consumer frozen, and the public pension system was surplus advantage from the reduction in reformed. The colon became fully convertible, prices and increase in Internet and cellular and a crawling peg policy was implemented to line access, increasing its contribution to GDP stabilize the real exchange rate and maintain (World Bank, 2013a). New private telecommu- external competitiveness. nication companies and ICE contributed more 6. Dypsky (2002). to the value added in telecommunications, as 7. Hornbeck (2012). they provided more services and added new 8. World Bank (2013a). clients, generating new revenues. 9. Monge-Gonzalez et al. (2005). 4. The Roya fungus affected 60 percent of coffee 10. OECD (2012). planted area, resulting in output reduction of 11. Barquero (2014). about 19 percent in 2013 (BCCR 2014). 12. Patton and Moore (2012). 5. World Bank. (2006a). Apart from reducing 13. Monge-Ariño (2011). industrial policy regulation and initiating a 14. Gereffi et al. (2013). gradual opening of the economy, it brought the 15. Based on a survey conducted by the World fiscal situation under control by both increas- Bank in 2014, “Facilitating Global Value Chain ing taxes and curtailing expenditures. On the Integration for Competitiveness in Costa Rica” revenue side, the value added tax rate was (World Bank 2014f). Other studies observing 98 The Nature of Costa Rica’s Growth and Its Constraints that backward linkages have been weak include 29. Some of the limitations of cross-country also Paus and Gallagher (2008), OECD (2012), growth regression include: omitted variables, Patton and Moore (2012), IDB (2010), and instruments used for the GMM estimation IDB (2014). may be mis-specified, explanatory variables 16. Groote (2005). Note that the evaluation may not capture the structural or stabiliza- focuses on inputs purchased from local suppli- tion policies that are most relevant, and risk of ers by MNCs through the “Costa Rica Provee” over-fitting the model. program. No later data is available. 30. These studies rely on cross-country regres- 17. Paus and Gallagher (2008), Patton and Moore sions, using five-year non-overlapping data, (2012), and IDB (2014). and generalized method of moments (GMM) 18. Patton and Moore (2012). estimators for dynamic models of panel data. 19. Monge-Gonzalez and Rodríguez-Álvarez The GMM estimation takes into account (2013), IDB (2014), and World Bank (2014f). unobserved country-specific effects and the 20. Monge González, Leiva, and Rodriguez- joint endogeneity of the explanatory variables Alvares (2012). (growth drivers) with the dependent vari- 21. Duhá Buchsbaum (2004) and Cornick, able (economic growth) in a dynamic model Jimenez, and Roman (2014). of panel data. Araujo et al. (2014) use a sam- 22. World Travel & Tourism Council (2014). ple of 126 countries and Swiston and Barrot 23. The tropical rainforests make Costa Rica one (2011) use a sample of 79 countries. Araujo et of the most biologically diverse areas of the al. (2014) classify the determinants of growth world, with five percent of the entire planet’s into structural and stabilization policies. biodiversity. Structural policies include education (human 24. Cornick, Jimenez, and Roman (2014). capital), financial depth, trade openness, gov- 25. World Bank. (2013a). ernment burden, infrastructure, and gover- 26. UCCAEP (2014). nance. Stabilization policies include measures 27. The 2007 WEF survey of business executives of fiscal, monetary, and financial policies, such identified the following top five constraints as price stability and real exchange rate over- to businesses: inefficient government bureau- valuation. Swiston and Barrot (2011) use a set cracy (21 percent), inadequate supply of infra- of structural indexes to gauge the extent of structure (21 percent), inflation (12.1 ­percent), reforms. inadequately educated workforce (7.3 per- 31. The analysis finds that, if Costa Rica were a cent), and restrictive labor regulations (6.8 top performance in government size (i.e., by percent). having a smaller government, as measured 28. Using a pooled sample of more than 10,000 by government consumption expenditure as firms from 16 Latin American countries, a share of GDP), its GDP per capita would be Fajnzylber et al. (2009) estimated the impact of 12 percent higher. That is because the bench- objective indicators of investment climate con- marking exercise, due to the estimated coeffi- straints (governance and institutional quality; cients of the model, views government size as infrastructure; access to finance; and education, something negative to improve upon. However, skills, and technology) on firm performance one has to consider that Costa Rica may have indicators, such as labor productivity, total large government consumption due to its social factor productivity, and wages. The authors compact, allocating around 23 percent of GDP benchmarked firm performance against five to public health and education. different scenarios: moving to Ireland’s invest- 32. While the real exchange rate affects interna- ment climate, moving to Chile’s investment cli- tional competitiveness, it is an endogenous mate, the 75th percentile of the entire sample of variable, which the authorities can use only in firms, the 75th percentile within same industry the short term. In the longer term, competi- and firm size, and the 75th percentile within tiveness is determined by the structural factors same industry, firm size, and country. elaborated later in this chapter. The Nature of Costa Rica’s Growth and Its Constraints 99 33. The widespread dollarization of the economy and for firms to take advantage of the agglom- adds vulnerability to the financial system and eration economies that infrastructure net- undermines the effectiveness of monetary pol- works can facilitate. icy, since it operates on a smaller monetary 46. Arvis et al. (2014). base. As of December 2014, forty-one percent 47. Arvis et al. (2014). of deposits and loans to the private sector were 48. World Bank (2012). denominated in dollars. 49. Kent (2011). 34. The real wage in agriculture in Honduras dra- 50. Judging by regional benchmarks, Costa Rica’s matically increased in 2011, with no apparent air transport is among the most competitive explanation. This data point is excluded from on the continent: it is served by more than 25 the graph. international carriers, it is in the top quintile of 35. See World Bank (2006a) and Lücke (2013). destinations in the Americas, and it is served 36. See Lustig, Lopez-Calva and Ortiz-Juarez very competitively in a highly de-concentrated (2013); Aedo and Walker (2012); and Cord et market. al. (2014). 51. ARESEP (2014). 37. Gindling and Trejos (2004). 52. The mechanism for the pass-through is com- 38. See Aedo et al. (2014). plex and subject to modifications. For exam- 39. This problem is more acute in private uni- ple, ARESEP introduced a mechanism for versities, which produce over half of the total partial pass-through during dry months to graduates. lower the impact on consumers; at the same 40. Some evidence also suggests that most stu- time it approved the transfer of fuel expendi- dents would not qualify for science programs tures incurred in past years to recent years. because their scores in the high school exami- 53. IMF (2013). nation are too low, so their only available 54. ARESEP (2014). options for study are in social sciences and 55. Rodriguez (2013). education. 56. Intel signed one of those agreements in 1998 41. Adduci, Pineda, and Villate (2013). and signed 45 new agreements in 2005 (World 42. World Bank (2006b). For example, one of the Bank, 2006a). objectives of the 2008 telecommunication law 57. Indigenous communities are opposing the is to ensure the provision of universal access, proposed Diquis hydroproject because it regardless of geographical location or eco- would inundate some villages. nomic status. 58. CEPAL (2014a). 43. Increased penetration in mobile services is 59. Loría and Umaña (2014) and World Bank (2008). explained by the introduction of a prepaid 60. Loría and Umaña (2014). mobile-cellular service and low tariffs. The 61. Loría and Umaña (2014). forces of competition provided an abundant 62. World Bank (2013a). supply of services, prices for Internet access 63. Dollarization of private credit increased from were reduced dramatically, and Costa Ricans five percent in 1992 to 50 percent in 2003. have responded by subscribing massively to Since then, it has decreased slowly to its cur- the new services. rent level of 41 percent. 44. World Bank (2013a). 64. In response, authorities have put in place pru- 45. This correlation does not address the under- dential measures to discourage lending in dol- lying causality: did better infrastructure and lars to non-foreign exchange generators and ICT follow the creation of an FTZ, or did FTZs to reduce the banks’ open foreign exchange choose to locate in cantons with better infra- positions. The authorities granted a four-year structure and ICT? Nonetheless, the fact that implementation period in 2013 to allow the FTZs exist in cantons with better conditions banks to gradually rebalance their loan port- provides an opportunity for service spillovers, folios and adapt to new regulations. 100 The Nature of Costa Rica’s Growth and Its Constraints Sustainability of Costa Rica’s 5.  Development Model The sustainability of its development model is at risk due to the deterioration of fiscal balances, tensions in the Social Compact, and various threats to the Green Trademark. The fiscal situation stands out as one of the most pressing development challenges facing Costa Rica. The recent deterioration stems from a combination of countercyclical measures undertaken during the crisis and structural forces. Without fiscal consolidation, the deficit could push public debt to unsustainable levels and threaten the country’s social and economic gains. Indeed, fiscal pressures, as well as rising inequality and increasing dissatis­ faction with social services, all combine to create tensions in the Social Compact. Costa Rica’s leader­ ship in environmental conservation faces the challenge of growing pressures on the use of resources and of urbanization. Changing economic incentives threaten the gains in reforestation, and the use of agro-chemicals is worrisome. Urbanization has increased energy use and air and water pollution, and the country has not developed a long-term plan to protect the environment from these threats. Finally, the current political landscape and institutional framework add an additional layer of complexity for approving and implementing key reforms needed to address these emerging challenges. Fiscal Sustainability percent of GDP in 2008 to 37 percent by 2013, with projections of 63 percent by 2019 The fiscal situation is arguably one of unless corrective measures are implemented the most pressing development challenges fac- (Figure 5.1).1 A study by the IMF estimates ing Costa Rica. Growing deficits could push that Costa Rica has to lower its primary def- the public debt to unsustainable levels and icit by 3.5 percent of GDP to achieve debt constrain resources for public investment, so- sustainability and by 4.5 percent of GDP to cial services, and environmental goals. bring its debt ratio to the pre-crisis level.2 Fiscal deterioration after the global crisis is not unique to Costa Rica. figure 5.2 and Fiscal deterioration since figure 5.3 present the overall and primary the global crisis: deficits and deficits, respectively, of Costa Rica and of public debt the structural peer countries. The fiscal bal- The fiscal situation has weakened substan- ance worsens for all countries. However, al- tially since the global crisis, with the overall though Costa Rica was among the better deficit of the Central Government growing performing countries before the crisis (third to 5.6 percent of GDP in 2013, and surpass- below Chile and Uruguay), after the crisis it ing six percent in 2014 and projected to was among the worst: its primary balance reach 6.5 in 2015 (figure 5.1). As a result, was only slightly smaller than Lithuania and public debt has increased by 12 percentage Croatia, and its overall deficit was the sec- points over the past five years, from 25 ond largest of the group. 3 Sustainability of Costa Rica’s Development Model 101 Figure 5.1  Central Government Fiscal Balance and Revenues Central government balance, 2006–14 6 4 3.7 2.7 2.4 2 0.3 0 % of GDP –0.3 –2 –1.4 –1.3 –1.9 –2.3 –3.0 –2.9 –3.1 –4 –3.6 –4.3 –4.6 –6 –5.4 –5.6 –6.3 –8 2006 2007 2008 2009 2010 2011 2012 2013 2014f Primary balance Overall balance Central government and public sector revenues, Government debt, 2006–14 2006–14 26 70 60 62.6 22 50 40 % of GDP % of GDP 18 36 30 24.8 20 14 10 10 0 08 20 6 20 9 20 0 20 7 20 4 12 18 20 6 19 20 3 15 20 7 20 1 f 06 07 08 09 10 06 12 13 0 14 0 0 1 1 1 1 1 1 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Non-financial public sector Central government Source: World Economic Outlook database and IMF (2015). 102 Sustainability of Costa Rica’s Development Model Figure 5.2  Primary Fiscal Balance Primary fiscal balance 5 4 3.7 3.1 3 2 1.9 1.6 % of GDP 1 0.7 0.0 0 –0.4 –0.5 –1 –0.7 –2 –1.6 –3 –2.6 –2.4 –2.9 –2.7 –4 Chile Costa Rica Croatia Dom. Rep. Lithuania Uruguay Panama 2002–07 2010–14 Figure 5.3  Overall Fiscal Balance Deteriorated Sharply After the Crisis Overall fiscal balance, 2002–14 4 3.4 2 0 0.0 % of GDP –1.1 –2 –1.5 –2.0 –1.9 –2.5 –2.2 –2.6 –4 –3.8 –4.0 –4.5 –5.2 –6 –5.9 –8 Chile Costa Rica Croatia Dom. Rep. Lithuania Uruguay Panama 2002–07 2010–14 Source: Calculations based on data from World Economic Outlook. Sustainability of Costa Rica’s Development Model 103 A consequence of weakening fiscal ac- And revenues, which had increased steeply counts is the downgrading of Costa Rica’s prior to the crisis, dropped back to average investment rating. In September 2014, Stan­ ­ levels from 2000–05 (figure 5.1). dard & Poor’s Ratings Services downgraded Public sector wages are particularly high the country’s rating due to fiscal trends and in Costa Rica, as compared with other coun- the foreseen difficulties in passing a compre- tries and the private sector. The public sector hensive tax reform that could put the debt wage policy adopted during the crisis aimed on a sustainable path. The longer the delay to to increase salaries in the central government ­ address the fiscal deficit, the larger will be to the 50th percentile (known as the percentil the adjustment needed to stabilize the public 50) of pay for similar jobs in the rest of the debt ratio and the potential impact on the public sector. Originally intended to adjust poorer segments of the population. the wages of some categories of public sector workers, it expanded throughout most enti- ties of the central government as well as de- The growing deficit: cyclical centralized institutions.4 As such, the wage measures and structural forces bill of the central government rose (increas- This recent fiscal deterioration stems from ing from 5.5 percent of GDP in 2008 to 7.4 countercyclical measures undertaken during percent in 2014), and the consolidated wage the crisis and structural forces. Moreover, bill of the overall public sector expanded sig- countercyclical measures adopted during the nificantly (climbing from seven percent of crisis were structural, rather than temporary, GDP to 10 percent in 2010). Public sector further adding to existing rigidities in gov- compensation is much higher in Costa Rica ernment finances and spending. than in other countries, given its level of GDP The set of policy actions taken to con- and government revenues—and public sector front the global economic crisis has long-­ wages are significantly higher than private lasting implications for fiscal accounts. As sector wages, in particular for state-owned in many other countries, Costa Rica faced enterprises (SOEs) (figure 5.4 and ­ figure 5.5). the global crisis by implementing counter- Another “big ticket item” on the expendi- cyclical fiscal policies. This was welcomed, ture side is pensions. Costa Rica’s public pen- given the magnitude of the crisis and the sion system includes the main pension fund fiscal space the country had at the time, (Seguro de Invalidez, Vejez y Muerte), a non- because the primary balance was positive contributory social pension, and special pen- and public debt was below 25 percent of sion regimes for teachers, civil servants, and GDP in 2008. However, unlike other coun- the judiciary. The first two are managed by tries, which implemented temporary fiscal the Caja Costarricense de Seguro Social policies, Costa Rica adopted policies with (CCSS), which also operates the universal long-­ lasting structural effects. Specifically, health insurance system. Although the CCSS it chose to implement steep, permanent in- pension accounts are currently roughly bal- creases in public salaries in 2008–10, and anced, with aging and demographic pres- the wage bill of the public sector continued sures they will likely face an imbalance in the to grow even after the crisis. Current trans­ foreseeable future. Even more pressing are fers also doubled to five percent of GDP. the balances of the special pension regimes. 104 Sustainability of Costa Rica’s Development Model Figure 5.4  Compensation of Central Government Employees 60 CRI 50 % of government revenue 40 CYP ARG IRE 30 DOM ESP USA SER BRA CRO CHL POR FRA ROM POL SLN PER TUR SLK NED 20 ITA COL CZE AUT BUL URU DEU 10 0 0 10,000 20,000 30,000 40,000 50,000 60,000 GDP per capita (US$) Source: Calculations with data from WEO. Excludes wages in public universities. Figure 5.5  Public-Private Income Premium by Occupation, 2012 Unqu lifi d l bor S rvic nd cl ric l Administr tiv T chnic l nd mid- l v l prof ssion ls Prof ssion l M n m nt positions Tot l 49.4 23.1 0 20 40 60 80 100 Public-priv t p pr mium Oth r public ntiti s C ntr l ov rnm nt Source: Ministerio de Hacienda (2013). This refers to relative percentage difference with private sector pay. Sustainability of Costa Rica’s Development Model 105 Over­ all government support to special pen- in any given year—and also limits its ability to sion regimes will cost Costa Rica about 59 control expenditures and the deficit. 8 percent of (2012) GDP over the next four de- Further, the budget process is fragmented cades.5 A recent study by the IMF projects and much of it is outside the control of the that the special regime for the judiciary will central government. Parliament approves start to run deficits very soon, and will de- only the budget of the central government, plete its assets by 2030; the special regime for which represents about one-third of consoli- the civil service is projected to run flat at 2.4 dated public expenditures. The budgets of percent of GDP through 2020, and will slowly decentralized institutions, public enterprises, diminish after that. Moreover, the special re- financial intermediaries, and de-concentrated gimes are far more generous than the CCSS entities is determined largely by their heads regime, which raises the question of equity, under budget caps and guidelines provided apart from their burden on the fiscal ac- by the Ministry of Finance and subject to counts.6 For instance, in 2012, the CCSS spent modification and monitoring by the Office 2.2 percent of GDP in general pension pay- of the Comptroller General (CGR). ments for 190,000 pensioners, against 2.6 Structural rigidities also prevail on the percent of GDP for 62,500 pensioners of the revenue side. Although overall revenue col- special regimes. And while the average annual lection is on par with upper-middle- and pension of a retired worker of the general high-income countries, at around 22 percent system of the CCSS amounts to around 50 ­ of GDP, tax revenues are low (13.5 percent percent of GDP per capita, the average pen- of GDP) and a large share of the remaining sion of a retired civil servant is about 170 revenues are earmarked to autonomous pub- percent of GDP per capita, and the average lic institutions (such as social security con- pension of a retired judge is about 270 percent tributions for pensions and health insurance, of GDP per capita. which are managed by the CCSS). Moreover, Managing expenditures is further compli- since 1953, Parliament has approved 1,259 cated by extensive rigidities built into the sys- tax exemptions, of which only 23 percent tem. These rigidities stem from the index­ ation had a time limit. Further, nearly half of these of public sector wages and pensions, as well as exemptions have no clear definition of the numerous spending mandates. These include tax that is exempted.9 The potential magni- both constitutional expenditure mandates tude of these exemptions for revenues is with­out corresponding financing sources, as quite large, amounting to 5.6 percent of GDP. well as legally required expenditures.7 For ex- Finally, the budgets of SOEs do not require ample, there is a constitutional mandate to in- Parliamentary approval, and SOEs do not crease expenditures in education from six to publish data. In an attempt to address the sit- eight percent of GDP. There are also legal uation, Congress approved a fiscal reform in mandates for allocations to municipalities, 2012, but it was deemed unconstitutional by housing subsidies, earmarked taxes, wages, the Supreme Court on the basis of proce- pensions, and interest payments on the debt. dural irregularities. The sum total of all of these rigidities leaves In addition, coordination between fis- the government with only five percent of the cal, monetary, and exchange rate policy is lim- annual budget for strategic allocative decisions ited. Fiscal policy has not been sufficiently 106 Sustainability of Costa Rica’s Development Model coordinated with monetary policy, and the fis- Compact goes back to Costa Rica’s early cal deficit is adding to inflationary pressures. days as a coffee-producing economy of The BCCR-stated objective is to move to a small landholders (see chapter 1). It was fully inflation-targeting system, and the recent consolidated during the middle of the 20th measure of floating the exchange rate, aban- century with the creation of the universal doning the pre-defined bands, is a move in the health and social security system Caja right direction. It is expected that this policy ­ Costarricense de Seguro Social (CCSS) in change will reduce dollarization of the finan- 1941, the establishment of the Labor Code cial system, reducing a source of vulnerability. in 1942, the guaranteeing of public educa- tion for all, and the 1949 Constitution, Sustainability of the which mandated basic social rights (com- plemented by extensive social welfare Social Compact legislation). To ensure sustainability of the Social This ambitious Social Compact has had the Compact, Costa Rica must confront the tri- backing of high and rising social spending. ple challenges of fiscal pressures on social With the abolition of the army, the country spending, increasing dissatisfaction with pub- had more resources to invest in the provision lic services, and rising inequality. First, the of basic public and social services to the entire deterioration of the fiscal system implies po- population. As a result, social spending started tential tradeoffs among social goals—and the to rise in the 1950s, reaching over three need for improved quality and efficiency of percent of GDP by 1958 (figure 5.6). It then in- social spending. Second, increasing dissatisfac- creased steadily over the decades, reaching al- tion with the quality of health care symbolizes most 10 percent by the end of the 1990s and the numerous “cracks in the system” that are 20.8 percent by the end of the first decade of emerging in the delivery of public services. the 2000s. This is close to the average level There is a sense that, although Costa Rica has spent by OECD countries,10 and significantly achieved impressive “levels” of performance higher than Central American neighbors (fig- for many indicators (low poverty, life expec- ure 5.7). Investment in education was initially tancy, literacy), it is increasingly constrained the highest category of social spending, but by on the “deltas” of adopting the changes the 1990s health and pensions had both sur- needed to make the transition to a modern passed education. In 2012, total social spend- economy with an effective “Welfare State” in ing amounted to 20.8 percent of GDP, with 7.4 the context of demographic transition. Third, percent for pensions, seven percent for health, rising inequality has exposed disparities in 5.3 percent for education, and 1.1 percent for the education system and the mismatch of social assistance. skills and jobs, as well as the ineffectiveness of social transfers to redistribute income to com- Some key achievements of the pensate for these inequities. Social Compact: universal The long rise of social spending coverage of basic services Costa Rica’s Social Compact has deep his- The Social Compact has achieved many torical roots. The construction of the Social successes, particularly in the delivery of Sustainability of Costa Rica’s Development Model 107 Figure 5.6  Social Investment Increased Steadily Over the 20th and 21st Centuries Costa Rica - Rise of social spending: 1950s, 1990s, and 2007–13 8 6 % of GDP 4 2 0 50 51 52 53 54 55 56 57 58 91 92 93 94 95 96 97 07 08 09 10 12 13 20 19 20 19 20 19 19 20 19 19 19 19 19 19 19 19 19 19 19 19 20 20 Education Health Pensions Source: Arroyo and Lindert (2014) and World Bank (forthcoming 2015). Figure 5.7  Costa Rica’s Social Spending Is the Highest in Central America Social spending, 2012 25 20.8 20 7.0 15.3 15 13.3 % of GDP 3.3 12.4 12.3 5.3 3.7 3.8 4.4 10 5.9 8.1 1.1 3.7 3.5 2.1 1.5 0.9 4.1 5 3.0 7.4 2.6 0.9 4.6 5.0 1.1 2.4 2.9 1.9 0 Costa Rica Honduras Panama El Salvador Nicaragua Guatemala Social Security Social Assistance and Labor Education Health Source: World Bank (forthcoming 2015). 108 Sustainability of Costa Rica’s Development Model universal services such as health care. Uni- The country has also invested heavily in versal access to the health care system has infrastructure, providing the population with led to outstanding health outcomes for the universal access to electricity and water, and population. The ­ country’s integrated and an extensive road network covers most of the universal health care system, managed by country. Early on, Costa Rica built an exten- the CCSS, has provided access to health sive network of infrastructure in nearly all care to the entire population, including the productive service areas (water, sanitation, bottom 40 percent (figure 5.8). Universal transport, electricity, and telecommunica- health care has been a key factor in im- tions), responding to strong social demand.11 proving key social indicators: life expec- Access to electricity went from 40 percent of tancy in Costa Rica is 82 for women and households in 1950 to 99 percent by 2013. Ac- above 77 for men, and child mortality is cess to improved water sources is also nearly low (about half of the Latin American and universal, with 96 percent of households hav- Caribbean average). Institutional births are ing access to piped water and 92 percent with virtually universal and maternal mortality access to drinking water. Costa Rica has twice has fallen over the last 20 years, reaching the road density and three times the rail den- levels that are less than half of the LAC sity of the average middle-income country, average. and is behind only Lithuania and the OECD Figure 5.8  Impressive Achievements of Costa Rica’s Social Compact a. Near universal access to health care (2006) b. OECD-level social indicators 100 100 90% 3.8 4.4 2.5 7.6 7.7 90 90 80% 25.7 22.9 25.5 80 80 25.1 25.2 70% 70 70 Outpatient visits (%) 60% World Ranking 60 60 Percentage 50% 50 50 40% 40 55.7 62.3 60.7 40 54.5 55.0 30% 30 30 20 20 20% 10 10 10% 12.8 14.8 10.4 11.2 12.0 0 0 0% Q1 Q2 Q3 Q4 Q5 m at ici o cy ur d 00 5 m at tr s t so ove 1,0 y < ra h– y ty to ale fe y ) e ce ec s rt nc h– nc al te Private Ebais Caja/INS Others el ce er it er pr Bi cta (p rtal Li rt ta Ac w Im Bi pec pe o m ex ex at ss ild fe fe ce Ch Li Li Ac Costa Rica LAC Avg OECD Avg World Avg CRI World Rank (right axis) Source: World Bank (2014b). Source: World Bank Health Stats/Find-My-Friends Tool. Sustainability of Costa Rica’s Development Model 109 on these indicators. Coverage of telephone average public spending on health in OECD lines increased dramatically, from 11 fixed countries (which averages 6.2 percent of telephone lines per 1000 inhabitants in 1950 GDP).14 to 203 in 2012. Almost all households have Moreover, recent increases in health access to a cell phone (94 percent, higher than spending have not been matched by gains in the OECD average) and 50 percent to a fixed service delivery. Between 2005 and 2010, the phone line in 2013.12 This large endowment CCSS experienced a rapid increase in spend- of infrastructure has historically supported ing without revenue growth, where 97 percent Costa Rica’s economic growth and standards of the extra spending corresponded to an in- of living. crease in the number and salary of employees. This was not accompanied by an increase in productivity: estimates suggest that between Health care: increasing concerns 2009 and 2011, average daily hospital produc- about quality and sustainability tion dropped from 69 to 61 days, and the Overall health spending has continued to number of annual outpatient visits per profes- grow during the 2000s, raising concerns sional declined from 758 to 654.15 Hence, re- about sustainability. Between 2000 and 2010, cent increases in spending have not translated total expenditure on health went from 6.5 into better services. percent of GDP to 10.1 percent. By 2012, Health budget allocations do not take public expenditure was around seven percent into account demographic and demand of GDP.13 This is far higher than public health changes, and thus leads to inefficiency and spending for countries of similar levels of inequality of care. Budget allocation for development (figure 5.9), and it exceeds health facilities across the country is done on Figure 5.9  Costa Rica’s Public Spending on Health Care Is on Par with OECD Countries Public health expenditure vs. GDP per capita, PPP 2012 20 Health expenditure, public 15 (% of GDP) 10 OECD Costa Rica 5 0 0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 GDP per capita, 2012 (constant 2005 US$) Source: World Bank (forthcoming 2015). 110 Sustainability of Costa Rica’s Development Model the basis of historical allocation, instead of some health centers, are paper-based or at current needs stemming from demographic best spreadsheets not linked to aggregated pressures and local morbidity patterns. As a databases). result, the resources that health centers re- As a result, private health costs are on the ceive are disconnected from the resources rise, and the perception of quality is declin- they need, and this is leading to inequity in ing. Between 1995 and 2010 the share of pri- the capacity to deliver care. vate out-of-pocket spending has gone from Furthermore, the current organization of 23.5 percent to 31.9 percent (of total health the system leads to increasing wait times and spending). This has occurred in parallel to patient frustration. The demographic change the aging of the country’s population, which in Costa Rica implies that as the population has naturally increased the demand for health ages, chronic conditions—which are more services. As a result, users of the CCSS face costly to treat—will become more prevalent. increasingly long wait times, in particular for Yet, the 3-level organization of healthcare surgery and specialized treatments. Not sur- (going back to the 1970s) is not well adapted prisingly, patients who can afford private in- to a growing number of patients with surance policies often use them, not to pay chronic conditions. Level-1 facilities are not for complementary services to those covered equipped to deal with such patients, thus by the CCSS, but to get the same services they refer them to level-2 or level-3 facilities, with a preferential treatment. This lowers the which also have to deal with more complex transparency of the system, and generates in- cases. This creates bottlenecks, whereas a equities in access and quality of treatment, more consolidated system would be able to which in turn fuels patient dissatisfaction deal with chronic cases (of low complexity). (figure 5.10). Likewise, an outdated infrastructure and information management system lowers quality, and decreases transparency and effi- Education: patterns of public ciency of resource management. The current spending versus the skills gap health infrastructure does not reflect the Similarly, Costa Rica has historically spent a high level of spending: Costa Rica is consid- large share of GDP on education. During the erably below the average of its peers both in sustained growth period between 1950 and density of hospital beds and medical staff, 1980, Costa Rica considerably expanded its and a recent assessment by the CCSS found investment in human capital, particularly in that only 26 percent of hospitals and clinics education. By the end of the 1950s, educa- were in good or excellent condition.16 In ad- tion expenditure was over two percent of dition, the information management system, GDP, and by 1980 it had reached 4.4 percent.17 which dates back to the 1970s, has not been The crisis of the early 1980s caused a sharp adequately modernized, and as a result it is reduction in public spending across the highly fragmented and information flows are board (figure 5.11); for example, education very difficult. This hampers efficient resource spending fell to 1.9 percent of GDP by 1982, management and lowers the accountability and by 1990 it was still only at 3.3 percent of of the entire system (for example, most in- GDP, one percentage point below the spend- formation systems, except clinical records in ing share of 1980. Nonetheless, education Sustainability of Costa Rica’s Development Model 111 Figure 5.10  Satisfaction with Public Health System Is Slowly Declining Would you say you are satisfied with the way the public hospitals work? 100 34.0 36.4 35.5 34.4 28.2 80 45.3 47.4 39.3 41.7 54.2 49.2 59.7 Percent 60 40 66.0 63.6 64.5 65.6 71.8 54.7 52.6 60.7 58.3 45.8 50.8 20 40.3 0 2007 2011 2007 2011 2007 2011 2007 2011 2007 2011 2007 2011 Costa Rica El Salvador Guatemala Honduras Nicaragua Panama Not satisfied Satisfied Source: World Bank (2014b). Figure 5.11  Costa Rica Has Heavily Invested in Education for Decades Education spending, 1980–2011 8 7.2 7.1 7 6.8 6 5.7 5.5 5.5 5.4 5.2 5.1 5.0 5.2 5 4.7 4.4 4.4 4.4 4.5 % of GDP 4.1 4 3.7 3.6 3.5 3.3 3.2 3.3 3.2 3.3 3.3 3.3 3.3 3.4 3.1 3 1.9 2 1 0 19 4 94 20 4 80 82 83 19 6 19 8 19 9 90 19 3 96 19 8 20 9 00 20 3 20 6 20 8 20 9 10 19 1 85 19 1 92 20 1 02 05 11 87 97 07 8 0 8 9 0 9 0 8 9 0 8 0 8 9 0 20 19 19 19 20 19 19 19 19 20 19 19 20 19 20 19 Source: Elaboration based on data from Jimenez (2014) and Programa Estado de la Nación (2014). spending increased systematically over the of GDP spent by OECD countries on 1990s and 2000s, as the Assembly approved average.19 spending targets of six percent of GDP in Education spending is in the “mid-range” 1997 (reached by 2009) and eight percent of between middle-income and high-income GDP in 2010 (projected to be reached by countries. Measuring education spending as a 2018).18 These levels of public spending as a share of GDP is useful for assessing its trend share of GDP are higher than the 5.4 percent over time, as well as its fiscal implications. 112 Sustainability of Costa Rica’s Development Model However, when comparing with other coun- budget goes to tertiary education (­figure tries, measuring per student spending as a 5.13), which responds in part to the coun- share of GDP per capita allows us to take into try’s pattern of growth and the shift towards consideration the country’s size and level of high-skilled, high value-added sectors and development.20 With this adjusted measure, also to the guaranteed allocation of the bud- Costa Rica spends relatively more than some get that is granted to autonomous public LAC countries and even on par with some universities. However, only 22 percent is al- high-income countries (HICs) like Australia, located to secondary education, despite the the Republic of Korea, Israel, or Singapore, demographic transition (trending towards but relatively less than Brazil, Mexico, and fewer younger children) and the large skills all other OECD countries (figure 5.12). gap for completion of secondary education Moreover, compared to the high-performing (with over half of young adults dropping out Scandinavian countries, Costa Rica spends before completing high school, many of significantly less per student, even relative to them as early as seventh grade, as discussed its much smaller GDP per capita. As dis- in chapter 3). Compared with other coun- cussed earlier, this investment in education tries, both the share of public spending on has translated into important achievements secondary and spending per secondary- in terms of literacy and primary education school student are low for Costa Rica, given attainment. its level of development (figure 5.14). Costa Rica’s education expenditure favors As a result of high dropout rates at the primary and tertiary, against spending on secondary level, public spending in educa- secondary schooling. One-third of the tion becomes regressive at higher levels. Figure 5.12  Costa Rica Spends More on Education Than Other MICs, but Less Than Most HICs Government expenditure per student as percentage of GDP per capita % GDP per capita circa 2009 120 100 80 60 40 20 0 Sp R) Sw Au den er ia hi to e (S a La ria ng ia ic u a K ia lo lia Si Isr ia Ch e Pe e e k re nds Ic ain M aly ico Ja ry ol n ic a A st ea ru lg K pu L en a Ne Fi and er nd Fr nd d h Br d pu il ap l , C Es nc or il na ni bl ith tin ng ae Sw ar pa itz str st US Hu tv of an b A rg Ric bl Re az an an Bu U a Co ra Au or a m ex It th nla la nm a la l el P De I A Co ec ng Cz Ko Re ng Ho Tertiary Secondary Primary Source: Elaboration based on data from UNESCO. Sustainability of Costa Rica’s Development Model 113 Figure 5.13  More than 70 percent quality and completion rates. As discussed of Education Spending Goes to in chapter 3, the low results in school attain­ Primary and Tertiary ment and quality are worrisome, given the high share of spending on public education Distribution of public spending in and the shift in labor demand towards high-­ education (%), 2012 skilled labor. Thus, the solution is not to spend more, but to spend better, by increas­ ing the efficiency of spending to achieve 19 better results. Some reform options are dis­ cussed in chapter 6. 33 Social protection: not so 26 effective for reducing poverty or inequality 22 Social protection spending, which includes social security and social assistance, is high by regional standards. Between 2007 and Primary Secondary Tertiary Other 2012 social protection spending rose from Source: World Bank (forthcoming 2015). 7.3 percent of GDP to 8.5 percent of GDP (left panel of figure 5.16). This is less than what is spent in OECD countries (which Given that a larger share of children from average 12.3 percent on old age and work- lower-income households drop out of sec- ing age benefits),22 and is considerably ondary education, those who remain en- higher than in neighboring countries, such rolled in the system and eventually access as El Salvador (5 percent), Honduras (6.1 tertiary education tend to come from mid- percent) or Panama (5.8 percent).23 The dle- and upper-class households. As figure largest share within social protection 5.15 shows, spending in tertiary education spend­ ing corresponds to social security goes mostly to upper-income quintiles, in (that is, contributory pensions), whereas contrast to spending in primary, which is social assistance spending represents only more heavily directed to lower-income quin- 1.1 percent of GDP, similar to the share in tiles. Given that tertiary education is allo- Guatemala (1.1 percent), Nicaragua, and cated 32 percent of the total education Panama (0.9 percent) and below El Salva- budget, the result is a regressive system dor (2.6 percent). During the crisis of 2009, where most Costa Rican households are sub- the government increased social protection sidizing the education of a minority of well- spending, particularly cash transfers, off children.21 which doubled in terms of GDP percentage Given fiscal constraints, rising inequal- from 0.3 percent to 0.6 percent.24 All other ity, and the mismatch of skills and jobs, spending categories remained largely un- Costa Rica needs to increase the efficiency changed during the period (right panel of of spending with a focus on improving figure 5.16). 114 Sustainability of Costa Rica’s Development Model Figure 5.14  Spending on Secondary Education Is Relatively Low, Given its Level of Development Secondary public expenditure as a % of GDP vs. GDP per capita, 2012 3.50 3.00 Public expenditure on education as % of GDP 2.50 LTU 2.00 1.50 URY CRI 1.00 PAN 0.50 0 0 10,000 20,000 30,000 40,000 50,000 60,000 GDP per capita, 2012, PPP (constant 2005 international $) Secondary public expenditure per pupil as a % of GDP per capita, and GDP per capita, 2012 60.00 Public expenditure per pupil as a % of GDP 50.00 40.00 per capita 30.00 OECD LTU 20.00 CHL SLV CRI 10.00 PAN 0 0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 GDP per capita, 2012, PPP (constant 2005 International $) Source: World Bank (forthcoming 2015). Sustainability of Costa Rica’s Development Model 115 Figure 5.15  At Higher Levels, Public Education Spending Becomes Regressive Public spending on education received by each quintile 100 6.3 10.0 14.1 80 17.7 37.3 18.3 23.4 Percent 60 25.6 27.9 40 24.1 16.6 20 33.4 24.8 12.4 8.0 0 Primary education Secondary education Tertiary education Q1 Q2 Q3 Q4 Q5 Source: World Bank (forthcoming 2015). Figure 5.16  Most Social Protection Spending Is on Social Security and Cash Transfers Social protection and security spending, Social assistance spending, 2007–2012 2007–2012 1.2 1.1 10 1.1 1.1 1.1 8.2 8.4 8.6 8.5 1.0 8 7.3 7.2 1.1 0.8 1.1 1.1 1.1 0.7 0.8 0.8 0.7 0.6 % of GDP 6 0.6 % of GDP 0.6 0.6 0.6 0.3 4 7.1 7.3 7.5 7.4 6.6 6.4 0.3 0.4 2 0.2 0.4 0.4 0.4 0.3 0.4 0.4 0 0.0 2007 2008 2009 2010 2011 2012 2007 2008 2009 2010 2011 2012 Social security Social assistance and labor Disability Other social assistance Cash transfers ALMPs Source: World Bank (forthcoming 2015). 116 Sustainability of Costa Rica’s Development Model Pension coverage is high, but different those provided by the CCSS for the general pension regimes exacerbate inequities. A population. This is not only a source of in- look at pension coverage among the elderly equity but also raises sustainability ques- reveals that, though not universal, coverage tions, as discussed in the previous section. is rather large for LAC standards.25 Indeed, At the same time, social assistance trans- by 2013, 58 percent of the elderly in the fers have had limited effects on poverty and bottom income quintile received a pension inequality due to low coverage of the poor. (of which 40 percent was non-contribu- This is explained by the fact that most social tory), and 67 percent of the elderly in the assistance programs in Costa Rica have fairly top quintile also received a pension (where low coverage among the poor. As figure 5.18 64 percent were contributory pensions). shows, even the most widespread program Thus, two-thirds of the elderly receive some (school meals) reaches only half of the pop- form of pension, and there is very little in- ulation in the lowest income quintile; and equality in terms of coverage across income the second largest program, the conditional quintiles (figure 5.17). Yet, the elderly poor cash transfer program Avancemos, reaches are much more likely to receive non-con- only 28 percent. In addition, there exist a tributory pensions, which are clearly smaller large number of programs, whose coverage than contributory pensions. Moreover, is not documented in the household survey among those that receive contributory pen- ENAHO and that reach a very small number sions, there is a stark difference between of beneficiaries (fewer than 20 thousand in pensioners of the public sector and private most cases). sector regimes. The special regimes (now Moreover, most major social assistance mostly closed to current workers) pay sig- programs are not well targeted. As the top nificantly more generous benefits than panel of figure 5.19 shows, even the “best” tar- geted programs, such as the non-contributory pension, have about half of their beneficiaries Figure 5.17  Pension Coverage in the bottom quintile. At the other end, a Among the Elderly Is High, Thanks to program like school meals provides most of the Non-Contributory Pension its benefits to non-poor children, and a simi- lar situation occurs with training programs Elderly covered by type of pension for from INA (Instituto Nacional de Aprendizaje, Q1 and Q5 2007–2013 (%) 100 National Learning Institute). Putting Costa Rica in an international context (bottom panel Percent of elderly 80 66.5 60.4 57.7 51.7 of figure 5.19) reveals that there is significant 60 18.2 room for improving the targeting accuracy of 40 22.7 53.8 63.7 its programs. For example, 74 percent of cash 20 29.0 39.6 transfers in the CCT Red de Oportunidades of 0 6.6 2.7 Panama (one of Costa Rica’s structural peers) Q1 Q5 Q1 Q5 go to the poorest quintile, against 44 percent 2007 2013 for Avancemos. Thus, it is unclear how effec- Non-contributory Contributory tive Costa Rica’s programs can be to help the Source: World Bank (forthcoming 2015). poor (and especially the poorest) bridge the Sustainability of Costa Rica’s Development Model 117 Figure 5.18  Coverage of Major Programs Reaches Half of the Poor, at Best Percentage of the poorest quintile receiving social assistance programs, 2013 60 50.7 % of poorest income quintile 50 40 28.2 30 20.0 19.6 18.6 20 15.6 9.2 10 2.3 2.1 0 School Avancemos INA Fonabe Social ECD School School Other school feeding CCT pension transport textbooks scholarships (food) Source: World Bank (forthcoming 2015). Figure 5.19  Most Major Programs Do Not Successfully Target the Poor a. Costa Rica: Share of beneficiaries of social programs by quintile, 2014 100 4.9 6.1 5.3 7.8 6.5 90 9.8 12.6 12.5 11.3 13.7 16.2 9.3 80 17.9 20.1 16.4 22.9 19.5 % of beneficiaries 70 18.1 31.8 21.2 27.3 32.7 60 33.6 22.2 23.7 24.5 30.8 32.7 50 37.0 27.6 40 19.8 23.6 30 25.4 52.8 52.3 20 44.0 41.6 40.5 39.9 33.1 30.2 29.6 26.9 10 21.4 0 Other CEN-CINAI Fonabe Social Transport Avancemos School School Other SA Food Other SAIMAS pension (education) feeding supplies pension scholarships Q1 Q2 Q3 Q4 Q5 b. Distribution of Beneficiaries of main CCTs-poorest quintile % of Benefitciaries belonging to 80 74 67 57 53 60 52 52 50 poorest 20% 45 41 40 39.9 38 37 36 40 32 20 0 ) 3) 9) ) 3) ) 9) 0) 3) 0) 2) 0) 0) 1) ov nd 14 09 13 01 01 01 01 01 00 00 01 01 01 01 20 eg a a 20 20 (2 in (2 (2 (2 (2 rz nia (2 (2 (2 (2 (2 (2 ( r( il ( a ca a a s as ico ca r sh He os u ile al do do ne bi am az r Ri ur ai de em Pe B Ch m ex a ua pi Br m n nd lv a la lo M lip at Pa st Ec Ja sa ng Co Ho i Gu Co Ph Ba El Source: World Bank (forthcoming 2015). 118 Sustainability of Costa Rica’s Development Model gaps created by low human capital and low ac- programs (Red de Cuido/CEN-CINAI), the cess to jobs. housing programs (BAHNVI), and other Furthermore, social assistance is widely smaller training and entrepreneurship pro- seen as “welfare assistance” instead of a grams. The lack of a culture of evaluation system to lift people out of poverty. Despite ­ and social accountability within the social a broad political and social support for so- sector (and beyond) has contributed to cial programs for the poor, there is a com- the view that social programs have “no mon view both in the government and in impact” and that they only serve as basic ­ the society, that social programs are not ef- safety nets. In the long term, this can under- fective in generating capabilities that would mine the legitimacy and the public support enable people to escape poverty. They are for them. rather seen as a retaining wall that prevents Finally, a significant challenge for the effi- further increases of poverty, especially dur- ciency and effectiveness of the social protec- ing crises, as seems to be the case with some tion system is its fragmentation and insti­​ programs like the CCT Avancemos (figure tutional complexity. Institutional arrange- 5.20), but they have little impact on over- ments in the management and financing of all poverty. Yet, many social programs social programs are highly complex. Before were conceived expressly to boost income-­ funds can be used by an implementing generating capabilities. This is the case of agency (for example IMAS (Joint Institute all school support programs (school meals, for Social Assistance), they flow through Fondo Nacional de Becas (National Scholar- various channels, in particular FODESAF ship Fund, FONABE) scholarships, and (National Development and Family Alloca- Avancemos), but also of the early childhood tions Fund, run by the Ministry of Labor), which distributes funds from the central government to various programs, based on Figure 5.20  Poverty Would an allocation formula set by law. The result- Marginally Increase Without ing earmarking makes budget allocation in- Avancemos flexible; and because more than 50 percent of FODESAF’s budget is earmarked, it is dif- 8 6.8 ficult to track financial flows and to assess 7 5.81 the cost effectiveness of programs.26 6 Percentage point change 5 4 3 Sustaining the Green 2 0.65 0.66 Trademark: Managing 1 0 0.01 Natural Resources and –1 –2 –0.99 the Environment Extreme Moderate Total-poverty Costa Rica is regarded as a world All Beneficiaries only leader in conservation and has made sig- Source: World Bank (forthcoming). nificant achievements in reforestation. Over Sustainability of Costa Rica’s Development Model 119 the last 30 years, forest cover has increased promoting forest conservation, reforesta- from 26 percent to 52 percent (figure 5.21). tion, and afforestation (chapter 1, box 1.2). This has been made possible through public These have proven critical to Costa Rica’s and private engagement and the use of eco- development strategy, given the country’s nomic incentives for conservation (under recognition as an international ecotourism the 1996 Forest Law and its Payment for desti­ nation. The tourism industry generated Environment Services [PES] framework), income of US$2.3 billion in 2013, repre- which have served as important drivers for senting around 44 percent of total service Knowledge gap 5.1  How Can Costa Rica Improve the Effectiveness and Coordination of its Social Programs? Social policy is fragmented across numerous institutions in Costa Rica, many of them autonomous agencies. Consequently, more than two dozen social programs operate in parallel, often with similar target groups. What is the mapping of social programs by agency, objectives, coverage, and performance? How can these duplications (and resulting gaps in coverage for some population groups) be reduced, either through consolidation of programs, integration of Social Registries (such as the SIPO, operated by IMAS, and the SINIRUBE [National System of Registration of Beneficiaries], operated by FODESAF), or more ambitious institutional consolidation? This is the topic of a new non-lending techni- cal assistance being requested by the Government of Costa Rica. Figure 5.21  Costa Rica’s Success in Reforestation Forest cover in Costa Rica 60 53 52 52 47 50 45 42 40 31 % of land 30 26 21 20 10 0 1961 1977 1983 1987 1997 2000 2005 2010 2013 Source: Websites from: Oficina Nacional Forestal (ONF, www.onfcr.org) and Fondo de Financiamiento Forestal de Costa Rica (FONAFIF, www​ .fonafifo.go.cr) 120 Sustainability of Costa Rica’s Development Model Knowledge gap 5.2  How Can Costa Rica Link Sustainable Production and Rural Landscapes to its Conservation Model? Costa Rica’s well-known Green Trademark has been centered on a conservation approach, supported by public sector economic incentives for conservation and reforestation (the PES mechanism) and eco-tourism resources. Fiscal and economic pressures, as well as scope, bring into question the sustainability of that model. How can Costa Rica adapt to the next generation of the green agenda, going beyond conservation and nature protection towards innovative ways for mainstreaming conservation into productive landscapes, and beyond carbon neutrality towards resilience? Further work under the programmatic NLTA on Green and Inclusive Growth seeks to bring further evidence to this important policy question. exports or 19 percent of merchandise ex- construction, and increasingly from gen- ports (chapter 4). Approximately 40 percent eral electricity consumption.27 of tourists visited national parks. Costa Rica The current environmental regulatory continues to strive for “green” leadership framework favors a strong conservation ap- with its international commitment in 2009 proach that has limited productive opportuni- (at the UN Summit on Climate Change) to ties. During the last four decades, the focus become a carbon neutral nation by 2021 on conservation has led to the development (chapter 1, box 1.2). of a nature-based tourism sector, on the one However, its Green Trademark and its hand, and a significant contraction of the do- commitment to carbon neutrality are at mestic timber industry, on the other. Costa risk. Being green and carbon neutral at the Rica has become a net importer of processed same time entails not only maintaining the and finished wood products, such as furni- current levels of forest cover, but also re- ture.28 As the demand for wood and processed ducing the current levels of Greenhouse timber has been steadily increasing over the Gas (GHG) emissions from all sectors. This last decade, underinvestment in the timber in- is a challenging task given the trade-offs dustry has led to a 36.4 percent drop in output built into the current economic incentive of processed timber between 2007 and 2012 structure for forest conservation and re- and an increase in the exports of low-value forestation, which is dependent on reve- unprocessed products (such as palettes [tari- nues generated from tax on gasoline. While mas], construction material, sawn wood). Ag- a large share of GHG emissions can be re- ricultural productivity of many of Costa Rica’s duced through better management of for- main crops has also seen a declining trend ested and agricultural land, the country (coffee, in particular) (chapter 4), which affects also needs to reduce carbon emissions the livelihood of rural populations engaged in from fossil fuels coming from transport, agriculture. The focus on conser­ vation has also Sustainability of Costa Rica’s Development Model 121 limited the country’s potential to meet increas- they may jeopardize Costa Rica’s achieve- ing energy needs through renewable energy ments in countering deforestation trends generation (such as hydropower, geothermal, and promoting a green image, unless land and biomass), which would require the ex- use change is adequately managed and po- ploitation of resources located within the tential negative environmental externalities boundaries of protected areas. (such as heavy use of agro-chemicals—see Efforts to maintain or further increase figure 5.22) are internalized through effec- the forest cover are undermined by rising tive mechanisms (such as organic produc- opportunity costs. The fiscal weight of tion). If in the medium to long term Costa Costa Rica’s conservation and reforestation Rica wants to continue increasing or main- program can prove too heavy to sustain in taining its current forest cover, the regula- the longer term, with more pressing social tory framework governing its management and economic issues that the country faces and associated incentives needs to be re- in the short term and an increasing oppor- viewed and adapted to a growth strategy tunity cost of alternative land-use options. that considers the role of forest conserva- More profitable non-traditional export crops tion in the context of a rural productive such as pineapples, bananas, or oil palm are space that is increasingly vulnerable to becoming more economically attractive, weather changes. with land owners opting out of the conser- Moreover, rapid urbanization has brought vation and reforestation programs once the new challenges for the country. Costa Rica’s contracts are phased out. Although these urban population has been growing rapidly, activities generate employment in rural areas, from 50 percent of the total population in Figure 5.22  Use of Insecticides, Herbicides, and Fungicides/Bactericides (kg/ha), Circa 2011 Insecticides, herbicides, and fungicides & bactericides use 8 6 kg/ha 4 2 0 Costa Rica El Salvador Panama France United States of America Fungicides & bactericides Herbicides Insecticides Source: FAOSTAT. Note: Figure shows use in kg/ha of the three largest groups of pesticides. Calculated as the ratio of pesticides use and agricultural land area minus organic agricultural land. Latest data point used between 2007 and 2011. (Costa Rica and Panama: 2011; El Salvador and France: 2010; United States: 2007). 122 Sustainability of Costa Rica’s Development Model 1990 to 75 percent in 2013. The first chal- solid-waste management. Although access lenge relates to the transportation area and to sanitation is almost universal (chapter 3), associated carbon emissions and air pollu- 96 percent of all urban wastewater collected tion. Increasing energy demand is boosting is discharged into rivers and receiving water the use of fossil fuels and raising air pollu- bodies without any treatment, generating tion levels, mainly through a significant in- public health risks and water resources con- crease in road-based transport. Growing tamination (figure 5.23). Only eight percent and aging vehicle fleets and increased con- of the population is connected to a sewage gestion are key drivers, especially in the San system and to a wastewater treatment plant. José Greater Metropolitan Area (GMA).29 In San José, only four out of 16 wastewater Energy subsidies, especially for transport treatment plants comply with state regula- fuels, further magnify this development by tions. The water contamination has been incentivizing inefficient use of vehicles. so severe in some cases as to result in five However, the public transportation system beaches being declared unsuitable for in the GMA is obsolete (all routes converge swimmers (Quepos, Azul, Tarcoles, Portete, to the city center) and contributes to grid- and Balneario municipal de Limon).30 lock. The economic costs of urban air pollu- Mounting unmanaged solid waste has con- tion in Costa Rica amount to about CRC sequences for public health and the sustain- 210 billion (about one percent of GDP). ability of the green model. In sum, an Around 350 premature deaths and close to emerging area of environmental sustain- 4,700 lost Disability-Adjusted Life Years ability is the brown agenda. (DALY) are attributable to urban air pollu- The country is also highly exposed to nat- tion (Particulate Matter (PM) exposure). ural hazards and faces significant challenges The second set of challenges emerging from from climate change. Costa Rica is one of urbanization includes sewage treatment and the countries with the highest exposure to Figure 5.23  Access to Water, Sanitation, Sewage, and Connection to Sewage Treatment, Circa 2012 Access to water source Access to sanitation Sewage Connection to sewage Country (% of urban pop) (% of urban pop) treatment (%) treatment (%) Costa Rica 99.6 94.9 4.01 8.01 Chile 99.6 100.0 72.0 (2011) 4 83.32 (2009) Croatia 99.8 98.6 81.64 (2011) 27.03 (2011) Dominican Republic 82.5 85.5 19.5 (2011) 4 12.02 (2005) Lithuania 99.3 98.7 48.94 (2011) 64.03 (2011) Germany 100 100 99.4 (2007) 4 95.03 (2010) Panama 96.8 79.7 n/a 55.02) (2007) Uruguay 99.9 96.5 15.0 5 n/a United States 99.4 100.0 87.0 (2008) 4 n/a Source: WDI database, 2012; 1) Programa Estado de la Nacion 2014; 2) UN Stat; 3) Eurostat; 4)Approximated based on Aquastat FAO data; 5) UNEP 2000. Sustainability of Costa Rica’s Development Model 123 hazards, including hydro-meteorological informal urban settlements.33 On the pub- (floods, cyclones, and landslides) and geo- lic infrastructure side, low levels of invest- physical (earthquakes and volcanoes).31 It is ment in disaster risk reduction are putting the second most exposed country to multiple critical transport infrastructure at risk of hazards based on land area, with 36.8 percent failure in the event of a disaster. An illus- of the total area exposed to three or more trative example is the need to replace and natural hazards. It is estimated that 77.9 per­ retrofit the bridges most crucial for trans- cent of the population and 80.1 percent of port and most vulnerable to seismic and GDP are subject to high risk from multiple hydro-­ meteorological events. Of all the hazards. While Costa Rica has been spared damages due to hydro-meteorological events by major disasters in recent years, it is not between 2005 and 2010, 51.6 percent were immune to the potential of a catastrophic recorded in the transport sector. event. Hurricane Thomas in 2010 caused Strengthening prevention and resilience damages estimated at US$292 million; Tropi- is complicated by low capacity of local gov- cal Depression 12E in 2011 caused US$60 ernments. Even though national policies for million in damages; and in 2012 the Samara- Disaster Risk Management (DRM) exist at Guanacaste earthquake generated US$100 the national level, a major constraint in the million in damages to public buildings, process of generating more knowledge and transport, health, education, and water and increasing preparation is the level of local sanitation infrastructure. Furthermore, ac- technical capacity to absorb information cording to the Intergovernmental Panel and implement the right measures.34 For ex- of Climate Change (IPCC), some of the cli- ample, the capabilities to use and process mate change projections for Costa Rica indi- information vary widely across municipali- cate up to a 32 percent decrease in overall ties: only a few have the right human and rainfall precipitation by 2050. Some areas, infrastructure resources, such as sophisti- especially in the North, have recently seen cated mapping systems. As a result, overall more frequent and prolonged drought peri- resilience and adaptation to climate-induced ods. Changes in storm regimes near the impacts has yet to be achieved at local coast may further erode coastal morphology, levels. disrupt fishing areas and agricultural lands, Sustaining and promoting a Green Trade- and salinize water sources.32 mark needs to extend beyond forest conser- Increasing urbanization and vulnerabil- vation and reforestation. The “green” asset ity of public infrastructure are two factors that Costa Rica has established with its forests contributing to risk exposure. The San Jose can be maintained only through an integrated GMA’s fast-­ growing population puts pres- approach that enables the conservation of sure on the limited natural resources and natural resources to be developed in parallel on public goods and services. Affordable to other productive activities and balances housing is a major socioeconomic con- the many financial and environmental trade- straint (see chapter 3) that forces many offs. This has been achieved in other coun- low-income families to relocate to higher-­ tries, such as New Zealand and Finland, risk areas. In fact, most of the people af- where the “green” image is supported with fected by disasters such as floods live in strong forestry and agricultural sectors and a 124 Sustainability of Costa Rica’s Development Model risk preparedness framework. However, the democracy, with a clear checks and balances challenge in Costa Rica will be to find the system between the executive and the legisla- right balance, if any, between “green” and car- tive (Assembly). Since 1953, elections have bon neutral while addressing increasing is- been peaceful and citizen participation in sues in the brown agenda in the country. the political process is high (chapter 1). Costa Rica also has high marks in the areas of voice and accountability and control of Governance Challenges corruption. The areas of government effec- tiveness and regulatory quality are the ones Hamper Government trailing behind, with the largest difference Effectiveness and with OECD countries. The current political landscape poses ad- Regulatory Growth35 ditional challenges for approving and im- Despite Costa Rica’s good standing on plementing needed reforms. Costa Rica’s governance relative to the LAC region, both legislative power has gradually shifted from perceptions and evidence suggest that its in- a bi-partisan to a multi-party system, ham- stitutions and procedures have not been able pering the adoption of reforms in many in- to adapt to the challenges of a new economic stances over the past ­ figure 5.25). decade(s) (­ and social environment. Costa Rica has bet- Parliamentary procedures, which histori- ter governance indicators than LAC, but lags cally did not impede decision making, give behind OECD countries (figure 5.24). After political minorities the power to delay votes the 1948 civil war, the abolition of the army, or to file injunctions, which, in a fragmented and the creation of the new Constitution and polarized environment, seems to be in 1949, Costa Rica became an exemplary hampering reform processes.36 One example Figure 5.24  Costa Rica Has Better Governance Indicators than LAC but Lags Behind OECD Countries 100 Percentile ranking, 2013 80 60 40 20 0 Regulatory Government Rule of law Control of Voice and Political stability quality effectiveness corruption accountability and absence of violence/terrorism OECD Costa Rica LAC Source: Worldwide Governance Indicators. Sustainability of Costa Rica’s Development Model 125 Figure 5.25  From a Two-Party to a Multi-Party System in the Last 25 Years 60 Number of sets in the assembly 50 40 29 17 11 13 30 5 6 8 20 25 25 24 10 18 0 1990 2006 2010 2014 PLN PUSC PAC ML FA Other Source: Costa Rican Legislative Assembly. PLN: Partido de Liberación Nacional; PUSC: Partido Unidad Social Cristiana; PAC: Partido Acción Ciudadana; ML: Movimiento Libertario; and FA: Frente Amplio. is the fiscal reform package that was ap- population. Finally, in the absence of clear proved by the Legislative Assembly, and political leadership and consensus, new in- later invalidated by the Constitutional terpretations of the existing laws take the Court following an injunction filed by a leg- place of passing new laws. islator. The leadership, in drafting new legis- A consequence of this gridlock in pass- lation, has also shifted slightly from the ing reforms has been the proliferation of an executive to the legislative, partly a reflec- increasing number of public (and often au- tion of parliamentary procedures. For ex- tonomous) institutions created to address ample, during the Chinchilla administration specific problems. Costa Rica is notorious in (2010–2014), most of the new legislation Latin America for the large number of pub- passed originated in the Arias administra- lic institutions. In 2011, the state comprised tion (2006–2010). Paradoxically, the Chin- 324 entities (table 5.1), followed by El Salva- chilla government experienced the most dor with only 162. Many of these institu- difficulty in passing comprehensive reforms tions were born out of a desire to make the in the Assembly during its first two years, government apparatus more efficient and when the political capital would seemingly agile—typically through the creation of “au- be the highest.37 During the second half of tonomous institutions,” which are run under the term, the “wear and tear” of parliamen- different regulations from the Central Gov- tary negotiations manifested itself, and the ernment and have a separate budget process magnitude of the new legislation passed (more on this below). Also, these institu- also diminished. The delays and difficulty in tions usually have much more generous pay passing comprehensive and meaningful re- scales than the Central Government, but the forms result in a mismatch between the same level of job security, and no real per- “political delivery” and the demands of the formance evaluation mechanisms. In the 126 Sustainability of Costa Rica’s Development Model Table 5.1  There Is a Large Number of Public Sector Institutions in Costa Rica, 2011 Type of institution No. Type of institution No. Branches of government 3 Semi-autonomous institutions 8 Electoral Supreme Court 2 Public enterprises 25 Legislative power 2 Non-state public agencies 47 Ministries 18 Public funds’ managing institutions 1 Agencies ascribed to ministries 79 Municipalities 81 Agencies ascribed to the Presidency 3 Municipal district councils 8 Autonomous institutions 34 Municipal enterprises 1 Agencies ascribed to autonomous institutions 12 Total     324 Source: Costa Rica’s Legislative Assembly – website (ttp://www.asamblea.go.cr). last two decades, more public entities were Current public procurement practices fail to created than in the entire period 1950–1979. consolidate purchases (taking advantage of Between 1990 and 2009, 118 were founded framework agreements, for example) or or restructured. Over time, these new bod- standard bidding documents, both of which ies are affecting the government’s efficiency could lead to important cost savings. Also, (and service ­ delivery), thus reinforcing the neither the public sector entity nor the gov- vicious cycle that leads to the creation of ernment makes good use of existing infor- new institutions. In some cases, some insti- mation technology to plan, manage, or tutional clusters have been effective in control procurement. In many cases, precise achieving positive outcomes (for example, purchasing statistics are not available and the attraction of FDI and tourism develop- there are only weak links between procure- ment discussed earlier); but in others there ment plans, the budget process, and various are failures, too, the most salient one being aspects of the financial management at the the case of the rice sector (box 5.1). With 28 budget execution stage. In addition, within institutions, the agricultural sector provides this overall legal and oversight framework, an example of the fragmentation of institu- each public sector entity is free to set its pro- tions that hinders the coordination and ef- curement rules and procedures, making it fectiveness of policies, decision-making, and difficult to do centralized planning or to quality of services for producers.38 As long consolidate purchases across institutions. as public sector action is needed to provide This absence of complete statistics on pro- services to foster growth, modernizing the curement and procurement methods makes public sector becomes a needed foundation it difficult to identify patterns, learn from that will impact all aspects of the economy. mistakes, carry out performance evaluation, Besides the proliferation of institutions, and design efficient procurement policies. current procurement practices are not con- Another complexity of the intricate pub- ducive to an effective control of resources. lic sector administration is the budget and Sustainability of Costa Rica’s Development Model 127 Box 5.1  The Rice Sector: Failure of a Productive Development Policy Rice, a staple item in the food basket of the poor, is one of the most protected commodities in Costa Rica. Rice tariffs in Costa Rica are high, at 62 percent for processed rice (having recently increased from 35 percent), and prices for locally produced rice are controlled at each step of the process. Rice subsidies (in the form of guaranteed prices), amount to 45 percent of the domestic price, higher than in the United States or the European Union (EU). Ironically, the policies that protect rice prices under a banner of food security and equity have brought them well above international prices, benefiting a handful of large producers and harming the budget of the poor. The Corporación Arrocera Nacional (CONARROZ), created in 2002, has been effective in protecting producers from international price shocks, but not in improving local produc- tion conditions. CONARROZ gathers a very diverse set of rice producers and processors, most of them very small producers. Its lobbying efforts are presented as a way of defend- ing these small producers and protecting jobs in the agricultural sector. However, most of the rice produced in Costa Rica (about 80 percent) comes from a minority of large producers and processors, who also benefit from tariff-free import quotas (assigned proportionally to their processing capacity). Thus, they can purchase rice at international market prices, and sell the processed rice domestically with a high profit. It is estimated that from 1995 to 2005, consumers transferred US$396 million to rice producers, which for the poorest households represents seven to eight percent of their per capita income. Price controls have done little to improve productivity among rice producers but have created incentives to maintain the status quo. Although consumer prices have continu- ously increased since 2008, producer prices have remained flat or decreased, mostly due to low quality. The average yield per hectare has gone down from 4.8 tons in 1999/2000 to 3.39 tons in 2011/2012. As a result, most domestic rice producers, particularly the smaller ones, would not be profitable at international prices, and thus have a strong interest to maintain the status quo. Source: Crespi (2014); Monge Gonzalez et al. (2010); Arroyo et al. (2013); and Cornick et al. (2014). planning system. The current budget system institutions outside the central government, limits the ability to have a comprehensive whose budget is approved by the Comptroller discussion on policy and spending priori- General of the Republic (CGR), and which ties.39 Costa Rica has three distinct budgetary doubles the size of the national budget. Of processes in the public sector. First, the cen- the total public sector budget, the national tral government budget or national budget, budget is about 33 percent, and the budget of which is approved by the legislative assembly. other institutions is 66 percent.40 The third Second, the budgetary process of the budget process is for municipality budgets, 128 Sustainability of Costa Rica’s Development Model Box 5.2  The CCSS Lacks an Integrated Management System Over Resources In terms of budget, the Costa Rican Social Security Institute (Caja Costarricense de Seguro Social, CCSS) is the largest decentralized autonomous entity in the Government. Its annual budget of about US$5.6 billion for 2015 is equivalent to 37 percent of SS: Accounting ProcessCycle the Central Government Budget. Information systems The CCSS’ operational processes are Financial consolidation Results characterized by complex and cum- • Central collection SIF Financial Statements system (SICERE). • Other collection posts. Health Pensions 299 Budget • Mortgage loans. • General accounting. bersome procedures. They rely on a insurance insurance “silo” approach that favors fragmenta- Executing • Revolving funds (FROs). • Statement of Units • Investments. financial position. • Pensions. • Income statement. Operating transactions • Treasury. • Inventories and tion and duplication of many pro- • Cash flow statement. cesses paired with multiple layers of supplies. • Statement of • Insurance. changes in equity. • Industrial costing systems. • Human resources. approvals. Moreover, the fragmenta- • Etc. tion of processes has resulted in the Manual procedures for development and use of multiple non- consolidation and accounts reconciliation. linked IT tools (including Excel sheets and databases) for the recording, control, and generation of different pieces of information. The information generated by these independent IT tools needs to be aggregated and reconciled. In a highly decentralized environment, having excessive manual controls is not effective and has consequences in terms of added transaction costs and difficulties in producing information that could assist with service delivery. The preparation of financial statements provides a good example of the challenges that CCSS faces to adequately monitor the use of its resources. Information from over 20 IT systems and databases is manually collected and aggregated Because these systems and databases are not linked, the CCSS has developed burdensome and manual reconciliations across the organization, as well as additional manual controls, in order to ensure that errors are prevented or detected in a timely manner. The lack of an integrated IT platform does not allow the timely and systematic recording of financial and accounting transac- tions. While these manual and compensating controls allow for the timely preparation of budgetary reports and financial statements (including budget monitoring, reconciliations of accounts, analyses of fluctuations, aging of accounts, and manual and supervisory reviews), given the nature of the CCSS more meaningful financial information—such as financial information by service delivery units or by cost centers—might be needed to support decision-making and monitoring financial performance. Sustainability of Costa Rica’s Development Model 129 which account for less than two percent of often resorted to cutting back on public in- public sector expenditures. The CGR verifies vestment, putting pressure on service deliv- the legality of proposed expenditures by the ery. One of the most evident results of municipalities. Thus, a large part of public public investments restrictions is the dete- expenditures is not under the direct control rioration in the quality of most of Costa of the executive or the legislative powers.41 Rica’s infrastructure services.”43 More im- The latest PEFA assessment (World Bank and portant, the executive has no power to di- IDB 2010), found that only 49.6 percent of rect or contain spending in autonomous expenditures were approved by the Congress institutions, since they operate with wide and in 2011 the central government ac- margins of budgetary and administrative counted for only 41 percent of expenditures independence, and are constitutionally and 27 out of a total of 229 entities.42 protected from political interference or The budget process and numerous ear- changes in government. marked expenditures reduce the margin for Moreover, sectorial planning is weak the executive to control public investment with limited medium-term costing projec- and current expenses. Budgets are increas- tions. Annual work plans are developed at ingly constrained by constitutional man- the institution level despite a provision dates and rules without corresponding made in Law 8131 to develop multi-annual financing, such as directives on minimum plans at the sectorial level (multi-institu- spending for education, municipalities, tional). The PEFA (ID-11 iii) shows a very housing subsidies, and community devel- weak environment for sectorial planning. opment. Additional laws stipulate that por- More analysis will be needed to fully under- tions of taxes and fees must be allocated to stand the consequences of weak sectorial certain activities and institutions, such as planning coupled with increased number of FODESAF (National Development and institutions (including special purpose insti- Family Allocations Fund) and CONAVI tutions, executive agencies) on service de- (National Roads Authority). Pensions, wages, livery, but clearly a fragmented planning and salaries account for three-­ quarters of system is not conducive to improving the central government expenditures. Public quality of public expenditures. sector employment is much higher in Costa The main effect of weaknesses in public Rica than in most other Latin American sector administration is on the diminished countries, comprising 14.4 percent of the ability of the public sector to deliver services labor force in 2012 compared to an average expected by the population and private sec- of 10.9 percent (figure 5.26). Public sector tor. Agencies affected are those in charge of wages are also significantly higher than the education, health, public infrastructure, and private sector ones (see chapter 5, figure social assistance, which have a direct effect on 5.5). Moreover, the central government tax the inclusiveness of growth. Not surprising, revenue is still insufficient to fund all these there is a growing perception of low effective- legal and constitutional mandates. A World ness of government institutions and civil ser- Bank Public Expenditure Review finds that vants. According to a 2011 Latinobarómetro “in order not to exceed spending limits and study, Costa Rica was rated by its own citizens given these rigidities, the government has at 5.5 out of 10 on the efficiency of the 130 Sustainability of Costa Rica’s Development Model Figure 5.26  The Public Sector in Costa Rica Is One of the Largest in LAC in Terms of Workers 25 agencies, by income quintile, 2012 Share of workers in public sector 20 15 10 5 0 r or a a C a ca as p. il ia ay ico ru ile do az al bi m LA liv .re ad Pe Ch ur Ri gu ex em m na Br ua Bo m nd lv ra a lo M Pa Ec at st Sa Do Co Pa Ho Gu Co El All Bottom 40% Top 60% Source: LAC Equity Lab. Figure 5.27  High Socioeconomic Groups Are More Satisfied with the Government What has the government done for you lately? (2011) 100 90 80 37 47 70 % of responses 60 50 20 40 18 30 22 20 19 10 19 13 0 Decile 1–5 Decile 6–10 A lot Something Little Nothing Don´t know Source: Calculation with data from Latinobarómetro. government and 5 out of 10 on the efficiency months,” citizens responded that government of civil servants. This perception was worse was doing a relatively good job for middle among people in the lower deciles than in the class and upper middle class but not so much top deciles. The Latinobarómetro study also for the poorer groups (figure 5.27). reports that, when asked about “what the The institutional complexity of the pub- government has done for you in the past 12 lic sector results in convoluted procedures Sustainability of Costa Rica’s Development Model 131 Knowledge gap 5.3  What Are the Key Governance Bottlenecks in Executing Infrastructure Projects—and What are Their Cost Implications? Various factors are named as obstacles to implementation of public infrastructure and social projects in the various line ministries, such as cumbersome processes in the Public Procurement Law, deficiencies in the environmental regulatory framework, cumbersome checks and reviews by the Contraloria, and deficiencies in the expropriations law. Further evidence on the trajectory of investment projects—and the time delays and costs of each step along the way—would help inform the debate about how to reduce the “chokes and strangleholds” on public investment in Costa Rica. that increase the transaction costs of inter- that red tape declined in 2014.44 Small and acting with public institutions and decrease medium enterprises are the most affected the quality of services. The poor ranking of by red tape. Firms in the FTZs depend on Costa Rica in the Doing Business Indicators CINDE to expedite solutions for complying (see figure 4.28) is partly attributed to bur- with business regulations. Furthermore, the densome procedures within and between quality and efficiency of public services can public institutions. Reducing red tape will be affected by cumbersome procedures. facilitate firms’ operations, reducing trans- In sum, we find that the public sector has action costs of interacting with public insti- moved very slowly to adapt to the increasing tutions. Only one quarter of firms report sophistication and needs of the economy. Notes 1. IMF (2015). 8. Ministerio de Hacienda (2013). Cornick and 2. Garza et al. (2012). Trejos (2009) show that the share of pre-com- 3. Although in 2002 Costa Rica had a fiscal defi- mitted revenues has been high since the 1990s. cit higher than four percent of GDP, almost all 9. Estado de la Nacion (2014). of it was due to interest payments. Nowadays, 10. Public social spending in OECD countries aver- interest payments are around half that amount, ages 21.4 percent of GDP on health, social ser- and the overall deficit reflects a growing pri- vices, and social protection benefits, plus another mary deficit, following the steep increases in 5.4 percent of GDP on education, for a total public expenditures. of 26.8 percent of GDP. Sources: OECD 2014 4. For example, the wage increase was applied to update, OECD Social Expenditure Database teachers because they should receive the same (SOCX) and OECD Education at a Glance 2013. salaries or benefits increases as other parts of the 11. World Bank (2006b) central government (Estado de la Nación, 2014). 12. The share of households with fixed phone lines 5. Net present value. Estimates by IMF (2013). has decreased since 2014 as households have 6. IMF (2013). gained access to cell phone services. 7. For more details, see Ministerio de Hacienda 13. World Bank (forthcoming 2015). (2013), Estado de la Nacion (2014), and 14. OECD Social expenditures database (http:// Cornick and Trejos (2009). www​.oecd​.org/social/expenditure.htm). 132 Sustainability of Costa Rica’s Development Model 5. World Bank (2014b). 1 33. An “informal” settlement is typically a piece 16. World Bank (2014b). of occupied land for housing where occupants 17. Arroyo and Lindert (2104) and Jimenez have no legal title. See Estado de la Nacion (2014). (2014). 18. The estimated education expenditure for 34. GFDRR (2011) and World Bank (2013c). 2012 (5.3 percent of GDP) shown in figure 35. Governance is broadly defined as “the tradi- figure, as it takes 5.6 differs from the official ­ tions and institutions by which authority in into account only executed budget. See World a country is exercised” (Kaufmann, Kraay, & Bank (forthcoming 2015). Mastruzzi, 2010) or “a government’s ability to 19. OECD (2013). make and enforce rules, and to deliver ser- 20. Another option is to compare expenditure vices” (Fukuyama, 2013). Numerous studies per student in PPP US Dollars; however, the across a wide set of countries point to good largest education expenditure item is teacher governance being beneficial for growth, salaries (a non-tradable good), thus, even the investment, and finance, and bad governance PPP correction places too large a weight on leading to detrimental effects on economic high-income countries’ expenditures. outcomes. 21. Primary education receives 41 percent and 36. Cornick and Trejos (2009) highlight the prob- secondary 27 percent. Figures are for 2012. lematic ­features of the rules and procedures 22. OECD Social Expenditure database (http:// of Costa Rica’s legislature in more detail: “A www​.oecd​.org/social/expenditure.htm). two-thirds quorum is required not only when 23. World Bank (forthcoming 2015). votes are cast but even during debate: written 24. Much of these increases were directly linked discussion and submissions for the record are to an increase in the public sector’s base sal- not allowed. Bills must be voted on once in ary, which had repercussions for pension committee and then twice by the full Congress. payments, including non-­ contributory pen- Questions concerning a law’s constitutionality sions. See also the previous section on fiscal are referred to the courts between the first and accounts. the second vote; a minority of members may 25. Rofman and Oliveri (2011). even send the bill to the courts before the first 26. See IMF (2013). vote. Every member may propose an unlimited 27. About 66 percent of total energy consumption number of amendments to a bill and is allo- is based on oil products such as gasoline and cated a certain amount of time to argue for each diesel (Blanco 2014). proposal. Each motion to amend is debated and 28. ONF (2013). voted on separately, may be raised and re-raised 29. The GMA accounts for only four percent in committee, and then raised and re-raised of the national territory but concentrates again before the full house” (p. 165). more than 60 percent of the population and 37. This is also true for other administrations. approximately 75 percent of the vehicle fleet. 38. Estado de la Nación (2014). Besides fragmen- There are approximately 1.4 million cars in tation of public sector institutions, there is also Costa Rica for roughly 4.6 million people a fragmentation of programs, projects, or ser- (INEC, 2011 Census). vices provided (p. 337). 30. Estado de la Nación (2014). 39. World Bank and IDB (2010, p. 52). 31. According to the Natural Disaster Hotspot 40. Estado de la Nacion (2014). study by the World Bank, Costa Rica is num- 41. World Bank (2008). ber eight worldwide in economic risk exposure 42. IMF (2013). to three or more hazards. See GFDRR (2011). 43. World Bank (2008). 32. World Bank (2014d). 44. UCCAEP (2014). Sustainability of Costa Rica’s Development Model 133 Priority Areas, Linkages, and 6.  Complementarities Costa Rica’s development model has educational attainment. Moreover, despite sustained impressive achievements. Centered high spending on social protection benefits on its long-standing democracy, ambitious and the broader Social Compact, taxes and Social Compact, open economy model, and transfers have not proven to be effective in re- Green Trademark, the country has sustained distributing income to compensate for these healthy growth rates, improvements in social disparities. Consequently, rising inequality indicators, environmental gains, and one of offset the poverty-reducing impact of growth the lowest poverty rates in the Latin America in the late 2000s, and reversed what should and Caribbean (LAC) region. have been a decline in poverty between 2010 However, despite these impressive and 2014, with poverty increasing by 0.4 per- achieve­ ments, there are a number of emerg- centage points instead of falling by a projected ing challenges that will need to be addressed three percentage points during the post-­ crisis to main­ tain the country’s successful develop- “recovery” period. ment path. Second, although GDP has grown steadily First, despite reasonable growth and a in recent years, especially compared to the strong commitment to the Social Compact, Latin American Region as a whole, Costa poverty reduction has stagnated and inequal- Rica performance is not comparable to the ity is rising. The long-term trend suggests ris- top regional performers. Unlike the top re- ing or stagnating inequality across most of the gional performers (Chile, Panama, and past 25 years, in stark contrast to the signifi- Uruguay) Costa Rica’s per capita GDP had ­ cant decline in inequality in the broader LAC not shown any signs of convergence towards region. More recently, the gap between the the US level in the last 25 years. For example, rich and poor has widened significantly since in 1990, Costa Rica per capita income was the global crisis. Although growth recovered 13 percent of the US level, approximately as promptly after the global crisis, above the re- it is today. At the same time, the per capita gional average, job creation for low-skilled GDP of Chile, Panama, and Uruguay have workers has been feeble, contributing to ris- increased form 13, 12 and 9 percent of the ing unemployment and pushing returns to US per capita GDP in 1990 to 21, 17 and higher education upwards. Growth has been 17 percent today respectively. Production uneven, with lower growth and job creation costs that weaken the country’s competitive- in sectors that are more likely to employ un- ness prevent Costa Rica from joining the top skilled workers (e.g., construction and agri- growth performers. These productions costs culture). Not surprisingly, inequality has are affected by relatively high wages that increased, with the widening gap between the limit the country competitiveness in low earnings of rich and poor workers mirroring value added sectors, as well as by a number large disparities in human capital and of investment-climate related factors such as Priority Areas, Linkages, and Complementarities 135 rising electricity costs, weak infrastructure, the public sector to implement policies and and burdensome regulations. execute public investment projects. And third, fiscal pressures threaten to un- Several threads weave across this “tril- dermine the sustainability of the country’s ogy” of challenges for inclusion, growth, Social Compact and Green Trademark, and and ­sustainability—and point to priority prevent it from undertaking much needed areas for action. One thread involves the in- investments in public infrastructure. With- teractions between inequality and growth, out fiscal consolidation, the deficit could which hinge on the mismatch of skills and push public debt to unsustainable levels and jobs. Another strand is the dual challenge of threaten the country’s economic, social, and maintaining competitiveness of high val- environmental gains. The fiscal situation has ue-added sectors, while enhancing the via- deteriorated substantially since the global bility of traditional low value-added sectors. crisis, with the overall deficit of the Central Mounting fiscal pressures threaten the sus- Government growing to 5.6 percent of GDP tainability of the Social Compact and Green in 2013, and is projected to have surpassed Trademark. Finally, the challenges of gover- six percent in 2014 and to reach 6.6 in 2015. nance also weave across the development In tandem, public debt increased from 25 agenda, limiting the capacity of the public percent of GDP in 2008 to 37 percent in sector to adopt reforms, deliver services, 2013, with projections of 63 percent by 2019 and execute infrastructure projects. These unless corrective measures are imple- inter-­connected challenges highlight a mented. This recent deterioration stems number of priority areas that Costa Rica from counter-cyclical measures under-taken needs to address to continue on a sustain- during the crisis and structural forces that able and inclusive growth path. will require actions on multiple fronts. These links highlight a number of comple- Moreover, the current political landscape mentary priority areas for Costa Rica to con- and institutional framework add an addi- tinue on a sustainable and inclusive growth tional layer of complexity for approving and path. As in other countries, the list of policy implementing key reforms needed to address areas that could potentially help make prog- these emerging challenges. The shift from a ress toward poverty reduction and shared two-party to a multi-party system in the last prosperity is long. Through a combination of fifteen years has resulted in more complex diagnostics, benchmarking, and internal and and lengthy reform processes. The delays and external consultations, this Systematic Coun- difficulties in passing comprehensive and try Diagnostic (SCD) refines the long list to meaningful reforms, particularly on sensitive identify a set of cross-cutting areas with issues such as tax reform, has resulted in a strong linkages and complementarities. These mismatch between the demands of the popu- include: strengthening the education and lation and the “political delivery”. This is an training systems, boosting competitiveness increasing concern given the growing need to and reducing the infrastructure deficit, react and adapt quickly to changing global strength­ening governance, and undertaking a developments. Likewise, capacity constraints, variety of measures to ensure the sustainabil- related to weak sectoral planning and bureau- ity of the fiscal situation, the Social Compact, cratic inefficiencies, also affect the ability of and the Green Trademark (figure 6.1). 136 Priority Areas, Linkages, and Complementarities Figure 6.1  Priority Areas, Linkages, and Complementarities Priority Areas, Linkages & Complementarities Inclusion & concerns Growth & bout rising inequality Economic Model Competitiveness Education & Skills: Adapting to the changing labor market opportunities Business Infrastructure Governance Climate Sustainability of the Development Model Fiscal Social Compact Green Trademark Education and skills. Costa Rica needs value-added sectors, and that more workers to build a skilled workforce to supports its can access these better paying jobs, includ- trajectory towards a high value-added ing those in the bottom 40 percent of the economy and to reduce the skills-income population. As these changes are structural, gap. With fewer than half of young adults for the most part they are also long term in graduating from secondary school, and nature. Workers cannot just “acquire an with performance on test scores falling, education overnight.” As such, tackling ­ Costa Rica’s labor supply does not appear these challenges will require a three-pronged to be well adapted to generate the skills approach: (i) strengthening the quality, re- needed for the labor market. Thus, building tention, and relevance of the education sys- a more skilled workforce will ensure that tem (from pre-school through secondary the country remains competitive in high school)—which will help build the skills of Priority Areas, Linkages, and Complementarities 137 “tomorrow’s workers”; (ii) improving the further contributing to the skills shortage in quality and relevance of tertiary education; high value-added sectors. Moreover, outdated and (iii) strengthening the technical train- and bureaucratic procedures for recognizing ing system for the workforce. foreign degrees create obstacles for Costa Bold actions are needed to overhaul Rica to “import” the skills needed to sustain Costa Rica’s education system. Given the its high value-added growth model. Quality country’s level of development and high ed- and accountability mechanisms are also ucation spending, the education system se- needed, with performance agreements with riously underperforms in quality (as public universities and quality accreditation demon­ strated by test results), retention (low standards for both university and non-univer- completion rates), and relevance (as indi- sity tertiary education. cated by low returns to training and lower Finally, the country needs to expand the levels of education). Although high rates of offering and relevance of technical training, secondary school dropout are a symptom of which is the most direct way to build the the broader challenges in the system, imbal- skills of the current workforce. Again, stron- ances in the allocation of public spending ger quality standards, certification of techni- favor primary (41 percent) and tertiary ed- cal programs, and accountability of training ucation (32 percent) with relatively little al- institutes could help. The National Learning located to the secondary level (27 percent). Institute (INA) could also improve coordi- Indeed, both the share of public spending nation with private-sector employers to de- and the allocation per student in secondary sign market-relevant curricula and course education are low by international stan- offerings so as to better respond to the needs dards and given Costa Rica’s level of devel- of the growing sectors of the economy. opment. Moreover, inequities in learning outcomes start early in life—and affect mo- Competitiveness and the business cli- tivation and abilities to learn throughout mate. Boosting growth and inclusion in the school years. In addition to rebalancing the labor market requires confronting the spending towards secondary school and dual challenge of maintaining competitive- early childhood devel­ opment, Costa Rica ness of high value-added sectors, while im- needs to strengthen teacher quality and im- proving the viability of low value-added prove accountability through regular moni- sectors. For instance, improving the integra- toring with standardized learning tion of export-oriented and domestic firms assessments, and a more effective gover- through backward linkages could sustain the nance and incentive framework. growth among small and medium enter- Given the high-skilled profile of job op- prises (SMEs), generating jobs in mid- and portunities in Costa Rica, systemic efforts are low-skilled occupations. This can be done by also needed to enhance the quality and rele- lowering operation costs to improve the pro- vance of tertiary education. Currently, the ductivity of labor and counterbalance the tertiary education system is heavily biased to- high labor costs in Costa Rica compared to wards social science and humanities, and its neighbors, for example by lowering the produces few STEM (Science, Technology, costs of doing business through regulatory Engineering and Mathematics) graduates, simplification. 138 Priority Areas, Linkages, and Complementarities Infrastructure. Reducing the infrastruc- for private sector participation in the main- ture deficit would increase competitiveness, tenance and upgrading of its infrastructure. growth, and environmental sustainability. Both growth and inclusion would benefit Costa Rica’s historical efforts to build an ex- from the many complementarities involved tensive network of infrastructure in nearly in improving education, competitiveness, all productive service areas (water, sanita- and infrastructure. A well-educated work- tion, transport, electricity, and telecommuni- force with relevant skills is fundamental for cations) are clear from the infrastructure sustaining economic growth and increasing stock: the country has two times the road productivity. In parallel, closing the educa- and three times the rail density of the aver- tion gap between the poor and non-poor is age middle-income country; access to elec- also highly relevant for inclusion by provid- tricity is nearly universal; and mobile ing opportunities those in the bottom penetration is higher than the OECD aver- 40 percent. Lowering the costs of doing busi- age. Yet, the near freeze in public infrastruc- ness will boost competitiveness across vari- ture investment until the 1990s, as well as ous sectors. Furthermore, increasing recent fiscal constraints, have taken a toll on infrastructure spending would stimulate the country’s ability to upgrade and maintain construction, thereby creating more jobs for its infrastructure. Further, the government the large stock of low-skilled workers. has faced significant challenges in executing At the same time, actions are needed to infrastructure investment in a timely man- ensure the sustainability of Costa Rica’s ner. As a result, today roads and ports have ­development model: among the lowest quality marks in the LAC region. Electricity prices have doubled since • Fiscal Sustainability: Improving the fiscal 2007 due to weather related variable hydro- stance to restore sustainability requires electric output, causing increased use of reforms to manage expenditures and thermal units and high operating costs, increase revenues. On the expenditure among other factors. This infrastructure def- front, these include containing the wage icit reduces the potential of local firms to bill of the consolidated public sector, as grow and create jobs, and this is true in par- public sector wages, both in government ticular for firms that operate outside Free and more so in state-owned enterprises Trade Zones (FTZs). In addition, the coun- and other public institutions, are well try needs to improve its waste management above the private sector at all employ- and clean energy production capabilities to ment categories; and reviewing the fiscal be able to reduce GHG emissions, and water sustainability of the pension system, and soil pollution. Infrastructure improve- particularly of special pension regimes ment poses a number of tradeoffs, including in the public sector. In addition, a com- the need to intervene in protected areas (in prehensive reform of the budgetary the case of clean energy production), as well process is needed to increase efficiency as the need to control the current fiscal defi- and transparency in all public sector cit. Given the necessity of continuing to in- entities. Finally, curtailing the earmark- vest in infrastructure, and the reality of fiscal ing of revenues, which cover more than constraints, Costa Rica must look for options half of primary spending, would make Priority Areas, Linkages, and Complementarities 139 the budgetary process more flexible. the PES mechanism for conservation; Comprehensive reforms are also needed modernizing water and solid waste (ii) ­ to increase revenues. For example, the treatment (infrastructure, service de- 1,259 tax exemptions approved since livery, regulatory framework, capacity); 1953 that comprise almost six percent (iii) increasing the supply of renewable of GDP need to be thoroughly reviewed. energy by making regulations on the use Curtailing those exemptions would of protected areas more flexible; (iv) im- make the tax system more rational and plementing a comprehensive transport progressive, as well as produce higher policy, including measures to reduce revenues. This reform is essential for growth of demand for energy ­ associated restoring sustainability to the fiscal ac- with transport; (v) reinforcing regulation counts, which is a necessary condition and oversight of agro-chemical use and for achieving Costa Rica’s economic and incentivizing the expansion of “green” social objectives. (organic) agriculture; and (vi) improving • Social Sustainability: In addition to territorial planning, land management, strengthening education, priority areas and management of natural and man- for sustaining the Social Compact in- made disasters. cluding health and social protection. Costa Rica needs to modernize its uni- Governance. Progress in all the priority areas versal health system to improve quality discussed above hinges on improving the ca- by: (i) strengthening the health care pacity of the public sector to plan and imple- model to enhance capacity to adapt ment policies, execute public investment to demographic and epidemiological projects, deliver services, and increase account­ change while ensuring quality and ability. Despite Costa Rica’s good governance timeliness of service delivery; (ii) im- levels compared to the rest of the LAC region, proving the financial model of budget there is a growing perception of low effective- and resource allocation; and (iii) im- ness of government institutions. Cumbersome proving the management model for the regulations, in many cases resulting from lack health system for accountability, effi- of coordination among institutions, make the ciency, and performance. The country process of starting and running a business—­ also needs to increase the effectiveness particularly a non-FTZ small or medium en- of social protection programs by (i) har­ terprise—more challenging. Low levels of monizing eligibility criteria and social transparency and accountability lower the effi- information systems; (ii) improving ciency of public spending. The current political performance monitoring and evalu- landscape, where political minorities have the ation; and (iii) reducing institutional power to delay votes, further reduces the mar- and program fragmentation. gin for approving and implementing needed • Environmental Sustainability: To sustain reforms. The need to improve governance is its celebrated “Green Trademark,” Costa apparent in all priority areas, for instance by Rica needs to balance environmental increasing accountability in the education sec- and natural resource management goals, tor (e.g., by tracking student achievement to by: (i) reviewing the sustainability of reward teacher and school performance). 140 Priority Areas, Linkages, and Complementarities Employment creation would benefit from tackle the issues presented above. For exam- streamlining business regulations as well as the ple, the specific factors driving secondary public procurement and investment processes dropout are still not well understood. In the to improve infrastructure. A more consoli- labor market, it is not clear whether the con- dated budget, fewer tax exemptions, and more traction of employment in agriculture, man- control over spending by autonomous institu- ufacturing and construction is of a cyclical tions could greatly help to reduce the current or structural nature, and this has important fiscal deficit, and would improve the capacity implications for low-skilled unemployment. to monitor results of public spending. In turn, Like­wise, although there are strong signs that results-based management would help to boost reservation wages are high, there could be the efficiency of public spending, for example, more studies to quantify this phenomenon enforcing the use of common information sys- better. To improve the efficiency of the pub- tems and modernizing the M&E frameworks lic sector, it is crucial to identify the specific in the social sectors. governance bottlenecks in executing infra- structure projects, as well as their cost impli- An agenda for knowledge. Finally, cations. Also, the articulation mechanisms to a number of knowledge gaps need to be improve the effectiveness of social programs filled to inform better policy decisions. Al- need to be based on a thorough institutional though there is a large and productive re- mapping of social programs. Finally, envi­ search and policy analysis community ron­ mental conservation needs to be better studying Costa Rica, there are a few issues linked with economic activity, and a key where having further research and data col- knowledge gap in this regard is how to link lection would provide more information to sustainable production and rural landscapes help design more concrete policy reforms to to conservation. Priority Areas, Linkages, and Complementarities 141 Appendix A: Team Engagement and Consultation Process The Systematic Country Diagnostic September/October 2014—Brainstorming (SCD) core team followed a highly inclusive sessions: the SCD team held two broad process in the development of the final prod­ brainstorming sessions in which the partici- uct and the elaboration of the diagnostics. In pants discussed: (i) the revised hypotheses this appendix, the collaborative steps fol- and proposed means of analyzing the know­ lowed towards the preparation of the draft ledge gaps; and (ii) a draft of the overall SCD document are detailed. storyline that translated the hypotheses into a full structure around growth, inclusiveness, Team collaboration steps within and sustainability following the SCD the World Bank Group guidelines. March 2014—Initial Country Team brain- October/November 2014—Bilateral storming: The initial task team organized a consultations: Several rounds of bilateral half-day country team meeting to discuss the consultations with sectorial teams were held guidelines for SCD and to consult on an to focus on the remaining knowledge gaps initial power-point on key challenges for that could be filled prior to the elaboration of adequate growth, inclusive growth, and the overall storyline. sus­tainable growth. December 2014—Quality Enhancement April 21, 2014—Concept Note Review Review: Chaired by the Latin American and Meeting: The review meeting took place with Caribbean (LAC) Chief Economist, the a concept note that laid out the rationale, Quality Enhancement Review allowed the objectives, and approach for the task, as well SCD team to received feedback on the as the work plan going forward. Some back­ storyline from peer reviewers. ground papers were commissioned after the February 2015—Prioritization work- concept note review. shop: A facilitated workshop with the broad August/September 2014—Formulation of Costa Rica Team was held with the objec- hypotheses: The SCD preparation process tive of soliciting inputs on identified prio­ started with the formulation of a set of 10 rity areas for growth, inclusion, and general hypotheses around key areas for sustainability. Key constraints and solution Costa Rica’s growth and poverty reduction areas were identified for each priority area. patterns based on the review of the literature. The discussion revolved around how to The country team and key specialists in simplify the list of priority areas and which specific areas provided feedback and com- priorities were considered but discarded. At ments on their validity as well as additional the end, there was consensus on the cross-​ information to substantiate, refine, or change cutting nature of the proposed priority these hypotheses, or to dismiss them. areas, and some of them could be combined Appendix A: Team Engagement and Consultation Process 143 to provide a better entry point with the A preparation mission was conducted during government counterparts. September 22–26, in which the SCD team March 2015—Regional Operation participated in the “Encerrona” event orga­ Committee Decision Meeting: Chaired by nized with the counterparts in Costa Rica to the LAC Vice President, the Decision discuss the World Bank engagement. The Meeting provided further feedback and SCD team held meetings with counterparts guidance to the team. in Costa Rica from the government, private sector, academia, and think tanks to get their Engagement with Costa Rica views on the poverty and inequality trends, The SCD preparation was accompanied by a as well as on the growth pattern of Costa consultation process in Costa Rica to ensure Rica. Two more consultation missions were that key stakeholders provided inputs into carried out in November 2014 and January the deliberations and shared early findings. 2015. 144 Appendix A: Team Engagement and Consultation Process Appendix B: Country Comparators To benchmark Costa Rica’s performance, tons emissions (the world average) per capita this report uses six comparable groups of were included in the selection process; (iv) peers: Central American countries, Latin geography, which often determines compara- American peers, upper-middle-income tive advantage and in some cases historical countries, the average for the World, struc- similarities; landlocked econo­ mies and small tural peers, and aspirational peers. island states were excluded, while several Organisation for Economic Co-operation Central American countries were included; and Development (OECD) countries are and (v) only non-fragile states were selected. used as “aspirational peers” given Costa The use of these criteria resulted in the Rica’s ongoing accession talks to the OECD. following set of countries: Chile, Croatia, The structural peers were selected using the Dominican Republic, Lithuania, Panama, “Find your friends” tool. and Uruguay. The group of structural peers includes countries that provide appropriate bench- GDP CO2 Population emissions marks for answering SCD-relevant questions Country per capita in 2014 (metric tons ($US) such as whether or not certain conditions, per capita) policies, or economic performances in Costa Costa Rica 9,722 4.86 1.7 Rica are adequate. The criteria and filters for Structural peers selection were the following: (i) GDP per Chile 15,205 17.71 4.2 capita between US$5,500 and US$15,500, as Croatia 13,638 4.28 4.7 Costa Rica is an upper-middle-income Dominican 5,710 10.6 2.1 country; (ii) population between 2.5 million Republic and 20 million people; (iii) CO2 emissions: Lithuania 14,801 2.96 4.4 Costa Rica is one of the most environmen- Panama 9,798 3.79 2.6 tally sustainable countries in the world, and Uruguay 15,054 3.40 2.0 only countries with less than 5 CO2 metric Source: Find my Friends Tool 2014. Appendix B: Country Comparators 145 References Adduci, R., E. Pineda, and R. Villate, R. 2013. Compete 2014: Trade Logistics in the Global “Networking Skills in Latin America.” White Economy. The Logistics Performance Index Paper, IDC, Mexico D.F. and Its Indicators. Washington, DC: The Aedo, C., and I. Walker. 2012. Skills for the 21st World Bank. Century in Latin America and the Caribbean. Azevedo, J. P., V. Sanfelice, and M. C. Nguyen. Washington, DC: World Bank. 2012. “Shapley Decomposition by Components Aedo, C., J. Hentschel, J. Luque, and M. Moreno. of a Welfare Measure.” Working Paper World 2014. “From Occupations to Embedded Skills: Bank, Washington, DC. a Cross-Country Comparison.” Policy Research Bamber, P., and G.Gereffi. 2013. Costa Rica in the Working Paper WPS6560, World Bank, Medical Devices Global Chain: Opportunities Washington, DC. for Upgrading. Durham, NC: Duke University, Andrés, L., J. Schwartz, and J. L. Guasch. 2013. Center for Globalization, Governance, and Uncovering the Drivers of Utility Performance: Competitiveness. Lessons from Latin America and the Caribbean BCCR (Banco Central de Costa Rica). 2014. Comen​ on the Role of the Private Sector, Regulation, and ­ tario sobre el desempeño de la economía nacional Governance in the Power, Water, and Telecom­ en el 2013 y el Programa Macroeconómico 2014–15. munication Sectors. Directions in Develop­ ment, San Jose, Costa Rica: BCCR. Infrastructure. Washington, DC: World Bank. Barquero, M. 2013. “43 empresas extranjeras in- Araujo, J., M. Brueckner, M. Clavijo, virtieron en Costa Rica.” La Nacion, December E. Vostroknutova, and K. Wacker. 2014. 19. Retrieved from: http://www.nacion.com​ Benchmarking the Determinants of Economic /economia/empresarial/empresas​-extranjeras​ Growth in Latin America and the Caribbean. -invirtieron-Costa-Rica_0_1385261486.html. Washington, DC: World Bank. Barquero, M. 2014. “Cinde espera atraer a Costa ARESEP (Autoridad Reguladora de los Servicios Rica 35 empresas nuevas en el 2015.” La Públicos). 2014. Situación de las tarifas eléctricas Nacion, December 19. Retrieved from: http:// en Costa Rica. San Jose, Costa Rica: ARESEP. www​.nacion.com/economia​/­empresarial​ Retrieved from http://www​.­dialogoenergiacr​ /Cinde​-inversiones-Jorge_Sequeira-servicios​ .com. -manufactura_0_1458254300.html. Arroyo, L., and P. Lindert. 2014. “Fiscal Redistri­ Barro, R., and J. Lee. 2010. “A New Data Set of bution in Latin America Since the Nineteenth Educational Attainment in the World, 1950– Century.” Chapter prepared for the conference 2010.” Journal of Development Economics, volume on Latin American Inequality in the 104:184–198. Long Run, edited by Luis Bértola and Jeffrey Barros, R. P., M. Carvalho, S. Franco, and G. Williamson. Washington, DC: Inter-American R. Mendoça. 2006. “Una Análise das Principais Development Bank and World Bank. Causas da Queda Recente na Desigualdade de Arroyo, N., R. Lücke, and L. Rivera. 2013. Análisis Renda Brasileira.” Revista Econômica 8 (1): sobre el mecanismo actual para la estimación y 117–147. determinación de los precios del arroz bajo el con­ Beck, T., A. Demirgüç-Kunt, and R. Levine. 2007. texto de la cadena de comercialización, Informe “Finance, Inequality, and the Poor.” Journal of Final PARA EL MINISTERIO DE ECONOMÍA, Economic Growth 12 (1): 27–49. INDUSTRIA Y COMERCIO (MEIC). San Jose, Bennet, K., and N. Henninger. 2009. “Payments Costa Rica: IICEE, Universidad de Costa Rica. for Ecosystem Services in Costa Rica and Arvis, J. F., D. Saslavsky, L. Ojala, B. Shepherd, Forest Law No. 7575: Key Lessons for C. Busch, and A. Raj. 2014. Connecting to Legislators.” World Resources Institute. References 147 Bilagher, M. 2014. “Resultados comparados CEPAL (Comisión Económica para América SERCE-­​TERCE.” Presentation, UNESCO Latina y el Caribe). 2014b. Centroamérica: Regional Bureau for Education in Latin Estadísticas del Subsector Eléctrico, 2013. America and the Caribbean, Santiago de Santiago de Chile: CEPAL. Chile. Contraloría General de la Republica. 2014a. Blanco, J.M. 2014. “Energía en Costa Rica: un Evolución Fiscal y Presupuestaria del Sector ­ desafío perenne.” Retrieved from: http://www​ Publico, Primer Semestre de 2014. San Jose, .academiaca.or.cr​/­wp-content/uploads/2014​ Costa Rica. /09/Presentacion-Energia.pdf. Contraloría General de la Republica. 2014b. Botero, J. 2014. Recognition of Foreign Higher Memoria Anual 2013 San Jose, Costa Rica. Education Degrees in Costa Rica: A Summary of Cord, L., O. Barriga-Cabanillas, L. Lucchetti, Models and Good Practices. Education Global C. Rodríguez-Castelán, L. Sousa, and Practice, Latin America and the Caribbean D. Valderrama. 2014. “Inequality Stagnation in Region. Washington, DC: World Bank. Latin America in the Aftermath of the Global Brown, Jessica, and Neil Bird. 2010. “Costa Rica’s Financial Crisis.” Policy Research Working Sustainable Resource Management: Successfully Paper 7146, World Bank, Washington, DC. Tackling Tropical Deforestation.” Overseas Cornick, J., and A. Trejos. 2009. “Political and Development Institute. Institutional Obstacles to Reform in Costa Bucheli, M., N. Lustig, M. Rossi, and F. Amábile. Rica.” In Growing Pains in Latin America: An 2014. “Social Spending, Taxes, and Income Economic Growth Framework as Applied to Redistribution in Uruguay.” In Fiscal Policy, Brazil, Colombia, Costa Rica, Mexico, and Poverty and Redistribution in Latin America, Peru, edited by L. Rojas-Suarez. Washington, Special Issue, Public Finance Review, edited by DC: Center for Global Development. N. Lustig, C. Pessino, and J. Scott. Cornick, J., J. Jimenez, and M. Román, M. 2014. Cadena, K., A. Cardozo Silva, L. Lucchetti, and “Public-Private Collaboration on Productive K. Scott. 2013. Central America in the New Development Policies in Costa Rica.” IDB Millen­ nium: Six Different Stories of Poverty Work­ ing Paper Series No. IDB-WP-480, and Inequality. Washington, DC: World Bank. Inter-­American Development Bank, Calderón, César, and Luis Servén. 2004.Trends in Washington, DC. Infrastructure in Latin America, 1980–2001. Crespi, Gustavo, Eduardo Fernández-Arias, and Office of the Chief Economist, Latin America Ernesto Stein, eds. 2014. Development in the and the Caribbean Regional Office, Finance, Americas (DIA) 2014. Rethinking Productive Private Sector, and Infrastructure Department. Development: Sound Policies and Institutions Washington, DC: The World Bank. for Economic Transformation. Washington, Calderon, Cesar, and Luis Serven. 2003. The Out­ DC: Inter-American Development Bank. put Cost of Latin America’s Infrastructure Gap. Cruces, G., P. Lanjouw, L. Lucchetti, E. Perova, Washington, DC: World Bank. R. Vakis, and M. Viollaz. 2011. “Intragenera­ Chomitz, K.M., E. Brenes, and L. Constatino. tional Mobility and Repeated Cross-Sections: A 1998. Financing Environmental Services: the Three-Country Validation Exercise.” Policy Costa Rican Experience and Its Implications. Research Working Paper 5916, World Washington, DC: World Bank. Bank,Washington, DC. CINDE (Coalición Costariccense de Iniciativas de Dang, H., and P. Lanjouw. 2014. “Measuring Desarrollo). 2015. Costa Rica: Medical Devices Poverty Dynamics with Synthetic Panels Based Hotspot. San Jose, Costa Rica: CINDE. on Cross-Sections.” Policy Research Working CEPAL (Comisión Económica para América Paper WPS6504, World Bank, Washington, DC. Latina y el Caribe). 2014a. Base de datos de in­ Dang, H., P. Lanjouw, J. Louto, and D. McKenzie. versión en infraestructura económica en América 2014. “Using Repeated Cross-Sections to Ex­ Latina y el Caribe, 1980–2012. Santiago de plore Movements in and out of Poverty.” Jour­ Chile: CEPAL. nal of Development Economics (107): 112–128. 148 References Datt, G., and M. Ravallion. 1992. “Growth and Fukuyama, F. 2013. What Is Governance? Redistribution Components of Changes in Washington, DC: Center for Global Poverty Measures: A Decomposition with Development. Applications to Brazil and India in the Garza, M., P. Morra, and D. Simard. 2012. “The 1980s.” Journal of Development Economics Fiscal Position: Prospects and Options for 38: 75–95. Adjustment.” In Central America, Panama, De la Torre, A., E. Levy Yeyati, G. Beylis, T. Didier, and the Dominican Republic - Challenges fol­ C. Rodríguez Castelán, and S. Schmukler. 2014. lowing the 2008-09 Global Crisis, edited by “Inequality in a Lower Growth Latin America.” M. Pinon, A. Lopez-Mejia, M. Garza, and LAC Semiannual Report (October). Washington, F. Delgado. Washington, DC: International DC: World Bank. Monetary Fund. DeShazo, J., and L. Monestel Vega. 1999. La Impor­ Gereffi, G., P. Gamber, S. Frederick, and tancia de las Áreas Protegidas Públicas en el K. Fernandez-Stark. 2013. Costa Rica in Global Desar­rollo del Turismo en Costa Rica: Un Análisis Value Chains: An Upgrading Analysis. Duke de su Visita. Harvard Institute for International Center for Globalization, Governance, and Development. Competitiveness. Duhá Buchsbaum, B. 2004. “Ecotourism and Sus­ GFDRR (Global Facility for Disaster Reduction tainable Development in Costa Rica.” Masters and Recovery). 2011. Disaster Risk Manage­ Thesis, Virginia Polytechnic Institute and State ment in Central America: Costa Rica. Global University. Facility for Disaster Reduction and Dypski, M. 2002. “The Caribbean Basin Initiative: Recovery Country Notes. Washington, DC: An Examination of Structural Dependency, World Bank. Good Neighbor Relations, and American Gindling T. H., and J. D. Trejos. 2004. “Account­ Investment.” Journal of Transnational Law and ing for Changing Earnings Inequality in Costa Policy 12 (1): 95–136. Rica, 1980-1999.” UMBC Department of Elbers, C., J. Lanjouw, and P. Lanjouw. 2003. Economics Working Paper 03-108, University “Micro-­ Level Estimation of Poverty and of Maryland Baltimore County, Baltimore, MD. Inequality.” Econometrica 71 (1): 355 –364. Gindling, T. H., and K. Terrell. 2004. “Legal El Financiero. January 8, 2015. “Gobierno eleva Minimum Wages and the Wages of Formal arancel y anuncia baja en precio del arroz al and Informal Sector Workers in Costa Rica.” consumidor.” http://www.elfinancierocr.com​ UMBC Department of Economics Working /economia-y-politica/Gobierno​-anuncia-­ Paper 04-102, University of Maryland precio-arroz-consumidor_0_661733827.html. Baltimore County, Baltimore, MD. Fajnzylber, P., J. Guasch, and J. Lopez. 2009. Does Gindling, T. H., and L. Oviedo. 2008. “Single the Investment Climate Matter? Microeconomic Mothers and Poverty in Costa Rica.” IZA DP Foundations of Growth in Latin America. No. 3286, Institute for the Study of Labor, Washington, DC: World Bank. Bonn, Germany. Fernández, R., S. Flórez Gómez, F. Estrázulas de Gindling, T. H., N. Mossaad, and J. D. Trejos. 2013. Souza, and H. Vaga. 2011. “Supply Chain Las consecuencias del aumento en la observación Analyses of Exports and Imports of Agricul­ del salario mínimo legal en Costa Rica: una eval­ tural Products: Case Studies of Costa Rica, uación del impacto de la Campaña Nacional de Honduras, and Nicaragua.” In Getting the Most salarios mínimos. San José, Costa Rica: Uni­ Out of Free Trade Agreements in Central versidad de Costa Rica, IICE. America, edited by J. Lopez and R. Shankar. Gonzalez-Vega, C., and V. Cespedes. 1993. “Part Washington, DC: World Bank. I: Costa Rica.” In A World Bank Comparative Ferreira, F., J. Messina, J. Rigolini, L. López- Study. The Political Economy of Poverty, Equity, Calva, M. Lugo, and R. Vakis. 2013. Economic and Growth. Costa Rica and Uruguay, 3-186, Mobility and the Rise of the Latin American edited by S. Rottenberg. Washington, DC: Middle Class. Washington, DC: World Bank. World Bank. References 149 Groote, R. 2005. “Costa Rica: Proyecto de Kolenikov, S., and A. Shorrocks. 2005. “A Desarrollo de Proveedores para Empresas Decomposition Analysis of Regional Poverty Multinacionales de Alta Tecnología.” ATN/ in Russia.” Review of Development Economics ME-6751-CR, Final evaluation. IDB. 9 (1): 25–46. Gutiérrez Saxe, M., and F. Straface. 2009. Krugman, Paul. 2004. The Great Unraveling: “Democracia estable ¿alcanza? Análisis de la Losing Our Way in the New Century. New gobernabilidad en Costa Rica.” Publicaciones York: W.W. Norton & Company. Especiales sobre el Desarrollo 4. Washington, Latinobarómetro. 2011. Informe 2011. Santiago DC: Inter-American Development Bank. de Chile: Corporación Latinobarómetro. Higgins, S., and C. Pereira. 2014. “The Effects of Latinobarómetro. 2013. Informe 2013. Santiago Brazil’s Taxation and Social Spending on the de Chile: Corporación Latinobarómetro. Distribution of Household Income.” In Fiscal Larrain, F., L. F. López-Calva, and A. Rodríguez- Policy, Poverty and Redistribution in Latin Clare. 2000. “Intel: A Case Study of Foreign America, Special Issue, Public Finance Review, Direct Investment in Central America.” CID edited by N. Lustig, C. Pessino, and J. Scott. Working Paper No. 58, Center for International Hornbeck, J. 2012. The Dominican Republic- Development at Harvard University. Central America-United States Free Trade Agree­ Lederman D., J. Messina, S. Pienknagura, and ment (CAFTA-DR): Developments in Trade and J. Rigolini. 2014. Latin American Entrepreneurs: Investment. Washington, DC: Congres­ sional Many Firms but Little Innovation. Washington, Research Services Report for Congress. DC: World Bank. IDB (Inter-American Development Bank). 2010. Loayza, N., P. Fajnzylber, and C. Calderón. 2005. “Productive Development Policies in Costa Economic Growth in Latin American and the Rica: Market Failures, Government Failures, Caribbean: Stylized Facts, Explanations, and and Policy Outcomes.” IDB Working Paper Forecasts. Washington, DC: World Bank. Series No. IDB-WP-157. Loría Sagot, M. A. 2013. El Sistema Financiero IMF (International Monetary Fund). 2013. Costarricense en los ultimos 25 anhos: Estructura Costa Rica—Options to Rationalize Central y Desempeno. San Jose, Costa Rica: Academia Government Expenditure October 2013. de Centroamérica. Washington, DC: IMF. Loría, M., and C. Umaña. 2014. La gestión de la IMF (International Monetary Fund). 2015. infraestructura pública en Costa Rica: el caso de Costa Rica—Staff Report for the 2014 Article la red vial. San Jose, Costa Rica: Academia de IV Consultation. Washington, DC: IMF. Centroamérica. Jaramillo, M. 2014. “The Incidence of Social Lücke, R. 2013. “Crecimiento, sector externo, Spending and Taxes in Peru.” In Fiscal Policy, empleo e ingresos.” Background paper for Poverty and Redistribution in Latin America, Vigésimo Informe Estado de la Nación en Special Issue, Public Finance Review, edited by Desarrollo Humano Sostenible. San José, Lustig, N., C. Pessino and J. Scott. Costa Rica: Programa Estado de la Nación. Jiménez, R. 2014. “Educación pública en Costa Lustig, N., and C. Pessino. 2014. “Social Spending Rica: políticas, resultados y gasto.” Análisis and Income Redistribution in Argentina in the Serie 6, Academia de Centroamérica, San Jose, 2000s: The Rising Role of Noncontributory Costa Rica. Pensions.” In Fiscal Policy, Poverty and Redis­ Kaufmann, D., A. Kraay, and M. Mastruzzi. 2010. tribution in Latin America, Special Issue, Public “The Worldwide Governance Indicators— Finance Review, edited by N. Lustig, Methodology and Analytical Issues.” Policy C. Pessino and J. Scott. Research Working Paper No. 5430, World Lustig, N., L. F. Lopez-Calva, and E. Ortiz-Juarez. Bank: Washington, DC. 2013. “Deconstructing the Decline in Inequal- Kent, P. 2011. “How Fit are Central America’s ity in Latin America.” Policy Research Work- Ports? An Exercise in Measuring Port ing Paper Series 6552, World Bank, Performance,” Working Paper, Washington, DC. ­Washington, DC. 150 References MEIC (Ministerio de Economía, Industria y Montenegro, F. 2013. Costa Rica Case Study: Comercio). 2014. Estudio sobre la caracteri­ Primary Health Care Achievements and zación de la oferta financiera y no financiera en Challenges within the framework of the Social apoyo a las PYME en 2013. Departamento de Health Insurance. Washington, DC: Política Pública de la Dirección General de World Bank. Apoyo a la PYME. San Jose, Costa Rica: Morales, N. 2013. “Aspectos relevantes en desi- MEIC. gualdad y pobreza al 2012.” Background paper MIGA. 2006. The Impact of INTEL in Costa Rica: for the Decimonoveno Informe Estado de la Nine Years after the Decision to Invest. Investing Nación en Desarrollo Humano Sostenible. San in Development Series. Washington, DC: José, Costa Rica: Programa Estado de World Bank. la Nación. Ministerio de Hacienda. 2013. En Ruta hacia la OECD (Organisation for Economic Co-operation Consolidación Fiscal: Agenda para un Diálogo and Development). 2012. Attracting Knowledge- Nacional. San José, Costa Rica: Ministerio de Intensive FDI to Costa Rica: Challenges and Hacienda. Policy Options. OECD Development Centre. Monge-Ariño, F. 2011. “Costa Rica: Trade Opening, OECD (Organisation for Economic Co-operation FDI Attraction, and Global Production Sharing.” and Development). 2013. Education at a Glance Staff Working Paper ERSD-2011-09, World 2013. Programme for International Student Trade Organization, Geneva. Retrieved from: Assessment. Paris: OECD. http://www.wto.org/english/res_e​/­reser_e​ OECD (Organisation for Economic Co- /­ersd201109_e.pdf. operation and Development). 2014. PISA Monge-Gonzales, R., J. Rosales-Tijerino, and 2012 Results in Focus: What 15-Year-Olds G. Arce-Alpízar. 2005. Cost-Benefit Analysis of Know and What They Can Do with What the Free Trade Zone System. The Impact of They Know. Programme for International Foreign Direct Investment in Costa Rica. Student Assessment. Paris: OECD. Washington, DC: Organization of American ONF (Oficina Nacional Forestal). 2013. Usos y States Office of Trade, Growth, and Aportes de la Madrea en Costa Rica. San Jose, Competitiveness. Costa Rica: ONF. Monge Gonzalez, R., L. Rivera, and J. Rosales Patton, D.B., and R. Moore. 2012. Linking Tijerino. 2010. “Productive Development Policies Manufacturing SMEs to Global Value Chains: in Costa Rica: Market Failures, Government The Case of Metalworking and Plastics in Costa Failures, and Policy.” IDB Working Paper Series Rica. Kennedy School, Harvard University. No. IDB-WP-157, Inter-American Development Paus, E., and K.P. Gallagher. 2008. “Missing Links: Bank, Washington, DC. Foreign Investment and Industrial Development Monge Gonzalez, J. Leiva, and J. Rodriguez- in Costa Rica and Mexico.” St Comp Int Dev Alvares. 2012. “Movilidad laboral y der- 2008 (43): 53–80. rames de conocimiento desde las compañías Paz Arauco, Molina, V., G. Pozo, W. and E. Aguilar. multinacionales en Costa Rica: nuevos em- 2014. “Explaining Low Redistributive Impact in prendimientos y externalidades positivas so- Bolivia.” In Fiscal Policy, Poverty and Redis­ bre las empresas locales.” Working Paper tribution in Latin America, Special Issue, Public Series, Inter-American Development Bank, Finance Review, edited by N. Lustig, C. Pessino, Washington, DC. and J. Scott. Monge-Gonzalez, R., and J.A. Rodriguez-Alvarez. Programa Estado de la Nación. 2012. Decimoc­ 2013. “Impact Evaluation of Innovation and tavo Informe Estado de la Nación en Desarrollo Linkage Development Programs in Costa Humano Sostenible. San José, Costa Rica: Pro- Rica: The Cases of PROPYME and CR grama Estado de la Nación. Provee.” IDB Working Paper No. IDB- Programa Estado de la Nación. 2013. Cuarto WP-461, Inter-American Development Bank, Informe Estado de la Educación. San José, Washington, DC. Costa Rica: Programa Estado de la Nación. References 151 Programa Estado de la Nación. 2014. Vigésimo Policy, Poverty and Redistribution in Latin Informe Estado de la Nación en Desarrollo America, Special Issue, Public Finance Review, Humano Sostenible. San José, Costa Rica: edited by N. Lustig, C. Pessino, and J. Scott. Programa Estado de la Nación. SEDLAC (Socio-Economic Database for Latin Ramirez, E. 2015. “Apollo Endosurgery fabricará America and the Caribbean). Center for dispositivos médicos en Costa Rica.” La Nacion Distributive, Labor and Social Studies January 16. Retrieved from http://www.nacion​ (CEDLAS) of Universidad de La Plata, .com/economia/empresarial​/Coyol-zona- Argentina, and World Bank, Washington, DC. franca​-ciencias-vida-dispositivos​-medicos_0​ http://sedlac.econo​.unlp.edu.ar/eng. _1463853683​.html. Straface, F., and J. Vargas Cullell. 2010. “Capítulo Rodriguez, O. 2013. “Costoso alquiler de plan- V - Gestión Pública.” In Democracia estable tas eleva factura a clientes de ICE.” La Nación ¿alcanza? Análisis de la gobernabilidad en October 1. Retrieved from: http://www​ Costa Rica, edited by M. Gutiérrez Saxe and .nacion.com/economia/consumidor/Costoso​ F. Straface, Washington, DC: Inter-American -alquiler-plantas-clientes-ICE_0_1371462854​ Development Bank. .html. Swiston, A., L. D. Barrot. 2011. The Role of Rodriguez Zunega, J.M. 2003. “Payment for forest Structural Reforms in Raising Economic Growth environmental services: the Costa Rican in Central America. Western Hemisphere. Experience.” Unasylva 54 (212): 31–33. Washington, DC: International Monetary Fund. Rofman, R., and L. Oliveri. 2011. “La cobertura de Trejos, A. 2008. “Country Role Models for los sistemas previsionales en América Latina: Development Success: The Case of Costa conceptos e indicadores.” Serie de Documentos Rica.” Helsinki. Retrieved from http://www​ de Trabajo sobre Políticas Sociales Nº 7, World .econstor.eu/handle/10419/45106. Bank, Washington, DC. Trejos, J. D., and D. Murillo. 2012. “El índice de Roller, L.H., and L. Waverman. 2001. “Telecom­ oportunidades educativas Un indicador resu- munications Infrastructure and Economic men de la equidad en la educación.” Development: A Simultaneous Approach.” The Background paper for the Cuarto Informe American Economic Review 91 (4). Estado de la Educación. San José, Costa Rica: Sáenz, M. 2014. “Finanzas del Gobierno Central Programa Estado de la Nación. en un Contexto de Creciente Deterioro.” Trejos, J. D., and L. Oviedo. 2012. “Cambios en la Background paper for the Vigésimo Informe Distribución del Ingreso Familiar en Costa Estado de la Nación en Desarrollo Humano Rica Durante la Primera Década del Siglo Sostenible. San José, Costa Rica: Programa XXI.” Ciencias Económicas 30 (2): 9–29. Estado de la Nación. Trejos, J. D., and M. Saenz. 2012. “Desigualdades Sáenz, M., Bermúdez, J., and M. Acosta. 2010. distritales en la formación y en el acervo de Universal Coverage in Middle Income Country: capital humano. 2000–2011.” Background pa- Costa Rica. World Health Organization. per for the Cuarto Informe Estado de la Sauma, P., and J. D. Trejos. 2014. “Social Public Educación. San José, Costa Rica: Programa Spending, Taxes, and Redistribution of Estado de la Nación. Income, and Poverty in Costa Rica.” UCCAEP (Unión Costarricense de Cámaras y Commitment to Equity Working paper No 18. Asociaciones del Sector Empresarial Privado). Schiffbauer, M., H. Sahnoun, and J. Araujo. 2014. 2014. Pulso Empresarial Encuesta Trimestral. Structural Change in Latin America: Does the San José, Costa Rica: UCCAEP. Allocation of Resources Across Sectors, UNESCO (United Nations Educational, Scientific Products, and Technologies Explain the Region’s and Cultural Organization). 2014. Third Re­ Slow Productivity Growth? Washington, DC: gional Comparative and Explanatory Study World Bank. (TERCE). Santiago, Chile: UNESCO Regional Scott, J. Forthcoming. “Redistributive Impact and Bureau for Education in Latin America and Efficiency of Mexico’s Fiscal System.” In Fiscal the Caribbean. 152 References Vakis, R., J. Rigolini, and L. Lucchetti. 2015. Resource Document. Washington, DC: Left Behind Chronic Poverty in Latin America World Bank. and the Caribbean. The World Bank: World Bank. 2014b. Costa Rica Health Policy Washington, DC. Note. Washington, DC: World Bank. World Bank. 2004. A Better Investment Climate World Bank. 2014c. Doing Business: Trading for Everyone. Washington, DC: World Bank. Across Borders. Washington, DC: World Bank. World Bank. 2006a. Costa Rica—Country World Bank. 2014d. Environment and Agriculture Economic Memorandum: The Challenges for Policy Note. Washington, DC: World Bank. Sustained Growth. Poverty Reduction and World Bank. 2014e. “Social Gains in the Balance: Economic Management Sector Unit. A Fiscal Policy Challenge for Latin America Washington, DC: World Bank. and the Caribbean.” Report 85162 (February), World Bank. 2006b. Costa Rica: Investment World Bank, Washington, DC. Climate Report. Finance, Private Sector, and World Bank. 2014f. Facilitating Global Value Infrastructure Unit. Washington, DC: The Chain Integration for Competitiveness in Costa World Bank. Rica: Recommendations to the Costa Rican World Bank. 2007. “Costa Rica Poverty Assessment: Government to Foster Productive Linkages be­ Recapturing Momentum for Poverty Reduction.” tween Domestic SMEs and Multinational Report No. 35910-CR, World Bank, Corporations. Washington, DC: World Bank. Washington, DC. World Bank. 2015. Global Economic Prospects, World Bank. 2008. Costa Rica: Public Expenditure January 2014. Washington, DC: World Bank. Review—Enhancing the Efficiency of World Bank. Forthcoming 2015. Central America Expenditures. Washington, DC: World Bank. Social Sector Expenditure and Institutional World Bank. 2009. Costa Rica: Competitiveness Review. Washington, DC: World Bank. Diagnostics and Recommendations. Poverty World Bank and Inter-American Development Reduction and Economic Management Bank. 2010. Informe de Desempeño de la Gestión Department. Washington, DC: World Bank. de Finanzas Públicas (PEFA). Washington, DC: World Bank. 2010. Disaster Risk Management in World Bank and Inter-American Development Central America: Costa Rica. Washington, DC: Bank. World Bank. World Development Indicators (database), World Bank. 2012. Logistics in Central America: World Bank, Washington, DC, http://data​ The Path to Competitiveness: Summary .­worldbank.org/data-catalog​/­world​ Document. Washington, DC: World Bank. -­development-indicators. World Bank. 2013a. Costa Rica: Five Years After World Economic Forum. 2008. The Global CAFTA-​ DR—Assessing Early Results for the Competitiveness Report: 2008–2009. Geneva: Costa Rican Economy. Washington, DC: World Economic Forum. World Bank. World Economic Forum. 2014. The Global World Bank. 2013b. “Shifting Gears to Accelerate Competitiveness Report: 2014–2015. Geneva: Shared Prosperity in Latin America and the World Economic Forum. Caribbean.” LCSPP Poverty and Labor Brief WorldWatch Institute. April 2015. “Costa Rica No. 5, Washington, DC. Aims to Become First Carbon Neutral World Bank. 2013c. World Development Report Country.” 2014: Risk and Opportunity—Managing Risk for World Travel & Tourism Council. 2014. WTTC Development. Washington, DC: World Bank. Travel & Tourism Economic Impact 2014, World Bank. 2014a. Analytical Approaches for a Costa Rica. London: World Travel & Tourism Systematic Country Diagnostic (SCD)—A Council. References 153 Costa Rica stands out for being among the most politically stable, progressive, prosperous, and environmentally conscious nations in the Latin America and the Caribbean region. Its development model has brought impor- tant economic, social, and environmental dividends, with sustained growth, upward mobility for a large share of the population, important gains in social indicators, and significant achievements in reforestation and conservation. However, there are a number of development challenges that need to be addressed to maintain the country’s successful development path. This Systematic Country Diagnostic takes stock of poverty, inequality, and growth trends, addressing the following questions: the ­ • To what extent has the Costa Rican development model been inclusive? • What has driven growth in Costa Rica in recent years, and what are the bottlenecks that need to be addressed? • How sustainable is the development model of Costa Rica economically, socially, and environmentally? SKU K8319