Systematic Country Diagnostic Mexico Systematic Country Diagnostic November 2018 Mexico © 2019 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 21 20 19 18 This work is a product of the staff of The World Bank. The findings, interpretations, and conclu- sions 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 privi- leges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions Attribution—Please cite the work as follows: World Bank. 2018. Mexico - Systematic Country Diagnostic (English). Washington, D.C. : World Bank Group. Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. 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 reuse a component of the work, it is your responsibility to determine whether per- mission is needed for that reuse 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 World Bank’s Senior Communica- tions Officer for Mexico; e-mail: bmmexico@worldbank.org. Cover design: www.sonideas.com / Alejandro Espinosa Contents List of abbreviations••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••9 Acknowledgements•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 11 1. Overview••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••15 1.1 Setting the stage •••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 15 1.1.1. Reform progress•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 15 1.1.2. Other key defining characteristics ••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 19 1.2 Growth dynamics••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 22 1.3 Poverty reduction, shared prosperity, and inclusion dynamics••••••••••••••••••••••••••••••••••••••••••••••••••••••• 24 1.4 Sustainability issues••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 28 1.5 Structural impediments ••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 29 1.5.1. Product and factor markets••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 30 1.5.2. Rule of law institutions•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 31 1.5.3. Resource allocation and institutional policy coordination•••••••••••••••••••••••••••••••••••••••••••••••••••••••32 1.6 Priorities •••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••33 2. Growth•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 37 2.1 Convergence, growth, and trade dynamics••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 37 2.2 Dynamics affecting growth rates••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••44 2.2.1. Business cycles and aggregate shocks••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 44 2.2.2. Limited capital accumulation••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 44 2.2.3. Significant regional growth disparities and limited regional convergence••••••••••••••••••• 47 2.2.4. Limited productivity growth and significant dispersion across regions, sectors, and firms••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 48 3. Inclusion••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 57 3.1 Social inclusion dynamics••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 57 3.2 Trends in productive inclusion••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 64 3.2.1. Human capital: access to education is near universal but student performance lags behind••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 64 3.2.2. Natural capital: uneven development in agriculture and forestry••••••••••••••••••••••••••••••••••••• 68 3.2.3. Social capital: low civil engagement and social cohesion prevents productive inclusion•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 70 CONTENTS 3.3 Unequal service delivery amplifies inequalities••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 71 3.4 Decreasing labor productivity across the skills distribution••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 75 3.5 Low female and youth labor force participation•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 75 4. Key sustainability challenges••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••79 4.1 Fiscal sustainability••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 79 4.2 Environmental sustainability••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 82 4.3 Social sustainability•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 89 5. Unlocking Mexico’s inclusive growth potential••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 93 5.1 Structural impediments to inclusive growth•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 93 5.1.1. Product and factor markets•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 95 5.1.2. Rule of law institutions••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 104 5.1.3. Resource allocation and institutional policy coordination•••••••••••••••••••••••••••••••••••••••••••••••••••• 108 5.2 Prioritization process: From structural impediments to priority policy areas••••••••••••••••••••••••••116 5.3 Knowledge gaps and data limitations••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 118 References••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 121 Annex 1. Identification of comparators/peer countries•••••••••••••••••••••••••••••••••••••••• 129 Annex 2. Official poverty methodology in Mexico •••••••••••••••••••••••••••••••••••••••••••••••••••••••• 131 Annex 3. Jobs diagnostic ••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••132 Annex 4. List of stakeholders that participated in consultations••••••• 140 Annex 5. Data sources••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 144 SYSTEMATIC Annex 6. Benchmarking and prioritization exercise•••••••••••••••••••••••••••••••••••••••••••••••••149 COUNTRY DIAGNOSTIC MEXICO 4 CONTENTS List of Graphics Figure 1. Inflation in single digits since 2000••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 16 Figure 2. Trade as a share of GDP has doubled since 1990••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 16 Figure 3. The economy has diversified away from oil••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 17 Figure 4. Normalized sectoral fitness rank, Mexico, 2000–15••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 17 Figure 5. Prospera coverage, 1997–2016 (millions of beneficiary families)••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••18 Figure 6. Structure of labor supply, by educational attainment, 1993–2013•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••18 Figure 7. General government fiscal balances, % of GDP•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 20 Figure 8. Gross general government debt, % of GDP••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 20 Figure 9. Export flows, Mexico–United States (actual and predicted by a gravity model, 2013)••••••••••••••••••••••••••••••••••• 21 Figure 10. Inbound international tourists (1,000s), 2009 and 2016••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 21 Figure 11. Average GDP growth rates for Mexico and comparators, 1980–2016 (%)••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 23 Figure 12. Per capita income as a share of U.S. per capita income••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 23 Figure 13. Monetary poverty rates (official), 1992–2016•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 25 Figure 14. Shared prosperity is one of the lowest in the region, 2004–14•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 25 Figure 15. Non-monetary components of the official multidimensional poverty measure, 2010–2016••••••••••••••••••••• 26 Figure 16. Share of population with simultaneous deprivation in several non-monetary components of poverty•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 26 Figure 17. Contribution of growth and redistribution to poverty reduction in Latin America and the Caribbean and Mexico, (moderate poverty at $5.5/day per capita), 2004–14•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 27 Figure 18. Structural impediments and growth••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 30 Figure 19. Average GDP growth rate, Mexico and comparators, 1980–2016 (%)••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 38 Figure 20. Per capita income as a share of U.S. per capita income•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 38 Figure 21. Contribution to GDP growth, by component (demand side)•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 39 Figure 22. Contribution to GDP growth, by component (supply side)••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 39 Figure 23. Total factor productivity (TFP) by sector (1990=100)••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 40 Figure 24. Contributions to growth in total wealth (%), 1995–2014•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••41 Figure 25. Cyclical component of GDP in the United States and Mexico (annual % change••••••••••••••••••••••••••••••••••••••••41 Figure 26. Mexico’s business cycles pre- and post-NAFTA•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 42 Figure 27. Mexico’s terms of trade compared with Latin American peers (1998=100)••••••••••••••••••••••••••••••••••••••••••••••••••• 42 Figure 28. General government fiscal balances, % of GDP••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 43 Figure 29. Gross general government debt, % of GDP••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 43 Figure 30. Annual GPD growth, Mexico (%), 1982–2016•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 44 Figure 31. Growth volatility in Mexico and selected comparators••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 45 Figure 32. Average annual public investment in infrastructure, 2008-2015 (% of GDP)•••••••••••••••••••••••••••••••••••••••••••••••• 45 Figure 33. Growth in capital stock and GDP (1961-2015)•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••46 Figure 34. Investment in the transport sector (public + private), % of GDP••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••46 Figure 35. Regional convergence (unconditional) in GDP per capita•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 47 Figure 36. Growth accounting, Mexico and comparator countries (% contribution), 1991–2016••••••••••••••••••••••••••••••••• 48 Figure 37. Growth accounting, Mexico (% contribution)••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 48 Figure 38. Gap between United States GDP per worker and GDP per worker in selected Latin America and Caribbean countries•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••49 Figure 39. Skilled workforce, Mexico and selected countries, 2010•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••49 Figure 40. Labor productivity by state (Mexico City = 100)•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 50 Figure 41. Labor productivity vs. informality, states••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 51 Figure 42. Economic units with more than 10 employees, share of total (%)•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 51 Figure 43. Change in employment share vs. labor productivity, by sector••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 52 Figure 44. Productivity dispersion in the manufacturing sector by firm size•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 53 Figure 45. Mexico’s manufacturing management score distribution vs. the United States•••••••••••••••••••••••••••••••••••••••• 53 Figure 46. Global Innovation Index•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 54 Figure 47. Urban population in 2018 (% total)••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 54 SYSTEMATIC COUNTRY Figure 48. Correlation between city size and productivity, Mexico•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 54 DIAGNOSTIC Figure 49. Poverty among indigenous and nonindigenous population groups (%), 2010–16•••••••••••••••••••••••••••••••••••••• 58 MEXICO Figure 50. Poverty, by place of residence (%), 2010–16•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 58 Figure 51. Population shares experiencing social deprivations, by area of residence and ethnicity, 2016•••••••••••••••• 59 5 CONTENTS Figure 52. Socioeconomic classes (%), 2000–16••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 61 Figure 53. Gini (per capita income), 2002–16••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 62 Figure 54. Contribution of household income sources to changes in poverty, 2004–14, $4/day (2011 PPP)•••••••••••• 62 Figure 55. Annual growth of remittances in Mexico, 2000-17••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 63 Figure 56. Share of households receiving remittances by quintile, 2008-16••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 63 Figure 57. Percentage of students in the bottom and top levels of performance in math, PISA 2015•••••••••••••••••••••• 65 Figure 58. Children-to-contact staff (teachers and teachers’ aides) ratios in pre-primary education services, 2015••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 65 Figure 59. Composition of the workforce, 2017••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 66 Figure 60. Intergroup cohesion index, 2010••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 71 Figure 61. Civic activism index, 2010••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 71 Figure 62. Percentage of students with sufficient reading skills (ENLACE), by spending source, 2006–13••••••••••••• 72 Figure 63. Spending per student at the basic education level by state, 2013••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 72 Figure 64. Insurance coverage, by income quintile, 2016•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 73 Figure 65. Out-of-pocket share of total current spending on health, circa 2015••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 74 Figure 66. Properties with sewerage service, by city size and type, 1990 and 2010•••••••••••••••••••••••••••••••••••••••••••••••••••••••• 74 Figure 67. Infrastructure access in the Guadalajara Metropolitan Area, 2000••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 74 Figure 68. Percentage of workers that is informal by firm size, 2008•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 75 Figure 69. Wage index for selected workers, 2006–16 (male age 15-yo-64)••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 76 Figure 70. Poverty rates and irrigation by state, 2016••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 76 Figure 71. Evolution of female labor force participation, 1992–2016•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 77 Figure 72. Gender participation gap across the life cycle, 2016••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 77 Figure 73: Tax structures in Mexico, Latin America and the Caribbean and the OECD (2016)•••••••••••••••••••••••••••••••••••• 80 Figure 74. Public sector revenue and expenditure, 2000–17 (percent of GDP)•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••81 Figure 75. Federal public sector debt, 2004–17 (percent of GDP)•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••81 Figure 76. Water stress and availability, by river basin, 2015••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 83 Figure 77. Trends in costs of water depletion and degradation in Mexico••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 84 Figure 78. GDP adjusted to environmental and water resources•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 84 Figure 79. Proportion of households with access to sewerage•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 85 Figure 80. Proportion of households with access to piped water. Lowest and highest quantiles of income••••••••• 85 Figure 81. Number of cars per capita by city size, 1990–2013•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 86 Figure 82. Air pollution – micrograms per cubic meter••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 86 Figure 83. Top 20 economies according to their carbon dioxide emissions, tons, millions, 2015•••••••••••••••••••••••••••••••• 87 Figure 84. Structural impediments and growth••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••94 Figure 85. Product market regulation indicator•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 97 Figure 86. Credit to GDP vs. per capita income••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 99 Figure 87. Commercial bank asset concentration (2017)••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 100 Figure 88. Fintech sector Mexico••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 102 Figure 89. States’ implementation of oral commercial lawsuit••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 105 Figure 90. The role of taxes and transfers on redistribution in OECD countries, latest available year•••••••••••••••••••• 109 Figure 91. Vertical fiscal gap, OECD and selected countries, 2016•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 112 Figure 92. Revenue and expenditure decentralization, OECD and selected federations•••••••••••••••••••••••••••••••••••••••••••• 112 Figure 93. Participaciones and GDP, by state•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 113 Figure 94. Aportaciones and GDP, by state•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 113 Figure 95. Prioritization process•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 116 Figure A.3.1. Share of youth not in employment, education or training (NEET) by gender, selected LAC countries•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 133 Figure A.3.2: Unpaid workers••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 134 Figure A.3.3: Employment by states and regions (percentage)•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 135 Figure A.3.4: Available population••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••136 SYSTEMATIC Figure A.3.5: Workers and minimum wage ranges•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 137 COUNTRY Figure A.3.6: Workers and minimum wage ranges••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 138 DIAGNOSTIC MEXICO 6 CONTENTS List of Tables Table of extended team members and input providers••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 12 Table 1: Priority policy areas••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 34 Table 2. Evolution of the social components, multidimensional poverty index, Mexico, 2010–16••••••••••••••••••••••••••••••• 58 Table 3. International comparison of the forestry sector: Employment, GDP, and balance of trade••••••••••••••••••••••••• 69 Table 4: Public Expenditure, 2014-2017 (% of GDP)•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••81 Table 5. Progress in the implementation of the structural reforms•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 95 Table 6. Main results of the structural reforms (2012-2017)•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 95 Table 7. Credit to the private and public sector, selected countries, 2017••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 100 Table 8. Main structural impediments to achieve sustainable inclusive growth in Mexico•••••••••••••••••••••••••••••••••••••••• 117 Table 9. Summary table of selected indicators (distance to top performers by structural impediment)•••••••••••••••• 117 Table 10. Priority policy areas•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 119 List of Boxes Box 1. Key reforms in Mexico 1997-2012••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••18 Box 2. Structural reforms 2012-2017: initial results•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 19 Box 3. The motor vehicle industry in Mexico••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 39 Box 4. The aerospace industry in Querétaro: A success story of public-private sector collaboration••••••••••••••••••••••• 40 Box 5. The high and rising incidence of noncommunicable diseases in Mexico is associated with substantial economic costs and particularly pernicious effects on the poor•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 66 Box 6. Potential distributional effects of automation in Mexico•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 67 Box 7. Agribusiness as an opportunity for productive inclusion•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 68 Box 8. The timber sector: incentives to balance conservation and economic objectives?••••••••••••••••••••••••••••••••••••••••••• 69 Box 9. Structural reforms 2012-2017••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 95 Box 10. The Fintech (financial technology) sector in Mexico•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 102 Box 11. Fragmentation and duality in Mexico’s health system•••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 110 Box 12. A tale of two Mexican states••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 115 SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 7 Abbreviations List of abbreviations ASERCA Agencia de Servicios a la Comercialización y Desarrollo de Mercados Agropecuarios (Agency for Marketing Services and Development of Agricultural Market) CDMX Ciudad de México (Mexico City) CEDLAS Centro de Estudios Distributivos, Laborales y Sociales (Center of Distributive, Labor and Social Studies) CFE Community-Based Forest Enterprise (National Foresrtry Comission) COFECE Comisión Federal de Competencia Económica (Federal Commission for Economic Competition) CONAFE Consejo Nacional de Fomento Educativo (National Council for Education Development) CONAFOR Comisión Nacional Forestal CONEVAL Consejo Nacional de Evaluación de la Política de Desarrollo Social (National Council for the Evaluation of Social Development Policy) ECOPRED Encuesta de Cohesión Social para la Prevención de la Violencia y la Delincuencia (Social Cohesion Survey for the Prevention of Violence and Crime) EMOVI Encuesta de Movilidad Social (Social Mobility Survey) ENIGH Encuesta Nacional de Ingresos y Gastos de Los Hogares (National Survey of Household Incomes and Expen- ditures) ENLACE Evaluación Nacional del Logro Académico en Centros Escolares (National Evaluation for Academic Achieve- ment in Schools) ENOE Encuesta Nacional de Ocupación y Empleo (National Survey on Occupation and Employment) ENSANUT Encuesta Nacional de Salud y Nutrición (National Health and Nutrition Survey) ESRU Fundación Espinosa Rugarcía (Espinosa Rugarcia Foundation) FAEB-FONE Fondo de Aportaciones para la Nómina Educativa y el Gasto Operativo (Contribution Fund for Educational Payroll and Operational Expenditure) FAO Food and Agriculture Organization of the United Nations FASSA Fondo de Aportaciones para los Servicios de Salud (Contributions Fund for Health Services) FDI Foreign Direct Investment FONDEN Fondo Nacional de Desastres Naturales (National Natural Disaster Fund) SYSTEMATIC COUNTRY GATT General Agreement on Trade and Tariffs DIAGNOSTIC MEXICO GDP Gross Domestic Product GHG Greenhouse Gas IDA International Development Association 9 Abbreviations IFC International Finance Corporation IFE Federal Elections Institute ILOSTAT International Labor Organization Statistics INEGI Instituto Nacional de Estadística y Geografía (National Institute of Statistics and Geography) LFPRH Ley Federal de Presupuesto y Responsabilidad Hacendaria (Federal Budget and Fiscal Responsibility Law) MCS-ENIGH Módulo de Condiciones Socioeconómicas – ENIGH (Module of Socioeconomic Conditions – ENIGH) MEC Modelo de Comparabilidad Estadísticas (Statistics Comparability Model) MIGA Multilateral Investment Guarantee Agency NAFTA North American Free Trade Agreement OECD Organisation for Economic Co-operation and Development PEMEX Petróleos Mexicanos PISA Programa Internacional de Evaluación de los Alumnos (Programme for International Student Assessment) PROAGRO Programa de Oferta Agropecuaria (Agricultural Offer Program) PROCAMPO Programa de Apoyos Directos al Campo (Direct Support for Farming Program) PSBR Public-Sector Borrowing Requirements SAR Special Administrative Region (China) SEDESOL Secretaría de Desarrollo Social (Secretariat of Social Development) SEDLAC Socio-Economic Database for Latin America and the Caribbean SPSS Social Protection System TFP Total Factor Productivity UDLAP Universidad de las Américas Puebla SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 10 Acknowledgements This Systematic Country Diagnostic (SCD) was prepared as part of a collaborative process of the Mexico Country Team with participation of all Global Practices of the World Bank and the International Finance Corporation, as well as of many stakeholders including government authorities, think tanks, international organizations, academia, civil society, and rep- resentatives of the private sector (see Annex 4 for a full list of participants in consultations). The team that produced this report is grateful for their inputs, knowledge and advice. The Concept Note Review of this SCD was held on February 23, 2018. This report was prepared by a team led by Carlos Rodríguez-Castelán (Sr. Economist and TTL, GPV04) and Jasmin Chakeri (Program Leader EFI and TTL, LCC1C). The core team included: Lourdes Rodríguez-Chamussy (Economist, GPV04), Fernan- do Blanco (Lead Economist, GMTLC), Jozef Draaisma (Senior Country Economist, GMTLC), Luciana Harrington (Strategy Officer, CCECE), Juan José Miranda Montero (Environmental Economist, GENGE), Jia Jun Lee (Research Analyst, GENGE), Kiyomi Cadena (Consultant, GPV04), Laura Moreno (Research Analyst, GPV04), Samantha Lach (Consultant, GPV04), Theo- dore Khan (Consultant, GPV04), James Sampi (Consultant, GMTLC) and Vic Duggan (Consultant, GMTLC). The team received outstanding administrative assistance from Desiree González (GPV04), Alejandra Ramón (LCC1C), Diana Martinez Ramirez and David Islas (LCC1C). The full list of extended team members who were consulted and who have contributed written inputs to the SCD are shown in the table below. The team gratefully acknowledges the overall guidance of Jorge Familiar Calderón (Regional Vice President, LCRVP), Pablo Saavedra (Country Director, LCC1C), Humberto López (Director, Strategy and Operations, LCRVP), Carlos Vegh (Chief Econ- omist, LCRCE), Gerardo Corrochano (Director, OPSRR and former Country Director LCC1C), Jutta Ursula Kern (Operations Manager, LCC1C), Daniel Lederman (Lead Economist, LCRCE), Jorge Araujo Thompson (Economic Adviser, LCRVP), Oscar Calvo-González (Practice Manager, GPV04), Arturo Herrera (Practice Manager, GG016), Eduardo Wallentin (Senior Manager, CCECE), Ary Naim (Country Manager, IFC Mexico), William Wiseman (Program Leader HD, LCC1C), Gregor Wolf (Program Leader SD, LCC1C), and Ambar Narayan (Lead Economist and SCD Central Team TTL, GPVDR). The team would also like to thank the peer reviewers: María de los Ángeles González (Practice Manager, GMTE2), Carter Brandon (Global Lead of SYSTEMATIC Environmental Economics and Lead Economist, GENDR), Roberta V. Gatti (Chief Economist, GGHVP), and Emily Sinnott COUNTRY (Program Leader, LCC7C). The team would also like to thank its advisory group including: Luis-Felipe López-Calva (Practice DIAGNOSTIC MEXICO Manager, GPV02), Emmanuel Skoufias (Lead Economist, GPV01), Jorge Ernesto López-Cordova (Practice Manager, GFCIS), Gladys López-Acevedo (Lead Economist, SARCE), and Samuel Freije-Rodríguez (Lead Economist, GPV03). 11 ACKNOWLEDGEMENTS Table of extended team members and input providers List of IFC and WB GPs Team consulted and extended team members Name GP Preeti Ahuja, Luz Díaz Ríos, Katie Kennedy, Svetlana Edmeades, Ashwini Rekha Sebastian Agriculture Reema Nayar, Rafael de Hoyos, Ciro Avitabile Education Guillermo Hernández González Energy and Extractives Valerie Hickey, Renan Poveda, Ernesto Sánchez-Triana, Angela Armstrong, Katharina Siegmann Environment and Natural Resources Zafer Mustafaoglu, Steen Byskov, Patricia Caraballo, Marialisa Motta, Leonardo Iacovone, Mariana Finance, Competitiveness and Innovation Pereira Arturo Herrera, Alberto Leyton, Jorge Luis Silva Méndez Governance Roberto Lunes, Ronald Gomez Health, Nutrition and Population Ary Naim, Eduardo Wallentin, Lucía Baltazar IFC Tanja Goodwin Macroeconomics, Trade and Investment Oscar Calvo-González, Freeha Fatima Poverty and Equity Jan Weetjens, Arelia Jacive López Castaneda, Carlos Pérez-Brito, Asli Gurkan, Dorothee Georg, Paula GPSURR - Social Development Andrea Rossiasco Pablo Gottret, Aylin Isik-Dikmelik, Maria Concepcion Stetta Social Protection and Labor Shomik Mehndiratta, Felipe Targa, Abel López Dodero Transport and ICTs Ming Zhang, Alexandra Ortiz, Oscar Ishizawa, Horacio Terraza, Ondina Rocca, Nancy Lozano GPSURR - Urban and Disaster Risk Management Rita Cestti, Christian Borja-Vega, Diego Rodríguez Water Eliana Rubiano Gender CCSA Eliana Carranza Jobs CCSA Raul Alfaro-Pelico Climate Change CCSA SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 12 1. Overview 1.1 Setting the stage 1.1.1. Reform progress 2. Mexico has strong macroeconomic institutions, with a 1. The development process in Mexico has been partly commendable policy track record. During the 1980s and framed by a set of defining characteristics as well as by 1990s, the country undertook deep structural reforms, significant economic reforms during recent decades. The including privatization, deregulation, and trade liberal- country has established a strong track record of prudent ization. These led to large capital inflows, a surge in credit macroeconomic policies. The Central Bank and the Min- to the private sector, and an appreciation of the Mexican istry of Finance have delivered stable and sustainable peso. This progress stopped abruptly in late 1994 when the monetary and fiscal policies, and garnered high credibility United States tightened monetary policy, which, together in international markets. Successive governments have with other factors, resulted in a currency crisis and a deep implemented a broad set of reforms that have opened the recession. The crisis motivated successive governments to economy to trade and liberalized domestic markets. In the implement prudent monetary and fiscal policies through social sectors, Mexico has led the way among emerging strengthened macroeconomic institutions, and these pol- and developing economies in reshaping social protection, icies proved to be long-lasting. Key reforms included the health care, and education policies. These reforms have adoption of a flexible exchange rate, an inflation-targeting helped transform the country over the last 30 years. Aside regime, an autonomous Central Bank, fiscal consolidation from reforms and policies that have improved outcomes, when needed, improved debt management, strength- Mexico has, at least, three key defining characteristics: ened macroprudential regulations, and an opening up (1) a privileged geography, (2)  abundant natural capital, of the financial sector to foreign participation. Inflation and (3) a democracy that has allowed the emergence of was brought down from an average of 39 percent a year more political parties over the last 20 years. These factors between 1985 and 2000 to 4 percent between 2001 and have each partly shaped the country’s development and 2016 (Figure 1). Over the years, the Mexican authorities delivered significant economic and social outcomes. Yet, showed a strong track record in macroeconomic manage- the question remains: how can Mexico grow more rapidly ment that is well regarded in international capital markets and become more inclusive along its development path? and that has made the Mexican peso the most highly trad- SYSTEMATIC These are the central issues covered in this systematic ed emerging market currency. COUNTRY country diagnostic (SCD). DIAGNOSTIC MEXICO 3. The economy has also been opened to foreign trade and private investment. The entering into force of the North American Free Trade Agreement (NAFTA) in 1994 had a 15 1. Overview Figure 1. Inflation in single digits since 2000 120 100 80 inflation 60 Average 1985-2009, 39 40 20 Average 2001-2016, 4 0 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 2015 Source: Data of WDI (World Development Indicators) (database), World Bank, Washington, DC, http://data.worldbank.org/products/wdi. Figure 2. Trade as a share of GDP has doubled since 1990 90 80 70 60 Trade (%GDP) 50 40 30 20 10 0 1990 1992 1994 1996 1998 2000 2002 2004 2006 2012 2014 2016 brazil OECD members Chile Mexico Source: Data of WDI (World Development Indicators) (database), World Bank, Washington, DC, http://data.worldbank.org/products/wdi. profound effect on the economy. NAFTA eliminated trade the launch of NAFTA (compared with an average GDP tariffs among Canada, Mexico, and the United States, but growth of 2.4 percent in real terms). In 2016, the total an- it also established the protection of intellectual property nual value of exports in agriculture was US$14.74 billion, rights, removed many barriers to investment, stipulated up from US$3.68 billion in 1993.1 mechanisms for investment dispute resolution, and intro- duced measures for environmental and labor protection. 4. The country has also diversified away from oil and devel- Exports from Mexico to the United States rose beyond oped a stronger productive structure in the private sector. expectations from US$18.5 billion in 1990 to US$327 bil- The growth in manufacturing transformed the export lion in 2017, while the value of imports to Mexico from the basket away from the dominance of raw materials, espe- United States increased from US$20 billion to US$195 bil- cially oil, toward manufactured products. These include lion. Mexico has become the world’s 15th largest exporter machinery, electronics, and transport vehicles (Figure 3). (in US$ terms). Moreover, its trade accounted for close to The country strengthened its productive capacities in the 80 percent of gross domestic product (GDP) in 2016, com- most complex sectors and is today a competitive exporter pared with less than 40 percent in 1990 (Figure 2). Foreign in products that have a high share of domestic value add- direct investment (FDI) expanded from 1 percent of GDP in ed (Figure 4). 1990 to 3.2 percent in 2016, as the country became deeply integrated into regional and global value chains (GVCs), 5. Mexico has led the way in reforms across emerging and de- especially in the motor vehicle sector, which, in 2016, ac- veloping countries in social protection and health. Between SYSTEMATIC counted for 70 percent of total inward FDI. The exports of 1997 and 2014, 6.1 million households were brought un- COUNTRY DIAGNOSTIC agricultural products grew by 3.9 percent a year following der the main social protection scheme, Prospera (Figure 5), MEXICO 16 1 The main export products include tomatoes, avocados, berries, broccoli, and zucchini. In 2014, Mexico accounted for more than 80 percent of the tomato imports to the United States according to the U.S. Department of Agriculture. There is some evidence that the growth in tomato exports also contributed to increased income and employment opportunities among the poor. 1. Overview Figure 3. The economy has diversified away from oil veg 6.7% Metals 6.7% Machinery 13.0% Fuels 37.5% 1980 Animal 2.8% Minerals 2.5% Chem 5.7% Food 2.4% Plastic 2.0% Transport 12.4% Stone 3.1% Textiles 2.2% Miscellaneous 9.1% Fuels 4.8% Machinery 36.9% Transport 24.7% Food 3.0% Stone 3.0% 2016 Metals 4.2% Plastic 2.9% veg 3.8% Chem 2.6% Source: Data of WITS (World Integrated Trade Solution) (database), World Bank, Washington, DC, http://wits.worldbank.org/WITS/. Figure 4. Normalized sectoral fitness rank, Mexico, 2000–15 Complexity increasing Normalized Sector Fitness Rank counter-clockwise from here 2000 2015 Forestry Oil & Gas Electronics 1 Apparel Chemicals Fishing 0.75 Machinery Beverage & Tobacco 0.5 Misc. Crops 0.2 Electronic Equipment 0 Wood Metal Products Leather Paper Food Transportation Textile Products Minerals Animal Products Metal Mining Petroleum & Coal Furniture Plastics Textiles Source: Data of the International Finance Corporation. as the country shifted from general subsidies to targeted 6. Since 2012, Mexico embarked on a new wave of struc- and conditional transfers starting in the 1990s. The Pros- tural reforms to tackle critical development challenges, pera Program conditions cash transfers on human capital which will leave a solid base for future growth. In partic- accumulation (attendance at school and health check- ular, these included reforms in education, tax policy, ups). It has been key to improving the number of years of energy, telecommunication, competition, labor, and the schooling among the poorest. Over the last two decades, financial sector (Box 1). The education reform of 2013 in- the share of the population with some college educational troduced improved teacher evaluations and merit-based attainment or more almost doubled, while the share of the entry, pay and promotions. In energy, a constitutional labor force with some high school education rose by more amendment, approved in 2013, opened the electricity, than 20 percentage points, to 55 percent (Figure 6). Oth- oil, and gas sectors to private sector participation to bol- er social programs, including the noncontributory health ster overall economic growth. With this reform, Mexico program Seguro Popular, were implemented and expand- aimed to tackle three challenges: (1) energy security, SYSTEMATIC COUNTRY ed, contributing to improved access to health care among (2) sectoral sustainability, and (3) energy efficiency. The DIAGNOSTIC the poor. Mexico is well-known for its innovations in social financial and telecommunication reforms, signed into MEXICO policy since the late 1990s, many of which (particularly law along with secondary legislation in 2014, aimed at 17 conditional cash transfers) have been adopted by dozens deconcentrating these two markets by lifting entry re- of countries around the world. strictions, thereby incentivizing more competition. The 1. Overview Figure 5. Prospera coverage, 1997–2016 (millions of beneficiary families) 7 6.2 6.1 5.8 5.9 6 5.0 5.2 Millions of beneficiary families 5 4.2 4.2 4 3.1 3 2.5 2.3 2 1.6 1 0.3 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Source: Data of Secretaría de Desarrollo Social (Secretariat of Social Development, SEDESOL). Figure 6. Structure of labor supply, by educational attainment, 1993–2013 60% 50% Labor supply by education 40% 30% 20% 10% 0% 1994 1996 1998 2000 2002 2004 2005 2006 2008 2010 2012 2014 2016 Primary or less high school College Source: Data of SEDLAC (Socio-Economic Database for Latin America and the Caribbean), Center for Distributive, Labor, and Social Studies, Facultad de Ciencias Económicas, Universi- dad Nacional de La Plata, La Plata, Argentina, and Equity Lab, Team for Statistical Development, World Bank, Washington, DC, http://www.worldbank.org/equityla Box 1. Key reforms in Mexico 1997-2012 • Reform of the pension system (1997) • Reform of the fiscal coordination Law (1997) • Reform of social protection programs (Prospera) (1998) • Education reform on mandatory preschool education (2002) • Noncontributory health insurance (Seguro Popular) (2002) • Judicial process reform (2008) • Education reform on mandatory high school education (2012) ultimate objective was to reduce prices and improve 7. Also, more recently, and in the context of the 2014 oil the quality of banking and telecommunication services. price drop, the authorities reacted with prudent policies, While there is still a significant reform agenda in these enabling the country to withstand the shock while still SYSTEMATIC areas, the reforms adopted have already started to bear achieving economic growth. Oil price collapses in the COUNTRY DIAGNOSTIC fruit through greater accountability, lower consumer past in Mexico had significant negative effects on both MEXICO prices, and investment (see Box 2). A focus on imple- output and the fiscal position. In this last episode, how- mentation will help realize the full economic impact of ever, the 2013 tax reform generated sufficient additional 18 the reforms over time; they also provide a solid basis for revenues to help offset the drastic drop in oil revenues. additional measures going forward. The authorities also applied an expenditure rationaliza- 1. Overview Box 2. Structural reforms 2012-2017: initial results Labor market reform (2012) • More than 3 percentage points of GDP in additional tax revenues • Informality dropped from 59.5 percent in 2012 to 56.9 percent in 2017 Tax policy reform (2013) • More than 3 percentage points of GDP in additional tax revenues • Oil revenues as share of total tax revenues from 39 to 17 percent • Tax base increased from 38 to 66 million taxpayers Education quality reform (2013) • Scholarships from 3 out of 10 students in public schools Energy market reform (2013) • Expected investment between 160 and 200 US$ bn • More than 70 new energy firms Competition policy and regulatory reform (2013) • Double in fines for monopolistic practices Financial sector liberalization (2014) • 13 million people gained access to financial services Telecommunication reform (2014) • 50 million additional subscribers to mobile broadband24 percent decrease in telecommunications prices between 2013 and 2017 Fiscal responsibility Law for subnational discipline (2016) • States’ debt to non-earmarked transfer ratio reduced from 88 to 80 percent in 2017 In addition to these structural reforms, the basis for significant changes in governance issues were also enacted, includ- ing reforms to: • Anticorruption and transparency (2015) • New criminal justice system (2016) Source: SHCP (2018) tion program in 2015-16 to improve the fiscal stance, and It has a significant share of arable land and abundant natu- after the initial shock, public debt started to stabilize and, ral capital, including substantial endowments of hydrocar- more recently, to decline, leading the way among emerg- bons and mineral deposits. Because of its rich and diverse ing and developing economies (Figure 7 and Figure 8). natural setting, it has great potential (even beyond what it All this occurred while the economy continued to grow has been realized) in agriculture and tourism.2 close to its average pace. The last years showed that the economy is more resilient to terms of trade shocks than 9. Its proximity to the United States (and the integration before, and that macroeconomic policies have continued with the U.S. economy through trade) has been a driver of to be prudent, safeguarding the country’s economic sta- the country’s economic transformation. Beginning with its bility and sustainability. entry into the General Agreement on Trade and Tariffs, the precursor of the World Trade Organization, in 1984, and later through its integration with North America through 1.1.2. Other key defining NAFTA in 1994, Mexico experienced a process of deep characteristics integration with the United States in terms of economic 8. Mexico has a favorable geography and abundant natural activity. A gravity model would have predicted export resources. Mexico is flanked by the Pacific Ocean and, on flows from Mexico to the United States equivalent to 33.3 SYSTEMATIC the Atlantic side, the Gulf of Mexico and the Caribbean Sea. percent (using 2013 trade data), while actual flows were COUNTRY It has a border with the United States (the largest economy equivalent to 71.5 percent, more than 2.1 times greater DIAGNOSTIC in the world) in the north and Central America in the south. than the predicted level of integration (Figure 9). MEXICO 2 See Annex 1 for the detailed description on the identification of comparators/peer countries for Mexico. 19 1. Overview Figure 7. General government fiscal balances, % of GDP 4 2 0 -2 -4 -6 -8 2012 2013 2014 2015 2016 2017 Mexico, primary balance LAC, primary balance Emerging and developing, primary balance Mexico, fiscal balance LAC, fiscal balance Emerging and developing, fiscal balance Source: IMF, World Bank. Figure 8. Gross general government debt, % of GDP 65 60 55 50 45 40 35 2012 2013 2014 2015 2016 2017 Mexico LAC Emerging and developing economies Source: IMF, World Bank. 10. Due to its geographic position, Mexico is also prone to a response institutions, becoming an example to the coun- wide range of natural hazards; 41 percent of its territory and tries of the Latin America and Caribbean region and other 31 percent of its population are exposed to hurricanes, other emerging countries in the world. storms, droughts, floods, earthquakes, and volcanic erup- tions. In economic terms, this translates into 30 percent of 11. Mexico’s natural resource base (agriculture, forests, GDP at risk from three or more hazards, and 71 percent fisheries, water, and coastal resources) is an important con- at risk from two or more hazards.3 Population growth and tributor to the economy. It represents approximately 11 the rising concentration of physical assets in urban areas percent of GDP. Natural resources provide important en- are contributing to the country’s increased vulnerability to vironmental benefits, including the regulation of the hy- hazards. These trends are likely to continue and, together drological regime and water quality, erosion control, and with a changing climate and increased climate variability, the provision of habitats for wildlife. Mexico accounts for are expected to result in growing losses from disasters, approximately 12 percent of the world’s biodiversity. Natu- particularly in the poorer regions of the country. Between ral resources directly support the livelihoods of more than 2000 and 2013, Mexico experienced the second and third 30 million people in rural areas, many of whom are poor. SYSTEMATIC costliest disasters in the country’s history after the 1985 Due its natural beauty and diversity, and cultural heritage, COUNTRY DIAGNOSTIC earthquake. To mitigate these threats and deal with the Mexico is also a top travel destination (Figure 10). In 2016, MEXICO events, Mexico has been developing innovative first-rate the country received more international tourists than Ar- 20 3 Fondo de Desastres Naturales: Una reseña. Banco Mundial, Secretaría de Gobernación y Secretaría de Hacienda de México (2012). 1. Overview Figure 9. Export flows, Mexico–United States (actual and predicted by a gravity model, 2013) JAP CAN CAN ChN Gravity Model Export Flows 2013 predicted export flows uSA uSA Source: World Bank calculations based on data of Bown et al. 2017. Figure 10. Inbound international tourists (1,000s), 2009 and 2016 40 35 30 Millions Int. Tourists 25 20 15 10 5 0 Colombia Peru Argentina Chile Brazil Romania Korea, Rep. Poland Malaysia Turkey Thailand Mexico 2009 2016 Source: Data of WDI (World Development Indicators) (database), World Bank, Washington, DC, http://data.worldbank.org/products/wdi. gentina, Brazil, Chile, Colombia, and Peru combined. In ico’s hydrocarbon sector has been controlled by a state- the same year, tourism directly contributed 8.5 percent owned company, Petróleos Mexicanos (PEMEX), which of GDP to the economy. However, natural resources are had a public monopoly on exploration, production, and also vulnerable to climate change and weather variations. refining activities in the national territory until 2013. But Moreover, disparities in the incidence of natural resources, oil production gradually dropped over the last decades such as water, combined with underinvestment in natural due to limited investment and other factors. It is, however, capital, carries the prospect of severe resource shortages expected to bounce back over the medium term owing to that can negatively impact regional development. the far-reaching energy reform of 2013 that enabled pri- vate sector participation in this area.4 12. Mexico also has a significant endowment of nonrenew- able resources, particularly oil. The country is a large pro- 13. Mexico’s democracy was dominated by a single party ducer, consumer, and exporter of energy. It is one of the until 20 years ago, and thus it is still in the process of de- 10 largest oil producers in the world and Latin America’s veloping its institutions. Mexico’s democracy dates to the second largest energy consumer. Mexico’s energy sector revolution of 1917. But between 1929 and 2000, it was has been of strategic importance to the economy and is dominated by one party, the Partido Revolucionario Insti- SYSTEMATIC COUNTRY a key driver of economic growth and productivity. Mex- tucional (PRI). Over these 71 years, in addition to the pres- DIAGNOSTIC MEXICO 21 4 The prolonged decline of oil prices has had important fiscal implications for Mexico, although not for the balance of payments. Oil revenue has been shrinking since 2011 (from 7.1 percent of GDP in 2011 to 6.8 percent in 2012, 6.1 percent in 2013, and a dramatic 4.9 percent in 2015). Although oil prices rose slightly in 2017, Mexico’s 2017 budget was still affected by low oil revenues. 1. Overview ident, most of the governors, mayors, senators, and rep- 2017, achieved higher income levels and had closed the resentatives belonged to this party or affiliated parties. In gap with respect to the United States. 2000, another party, the Partido Acción Nacional (PAN), was elected for the first time at the federal level. By 2016, the 17. On the supply side, growth in the private sector was pri- electoral map had changed substantially with respect to marily driven by the expansion in services, although some 1990. Most of the states are now under a party that has not manufacturing subsectors also showed rapid growth rates. been historically dominant. The more competitive democ- Telecommunications and financial services were the most racy has been effective at redistributing power. But some rapidly growing service subsectors over the last two de- local observers have argued that, at the same time, it has cades. Despite the overall decrease of manufacturing in maintained or even expanded rent seeking, particularly at value added, some higher value added manufacturing local levels, limiting the effectiveness and accountability subsectors have become central to regional and global of institutions and control mechanisms across the govern- value chains and important growth drivers, boosted by ment. The electoral map changed even more dramatically trade liberalization. The production of transport equip- recently (in July 2018) with the land slide victory of the ment, electronics, and computers grew well above GDP, as Movimiento de Regeneración Nacional (MORENA) party. Mexico joined regional and GVCs in the auto, electronics, and aerospace subsectors. Other manufacturing indus- 14. Despite policy reforms and many favorable defining tries, however, especially those that tend to be more labor characteristics, the country has underperformed in terms intensive, such as wood, textiles, and clothing, declined of growth, inclusion, and poverty reduction compared to its in importance over the same period, dragging the overall peers. This SCD argues that the country’s defining charac- share of manufacturing in value added down from 17.2 teristics alone, that is, geography, natural capital, and the percent in 1993 to 16.6 percent in 2017. On the demand young institutions in its more competitive democracy, side, private consumption has been the main growth driv- cannot explain the performance of the country in growth er (expanding an average of 2.8 percent a year between and inclusion. Instead, it highlights that the interaction be- 1994 and 2017), contributing more than two-thirds to tween these characteristics (some with positive and some GDP growth. Investment rose 2.3 percent a year, below with negative impacts) and structural impediments aris- the overall GDP growth rate, as public capital spending ing from incomplete (or not fully implemented) reforms, entered a secular decline. Net exports, on the other hand, limitations in the development of institutions, and other contributed little to growth. While gross exports grew by factors, explain the sluggish dynamics of growth and in- as much as 6.2 percent an annum on the back of the boom clusion. in manufactured exports, imports expanded at a similar pace. 15. Lifting these impediments can help support an accel- erated and more inclusive growth in Mexico over the next 18. As in many countries in the region, growth in Mexico years. The report also argues that key aspects of fiscal, has been affected by a series of frequent negative shocks environmental and social sustainability are critical to con- to the economy. The economy experienced greater vol- sider in order to achieve sustained growth and poverty re- atility since the early 1980s, particularly through several duction over the medium and long term. It is important to successive external and domestic shocks that included highlight at the outset that significant progress has been the 1982 debt crisis, the 1985 earthquake in Mexico City, made. As highlighted earlier, the past decade has seen the financial crisis of 1994/95, the entry of China into the important and positive reforms across a number of policy World Trade Organization in 2001, the bursting of the areas that are starting to render fruits and are likely to help U.S. stock market bubble in 2001–02, the global crisis in foster growth over the medium term. Thus, the report fo- 2008–09, and (to a lesser extend) the collapse in oil prices cuses more deeply on the pending agenda ahead. in 2014. Growth volatility in Mexico seems to be relatively high compared with the rest of the region and the world (see chapter 2). 1.2 Growth dynamics 19. Capital accumulation has not been sufficient to support 16. Mexico’s economic growth averaged only 2.4 percent higher growth over the last four decades. While overall between 1980 and 2017, limiting progress in convergence investment is not low by Latin America and Caribbean relative to the U.S. per capita income (Figure 11). On a per standards (total investment has averaged 20 percent of capita basis, average growth was close to 1 percent. The GDP since 1990), the annual investment and capital stock country’s per capita GDP today stands at 34 percent of U.S. growth has been lower in Mexico than in rapidly growing per capita GDP, compared with 49 percent in 1980 (Fig- emerging economies that are converging to higher in- ure 12). The Republic of Korea had, by 2017, reached 66 come levels: 33 percent in the Republic of Korea and 28 SYSTEMATIC percent of U.S. per capita income, even though, in 1980, it percent in Malaysia, for example. The growth in capital COUNTRY stock began to slow in the early 1980s, coinciding with the DIAGNOSTIC had less than half the per capita income of Mexico. Other MEXICO countries, such as Chile, Malaysia, and Poland, which all slump in GDP growth. had lower per capita incomes than Mexico in 1980, had, by 20. Mexico’s public investment in infrastructure only 22 reached an average of 3.2 percent of GDP between 2008 1. Overview Figure 11. Average GDP growth rates for Mexico and comparators, 1980–2016 (%) 7 6 5 4 3 2 1 0 Korea, Rep. Malaysia Thailand Turkey Chile Colombia Peru Mexico Uruguay Brazil Argentina Source: Data of WDI (World Development Indicators) (database), World Bank, Washington, DC, http://data.worldbank.org/products/wdi. Note: Growth rates of GDP measured in constant local currency unit. Figure 12. Per capita income as a share of U.S. per capita income 0.7 0.6 0.5 0.4 0.3 0.2 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 2013 2016 Malaysia South Korea Poland Chile Mexico Source: November 2017 data of Total Economy Database, Conference Board, New York, https://www.conference-board.org/data/economydatabase/. and 2015. 5 Excluding PEMEX, that number would be ment in the sector has trailed that among regional peers.7 1.7 percent of GDP in 2017. This level falls short of those Moreover, the transport, logistics, and facilitation services fast-growing Latin American countries and emerging to support export markets other than the United States, economies that spend above 5 percent of GDP in this area. such as Asian markets through export corridors toward Since 2008, private investment in four key sectors—water, the Pacific Ocean ports, are relatively weak. The country roads, energy, and telecommunication—accounted, on also continues to experience deficiencies in electricity average, for one-third of total investment in these sectors. (particularly in transmission capacity and distribution) and Public-private partnerships have grown in importance. telecommunication. These gaps suggest that an increased Since 1990, 296 projects have been undertaken as pub- role is needed for the public and private sectors in infra- lic-private partnerships, representing a value of US$83 structure financing. billion, half of which since 2006.6 Yet, insufficient overall in- vestment resulted in infrastructure bottlenecks that have 21. Average growth rates in Mexico mask significant region- ben also hampering private sector growth. Even though al income and growth disparities, with very limited domestic transport infrastructure is better in Mexico than in many convergence among states. There are large differences be- SYSTEMATIC other Latin American countries, it is aging, and new invest- tween the industrialized north and center-north and the COUNTRY DIAGNOSTIC MEXICO 5 Infralatam (Economic Infrastructure Investment Data, Latin America and the Caribbean) (database), Economic Commission for Latin America and the Caribbean, the Development 23 Bank of Latin America, and the IADB, Washington, DC, http://infralatam.info/. Includes PEMEX. 6 PPI Project Database (Private Participation in Infrastructure), World Bank, Washington, DC, http://ppi.worldbank.org/. 7 Infrastructure pillar (2017/18), GCI (Global Competitiveness Index) (database), World Economic Forum, Geneva, http://reports.weforum.org/global-competitiveness-index/. 1. Overview less developed south. In 2016, the average GDP per capita between sectors alone does not explain the sluggish pro- of one of the richest states (Nuevo León) was close to the ductivity trend, as maintaining sectoral labor shares con- average in Poland, while that of the poorest state (Chiapas) stant at 1990 level would have only marginally improved was similar to that of Honduras or Timor-Leste. Moreover, productivity growth. It is misallocation within sectors that there has been very limited domestic regional converge is one of the main culprits of low productivity growth. In over the last 20 years. The more developed regions are fact, comparing the distribution of productivity across sec- home to industries that have been able to take advantage tors with the distribution of productivity across firms with- of the market opportunities that NAFTA fostered. These in each sector reveals that firm-level productivity is signifi- high-growth industries and states have not built signifi- cantly more dispersed and has a larger standard deviation, cant backward linkages to other parts of the economy and with most firms below the sectoral average and a fat left other states. tail of unproductive firms. Productivity dispersion is much higher in Mexico than it is in the United States or in any 22. Aggregate productivity growth has been historically other Latin American country for which comparable data low, and the dispersion in labor productivity growth across are available.8 These productivity differences persist even states has been substantial. The contribution of total factor within narrowly defined sectors, such as cut-and-sewn ap- productivity (TFP) to growth was negative (−1 percent) parel manufacturing, where the most-productive firms are between 1991 and 2016, representing the weakest perfor- about 8 times more productive than the least-productive mance among Mexico’s structural and aspirational peers firms. One way in which this is manifested is that firms do (TFP growth had been marginally positive in the early not grow: firms that are in business for 10 years or more, 1990s). The contribution of labor has been driven mostly employ only marginally more labor than those in business by quantity, that is, a growing labor force, rather than la- for 5 years. Within-firm growth, on the other hand, drives bor quality, which has improved only moderately. A sim- productivity growth. In the manufacturing sector, the ple calculation suggests that, if Mexico’s productivity had within-firm component represents about 80 percent of grown at a pace similar to the rate during the high-growth productivity growth. However, very few firms are produc- period between 1950 and 1970 (1.3 percent a year), GDP tive at the global level: the domestic frontier is far below per capita would have been 128 percent higher than the the global frontier in all industries, and there has been lim- current level. Similarly, if TFP growth were similar to that of ited improvement in the past decade.9 At the same time, the United States, Mexico could reduce its GDP per work- high levels of informality persist. At the national level, 56.5 er gap with the United States by 22 percent. However, percent of total employment was in the informal econo- capital per worker would still be far too low to reach the my in 2017, although there is significant variation across same level, suggesting that improvements in productivity states.10 By some estimates, up to 97 percent of firms in the would need to be accompanied by increased and faster manufacturing and services sector rely partially or wholly capital accumulation. Labor productivity in the states of on informal labor and absorb 80 percent of labor and 70 Aguascalientes, Chihuahua, Nuevo León, Querétaro, and percent of capital.11 This large concentration of factors in Zacatecas increased at an average annual rate of over 1 informality contributes to low aggregate productivity. Es- percent between 1993 and 2015, while labor productivi- timates suggest that Mexican formal firms are on average ty in Chiapas, Hidalgo, Oaxaca, Quintana Roo, and Tlax- 84 percent more productive than informal firms.12 cala declined. As a result, the divergence in productivity growth has become more pronounced in the last two de- cades. Today, the differences are significant. Measured by 1.3 Poverty reduction, shared value added per worker, productivity is five times higher in prosperity, and inclusion Mexico City than in Chiapas. If labor productivity in states at the bottom had grown at the same rate as in Aguascali- dynamics entes (one of the top performers), their current GDP per 24. Progress towards poverty reduction and shared prosper- capita would be 81 percent higher. ity has been moderate. Starting in 1992, official monetary poverty rates rapidly increased due to the Peso Crisis of 23. Factor misallocation and informality are a drag on 1994-95 (by 1996, assets poverty and food poverty rates productivity growth in Mexico. Productivity differences reached their peak at 69 percent and 37.4 percent, respec- between sectors are large, but productivity dispersion be- tively).13 And although Mexico has made considerable tween firms within sectors is even larger. In recent years, as progress since 1996, the Global Financial Crisis of 2008–09 labor productivity in the manufacturing sector declined, and other prior shocks in output, encumbered poverty its share of total labor fell. At the same time, productivity reduction until 2014 (assets poverty in 2014 was 53.2 per- also decreased in the services and commerce sectors, yet cent versus 53.1 percent in 1992, while food poverty rate their share in total labor increased. Overall, reallocation SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 8 Busso et al.(2013) 9 Araujo et al. (2016). 10 Data of the National Institute of Statistics and Geography (INEGI). 24 11 Busso, Fazio, and Levy (2012). Estimates based on 2008 data. 12 Busso, Fazio, and Levy (2012). 13 For more detail on the official methodology for poverty measurement, see Annex 2. 1. Overview Figure 13. Monetary poverty rates (official), 1992–2016 80 69.0 70 60 51.6 53.2 49.0 52.0 50 53.1 52.4 51.1 52.3 50.6 37.4 47.0 47.8 40 42.9 30 19.4 20.0 20.6 16.8 20 21.4 21.2 19.7 17.5 18.2 18.8 10 18.6 14.0 0 1992 1994 1996 1998 2000 2002 2004 2005 2006 2008 2010 2012 2014 2016 Food poverty line Assets poverty line Minimum well-being line Well-being line Sources: Data of Consejo Nacional de Evaluación de la Política de Desarrollo Social (National Council for the Evaluation of Social Development Policy, CONEVAL); INEGI 2016a. Note: Numbers for 2016 are not strictly comparable. Figure 14. Shared prosperity is one of the lowest in the region, 2004–14 8 6 4 2 0 -2 -4 2011-2015 2008-2016 2006-2014 2006-2016 2006-2016 2006-2016 2006-2915 2008-2016 2006-2016 2006-2016 2007-2016 2006-2016 2006-2016 Chile Panama Uruguay Paraguay Brazil Colombia Bolivia Argentina Ecuador El Salvador 2006-... Honduras Mexico Peru Dom. Rep. Growth total population Growth bottom 40 Source: SEDLAC (Socio-Economic Database for Latin America and the Caribbean), Center for Distributive, Labor, and Social Studies, Facultad de Ciencias Económicas, Universidad Nacional de La Plata, La Plata, Argentina, and Equity Lab, Team for Statistical Development, World Bank, Washington, DC, http://www.worldbank.org/equityla. was 20.6 percent in 2014 versus 21.4 percent in 1992).14 25. Mexico’s bottom 40 percent have benefitted more from Yet, monetary poverty rates—also called wellbeing pov- economic growth than the overall population, nevertheless erty and measured according to the new methodology the country lags behind its peers in terms of shared pros- for income poverty measurement, dropped from 53.2 per- perity. Whereas overall average income contracted by 0.3 cent in 2014 to 50.6 percent by 2016. And over the same percent between 2006 and 2014, the first bottom quintiles period, monetary extreme poverty rates, also called min- of the income distribution (the World Bank’s indicator of imum wellbeing poverty, declined from 20.6 percent to shared prosperity) grew by 0.2 percent, according to the 17.5 percent (Figure 13). This is consistent with the recent main household survey ENIGH (Figure 14).15 Yet, even recovery of labor income shown in the labor force survey though growth has been pro-poor, it is far behind its peers data (since 2014 Q4). Nevertheless, even though monetary in Latin America.16 There is also significant heterogeneity poverty declined between 2014 and 2016, the rates corre- among regions and states of the country. Three states had sponding to 2016 are still higher than those in 2008. a lower average growth rate among the poorest 40 per- SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 14 In 2014, the well-being line was Mex$2,542 (about US$137) per capita in urban areas and Mex$1,614 (US$87) in rural areas, and the minimum well-being line was Mex$1,242 (US$67) and Mex$ 868 (US$46) per capita, respectively. 25 15 Numbers corrected for inflation (real growth). This growth is lower than the GDP per capita because the ENIGH survey underestimated higher incomes, which affected the average income. 16 Tabulations of Equity Lab, Team for Statistical Development, World Bank, Washington, DC, based on data in SEDLAC (Socio-Economic Database for Latin America and the Caribbean). 1. Overview Figure 15. Non-monetary components of the official multidimensional poverty measure, 2010–2016 70 60 Percent of population 50 40 30 20 10 0 Gaps in Lack of Lack of Low quality Lack of access to Lack of educational acces to access to and space in basic services access to attainment health care social the dwelling in the dwelling food services security 2010 2016 Figure 16. Share of population with simultaneous deprivation in several non-monetary components of poverty 80 60 Percent of population 40 20 0 2010 2012 2014 2016 Year Population with one social deprivation or more Population with three social deprivations or more Source: Data of Consejo Nacional de Evaluación de la Política de Desarrollo Social (National Council for the Evaluation of Social Development Policy, CONEVAL); INEGI 2016a. cent of the population than the average growth rate for 27. Moderate improvements in poverty reduction seem to their entire population between 2005 and 2014.17 be a consequence of insufficient growth and redistribution. Between 2004 and 2014, moderate poverty at US$5.5/day 26. Reduction in monetary poverty has been limited, how- per capita (2011PPP) in Mexico declined by 4.0 percentage ever there has been a consistent improvement in nonmone- points, from 37.6 percent to 33.6 percent. In contrast, over tary dimensions of poverty since they were first established the same period, moderate poverty in the Latin America and as components of the official multidimensional poverty Caribbean region declined 16.8 percentage points, from measure in 2010. The share of population experiencing 41.3 percent to 24.4 percent (Figure 17). Economic growth deprivation in one or more of the six nonmonetary dimen- was the main factor behind both the decline in poverty ex- sions of poverty included in the official multidimensional perienced across Latin America and Mexico over the past de- poverty measure, decreased from 74.2 to 70.4 percent cade. However, the relative importance of economic growth from 2010 to 2016. The share of population experiencing compared to redistribution was higher for Latin America SYSTEMATIC deprivation in three or more of these dimensions went and the Caribbean (79 percent versus 21 percent), than for COUNTRY down from 28.2 to 18.7 percent in the same period (Fig- Mexico, (55 percent versus 45 percent). Because of these DIAGNOSTIC ures 15 and 16). dynamics, the size of the middle class now lies below the MEXICO 26 17 Differences of income growth across the distribution at state level are calculated from poverty maps that use traditional ENIGH survey and Intercensal data, given that representative- ness at state level was one of the improvements in the MCS-ENIGH. 1. Overview Figure 17. Contribution of growth and redistribution to poverty reduction in Latin America and the Caribbean and Mexico, (moderate poverty at $5.5/day per capita), 2004–14 45 (a) Latin America and the Caribbean 45 (b) Mexico 41.3 40 40 37.6 35 35 33.6 -2.2 -1.8 30 30 -13.3 24.4 25 25 -3.6 20 20 15 15 10 10 5 5 0 0 Poverty 2004 Growth Redistribution Poverty 2014 Poverty 2004 Growth Redistribution Poverty 2014 Source: World Bank calculations following a Datt-Ravallion decomposition and based on SEDLAC (Socio-Economic Database for Latin America and the Caribbean), Center for Distribu- tive, Labor, and Social Studies, Facultad de Ciencias Económicas, Universidad Nacional de La Plata, La Plata, Argentina, and Equity Lab, Team for Statistical Development, World Bank, Washington, DC, http://sedlac.econo.unlp.edu.ar/wp/en/estadisticas/sedlac/estadisticas/. Latin America and Caribbean average (22.3 percent versus remains subpar: about half of students graduating from 36 percent). And the share of people who are vulnerable to primary education have insufficient proficiency in math poverty has grown, reaching nearly 43 percent of the popu- and language/communication.19 Poor educational qual- lation in 2014 (up from 39 percent in 2004), and above the 37 ity is a key challenge towards curbing poverty, reducing percent average in Latin America and the Caribbean.18 inequality and promoting mobility. Disparities in learning outcomes by type of school are telling: the proportion of 28. Challenges remain in terms of access to basic services insufficient proficiency among private schools is 13 per- and markets, this is particularly the case in the south of the cent, 51 percent in general public schools, 70 percent in country. Rural areas suffer a vicious cycle of low produc- community schools, and 80 percent among indigenous tivity, low investments in physical and human capital, and schools.20 Educational outcomes also vary across gen- high poverty rates, particularly in the south of the country. ders—boys outperform girls in science by 8 points—and At the same time, despite the positive effects of urbaniza- geographical area.21 For example, the populations of tion, most of Mexico’s poor live in urban areas with chal- Mexico City and Nuevo León have about 10 years of ed- lenges in the provision of services. And regional disparities ucational attainment on average, compared with only persist: in 2016, 68 percent of the extreme poor lived in slightly more than 6 years in Chiapas. These results have only six of Mexico’s thirty-two states. In addition, although implications for inclusive growth. Only half of 15-year-olds it has narrowed, income inequality remains high. There is in Mexico obtain the necessary skills to participate effec- evidence of intergenerational mobility in education and tively in society and the labor market.22 Taking advantage occupational status (though with unequal patterns by of digital platforms in education is gradually taking place geographic region). Yet, intergenerational mobility pro- (mostly at higher levels of education), but with significant cesses appear to be unequal for men and women, and for room for improvement at all levels. rural and urban populations. 30. Access to financial capital for poor and vulnerable house- 29. Access to education—a critical component of human holds is limited. In 2017, only 35 percent of adults in Mexico capital—is near universal, yet, student performance lags be- had an account at a formal financial institution, compared hind, quality remains an issue, and large regional disparities to 92 percent in the OECD, 73 percent in upper-middle in- persist. From near universal in the primary and lower sec- come countries, and 54 percent in Latin America and the ondary levels, enrolment falls to roughly over half of 15 to Caribbean. There is also a large gap in access between the 19 year-olds in upper secondary—the lowest rate among poorest 40 and the richest 60 percent of the population— countries of the Organisation for Economic Co-operation a difference of about 18 percentage points (compared to and Development (OECD). Student achievement also 5 points in the OECD).23 The cost of exclusion from the tra- SYSTEMATIC COUNTRY DIAGNOSTIC 18 Tabulations of Equity Lab, Team for Statistical Development, World Bank, Washington, DC, based on data in SEDLAC (Socio-Economic Database for Latin America and the Caribbean). MEXICO 19 Per the national PLANEA-ELSEN, 2017. 20 Educational Indicators Bank (Banco de Indicadores Educativos), RE01a. Share of students that obtain the educational achievement level (insufficient) in the areas evaluated in the PLANEA-ELSEN (2015) tests. 27 21 Refers to the scores of 15-year-old students (girls and boys) on the PISA 2015 science literacy scale. 22 According to results from PISA 2015. 23 Global Findex (Global Financial Inclusion Database), World Bank, Washington, DC, https://globalfindex.worldbank.org/. 1. Overview ditional financial system can be higher for the poor, who (below the OECD average, 89 percent). Low levels of social rely more on informal institutions. For instance, despite capital accumulation affect the poor disproportionately, reforms to increase the availability of access to credit, the as they rely more on informal networks. For instance, evi- cost of borrowing remains significantly higher in the retail dence shows that poor farmers in peri-urban areas Mexico banks, which tends to target low income consumers (in expand their chances to improve their poverty status as 2017, the total annual cost of personal loans ranged from social capital increases.30 30 to 60 percent in traditional banks compared to 80 to 120 percent in retail banks).24 33. Low productivity is common in microenterprises and smallholder agriculture. The productivity of microenter- 31. Uneven development and underinvestment in agri- prises—a key source of employment—is low, and their culture, water resource management and forestry limit contribution to GDP is shrinking (from 15 to 9.8 percent the potential of abundant natural capital. A large share of of GDP between 2008 and 2014).31 Most workers in micro- land in Mexico is highly fragmented, with implications firms are informal.32 Lagged regions are particularly vul- for agricultural productivity. Nearly 52 percent of land nerable: for example, labor productivity in southern states holdings are small (0 to 2 hectares).25 The numerous, is about 53 percent the national average.33 In agriculture, low-productivity, small-holder sector coexists with sig- low productivity is a critical constraint to inclusion. In nificant land concentration among large-scale produc- Mexico, over 75 percent of producers are semi-subsistence ers, vertically integrated in agribusiness supply chains smallholders (under 5 hectares), concentrated in the cen- chiefly located in the northern states.26 Thus, large dis- ter and south, employing traditional production practices. parities in productivity and growth between North and South remain. In 2015, five states in the North/Center 34. Female and youth labor force participation remain low, contributed 50 percent of Mexico’s agricultural GDP, with significant implications for growth. Only 45.5 percent while that of Southern states—home to a large number of working age Mexican women are part of the labor force, of traditional agriculture producers—was much lower. below the rate in Latin America and Caribbean countries For instance, the primary sector in Oaxaca had little par- (53 percent), and the OECD (51 percent). Mexico’s gender ticipation in the state’s GDP (6 percent), even though 32 gap in labor market participation has an associated with a percent of employment in the state is concentrated in potential loss up to 25 percent of income per capita.34 Ad- this sector. Gender inequality also persists in the sector, ditionally, a large share of Mexico’s youth is not in employ- whereby women farmers own only 10 percent of the ment, education or training (NEET), namely, 25 percent of land they work and are subject to higher productive 20- to 24-year-olds and 14 percent of 15- to 19-year-olds. exclusion.27 Natural capital is a particularly important This has implications for earnings—a 1 percentage point source of income for the poorest population, but it is increase in the proportion of NEET predicts a 7 percent increasingly at risk due to underinvestment and mis- reduction in earnings 20 years later–and equity—close to management. Twelve million people live in forest areas; 60 percent of the NEET population is in the bottom 40 per- of which 88 percent live in highly marginalized locali- cent.35 Mexico also has one of the biggest gaps between ties, and 62 percent are poor. There is a concomitance male and female NEETs, and the highest adolescent preg- between areas with high poverty and those that show nancy rate in the OECD. high rates of deforestation.28 32. Social capital is comparatively low, with implications 1.4 Sustainability issues for productive inclusion. Mexico lags behind in terms of group cohesion, and measures of safety and trust, based 35. Building public spending pressures over the medium on Social Development Indices. Intergroup cohesion, term will weigh on fiscal sustainability. Pressures stem from which measures ethnic tensions and discrimination, puts mandatory spending items such as pensions, as Mexico’s the country among the worst rankings compared to its population ages and the costs of pension reform transi- peers.29 Social support appears to have fallen in Mexico tion build up. Healthcare spending also faces upward pres- over the past decade. The share of people who reported sures stemming from demands for further coverage, grow- having a friend or relative whom they could count on fell ing costs from NCDs, and an aging demographic profile from 88 percent in 2005–07 to 80 percent by 2014–16 that would require higher spending on long term care.36 24 CONDUSEF (2017). See more details on the interest rates per type of bank in Section 3, below. 25 As shown in the 2016 INEGI agricultural census update. 26 Fox and Haigh (2010). 27 World Bank (forthcoming). 28 World Bank (forthcoming). SYSTEMATIC 29 International Institute of Social Studies, Indices of Social Development, 2010. COUNTRY 30 Méndez-Lemus and Vieyra (2017). DIAGNOSTIC 31 INEGI (2017). MEXICO 32 According to the economic census they amount to 93.2 percent, and 96.6 percent from the labor force survey. 33 Deichmann et al. (2004). 34 Cuberes and Tiegnier (2016). According to their estimations, the countries with the largest average total losses are Turkey (33.1 percent), Mexico (25.5 percent), and Italy (21.2 percent), while Iceland (9.2 percent), Norway (9.7 percent), and Finland (9.7 percent) display the smallest losses. 28 35 De Hoyos et al. (2016). 36 Simulations suggest that, if current technology-related spending trends continue and without considering income growth, health spending will increase from 8.9 percent of GDP in 2006 to 24.2 percent of GDP in 2050. With an income growth rate of 1 percent, the figure would be 15.5 percent, and of 3 percent, it would be 6.6 of GDP (Glassman and Zoloa 2014). Pressures also arise from people’s demand for improved 37. The social sustainability dimension of Mexico’s develop- 1. Overview education and multiple public services at the subnational ment is defined by several interconnected challenges that level. At the same time, decades of underinvestment in include lack of social mobility, unemployed youth, and crime infrastructure, including for public service delivery, call coupled with rule of law challenges. Inequality of oppor- for additional public investment to avoid compromising tunity is prevalent in the country and related to gender future medium-term growth. There is indeed space for and ethnicity—indigenous populations fare worse than efficiency gains and fiscal savings across categories of other race groups in all categories of social outcomes. spending, but the growing spending needs will require For instance, 64.6 percent of the indigenous population measures on the revenue side as well. Mexico still has a experience deprivations in terms of access to basic ser- relatively low tax-to-GDP ratio. While this ratio increased vices in the dwelling, compared to 16.3 percent for the significantly after the 2013-4 reform (by more than 3 per- non-indigenous population. In terms of learning out- centage points), it was 17.2 percent in 2016, with a base comes, the proportion of insufficient proficiency among still eroded by tax expenditures. Moreover, tax evasion students at private schools is 13 percent and at general (although declining) continues to impose significant rev- public schools 51 percent, while in indigenous schools it enue losses. Future reforms will need to tackle a program is 80 percent. In addition, the social and economic integra- of rationalization and efficiency gains on the expenditure tion of millions of young people who lack opportunities side. Further reforms to expand the tax base (and rates in is one of the highest priority social inclusion issues fac- some cases) will also be needed. Overall, policies will need ing the country. The NEETs phenomenon is one element to be earmarked in the context of a strong medium term in a chain of subsequent long-term consequences given fiscal framework. its several intergenerational dimensions. By definition, it implies that a sizeable part of today’s youth population is 36. There is high vulnerability to climate change and envi- not accumulating the human capital necessary to effec- ronmental degradation, but Mexico has been a leader in us- tively contribute to and benefit from labor and economic ing instruments to reduce risk. Mexico’s forests represent an opportunities. Exclusion and youth disenfranchisement is important natural asset, particularly for rural communities. also correlated with patterns of crime and violence. Low But the country faces the challenge of conserving and sus- trust in institutions is associated with weaker incentives tainably managing them while also meeting a growing de- for cooperation and coordination creating a dynamic of mand for timber products. Its payments for conservation low compliance with norms and free-riding problems. and environmental services program, has shown positive Levels of indicators such as civic engagement, trust in results in curbing deforestation and poverty alleviation, institutions and confidence in the government are low and it would need to be fostered moving ahead. Disparate in Mexico compared to its peers. Comprehensive policies availability of water in terms of both quantity and qual- to tackle these challenges will need to be designed and ity curtails the country potential for evenly distributed implemented. growth and inclusion, with a strong connotation of region- al, and urban/rural inequality. More than 35 million Mex- icans have limited access to water or receive low-quality 1.5 Structural impediments water services. The economic costs of water depletion and degradation have increased over the last 15 years. The wa- 38. The persistence of well-known structural issues seems ter sector would require significant investments to achieve to have hampered the long-term path of growth, inclusion efficiencies in the system and provide water to the popula- and sustainability. A key argument in this report is that a tion in a sustainable way. In the period from 2000 to 2005, misallocation of resources in the economy is taking place, natural disasters, exacerbated by climate change, were within and across sectors in the economy, across regions, more frequent and severe, increasing poverty between firms and individuals, and that it is harming productivity, 1.5 and 3.7 percent (depending on the poverty measure growth and inclusion.38 In this context, this part of the re- used). Research using the Human Development Index in- port looks to identify the most critical impediments that dicates that weather events could reduce up to two-year prevent the assignment of resources to their highest-val- achievements in the HDI index for affected municipalities ued use.39 Through a filtering process (described in detail in Mexico.37 Progress has been made in disaster risk man- in section 5.2 of this report), the SCD arrives at three key ar- agement, for example through the creation of the Fund for eas of structural impediments: (i) product and factor mar- Natural Disasters (Fondo de Desastres Naturales, FONDEN) kets, (ii) rule of law institutions; and (iii) resource allocation through which the federal government allocates budget and institutional policy coordination. These structural ex-ante for post-disaster response and reconstruction. impediments appear to be associated with unfinished or Nonetheless, climate change mitigation and adaptation incomplete reforms, as well as policies and programs that policies will need to continue. could be adjusted. It is argued that these areas of struc- tural impediments are linked to the limited growth experi- SYSTEMATIC COUNTRY enced and high economic disparities (among households, DIAGNOSTIC MEXICO 37 Rodríguez-Oreggia et al. (2013). 29 38 See Chiquiar and Ramos Francia (2006), Levy and Walton (2009), Hanson (2011) and Levy (2018). 39 For more detail on economic theory of allocative efficiency based on firm dynamics, see Lucas (1978), Hopenhayn (1992), Caballero and Hammour (2001), Haltiwanger (2011), and Bartelsman, Haltiwanger and Scarpetta (2013), among others. 1. Overview Figure 18. Structural impediments and growth Growth Inclusion challenges challenges • Convergence, growth and trade dynamics Structural • Limited productive inclusion (human, • Business cycles and aggregates shocks impediments financial, natural and social capital) • Limited capital accumulation • Unequal service delivery • Regional growth disparities • Decreasing labor productivity • Low productivity growth and high • Low female and youth labor force dispersion across regions, sector and firms participation Sustainability challenges • Fiscal and debt sustainability • Environmental sustainability • Social sustainability firms, and regions), which in turn seem to reinforce and by 1.4 percentage points over a period of twenty years, perpetuate these same structural impediments creating where the greatest potential gains are found in some of a vicious circle (Figure 18). They are also linked to be- the poorest states. low-potential levels of investment as well as to economic vulnerability to external shocks. Moreover, they are also 40. Limited access to finance hampers private sector growth connected to the suboptimal provision of quality inputs and household inclusion. In the context of the financial for human capital accumulation and to low levels of labor sector reform, significant progress has been made over force participation, particularly for women and youth. the last four years. For example, Mexico’s position in the World Economic Forum’s Global Competitiveness Report improved from 63 in 2013 to 36 in 2017. Credit (as a share 1.5.1. Product and factor markets of GDP) to the private sector increased and more than 13 39. Product market inefficiencies affect private sector million people gained access to financial services due to growth. There are weak linkages between the NAFTA ex- the reforms. Yet further and more accelerated progress port-oriented firms in the northern and central states of would be needed. Mexico continues to have one of the the country, and the large share of low-productivity firms lowest credit-to-GDP ratios among peers (including coun- not linked to those GVCs. To strengthen those linkages, it is tries at similar levels of income). The country’s unbanked of critical importance to reduce any hurdles to firms’ entry, population is far larger than the country’s level of econom- competition, exit, efficient costs of operation and flow of ic development would predict. The share of adults with an labor to productive and formal activities, as well as to their account at a formal financial institution decreased from 39 access to finance. While progress has been made through percent in 2014 to 35 percent in 2017, far below the level a number of recent reforms to strengthen competition of most comparator countries (for example, 80 percent of (such as telecommunication, energy, and competition adults have a bank account in Brazil or China). In addition, policy) and institutions (such as the Federal Commission while that proportion remained the same for men (39 for Economic Competition, COFECE), there is significant percent) between 2014 and 2017, it decreased for women room to reduce limited competition and concentration from 39 to 33 percent over these years. Moreover, barely at the federal and especially at the subnational levels. one third of SMEs have loans, and only one in four of the Regulatory barriers to competition at the local level tend poorest 40 percent of Mexicans have an account in a fi- to be dispersed across legal instruments, sectors and ju- nancial institution. In lieu, the unbanked population often risdictions, with significant negative effects. For example, use stores and retailers as their primary source of credit.40 high intermediation in the agribusiness chain is associated Moreover, if the unbanked are reached by formal credit with consumer loss and reduced benefits for producers. institutions, interest rates are often higher than warranted Moreover, reducing barriers to doing business can also by the risk. Faster progress in traditional access to finan- bring productivity gains for private firms and welfare gains cial services, together with new innovative approaches to households, especially in the south of Mexico. Recent through technology, present a great opportunity to boost SYSTEMATIC COUNTRY evidence shows that improving the efficiency of factor al- investments and broad-based growth. Despite positive re- DIAGNOSTIC location by a quarter could increase the annual growth of forms, the banking sector remains concentrated (the top MEXICO the output of private sector manufacturing and services seven banks control 78 percent of the sector’s assets). The 30 40 See Skelton (2008). predominance of banking impacts the development of timeliness of economic disputes, for instance, those relat- 1. Overview capital markets, through the control of distribution chan- ed to contract enforcement. Nevertheless, most civil and nels and intermediaries. Limited competitive pressures commercial cases are still handled using the unreformed could explain why overhead and administrative costs in justice system. In terms of the application of the rule of commercial banks have not declined and intermediation law, impunity continues to be a major challenge. Mexico margins have increased over recent years. Moreover, sub- ranked 58 of 59 countries in the Global Impunity Index sectors of the economy with high levels of concentration (GII) published by UDLAP in 2015. According to the MCCI, appear to obtain a relatively high proportion of financial the probability of a crime being reported, investigated, resources. Despite positive reforms, the banking sector re- prosecuted and resolved in the Mexican criminal system mains concentrated (the top seven banks control 78 per- is only 2.95 percent. Beyond criminal offences, impunity in cent of the sector’s assets). The predominance of banking Mexico extends to civil and administrative crime. impacts the development of capital markets, through the control of distribution channels and intermediaries. Lim- 43. Uneven application of the rule of law and shortcom- ited competitive pressures could explain why overhead ings in the control of corruption increases the cost of doing and administrative costs in commercial banks have not business affecting private sector growth and impacts poorer declined and intermediation margins have increased over households through higher unproductive out-of-pocket recent years. Moreover, subsectors of the economy with expenses. Transparency International’s Corruption Percep- high levels of concentration appear to obtain a relatively tions Index, placed Mexico in 123rd out of 176 countries, high proportion of financial resources. scoring last among OECD nations. Furthermore, Mexi- co’s relative place in performance in control of corrup- 41. Well-intentioned labor legislation aimed to protect tion rankings has been worsening in relative terms over workers could be unintentionally contributing to foster in- time compared to peers in the LAC region. WEF’s Global formality and labor misallocation. Recent reforms made Competitiveness index (2017-18) ranks Mexico in 123rd an impact, albeit moderate, on the persistent problem place among 137 countries with a negative trend, with informality (informality dropped from 59.7 percent in corruption seen as the most problematic factor for doing 2012 to 56.9 in 2017).41 Yet, still a large portion the cur- business by the private sector. About 65 percent of entre- rent employment in Mexico is informal, including wage preneurs in Mexico report having missed a business op- earners that do not have access to social security and portunity due to unfair competition, where competitors self-employed workers. To the extent that social securi- use political influence or handouts.44 Some studies argue ty is primarily financed by wage-based contributions, that Mexican households spent 14 percent of their income and is not fully valued by affiliated workers, it acts as a on unofficial payments, but that share could be higher for tax on salaried employment.42 This incentivizes firms to low-income households.45 The poor and vulnerable are move towards non-salaried and informal contracts, and also particularly susceptible to clientelism—the exchange the illegal evasion of social security, with consequences of political support for typically short-term benefits—giv- on productivity and growth. Moreover, settling labor dis- en their liquidity constraints and higher time preference putes when firing occurs (based on formal employment for the present.46 contracts) is also a long and expensive process, which together with other factors, deters formal labor demand 44. Shortcomings in institutional effectiveness may also be and does not help workers. reflected in the rise in crime and violence, which also deters economic activity and household welfare. The WEF’s Global Competitiveness index (2017-18) ranks Mexico in 131th 1.5.2. Rule of law institutions place among 137 countries in terms of the business cost 42. Mexico ranks low on international indicators for the rule of crime and violence, and in 131th place both in terms of of law. In the Rule of Law Index 2017–2018, for example, organized crime and the reliability of police services. Inse- Mexico ranked in 92nd place out of 113 countries.43 Mex- curity and crime are consistently rated as the top problem ico is at the bottom of the pack of upper middle-income for firms to operating in Mexico and by Mexican citizens. countries (34 out of 36 countries), only above Turkey and Increased crime and violence can contribute to a worse Venezuela and below China, Russia, and the DR. A second allocation of productive assets, including notably through wave of legal reform to the civil and commercial justice their effect on labor market outcomes and hampered in- could help significantly. To incentivize a faster and more vestment, with implications for equity and growth. Crime transparent resolution of lawsuits, an initiative approved and violence also seem to be affecting the accumulation by Congress in October 2017 increased the scope of oral and use of human capital, diverting away resources from procedures to resolve commercial disputes. Moving from their highest valued use, via an unskilled young labor force. written to oral trials can help improve the outcomes and SYSTEMATIC COUNTRY DIAGNOSTIC 41 INEGI, informality rate based on ENOE survey data. Indicators of employment and occupation published May, 2018. MEXICO 42 Levy (2009). 43 WJP (2018). 31 44 WJP (2018). 45 WJP (2018). 46 Díaz-Cayeros et al. 2016. 1. Overview Recent evidence has found a high correlation between the inefficiencies is particularly important as social spending share of NEETs and murder rates between 2008 and 2013. pressures are likely to continue to grow. Public sector pro- curement presents a good avenue to achieve fiscal savings while increasing expenditure efficiency, but there are sev- 1.5.3. Resource allocation and eral others in the accounts of the economic and functional institutional policy coordination classifications of the budget. Mexico has a reasonable plat- 45. Inefficiencies in public resource allocation may result form for providing social protection, and social assistance from tax and expenditure policy issues. In general, issues of programs are generally well-targeted. However, Mexico’s tax structure, and tax policy and administration may lead large number of social assistance programs has resulted to inefficiencies in the economy or limitations in revenue in some degree of duplication, overlaps (of programs and mobilization. Inefficiencies may also be linked to labor beneficiaries), and fragmentation. There are at least 5,491 and social insurance revenues or expenditure regulations social development programs across government levels. and policies (and their enforcement), special subsidies, This complexity is heightened by the lack of a single/ the distribution of resources to subnational governments universal personal (beneficiary) identification that help through the intergovernmental transfers system, as well as policy makers avoid overlaps. One of the main obstacles technical and allocative inefficiency across spending cate- to enhancing spending efficiency and equity is the strong gories in the budget that affect public service delivery, in- fragmentation observed in health care and allocation effi- cluding investments. This sub-section looks at these issues ciency issues across sectors. in the case of Mexico given their relevance. 48. The intergovernmental transfer system has a weak 46. The tax structure is tilted towards direct taxes, partly equalization power to reduce horizontal disparities of fiscal reflecting significant tax expenditures in indirect taxes. In- capacity and the higher needs for services in poorer states. come tax revenue makes up nearly 42 percent of Mexico’s The spatial concentration of tax bases associated with so- total tax revenue, well-above the averages for both the cioeconomic regional characteristics results in horizontal Latin America and Caribbean region and the OECD. By fiscal imbalances and fosters further regional inequalities contrast, Mexico derives less than 40 percent of its reve- in service delivery.47 This is not compensated for by the nue from indirect taxes, whereas these taxes account for intergovernmental transfer system. While the positive over half of revenues in Latin America. Mexico’s tax struc- reforms of 2007 to the distribution formula for Participa- ture reflects differences in its revenue-generating capacity ciones (the unconditional federal transfer to states and and has economic and distributional implications. For ex- municipalities) enabled a better allocation of resources ample, due to exemptions and zero-rating in the VAT re- and introduced some, albeit limited, incentives to improve gime (among other factors), the country collects only 31.5 subnational government tax collection effort, its equaliza- percent of the revenue that it could theoretically collect tion effect is limited, as it does not give higher per capita if VAT was applied at the standard rate to all goods and transfers to less developed regions. The distribution crite- services. By contrast, this VAT revenue ratio is 42.6 percent ria of the Aportaciones transfer for education (Fondo de in Colombia, 55.1 percent in Peru and 64.4 percent in Chile. Aportaciones para la Educación Básica) was modified in Despite progress in streamlining tax expenditures in the 2007 to introduce demand-side considerations. In terms of 2013-14 reform, they remain high (at 3.7 percent of GDP in health, there are improvements in the allocative efficien- 2017). Tax expenditures not only reduce public resources, cy of the federal fund earmarked for health expenditures, they can also have broader distributional and economic though challenges remain.48 Despite these positive devel- effects. opments, regional disparities in per capita spending are substantial, with poorer states that have the larger gaps 47. Public spending is highly rigid, with significant space in service delivery and outcomes most affected. Per capita to reduce inefficiency, and with a relatively low share of in- accumulated spending of state and local governments in vestment. By some measures, rigid spending represents 80 the northeast states is more than 20 percent higher than percent of total spending. This reduces fiscal space for in- in the southwest. Other intergovernmental transfers have vestment in infrastructure and the capacity of the author- regressive rather than equalizing effects. Federal transfers ities to react to economic shocks with adequate counter earmarked to promote investment in social infrastructure cyclical policies and to changing priorities over time. in poorer municipalities have had limited impact on re- During the needed fiscal consolidation of the last years, gional development and welfare effects. Other transfers all discretionary spending was squeezed given the inflexi- (such as Ramo 23), which are smaller in size (and represent bility of other spending categories, reducing investments less than 7 percent of total transfers) are currently used as to historically low levels, which can affect medium-term gap fillers (and are sometimes over executed) without a growth. Public spending inefficiencies have been re- clear developmental impact. A more rules-based alloca- SYSTEMATIC duced recently, but much work remains ahead. Reducing tion of these transfers, along with other adjustments in COUNTRY DIAGNOSTIC MEXICO 47 State and municipal tax collections per inhabitant in the northeast and center north regions is 4.5 times the collection of state and local governments in the southwest. Mexico City collects 15 times more state and municipal revenues per capita than Chiapas, Oaxaca and Tlaxcala and more than 10 times than in Guerrero and Zacatecas, the states with the lowest 32 subnational tax collections. 48 The fund is the Fondo de Aportaciones para los servicios de salud. The 2003 reform to the General Health Law revised the fund allocation rule, linking the transfer to the number of beneficiaries, and a ponderation to health needs, spending, and spending efficiency. this area, can have significant impact in reducing regional was undertaken. The SCD applied two first filters to identi- 1. Overview inequalities and fostering long term growth. Notwith- fy the main structural impediments (discussed in detail in standing the needed adjustments in the transfer system, section 5.1) and other key constraints (that also arise from sub-national governments would also need to be encour- chapters 2, 3, and 4) that are holding back faster growth aged (using different mechanisms) to collect more own and inclusion with sustainability in Mexico. The first filter source revenues. used a research and policy review of the extensive work done on Mexico by local and international scholars and 49. Limited coordination among public entities (includ- practitioners, analytical work conducted by the World ing among levels of government) and between public and Bank, and other international organizations, and research private sector, constraint the strategic planning of invest- papers and policy reviews published in peer-reviewed ments (including PPPs) and may contribute to suboptimal journals and academic and policy outlets. The second filter outcomes. The planning and prioritization of investments relied on several rounds of consultations with local and across the country is an area where further improvements international scholars in academia and experts on Mexico, to coordination could make a big difference. Over the practitioners, government authorities in various sectors, last years investment planning has improved significant- private sector representatives, civil society, development ly under the leadership of the Ministry of Finance, but partners, and other external experts50. Additionally, two further work would be needed. Strategic investments to broad rounds of consultations with all stakeholders were support growth and inclusion could be better set, build- held in-country, in October 2017 and April 2018, including ing long-term pipelines of key development projects and field visits to several states. After applying these two initial bringing-in more private sector resources, under a stra- filters for prioritization, twelve priority areas of structural tegic plan that could go beyond administration periods. impediments and key constraints to growth were selected. This would also help the maximization of private sector Within these (already) priority issues, the SCD used anoth- financing of infrastructure in the country, while guarding er two filters to further prioritize specific policy levers. A for fiscal risks. Problems of coordinated policies can be ob- data-driven benchmarking exercise was used as a third fil- served among federal entities as well as between federal ter with the aim to measure Mexico’s performance in com- and sub-national entities (where there is the highest need parison to the world and selected structural peer groups for investments). Differences in institutional effectiveness of countries on a number of specific issues. Additionally, across states may have contributed to the duality of Mexi- the SCD used the World Bank country expert knowledge to co’s production system. Case studies on sub-national gov- calibrate the prioritization of policy areas as the fourth fil- ernments show that the process through which actors and ter. Table 1 summarizes the more detailed policy priorities. groups interact to influence the allocation of resources and the design of policies is an important factor that con- 51. It is important to point out that this list of priorities does tributes to this divergence.49 For example, the divergence not mean that other issues that were excluded from the list in economic performance between the states of Queréta- are not important. Rather, the aim is to provide a sense of ro and Puebla may be, at least partly, linked to differences priorities and policy direction, based on existing knowl- in the interaction between policymakers, chambers of edge and the filters of prioritization used and derived from businesses and politicians, which led to distinct institu- the diagnostic undertaken. This diagnostic also acknowl- tional arrangements and business development in each edges a number of knowledge gaps that require further state. In this context, limited local capacity for planning study (section 5.3) and may affect the views on priorities. may exacerbate the challenges of coordination. Munici- palities often lack planning capacity, or a strategic vision 52. Moreover, the priority policy areas do not include strictly that considers a territorial and interjurisdictional planning macroeconomic stability policies, given the excellent track approach, and instead focus on sectoral, frequently siloed record of Mexico in this area. However, the diagnostic as- and low-return investments. sumes and sets prudent and sustainable fiscal and mon- etary policies as a critical pre-condition for growth, inclu- sion, and sustainability. 1.6 Priorities 50. A process of prioritization of structural impediments and other key constraints to growth, inclusion, and sustainability SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 49 World Bank (2017). 50 Teams also provided written inputs, focusing on (i) the most important analytical pieces and sources of their sector, (ii) the key development challenges for sustainable, inclusive 33 growth, and (iii) the main challenges in their sectors. Consulted Global Practice teams included: Agriculture; Education; Energy Extractives; Environmental and Natural Resources; Finance, Competitiveness and Innovation; Governance; Health, Nutrition and Population; International Finance Corporation; Macroeconomic, Trade and Investment; Poverty and Equity; Social Development; Social Protection and Labor; Transport and ICT’s; Urban and Disaster Risk Management; Water; Gender CCSA; Jobs CCSA; and Climate Change CCSA. 1. Overview Table 1: Priority policy areas Policy areas Structural Impediments 1. Concentration (and market power) in critical input markets and barriers to entry at the local level ―― Promote (and strengthen) regulation and supervision to curb concentration (and market power) in critical input markets and complete implementation of product market reforms. ―― Address regulatory failures particularly at the local level (reducing the costs of firm establishment and operation), reduce protection for incumbents, remove preferential treatment for politically-connected firms, improve business environment (including protecting SMEs from crime and extortion, removing informal fees). ―― Strengthen SME capabilities to link with large companies that work within GVCs. 2. Access to finance ―― Incentivize lending to MSMEs, including by establishing the legal framework for instruments such as Asset-Based Lending. Product and factor markets ―― Strengthen mechanisms to foster financial inclusion, including through new technologies (e.g., fintech) and with emphasis in rural areas. ―― Promote the development of the domestic capital market (expanding to a broader set of firms) by fostering competition. ―― Reform state banks to provide them more developmental oriented goals. 3. Labor market rigidities and informality ―― Reduce the costs of formalization for firms and workers, gradually severing the link between payroll taxes and social insurance programs, cutting the costs for hiring and firing, and reducing the length of legal procedures in labor courts. ―― Strengthen programs targeted at firms to improve formalization and entrepreneurship. ―― Strengthen the relationship between the education system and the private sector to equip workers with the skills demanded by employers. 4. Access to justice ―― Increase access to justice for vulnerable populations. ―― Improve litigation times of most frequent cases, such as debt cases, wrongful dismissals, non-violent drug-related offences. ―― Accelerate the implementation of reforms to enhance commercial justice. ―― Improve contract enforcement and enforcement of property rights. 5. Control of corruption Rule of law institutions ―― Implement aggressive legislation to fight corruption (e.g., public-private contracts, public procurement, reduce cash transactions between citizens and public servants). ―― Fully implement the National Anti-Corruption System and extend reforms to the subnational level. ―― Fully apply OECD’s anti-corruption convention. 6. Crime and violence ―― Implement programs to promote social cohesion and support youth-at-risk, including youth employment programs. ―― Strengthen and hold accountable institutions in charge of providing public safety and preventing crime. 7. Tax structure and tax expenditures ―― Gradually increase revenue mobilization through: base broadening, tax rates where needed, and tapping undertaxed bases (digital economy and subnational), while considering distributional impacts. ―― Adjust the tax structure and reduce tax expenditures to gradually increase the share of indirect taxes (while reducing payroll taxes). ―― Reduce collection gaps, through the modernization of tax administration tools and stronger voluntary compliance measures. 8. Public spending: rigidities, inefficiencies and distributional issues Resource allocation and ―― Reduce public spending inefficiencies to create fiscal space for infrastructure (e.g., room for efficiencies could be found in institutional policy public procurement, wage bill, consolidation of public sector programs; reducing fragmentation in the health system). coordination ―― Reduce existing spending rigidities (pensions, wages, other entitlements). ―― Reduce dependency of payroll taxes for social insurance programs. ―― Reform the current pension systems to ensure sustainability and promote adequacy and equity. ―― Explore ways to reduce overlaps and expand reach of the social protection system to the poorest and most marginalized. ―― Reduce vertical and horizontal fiscal gaps in subnational governments. ―― Strengthen equalization capacity of the intergovernmental transfer system (while applying stronger incentives for fiscal effort and curtailing ad hoc transfers) to reduce regional inequalities in service delivery and outcomes. ―― Build larger fiscal buffers to use during difficult times. SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 34 1. Overview Policy areas Structural Impediments 9. Institutional coordination and investment planning shortcomings Resource ―― Enhance coordination between the public and private sector, especially at the subnational level, with emphasis on productive allocation and investment. institutional policy ―― Enhance public investment management process, starting with long term strategic planning (beyond a government period), the coordination development of a solid pipeline of projects with an enhanced feasibility analysis process and cross-institutional coordination. ―― Promote integrated multisector urban planning and service provision. 10. Investment in infrastructure ―― Raise investment in infrastructure, including through private sector participation. ―― Further strengthen and streamline the PPP framework, while managing fiscal risks. ―― Invest in transport, logistics and trade facilitation to strengthen the Pacific side export corridors (including gulf to pacific corridors), as well as to achieve higher efficiencies domestically. ―― Reduce vulnerability of existing infrastructure and integrate concept of resilience into new investments. ―― Expand access to services in lagged regions. ―― Invest in water infrastructure modernization and electricity transmission capacity. 11. Quality and utilization of human capital Other structural constraints to ―― Support youth in making effective school-to-work transitions. growth, inclusion ―― Ensuring basic learning with a particular focus on closing gaps in attainment outcomes. and sustainability ―― Promote universal health care reform based on a standard benefit package that promotes horizontal and vertical integration of services. ―― Strengthen the primary health care system with a focus on prevention and promotion. ―― Eliminate barriers that hinder participation of women in the labor market focusing on access to quality childcare and promoting gender-neutral flexible work arrangements. 12. Management of natural capital ―― Adopt long-term planning and prioritization of investments in water security. ―― Strengthen effectiveness of current support to agriculture, forestry and other productive, resource-based sectors by focusing on long-term productivity and competitiveness. ―― Build resilience to deal with climate change and extreme events and foster climate-smart growth. SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 35 2. Growth 2.1 Convergence, growth, 55. On the supply side, growth was primarily driven by the and trade dynamics expansion in services. Overall, the services sector recorded annual average growth of 2.9 percent over this period and accounted for 65.7 percent of value added by 2017. This 53. Mexico’s economic growth averaged only 2.4 percent underpinned a further structural transformation of the between 1980 and 2017, limiting its convergence progress Mexican economy, which saw a shift away from the pri- in convergence relative to U.S. per capita income (Fig- mary (−0.4 percentage points) and secondary (−6.8 per- ure 19, Figure 20). The country’s per capita GDP today centage points) sectors toward the services sector (+7.2 stands at 34 percent of U.S. per capita GDP, compared percentage points) between 1993 and 2017. Telecom- with 49 percent in 1980. The Republic of Korea had, by munications and financial services were the most rapidly 2017, reached 66 percent of U.S. per capita income, even growing service subsectors. Between 1993 and 2017, the though, in 1980, it had less than half the per capita in- telecommunications subsector increased its share of value come of Mexico. Other countries, such as Chile, Malaysia, added from 0.3 to 2.4 percent, while the financial sector in- and Poland, which all had lower per capita incomes than creased its share from 1.2 to 4.9 percent. Over the last two Mexico in 1980, had, by 2017, achieved higher income decades, these subsectors average growth rates reached levels and had closed the gap with respect to the United 11.2 percent and 8.8 percent, respectively, as first mobile States. telephony and then the Internet became central com- ponents of the economy; and as the financial sector was 54. Private consumption has been the main driver of growth opened up to foreign participation after the 1995 crisis. on the demand side. Private consumption grew an average 2.8 percent a year between 1994 and 2017, contributing 56. Despite the overall decrease of manufacturing in value more than two-thirds to GDP growth (Figure 21). Invest- added, some private sector manufacturing subsectors also ment expanded 2.3 percent a year, below overall GDP showed rapid growth rates on the back of trade liberalization. growth, as public capital spending entered a secular de- The production of transport equipment, grew at an average cline. Net exports, on the other hand, contributed little to rate of 5.6 percent over the last 20 years, as Mexico joined SYSTEMATIC growth: while gross exports grew by as much as 6.2 per- regional and global value chains in the auto and aerospace COUNTRY cent per annum on the back of the boom in manufactured subsectors (Box 3 and Box 4). Other technology manufac- DIAGNOSTIC exports, imports expanded at a similar pace. This is partly turing subsectors such as electronics and computers have MEXICO a result of the high import content of Mexico’s manufac- also become central to GVCs and important growth drivers 37 tured exports (40 percent). The trend started to marginally (growing at an annual rate of 7.8 percent over the same pe- change in 2010–17 (Figure 22). riod). Other manufacturing industries, however, especially 2. Growth Figure 19. Average GDP growth rate, Mexico and comparators, 1980–2016 (%) 7 6 5 4 3 2 1 0 Korea, Rep. Malaysia Thailand Turkey Chile Colombia Peru Mexico Uruguay Brazil Argentina Source: Data of WDI (World Development Indicators) (database), World Bank, Washington, DC, http://data.worldbank.org/products/wdi. Note: GDP growth rates measured in constant local currency units. Figure 20. Per capita income as a share of U.S. per capita income 0.7 0.6 0.5 0.4 0.3 0.2 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 2013 2016 Malaysia South Korea Poland Chile Mexico Source: Conference Board, Total Economy Database, November 2017. those that tend to be more labor intensive, such as wood, forestry) was estimated at 3.2 percent. Following a period of textiles, and clothing, declined in importance over the same modest growth during the 2000s, the performance of the period, dragging the overall share of manufacturing in val- primary agriculture sector has significantly improved in the ue added down from 17.2 percent in 1993 to 16.6 percent in present decade, with an average annual growth in agricultur- 2017. Oil, gas, and mining, which accounted for over 9 per- al value added of 2.8 percent during 2010–16, versus 1.5 per- cent of GDP in 1994, contributed positively but marginally cent during 2000–09. Between 2012 and 2016, TFP growth in to growth after 2004, as oil production began to decrease, the primary sector increased by an accumulated 9.1 percent, falling below 2 million barrels a day by the end of 2017 compared to a decrease in the secondary and tertiary sectors (from a 2004 peak of 3.4 million barrels a day). This reduc- (Figure 23). The sector employs a significant share (13.5 per- tion was the result of years of declining investment by the cent) of the labor force nationally; in the southern states, it state-owned oil company, Petróleos Mexicanos (PEMEX). provides jobs for 20 to 40 percent of the labor force. However, through its backward and forward links, the sector’s contribu- 57. The structural transformation of Mexico’s economy has been tion to GDP and employment are more substantial. In 2017, SYSTEMATIC accompanied by a declining share of primary agriculture in GDP, the share in total GDP of the agricultural and agribusiness sec- COUNTRY although its employment share remains significant. In 2017, tors51 stood at 7.6 percent, more than double the contribution DIAGNOSTIC the share of total GDP contributed by the primary agriculture of primary agriculture alone. For example, in Mexico’s north- MEXICO sector (including crop and livestock production, fisheries, and ern and central states, export-oriented agriculture provides 38 51 The agribusiness sectors included in this share are food, beverages and tobacco. 2. Growth Figure 21. Contribution to GDP growth, by component (demand side) 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 Average Average Average 1994-2001 2002-2009 2010-2017 Private Consumption Government Consumption investment variation in Stocks Net Esports Statistical Discrepancy Source: World Bank calculations based on data of the National Institute of Statistics and Geography. Figure 22. Contribution to GDP growth, by component (supply side) 3.5 3.1 3.0 2.5 1.5 1.2 0.5 0.0 -0.5 Average 1994-2001 Average 2002-2009 Average 2010-2017 Taxes Construction & utilities Services Manufacturing Oil, Gas & Mining Agriculture, forestry, fishing & hunting Source: World Bank calculations based on data of the National Institute of Statistics and Geography. Box 3. The motor vehicle industry in Mexico One of the industries that has benefited the most from the rise in manufactured exports following the launch of NAFTA is the motor vehicle industry. The industry’s share in total GDP more than doubled between 1993 and 2017, from 1.4 to 2.9 percent; and its share in manufacturing GDP increased by 10 percentage points, from 8.3 to 18.3 percent. By 2015, Mexico had become the seventh largest producer of motor vehicles in the world. Mexico’s share of U.S. vehicle imports rose from 10 to 26 percent between 1995 and 2015; for auto parts, the share increased from 23 to 35 percent in the same period. Mexico’s auto parts and vehicle assembly subsectors attracted US$6.9 billion in FDI in 2017, an amount equivalent to 23 percent of total FDI in that year. As of December 2015, 875,382 people were directly employed in the automotive industry: 81,927 in the manufacture of automobiles and trucks, and 793,456 in the auto parts sector. Average salaries in the assembly and auto parts industry are higher than in other manufacturing sectors in Mexico (almost triple in the case of auto assembly). Sources: ProMexico 2017; Secretaría de Economía. SYSTEMATIC COUNTRY jobs on farms to large numbers of Mexicans, but also off-farm natural, and human capital, increased by almost 40 percent DIAGNOSTIC activities such as packaging and agrifood processing. between 1995 and 2014, but the growth rate is lower than MEXICO the average in Latin America (65.5 percent) and among up- 39 58. The increase in total wealth is driven by produced capital. per-middle-income countries (161 percent). In per capita Mexico’s total wealth, measured as the sum of produced, terms, Mexico’s wealth only rose 5.2 percent, the second 2. Growth Box 4. The aerospace industry in Querétaro: A success sto- ry of public-private sector collaboration The aerospace industry is one of the most rapidly growing industries in Mexico. Aerospace-related manufacturing ac- tivities grew by 178 percent in real terms, compared with 22 percent for manufacturing as a whole, between 2005 and 2017. In 2017, it accounted for 0.8 percent of Mexico’s manufacturing GDP. The total number of aeronautical companies rose from 60 in 2005 to 330 in 2016, of which more than two-thirds are engaged in manufacturing; 11 percent in main- tenance, repair, and overhaul; 13 percent in the design and engineering side; and the remainder in academic activities (universities and research centers). These companies are concentrated in the states of Baja California, Chihuahua, Nuevo León, Querétaro, and Sonora, states in which the public, private, and educational sectors have worked together to form specialized clusters that have fostered the development of the industry on a regional scale. One of the most prominent examples of this cooperation is the case of Querétaro, home to Mexico’s only university specialized entirely in the aerospace sector. The Aeronautical University of Querétaro, a public university financed in equal parts by the federal and the state governments, was founded in 2007 shortly after the arrival of the Canadian company Bombardier that initiated the establishment of the aeronautics cluster in the state. Private sector firms and the university collaborated in defining the skills and profiles required for the workforce in the cluster, as well as the qualifications needed among teaching staff. In the initial phase of the cluster, the university’s curriculum focused on the basic technical skills needed for the simple assembly of parts. In a second phase (2009 onward), reflecting the growing sophistication of the cluster, the university introduced a tertiary-level program to train technicians in more advanced manufacturing skills. Beginning in 2010, the university offered the full master and doctoral degrees required for the assembly of airplanes, as well as design and innovation. Today, the aerospace industry in Querétaro directly employs approximately 10,000 staff, of whom 7,000 were trained at the university, in 53 companies in the cluster. It accounts for 4.5 percent of the state’s GDP and 37.0 percent of all aerospace exports in Mexico. Sources: ProMéxico 2017; data of Instituto Nacional de Estadística y Geografía (National Institute of Statistics and Geography, INEGI); interview with university management. Figure 23. Total factor productivity (TFP) by sector (1990=100) 120 119 115 110 105 100 95 93 90 91 85 85 80 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 Total Primary Secondary Tertiary Source: INEGI lowest growth rate among the comparator countries after 20 percent between 1995 and 2014. Nonrenewable natural Turkey (Figure 24). Mexico experienced an erosion in hu- capital, including oil, gas, and minerals, increased as a share man capital (the value of earnings over a person’s lifetime), in total natural capital to 40 percent of this asset class (up which fell by 7.3 percent over this period and, in 2014, only from 28 percent in 1995). Not captured in the data is the accounted for 54 percent of total wealth, compared with significant water depletion and degradation, the costs of 61 percent in 1995. This was primarily caused by stagnant which rose from 0.5 to almost 0.8 percent of GDP between wages. Natural capital also declined slightly due to unsound 2003 and 2015. On top of these costs, there were signif- management and underinvestment in renewable resourc- icant reductions in asset values because of soil and water SYSTEMATIC COUNTRY es, including water, forests, cropland, and pastureland, contamination. The trend over the last 15 years shows the DIAGNOSTIC suggesting a lack of investment in rural areas in particular. loss associated with the depletion of these natural resources MEXICO For example, the per capita value of agricultural land fell by increasing 10-fold.52 40 52 INEGI (2016b). 2. Growth Figure 24. Contributions to growth in total wealth (%), 1995–2014 120 100 80 60 40 20 0 -20 Chile Peru Uruguay Korea, Rep. Poland Malaysia Thailand Romania Colombia Mexico Turkey Produced capital Natural capital human capital Net foreign assets Source: World Bank calculations based on World Bank 2018. Figure 25. Cyclical component of GDP in the United States and Mexico (annual % change 5 3 1 -1 -3 -5 -7 -9 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 q1 q1 q1 q1 q1 q1 q1 q1 q1 q1 q1 q1 q1 q1 q1 q1 q1 q1 q1 Mexico uS Source: World Bank calculations based on data of U.S. Federal Reserve and INEGI. Note: Cyclical component estimated using the Hodrick-Prescott filter 59. As a result of close trade integration in North America while recessions increased by 1 quarter on average. Mean- through NAFTA, the Mexican and U.S. economies have be- while, the cumulative growth during expansions increased come more synchronized. The business cycles in the two 12 percentage points, to 23 percent, but expansions were countries have co-moved more closely since the mid-1990s slower because of lower average annual growth rates. Reces- (Figure 25). This became particularly pronounced following sions, on the other hand, deepened by 5 percentage points the launch of NAFTA in 1994, given the close trade relation- to −7 percent. While a similar trend can be observed in the ship that developed between the two countries: in 1990, the other NAFTA partners, the effect was far larger in Mexico. United States was the market for 70 percent of total Mexican exports; by 2016, the share had risen to 81 percent. Similarly, 61. The share of Mexico in world trade has expanded apprecia- 47 percent of Mexico’s imports originate in the United States. bly in recent decades. Between 1990 and 2016, Mexico’s ex- ports increased 14-fold as world trade expanded 10-fold, so 60. At the same time, business cycles have become more pro- that the share of Mexico’s exports in world exports rose from nounced. As economic integration with the United States 1.9 to 2.6 percent. The country’s export basket also under- SYSTEMATIC deepened and the business cycles of the two countries went significant change: fuels accounted for 37.5 percent of COUNTRY became more closely synchronized, both the magnitude exports in 1990, but only 4.8 percent by 2016, while the coun- DIAGNOSTIC and the duration of economic expansions and contractions try’s share in the global fuel trade fell from 12.8 to 1.7 percent. MEXICO increased in Mexico (Figure 26). A business cycle lasted near- The share of transport equipment in the export basket dou- 41 ly 14 quarters before NAFTA, increasing to 32 quarters after bled from 12.4 to 24.7 percent, while the share of machinery NAFTA. The expansion phases rose from 11 to 27 quarters, and electronics nearly tripled, from 13.0 to 36.9 percent. Mex- 2. Growth Figure 26. Mexico’s business cycles pre- and post-NAFTA 20 10 0 -10 1980q1-1993q4 1994q1-2017q3 Duration, expansion (in quarters) Cumulative growth, expansion (%) Duration, recession (in quarters) Cumulative growth, recession (%) Source: Calculations of the Organisation for Economic Co-operation and Development and World Bank. Figure 27. Mexico’s terms of trade compared with Latin American peers (1998=100) 200 180 160 140 120 100 80 60 40 Mexico brazil Argentina Colombia Peru Chile Source: Data of WDI (World Development Indicators) (database), World Bank, Washington, DC, http://data.worldbank.org/products/wdi. ico’s share of global trade in transport equipment rose from cific Partnership, together with 10 other countries.53 Mean- 0.4 percent in 1990 to 2.6 percent in 2016, while the share in while, negotiations are at an advanced stage between machinery and electronics increased from 0.8 to 3.3 percent Mexico and the European Union on an updated global over the same period. Today, Mexico ranks eighth globally in agreement, and with Brazil on the Agreement on Eco- exports of transport equipment (behind the United States, nomic Complementarity no. 53. Mexico has also pursued Germany, the United Kingdom, France, China, Canada, and It- closer economic integration with Chile, Colombia, and aly) and fifth in machinery and electronics (behind the United Peru through the Pacific Alliance, through which the vast States; China; Hong Kong SAR, China; and Germany). majority of traded goods are no longer subject to tariffs. 62. Efforts to diversify export markets through trade agree- 63. Trade agreements are one of the factors that have made ments have been under way. Mexico has an extensive net- Mexico an attractive market for FDI. Since 2005, Mexico work of 12 free trade agreements, covering 46 countries, has received, on average, US$28.7 billion in inward FDI a as well as another 32 investment agreements covering year, accounting for 3 percent of GDP and representing 12 33 countries. In the context of NAFTA renegotiations, percent of total investment. As a result, the country has SYSTEMATIC Mexico has continued to extend and upgrade its trade ranked among the global top 20 inward FDI recipients in COUNTRY agreements, signing on March 8, 2018, for example, the recent years.54 During this period, the United States has ac- DIAGNOSTIC Comprehensive and Progressive Agreement for Trans-Pa- counted for 46 percent of total FDI, followed by Spain (10 MEXICO 42 53 54 Australia, Brunei Darussalam, Canada, Chile, Japan, Malaysia, New Zealand, Peru, Singapore, and Vietnam. UNCTAD (2017). 2. Growth Figure 28. General government fiscal balances, % of GDP 4 2 0 -2 -4 -6 -8 2012 2013 2014 2015 2016 2017 Mexico, primary balance LAC, primary balance Emerging and developing, primary balance Mexico, fiscal balance LAC, fiscal balance Emerging and developing, fiscal balance Source: IMF, World Bank. Figure 29. Gross general government debt, % of GDP 65 60 55 50 45 40 35 2012 2013 2014 2015 2016 2017 Mexico LAC Merging and developing economies Source: IMF, World Bank. percent), Canada (8 percent), and Japan and Belgium (5 goods, such as electronics. This competition became more percent each). Manufacturing accounted for half of all FDI significant after China joined the World Trade Organization (of which a quarter went to transport equipment alone), in late 2001. The prevalence of manufactured exports also while the remainder was divided between financial ser- meant that Mexico’s terms of trade did not experience the vices (10 percent), mining (7 percent), retail and wholesale same sharp strengthening that large commodity exporters trade (7 percent), and others. Geographically, the largest such as Argentina, Brazil, and Chile experienced between shares of FDI went to Mexico City (19.0 percent), Nuevo 2003 and 2013 (Figure 27). However, the drop in oil prices León (8.8 percent), and Chihuahua (6.6 percent). caused a significant reduction in oil revenues from PEMEX on the order of 4.5 percent of GDP between 2013 and 2017. 64. The tail wind of the commodities super cycle (2003–13) was limited in Mexico, even though the oil price drop in 2014 65. The authorities reacted with prudent policies to the oil had significant impacts. Despite Mexico’s considerable wealth price impacts, enabling the country to withstand the shock in nonrenewable resources, the reduction in oil production, while still achieving economic growth. Oil price collapses which set in at a time of rising oil prices, meant that Mexico in the past had significant negative effects on output (and SYSTEMATIC did not benefit as much from the commodities super cycle the fiscal position). In this last episode, however, the 2013 COUNTRY as its South American peers. Moreover, Mexico now relies tax reform generated sufficient additional revenues to help DIAGNOSTIC increasingly on manufactured exports and trades mainly offset the drastic drop in oil revenues. The authorities also MEXICO with the United States, rather than with China. Instead of applied an expenditure rationalization program in 2015-16 43 benefiting from China’s demand for raw materials, Mexico to improve the fiscal stance, and after the initial shock, pub- has been competing with it in the export of manufactured lic debt started to stabilize and, more recently, to decline, 2. Growth Figure 30. Annual GPD growth, Mexico (%), 1982–2016 8 China joins WTO 2001 NAFTA 1994 6 CDMX Earthquake 1985 4 2 0 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 -2 -4 Mexican debt crisis 1982 Great recession 2008 -6 Peso crisis 1994/95 -8 Source: Data of WDI (World Development Indicators) (database), World Bank, Washington, DC, http://data.worldbank.org/products/wdi. leading the way among emerging and developing econ- erage growth since the early 1980s (Figure 30).55 In this pe- omies (Figure 28 and Figure 29). All of this occurred while riod, Mexico experienced several successive external and the economy continued to grow close to its average pace. domestic shocks that resulted in large swings in econom- Recent years showed that the economy is more resilient to ic output. The 1982 debt crisis was followed by the 1985 terms of trade shocks than before, and that macroeconom- earthquake in Mexico City, the financial crisis of 1994/95 ic policies have continued to be prudent safeguarding the that followed on the heels of NAFTA, the entry of China country’s economic stability and sustainability. into the World Trade Organization in 2001, the bursting of the U.S. stock market bubble in 2001–02, the Great Recession in 2008–09, which coincided with the spike in 2.2 Dynamics affecting drug-related crime and the outbreak of the A(H1N1) flu growth rates epidemic, and the collapse in oil prices in 2014 (Figure 30). All these episodes (except, to some extent, for the lat- 66. The convergence, growth, and wealth dynamics pre- ter) meant output downswings and in some cases, severe sented in this section raise the question of why growth recessions. Overall, growth volatility in Mexico historically rates have been moderate, despite the country’s signifi- seems relatively high compared with comparators re- cant reforms undertaken. At the outset, it is important to gionally and globally (Figure 31). Moreover, the country highlight that structural reforms, like the ones undertak- did not experience a single spell of growth, that is, over 5 en over the last years, take time to render the outcomes percent per annum, over a sustained period. expected. In the case of Mexico, they have started to pro- duce fruits and may generate higher growth rates over 2.2.2. Limited capital the medium term. This sub-section examines the past accumulation dynamics that have affected growth from an aggregate perspective (section 5.2 looks at micro-structural roots 68. Mexico’s capital accumulation does not seem to have of the growth performance). It focuses on the role of been sufficient to support higher growth rates. While the shocks and economic volatility, subdued capital invest- investment level is not acutely low relative to LAC (total ment growth, regional disparities and the lack of internal investment has averaged 20 percent of GDP since 1990), it convergence, and productivity disparities across regions, has been much lower than that in rapidly growing emerg- sectors, and firms. ing economies that are converging to higher income lev- els, for example: 33 percent in Korea and 28 percent in Ma- laysia. Moreover, annual growth rates in capital stock have 2.2.1. Business cycles and been below fast growing emerging and relevant compar- aggregate shocks ators over the last decades. The growth in capital stock 67. As in many countries in the region, growth in Mexico began to slowdown in the early 1980s, coinciding with the has been impacted by a series of negative shocks. Mexico’s slump in GDP growth (Figure 33). Evidence suggests that SYSTEMATIC COUNTRY economy has experienced greater volatility and lower av- the collapse in capital accumulation was not primarily a DIAGNOSTIC MEXICO 55 Between 1950 and 1981, Mexico’s economy expanded at 6.6 percent annually. Rapid industrialization led to the migration of labor away from rural agriculture into the cities, and the 44 expansion of education contributed to significant improvements in human capital. However, the inward-looking nature of the development model, supported by import substitution, began to reach its limits in the late 1970s. By 1981, the economic imbalances had resulted in high inflation, fiscal deficits, and unsustainable increases in public debt, culminating in the debt crisis of 1982 (for instance, see Esquivel and Hernández-Trillo 2009; Kehoe and Meza 2011). 2. Growth Figure 31. Growth volatility in Mexico and selected comparators 2.5 2.0 Coeff. of variation 1.5 1.0 0.5 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Mexico Advanced Economies Emerging Market and Developing Economies Latin America Brazil Colombia Chile South Africa Source: World Bank calculations based on data of WEO (World Economic Outlook Database), IMF, Washington, DC, https://www.imf.org/external/pubs/ft/weo/2016/01/weodata/index. aspx. Note: Coefficient of variation is defined as the ratio of the standard deviation to the average growth rate. Figure 32. Average annual public investment in infrastructure, 2008-2015 (% of GDP) 10 9.0 8 7.2 6 5.1 3.9 4 3.2 3.3 2 0 Mexico Argentina Brazil Chile Colombia Peru Source: Infralatam (Economic Infrastructure Investment Data of WDI (World Development Indicators, Latin America and the Caribeean) (database), World Economic Comission for Latin America and the Caribbean, the Development Bank of Latin America, and the IADB, Washington, DC, http://data.worldbank.org/products/wdi.http://infralatam.info/. result of a low savings rate, but rather of the increase in the short of those fast-growing Latin American and emerging capital-output ratio following the debt crisis of the early economies that spend above 5 percent of GDP in this area. 1980s.56 70. Insufficient investment has resulted in infrastructure bottle- 69. Public investment, and public spending in infrastruc- necks. Even though transport infrastructure is better in Mexico ture, has been particularly low compared to fast growing than in many other Latin American countries, it is aging, and economies. Mexico’s public investment in infrastructure new investment in the sector has trailed that of regional peers only reached an average of 3.2 percent of GDP between (Figure 34).58 Moreover, the transport, logistics, and facilitation 2008 and 2015 (Figure 32).57 In fact, public investment services to support export markets other than the United in infrastructure fell to 2.6 percent of GDP in 2017 in the States, such as Asian markets through export corridors toward context of the needed and commended fiscal consolation the Pacific Ocean ports, are relatively weak. It also ranks in the process of recent years. Excluding PEMEX, that number middle of the pack in terms of the quality of railroad, port, and would be 1.7 percent of GDP for 2017. These levels fall airport infrastructure, which all require significant investments SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 56 Bacha and Bonelli (2015). 45 57 Infralatam (Economic Infrastructure Investment Data, Latin America and the Caribbean) (database), Economic Commission for Latin America and the Caribbean, the Development Bank of Latin America, and the IADB, Washington, DC, http://infralatam.info/. Includes PEMEX. 58 Infrastructure pillar (2017/18), GCI (Global Competitiveness Index) (database), World Economic Forum, Geneva, http://reports.weforum.org/global-competitiveness-index/. 2. Growth Figure 33. Growth in capital stock and GDP (1961-2015) 15 10 5 0 -5 -10 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 capital stock GDP Source: IMF Investment and Capital Stock Dataset, 2017. Figure 34. Investment in the transport sector (public + private), % of GDP 12 10 8 6 4 2 0 2008 2009 2010 2011 2012 2013 2014 2015 Argentina Mexico brazil Chile Peru Colombia Source: Infralatam (Economic Infrastructure Investment Data, Latin America and the Caribbean) (database) as Mexico attempts to diversify trade markets and support private investment across four key sectors (water, roads, growth across sectors.59 Overall, the country invests barely energy, and telecommunication) accounted, on aver- one-quarter of what is needed in the sector, opening a wide age, for one-third of total investment in these sectors.61 transport infrastructure gap.60 The country also continues to Public-private partnerships, in particular, have grown in have important deficiencies in electricity (particularly in trans- importance. Once largely focused on toll roads, these in- mission capacity and distribution) and telecommunication. vestments have become more diverse. Since the energy These sectors would benefit from investments through the reform in 2013 and 2014 and the telecommunications full implementation of recent structural reforms. Similar short- reform in 2013, electricity and natural gas projects have comings may be observed in the water sector infrastructure. become increasingly important, while 2017 saw the fi- These gaps suggest that an increased role is needed for the nancing of the Red Compartida Project to develop back- public and private sectors in infrastructure financing. bone telecommunication infrastructure.62 Since 1990, 296 projects have been undertaken as public-private 71. The private sector plays an increasingly important partnerships, amounting to US$83 billion, half of them role in investment, including in infrastructure. Since 2008, since 2006.63 SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 59 Ports on the Pacific side, for example, would be critical for export markets in the Asia-Pacific region, but would need significant enhancements to handle large volumes. 60 Global Infrastructure Outlook (database), Global Infrastructure Hub, Sydney, Australia, https://outlook.gihub.org/. 61 Infralatam (Economic Infrastructure Investment Data, Latin America and the Caribbean) (database), Economic Commission for Latin America and the Caribbean, the Development 46 Bank of Latin America, and the Inter-American Development Bank, Washington, DC, http://infralatam.info/. 62 In addition, the full impact on investment of the 2013-14 reforms will only materialize in the medium term, raising the prospect of further increases in private sector investment. 63 PPI Project Database (Private Participation in Infrastructure), World Bank, Washington, DC, http://ppi.worldbank.org/. 2. Growth Figure 35. Regional convergence (unconditional) in GDP per capita a. 1980-94 Camp b. 1994-2016 6.0 5.0 Tab Camp 5.0 average growth rate, 1980-2016 Chih average growth rate, 1980-1994 3.0 Qr Ags 4.0 SLP CDMX Mor Sin Dgo 1.0 Zac Qro Son Coah NL 3.0 Zac Oax Gro Yuc Gto Qro Pue Col Tamps bC SLP 0.0 Mich Mich Nay Son Tlax 11.0 Nay Jal bCS 2.0 hgo ver Sin Jal 10.5 11.5 12.0 12.5 Oax Ags CDMX -1.0 hgo ver EDOMEX Tlax Yuc Col Dgo bCS NL 1.0 Gro Pue Coah Chis Mor Tamps -3.0 0.0 bC Qr 10.5 11.0 11.5 12.0 12.5 Chis EDOMEX Chih -1.0 -5.0 log GDP per capita 1980 log GDP per capita 1996 Source: World Bank calculations based on data of the National Institute of Statistics and Geography. Map 1. Average annual growth rate, by region (%), 1980–2016 3.3 3.3 2.8 4.2 3.2 2.6 2.3 2.8 2.4 3.4 2.1 3.1 3.4 2.5 2.6 6.3 2.5 1.8 8.2 2.5 1.9 1.7 1.9 2.6 1.1 Source: World Bank calculations based on data of the National Institute of Statistics and Geography. ing slightly higher GDP per capita growth between 1980 2.2.3. Significant regional growth and 1994, the convergence since then has been very limit- disparities and limited regional ed. The GDP per capita of one of the most rapidly growing convergence and richest states (Querétaro) grew an average 2.5 percent 72. Average growth rates in Mexico mask, and are to some a year in 1996–2016, compared with 0.8 percent a year in extent dragged down, by significant regional income and Chiapas. Indeed, tests of unconditional convergence in growth disparities. There are large differences between GDP per capita suggest that the rhythm of convergence the industrialized north and center-north of Mexico (per has slowed and that regions in the north, center-north, capita GDP of MXN181,100 or US$9,800) and the less and center are benefiting the most from Mexico’s econom- well developed south (per capita GDP of MXN67,800 or ic transformation (Figure 35, Map 1). These more well-de- US$3,300). In 2016, the average GDP per capita of one of veloped regions are home to industries that have been the richest states (Nuevo León) was close to the average of able to take advantage of the market opportunities that Poland, while that of the poorest state (Chiapas) was simi- NAFTA created, including in automobile and associated lar to that of Honduras or Timor-Leste. manufacturing, export-oriented agriculture (high-value fruits and vegetables), machinery, and electronics. How- SYSTEMATIC 73. There has been almost no domestic regional conver- COUNTRY ever, these high-growth industries and states seem not to DIAGNOSTIC gence over the last 20 years. While there was some conver- have built backward links to other parts of the economy MEXICO gence before the mid-1990s, with poorer states experienc- and other states. 47 2. Growth Figure 36. Growth accounting, Mexico and comparator countries (% contribution), 1991–2016 6 5 4 3 2 1 0 -1 -2 MEX ROU THA TUR MYS KOR POL CHL GDP growth Labor quantity Labor quality Capital TFP Source: November 2017 data of Total Economy Database, Conference Board, New York, https://www.conference-board.org/data/economydatabase/. Figure 37. Growth accounting, Mexico (% contribution) 6 4 2 0 -2 1991-1994 1995-2003 2004-2012 2013-2016 Capital Labor quantity Labor quality intermediate inputs TFP GDP Source: World Bank calculations based on data of INEGI; World KLEMS Data (database), World KLEMS Consortium, University of Groningen, Groningen, the Netherlands, http://www. worldklems.net/data.htm. ilarly, if TFP growth had been similar to that of the Unit- 2.2.4. Limited productivity growth ed States, Mexico could reduce the GDP per worker gap and significant dispersion across with the United States by 22 percent. However, capital per regions, sectors, and firms worker would still be far too low to reach the same lev- 74. The absence of aggregate productivity growth is one el, suggesting that improvements in productivity would of the factors explaining Mexico’s lack of convergence. The need to be accompanied by more rapid capital accumu- contribution of total factor productivity (TFP) to growth lation (Figure 38). was negative (−1 percent) between 1991 and 2016, rep- resenting the weakest performance among Mexico’s 75. Even though the supply of educated labor has in- structural and aspirational peers (Figure 36). TFP growth creased, labor quality remains insufficient. The share of was marginally positive in the early 1990s, but negative workers with a high school education rose from 35.2 from 1995 to 2016 (Figure 37). Falling TFP has partially percent in 1994 to 55.5 percent in 2014, and the share offset the modest gains from factor accumulation. The of college educated labor doubled from 12.5 percent to contribution of labor has been driven mostly by quantity 24.3 percent. However, given the increased importance SYSTEMATIC (a growing labor force) rather than labor quality, which has of high-complexity sectors in the economy (which tend COUNTRY DIAGNOSTIC improved only moderately. A simple calculation suggests to require more technical skills), there is evidence that MEXICO that, if Mexico’s productivity had grown at a pace similar employers face difficulty filling jobs. In a survey by the to the pace during the high-growth period between 1950 Manpower Group, 54 percent of employers stated they 48 and 1970 (1.3 percent a year), the country’s GDP per capita had difficulty filling vacancies. Similarly, 31 percent of re- would be 128 percent higher than the current level. Sim- spondents to the most recent Enterprise Survey in Mexico 2. Growth Figure 38. Gap between United States GDP per worker and GDP per worker in selected Latin America and Caribbean countries 120,000 uSA GDP oer worker 100,000 37% 80,000 57% 51% 44% 66% USD 60,000 20% 22% 40,000 24% 20% 13% 20,000 0 GDP per GDP per GDP per GDP per GDP per GDP per GDP per GDP per GDP per GDP per worker with worker worker with worker worker with worker worker with worker worker with worker USA TFP USA TFP USA TFP USA TFP USA TFP Chile Colombia Costa rica Mexico Peru Source: World Bank calculations using data of Penn World Table (database version 9.0), Groningen Growth and Development Centre, Faculty of Economics and Business, University of Groningen, Groningen, the Netherlands, https://www.rug.nl/ggdc/productivity/pwt/. Figure 39. Skilled workforce, Mexico and selected countries, 2010 90 79.4 80 74.4 68.0 70.9 70 60 Percentage 50 40 30.9 31.6 30 21.2 20 17.3 10 0 Proportion of skilled workers out % of firms identifying an inadequately of all production workers* educated workforce high income OECD All countries Mexico Latin America & Caribbean Source: 2010 data of the Enterprise Surveys (database), International Finance Corporation and World Bank, Washington, DC, http://www.enterprisesurveys.org. Note: This indicator is computed using data only on manufacturing firms. Regional and all-country averages among the indicators are computed by taking a simple average of country point estimates. For each economy, only survey data of the latest available year are used in the computation. (2010) identified an inadequately educated workforce as Querétaro, and Zacatecas rose at an average annual rate a major constraint (Figure 39). Only 68 percent of all pro- of over 1 percent between 1993 and 2015, while labor duction workers are skilled, compared with 79 percent in productivity in Chiapas, Hidalgo, Oaxaca, Quintana Roo, countries of the Organization for Economic Co-operation and Tlaxcala declined. As a result, the divergence in pro- and Development (OECD). Labor quality – a function of ductivity growth has become more pronounced in the education and experience – continues to be low, and its last two decades. Today, the differences are significant: contribution to growth has been minimal. Improving the measured by value added per worker, productivity is five quality of labor is particularly important considering that times greater in Mexico City than in Chiapas (Figure 40). the share of the 18- to 64-year old population is projected If labor productivity in states at the bottom had grown at to begin declining in 2027. the same rate as in Aguascalientes (one of the top per- SYSTEMATIC formers), their current GDP per capita would be 81 per- COUNTRY 76. There is substantial dispersion in labor productivity cent higher. DIAGNOSTIC growth across states. States that have experienced high- MEXICO er overall GDP growth rates since 1993 also experienced 77. High levels of informality persist, especially in some 49 improvements in labor productivity. Labor productivity regions, and contribute to low productivity. At the national in the states of Aguascalientes, Chihuahua, Nuevo León, level, 56.9 percent of total employment was in the infor- 2. Growth Figure 40. Labor productivity by state (Mexico City = 100) Ciudad de México Nuevo León Querétaro Coahuila de Zaragoza Sonora Aguascalientes Baja California Sur Quintana Roo Chihuahua Baja California Jalisco Tamaulipas San Luis Potosí Colima Sinaloa Guanajuato Durango Veracruz de Ignacio de la Llave Zacatecas Morelos Yucatán Hidalgo Nayarit Michoacán de Ocampo México Puebla Tlaxcala Guerrero Oaxaca Chiapas 0 20 40 60 80 100 Source: World Bank calculations based on the economic census data of the National Institute of Statistics and Geography. Note: Labor productivity is defined as value added per worker. mal economy in 2017.64 By some estimates, up to 97 per- because, as the shift-share analysis shows, the strong con- cent of firms in manufacturing and services rely partially tribution of shifting labor resources to sectors with higher or wholly on informal labor and absorb 80 percent of labor levels of productivity (+58.1 percent) is offset to a large and 70 percent of capital.65 This large concentration of degree by the impact (−28.9 percent) of shifts to sectors in factors in the informal sector contributes to low aggregate which productivity growth is slower or even negative.68 A productivity because there is ample evidence that formal prime example is the oil and gas extraction sector, which, firms tend to be more productive than informal ones.66 In though it is the sector with the highest level of productivi- Mexico, the difference is not trivial: estimates suggest that ty in value added per hour worked, saw productivity fall by formal Mexican firms are an average of 84 percent more more than half over 1990–2016. At the same time, the sec- productive than informal firms.67 There is, however, great tor experienced a modest 0.2 percentage point increase regional variation: states with lower labor productivity in its share of hours worked across the economy. In aggre- also record the highest informality rates, reaching up to gate, the manufacturing sector experienced a significant 80 percent (Figure 41). The prevalence of micro, small, and decline in labor share (−4.4 percentage points), even as its medium enterprises is also associated with higher infor- average annual labor productivity growth (+1.4 percent) mality rates in states (Figure 42). outpaced that of the economy as a whole. The transport equipment subsector is an important exception, expe- 78. Structural transformation accounts for only one-third of riencing an increase (from 1.4 to 2.4 percent) in its labor the growth in labor productivity. The correlation between share, while also experiencing annual average growth in labor productivity and the share of hours worked was pos- labor productivity amounting to 1.9 percent, bringing its itive but weak between 1990 and 2011 (Figure 43). This is labor productivity above the economy-wide average. The SYSTEMATIC COUNTRY 64 Data of the National Institute of Statistics and Geography. DIAGNOSTIC 65 Busso, Fazio, and Levy (2012). Estimates based on 2008 data. MEXICO 66 See La Porta and Shleifer (2008, 2014); Fajnzylber et al. (2011). 67 Busso, Fazio, and Levy (2012). 68 At the 3-digit North American Industry Classification System–level of disaggregation. By comparing the sectoral allocation of both hours worked and value added in 1990 and in 2016, a shift-share analysis disaggregates the impact on economy-wide value added per hour worked of (a) intersectoral (between-sector) changes and (b) intrasectoral (within-sector) 50 changes. The intersectoral element captures the role of structural transformation and can be disaggregated into static and dynamic effects. The static effect results from shifting labor resources between sectors with different levels of productivity, while the dynamic effect results from the shifting of labor resources between sectors with different rates of productivity growth. 2. Growth Figure 41. Labor productivity vs. informality, states 13.8 CDMX 13.3 NL COAH QRO TAB ln(VA/worker) 2016 SON AGS QROO 12.8 JAL BCS SLP CHIH BC TAMPS COL VER SIN GTO ZAC DUR YUC HGO 12.3 MOR MICH MEX NAY PUE GRO TLAX CHIS OAX 11.8 35 45 55 65 75 85 informality rate 2017 (%) Source: World Bank calculations based on data of the National Institute of Statistics and Geography. Figure 42. Economic units with more than 10 employees, share of total (%) 10 9.6 8 6 4.2 4 2 0 Six states with highest rate of formality Six states with lowest rate of formality Source: IMCO 2015. services sector saw its share of hours worked increase by 80. Indeed, a core source of factor misallocation seems 4.5 percentage points (from 55.7 percent in 1990 to 60.2 to occur among firms within the same sector. Comparing percent in 2016), while its relatively strong labor produc- the distribution of productivity across sectors with the tivity growth saw productivity levels move farther ahead distribution of productivity across firms within each of the economy-wide average. sector reveals that firm-level productivity is significant- ly more dispersed and has a larger standard deviation, 79. Large productivity dispersion signals factor misallocation with most firms below the sectoral average and a fat in the economy. Productivity differences between sectors left tail of unproductive firms. These productivity dif- are large, for example, the distribution of TFP across sectors ferences persist even within narrowly defined sectors, shows that productive sectors (the top 5 percent) are up such as cut-and-sewn apparel manufacturing, where the to 500 percent more productive than the least productive most-productive firms are about 8 times more produc- ones. In recent years, as labor productivity in the manu- tive than the least-productive firms. The productivity facturing sector declined, its share of total labor fell. At the gap between the most and least productive firms in the same time, productivity also decreased in the services and same sector is considerably larger in Mexico than in the SYSTEMATIC commerce sectors, yet their share in total labor increased. United States. Within the same sector (at 4 digits), pro- COUNTRY But reallocation between sectors alone does not explain the ductive firms (the top 5 percent) are up to 1,200 percent DIAGNOSTIC sluggish productivity trend (e.g., maintaining sectoral labor more productive than the least productive ones. There MEXICO shares constant at 1990 level would have only marginally is also considerable heterogeneity across firms: most 51 affected productivity growth). Productivity dispersion be- have a below-average level of TFP (Figure 44). Despite tween firms within sectors, on the other hand, is large. these large differences, labor does not move from less 2. Growth Figure 43. Change in employment share vs. labor productivity, by sector 1.2 1.0 Oil, Gas & Mining Real estate 0.8 Log Sectoral Productivity / Average Productivity (2016) Financial services 0.6 Manufacturing: transport equipment Utilities 0.4 Manufacturing: Business services ICT equipment 0.2 Education 0.0 Manufacturing: other -0.2 Hotels & Restaurants Retail & Wholesale -0.4 Construction Agriculture -0.6 Other services -0.8 -8% -6% -4% -2% 0 2% 4% 6% Change in Employment Share 1990-2016 Source: World Bank calculations based on data of the National Institute of Statistics and Geography. Note: Labor productivity is defined as value added per hour worked. productive firms to more productive ones. One way in south, where traditional production practices continue which this is manifested is that firms do not grow: firms to dominate; productivity is low; land fragmentation is that have been in business for 10 years or more employ common, and market integration is limited because of only marginally more labor than those in business for scale and the terrain. 5 years. High-growth firms tend to be concentrated in few sectors that have a high share of value added, FDI 81. At the same time, high levels of informality persist. At penetration and trade volume, and are more prevalent the national level, 56.5 percent of total employment was in the north of the country.69 According to Mexico’s 2014 in the informal economy in 2017, although there is sig- Economic Census, micro, small, and medium enterprises nificant variation across states.70 By some estimates, up account for 99.8 percent of the total number of firms to 97 percent of firms in the manufacturing and services in Mexico and absorbed 76.4 percent of employment. sector rely partially or wholly on informal labor and absorb However, they only contributed 31.5 percent of total val- 80 percent of labor and 70 percent of capital.71 This large ue added. Given the importance of regional value chains concentration of factors in informality contributes to low and of the geographical proximity to the United States, aggregate productivity. Estimates suggest that Mexican within-sector dispersion also corresponds, to a large formal firms are on average 84 percent more productive extent, to between-state dispersion: in the agricultural than informal firms.72 sector, for example, the north and center of the country are home to modern agrifood systems, many of which 82. The shift-share analysis shows that more than two- are global leaders in terms of scale, level of technolog- thirds of the limited labor productivity gains have come ical sophistication, productivity, integration into mar- about via within-firm dynamics. In the manufacturing SYSTEMATIC kets, and value added. This is in stark contrast with the sector, the within-firm component represents about COUNTRY DIAGNOSTIC MEXICO 69 Background Paper for the Flagship on “High Growth Enterprises in Developing Countries” (World Bank, forthcoming). 52 70 Data of the National Institute of Statistics and Geography (INEGI). 71 Busso, Fazio, and Levy (2012). Estimates based on 2008 data. 72 Busso, Fazio, and Levy (2012). 2. Growth Figure 44. Productivity dispersion in the manufacturing sector by firm size 0.30 Distance between NAiCS 4-digit median 0.25 and national median Proportion of establishments 0.20 0.15 0.10 0.05 0.00 0.0001 0.001 0.01 0.1 1 10 100 Productivity relative to manufactring average All firms [0-10] [11-100] [101-500] [+500] uSA Distribution1997 Source: World Bank calculations based on data of the National Institute of Statistics and Geography; Pages 2010. Note: NAICS = North American Industry Classification System. Figure 45. Mexico’s manufacturing management score distribution vs. the United States 0.80 0.60 Management score 0.40 0.20 0.00 p10 p25 p50 p75 p90 Mexico uS Source: Bloom et al. forthcoming. 80 percent of productivity growth. However, few firms Mexico also underperforms compared with peers in in- are productive at the global level: the domestic fron- novation: the global innovation index, which measures tier (computed based on the top quartile in the distri- the capacity of countries for and their performance in bution of value added per employee) is far below the innovation, puts Mexico below what is expected based global frontier in all industries, and there has been little on its level of economic development and also indicates improvement in the past decade.73 An important factor that this performance is produced in a relatively ineffi- is product and process innovation, including quality im- cient way (Figure 46). provement and management quality. Low management quality explains about one-third of the cross-country 83. Private sector research and development expenditure difference in TFP.74 In the manufacturing sector, Mexican is well below that of most OECD countries as well as Bra- firms in the 90th percentile on management scores have zil, China, India, the Russian Federation, and South Africa. roughly the same management quality as the U.S. medi- Innovative firms are more likely than noninnovative an (Figure 45). However, Mexican managers consistently firms to participate in international markets.75 In the SYSTEMATIC COUNTRY overestimate the quality of management in their firms. case of Mexico, evidence supports the relation between DIAGNOSTIC MEXICO 53 73 Araujo et al. (2016). 74 Bloom et al (2016). 75 OECD (2008, 2015). 2. Growth Figure 46. Global Innovation Index 70 Switzerland 65 uK 60 uSA South Korea Singapore China Germany 55 Finland 50 Global Innovation 45 Malaysia Index Score 2017 romania 40 Poland Thailand Turkey Colombia Mexico Chile 35 uruguay Peru brazil 30 Argentina 25 10,000 100,000 GDP per capita 2016 , PPP 2011 (logarithmic scale) Source: Dutta, Lanvin, and Wunsch-Vincent 2017. Figure 47. Urban population in 2018 (% total) 100% 80% 60% 40% 20% 0% Uruguay Argentina Chile Costa Rica Venezuela Brazil Mexico Colombia Bolivia Ecuador-2000 Colombia-1977 Peru El Salvador Panama Ecuador Bolivia-1996 Paaraguay Honduras Guatemala Sources: UN, Dept. of Econ. and Social Affairs, Population Division (2018). Figure 48. Correlation between city size and productivity, Mexico 0.4 Puerto vallarta 0.3 Ensenada Tijuana Mexicali La Paz benito Juárez Monterrey 0.2 Guadalajara León Productivity differential 0.1 Quéretaro 0.0 San Luis Potosí Mexico City Toluca Puebla -0.1 Mérida -0.2 Tuxtla Gutiérrez Tlaxacala Xalapa -0.3 Córdoba San Cristobal de las Casas -0.4 Tapachula SYSTEMATIC 50,000 200,000 800,000 3,200,000 12,800,00 COUNTRY DIAGNOSTIC Number of inhabitants MEXICO Source: Ahrend et al. 2014. 54 innovation and spending in research and development 84. Although urbanization has been key for economic and pro- 2. Growth in industry. The relatively low level of private research ductivity growth, its pace has slowed in recent years. Mexico and development is partly a result of Mexico’s industrial is at an advanced stage of urbanization; 80.2 percent of the structure because more than one-third of manufacturing population was living in urban areas in 2018 (Figure 47). Ur- research and development is carried out in low- and me- banization and cities play an important role in economic and dium-technology sectors. However, obstacles to boost- productivity growth: 87 percent of Mexico’s gross value add- ing the country’s innovative potential include a weak ed is produced in cities with a population of over 100,000 in- domestic research and skills base, an underdeveloped habitants. The results show a clear north-south divide. There knowledge-based start-up environment and institution- are negative outliers in the south and positive outliers in the al challenges. north, close to the border with the United States (Figure 48). SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 55 3. Inclusion 3. Inclusion 85. Progress has been made in recent years in social and from 11.3 percent in 2010 to 7.6 in 2016.79 Yet, progress in productive inclusion in Mexico, but there is significant room reducing monetary poverty overall has been limited. The to improve. The diagnostic of inclusion in this report is total share of the population living in poverty in 2014 was based on the key notion that equity is a necessary condi- around 53 percent (the same level as in 1992). Approxi- tion to sustaining a robust process of inclusive growth.76 mately two Mexicans in five were still considered poor in Social inclusion is understood for the purposes of this re- 2016. In contrast, peer countries have all experienced pov- port as the combination of intragenerational mobility and erty reduction. Most have cut poverty by half in the last 10 voice (freedom and influence in the social, economic, and years. Nonmonetary measures of poverty have, however, political domains). Productive inclusion is framed in this consistently improved in Mexico since they were first es- diagnostic as equal opportunities to access markets and tablished in 2010 (Table 2). The share of the population contribute to economic activities based on the four main with one or more social deprivations decreased from 74.2 productive assets available to households: human capital, percent in 2010 to 70.4 percent in 2016, while the share of financial capital, natural capital (such as land, soil, forest, the population experiencing three or more social depriva- and water), and social capital.77 tions had declined from 28.2 percent to 18.7 percent by 2016. Lack of access to social security is the most prevalent deprivation, at 55.8 percent, followed by lack of access to 3.1 Social inclusion dynamics food at 20.1 percent in 2016. Lack of access to health care services improved by an expansion of a noncontributory 86. Mexico has experienced improvements in nonmonetary health insurance program, Seguro Popular. This was the measures of poverty in recent years, but progress in mon- dimension of nonmonetary measures of poverty showing etary poverty reduction and shared prosperity has been the largest decline between 2010 and 2016. slower. The official multidimensional poverty rate, which combines income poverty with six indicators of social 87. Poverty incidence is higher among indigenous popula- deprivation, has declined in recent years, from 46.2 per- tions, households with dependents, and households with cent in 2014 to 43.6 percent in 2016, after being stagnant a majority of women among adult members. In 2016, the since 2010.78 The official extreme poverty rate declined extreme poverty rate was six times higher among indig- SYSTEMATIC COUNTRY DIAGNOSTIC 76 Limited poverty reduction, coupled with relatively high and persistent inequality, usually contributes to low economic growth. See, for instance, Perry et al. (2006). Also, while eco- MEXICO nomic growth is empirically associated with poverty reduction, equity-driven policies enhance the capacity of all groups in society to contribute actively to the growth process, making growth more inclusive and sustainable (Dollar et al. 2016). Growth and its distribution are jointly determined processes (Ferreira 2010). 77 Social capital refers to the preferential treatment and social cooperation among individuals and groups that can contribute to the economic gains of these individuals and groups. 57 78 The official multidimensional poverty rate declined from 46.1 percent in 2010 to 45.5 percent in 2012, then increased to 46.2 percent in 2014. Annex 2 describes the official methodol- ogy and data sources for poverty measurement in Mexico. 79 Data from Consejo Nacional de Evaluación de la Política de Desarrollo Social (National Council for the Evaluation of Social Development Policy, CONEVAL) using ENIGH 2016. 3. Inclusion Table 2. Evolution of the social components, multidimensional poverty index, Mexico, 2010–16 Components 2010 2012 2014 2016 Gaps in educational attainment 20.7 19.2 18.7 17.4 Lack of access to health care services 29.2 21.5 18.2 15.5 Lack of access to social security 60.7 61.2 58.5 55.8 Low quality and space in the dwelling 15.2 13.6 12.3 12.0 Lack of access to basic services in the dwelling 22.9 21.2 21.2 19.3 Lack of access to food 24.8 23.3 23.4 20.1 Sources: Data of CONEVAL; INEGI 2016a. Figure 49. Poverty among indigenous and nonindigenous population groups (%), 2010–16 100 100 79.5 78.4 77.6 80 76.8 80 60 34.8 60 44.7 38.0 39.9 Percentage Percentage 43.3 43.0 43.6 41.0 40 7.4 40 8.9 7.9 5.8 20 38.8 38.5 42.8 20 34.8 34.4 35.1 36.2 35.2 0 0 2010 2012 2014 2016 2010 2012 2014 2016 Indigenous Non-indigenous Poor Extreme poor Sources: Data of CONEVAL; INEGI 2016a. Note: The indigenous population is classified based on mother tongue. Figure 50. Poverty, by place of residence (%), 2010–16 100 100 90 90 80 80 70 64.9 70 61.6 61.1 60 58.2 60 26.5 21.5 20.6 Percetange 50 17.4 50 Percetange 40.4 40.6 41.7 39.2 40 6.2 40 6.7 6.3 4.7 30 30 20 38.5 40.1 40.5 40.8 20 33.7 34.3 35.4 34.4 10 10 0 0 2010 2012 2014 2016 2010 2012 2014 2016 Rural Urban Poor Extreme poor Sources: Data of CONEVAL; INEGI 2016a. Note: Rural localities are defined as those with a population below 2,500. SYSTEMATIC COUNTRY DIAGNOSTIC enous populations than among nonindigenous popula- 48 percent of the population living in households with MEXICO tions. As of 2016, three indigenous people in four were dependents are poor; the corresponding share among poor based on the official multidimensional poverty rate, households without dependents is 25 percent. In house- 58 while about 40 percent of the nonindigenous population holds with an equal or a larger share of adult women in were living in conditions of poverty (Figure 49). About the household, the poverty headcount is slightly larger 3. Inclusion Figure 51. Population shares experiencing social deprivations, by area of residence and ethnicity, 2016 100 100 90 90 79.9 80 77.1 80 70 70 64.6 60 60 53.1 50 50 43.2 54.0 40 49.4 40 33.0 31.5 29.1 30 24.7 30 16.2 21.4 20 20 15.5 10 18.7 10 16.7 16.3 19.3 13.9 13.2 13.3 9.2 9.1 10.4 0 0 Gaps in educational attainment Gaps in educational attainment Access to health care services Access to social space in the Access to Access to food Access to health care Access to space in the Access to Access to food security Quality and basic services services social basic services dwelling in the dwelling security Quality and dwelling in the dwelling rural urban indigenous Non-indigenous Source: World Bank calculations using data of CONEVAL; INEGI 2016a. Note: Rural localities are defined as those with a population below 2,500. The indigenous population is classified based on mother tongue. than the national average, while, in households with which are located in southern Mexico, 25 percent of total more adult men, the poverty incidence is 6 percentage employment is in agriculture, where access to inputs such points lower. as irrigation and formal credit is limited. 88. Although most of the poor live in urban areas, there is 90. Particularly in southern Mexico, rural areas suffer a vi- a higher incidence of poverty and extreme poverty in rural cious cycle of low productivity, low investments in physical areas. In 2016, 58.2 percent and 17.4 percent of the pop- and human capital, and high poverty rates. Two decades ulation in rural areas were living in poverty or extreme after the 1994 land reforms, about half of the land in Mex- poverty, respectively, in contrast to 39.2 percent and 4.7 ico is under some form of communal or ejido ownership. percent in urban areas, although there has been some Furthermore, small land-holding -those with fewer than progress in poverty rates in rural areas over recent years 5 hectares of land- dominate Mexico’s agricultural sec- (Figure 50). Moreover, the disparities in access to social tor, owning or managing more than 75 percent of rural rights between urban and rural population groups are property. Typically, these are traditional or subsistence large, and they are even wider between indigenous and farmers with limited access to improved seeds, irrigation, nonindigenous groups (Figure 51). In particular, the share formal credit, better training or marketing infrastructure. of the population without access to basic services in their Southern states with the highest rate of poverty have the dwelling is almost 50 percentage points higher among highest levels of ejidos and incidence of small plot size. In indigenous populations and 44 percentage points higher addition, this region hosts a majority of the indigenous in rural areas compared with nonindigenous and urban people, and is primarily rural. Land fragmentation and areas, respectively. land titling correlate with the proportion of subsistence agriculture and of indigenous population. With a large 89. Poverty in rural areas decreased, but they continue to proportion of smallholders involved in producing low val- host most of the extreme poor, who tend to be engaged in ue-added crops, Mexico’s rural poverty pockets risk being low-productive primary activities. The poverty rate in rural poverty traps. areas decreased from 64.9 percent in 2010 to 58.2 percent in 2016, while the extreme poverty rate declined from 26.5 91. Female-headed households are poorer and face high- percent to 17.4 percent. Yet, 31 percent of the poor and er risks of falling into a poverty trap.80 About 25 percent 53 percent of the extreme poor live in rural areas. In 2016, of all households in Mexico are headed by females (this 62 percent of the household heads among the bottom 40 percentage is higher than in other countries due to migra- percent of the distribution in rural areas were employed in tion, among other factors).81 Female-headed households primary activities. Primary activities exhibit low productiv- are also less likely than male-headed households to have ity. Although the primary sector employs around 13 per- access to energy, infrastructure, formal credit, and land.82 cent of the total working population nationwide, it only Twenty-three percent of Mexico’s women live in rural areas SYSTEMATIC produced 3.3 percent of GDP in 2017. In the poorest states, and in vulnerable conditions with limited access to human COUNTRY DIAGNOSTIC MEXICO 59 80 World Bank (forthcoming) 81 INEGI (2008). 82 INEGI (2008). 3. Inclusion Map 2. Changes in the proportion of households with access to drinking water, 2005-15 >95% 90-95% <85% 85-90% >95% 90-95% <85% 85-90% Source: Data of CONEVAL-INEGI: Social Gap Index 2005-2015; World Bank calculations endowments such as education, health, infrastructure, south regions, and by income quintiles. In 2014, 86.3 per- and employment. In 2015, nearly half of all Mexican wom- cent of the richest households had access to sewerage, en were living in households with a per capita income while only 33 percent of the poorest households did. The below the poverty line, a poverty rate above the Mexican limited resources in the sector are raised through tariffs, average in rural areas. Women are also 6 percent more limiting investments in areas with lower demand. Invest- likely to fall into poverty traps, have reduced educational ments rather depend on government’s subsidies. Mexi- levels, and suffer food insecurity if they are also heading cans pay relatively little for water services, compared with the household. the international standard of 4-5 percent of household’s income spent on water. Due to subsidized tariffs, water 92. Notwithstanding the positive effects of urbanization, payments in urban areas range from 0.25 to 2.19 percent most of Mexico’s poor live in urban areas with challenges in of household’s income. In rural areas, the poorest are pay- the provision of services. Mexico is at an advanced stage ing a higher proportion of their income for water services. of urbanization, with 80.2 percent of its population living Yet, basic water services are essential for human capital in urban areas in 2018,83 and 87 percent of its gross value accumulation. Delivering safe water supply sanitation and added produced in cities with population over 100,000. hygiene services contribute to improved health condi- Although the pace of urbanization has slowed down, the tions. It is a cost-effective way to improve the economic population in cities continues to grow at an average rate outcomes of vulnerable populations, given the long-term of 1.6 percent per year.84 Yet, the magnitude in terms of consequences that poor health status and malnutrition the number of people who are poor and face social depri- during childhood have on the individuals’ cognitive devel- vations poses a difficult challenge in urban areas. Overall opment, years of education, adulthood’s wages and labor in 2016, 36.9 million poor lived in urban areas, more than force participation.86 There is also a significant north-south double the number in rural areas (16.5 million). Moreover, gap in access to water and sewage services. Moreover, close 19 million people in urban areas are considered progress in the last decade has been achieved mostly in unable to cover basic food needs in contrast to 8 million the northern (and some central) states (Map 2). in rural areas. And, even though cities have reached near universal access to services, issues in the quality of the 94. Besides the rural-urban inequalities, national averages provision of basic services, and heterogeneity within cities for poverty rates hide large territorial disparities that have persist. Particularly, as cities continue to sprawl without persisted over time. In 2016, 68 percent of the extreme the corresponding extension of infrastructure and service poor lived in only six of the thirty-two states: Chiapas networks, there are serious implications for the most vul- (which held 16 percent of the total number of extreme nerable. For instance, it has been shown that the adapta- poor), Veracruz (14.2 percent), Oaxaca (11.6 percent), tion actions that households in marginal areas of Mexico State of Mexico (11.3 percent), Guerrero (8.8 percent), and City implement to deal with water scarcity, low quality, Puebla (6 percent). In terms of extreme poverty rates, the and flooding imply high financial and opportunity costs, State of Mexico concentrated 13.5 percent of those with with further consequences on poverty.85 an income lower than the minimum well-being line, while Chiapas, Oaxaca, Guerrero, Veracruz and Puebla housed SYSTEMATIC 93. The provision of water and sanitation services differ 44.3 percent (Map 3). According to the 2017 Territorial Eq- COUNTRY greatly between rural and urban communities, north and uity Index for the SDGs, Mexico has high territorial dispari- DIAGNOSTIC MEXICO 83 World Urbanization Prospects: The 2018 Revision (database), Population Division, Department of Economic and Social Affairs, United Nations, New York, https://esa.un.org/unpd/wup/. 60 84 UN (2014). 85 Eakin at al. (2016). 86 Conti and Heckman (2014); Hoddinott et al. (2008); Currie and Almond (2011). 3. Inclusion Map 3. Extreme poverty by municipality, 2014 Mayor a 75% 50% - 75% 25% - 50% Menor 25% No data Source: World Bank 2014 poverty map base on ENIGH2014 and Intercensal 2015. Extreme poverty refers to food poverty (pobreza alimentaria). Figure 52. Socioeconomic classes (%), 2000–16 50 44.0 44.6 40 35.5 28.4 Percentage (%) 30 21.4 20 10 7.8 0 2000 2002 2004 2006 2008 2010 2012 2014 2016 Middle Class $13-$70 (2011 PPP) Poverty $3.2 (2011 PPP) Poverty $5.5 (2011 PPP) vulnerable $5.5-$13 (2011 PPP) Source: Tabulations of Equity Lab, Team for Statistical Development, World Bank, Washington, DC, based on data in SEDLAC (Socio-Economic Database for Latin America and the Caribbean). ties, presenting the highest index value out of seven Latin households at risk of falling into poverty in Mexico is the American countries.87 The evolution of poverty shows ter- largest economic group. For many Mexicans, escaping ritorial disparities. poverty does not mean that they have reached econom- ic security. In 2004, about 39 percent of the population 95. In transitioning the poor and the vulnerable into the were vulnerable to fall back into poverty, and by 2014 middle class, Mexico underperformed with respect to its this number reached almost 42.8 percent (Figure 52). 89 regional peers over the past decade. The middle class in This contrasts with the Latin America and Caribbean re- Latin America and the Caribbean, (those living with an gional average, where 37 percent of the population was income of $13 to $70 a day per capita in 2011 purchasing vulnerable in 2014.90 power parity) achieved overall positive growth between 2002 and 2014; and, for the first time in 2010, the region 96. While income inequality has narrowed since the 1990s, had more people in the middle class than in poverty. income distribution remains highly unequal. More recently, By 2015, however, the region experienced a slowdown the positive trend seen over the last decades has stag- in the trends of declining poverty and a growing mid- nated, particularly due to the negative effects of the con- dle class.88 By 2014, Mexico had a smaller middle class tinuous exposure to aggregate shocks. Measured by the than the Latin America and Caribbean average (22.3 Gini coefficient (in per capita terms and after government percent versus 36 percent). The number of vulnerable transfers), inequality fell from 48.5 in 2002 to 47.0 in 2014 SYSTEMATIC COUNTRY DIAGNOSTIC 87 The value for Mexico’s Territorial Equity Index for the SDGs lies at 0.132, above Guatemala (0.106), Colombia (0.094), Bolivia (0.090), Peru (0.086), Ecuador (0.076), and Chile (0.062). MEXICO It is worth noting that Mexico is the only country where municipal-level data was used for the calculation of the index—whereby territorial differences tend to be larger than in less disaggregated data (Rimisp-Centro Latinoamericano para el Desarrollo Rural, 2018) 61 88 World Bank (2013). 89 Tabulations of Equity Lab, Team for Statistical Development, World Bank, Washington, DC, based on data in SEDLAC (Socio-Economic Database for Latin America and the Caribbean). 90 Tabulations of Equity Lab, Team for Statistical Development, World Bank, Washington, DC, based on data in SEDLAC (Socio-Economic Database for Latin America and the Caribbean). 3. Inclusion Figure 53. Gini (per capita income), 2002–16 51 49 48.3 48.5 47.0 47.1 47 48.2 47.9 45 46.0 43 41 39 37 35 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Source: INEGI 2016a. Note: The numbers for 2016 are not strictly comparable. The Gini coefficient is shown with transfers. Figure 54. Contribution of household income sources to changes in poverty, 2004–14, $4/day (2011 PPP) 1.2 0.9 Percent Points 0.5 0.0 -0.5 -0.4 -1.5 -1.1 -1.3 -2.5 -6.3 -9.1 LAC Mexico Women - shared who are employed Women - labor income Men - shared who are employed Men - labor income Non labor income Shared of individuals 15-69 years old Source: Tabulations of Equity Lab, Team for Statistical Development, World Bank, Washington, DC, based on data in SEDLAC (Socio-Economic Database for Latin America and the Caribbean). Note: PPP = purchasing power parity. (Figure 53).91 In 2016, the richest 10 percent of households 97. Public transfers have played a role in reducing pover- received 50.2 percent of the incomes, while the bottom 60 ty and inequality over the last decade, but they were not percent received 17.2 percent.92 The persistence of income able to compensate for the negative effect of falling labor inequality in Mexico is largely explained by the persistence income. Between 2004 and 2014, labor income in Mexico in labor income inequality, which in turn reflects an un- fell, explaining 10.6 percent of the increase in poverty equal distribution of skills: over 46 percent of individuals (Figure 54). Social programs, including Prospera, Seguro from the bottom quintile are unskilled (those that have Popular, and Programa de Acceso al Financiamiento para less than primary education) relative to only 15.5 percent Soluciones Habitacionales have expanded throughout in the top quintile. In 2016, the poorest quintile had only 5 the years, significantly contributing to improved access percent of total labor income whereas the richest quintile to health care, affordable housing and education for the had 51 percent.93 poor, and providing financial support through cash trans- SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 91 National Institute of Statistics and Geography data. The Gini value for 2016 equivalent to 44.8 is not strictly comparable with the historical series because of a methodological change in the household survey of 2016. In 1996 the average income of the richest quintile was 30.3 times that of the lowest quintile before transfers; by 2014, the difference had dropped to 14.4 times. When taking transfers into account, the 5th quintile took home 12.3 times the income of the poorest quintile in 2014, in comparison with 20.3 times in 1996. This reflects positively on the progressive impact that cash transfers have had in slightly decreasing inequality since the early 2000s. 62 92 See INEGI (2016a). 93 2014 data from tabulations of Equity Lab, Team for Statistical Development, World Bank, Washington, DC, based on data in SEDLAC (Socio-Economic Database for Latin America and the Caribbean). 3. Inclusion Figure 55. Annual growth of remittances in Mexico, 2000-17 70 60 50 40 Annual growth rate (%) 30 20 10 0 -10 -20 -30 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Year, by quarter Source: Banco de Mexico 2017. Figure 56. Share of households receiving remittances by quintile, 2008-16 8% Share of households receving remittances 7% 6% 6% 6% 5% 6% 5% 5% 5% 5% 4% 4% 3% 2% Q1 Q2 Q3 Q4 Q5 2008 2012 2016 Source: World Bank calculations using data of INEGI. fers. They have played a mitigating role, driving most of mittances in the bottom 40, this flow of income represent- the poverty reduction (63.3 percent of the reduction). An ed 35 percent of their total income in 2008, but in 2016 increase of the proportion of households with members this share was reduced to 27 percent. This had negative in the working age population, and a lower dependency repercussions in terms of poverty reduction. ratio have also contributed to a decrease in poverty. 99. There is evidence of intergenerational mobility in educa- 98. Remittances played an important role in poverty reduc- tion and occupational status, although with unequal patterns tion during previous decades, however, their importance has by geographic region. A recent study using data of the ESRU diminished since the global crisis in 2009. The flow of remit- Social Mobility Survey (EMOVI) shows that 48 percent of the tances in Mexico is strongly associated to the economic children of parents in the lowest quintile, do not move to performance of the U.S economy. Until 2007 remittances the next quintile when they grow up. There is also a region- rose significantly, but with the global financial crisis the al pattern: the degree of social mobility is higher than the annual growth rate of remittances fell sharply -reaching national average in the North and North-Central regions negative levels- and it is only after 2013 that they have ex- of Mexico, similar to the national average in the Central re- SYSTEMATIC perienced sustained recovery. Even then, they are far from gion, and lower than average in the South. The children of COUNTRY the growth experienced in the early 2000s (Figure 55.). poor parents (i.e. parents in the 25th percentile rank in the DIAGNOSTIC The share of households receiving remittances dropped national distribution of wealth of their generation) achieved MEXICO between 2008 and 2016 for every income quintile. In greater than average progress if they grew up in the North 63 particular, households in the bottom 40 experienced the region, and smaller than average progress if they grew up in largest reduction (Figure 56). For households receiving re- the South region. Relative intergenerational social mobility 3. Inclusion in education is somewhat larger than that estimated for the opportunity cost of remaining in school. As a result, tertiary case of wealth, but it shows the same regional pattern. The education participation rates remain low (only 16 percent of same results are found for the case of occupational status. the population between 25 and 64 years) and concentrated in the households at the top of the distribution. While 44 per- 100. Intergenerational mobility processes are unequal for cent of the children in the top income quintile of the distri- men and women, and for rural and urban population as well. bution are enrolled in tertiary education, only 15 percent of For example, parents’ education and occupation, as well as the children in the bottom 40 are enrolled. Mexico has one of living in rural areas are the circumstances that explain most the lowest completion rates in higher education of the OECD of the inequality in access to quality education in Mexico, countries, with only 25 percent of the population obtaining suggesting that important barriers remain for intergenera- a degree. tional mobility. In terms of quality of education, Mexico is among the lowest ranking OECD countries, especially in 102. Low and highly unequal outcomes in learning at all lev- math and science with large differences in performance els of the system are key challenges ahead. Approximately based on gender and socio-economic status. In terms of the half of the students graduating from primary education wealth dimension, women have a greater probability of re- have insufficient proficiency in math and language/com- maining in lower strata when they originate from that strata munication as measured by the national PLANEA-ELSEN and have fewer options than men for remaining in higher test. Differences in learning outcomes by type of school strata, even when those women are born in the upper strata. are stark: the proportion of insufficient proficiency among Unlike men, the participation of women in the labor market private schools is 13 percent, among general public school seems to be conditioned by the socioeconomic conditions it is 51 percent, among community schools it is 70 percent, of their households or origin. In particular, a higher educa- and among indigenous schools it is 80 percent.97 The edu- tional attainment of a woman’s father is correlated with a cation sector remains characterized by disparities between greater likelihood of her participation in the labor market. students from households at different income levels.98 103. The results from PISA 2015 show that only half of the 3.2 Trends in productive 15-year olds in Mexico obtain the necessary skills to partic- inclusion ipate effectively and roductively in society and in the labor market. Although the large majority of the 15 year olds in Mexico are enrolled in the education system, Mexico per- 3.2.1. Human capital: access to forms below the OECD average in science, reading, and education is near universal but mathematics.99 Mexico has a larger male-female score gap student performance lags behind than the OECD average: boys outperformed girls in science 101. Mexico is close to achieving universal enrolment in prima- by 8 score points compared with 4 points in the OECD av- ry and lower secondary education, however, enrolment rates in erage100; while similar percentages of boys and girls are low upper secondary education remain low. During the transition and top performers in science. In all three domains, less from lower to upper secondary, the enrolment rate of the than 1 percent of students in Mexico are top performers population between 15 and 19 years old falls to 55.6 percent while around 8 percent of students across OECD countries – the lowest among OECD countries, at almost 30 percent- are top performers in science, 10.7 percent in math and age points below the OECD average94. About 55 percent of 8.3 in reading (Figure 57). Educational outcome also varies dropouts are male, and dropout rates are larger in urban ar- substantially across regions, e.g. the populations of Mexi- eas at 15.2 percent (while in rural areas it is 11.9 percent).95 A co City and Nuevo León have an average of 10.5 and 9.8 potential explanatory factor behind the fall in enrolment and years of education respectively, while the average for Chi- drop-out rates has been pointed out as the limited ability of apas is just 6.3 years. Children in southern Mexico not only the basic education system to develop minimum cognitive spend fewer years in school, they also learn less than their skills, which, at some point, make advancing throughout the peers. In the 2017 PLANEA test, roughly three-quarters of system progressively challenging.96 A reduction of a standard students finishing lower secondary (grade 9) in Guerrero variation in test scores in the sixth grade reduces the proba- and Tabasco were not adequately proficient in math, com- bility of graduating from upper secondary by 5.5 percentage pared with about half in Puebla and Mexico City. points; while higher test scores have a large relationship with student’s likelihood to finish secondary school on time. In ad- 104. Improvements in early childhood education access dition, pressure to generate income soon and a relative high and quality will also be critical. Mexico shows one of the demand for low-skilled labor in the informal sector, raises the highest average ratio of pupils to teaching staff and to all contact staff (teachers and teaching aides) in pre-primary SYSTEMATIC COUNTRY 94 OECD (2017). DIAGNOSTIC 95 National Survey about Dropouts in Upper Secondary Education 2010–11, Encuesta Nacional de Deserción en la Educación Media Superior. MEXICO 96 de Hoyos et al. (2018). 97 Educational indicators Bank (Banco de Indicadores Educativos), RE01a. Share of students that obtain the educational achievement level (insufficient) in the areas evaluated in the PLANEA-ELSEN (2015) tests. 98 World Bank (2016). 64 99 In science, Mexico scores 416 points, compared with the OECD average of 493; in reading the country scores 423 points, below the OECD average of 493; while in mathematics Mexico’s score is 408, below the OECD average of 490. 100 Refers to scores of 15-year-old female and male students on the PISA 2015 science literacy scale. 3. Inclusion Figure 57. Percentage of students in the bottom and top levels of performance in math, PISA 2015 % of low achievers % of top performers (below level 2) (level 5 and above) 0 36 5 34 Singapore Singapore Korea rep. 10 32 15 Malasya 30 Poland Spain 20 28 25 OECD average 26 30 24 35 Korea rep. 22 40 20 45 18 Chile 50 16 55 Turkey 14 60 Mexico Poland 12 Peru 65 10 OECD average Colombia 70 brazil 8 75 Spain 6 Algeria 80 4 Malasya 85 2 Mexico, Peru, Colombia Dominican rep. brazil, Chile, Turkey 90 0 Dominicana rep. 95 100 Source: World Bank calculations based on PISA 2015. Figure 58. Children-to-contact staff (teachers and teachers’ aides) ratios in pre-primary education services, 2015 30 25 20 15 10 5 0 Norway Germany Slovenia Chile Slovak republic Hungary Czech Republic Italy Korea Netherlands Japan France Brazil Mexico Sweden Lithuania Austria Luxembourg Source: Education at a Glance 2017, OECD. education services (public and private) among OECD and (ages 15 to 65 years old) is composed roughly of 82 mil- peer countries (Figure 58). Moreover, the child-to-contact lion people, of which 63.2 percent (52 million) participate staff ratio in early childhood educational development in the labor market, while the remaining 36.8 percent (30 services is 25, while the average among OECD countries million) does not participate. Unemployment in Mexico is around 14.101 Although imperfect and restricted to only only reaches 3.6 percent (1.8 million) of the labor force. one single aspect of quality, the child-to-staff ratio gives However, 16.1 percent (5 million) of those who are out- at least some quantitative indication of the frequency of side the labor force, showed their discouragement in the contact between staff and children. Quality provision of search for employment in the same period. The employed early childhood education and care is fundamental to re- population in the country is around 50 million, comprised duce the impact of family background on an individual’s mostly by salaried workers (35 million) and self-employed life chances to break the cycle of transmission of disadvan- workers (10 million). The remaining 9.1 percent is made up tages from one generation to the next. by employers and unpaid family workers, who share this rate in equal proportions (Figure 59). SYSTEMATIC 105. Despite low unemployment rates in Mexico, there is a COUNTRY DIAGNOSTIC considerable group of the population that remains available 106. Seven out of ten Mexicans who migrate abroad report MEXICO but discouraged to search for a job. The working population doing so because of work reasons. The strength of the U.S 101 https://www.oecd.org/els/soc/PF4-2-Quality-childcare-early-education-services.pdf. 65 3. Inclusion Box 5. The high and rising incidence of noncommunicable diseases in Mexico is associ- ated with substantial economic costs and particularly pernicious effects on the poor An aging population and changing lifestyle factors are driving an increase in chronic noncommunicable diseases and degenerative conditions, which account for more than 75 percent of total mortality in Mexico. Diabetes and cardiovas- cular disease alone are responsible for more the 33.4 percent of mortality in adult population. In 2015, the prevalence of diabetes among the population ages 20–79 years was 15.8 percent, which is the highest rate in the OECD and more than twice the OECD average (7 percent). The north and central regions, along with urban areas, have the highest prevalence of diabetes. Under current trends, the population ages 50 and older diagnosed with diabetes is expected to increase from 19.3 percent to 34.0 percent by 2050. A risk factor associated with diabetes and cardiovascular disease is excessive weight. Mexico has one of the highest overweight rates and the largest proportion of obese children in the world. Estimates from the 2016 National Health and Nutrition Survey reveal that 72.5 percent of the population older than 20 and 33.2 percent of children ages 5–11 are overweight. Sedentary behavior and physical inactivity have grown among adults. And in 2016, 83 percent of children and adolescents (10-14 years) did not fulfill the weekly amount of physical activity recommended by the World Health Organization. In addition, evidence suggests that hypertension, a significant risk factor for cardiovascular disease, is increasing, in parallel with obesity among children and adolescents. Children and adolescents ages 12–18 show higher prevalence rates of hypertension (14.1 percent) compared with their counterparts in China and India (5.2 percent and 10.1 percent, respectively). The costs associated with primary obesity–related diseases are projected to increase from US$880 million in 2013 to US$1.2 billion by 2030, with intensifying pressure on the health care system. The direct and indirect costs of diabetes and its complications have been estimated to account for 2.3 percent of GDP. A related study projects the annual costs of diabetes to the Mexican economy (including direct and indirect costs) at 2.6 percent of GDP in 2018. The rising incidence of noncommunicable disease also represents a risk for human capital formation. Noncommunica- ble diseases have direct and indirect costs in time, productivity, and income loss by patients and caregivers, as well as negative impacts on household members, such as diminished food consumption, increases in school absenteeism, and decreased expenditures on education, which can have long-lasting impacts on human capital formation. People with diabetes are prone to additional health complications and loss in productivity and earnings. In Mexico, dia- betes is the leading cause of premature retirement, blindness, and renal failure, while mortality rates tend to be higher among women.a Among the poor, diabetes and cardiovascular disease represent a particularly acute employment con- straint. Men blue-collar workers who suffer from cardiovascular disease experience an average 1.17 sick days more per month. Not only is the prevalence of diabetes greater in low-income groups, but it affects their labor market outcomes more negatively. Self-reported diabetes is associated with a lower probability of employment, and the association is stronger among the poor, particularly among poor men.b In addition, the negative impact of diabetes on the proba- bility of employment is stronger within the agricultural sector and self-employment, where the poor are concentrated. Sources: González-Pier et al. (2016); OECD (2015); González-González et al. (2016); Medina et al. (2018); Dyson et al. (2013); Barraza-Lloréns et al. (2013); Arrossi et al. (2007); Rull et al. (2005). a. The leading cause of mortality in Mexico in 2005–09 was diabetes mellitus (PAHO 2012; data from Secretaría de Salud, Mexico). The leading cause of death among women is diabe- tes; among men, it is cirrhosis and other chronic liver diseases (INEGI 2010). While the mortality rate tends to be higher among women, rates were higher among men in Baja California Sur, Colima, Mexico City, Michoacán, and Quintana Roo (Secretaría de Salud 2010). b. Seuring et al. (2016) find that diabetes in Mexico decreases employment probabilities by 10.0 percentage points among men and 4.5 percentage points among women. Figure 59. Composition of the workforce, 2017 Working age population (15-65 years old) 81,830,389 in the labor force Not in the labor force 51,675,825 30,154,564 (63.2%) (36.8%) Employed unemployed Available Not available 49,826,135 1,849,690 4,835,663 25,288,901 (96.4%) (3.6%) (16.1%) (83.9%) SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO Wage worker Employer Self-employed unpaid 34,971,016 2,183,063 10,299,629 2,372,427 (70.2%) (4.4%) (20.7%) (4.7%) 66 Source: ENOE 2017-II. Note: The working-age population is defined as the population ages 15–64. 3. Inclusion Box 6. Potential distributional effects of automation in Mexico Mexico, among a few other countries, did not experience real wage growth associated with increased use of robots. Automatization in Mexico has been linked with a drop in unit labor costs and a more rapid decline in activities relying relatively more on robotic automation than in industries with low robot density. In consequence, the automatization phenomenon has mostly rewarded capital and contributed to the downward trend in the share of labor income in the country.a Moreover, real wages in the highly automated automotive sector dropped by 1.6 percent between 2011 and 2015, while real wages expanded by 1.5 percent in manufacturing as a whole. These findings suggest that the overall distributional impact of robots may be adverse.b Significant output effects because of technological adoption can lead to an overall positive expansion in employment among both lower-skilled and higher-skilled workers in Mexico.c These output effects are greater in more highly tradable sectors that provide larger output expansion opportunities.d However, sectoral effects determine the degree of inclu- siveness of technological adoption by Mexican firms on wages. Wages increased in manufacturing.e Wages in the service industry declined among both white-collar and blue-collar workers, but especially among white-collar workers.c These differences can be explained by the higher risks of the automation of jobs in the service sector relative to the manufactur- ing sector.f Through the continued use of technology by firms, wage inequality was reduced in both sectors as low-skilled workers, rather than being replaced by technology, have become stronger complements of technological adoption. a. ILO and OECD (2015). b. UNCTAD (2017). c. World Bank (2018). d. For example, in the manufacturing sector, a 10 percentage-point increase in the share of labor using the Internet led to a 11 percent increase in white-collar workers and a 6 percent increase in blue-collar workers relative to a 7 percent increase in white-collar workers and an 11 percent increase in blue-collar workers in the service sector. e. A 10 percentage point increase in Internet use in manufacturing was associated with a 14 percent increase in white-collar wages versus a 16 percent increase in blue-collar wages. f. World Bank (2016). economy in the 1990s provided an escape valve for workers compared with the OECD average, as well as with respect who could not find a job in Mexico, and the remittances gen- to Latin America and the Caribbean and upper-middle-in- erated by this outflow a lso s upported t he l ocal e conomy. come countries (on average standing at 54 percent and 73 But after the global economic crisis of 2008 and its aftermath percent, respectively). The share of adults with an account at -combined with stricter enforcement of U.S immigration a formal institution in fact decreased in recent years, from 39 laws- a significant decline in net migration flows occurred. percent in 2014 to 35 percent in 2017. The disaggregation Although net migration is expected to increase over the by gender reflects growing exclusion: while the percentage coming years, the flows are unlikely to reach the levels reg- of adults without a bank account remained constant for istered during the 1990s.102The estimates of international men between 2014 and 2017 (at 39 percent), it decreased migration show a declining trend in the past decade; while for women during the same period: from 39 percent to 33 an average of 140 migrants per 10 thousand residents was percent. Moreover, differences in participation rates by peo- registered in 2006, for 2010 the number dropped to around ple in the top 60 percent income bracket and the bottom 35. The main reason to migrate abroad is to search for a job 40 percent are relatively small in OECD countries (5 percent or because of having already found a job in a foreign country points of difference; 93.7 percent versus 88.6 percent re- (68 percent), followed by those who report family reunion spectively). In Mexico, on the other hand, there is a signifi- motives (14.4 percent) or study (12.4 percent).103 cant gap in access between the poorest 40 percent and the richest 60 percent of the population—about a difference of Financial capital: absence of service and high 18 percentage points.104 costs for poor households 107. The poor and vulnerable in Mexico have limited ac- 108. The cost of being excluded from the traditional financial cess to financial services compared with the same economic system is high for the poor and vulnerable households as they groups in OECD and Latin American countries. In the context rely on informal institutions and pay higher interest rates. Ac- of the financial sector reform significant progress has been cess to financial services in formal banks tends to be low made over the last four years. Credit (as a share of GDP) to in countries with a high level of informality, among other the private sector increased and people with bank account factors, because of lack of formal documentation. The poor increased by more than 10 percentage points due to the re- rely on informal money lenders or pawn shops to obtain forms. Yet further and more accelerated progress would be loans, which charge higher interest rates. Recent reforms in needed. As of 2017 in OECD countries, close to 92 percent Mexico increased the availability of access to credit. How- of adults had an account at a bank, credit union or other ever, the cost of borrowing remains substantially higher in type of financial i nstitution. I n M exico, o nly 3 5 p ercent o f the retail banks targeting low income consumers. For in- adults had an account at a formal financial i nstitution i n stance, across the main traditional banks, the total annual 2017, placing the county toward the lower end in this metric cost of personal loans105 in 2017 ranged between 30 and SYSTEMATIC COUNTRY DIAGNOSTIC 102 Chiquiar and Salcedo (2017). MEXICO 103 Based on information from the Encuesta Nacional de la Dinámica Demográfica. 104 Global Findex (Global Financial Inclusion Database), World Bank, Washington, DC, https://globalfindex.worldbank.org/. 67 105 The Bank of Mexico’s total annual cost (Costo Anual Total, CAT) indicator, which must be disclosed by banks, was established as a measure to facilitate the comparison of financial products. The indicator provides the total financing cost of a loan, integrating not only the interest rate but all the elements the consumer will be charged such as annual fees and commissions, among others (Banxico, n.d.) 3. Inclusion 60 percent; whereas in the largest retail banks, the rate was 5 hectares.110 More recently, the 2016 INEGI agricultural significantly higher, ranging from 80 to 120 percent.106 census update shows that 51.9 percent of agricultural land holdings are small (0–2 hectares).111 Land fragmen- tation has been associated with inefficiencies in crop 3.2.2. Natural capital: uneven productivity due to factors associated with inefficient development in agriculture and resource allocation, lack of economies of scale, among forestry other issues (see also Box 7). 109. A large share of land in Mexico is highly fragmented, with implications for agricultural productivity. As classified 110. On the remaining land, there is high concentration. by the National Institute of Statistics and Geography (IN- In 2007, the largest one percent of farms occupied over EGI), about 17 percent of total land (31.2 million hectares) half (56 percent) of the agricultural land.112 This was the in Mexico is agricultural land.107 A large share of this land fifth largest land concentration in the region after Peru is still used for traditional low-value commodities (maize, (77 percent), Chile (75 percent), Paraguay (71 percent), wheat, beans, etc.). Less than half (43.6 percent) of the and Bolivia (66 percent). According to the 2016 “marco agricultural land is privately owned, almost 9 percent is censal” update, 2.4 percent of the large producers own communally owned, and the remaining 47.3 percent is 15 percent of the land in Mexico. Large-scale producers, under ejidos.108 The agricultural sector in Mexico shows vertically integrated in agribusiness supply chains, are a dualistic structure.109 On one hand is the numerous, mainly located in the northern states.113 In the middle low-productivity, small-holder and subsistence sector. of the spectrum are the small to middle-sized producers According to the 2007 INEGI agricultural census, 63 per- (5–20+ ha.). Earlier research has argued that this middle cent of agricultural employment is on farms of less than group could benefit the most from agricultural subsidies, Box 7. Agribusiness as an opportunity for productive inclusion Agribusiness represents approximately 7.5 percent of Mexico’s GDP, considering the contribution of primary agriculture, and the food and beverages industries (the agriculture sector alone accounted for 3.2 percent of GDP). In the third quarter of 2017, it employed 9.3 million people with a mean monthly labor income of $4,285 current pesos.a Agriculture and agri-food exports have been a very important source of growth in Mexico, as a result of export expansion and increased domestic com- mercial production (grains and oil-seeds). Mexico’s main agricultural trading partner is the US, receiving almost 80 percent of agricultural exports. Due to its diverse geographic, varied climate, strategic geographic location and labor availability. Mexi- co is highly competitive in the production of fruits and vegetables and has become the largest exporter of beer worldwide. Despite its competitive advantages, several impediments constrain growth of the agriculture and agri-food business sector. One of the main structural constraints is the issue of land fragmentation. Although highly competitive farms and agribusiness companies drive sectoral growth, 81.3 percent of the agricultural units (economic rural units) in Mexico are actually equal/smaller than 5 hectares.b Many of these units are worked by semi-subsistence farming households, employing traditional, rainfed production practices, particularly in the central and southern parts of the country. The small size of plots prevents the formation of economies of scale and limits market integration, except in cases where effective farmer organizations are in place. Technological factors also play a major role in determining large differences in agriculture productivity among producers and geographical regions. Furthermore, land under communal ownership (ejido) has encouraged conservation of forest but has been less successful on driving higher levels of entrepreneurship around the sustainable use of forest and agriculture land. Increased crime rates in some states, with effects on increas- ing operational costs, are discouraging investment or even displacing production. Geographic differences in infrastruc- ture, such as between northern and southern states, limit access to markets and reduce competitiveness. Agriculture is highly vulnerable to weather extremes, such as water scarcity in the north, or tropical storms in the southern parts of the country, which can cause extensive damage to crop and livestock production. Overcoming these set of challenges is critical given the importance of the sector for productive inclusion in the country.  a. ENOE (2017). Based on the (NAICS), Agribusiness includes: Agriculture, Forestry, Fishing and Hunting (Sector 11); Food Manufacturing (Sector 311); and Beverage and Tobacco Product Manufacturing (Sector 321). b. SAGARPA/FAO (2013). 106 As of June 2017, the average total annual cost in the following banks was: Scotiabank (32.8 percent), HSBC (34.4 percent), Santander (55.5 percent), and Banamex (56.5 percent), while in the retail banks it was: Bancoppel (80.9 percent), Banco Azteca (104.3 percent), and Banco Famsa (119.3 percent) (CONDUSEF 2017). These rates are, however, lower than those charged by money lenders or pawn shops; e.g. the CAT for a valuation of a piece of gold (4 gr) in Prendamex, a large national pawn shop chain, was 249.8 percent in 2009 (PROFECO 2009). 107 Under the FAO global definition, 53 percent of land in Mexico (106 million hectares) is agricultural land. This definition, however, is less accurate for the country given its type of vege- tation and crops; in addition to the fact that the FAO definition groups pastures, crops, and other vegetation together, making its interpretation less straightforward. The SCD thus uses the national official numbers based on the land use and vegetation maps, that is, the INEGI series (see, for example, INEGI 2008, 2016b). The INEGI definition includes the different SYSTEMATIC types of agriculture developed in Mexico, organizing the information under three criteria: land occupation (permanent or nomadic agriculture), temporality of crop, and water supply. COUNTRY The use of the INEGI definition is also consistent with the objective of alignment with national priorities. DIAGNOSTIC 108 The ejido is an association of peasant farmers, who are owners of common property assigned to them by the State. This form of collective land ownership is recognized by the Polit- MEXICO ical Constitution of the United Mexican States and by the Agrarian Law. The communal land is used for agriculture, where community members individually farm designated parcels and collectively maintain communal holdings. 109 Scott (2010). 110 INEGI (2008). 68 111 INEGI (2016b). 112 OXFAM (2010). 113 Fox and Haigh (2010). 3. Inclusion Box 8. The timber sector: incentives to balance conservation and economic objectives? Mexico faces the challenge of conserving and sustainably managing its forests while also meeting a growing de- mand for timber products, which currently exceeds production by a factor of three. Forestry production in Mexico accounted for 0.6 percent of GDP in 2015.a Timber is produced in Mexico from managed natural forests (93.2 per- cent of the total value), and, to a lesser extent, from commercial forest plantations (6.8 percent). Forest management is an important activity carried out primarily by the owners and holders of forest lands. An estimated 61 percent of Mexico’s 66 million hectares of temperate and tropical forests are owned by núcleos agrarios (agrarian forest com- munities). Approximately 53 percent of forest lands with a high productivity potential and 52 percent of forests with average productive potential are found within núcleos agrarios. Mexico’s system of collective production is unique in scope, maturity, and the impact on local communities. In 2015, approximately 992 community-based forest en- terprises (CFEs) were active in the country. Many were located in highly marginalized areas. Over the last two de- cades, CFEs have flourished. Studies have found that well-run CFEs contribute substantially to local development by creating jobs and economic opportunities, while CFE earnings are reinvested in public services and infrastructure. The success of CFEs reflects the soundness of the legal framework for local ownership and the strong social capital of rural communities, although various obstacles can limit the domestic and international competitiveness of these communities. Mexico has the capacity to more than double the current production of standing timber (in cubic meters). The rising number of CFEs has not yet been matched by a commensurate increase in timber production. Instead, legal timber production fell from 9.4 million cubic meters in 2000 to 6.1 million cubic meters in 2015, even as domestic de- mand for timber rose. This trend derives in part from the failure of sectoral incentives to balance conservation and economic objectives. In 2015, with almost 1,500 harvesting permits granted for forest exploitation in the country, extraction occurred in only 46 percent of the approved forest land area. Pinewood accounts for 74.8 percent of the total volume of timber produced in natural forests. In the areas that are most accessible, Mexico has the potential to produce 60 million cubic meters of roundwood at the national level, while the priority areas identified by CONAFOR (watersheds, as well as the states of Campeche and Quintana Roo), could produce up to 16 million cubic meters of roundwood.a CFE production costs in Mexico have been too high compared with other countries, especially low-cost producers, such as Brazil and Chile, or even small private landowners in the U.S. south and Pacific Northwest. Mexican timber exports have potential competitors in these three countries, as the types of timber production they have are com- parable to Mexico’s.b Chile and the United States have coniferous forests, while Brazil has tropical forests as well as extensive, high yield plantations.c Table 3 presents the macroeconomic data of the forestry sector in the four countries. Table 3. International comparison of the forestry sector: Employment, GDP, and balance of trade Indicator Mexicod Brazile Chilef United Statesg Forest GDP, US$, million, 2012 prices 6,135 22,938 5,553 102,600 Trade balance, US$, million, 2012 prices −5,973 5,520 3,138.52 −14,041 373,873 508,084 78,668 802,050 Employment in the sector, total (2015) (2012) (2014) (2015) Productivity per worker (US$/worker) 16,409 45,146 70,587 127,920 39 7–15 7–15 12 Exploitation costs, US$ per cubic meter (max. 119, min. 9) (South America) (South America) (south) Forest area, 1,000 hectares 65,205 543,905 15,536 225,993 Wood volume from forests, million cubic metersh 2,871 71,252 2,486 30,838 Total forest productivity, US$, hectare 570.23 359.65 941.18 1,123.28 Sources: World Bank data; IMCO 2015. a. IMCO (2015) and World Bank (forthcoming). b. Cubbage et al. (2015). c. IMCO (2015); World Bank (forthcoming). d. INEGI. SYSTEMATIC e. ABRAF (2013); f. Forestry Institute, Chile (2015). COUNTRY g. US Bureau of Economic Analysis, and US Bureau of Labor Statistics. DIAGNOSTIC h. Total amount of wood in each country according to the density of forests, that is, the total stock of wood in cubic meters. MEXICO 69 3. Inclusion often unable to benefit from programs such as Progresa, territory—around 88 million hectares.119 About 60 percent while not being large enough to attract the larger agri- of forests belong to rural communities. About 12 million cultural subsidies.114 These producers receive some sup- people live in forest areas; of this population, 88 percent port from programs such as PROAGRO115 but often lack live in highly marginalized localities, and 62 percent live the complementary inputs, which tend to be received by in poverty.120 More than half of forest dwellers live in con- larger producers.116 ditions of extreme poverty, with limited access to health services, education, and accessible forest tracks. In 2010, 111. The disparities in agricultural development between the percentage of inhabited private dwellings without the North and the South are increasing; while gender in- sewage systems was four times higher in forested areas equality remains. The contribution of primary agriculture than in the rest of the country.121 Areas with high poverty to total GDP was 3.2 percent in 2017, but the national and marginalization also show high rates of deforestation average conceals tremendous regional disparities. In and, at the same time, dependency on natural resources. 2015, five states in the North/Center (Jalisco, Michoacán, In 2008, 57 percent of the rural household in the poorest Sinaloa, Chihuahua and Sonora) contributed 50 percent of quintile obtained income from extracting natural resourc- Mexico’s agricultural GDP, while states in the South, which es. However, deforestation has reduced the extension of are home to a large proportion of traditional agriculture forest over the past five decades, which together with producers (in number) contribute minimally to total agri- degradation has been driven by proximity to cities and ru- cultural GDP. In Oaxaca, for example, agriculture (primary ral population centers; topography and soils appropriate sector) has relatively little participation in state GDP (6 for agriculture are the main correlates. Especially in south- percent) and little contribution to the national agriculture ern Mexico, where poverty levels are highest, the highest GDP (2.7 percent). Yet, 32 percent of employment in Oax- rates of forest loss (tropical dry and tropical rain forest) can aca falls within the primary sector. This signals the impor- be observed (with 32 percent of deforestation of the total tance of agriculture to livelihoods, but also the challenge of 9.5 million hectares lost between 1993 and 2012).122 of enhancing the value added generated by the sector. Moreover, women farmers own only 10 percent of the land that they work and are subject to higher productive exclu- 3.2.3. Social capital: low civil sion.117 Indeed, an important factor affecting rural women engagement and social cohesion is their disadvantage in land tenure, inheritance rights, prevents productive inclusion limited access to decision making processes, and use and 113. In Mexico’s society, civic engagement, trust, social control over natural resources. Although inheritance laws network support and other measures typically used to ap- in Mexico treat men and women equally, land inheritance proximate social capital123 are comparatively low. Based on in 2011 was about four times higher among sons than different measures used in the literature, it is often argued among daughters.118 that social capital is weak in Mexico. According to an OECD report, social support has fallen in Mexico over the past 112. Renewable natural resources, such as forests, provide a decade. While 88 percent of people reported having a particularly important source of income for the poorest pop- friend or relative whom they could count on in 2005-07, ulation. Due to its geographical location and topography, by 2014-16 this share had dropped to 80 percent -which Mexico has a great variety of ecosystems and biological is considerable below the OECD average of 89 percent. diversity. Mexico’s forests contribute to the overall econ- Moreover, voter’s turnout, a measure of citizens’ partic- omy through the provision of critical ecosystem services ipation, was 63 percent in the recent election. Based on that sustain key sectors and rural and urban areas, such as the Indices of Social Development124, Mexico lags behind mitigating soil erosion, maintaining soil fertility, support- in terms of group cohesion, and measures of safety and ing the infiltration of water, and providing raw material for trust. Intergroup cohesion, which measures ethnic ten- productive sectors and natural habitats for biodiversity. sions and discrimination, puts Mexico among the worst The country’s forests are also an important sink for carbon rankings compared with its peers (Figure 60). In terms of dioxide. Forests cover around 45 percent of the national civic activism, measuring the use of media and protest 114 Scott, 2010 115 In 2014, the Productive PROAGRO program succeeded PROCAMPO, maintaining the beneficiaries who were registered in PROCAMPO but re-coupling payments to production (payments are to be used to cover production expenses). PROAGRO payments are differentiated by producer type, with subsistence farmers receiving a higher payment compared to transitional and commercial farmers (OECD, 2017). For example, in the Yucatan Peninsula as a whole, 87 percent of the agricultural producers who benefited from PROAGRO cultivate 5 hectares or less (Mardero et al. 2018). 116 ASERCA, for example, promotes the commercialization of crops through contract farming, production guarantees, risk coverage and incentives for storage. The agency has a budget of about MX$10 billion per year, and reaches around 260,000 beneficiaries yearly, many of whom are large farmers. The poorer regions in the South often have operated at scales too small to participate in ASERCA’s incentive programs. A recent World Bank initiative is looking to expand ASERCA’s role in providing support for the productive and financial inclusion of small and medium farmers (World Bank, 2016). 117 World Bank (forthcoming) SYSTEMATIC 118 World Bank (2011). COUNTRY 119 This definition of forests covers low dry deciduous forests, as well as coniferous forests and some types of xeric shrublands DIAGNOSTIC 120 World Bank (forthcoming). MEXICO 121 World Bank (forthcoming). 122 World Bank (forthcoming). 123 Built by the combination of formal and informal social engagement and exchange, social capital is frequently conceptualized as a mechanism that “gives rise to stocks on which people can draw” (Rakodi 2002). 70 124 The International Institute of Social Studies produces the Indices of Social Development (ISD), which brings together measures of 200 indicators synthetizing them along 6 dimen- sions of development: (i) civic activism, (ii) clubs and associations, (iii) intergroup cohesion, (iv) interpersonal safety and trust, (v) gender equality, and (vi) inclusion of minorities. The indices are produces for 193 countries from 1990 to 2010 and uses a method of ‘matching percentiles’ to aggregate the indices. 3. Inclusion Figure 60. Intergroup cohesion index, 2010 Figure 61. Civic activism index, 2010 0.8 0.58 0.7 0.56 0.58 Intergroup cohesion index (0 to 1) 0.6 Civic activism index (0 to 1) 0.54 0.5 0.4 0.52 0.52 0.3 0.50 0.2 0.1 0.48 0.0 0.46 Turkey Malaysia Chile Uruguay Poland Peru Colombia Mexico Brazil Argentina Korea, Rep. Colombia Mexico Chile Malaysia Argentina Turkey Uruguay Poland Peru Brazil Korea, Rep. Source: International Instituto of Social Studies, Indices of Social Development, 2010, Source: International Instituto of Social Studies, Indices of Social Development, 2010, http://www.indsocdev.org/data-access.html. http://www.indsocdev.org/data-access.html. behavior, Mexico ranks much below its aspirational peers gap between Mexico’s urban and rural areas. Moreover, (Figure 61). the allocation of investment in school facilities, teacher training and educational supplies across states com- 114. Low levels of social capital accumulation affect the pounds Mexico’s already substantial regional inequalities. poor disproportionately, as they rely more on informal net- States efforts to channel their own resources to education works. The implications of having relative low levels of are an important component in actual spending that gives social capital affects the whole population but especially them the ability to increase the expenditure by pupil. This the poor as networks and social relationships could serve has contributed to polarize public spending in education as mechanisms to reduce uncertainty in the presence of among states.128 A regressive distribution of financial re- market failures. For example, under asymmetries of infor- sources contributes to the positive correlation between mation people resort to their network to access job op- school quality and household income level. In relatively portunities, credit and financial resources and so on. For affluent areas such as Mexico City and Nuevo León, invest- instance, there is evidence that poor farmers in peri-urban ment per student at the basic education level is above the areas in Mexico, expand their chances to improve their national average (MX$20,000 per year), whereas in Guerre- poverty status as social capital increases125. ro, Chiapas, and Oaxaca, investment per student is signifi- cantly below the national average (Figure 63). 3.3 Unequal service delivery 116. Low efficiency and limited coverage of secondary amplifies inequalities education contributes to the uneven distribution of edu- cation services, perpetuating socioeconomic inequalities. 115. Mexico’s current education and health spending pat- Although the primary and lower secondary enrollment terns exacerbate existing socioeconomic inequalities. Be- rates have improved over the last 30 years, retention rates ginning with the former, richer and larger states in terms at the upper secondary level and overall educational at- of population and number of students tend to have higher tainment indicators continue to lag those of comparator public education spending.126 Spending is also largely di- countries.129 Spending in primary education represents rected toward urban centers and subsystems, where urban the greatest proportion of public spending in education schools exhibit significantly better outcomes than their across states. In contrast, total expenditure and per-pupil rural peers (Figure 62). Results from the 2017 Planea test expenditure in secondary and tertiary levels is low com- show that upper secondary students in urban areas score pared with the needs and coverage goals set for coming significantly higher than students in rural areas in math years.130 Moreover, while tertiary education is intrinsically and language.127 These disparities widen the economic expensive, and its distribution is almost always regressive, SYSTEMATIC COUNTRY 125 Méndez-Lemus and Vieyra (2017). DIAGNOSTIC 126 CIEP (2016). MEXICO 127 Students in the third year of secondary school in urban areas scored on average 507 in language and communication and 503 in math, compared with students in rural areas, who scored 452 and 475, respectively. INEE, PLANEA 2017. 128 Students in the third year of secondary school in urban areas scored on average 507 in language and communication and 503 in math, compared with students in rural areas, who 71 scored 452 and 475, respectively. INEE, PLANEA 2017. 129 Only 6 out of 10 students who enroll in upper secondary ultimately graduate, and graduation rates are even lower among students from poor and vulnerable households. 130 World Bank (2016). 3. Inclusion Figure 62. Percentage of students with sufficient reading skills (ENLACE), by spending source, 2006–13 80 70 60 50 40 30 20 10 0 2006 2007 2008 2009 2010 2011 2012 2013 CONAFE (MXN ~$9.846 per student in 2015) SEP (MXN ~$15.300 per student in 2015) Private Source: World Bank calculations using ENLACE. Note: SEP (Secretaría de Educación Pública) refers to Mexico’s Secretariat of Public Education, CONAFE (National Council for Education Development) is a division of SEP which provides educational services to rural, small and indigenous areas. Figure 63. Spending per student at the basic education level by state, 2013 50,000 60% 45,000 50% 40,000 % of State Budget 35,000 40% 30,000 MXN 25,000 30% 20,000 20% 15,000 10,000 10% 5,000 0 0% Estado De México Oaxaca D.F. b.C. b.C.S. Durango Campeche Zacatecas Colima Tamaulipas Nuevo León Sinaloa Nayarit Sonora Coahuila hidalgo Quintana roo Aguascalientes Morelos Tabasco veracruz Jalisco Guanajuato Puebla Querétaro Chihuahua Yucatán Michoacán Guerrero Tlaxcala San Luis Potosí Chiapas Spending per Student Educación básica State Spending per Student Educación básica Federal National Average Share of State budget Source: World Bank 2016b. it appears that tertiary spending is becoming increasingly than 5,000 post-secondary institutions in Mexico, of which regressive over time.131 around 3,000 are privately operated.132 Student lending (cur- rently with a penetration rate of around 1 percent of the total 117. The gross enrollment rate in tertiary education in Mexico is market), has potential to reduce liquidity constraints for mid- one of the lowest among relevant comparators. Estimated at 30 dle-class students willing to enroll into private universities. Giv- percent in 2014, gross enrollment rate in tertiary education is en the limited role of the government in student lending and if significantly lower than Argentina (70 percent) and Chile (83 enrollment in tertiary education were to continue its historical percent). Among the many reasons of why about half of the trend of around 5 percent per year, the student lending market students completing upper-secondary education enroll into would be insufficient to increase enrollment into private tertia- tertiary education is the fact that good quality public univer- ry education in the margin. Mexico has been classified, along SYSTEMATIC sities, which are typically tuition free, are close to saturation, with countries such as South Africa, in the middle rung in terms COUNTRY while a significant percentage of the students face liquidity of the student loan market maturity, with growing awareness DIAGNOSTIC constraints to enroll into private universities. There are more and uptake/utilization, but behind mature markets.”133 MEXICO 72 131 Scott (2010). 132 Parthenon Consulting based on National Sources; IFC analysis. 133 IFC (2015). 118. Access to health services has expanded, although Mex- 3. Inclusion Figure 64. Insurance coverage, by income ico’s health sector remains under-resourced. Between 2004 quintile, 2016 and 2014, public health spending as a share of GDP rose by approximately 43 percent, including in terms of catastrophic expenditure. This contributed to a reduction in out-of-pocket 20.8 spending, which fell from 55 to 44 percent of total spending 43.7 50.4 between 2003 and 2014—although it remains a source of 61.2 inefficiency and barrier to health among poor households. 76.4 Notwithstanding its expansion, health spending in Mexico 87.3 is still low by international standards. Despite reaching 6.3 percent in 2014 (up from 4.9 percent in 2000), total health 75.5 expenditure in Mexico lies at half of the OECD average, and below the Latin American average of 7.2 percent. Public sec- 54.5 47.6 tor health spending accounted for 52 percent of total health 37.4 spending, versus the OECD average of 70 percent and be- 23.0 low most regional comparators. Also, as a result of reforms, 11.4 insurance coverage expanded substantially. By 2012, Seguro Q1 Q2 Q3 Q4 Q5 Total Income quintile Popular (which was established in 2003) covered 52.6 million Social security Seguro Popular Mexicans previously lacking health insurance (Figure 64). Private insurance Other The trend in increased access has continued since then. The Source: World Bank. share of population without access to health care was halved in less than a decade, from 38.4 to 15.5 percent between effective insurance and high-quality care (Figure 65). This is a 2008 in 2016, according to national statistics.134 Since 2000, regressive outcome that poses an obstacle to care, especially maternal mortality has fallen by 50 percent, and infant mor- among lower-income households. About 45 percent of total tality by over 40 percent. Moreover, between 2004 and 2014, ambulatory OOP health costs are borne by households in catastrophic health spending by households in the poorest the first three income quintiles, and 28 percent are borne by income quintile declined by 11 percent, and impoverishing noninsured households across all income quintiles, especially health spending fell by 54 percent.135 noninsured households in quintile four.137 Also, the poorer quartile of the population spends a significantly higher (37 119. Yet, the available resources for health are not spent effi- percent) proportion of their income on health than the richest ciently. The impact of the budget tends to be weakened by quartile. As a large share of total health spending is financed a combination of allocative and technical inefficiencies, often OOP, the need for expanded risk pooling becomes even more resulting in disparity in health outcomes. The fragmentation critical to improve the distributional equity of the health care of the health care system is a key issue, limiting the ability of system.138 In addition, a more efficient use of resources is providers to exploit economies of scale, and contributing to made even more urgent by the fact that expenditures are ex- an unequal service delivery, expenditure inefficiencies and pected to grow organically in the country due to factors such the suboptimal use of sectoral assets. The quality of service of as population ageing and epidemiological transition. the different public health insurance scheme varies, offering different entitlements to their beneficiaries, which results in 120. Regional disparities in health continue to exist in Mexico, highly uneven access to care that often exacerbates under- including in the private health sector. The density of doctors lying socioeconomic inequalities. Differences in health out- is 3.9 per 1,000 in Mexico City, but just 1.3 in Chiapas. Mexico comes are reflected, for instance, in the gaps in the quality of City has a ratio of over 11 health care workers, physicians care for myocardial infarction mortality.136 Moreover, the re- and paramedics per 1,000 people, compared with less than sources that finance each insurance scheme are pooled sep- 4 per 1,000 in Chiapas. Only 10 of 31 states in 2015 had a arately, limiting the possibilities of distributing risks between nursing density above the national average. Regarding the rich and the poor. A significant proportion of resources private health care, the sector is small but increasing. Al- in the health sector goes to administrative and insurance though it still shows low penetration, the expenditure on costs (around 9 percent, the largest proportion in the OECD private health insurance is rising, at around 6-7 percent of countries). Additionally, Mexico has one of the highest out- the population. The private sector accounts for two-thirds of-pocket (OOP) share of total health care spending among of hospitals but only one-third of all beds (just 6 percent of OECD countries (41 percent versus the OECD average of 20.3 private hospitals have more than 25 beds). Private hospitals percent), which may signal a failure of the system to provide are concentrated in larger cities, and wealthier areas, which 134 See “Medición de la Pobreza: Evolución de las Dimensiones de la Pobreza, 1990-2016,” Consejo Nacional de Evaluación de la Política de Desarrollo Social (National Council for the Evaluation of Social Development Policy), Mexico City, https://www.coneval.org.mx/Medicion/Paginas/Evolucion-de-las-dimensiones-de-pobreza.aspx; “Medición de la Pobreza: SYSTEMATIC Evolución de las Carencias Sociales 2015 y su Comparativo con la Serie 2010-2014,” Consejo Nacional de Evaluación de la Política de Desarrollo Social (National Council for the COUNTRY Evaluation of Social Development Policy), Mexico City, https://www.coneval.org.mx/Medicion/EDP/Paginas/Datos-del-Modulo-de-Condiciones-Socioeconomicas.aspx. DIAGNOSTIC 135 González Block and Martínez (2015). MEXICO 136 Mortality rates within 30 days of hospitalization for cardiac infarction vary from 3 percent among high-income, mostly privately insured patients with access to private care to 6.5 percent among IMSS-insured patients but reach 59 percent among patients cared for elsewhere. This disparity is due largely to the fact that Seguro Popular only covers cardiac infarction cases in patients under age 60. Hospitals are often unwilling to devote substantial resources to treating uninsured patients, as they are unlikely to be reimbursed (World Bank 2013). 73 137 González Block (2015). 138 Mexico has the highest out-of-pocket share of total health care spending among OECD countries, which can be explained partially by the relatively low levels of government health spending (about 3 percent of GDP). 3. Inclusion Figure 65. Out-of-pocket share of total current spending on health, circa 2015 45 % of current health expenditure 40 35 30 25 20 15 10 5 0 OECD average Estonia Slovenia Germany Japan Denmark Norway Canada United Kingdom Czech Republic Ireland Sweden Turkey Iceland Austria Slovak Republic Australia Finland Italy Poland Spain Israel Portugal Switzerland Hungary Chile France United States Netherlands New Zealand Belgium Greece Korea Mexico Latvia Luxembourg Source: OECD Health Statistics 2017 (database), Organisation for Economic Co-operation and Development, Paris, http://www.oecd.org/els/health-systems/health-data.htm. Figure 66. Properties with sewerage service, by city size and type, 1990 and 2010 93% 93% 94% 94% 70% 66% 61% 63% 1990 2010 1990 2010 1990 2010 1990 2010 Small cities (>100k-500k) Medium cities (>500K-1m) big cities (>1m-10m) Mega cities (>10m) Source: Kim and Zangerling 2016. Figure 67. Infrastructure access in the Guadalajara Metropolitan Area, 2000 infrastructure index Guadalajara, Mexico - 2000 very high high Low veryLow No Data Sources: Kim and Zangerling 2016; data of National Institute of Statistics and Geography. tend to have a health status and profile closer to OECD av- social and productive inclusion. The coverage of public erages. services can vary greatly within cities, particularly as they continue to sprawl without corresponding extension of 121. Even though cities have reached near universal access infrastructure and service networks, with serious impli- to municipal services, issues in quality in the provision of ba- cations for the most vulnerable in the periphery (Figure sic services persist. Between 1990 and 2010, there has been 67). For instance, it has been shown that the adaptation a substantial expansion of basic services including access actions that households in marginal areas of Mexico City SYSTEMATIC to water and sewerage in cities of more than 100,000 in- implement to deal with water scarcity, low quality, and COUNTRY DIAGNOSTIC habitants (Figure 66). In addition, heterogeneity within flooding imply high financial and opportunity costs, with MEXICO cities in access to services such as water and sewage limits further consequences on poverty.139 74 139 Eakin et al. (2016). 3. Inclusion Figure 68. Percentage of workers that is informal by firm size, 2008 Economic Census Labor force survey 99.8 99.6 100 93.2 95.0 86.8 80 57.2 64.3 60 55.5 48.0 40 28.0 20 12.8 0 0-5 6-10 11-50 50+ Self- 2-5 6+ Agriculture employment and other rural Firm size/sectort Average private urban employment Average total employment Source: Levy et at (2012), based in Economic Census of 2008, and ENOE pressed the earnings’ distribution and lowered returns to 3.4 Decreasing labor education. This is less due to a lack of firms than to the na- productivity across the ture of the firms that are hiring, which are not demanding skills distribution higher skilled workers. Indeed, the percentage of wage/ salaried employees registered with the social security sys- 122. Low productivity of microenterprises poses concerns tem decreased between 2000 and 2012, and labor earnings for formal employment growth, more acutely in the south. fell.144 Earnings have declined since the Global Financial Although microenterprises are the most common firm crisis especially for workers in low-skilled services sectors category and source of employment, their productivity (Figure 69). is low and their contribution to the GDP is shrinking. The contribution of microenterprises with premises to GDP fell 124. The low productivity in agriculture remains a critical from 15 percent of GDP in 2008 to 9.8 percent in 2014.140 constraint to social and productive inclusion in Mexico. Most Lagged regions are particularly vulnerable. Labor produc- agricultural producers in the country (over 75 percent) tivity in the Southern states is about 53 percent of the are smallholders, having less than five hectares each and national average, while the economic base is dominated semi-subsistent, employing traditional, rain-fed produc- by micro enterprises, in low productivity activities.141 Low tion practices, and concentrated in the Center and South- skill levels and less skill upgrading opportunities further ern parts of the country. These producers typically have reduce the Southern firms’ performance. Women in Mexi- no access to improved seeds, irrigation, access to credit, co, on average, are about 56 percent less likely to be entre- insurance, technology, or marketing infrastructure. About preneurs in the formal sector and 63 percent more likely 70 percent of Mexican agriculture is still harvested through to be entrepreneurs in the informal sector, compared with manual labor; while less than 20 percent of croplands are men.142 Most of the workers in micro-firms are informal irrigated, leaving crops dependent on seasonal rains or (93.2 percent economic census, and 96.6 percent from the irrigation through mobile water pumps.145 There is an as- labor force survey) (Figure 68). sociation, indeed, between poverty and worst inputs for agriculture (such as in terms of irrigation, credit) (Figure 70). 123. Factor misallocation helps to explain why increases in human capital over the past two decades have not necessarily translated into higher wages. Resources are being pushed 3.5 Low female and youth toward firms that hire workers through non-salaried con- tracts or with illegal salaried contracts (informal). Even labor force participation though workers are coming into the labor market with 125. Mexico still has one of the lowest rates of female labor more years of schooling, the workplace is not fully valuing participation among OECD and regional peers. Only 45.5 their additional education.143 There is evidence of declining percent of working age Mexican women are part of the earnings for workers with more schooling that have com- labor force, compared with an average of 53 percent for SYSTEMATIC COUNTRY DIAGNOSTIC 140 INEGI (2017a). MEXICO 141 Deichmann et al. (2004). 142 Fareed et al. (2017). 75 143 Levy and López-Calva (2017) 144 Cruces et al. (2015). 145 ITA (2017). 3. Inclusion Figure 69. Wage index for selected workers, 2006–16 (male age 15-yo-64) 110 105 100 Index (Q3 2008 = 100) 95 90 85 80 75 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 Q1 Q3 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Agriculture + Low Tech Manufacturing high Tech Manufacturing + Construction Commerce 4 per. Mov. Avg. (Agriculture + Low Tech ind) 4 per. Mov. Avg. (high Tech ind + Construction) 4 per. Mov. Avg. (Commerce) Source: World Bank based in ENOE Figure 70. Poverty rates and irrigation by state, 2016 80 Chiapas Oaxaca Guerrero veracruz Puebla 60 Tlaxcala Michoacán de Ocampo hidalgo México Morelos Tabasco Poverty 2016 Zacateca San Luis Potosí Guanajuato Campeche Durango 40 Yucatán Nayarit Quintana roo Chihuahua Colima Tamaulipas Sonora Jalisco Aguascalientes Sinaloa Ciudad de México baja California Querétaro Coahuila de Zaragoza 20 Nuevo León baja California Sur 0 0 .2 .4 .6 .8 1 irrigated area as a proportion of cultivated area Sources: INEGI 2016a, 2016b; 2016 data of CONEVAL. Latin America and Caribbean countries, 51 percent for the the labor market: if working-age women that are not par- OECD, and levels of around 58 percent in Colombia, and ticipating in the labor market were to do so at the same 52 percent in Chile – two regional peers whose levels of fe- rates as their male counterparts there would be a gain of male labor force participation in the 1990s were similar or 25 percent of Mexico’s GDP per capita.147 This is one of the lower to those of Mexico (Figure 71).146 A significant driver largest average total missed gains, comparable to 33 per- of female LFP is education; yet, in Mexico, there is still an cent in Turkey and 21 percent in Italy. important proportion of women with tertiary education who remain outside the labor market. 127. Multiple barriers hinder women’s entry into the labor force. In Mexico, women shoulder nearly 77 percent of all 126. On average, the loss associated with the gender gap in unpaid housework. The average woman spends six hours labor market participation for Mexico is around 25 percent each day doing unpaid housework, compared with an of income per capita. This inequality in labor participation average of two hours for men. A large household labor comes at a high cost. A recent study has estimated a sig- burden presents a serious challenge for women attempt- nificant GDP per capita loss from existing gender gaps in ing to attend school or work a full- or even part-time job. SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 146 Notwithstanding that Mexico has one of the weakest female labor market performance in the region, the latest evidence shows signs of a positive trend. A recent study finds that the difference in labor force participation rates by gender are narrowing (from 38 percentage points before the global financial crisis of 2008 to 35 points in 2017), while the unem- 76 ployment rate gender gap is closing. A higher propensity of women to be employed after the crisis is found to be partly related to an increase in the availability of daycare facilities (Cardozo et al. 2018). 147 Cuberes and Tiegnier (2016). 3. Inclusion Figure 71. Evolution of female labor force participation, 1992–2016 65 60 55 % of female population ages 15+ 50 45 40 35 30 25 20 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 Mex bra Chl Col OECD LAC Source: 2017 data of WDI (World Development Indicators) (database), World Bank, Washington, DC, http://data.worldbank.org/products/wdi. Figure 72. Gender participation gap across the life cycle, 2016 35 30 25 25 20 15 14 10 5 0 Czech Republic Denmark Slovenia Germany Latvia Netherlands Estonia Switzerland Hungary Poland Slovak Rep. France Israel Australia United Kingdom Italy Colombia Costa Rica 15-19 years-old 20-24 years-old Source: World Bank calculations based on ILOSTAT and ENOE 2016. Moreover, almost 30 percent of employees in Mexico 20 years later.150 The negative income effect of not being work long hours (more than 40 hours in a usual week),148 engaged in education or work also harms equity since far above the OECD average of 13 percent, which com- close to 60 percent of the population NEET in Mexico is in pounds the challenge of balancing multiple obligations the bottom 40 percent of the income distribution. 151 and acts as a barrier to women’s entry in the labor mar- ket.149 129. Mexico also experiences one of the biggest gaps be- tween male and female NEET, and has the highest adoles- 128. A large share of Mexico’s youth is not in employment, cent pregnancy rate in the OECD. The majority of youth education or training (NEET). Twenty-five percent and 14 NEET are urban women. Mothers in particular, face high percent of the 20-24-year olds and the 15–19 year olds barriers to paid work. This has implications not only for population are NEET (Figure 72). For any given cohort, a 1 gender equality, but also for children. Maternal employ- percentage point increase in the proportion of youth NEET ment is strongly negatively correlated with child poverty predicts a 7 percent reduction in earnings for that cohort across countries.152 SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 148 According to OECD statistics, in Mexico the average usual weekly hours worked is 45.6, one of the highest among OCDE and Latin America and Caribbean countries. 149 OECD, Better Life Index, 2016. 77 150 De Hoyos et al. (2016). 151 De Hoyos et al. (2016). 152 OECD (2017) 4. Key sustainability challenges 130. Sustainable growth requires balancing the needs of 132. Nevertheless, tax collection in Mexico remains below present and future generations. Fiscal, environmental and the regional average and is the lowest among OECD coun- social sustainability are central to maintain high and sus- tries. International comparisons of tax collection employ a tained rates of inclusive growth. This section identifies key broad concept of tax revenue, which includes social secu- constraints from this medium to long term perspective. rity contributions and subnational taxes. Under this defini- tion, Mexico’s tax-to-GDP ratio increased from 13.4 percent in 2010 to 17.2 percent in 2016. However, during this peri- 4.1 Fiscal sustainability od average tax-to-GDP ratios also increased among other countries in the region, and Mexico’s tax burden remains 131. In 2013, a comprehensive tax-reform program signifi- below the Latin America and the Caribbean (LAC) regional cantly increased nonoil tax revenue. The reforms eliminat- average of 22.7 percent. Mexico’s tax-to-GDP ratio is also ed tax benefits and preferential tax regimes, limited tax the lowest among OECD countries, amounting to just half deductions, introduced dividend and capital gains taxes, the OECD average of 34.3 percent (Figure 73). raised the top marginal income tax rate, and replaced the tax regime for small businesses. The reforms also eliminat- 133. Even after the positive 2013 reforms, extensive tax ed a preferential VAT rate that had been applied in border expenditures continue to impose major fiscal costs. Tax ex- regions. They introduced new excise taxes on carbon, sug- penditures include exemptions, deductions, deferrals, and ar-based beverages and high-calorie foods. The negative preferential rates applied to specific activities or types of excise tax on domestic fuel sales was phased out in the taxpayers. Forgone revenue through tax expenditures context of low international oil prices and energy-sector is estimated at 3.7 percent of GDP. Tax expenditures are reforms. The federal government’s non-oil tax revenue rose incurred, inter alia, through exemptions and zero-rating from 10 percent of GDP in 2013 to 13.1 percent in 2017. in the VAT regime (1.5 percent), since 2017, through a dis- Rising income tax collection contributed 1.4 percentage count to the excise tax on fuel (0.7 percent), exemptions SYSTEMATIC points of GDP to the increase in non-oil tax revenue, fuel to wage income (0.4 percent of GDP) and pension income COUNTRY excise taxes contributed 1.0 percentage point, VAT con- (0.2 percent), a negative income tax designed to encour- DIAGNOSTIC tributed 0.3 percentage points, and other excise taxes age formal employment (0.2 percent). MEXICO contributed the remaining 0.2 percentage points. The total 79 increase in non-oil tax revenue largely offset a decline in oil 134. In addition, tax evasion continues to inflict significant revenue equal to 4 percentage points of GDP. revenue losses, although they have been declining in recent 4. KEY SUSTAINABILITY Figure 73: Tax structures in Mexico, Latin America and the Caribbean and the OECD (2016) CHALLENGES % of GDP % of total tax revenue 35 100 2.9 5.8 6.3 7.7 30 14.7 12.4 4.2 21.2 80 25 1.6 6.7 23.9 20.0 20 60 23.9 1.2 5.2 9.0 13.9 25.8 15 2.4 6.2 40 4.1 15.9 10 2.2 3.7 11.5 20 41.7 34.1 5 7.3 27.3 6.0 0 0 Mexico LAC OECD Mexico LAC OECD Other taxes Other goods and services vAT Social secutiry income Source: OECD et al. (2018). years. These losses were estimated to be close to 2.5 per- are obliged to collect VAT on their sales, they may be at a cent of GDP in 2016, of which income tax evasion amount- pricing disadvantage vis-à-vis non-resident competitors. ed to 1.5 percent of GDP and VAT evasion amounted to 1.0 percent. However, tax evasion appears to have fallen 136. At the same time, Mexico’s federal public sector ex- significantly over the past few years, likely due to improve- penditures have increased substantially over the years, and ments in tax administration. pressures will continue to mount. The country experienced a secular increase in public spending driven by a combina- 135. Technological progress poses the new challenge of tion of policy decisions, demographic trends, legacies, and including a growing digital economy in the country’s tax other spending rigidities (Figure 74). This has resulted in base. Following global trends, Mexico’s digital economy is a decade-long rise in health, education, social protection, growing rapidly, supporting economic development more and public security spending. The number of people aged broadly. Although from a low base, between 2010 and 65 and older will more than double to 18.4 million over 2016, the share of the adult population that had ordered the next two decades and pension costs on the federal goods or services online increased by a factor of 4.5, the public-sector budget, due to this increase in the number largest such increase of all OECD countries. Meanwhile, it of elderly combined with the favorable treatment of this has been estimated that Mexico’s business-to-consumer cohort in past pension reform, is projected to increase (B2C) e-commerce market could grow to US$40.8 billion by about 1 percent of GDP per decade before ultimately by 2019, of which nearly two-thirds would consist of digi- peaking between 2040 and 2050. Population aging is also tal services.153 New market entrants armed with disruptive intensifying pressure on Mexico’s public healthcare sys- technologies can increase the choice, improve the qual- tem by shifting the disease burden to chronic degenera- ity, and reduce the price of inputs to firms and of goods tive diseases, with population aging accounting for nearly and services to final consumers in Mexico. Coupled with 60 percent of the increase in public health spending to 3.1 rapid increases in fixed internet and mobile broadband percent of GDP in 2015 and an increase by 0.5 percent of penetration, these market dynamics mean that more and GDP over the previous decade. These long-term trends un- more consumers are in a position to reap the benefits of derscore the importance of improving the efficiency and the digital economy. Typically, digital services consumed effectiveness of public expenditures. in Mexico are subject to VAT, whether provided by a for- eign or domestic supplier, with the obligation to collect 137. Expenditure rationalization over the past few years took and remit the VAT on the seller or the resident importer place across several categories, including public investment. in the case of purchases from abroad. Since in the case of Subsidies, transfer programs and the public sector wage digital services there may be no business need for foreign bill were all affected by spending containment and cuts. suppliers to establish a physical or legal presence in the However, general government public investment (includ- country, there is not necessarily an entity that charges ing federal investment projects and transfers for state and and remits the VAT owed on digital services imported by municipal investments) was affected significantly, drop- SYSTEMATIC final consumers, and that revenue is effectively foregone ping from 4.7 percent of GDP in 2014 to 2.6 percent in 2017 ( COUNTRY DIAGNOSTIC by the Mexican authorities. Moreover, since Mexican-res- Table 4). These cuts also included PEMEX capital expendi- MEXICO ident firms selling digital services in the domestic market tures, potentially affecting its future production capacity. 80 153 2016: The Tipping Point for E-Commerce in Mexico, A.T. Kearney, 2016. Table 4: Public Expenditure, 2014-2017 (% of GDP) 4. KEY SUSTAINABILITY CHALLENGES   2014 2017 Change Total 25.9% 23.8% -2.1% Wages and salaries 5.8% 5.3% -0.6% Other operating costs 5.8% 5.7% -0.1% Subsidy and transfer programs 3.8% 3.1% -0.7% Physical investment 4.7% 2.6% -2.1% Other capital expenditures 0.4% 1.0% 0.6% Participaciones 3.3% 3.5% 0.2% Interest payments 2.0% 2.5% 0.5% Source: Bank staff estimates based on SHCP and INEGI Figure 74. Public sector revenue and expenditure, 2000–17 (percent of GDP) 28 26 24 22 20 18 16 2000 2002 2004 2006 2008 2010 2012 2014 2016 Expenditure Average revenue Average Source: World Bank estimates based on data of the INEGI; SHCP. Figure 75. Federal public sector debt, 2004–17 (percent of GDP) 65% 55% 45% 35% 25% ! 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Net Public Sector Debt Gross Public Sector Debt Source: INEGI; WEO (World Economic Outlook Database), International Monetary Fund, Washington, DC, https://www.imf.org/external/pubs/ft/weo/2016/01/weodata/index.aspxn. SYSTEMATIC COUNTRY DIAGNOSTIC 138. Overall, fiscal policy in Mexico has been prudent and a indicated the need for a gradual fiscal consolidation to sta- MEXICO gradual consolidation process has been underway with the bilize debt and start downward-sloping path (Figure 75). aim of stabilizing and then reducing public debt levels. The Under certain assumption Mexico’s current debt-stabiliz- sustained increase in the public debt burden up to 2016 ing primary surplus would be equal to at least 0.5 percent 81 4. KEY of GDP. In terms of the overall balance, assuming an aver- peers, following Turkey (26 percent), Peru (16 percent) and SUSTAINABILITY CHALLENGES age nominal GDP growth rate of 6 percent, the debt-stabi- Brazil (15 percent). Of course, natural capital also includes lizing PSBR would equal 2.5 percent of GDP. non-renewable resources such as oil, gas, coal and miner- als. 139. Moreover, the country has established a solid fiscal framework, though some adjustments to the framework 142. Poorly managed natural capital can represent a social, may signal a continued strong commitment to fiscal dis- environmental and economic liability. Deforestation and cipline. The 2014 LFPRH called for an annual numerical forest degradation are among the principal drivers behind target and an indicative multi-year path for the PSBR to the loss of natural capital in Mexico, with agricultural ex- complement the traditional budget balance. Without pansion playing an important role. Mexico’s forests repre- changing the current framework, a positive development sent an essential source of employment, income and live- would be to establish the fixed numerical cap of the PSBR lihood for approximately 12 million people. Despite their (consistent with a downward-sloping debt path) in legisla- strategic environmental and social role, forests in Mexico tion, as a signal of a continued strong fiscal policy anchor are subject to high degradation. Forest cover155 has fall- moving forward. This will continue to be complemented en by 5.4 percent in relation to 1990, with more than 3.7 with the existing escape clause, which could tighten its million net hectares lost over that period. However, these criteria. A deeper adjustment to the fiscal rule framework, figures mask the heterogeneous nature of deforestation with the aim to enhance its pro-cyclicality while preserv- and forest degradation patterns, as their incidence is ing sustainability would be setting a combination of an significantly higher in some types of ecosystems, such as adjusted expenditure rule (i.e, some adjustments to the tropical dry forests and tropical rain forests. existing expenditure rule) and an explicit ceiling on the public debt-to-GDP ratio. This could be complemented 143. Mexico’s forests represent an important natural asset with corrective measures at certain thresholds of debt to for the country, particularly for rural communities. Forests GDP below the ceiling. cover 45 percent of the national territory, and 61 percent of forests are the communal property of ejidos and commu- 140. Mexico’s fiscal framework could be complemented by nities. As discussed in Section 3, it is estimated that forests the establishment of an independent fiscal council that pro- are home to more than 12 million people, 88 percent of motes sound fiscal policies through independent oversight. whom live in highly marginalized localities156, and directly Fiscal councils have been adopted by different countries depend on local natural resources. More than 1.5 million around the world in addition to fiscal rules over the past of this population is indigenous, 62 percent live in poverty, few decades as an institutional device to strengthen the and more than half of all forest dwellers live in conditions credibility of governments’ commitment to sustainable of extreme poverty, with limited access to health services, public finances. Despite a diversity of institutional arrange- education, and accessible forest tracks. In 2010, the per- ments, a fiscal council is generally defined as a permanent centage of inhabited private dwellings without sewage agency with the mandate to monitor publicly and inde- systems was four times higher in forested areas than in the pendently government’s fiscal policies against macroeco- rest of the country.157 nomic objectives related to long-term fiscal sustainability and medium-term macroeconomic stability. The design 144. Mexico faces the challenge of conserving and sustain- features of these institutions are critical to ensure trans- ably managing its forests while also meeting a growing de- parency, independence, and sharp focus on the objectives mand for timber products. In 2015, forestry production in established, while avoiding policy-making interference. Mexico accounted for 0.6 percent of the GDP. The country has the capacity to more than double the current produc- tion of standing timber; in the areas that are most accessi- 4.2 Environmental ble, Mexico has the potential to produce about 60 million sustainability (in cubic meters) of roundwood. Yet, demand exceeds production by a factor of three. The increase in communi- 141. Natural capital in Mexico includes agricultural soils and ty-based forest enterprises (CFEs) has not been translated pastures; water; forests; fisheries; air; strong winds and solar in an increase in timber production. Instead, legal timber potential; and subsoil assets (oil, gas, coal, and minerals). production fell from 9.4 million cubic meters in 2000 to 6.1 Conservative estimates154 suggest that renewable natural million, even as domestic demand for timber increased. capital, captured in the value of agricultural land, forest This trend is due in part to the failure of sectoral incen- land and protected areas (thus excluding, given lack of tives to balance conservation and economic objectives. data, many other resources such as fisheries) represents In 2015, with almost 1,500 harvesting permits granted for about 14 percent of Mexico’s total wealth. This puts Mexi- forest exploitation in the country, extraction occurred in SYSTEMATIC COUNTRY co on the top quartile of the list of regional and structural only 46 percent of the approved forest land area. Mexico’s DIAGNOSTIC MEXICO 154 The Changing Wealth of Nations (2018). 155 This estimate is based on 2015 Forest Resource Assessment (FRA) data. FRA’s forest definition: Land spanning more than 0.5 hectares with tree higher than 5 meters and a canopy 82 cover of more than 10 percent, or trees able to reach these thresholds in situ. It does not include land that is predominantly under agricultural or urban land use. 156 World Bank (forthcoming). 157 World Bank (forthcoming). 4. KEY Figure 76. Water stress and availability, by river basin, 2015 SUSTAINABILITY CHALLENGES % 160 140 120 100 80 60 40 20 0 valle de México río bravo Noroeste baja California Cuencas Centrales del Norte balsas Lerma-Santiago-Pacífico Pacífico Norte Golfo Norte Península de Yucatán Golfo Centro Pacífico Sur 0 5000 10000 15000 20000 25000 (m3/capita) Water availability (lower axis) Water stree (upper axis) Source: Sistema Nacional de Información de Agua http://sina.conagua.gob.mx/ CFEs exploitation costs (US$39 per cubic meter) have been such as patrolling, building fire breaks, or promoting soil too high compared with other countries, especially low- conservation—by around 50 percent.162 There is evidence cost producers like Chile and Brazil (US$7–US$15 per cubic that the PSAH program generated an approximate 20 per- meter), or even small private landowners in the U.S. south cent reduction in expected forest cover loss at the locality and Pacific northwest (US$12 per cubic meter).158 level between 2000 and 2012, when controlling for char- acteristics affecting selection and deforestation risk; while 145. Mexico’s payments for conservation and environmen- creating a small impact on poverty alleviation.163 As a com- tal services, one of the largest national programs of its kind, parison, traditional conservation policies, such as protect- has shown positive results in curbing deforestation and pov- ed areas establishment led to a reduction of 24 percent of erty alleviation. Mexico’s federal payments for hydrological forest loss over the same period, with no effect on poverty services program (PSAH) began in 2003, paying landown- alleviation.164 Other studies have questioned whether hav- ers to maintain forest cover under five-year contracts.159 ing the double objective of deforestation and poverty has Between 2003 and 2011, the Mexican National Forestry made the PSAH program less effective.165 Commission (CONAFOR) allocated approximately US$450 million to enroll more than 2.6 million hectares of land in 146. More than 35 million Mexicans have limited access the program. By 2013, over 4.27 million hectares had been to water or receive low-quality water services. More than enrolled into the scheme, benefiting a total of 7,350 com- a hundred of the country’s 731 watersheds face severe mon- or private-property lands, and representing an in- shortages (Figure 76), and the number of overexploited vestment of Mex$8,586 million.160 Between 2013 and 2016, aquifers tripled between 1975 and 2013.166 Overall, water CONAFOR’s payment for environmental services program availability per capita has reduced drastically, from 18,035 destined Mex$3,355 million pesos for the conservation of to 3,982 m3/inhabitant per year between 1950 and 2013. 1.55 million hectares.161 The financial incentives have been Groundwater provides more than 65 percent of all water found to increase land cover management activities— used by Mexican cities. Pumping out underground water 158 Cubbage et al. (2015). 159 The PSAH, managed by CONAFOR, targets the peasant communities of ejidos in sites with hydrological importance. In 2006 the objectives of the program were modified to include poverty alleviation. To secure payments under the five-year contracts, landowners agree to carry out a series of conservation activities such as surveillance, fence establishment, or invasive species extraction (DOF 2013). 160 CONAFOR (2014). 161 CONAFOR (2016). 162 Alix-García et al. (2018). The study also finds that increases in paid activities do not lead to a decline in the contribution of households to unpaid land cover work, and that community social capital rose by 8 to 9 percent. 163 Sims and Alix-Garcia (2017). Using different metrics (unit of analysis, outcome variable, and time frame) Alix-Garcia et al. (2015) find that PSAH reduced the downward trend in forest SYSTEMATIC cover by 40-50 percent, and small but positive effects on poverty. Less impacts on poverty have been found by other studies (Rico García-Amado, et al. 2011; Alix-Garcia, et al. 2012). COUNTRY 164 Sims and Alix-Garcia (2017). DIAGNOSTIC 165 Shapiro (2013) notes that the program has become hybridized, as the targeting has been weighed toward marginalized communities and away from risk of deforestation; while its MEXICO evolution has also been influenced through contestation and engagement with social movements. Alatorre-Troncoso (2014) suggests that in the quest to target poverty alleviation and conservation, neither objective has been adequately fulfilled: implementing the programme in extreme biological importance locations would have achieved protection of the same area (4.3 million hectares) at a fraction of the cost (18.5 percent). The study also notes that most lands enrolled in 2010 were in municipalities with a Medium Human Develop- 83 ment Index, while low HDI areas were only marginally represented, despite being the most extensive. 166 Overexploited aquifers (where groundwater provides more than 65 percent of the volume required by cities) increased from 32 in 1975 to 126 in 2013, out of a total of 653. See Siste- ma de Cuentas Nacionales de México: Cuentas Nacionales Ambientales (database), INEGI, Aguascalientes, Mexico, http://www.inegi.org.mx/geo/contenidos/recnat/default.aspx. 4. KEY SUSTAINABILITY Figure 77. Trends in costs of water depletion and degradation in Mexico CHALLENGES (2003 = 100, current prices) 200 180 160 140 120 100 80 60 40 20 0 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Costs of water scarcity (2003=100) Depletion costs of water resources (2003=100) Source: Sistema de Cuentas Nacionales de México: Cuentas Nacionales Ambientales (database), INEGI, Aguascalientes, Mexico, http://www.inegi.org.mx/geo/contenidos/recnat/ default.aspx. Figure 78. GDP adjusted to environmental and water resources (2003 = 100, current prices) 80.5 1.00 80.0 0.90 79.5 0.80 79.0 0.70 78.5 0.60 78.0 0.50 77.5 0.40 77.0 0.30 76.5 0.20 76.0 0.10 75.5 0.00 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Percentage of GDP adjusted with environmental degradation and depletion Percentage points of GDP lost due to water resources depletion (right axis) Source: Sistema de Cuentas Nacionales de México: Cuentas Nacionales Ambientales (database), INEGI, Aguascalientes, Mexico, http://www.inegi.org.mx/geo/contenidos/recnat/ default.aspx. causes land subsidence (or sinking), which makes flooding America—i.e. the lowest availability among all mainland worse and significant structural damages to urban infra- countries in the region. Given the relatively limited volume structures. Subsidence levels in Mexico City are observed of renewable water, highly populated regions (like Mexi- over 30 cm/yr.167 co’s City Valley) must increase efficiency and resilience of current water exploitation and distribution infrastructures. 147. The costs of water depletion and degradation have in- Yet, inefficiencies in water resource management such as creased over the last 15 years. Between 2003 and 2015, wa- overexploitation, pollution, leakages, lack of (or inefficient) ter depletion and degradation costs increased by 60 per- infrastructure, unfulfilled legal and institutional reforms, cent, from 0.5 to almost 0.8 percent of GDP (Figure 77 and and insufficient investments, have put additional pressure Figure 78). In addition to these costs, the value of assets on water resources, already under stress from rapid urban- lost due to soil and water contamination has increased ization, population and economic growth. Furthermore, 10-fold over the last 15 years.168 At the same time, Mexico water scarcity and climate uncertainty combined with SYSTEMATIC COUNTRY is endowed with only a fifth of fresh water resources per generally poor water services represent a serious threat to DIAGNOSTIC capita compared with the average endowment in Latin green growth prospects in Mexico. MEXICO 84 167 Chaussard et al. (2014). 168 Sistema de Cuentas Nacionales de México: Cuentas Nacionales Ambientales (database), INEGI, Aguascalientes, Mexico, http://www.inegi.org.mx/geo/contenidos/recnat/default. aspx. 4. KEY Figure 79. Proportion of households with access to sewerage SUSTAINABILITY CHALLENGES 80% 70% 60% 50% 40% 30% 20% 10% 0% 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 Mexico average urban rural Source: Elaborated based on 2018 data of SEDLAC (Socio-Economic Database for Latin America and the Caribbean), Center for Distributive, Labor, and Social Studies, Facultad de Ciencias Económicas, Universidad Nacional de La Plata, La Plata, Argentina, and Equity Lab, Team for Statistical Development, World Bank, Washington, DC, http://sedlac.econo.unlp. edu.ar/wp/en/estadisticas/sedlac/estadisticas/. Figure 80. Proportion of households with access to piped water. Lowest and highest quantiles of income 100% 80% 60% 40% 20% 0% 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 Mexico average bottom 20 Top 20 Source: Elaborated based on 2018 data of SEDLAC (Socio-Economic Database for Latin America and the Caribbean), Center for Distributive, Labor, and Social Studies, Facultad de Ciencias Económicas, Universidad Nacional de La Plata, La Plata, Argentina, and Equity Lab, Team for Statistical Development, World Bank, Washington, DC, http://sedlac.econo.unlp. edu.ar/wp/en/estadisticas/sedlac/estadisticas/. 148. The uneven spatial distribution of water and the as- of the urban population, and are partly to blame for en- sociated availability in terms of both quantity and quality vironmental challenges such as water pollution and over- impacts Mexico’s potential for growth and inclusion, with a exploited aquifers. The north-center region of the coun- strong connotation of regional, and urban/rural inequality. try, with only 32 percent of the water resources, houses In 2014, 73 percent of households in urban settlements 77 percent of the Mexican population and contributes had access to sewerage services, compared with 21 per- to over 80 percent of its GDP. In contrast, the south and cent of rural ones (Figure 79). Due to subsidized tariffs, south-east region has 68 percent of the water resources water payments in urban areas range from 0.25 to 2.19 but only 33 percent of Mexico’s population, contributing percent of household’s income, while in rural areas, the about 21 percent of GDP. The north-center region, despite poorest are paying a higher proportion of their income for its low natural water endowment, provides most of its water services. And despite improvements in recent years, inhabitants with water and sanitation services, while also SYSTEMATIC disparity in coverage access across income groups remain: being responsible of a thriving agricultural sector—partly COUNTRY 98 percent of households in the top quintile had access resulting from both a largely inefficient irrigation sector DIAGNOSTIC to piped water compared with 86 percent of households (which consumes almost 70 percent of water available), MEXICO in the poorest quintile (Figure 80). The Government’s and the overexploitation of underground water sources 85 incentive schemes, such as Apoyos Compensatorios and (with limited fines to polluting industries). Conversely, the PROAGRO, tend to favor agricultural interests over those south region, rich in natural and hydric resources, fails at 4. KEY SUSTAINABILITY Figure 81. Number of cars per capita by city size, 1990–2013 CHALLENGES 0.60 0.56 0.50 0.39 0.39 0.40 0.37 0.30 0.20 0.18 0.17 0.15 0.10 0.2 0.00 1990 2013 1990 2013 1990 2013 1990 2013 Small cities (100k-500k) Medium cities (500K-1m) big cities (1m-10m) Mega cities (>10m) Source: Kim and Zangerling 2016. Figure 82. Air pollution – micrograms per cubic meter 100 80 Micrograms per cubic metre Thailand, 28.9 Korea, 32.0 60 Poland, 23.4 romania, 18.2 Turkey, 21.8 Argentina, 11.1 Mexico, 13.3 Chile, 17.2 Colombia, 8.4 brazil, 11.3 40 20 0 Qatar Saudi Arabia Mali Djibouti Azerbaijan Rwanda Uganda Dem. Rep. of the Congo South Africa San Marino Ukraine Austria Malta Namibia Brazil Denmark Colombia Dominican Rep. Finland Micronesia Timor-Leste Exposure to PM2.5 fine particles, 2015 Source: OECD 2018, “Air quality and health: Exposure to PM2.5 fine particles - countries and regions” providing piped water and sewerage to about 15 percent increased losses of productivity as well as higher premature of its population, despite large government investments mortality and morbidity. These trends are in line with those in the sector and water distribution capabilities to semi-ur- found by other international studies. Recent estimates ban and rural areas. suggest that 39,000 people died in Mexico in 2013 of dis- eases associated with environmental degradation; while 149. Environmental degradation in urban areas has direct environmental health risks represented 2.5 – 3.6 percent of impacts on the economy due to the loss of productivity asso- Mexico’s GDP in 2013.170 Around 83 of the deaths recorded ciated with health costs. The main contributors to environ- were associated with outdoor and household air pollution. mental degradation include: outdoor and indoor air pol- lution; water pollution; soil degradation; and inadequate 150. Urbanization combined with a weak delivery of urban hazardous and solid waste management.169 Between 2003 environmental services (water supply, sanitation, solid waste and 2016, environmental investment and expenditures management among others) can reduce human and natural increased from 0.5 percent to 0.7 percent of GDP. Despite capital. Approximately 80.2 percent of Mexico’s population these positive trends, environmental degradation and lives in urban areas171 which produce more than 84 per- SYSTEMATIC natural resources depletion continue to grow, resulting in cent of the GDP, as well as an estimated 75 percent of GHG COUNTRY DIAGNOSTIC MEXICO 169 Some estimations of the costs of environmental degradation and depletion of natural resources amounted to 4.7 percent of GDP in 2016. See Sistema de Cuentas Nacionales de México: Cuentas Nacionales Ambientales (database), INEGI, Aguascalientes, Mexico, http://www.inegi.org.mx/geo/contenidos/recnat/default.aspx. 86 170 World Bank (2015). 171 World Urbanization Prospects: The 2018 Revision (database), Population Division, Department of Economic and Social Affairs, United Nations, New York, https://esa.un.org/unpd/ wup/. 4. KEY Figure 83. Top 20 economies according to their carbon dioxide emissions, tons, millions, 2015 SUSTAINABILITY CHALLENGES 10000 9000 8000 7000 6000 5000 4000 3000 Mexico, 442 Poland, 282 Turkey, 317 Korea, 586 Brazil, 451 2000 1000 0 China Indonesia USA India Russia Japon Germany Korea Iran Canada Brazil Mexico UK Australia Italia Turkey Saudi Arabia South Africa Francia Poland Source: OECD 2018a. emissions. Environmental issues are largely related to the 152. Regarding the efficiency of domestic logistics and challenges of rapid urbanization processes, including the competitiveness, most products (56 percent of domestic accelerated growth of a vehicle fleet that disproportion- cargo)173 are carried by road-based freight transport, also ately contributes to air pollution; insufficient wastewater known as auto-transport. Freight transport is one of the treatment; and inadequate disposal of solid waste (Figure least-efficient freight transport modes in terms of energy 81). Solid waste collection is deficient in Mexican cities: only use and emissions (accounting for over 90 percent of total 83 percent of the solid waste produced is collected as com- freight-related emissions in the country).174 Moreover, die- pared with the OECD standard of 98 percent. The urban sel-fueled heavy-load vehicles account for 51 percent of poor, on the other hand, are disproportionately affected by PM10 emissions, PM2.5 emissions, and NOx emissions.175 exposure to air, water, soil and chemical pollution. As dis- These emissions have impacts on public health, especial- cussed above, air pollution continues to be a major environ- ly among the most vulnerable groups (children, the sick mental concern of the population and imposes significant and the elderly). However, freight transport is a strategic costs on the economy (Figure 82). sub-sector of Mexico’s economic activity: it represents 4.9 percent of the national gross domestic product (GDP) 151. The land use and transport model in Mexican cities is and generates more than 2 million direct jobs.176 Since increasingly dependent on cars, with implications on the 2010, SEMARNAT and the Ministry of Communications environment and equitable access. While public transit re- and Transportation (SCT) has supported a Clean Trans- mains the main mode of transport in Mexico’s urban areas, port Program to reduce these air-quality externalities in with an average of 60 percent of total trips, 90 percent still the freight transport sector. Despite a low level of partic- use conventional low-quality bus service. The mass-transit ipation (375 participating companies177 corresponding to systems (i.e., Bus Rapid Transit - BRT, metros and Light Rail roughly 7 percent of the entire fleet – baseline 2017)178179, Transit - LRT) implemented or upgraded recently in large official figures show a potential reduction of more than cities still represent less than eight percent of the total 7 MtCO2 since the program’s inception in 2010.180 There transit modal share. A factor behind the increasing depen- is great potential to improve awareness of the program, dency on cars is related to the expansion of new real-estate provide more incentives for adhesion and build capacity developments that are disconnected from urban centers, for monitoring and evaluation to attain greater emission resulting in low transport accessibility to opportunities for reductions in the sector. new low-income dwellers.172 The strategies behind urban road and parking infrastructure further reinforce the trend 153. Mexico is highly vulnerable to climate change, which through high implicit subsidies for private car users and can exacerbate the country’s development challenges. While a regressive policy in terms of budget and public-space Mexico is a global climate change leader and has assigned allocation in cities. a political priority to tackling emissions, it is the world’s SYSTEMATIC 172 See further details in Section 4: Coordination failures and limited local capacity for long-term planning. COUNTRY 173 Secretaría de Comunicaciones y Transporte (2016). DIAGNOSTIC 174 Solís Ávila y Sheinbaum Pardo (2016). MEXICO 175 INECC (2017). 176 Secretaría de Comunicaciones y Transporte (2016). 177 Secretaría de Medio Ambiente y Recursos Naturales (2017). 87 178 https://canacar.com.mx/stat/parque-vehicular-clase-vehiculo/ 179 Secretaría de Medio Ambiente y Recursos Naturales (2017). 180 Ibid. 4. KEY 12 highest emitter of greenhouse gases (Figure 83). This tion. Mexico also has access to catastrophe insurance and SUSTAINABILITY CHALLENGES large share of GHG emissions stem from the energy sector, has issued Cat-bonds to transfer part of the natural disas- which correlate with both the sector’s size and its reliance ter-related financial risk to the private sector. Use of these on fossil fuels (a total share of 67 percent in 2010), fol- instruments is based on solid risk assessments, including lowed by transport and agriculture. Mexico also remains the National Atlas of Risk and modelling of catastrophic highly vulnerable to extreme weather events. events. 154. Extreme weather events have adverse effects on pov- 156. Mexico also has a comprehensive and evolving in- erty.181 Research using the Human Development Index stitutional framework to address climate change. Key in- indicates that weather events could reduce up to two-year struments include: (i) the 2012 General Law on Climate achievements in the HDI index for affected municipalities Change (LGCC), which established a series of non-bind- in Mexico.182 Further, floods impacts are long lasting, while ing, aspirational goals and defined the scope of Mexico’s current social protection strategies often fail to mitigate national climate change actions, the responsibilities of their negative effects. Evidence shows that El Niño South- federal, state and local governments, and the institution- ern Oscillation (ENSO) had important negative impact on al arrangements to meet the law’s objectives187; (ii) the the development of Mexican children, including on their 2013 National Climate Change Strategy, which provided cognitive functions.183 Additionally, children affected guidance to achieve México’s climate change objectives during their early life stages exhibit lower height (0.42 to for 10-, 20-, and 40-year periods188; (iii) the Special Climate 0.71 inches), higher likelihood of stunting (11 to 14 per- Change Program (PECC) 2014-18, in which the national centage points), and lower weight (0.84 pounds) than government established priority climate change mitiga- same-aged children who were not affected by the flood tion actions for its term of office189; (iv) policy instruments shocks. Forecasts indicate that the consequences of cli- developed under the global climate change regime, such mate change are likely to increase, with more frequent and as Nationally Appropriate Mitigation Actions (NAMAs); severe extreme weather events, with further impacts on (v) the Nationally Determined Contributions (NDC), ad- poverty. In the case of extreme dry events, for example, it opted under the framework of the Paris Agreement, that is expected that climate change will generate an addition- include Mexico’s conditional and unconditional climate al share of the population in Mexico to be impoverished in change goals190; and (vi) the Mid-Century Climate Change the amount of 1.76 percentage points, that is, 1.78 addi- Strategy.191 tional millions of poor people.184 Extreme weather events and other impacts of a changing climate are expected to 157. Mexico’s legal and policy framework recognizes the lead to an increase in climate migrants. Recent analysis essential role of market instruments in achieving mitigation suggests that in an extreme case, Mexico could have 1.7 goals cost-effectively. The LGCC provides the legal basis million climate migrants – or 11 percent of all internal mi- for market-based instruments, such as emissions trading grants – by 2050. Most of these people will be migrating to and carbon taxes. It also creates the Climate Change Fund the central plateau around Mexico City, placing increased with the purpose of attracting and channeling public, and demand on the city’s resources and institutions.185 Other private, national, and international financial resources to research finds that, by 2030, nearly 3 million people will support the implementation of actions to combat climate remain in poverty as a result of climate change.186 The rise change.192 At the international level, and in alignment with in the sea level is particularly worrisome in the coastal the LGCC provisions on market-based instruments, Mexico states of Quintana Roo, Tabasco, Campeche and Sinaloa. acknowledges the role of a robust global carbon market to Agriculture is highly vulnerable to climate change, direct- achieve rapid and cost-efficient mitigation.193 ly affecting food security and the livelihoods of rural and urban populations. 158. In 2013 Mexico established an excise tax based on the carbon content of fossil fuels. As a component of the 155. Mexico has been at the forefront of using innovative 2013 fiscal reform, this environmental tax aims to send instruments to reduce the financial risk associated with a price signal towards a lower-carbon economy and ob- natural disasters. In 1996, the country established the taining revenues from externalities.194 In this context, it FONDEN, through which the federal government allocates complements the effect of fossil fuel subsidies reductions budget ex-ante for post-disaster response and reconstruc- that started in previous years as well as the effect of fuel 181 Rodríguez-Oreggia et al. (2013) find that the occurrence of natural disasters during 2000-05 increased food poverty, or extreme poverty, by about 3.7 percent; capacities poverty by 3 percent and assets poverty by 1.5 percent. 182 Rodríguez-Oreggia et al. (2013). 183 Aguilar and Vicarelli (2011). 184 Ahmed et al. (2009). 185 Rigaud et al. (2018). SYSTEMATIC 186 De la Fuente et al. (2013). COUNTRY 187 GoM (2012). DIAGNOSTIC 188 GoM (2013). MEXICO 189 GoM (2014). 190 GoM (2015). As the first country to submit its NDCs to the UNFCCC, Mexico committed to unconditionally reduce its greenhouse gases and Short-Lived Climate Pollutants (SLCPs) emissions by 25 percent below business-as-usual (BAU) by 2030. 191 GoM (2016). 88 192 GoM (2012). 193 SEMARNAT (2017b). 194 Muñoz-Piña (n.d.). excise tax increases.195 As with all taxes in Mexico, carbon social sustainability risks requires promoting social and 4. KEY SUSTAINABILITY tax revenues cannot be earmarked towards a specific pur- economic inclusion of traditionally excluded groups. In- CHALLENGES pose. Instead, the carbon tax has contributed to the gen- equality of opportunity based on gender, race and ethnic- eral expenditure of the federal government. In addition to ity is an important determinant of the observed inequality the carbon tax, Mexico is exploring the development of a in the distribution of wealth and educational outcomes in cap and trade system that would provide the means for Mexico. Empirical evidence shows, for instance, that strati- cost-effective emission reductions in certain economic ac- fication by skin color persists in Mexico. People with light- tivities. The development of the technical and regulatory er skin show more upward social mobility, independent of components of such a system is under way.196 In addition, their starting wealth. Individuals with darker skin rank 20 in collaboration with the Mexican stock exchange, Mexi- percentiles lower in the current wealth distribution than co is implementing a voluntary web-based simulation of their lighter-skinned counterparts—even when having an emissions trading scheme (ETS) with the objective of the same level of parents’ wealth—, and also show higher providing companies with insights about how the ETS op- downward mobility than other groups.202 erates and of building their capacity to participate in it.197 Mexico has also been collaborating with other govern- 161. The social and economic integration of millions of ments for the development of an offsets generation and young people who lack opportunities is one of the highest validation system particularly for those activities or sectors priorities. The NEETs phenomenon is one element in a that may not be incorporated into a cap and trade system chain of subsequent long-term consequences given its but that may equally contribute to mitigation actions that several intergenerational dimensions. By definition, it im- also support adaptation to climate change.198 plies that a sizeable part of today’s youth population is not accumulating the human capital necessary to effectively 159. To achieve its ambitious climate change goals, as de- contribute to and benefit from labor and economic oppor- scribed in its Nationally Determined Contribution, Mexico tunities. At the same time, there is the concern that the will have to overcome several challenges. While the NDC incidence of NEETs is higher among youth coming from states the country’s overall goals and there are preliminary households in the lowest quintiles of the income distribu- estimates of sectoral contributions needed to achieve tion, perpetuating thus inequality in the long-run. More- them, robust analytical tools to estimate the feasibility and over, NEETs are associated with other social problems that cost-effectiveness of alternative interventions have yet to increase the probabilities of poverty traps, for example be developed. Mexico has phased out government subsi- deficiencies in the quality and relevance of education that dies to gasoline and diesel between 2010 and 2015, and may lead to early dropout, high incidence of teen child- implemented a carbon tax.199 However, challenges remain bearing, high levels of youth unemployment, crime and regarding the establishment of an Emission Trading Sys- violence.203 tem (ETS), such as building consensus among stakeholders and ensuring consistency of Mexico’s markets with those 162. Exclusion and youth disenfranchisement present dis- of other jurisdictions. Also, energy subsidies are still signif- tinctive correlation patterns with crime and violence. Crime icant (e.g. in the electricity sector) and the present carbon and violence tend to be higher in poorer and more un- tax is too low to motivate de-carbonization at scale.200 equal areas; and have a strong gender, age and income There is also recognition of the need to leverage private, dimensions, whereby young, male NEETs are dispropor- public, and international financial resources to achieve the tionately affected. Homicide rate among young boys national mitigation goals.201 Linking these goals with other (ages 10-14, 15–19 and 20-24) are significantly higher development challenges such as health co-benefits (as the than those of the general population (2.1, 28.0, and 58.6 LGCC mandates) is another hard task, as is the goal to align per 100,000 for boys in the respective age groups com- national and sub-national climate change actions. pared with 1.7, 16.4, and 32.5 for the general population). A positive and significant correlation between exclusion (incidence of NEETs) and crime approximated by homicide 4.3 Social sustainability rates has been found in states along the border with the United States—a region afflicted by organized crime and 160. The social sustainability dimension of Mexico’s devel- the economic crisis of 2008–09.204 Changes in internation- opment is defined by two interrelated challenges: lack of so- al migration dynamics in recent years are affecting the cial mobility and inequality of opportunity, often stemming probabilities and mechanisms for a significant proportion from gender and ethnicity, among other factors. Reducing of youth—particularly boys and young men—to fulfill as- 195 Montes de Oca and Muñoz-Piña (2016). 196 Prieto (2017). 197 SEMARNAT et al. (2018). SYSTEMATIC 198 SEMARNAT et al. (2018). COUNTRY 199 Enríquez et al. (2018). DIAGNOSTIC 200 Montes de Oca and Muñoz-Piña (2016). MEXICO 201 GoM, 2015; 2016. 202 Campos-Vázquez and Medina-Cortina (2017). 203 For example, as discussed in Section 4: Crime and Violence, early labor experiences away from formal employment can have damaging effects over adult life employability and wages. 89 The limited ability of youth to acquire and use human capital thus constitute an exclusionary force to productive activities, leading to misallocation, and perpetuating the negative feedback loop. 204 De Hoyos et al. (2015). 4. KEY pirations of higher socioeconomic status, which may also al stresses such as growing regional gang activity, or cli- SUSTAINABILITY CHALLENGES be contributing to their increased participation in drug mate change-related migration and forced displacement, related crimes and violence.205 Moreover, 66 percent of among others. women in Mexico aged 15 and older have experienced some gender violence, which includes physical, sexual, 164. Low trust in institutions is associated with weaker economic, emotional or psychological violence. Leveling incentives for cooperation and coordination creating a dy- the playing field might have a positive impact in terms of namic of low compliance with norms and free-riding prob- growth and social cohesion, providing returns for society lems. Levels of indicators such as civic engagement, trust as a whole. in institutions and confidence in the government are low in Mexico compared with its peers. For instance, only 8 163. Inclusive violence prevention, effective implementa- percent of the population believes that the country is tion of the law and a shift of social norms are central to over- ‘governed for the good of all’; in contrast with 32 percent coming the current challenges that Mexico is facing, and mit- in Uruguay, 23 percent in Argentina, 17 percent in Peru, igating risk factors. Addressing a ‘continuum’ of violence and 14 percent in Chile. Moreover, Brazil and Mexico have that ranges from homicides, to Gender Based Violence, to experienced eroding indicators of trust reaching the low- gang activity requires a multifold policy approach, which est levels in the region.206 Ensuring access to efficient and combines social and situational prevention with effective effective implementation of the law is crucial not only to implementation of criminal justice. This approach should improve outcomes but also to increase compliance: be- account not just for the existing risk factors behind crime lieving that processes follow the rules leads to higher com- and violence in Mexico, but also look to mitigate risk asso- pliance with the law, even if the outcomes do not always ciated with mega-trends that are likely to create addition- favor individuals.207 SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 90 205 Meza González and Ramírez García (2012). 206 Latinobarometer 2006–17. 207 WDR 2017. 5. Unlocking Mexico’s inclusive growth potential 5.1 Structural impediments market. But what are the root factors linked to Mexico’s to inclusive growth growth and inclusion dynamics? Consolidating the signif- icant body of research in Mexico from different sources, 165. Chapters 2–4 highlight the growth, inclusion, and sus- past work of the World Bank, and work under this report, tainability dynamics and challenges faced by Mexico. On this section argues that structural/microeconomic imped- the economic growth side, limited growth dynamics seem iments could be hindering growth and inclusion. linked to limited capital accumulation over time, deep regional growth disparities with limited regional con- 166. The persistence of well-known structural impediments vergence, low productivity growth also with significant seems to have hampered the long-term path of growth, in- dispersion across regions, sectors and firms, and several clusion and sustainability. At the same time, low growth successive external and domestic shocks that resulted in and high economic disparities seem to reinforce and per- large downswings in economic output. On the inclusion petuate structural impediments. A key argument in this side, we observed that progress may have been slowed by report is that a misallocation of resources is taking place, low quality of inputs for human capital formation, insuffi- within and across sectors in the economy, across regions, cient access to basic services for the poor and vulnerable, firms and individuals, and that it is harming productivity, decreasing labor productivity in the bottom of the skills growth and even inclusion.208 209 High-performing market distribution, and low female labor force participation and economies exhibit a high degree of allocative efficiency. In high proportion of youth neither on school or in the labor this context, this report looks to identify the most critical SYSTEMATIC COUNTRY 208 See Chiquiar and Ramos Francia (2006), Levy and Walton (2009), Hanson (2011) and McKinsey (2013). DIAGNOSTIC 209 Separating misallocation between and within sectors can help identify improvements in intersectoral allocative efficiency—such as the movement of labor across sectors—as well MEXICO as changes in within-industry productivity, e.g., as a result of specific investments (Mc Millan et al. 2018; Herrendorf et al. 2013). A better understanding of the role of structural impediments is useful to explain the ambiguous effect of structural transformation on productivity; what factors are contributing to the low productivity of sectors, and the incentives/ disincentives to the movement of labor across them (Duarte and Restuccia 2009). Allocative efficiency—both within and between sectors—interacts with the role of the law and the 93 general political economy. E.g., a source of disparity within sectors—even in narrowly defined industries—could be due to misallocation of resources across firms (Bartelsman et al. 2013). This, in turn, may be explained by a differential implementation of rules due to privileged relationships with the authorities, among other distortions. Externalities, public goods, market power, or other factors associated with inefficient equilibrium outcomes also shape the process of structural change. 5. UNLOCKING MEXICO’S Figure 84. Structural impediments and growth INCLUSIVE GROWTH POTENTIAL Growth Inclusion challenges challenges • Convergence, growth and trade dynamics Structural • Limited productive inclusion (human, • Business cycles and aggregates shocks impediments financial, natural and social capital) • Limited capital accumulation • Unequal service delivery • Regional growth disparities • Decreasing labor productivity • Low productivity growth and high • Low female and youth labor force dispersion across regions, sector and firms participation Sustainability challenges • Fiscal and debt sustainability • Environmental sustainability • Social sustainability impediments that prevent the assignment of resources to raise consumer and financing prices and costs, create a their highest-valued use.210 Lifting these impediments or less competitive business environment for the private bottlenecks would help support an accelerated and more sector, and encourages informality. The second refers to broad-based, inclusive growth. Moreover, it also argues the role of the Rule of Law in leveling the playing field for that key aspects of fiscal, environmental and social sustain- private individuals, households and firms, including on ability are critical to consider in order to achieve sustained issues related to uneven access to justice and impunity, growth and poverty reduction over the medium and long crime, violence, and corruption. The third set of structur- term (Figure 84). Addressing these structural constraints al impediments refers to the allocation and utilization of and sustainability challenges would help Mexico realize public resources, particularly as regards to its impact on its growth potential more fully and catch up with US per economic efficiency, the redistributive capacity of the capita income levels. fiscal system, quality and equity of service delivery, and long-term planning of investments. 167. These structural impediments appear to be associated with unfinished or incomplete reforms, as well as policies 168. The three sets of structural impediments are linked and programs that could be adjusted. Recent evidence with each other and to the issues of horizontal and regional from Mexico shows that main distortions in the econo- disparities in income, income growth, and productivity in my result from the interaction of policies related to tax- the private sector. They are also linked to below-potential ation, credit, labor and social insurance regulations; the levels of investment and growth as well as to econom- absence of policies to correct for market failures; artificial ic vulnerability to external shocks. Moreover, they are barriers to entry or subsidies; as well as from registration also connected to the suboptimal provision of quality or transactions costs.211 Another recent study suggests inputs for human capital accumulation and to low lev- that Mexican states with better business regulations (less els of labor force participation, particularly for women impediments) lost fewer formal jobs during the 2008 and youth. The impediments have also been associated financial crisis and created more formal jobs in its after- with factor misallocation, which has driven decreasing math, than states facing worse business regulations.212 productivity at the bottom of the skills distribution, as Drawing on an extensive literature, and consultation well as systematic exclusion of vulnerable groups from with external and World Bank experts, this diagnostic has productive activities. identified three sets of structural impediments, which, if addressed, could have the highest potential to break the 169. It is important to highlight at the outset that significant vicious cycle or negative dynamics taking place between progress has been made in these three areas. As highlighted these impediments and moderate growth and inclusion in the overview chapter, the past decade has seen import- in the economy. The first refers to distortions in selected ant and positive reforms across a number of policy areas, product and factor markets, which deter entry of new including those where impediments continue to exist. Box SYSTEMATIC firms into certain economic subsectors, increase costs of 9 provides a brief summary of these reforms. Thus, the rest COUNTRY operation, delay exit of nonperforming economic agents, of the section focuses only on the pending issues ahead. DIAGNOSTIC MEXICO 210 For more detail on economic theory of allocative efficiency based on firm dynamics, see Lucas (1978), Hopenhayn (1992), Caballero and Hammour (2001), Haltiwanger (2011), and 94 Bartelsman, Haltiwanger and Scarpetta (2013), among others. 211 Levy and López-Calva (2017). 212 Specifically, better employment performance was driven by regulations that facilitated access to credit and simplified the process of starting a business (Iacovone et al. 2018). 5. UNLOCKING Box 9. Structural reforms 2012-2017 MEXICO’S INCLUSIVE As part of the Pacto por México, a political agreement across the main parties, the government started rolling out major GROWTH POTENTIAL structural reforms in 2012. This process entailed approving legislation and regulation aiming to boost productivity, strengthen rights and support the democratic system. Some reforms have been fully enacted, while others remain to be completed (Table 5). Table 5. Progress in the implementation of the structural reforms Well-advanced implementation Gaps in implementation Insufficient progress •• Tax policy reform •• Labor market reform and tackling •• Financial sector liberalization informality •• Education quality reform •• Agricultural transformation •• Telecom deregulation •• Anti-corruption and transparency reform •• Unemployment insurance, pensions and social •• Election system reform benefits •• Competition policy and regulatory •• Judicial process reform •• Health system reform reform •• Innovation system reform •• Energy market openness •• Fiscal federalism Source: OECD 2018. Table 6. Main results of the structural Overall, the growth impact of the reforms has not yet reforms (2012-2017) matched expectations, which range from an additional 0.75 to 1 percent a year (IMF estimates) to 4–5 percent Labor market reform (2012) (government estimates). It has been argued that part of •• More than 3 percentage points of GDP in additional tax revenues the observed gap is due to a lag, where reforms take a •• Informality dropped from 59.5 percent in 2012 to 56.9 percent in delayed time to yield results, and have short-term costs.a 2017 OECD calculations suggest that a subset of the reforms Tax policy reform (2013) in Mexico could add one percentage point to GDP growth after five years.b Reforms, in this sense, would •• More than 3 percentage points of GDP in additional tax revenues begin to show results about now, as most legislation was •• Oil revenues as share of total tax revenues from 39 to 17 percent approved by 2013–14. The main achievements of these •• Tax base increased from 38 to 66 million taxpayers reforms so far are listed in the Table 6 below. Education quality reform (2013) Political economy factors may have contributed to the •• Scholarships from 3 out of 10 students in public schools lack of results to date. According to this perspective, the Energy market reform (2013) reforms have not been able to shift the political econ- omy equilibrium (and associated incentives structure), •• Expected investment between 160 and 200 US$ bn making legislative changes largely ineffectual. This could •• More than 70 new energy firms be reflecting that, in the absence of a systematic imple- Competition policy and regulatory reform (2013) mentation of the rule of law, changing laws is insufficient. It has been argued that as long as corruption, insecurity •• Double in fines for monopolistic practices or extortion practices continue to influence production Financial sector liberalization (2014) processes and competitiveness, the potential impact of reforms will remain incomplete.c The present diagnostic •• 13 million people gained access to financial services accounts for both the time lag and the political economy Telecommunication reform (2014) factors. •• 50 million additional people subscribed to mobile broadband a. Evidence suggests that the growth impact of product market reforms in OECD •• 24 percent reduction in telecommunication prices between 2013 countries becomes statistically significant after three years and does not fully material- izes until after seven years (raising GDP on average by a cumulative 1.5 percent (Duval and 2017 and Furceri 2016; IMF 2016). b. OECD (2015). c. Rios and Wood (2018). Fiscal responsibility Law for subnational discipline (2016) •• States’ debt to non-earmarked transfer ratio reduced from 88 to 80 percent in 2017 chains. In that context, it is of critical importance to re- 5.1.1. Product and factor markets SYSTEMATIC duce any hurdles to entry, competition, exit, efficient costs COUNTRY 170. Product and factor inefficiencies affect private sector of operation and flow of labor to productive and formal DIAGNOSTIC growth. As discussed in Chapter 2, there are weak links activities, as well as to their access to finance. Further im- MEXICO between the NAFTA export-oriented firms in the northern plementation and reforms in these areas can support the 95 and central states of the country, and the large share of strengthening of those linkages. While progress has been low-productivity firms not linked to those global value made, there is significant room to improve across the 5. UNLOCKING MEXICO’S Map 4. Variation in resource misallocation across states INCLUSIVE GROWTH POTENTIAL high low medium-high medium-low Source: IMF (2017). board on these issues at the federal and especially at the for start-ups—represented the largest share in Mexico. subnational levels. Barriers to trade and investment, which refer to barriers to FDI, tariff barriers and differential treatment of foreign 171. Reducing barriers to doing business can also bring suppliers, were also relatively high in Mexico. Initial results substantial welfare gains to households and boost the pri- from an analysis of changes in profits in response to chang- vate sector in the poorest states, especially in the south of es in costs (as a proxy measure of competition) suggest Mexico. There is evidence that the benefits of alleviating that the services sector is characterized by a significantly distortions, which prevent the allocation of capital and lower degree of competition, and that competition has labor to their most productive use, are higher in Mexico actually decreased in the sector between 1999 and 2014, than in other peer countries.213 Recent evidence shows driving an overall decrease in competition in the economy that improving the efficiency of factor allocation by a during that period.216 Lack of competition is estimated to quarter could increase the annual growth of the output of cost Mexico 1 percent of GDP each year and dispropor- manufacturing and services by 1.4 percentage points over tionally affects households in the lowest income decile.217 20 years.214 The greatest potential gains are found in some The recent reforms in energy, telecommunications and fi- of the poorest, southern states (Map 4).215 Factor misallo- nancial services218, as well as those to the competition law cation is correlated with sectors with high entry barriers, itself, have started having some positive effects, including low access to financial services and informality. lower prices in telecommunications and increased entry of foreign competitors (energy). In the energy sector, con- cessions of US$80 billion had been awarded as of August Concentration in critical input markets and 2017, bringing in 66 private market entrants.219 Mobile barriers to entry at the local level broadband prices, which were comparatively high before 172. There are significant obstacles to competition, al- the reform, are now among the lowest in the OECD.220 though major federal-level reforms have begun to reduce these barriers in key industries. As of 2013, product market 173. The inefficiencies resulting from market concentra- regulations were relatively restrictive in Mexico compared tion and obstacles to competition are well-documented, with other OECD countries (Figure 85). Such barriers, cou- especially at the local level. A growing body of evidence pled with other noncompetitive arrangements, protect shows that subnational regulations often limit entry, rein- incumbent firms and prevent the entry of newcomers. The force dominance, facilitate collusive outcomes or distort extent of regulatory protection to incumbents was higher the level playing field, thus leading to anti-competitive than in other OECD member countries. Within regulatory behavior.221 For example, while private competitors to barriers, those related to entrepreneurship—including li- PEMEX may reduce upstream gasoline prices, subnation- cense and permit systems and administrative procedures al regulations that establish minimum distances—such 213 Busso, Fazio, and Levy (2012) find TFP gains of 95 percent from eliminating resource misallocation in Mexico (more than that of other Latin America and Caribbean countries considered). 214 Using World Bank Enterprise Survey data for the manufacturing sector, IMF (2017) reports that TFP gains in Mexico could be between 93 - 130 percent for 2010 and 2006. SYSTEMATIC 215 Using World Bank Enterprise Survey data for the manufacturing sector, IMF (2017) reports that TFP gains in Mexico could be between 93 - 130 percent for 2010 and 2006. The gains COUNTRY are 25 percentage points in the Southern and South-Eastern states than in the rest of Mexico. DIAGNOSTIC 216 Background paper produced for the SCD. The analysis follows the methodology proposed by Boone (2008). MEXICO 217 OECD Economic Survey (2015); Chiquiar and Ramos-Francia (2009). 218 A comprehensive financial reform was approved in 2014 with the objective of increasing the sector’s contribution to economic growth, promoting development opportunities in strate- gic sectors including through development banks; increasing competition; improving legal certainty for private sector actors in order to expand credit and make it more affordable; and maintaining a solid and stable financial system (Gobierno de México, 2014). 96 219 Agencia EFE (2017). 220 OECD (2017). 221 Licetti et al. (2016). 5. UNLOCKING Figure 85. Product market regulation indicator MEXICO’S INCLUSIVE 3.5 GROWTH 31% POTENTIAL 3.0 29% 31% 2.5 2.0 24% 23% 27% 34% 5% 31% 9% 1.5 11% 33% 38% 30% 34% 33% 38% 1.0 45% 36% 38% 62% 38% 0.5 42% 44% 33% 39% 51% 46% 35% 30% 0 OECD Top 5 average Chile Poland Colombia South Korea Mexico Peru Brazil Uruguay Argentina State control barriers to entrepreneurship barriers to trade and investment Source: World Bank calculations based on OECD data. as more than 1km—between gas stations may annul the lations that limit competition in key sectors in Oaxaca, benefits, granting significant local market/pricing power Tabasco, and the State of Mexico, in addition to 17 other to the incumbent. In the telecoms sector, the dominant states. Of the 520 restrictive regulatory provisions identi- operator, Telcel, still held 71 percent of subscriptions in fied, most of them are in the retail, construction and man- the mobile broadband market as of end-2016. In Tabasco, ufacturing sectors. recent evidence suggests that unions have a monopoly in the transport of construction materials, which lead to road 175. Evidence shows significant and regressive social wel- works that are 15 percent more expensive as a result of fare losses in Mexico due to the exercise of monopoly and overcharges.222 In the State of Mexico, stringent limits to oligopoly power.227 Increased market concentration usually market entry are associated with fewer competing super- raises consumer prices, and can lead to lower employment market chains and higher prices for consumers.223 Since and wages, creating an economy-wide welfare loss, and 2005, the Federal Commission for Economic Competition reducing the relative incomes of households, particularly (COFECE) has uncovered at least six tortilla cartel agree- of the poor. Concentration in input markets often prices ments, which can have a significant impact on households’ out the poorest and vulnerable or implies higher relative expenditures considering the average household spend- out of pocket costs. Until recent years, the regulatory bod- ing on this product.224 Market concentration and lack of ies of key sectors such as finance and telecommunications competition is reflected throughout the value chain. In the have struggled to connect the efficiency gains of banks production of hybrid seeds for corn, for instance, fertilizers and telecommunication firms into improved access and and access to warehouse and storage facilities are only better services for households and enterprises.228 Even available to large producers, which further limits competi- though recent reforms have helped address concentra- tion in the agricultural value chain, and which can impact tion, telecommunication and digital services markets the consumer price burden.225 remain concentrated, margins are high, and access is un- evenly distributed across the country.229 Similar situation 174. Regulatory barriers to competition at the local level with banks which benefit of high profitability partially due tend to be dispersed across legal instruments, sectors and to high commissions compared with peer countries.230 jurisdictions. Their potential negative impact depends Recent evidence shows the welfare-enhancing effects of on how these barriers are applied and on the features of market competition. Increased competition through re- the market that are affected, some of which are linked to tail globalization in Mexico in the 2000s had positive net strong local vested interests. Results from a recent study226 effect on average household welfare —even though the corroborate the proliferation of anti-competitive regu- richest households experienced higher relative gains.231 222 Based on surveys conducted with market actors during implementation of reforms proposed in Licetti et al. (2016). 223 Licetti et al. (2016). 224 Gobierno de México (2016) 225 UNCTAD (2014) SYSTEMATIC 226 Subnational application of the World Bank’s Markets and Competition Policy Tool, in partnership with COFEMER. (World Bank 2017). COUNTRY 227 Urzua (2008). DIAGNOSTIC 228 Levy and Walton (2009). MEXICO 229 For instance, while in Baja California Sur and Sonora, Internet access reached over 70 percent of households in 2016, in Chiapas and Oaxaca, only 13.3 percent and 20.6 percent of households had Internet access (INEGI 2017b). 230 It has been argued that during the 2000s, credit and debit cards were used relatively less in Mexico compared with other countries due to the high consumer fees—where fees for the 97 use of electronic methods were greater than for using checks, while the costs of the latter were higher than those in developed countries (Castellanos, Garrido and Mendoza (2008). The Law to Promote Transparent Financial Services was enacted in 2004, and revised in 2007, to promote transparency and to protect the interests of consumers. 231 Atkin et al. (2015), on the welfare effects of the entry of Walmart stores in Mexico. 5. UNLOCKING Analogously, higher local concentration in the retail sector segment of the population.241 Two companies accounted MEXICO’S INCLUSIVE has been found to raise the poverty headcount at the mu- for 90 percent of the total production volume of corn flour GROWTH nicipal level, also via higher prices.232 A recent study also in Mexico in 2010. While lack of data precludes more re- POTENTIAL provides evidence that markets for common, well-defined cent analysis, historical data show that large companies products in Mexico do not function like a well-integrated in this supply chain appear to be benefiting directly and market, namely, while there is heterogeneity across prod- indirectly from subsidies. Concentration is not exclusive to ucts, locations matter greatly.233 the marketing and processing of grains but also exists in the sale of inputs and seeds to farmers: in 2009, 95 per- 176. High intermediation in the agribusiness chain is associ- cent of planted hybrid seeds were produced only by two ated with consumer loss and reduced benefits for producers. corporations.242 Waste, intermediation and inefficiencies have implications for both producers and consumers. There is indication that Access to finance the level of intermediation in Mexico, on average, increas- es the price of food by 400 percent.234 Part of the price 178. Mexico has one of the lowest credit-to-GDP ratios differential is explained by food losses. Evidence points to among peers (including countries at similar levels of in- a 20 percent of food loss and waste alongside the agribusi- come), suggesting obstacles in the efficient allocation of ness chain (over 20 million tons235), most of which ends up financial capital (Figure 86). In the context of the financial being transferred to the consumer.236 A lag in the availabil- sector reform, significant progress has been made over ity of storage centers to collect grain, fruits and vegetables the last four years. For example, Mexico’s position in the has been pointed at one of the impediments to achieve World Economic Forum’s Global Competitiveness Report a better regulation of prices.237 Another factor behind the improved from 63 in 2013 to 36 in 2017. Credit (as a share high intermediation in the country is the highly polarized of GDP) to the private sector increased and more than 13 structure of the final sales market for food, which ranges million people gained access to financial services due to from large retail and supermarket chains to an extensive the reforms. Yet, further progress is needed. Credit to the number of small stores and informal markets.238 Some private sector and deposits remain low, not just when com- early reports have argued that NAFTA has contributed to pared to peers at the same level of income, but within the market concentration in the food and agriculture sector Latin America region as well. Just one third of SMEs have by incentivizing fewer and larger farms in monocultural access to loans, and only 37 percent of adults had access systems and increased export dumping by agribusiness to services in a financial institution in 2017 (Map 5, Global firms, where overproduction often resulted in downward Global Financial Inclusion Database- Findex) or 44 percent pressure on prices.239 as per the forthcoming Mexico’s Encuesta Nacional de In- clusión Financiera (ENIF 2018).243 The scarcity of long-term 177. The agricultural and food sector is marked by varying financial resources hinders the development of housing degrees of market concentration. Some evidence suggests and infrastructure finance. The credit to the private sector that some economic agents may be benefiting without a and deposits for Mexico is 25.6 percent, compared with strong rationale from agricultural subsidies. Several mar- the Latin America and Caribbean median of 45 percent, ket value chains in the agrifood sector Mexico are concen- and the income group median of 47.2 percent.244 Domes- trated among a handful of medium- to large-scale private tic bank credit to the private sector contracted consider- firms. These firms often display oligopolistic behavior, ably following the 1995 crisis.245 Although its growth final- with impacts on the price-taking producers and small-to- ly resumed in 2006, the current penetration of bank credit medium-scale agritrade entrepreneurs.240 It has also been to the private sector remains low by regional standards. argued that some of these firms receive a large share of The ratio is low even when considering nonbank financ- the agricultural subsidy programs. One example of a sec- ing sources, which have been institutionalized, such as tor with a high concentration is the subsector of corn flour housing credit through public housing institutes—Infon- milling. This product is of particular relevance for final avit and Fovissste—, financing to companies from banks consumption in Mexico, especially for the lowest income abroad, and financing from local financial markets.246 232 Rodríguez-Castelán and Rodríguez-Chamussy (2015). 233 Sánchez et al. (2018). 234 IPD (2014); Chávez (2014). The increase is higher in some products, including staple crops, e.g., the price of lemons increases from Mex$1.10 a kilogram, as bought from the produc- er, to Mex$8 to the consumer, where the producer only receives 14 percent of the final price. 235 Approximately 20.4 million tons of food are lost and wasted, 72 percent during the first stages of production and 28 percent in the retail selling and consumption stages. 236 IPD (2014). 237 As pointed out by Emilio Romero Polanco, specialist at the Instituto de Investigaciones Económicas de la UNAM, in Chávez (2014). 238 As pointed out by Emilio Romero Polanco, specialist at the Instituto de Investigaciones Económicas de la UNAM, in Chávez (2014). 239 IATP (2010). 240 UNCTAD (2014). SYSTEMATIC 241 In 2010, households in lowest income decile devoted about 10 per cent of their total food and beverages expenditures to tortilla, compared with only 3.1 per cent among households COUNTRY in the highest income decile. Also, tortilla consumption is significantly higher in rural areas (Secretaría de Economía, 2012). DIAGNOSTIC 242 Namely, Monsanto and Pioneer. See Luna et al. (2012). MEXICO 243 Global Findex (Global Financial Inclusion Database), World Bank, Washington, DC, https://globalfindex.worldbank.org/. The Global Findex is based on demand-side data and measures people’s perception about ownership and use of financial services. The margin of error for Mexico is 3.7. The methodology used for Findex may differ from country-based surveys. For instance, Mexico’s Encuesta Nacional de Inclusión Financiera (ENIF) suggests that from 2012-2015, the share of adults who own a transaction account (“cuenta de depósito” or “cuenta de ahorro”) with a regulated financial service provider increased from 35.5 to 44.1 percent. Results from the ENIF 2018 will be published in November 2018. 98 244 2018 data of FinStats (internal database), World Bank, Washington, DC. 245 Domestic bank credit to the private sector fell from 35.2 of GDP in 1995 to 12 in 2000, and 9.8 in 2005, only showing signs of recovery in 2006 (Albo and Morales 2012). 246 Accounted under II. Non-Bank Entities of the Private Sector (NBEPS or ENBSP in Spanish) in Table 4 5. UNLOCKING Figure 86. Credit to GDP vs. per capita income MEXICO’S INCLUSIVE 200 GROWTH POTENTIAL 180 ChN DNK ChE 160 JPN KOr Gbr AuS 140 MYS ESP NOr Credit as percentage of GDP PrT SWE ThA 120 iTA NLD GrC FrA brA! LuX 100 OECD FiN AuT iSL DEu PAN ChL bEL 80 ZAF LCATur iND EMEs POL SvKEST iSr bGr Cri SvN uSA 60 huN irL LvA PAKSiv MEX 40 PEr DOM ECu urY 20 hTi rOu rG 0 0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 income per capita at current uSD prices Source: FinStats (internal database), World Bank, Washington, DC. Map 5. Share of adult population with an account (%) 0-19 20-39 40-64 65-89 90-100 Source: Findex database. 179. An important fraction of total financing in Mexico goes higher than in Chile, Argentina, and Peru and slightly high- to the public sector (federal, state or municipal government, er than in Colombia.247 or public agencies). In 2017, the public sector accounting for roughly 54 percent of total private credit to the econ- 180. Mexico’s financial sector is concentrated, and stronger omy (Table 7). Under some circumstances, the large share competition may lead to more efficient intermediation. The of resources destined for the public sector may inhibit the financial sector includes a diverse range of financial inter- expansion of credit and financing to the private sector, as mediaries and many small institutions, but assets remain domestically, both sectors compete for the same savings concentrated in large banks and financial conglomerates, pool. Mexican banks may have little incentive to lend to and intermediation margins have been rising. The five SYSTEMATIC riskier borrowers when they can, instead, lend to the gov- largest banks account for about 70 percent of total bank COUNTRY ernment. The share of credit and financing to the public assets (Figure 87), and the sector is more concentrated DIAGNOSTIC MEXICO sector in Mexico is lower than in Brazil but considerably relative to Latin American peers. With economies of scale, 247 Based on 2011 data, Albo and Morales (2012). 99 5. UNLOCKING Table 7. Credit to the private and public sector, selected countries, 2017 MEXICO’S INCLUSIVE GROWTH Credit to the Private and South POTENTIAL Argentina Brazil Chile Colombia Mexico Turkey Russia Public Sectors Africa Domestic Bank Credit to the Private 13.60% 60.10% 77.70% 47.80% 25.70% 63.60% 52.10% 64.40% Sector Total Credit to the Private Sector 20.70% 63.00% 141.70% 64.70% 41.60% 84.70% 66.10% 72.50% Domestic Bank Credit to the Public 7.60% 1.90% 4.20% 5.80% 15.00% 11.50% 6.30% 12.20% Sector Total Credit to the General Government 52.60% 84.00% 23.60% 49.40% 54.20% 28.50% 17.40% 52.70% Total 73.30% 147.00% 165.30% 114.10% 95.80% 113.20% 83.50% 125.20% Sources: BIS (Total Credit, Q3 2017), IFS (Domestic Bank Credit, Q3 2017). Figure 87. Commercial bank asset concentration in Mexico tend to obtain a high share of concentration (2017) credit from banks, relative to firms in less concentrated sectors, with little regard to productivity growth.248 Banks 5 banks Asset Concentration appear to allocate credit more based on features such as strong balance sheets and stable profits, common in firms 3 banks Asset Concentration with market power, and less so on characteristics such as productivity growth, output, or wage growth rates. This 2 banks Asset Concentration means that more credit may be being allocated to less productive sectors, limiting the growth of the less concen- 37% trated sectors; while more concentrated sectors could be producing less than the socially optimal. The misallocation 50% of resources due to credit constraints is estimated to lead to a loss in TFP of 10 percent.249 Moreover, smaller SMEs (those with fewer employees and with lower overall debt 69% levels)—and those in the initial stages of development— face significantly less favorable conditions than relatively larger SMEs.250 The prevailing high levels of informality in smaller companies further hinder access to finance. Source: Estimated with end of year data from CNBV, R0. 182. As a result, lending to segments, such as micro, small, and medium enterprises, is particularly scarce. Only 12 more concentration does not necessarily lead to less effi- percent of microenterprises received finance, while 32 cient outcomes, and various measures of competition es- percent of SMEs needed to invest but could not due to timated over 2012-15 illustrate the level of competition in financial constraints.251 Credit to micro, small, and medium the sector. A Lerner index higher than in other major Latin enterprises, which account for 95 percent of all Mexican American countries suggests that mark ups are high. The companies and employ 76 percent of Mexican labor (ac- Boone indicator, a measure of profit elasticity with respect cording to the 2014 Economic Census), represents only to marginal costs, is not as strong as in the peer countries 11 percent of commercial banks’ loan portfolio.252 Even suggesting that competition can improve. The country though there are high returns to capital in Mexican mi- has liberalized entry both by allowing foreign banks into croenterprises, capital is not flowing to these firms.253 This the market, allowing small non-bank intermediaries to op- could bear relation to the cost of servicing small loans, lack erate, and by introducing niche bank licenses that allow of collateral, a weak contracting environment, and large for the entry of less complex banks with commensurate- information asymmetries.254 State development banks ly lighter regulatory burdens. These reforms may render have introduced significant guarantees (including loss more competition over the medium term. coverage, lower capital requirements and loan loss pro- visions) to stimulate commercial credit to this segment. 181. Sectors with high levels of concentration appear to Commercial bank lending, on the other hand, tends to be obtain a relatively high proportion of financial resources. short term, costly and increasingly focused on the more Recent evidence shows that sectors with high levels of profitable consumer credit. Access to finance is the sixth SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 248 Ramos-Francia and García-Verdú (2017). 249 Lopez (2017). 250 “Reporte sobre las condiciones de competencia en el otorgamiento de crédito a las pequeñas y medianas empresas (PYME)”, BANXICO, (2015). 251 According to data of ENAPROCE (2015). 100 252 From IFC. 253 McKenzie and Woodruff (2008). 254 Negrete (2017). most important obstacle for companies (after corruption, remained unbanked in 2017.260 Gender gaps also remain 5. UNLOCKING MEXICO’S crime, inefficient bureaucracy, tax rates and regulations), in terms of savings accounts, possession of assets, savings INCLUSIVE according to the World Economic Forum’s global competi- for retirement, insurance and credit for housing, despite GROWTH POTENTIAL tiveness index.255 In addition, while Mexico ranks closer to efforts such as the Programa Integral de Inclusión Finan- the top in items such as the soundness of banks (in 47th ciera.261 The gender gap in account ownership in Mexico place out of 137 countries), it scores substantially worse in is 8 percentage points, while that between the rich and availability and affordability of financial services and ease the poor is 18 percentage points; compared with 6 and of access to loans (in 85, 75, and 78th place, respectively, 12 percentage points in Chile, respectively.262 Collateral among 137 countries).256 constraints among the poor, among other obstacles, limit access to capital. In lieu, the unbanked population often 183. A large proportion of the population—mainly lower use stores and retailers as their primary source of credit.263 middle- to lower-income individuals—still lack access to When the unbanked are reached by formal credit institu- formal financial services. Mexico’s unbanked population tions, interest rates are often higher than the average. 264 is far larger than the country’s level of economic and fi- New technologies and their regulatory frameworks offer nancial-sector development would predict. The share of another option to expand access to credit while guarding adults with an account at a formal financial institution for the needed prudency in credit growth (Box 10). decreased from 39 percent in 2014 to 37 percent in 2017, far below the level of most comparator countries (80 per- Labor market imperfections and informality cent of adults have a bank account in Brazil or China by comparison).257 Some of the potential barriers to account 185. Nearly two-thirds of the current employment in Mexico ownership are: the lack of unique identification making is informal. The informal sector is comprised by the more complying with know-your-customer requirements costly usual categories of self-employed, agricultural workers (19 percent of adults), financial services are too expensive and wage employees in unregistered (informal) firms, but (37 percent of adults) and distance of point of service (21 also by an important share of workers not-covered by so- percent of adults).258 Agent networks have not significant- cial insurance but employed in registered (formal) firms. ly enhanced the coverage of banking infrastructure to Presence of restrictive labor legislation may be linked to underserved areas (agents cover 19 percent of rural areas). segmentation between the formal and informal sectors; at Mexico’s agent-banking model remains focused on urban the same time, there is evidence of voluntary (self-select- areas and relies heavily on large retail chains rather than ed) informal employment reflected for example in high small shops. Agent-banking services also tend to be lim- mobility between formal and informal employment and ited to taking deposits and accepting loan repayments. procyclical informal employment.265 Recent evidence esti- There is also a large gap in coverage of agents between mate that between 10 and 20 percent of informal workers northern Mexico and the center and south of the country. would prefer to have a formal job, while an important pro- Within the five states with the largest number of com- portion of workers in the informal sector self-select into mission agents per adult, four are located in the north of it.266 the country. In contrast, the five states with the smallest number of agents per adult are located in the south and 186. Well-intentioned reforms to labor laws and social pro- center of the country. Among the pitfalls of the Mexican tection schemes aimed in principle to protect workers could model, agent banking relies to a great extent on large re- be unintentionally contributing to foster informality. The tailers rather than on small shop owners and the number modifications to the Federal Labor Law, passed in Sep- of services offered is limited to mostly deposits and loan tember 2012, aimed at increasing flexibility to the labor repayments. market and facilitating the formalization of firms and jobs; however, it fell short of creating unemployment insurance, 184. Increased access to financial services can lead to a sig- increasing the share of temporary workers and improv- nificant increase in income, particularly among low-income ing the quality of jobs.267 268 The reforms also increased individuals and those located in areas with lower preexisting enforcement by adding to the resources available to in- bank penetration.259 Among households in the bottom 40 spection of employers’ fulfilment of the regulation, and percent of the income distribution, 74 percent of adults increasing the penalties for noncompliance. Since 2014, 255 Schwab (2017). 256 Schwab (2017). 257 Global Findex (Global Financial Inclusion Database), World Bank, Washington, DC, https://globalfindex.worldbank.org/. 258 Global Findex (Global Financial Inclusion Database), World Bank, Washington, DC, https://globalfindex.worldbank.org/. 259 As shown by Bruhn and Love (2014), on the basis of a natural experiment—the opening of 800 branches of Banco Azteca. Ruíz (2010) suggests that informal households in municipal- ities with Banco Azteca were better able to smooth their consumption. 260 The provision of financial services is concentrated in urban areas, with only 26 percent of the poorest Mexican population having a bank account, compared with an average of 42 percent in Latin America. See Global Findex (Global Financial Inclusion Database), World Bank, Washington, DC, https://globalfindex.worldbank.org/. SYSTEMATIC 261 Details in Gobierno de la República (2014) (OECD, 2017b). COUNTRY 262 Global Findex (Global Financial Inclusion Database), World Bank, Washington, DC, https://globalfindex.worldbank.org/. DIAGNOSTIC 263 See Skelton (2008). MEXICO 264 See Ruíz (2010). 265 See for example Heckman and Pages (2004), Levy (2007), Maloney (1998). 266 Alcaraz, Chiquiar and Salcedo (2015). 101 267 Approval of the unemployment insurance element by the legislature is still pending. 268 The revisions to allow employers to hire workers for a three-month training period, could be contributing to lower-quality employment; while other provisions appear to be, in fact, introducing new barriers that hinder entrepreneurship. 5. UNLOCKING MEXICO’S Box 10. The Fintech (financial technology) sector in Mexico INCLUSIVE GROWTH The Fintech sector has the potential to contribute to close the gap in access to financial services in Mexico. Financial POTENTIAL technology companies leverage online, mobile and information technologies to deliver financial services. According to Finnovista’s Radar, Mexico ranked as the largest Fintech market in Latin America in 2017, surpassing Brazil for the second year in a row.a The August 2017 version of the Fintech Radar Mexico reports 238 fintech startups across 11 dif- ferent segments—a growth of 50 percent in the last year. Startups in Mexico are concentrated in lending (23 percent of startups), payments and remittances (22 percent), enterprise and personal financial management (15 and 10 percent, respectively), crowdfunding (9 percent), and insurance (6 percent). The latter is on the rise, leading the highest growth segments in 2017 (Figure 88). Figure 88. Fintech sector Mexico Number of employees per firm Growth per segment, 2017 114 75 35% 59 55 57 39% 50 38 20 20 13 5% Lending Payments Payments Crowdfunding Insurance Fin Education & Savings Wealth Enterprise Fin Mgt Personal Fin Mgt Scoring Indentity & Fraud Trading & Markets 17% 4% 1-10 11-25 26-50 51-100 >100 Source: Finnovista 2017. Particular features of the Mexican market that make it a fertile environment for Fintech include (a) a high Internet and smart mobile devices penetration; (b) a strong ecosystem of entrepreneurship and e-commerce; (c) a low banking penetration, and (d) an undeveloped consumer lending offer, according to the Finnovista report. In particular, Fintech startups in Mexico offer more efficient and less costly alternatives compared with traditional credit and remittances ser- measures designed to reduce the cost of formalization, with labor productivity. Furthermore, informality reduces such as simplified tax procedures, have played a role in accumulation of human capital, which compounds the encouraging firms to join the formal sector. As part of the problems of inefficient inputs markets.271 Besides self-se- fiscal strategy “Crezcamos juntos”, the efforts to increase lection into the informal market, the root cause behind the tax base have been successful in formalizing firms informality is on one side burdensome and distortionary and their employees as well as increasing revenue. On the labor regulations272 (including taxes, social security and social security side, IMSS’ strategy against informality im- firing costs) which, along with social programs, generate plied a significant reduction of the administrative burden an implicit subsidy to informality and at the same time, for enrollment, reaching 2.9 million additional formal jobs an implicit tax on formal workers. This in turn slows down enrolled between 2012 and 2017.269 formal firms’ growth, affecting the life-cycle of firms and their investment decisions about innovation, technology 187. A large body of research points to informality as the and labor training. main encumbrance to labor productivity in Mexico. Infor- mality is a drag on growth as for firms of the same size, 188. Incentives to formalization of firms combined with formal ones are 84 percent more productive than the access to credit is key to reduce labor market segmenta- informal.270 State-level informality is highly correlated tion. Research has found positive but modest effects of SYSTEMATIC COUNTRY DIAGNOSTIC 269 IMSS 2017, Estadísticas de asegurados. MEXICO 270 Busso, Fazio, and Levy (2012). 271 Hanson (2010), Bolio et al. (2014) and Arias et al. (2010). 272 Mexico ranks above comparator countries in the Employment Protection Legislation (EPL), which looks at the strictness of regulation of collective dismissals, ranging from 0 (very loose) to 5 (very strict): Mexico (2.6), Turkey (2.5), Poland (2.4), Colombia (2.3), South Korea (2.2), Uruguay (2), Chile (1.8) (OECD, 2015). Beyond this indicator, this SCD argues that 102 distortions arise not only from labor regulations but also from inefficiencies in taxation, credit, and social insurance policies, and to the interaction between them, as well as from lack of policies correcting market failures; barriers to entry; and transactions costs. For instance, market distortions can deter firms’ entry, increase operation costs, delay exit of nonper- forming agents, raise prices and costs, create a less competitive environment, and encourage informality. 5. UNLOCKING vices. Most Fintech startups are currently concentrated around Mexico City, Monterrey and Guadalajara. The majority MEXICO’S are incorporated in Mexico City (71 percent) and operate domestically (90 percent). INCLUSIVE GROWTH POTENTIAL Since 2010, venture capital funds have invested US$280 million in Mexican Fintech startups, which have provided over 10 million credit lines among private individuals. The total transaction value of Fintechs is estimated to reach US$36.4 million in 2018, with an annual growth rate of 17.3 percent up to 2022.b Interest in the financial services industry is growing. Optimism in the sector is reflected in the investment trend, where, beyond the seed-stage, venture capital investors are leading large rounds in the Series A stage.c Recently, Kueski, Konfio and PayClip have closed $10 million, $8 million and $8 million rounds, respectively. Most Fintechs in Mexico are either small, or large: 39 percent are small firms (with less than 10 employees), and over a third are companies with over 100 employees; while the number of firms in the mid-size range is smaller. Most compa- nies are young—69 percent were created over the last three years. Digital innovation holds potential impact for financial inclusion in Mexico, particularly for bridging underserved and hard-to-reach communities. Technology can help improve authentication of user identity, a typical challenge faced by financial institutions, through blockchain, biometrics, behavioral data, or open networks. Innovations in credit scoring also have the potential to improve affordable access to credit. Fintechs in Mexico already appear to be tapping this sec- tor, as a potential market opportunity. While in Brazil only 28 percent of Fintech startups are focusing on underbanked and unbanked consumers and SMEs, 46 percent of Fintech startups in Mexico are targeting this market. In addition to smartphone-enabled financing, cross-border credit history is one of the new business models that startups in other countries are putting forward. A Fintech Law was approved in March 2018, placing Mexico among a small group of countries to establish regulation for the industry.d The law aims at regulating the activities of a number of disruptive financial service providers, focusing on nonbank e-money issuers and operators of peer-to-peer lending (that is, crowdfunding) platforms. The law also foresees adopting, through secondary/implementing regulations, a “regulatory sandbox” approach for other types of innovative entities. Moreover, the law includes a provision to recognize virtual assets and regulate their usage and operations in Mexico. Open banking, one of the aspects of the law, also holds potential benefits for inclusion. The law introduces the con- cept of open data for nonconfidential aggregate data and for transactional data with consumers’ consent through the Application Programming Interfaces (APIs). Recognizing that information collected and stored by financial institutions allows consumers choice and can help boost competition, as startups and small and medium-sized banks are able to use information from the clients of large banks.e Enhanced competition, in turn, can have implications for inclusion in terms of access and cost of services. a. According to the Fintech Radars conducted by Finnovista in 2017, the main markets in terms of Latin Fintech startups are Mexico (238 startups), Brazil (219), Colombia (124), Chile (75), Argentina (60), Peru (47) and Ecuador (31). b. Statista (2017). c. Antoni (2016). d. Ley para Regular las Instituciones de Tecnología Financiera. e. Espejo (2018). the SARE program—a reform to speed up firm registra- employment) and of voluntary movement into and out of tion—on the creation of new start-ups and the number of informality. To the degree that social security in Mexico is registered business owners in municipalities in Mexico.273 primarily financed by wage-based contributions, and not However, if credit remains scarce for formal firms and the fully valued by affiliated workers, it acts as a tax on salaried cost of enforcing contracts is high, the cost of paying taxes employment.275 This incentivizes firms to move toward may outweigh the benefits of registering formally for com- non-salaried contracts, and the illegal evasion of social panies. Indeed, modeling an economy resembling Mexico, security, with consequences on productivity and growth. research shows that the impact of completely eliminating Settling labor disputes (based on formal employment con- registration is small, while improving access to credit for tracts) is also a long and expensive process. formal sector firms increases wages, TFP and output per worker while reducing the size of the informal sector.274 190. The valuation of social security contributions by work- ers also affects the level of formalization. Another source of 189. Other sources of resource misallocation are connected labor market segmentation is related to ancillary benefits, to the interaction of policies related to labor and social in- currently financed by social security contributions and surance regulations. There is evidence that the high share payroll taxes. Decisions on whether to seek a formal job of informal employment in Mexico is a result of both la- depend on the extent to which workers value additional bor market segmentation (preventing entry into formal income more than the benefits provided by social secu- SYSTEMATIC COUNTRY DIAGNOSTIC 273 See Kaplan et al. (2011) and Bruhn (2011). In addition, wage earner types are less likely to register their business (but more likely to become wage workers in the newly-created MEXICO jobs) while those with characteristics similar to formal business owners are more likely to do so (Bruhn, 2013). Yet, most informal business owners remain informal after the reform, suggesting that deregulation is not enough. Aparicio (2014) finds larger impacts of the reform on formalization, in addition to increased profits for some firms, particularly in munici- 103 palities that adopted the reform early on. 274 López-Martín (2015). 275 Levy (2009). 5. UNLOCKING rity contributions and payroll taxes. Research shows that tors. At the state level, contract enforcement ranges from MEXICO’S INCLUSIVE low-income workers are especially likely to favor imme- 160 days in Campeche to 455 days in the case of Tlaxcala. GROWTH diate income over long-term benefits.276 Ensuring good According to a World Bank study, on average, it takes 209 POTENTIAL quality of these services and aligning the services provid- days from the presentation of a commercial complaint to ed with their perceived cost can also help encouraging the final hearing, while commercial justice is not avail- formality. able outside of major urban areas. 279 A recent study finds that a 100-day decrease in the time it takes to resolve a commercial dispute is associated with an increase of 0.6 5.1.2. Rule of law institutions percent in the average GDP per capita growth rate in 191. The persistence of factor misallocation in Mexico is Mexican states.280 While this calculation may be overesti- closely linked with its poor performance in areas related to mated, the study provides evidence of the importance of rule of law and access to justice, control of corruption, and further investigating the links between the efficiency of crime and violence. Distortions in allocative efficiency can the judiciary system and growth and inclusion. The qual- result from the interaction of outcomes associated with ity of the legal system has also been shown to have an weak institutions. Diverse indicators suggest that the gov- impact on firm size in Mexico due to contractual uncer- ernance environment in Mexico is worsening, even as it is tainty: in states with low quality legal systems, firms tend improving in comparator countries.277 Strengthening the to be smaller, especially in capital intensive industries.281 rule of law, improving effectiveness and inclusiveness of Lack of rule of law has also been associated with limiting the judicial system, and ensuring citizen security will be business confidence, such as by inhibiting the definition key to set the foundations to reduce misallocation and of bankruptcy procedures in Mexico, with consequences foster productive inclusion. on investment and growth.282 194. A second wave of legal reform to civil and commercial Access to justice justice could help significantly. To incentivize a faster and 192. Uneven access to justice and the application of the rule more transparent resolution of lawsuits, an initiative ap- of law can have severe adverse effects on individuals and pri- proved by Congress in October 2017 increased the scope vate sector development. Mexico ranks low on internation- of oral procedures to resolve commercial disputes. Mov- al indicators for the rule of law: in the Rule of Law Index ing from written to oral trials can help improve the out- 2017–2018, for example, Mexico ranked in 92nd place out comes of economic disputes, e.g. those related to contract of 113 countries.278 Mexico is at the bottom of the pack of enforcement. Currently 26 states use oral trials for larger upper middle-income countries (34 out of 36 countries), commercial cases, and four states use them in civil cas- only above Turkey and Venezuela and below China, Rus- es.283 Nevertheless, most civil and commercial cases are sia, Colombia, and the DR. The country ranks particularly still handled using the unreformed justice system. Clear poorly in terms of absence of corruption (102/113), order differences, in addition, exist across states in the elements and security (99/113), civil justice (100/113), and criminal required for the implementation of the oral commercial justice (105/113). lawsuit (Figure 89).284 Evidence suggests, additionally, that the timeline for the resolution of cases by oral procedures 193. Mexico performs better on the specific measure of remains much higher than that established under the law; contract enforcement than many peers, though within and causes for the delays area similar to those under the the country, subnational rates vary considerably. A more written-based procedure.285 effective judiciary system where disputes are resolved in shorter periods can have substantial economic bene- 195. Impunity is also a challenge. Mexico ranked 58 of 59 fits. Weak rule of law has consequences for productivity countries in the Global Impunity Index (GII) published growth: uncertainty can lead to higher transaction costs, by UDLAP in 2015.286 There is some regional variation in reduced competition and barriers to entry of new firms. levels of impunity across states, though the high failure In 2016, the average number of days required to enforce to report crime—an indirect measure of impunity—cuts a contract in Mexico was 341, compared with 779 for Lat- across the country.287 According to the MCCI, the proba- in America, based on World Bank Doing Business Indica- bility of a crime being reported, investigated, prosecuted 276 Cunningham and Maloney (2001); Levy (2008); and, Antón, Hernández and Levy (2013). 277 “Governance constraints on Growth in Mexico” background paper to the SCD. 278 WJP (2018). Mexico scored 0.45 on the index, which measures adherence to the rule of law based on eight factors: constraints on government powers, absence of corruption, open government, fundamental rights, order and security, regulatory enforcement, civil justice and criminal justice. 279 Silva Méndez (2016). 280 Chávez et al. (2017). 281 Dougherty (2013); Laeven and Woodruff (2004). SYSTEMATIC 282 Bergoeing et al. (2007). COUNTRY 283 OECD (2016). DIAGNOSTIC 284 World Bank in collaboration with COFEMER, based on a set of 18 indicators (grouped along three dimensions: court performance, courtroom design and technology, and organiza- MEXICO tional framework) to measure the degree of implementation of the oral commercial lawsuit. In addition to differences across states (especially when comparing the upper and lower echelon), important within state variation exists, e.g. Puebla, which scores at the top in courtroom conditions and equipment and near the bottom in organizational framework. 285 Silva Méndez (2016); similar results are found by Silva Méndez and Caballero (2012). 286 The index considers three components of impunity. A structural component captures the institutional capacities of the police and judicial systems; the functional one reflects de facto 104 outcomes of the criminal justice system, while the human rights component assesses the protection of physical integrity. 287 In 2014, the GII ranged from 47.2 in the case of Campeche to 76.5 in Quintana Roo (with higher scores indicating greater impunity). The failure to report a crime ranged from a low of 82.6 percent in Coahuila to 94.2 in Guerrero. been associated with the rise and fall in violence in the 5. UNLOCKING Figure 89. States’ implementation of oral MEXICO’S country.291 INCLUSIVE commercial lawsuit GROWTH 0% 20% 40% 60% 80% 100% POTENTIAL 196. Beyond criminal offences, impunity in Mexico extends Guanajuato Nuevo León to civil and administrative crime. Mexico ranks 127 out Ciudad de México of 137 countries in diversion of public funds under the Zacatecas Morelos institutions pillar of the World Economic Forum’s global Nayarit competitiveness Index (2017–18).292 Less than 2 percent of Tabasco the denunciations of improper diversion of public funds Puebla Sinaloa issued by Mexico’s government auditing agency, the Su- Quintana Roo preme Federal Auditor (ASF), led to a conviction between Aguascalientes Querétaro 1998 and 2012, despite the fact that the number of cases Sonora referred by the auditor to the federal prosecutors office Baja California Sur Estado de México jumped from 2 to 134 a year during the period.293 Mexico Yucatán also scores well below its peers in terms of the constraints Oaxaca on government power.294 Between 2000 and 2013, 41 Chihuahua Hidalgo Mexican governors faced 71 corruption scandals; of these, Campeche 16 were investigated and only four were properly prose- Chiapas Michoacán cuted. In contrast, all nine U.S. state governors accused of Durango corruption during the same period were prosecuted.295 Tamaulipas Baja California Tlaxcala 197. Mexico performs worse than its peers in the extent San Luis Potosí to which access to justice is differential (rather than uni- Performance Organizational Courtrooms versal).296 Unequal application of justice benefits those Source: World Bank – COFEMER, 2017. Note: Graph shows distance to the reference. connected, typically at the top of the income distribution, further perpetuating inequality. In a field experiment con- ducted in Mexico City it was shown that officers requiring and resolved in the Mexican criminal system is only 2.95 a bribe are more likely to target lower income individuals, percent. Sixty-nine percent of young people aged 12 to 29 letting off more affluent drivers off with just warnings.297 who were witness to, victim of or involved in an act of gun This is likely the result of the association of wealth with violence reported that neither their community members the capacity to exert revenge on the officers, explaining or local police did anything in response to the incident.288 why officers are more likely to demand bribes from poorer Some raw data suggest that 99 percent of crimes in the individuals. In terms of guaranteeing fundamental rights, country do not lead to a criminal sentence.289 Insecurity the Rule of Law Index 2017–18 ranks Mexico 24 out of 30 and lack of accountability lead to citizens’ dissatisfaction, countries in the region.298 A key aspect of the unequal who, in turn, do not report crimes, leading to a vicious cy- application of justice refers to the enforcement of labor cle. Over one-third (38 percent) of individuals surveyed in rights, where improving the provision of services provided urban areas in 2017 by the World Justice Project reported by the labor courts is one of the main challenges faced by experiencing a legal problem in the last two years; yet only the 2016 labor justice reform.299 11 percent turned to an authority or third party to seek help resolving the issue, while 89 percent did not take 198. Despite reforms, issues persist in the capacity of ju- any action.290 As perceptions of corruption in the judicial dicial institutions to ensure order and security.300 A recent and law enforcement agencies increase, trust in these in- assessment prepared by the Procuraduría General de la tuitions declines: 90 percent of Mexicans believe that the República (Office of the Attorney General) outlined some police is highly corrupt. The public’s trust in judges has of the main challenges related to the design of a ‘single 288 ECOPRED, 2014. 289 “Governance constraints on Growth in Mexico” background paper to the SCD. 290 WJP General Population Poll 2017; survey conducted in Mexico City, Guadalajara, and Monterrey. 291 IEP (2018). 292 Schwab (2017). 293 Casar (2015). Between 1998 and 2012, only seven rulings were issued out of 444 criminal cases that were filed for corruption by the Supreme Federal Auditor to the Attorney General (PGR by its acronym in Spanish). 294 Mexico scores 0.47 on the constraints on government powers indicator of the WJP Rule of Law Index, lower than the regional average (0.55), and much behind the OECD (0.75). SYSTEMATIC 295 Montes (2015). COUNTRY 296 “Governance constraints on Growth in Mexico” background paper to the SCD. DIAGNOSTIC 297 Fried et al. (2010). MEXICO 298 Including the right to equal treatment and absence of discrimination, security, due process of the law, freedom of belief, right to privacy, freedom of assembly and association, and labor rights. Global Insights on Access to Justice: Findings from the World Justice Project General Population Poll in 45 Countries (WJP, 2018). 299 In 2016, as part of the Justicia Cotidiana initiative, Congress approved a presidential proposal to eliminate labor courts (the federal and state-level Juntas de Conciliación y Arbitraje) to 105 let the judiciary at the state and federal level resolve labor issues. Following approval, a special unit was created to oversee the transition. The unit and CONATRIB are currently on the process of designing the institutional framework for implementing the reform. 300 Ríos and Wood (2018). 5. UNLOCKING model of justice procurement’.301 These include the need is worsening over time compared with peers in the Latin MEXICO’S INCLUSIVE to improve the investigation stages of criminal procedures America and Caribbean region. 309 The World Economic Fo- GROWTH (where operators often fall back on outdated practices); rum’s global competitiveness index (2017–18) ranks Mex- POTENTIAL the lack of coordination between different subsystems; ico in 123rd place among 137 countries with a negative pervasive corruption, irregularities and malpractices (such trend. Corruption is seen as the most problematic factor as from abuse of office, omission of due diligence and un- for doing business by the private sector.310 While there are lawful detention); and a lack of appropriate skills and ca- differences within states, high levels of corruptions appear pacities. A particular area of concern is the criminal justice to be widespread throughout the country.311 system, where serious problems persist despite the recent reform process302, including arbitrary arrests, forced con- 201. Corruption increases the cost of doing business for fessions, falsification of evidence, wrongful conviction303 the private sector and constraints productive investments. and human rights abuses by police, prosecutors, and pris- Corruption is widely perceived to play a role in the daily on officials.304 conduct of business: 63 percent of business people “agree” or “totally agree” that corruption is part of the business 199. Unequal access to justice seems to be frequent in the culture in Mexico, while 54 percent affirm that corruption criminal justice system, contributing to maintain the cycle affects their daily business operations.312 About 65 percent between crime and poverty. Research suggests that the of entrepreneurs in Mexico report having missed a busi- system tends to over target the poor and vulnerable.305 ness opportunity due to undue competition, where com- Evidence shows that individuals who pay bribes receive fa- petitors use political influence or handouts.313 Fifty-seven vors (even including release); while those who are able to percent of these same business people concede having pay legal representation obtain better treatment.306 Long employed go-betweens (“gestores”) that have access to pretrial detention times are an example of how poverty is information or political connections, to intervene with au- perpetuated in the criminal justice system. Close to half of thorities on their behalf. A recent national study on regula- all “suspects” in Mexico are incarcerated awaiting trial.307 tory quality on enterprises shows that the number of firms This number appears to be growing, possibly influenced perceiving that corruption acts performed by public ser- by the pressure on the criminal justice system to show re- vants in 2016 were frequent or very frequent range from sults in the fight against crime. Yet, in addition to being a low of 62.2 percent in Colima to 92.7 in Tabasco. Of the costly, evidence suggests that these incarcerations are dis- corruption acts, 65 percent are realized to move bureau- proportionately imposed on the most disadvantaged and are often unjust. A large share of those awaiting trials over cratic processes along; 39 percent to avoid penalties or long periods are acquitted of all charges, released without sanctions, and 31 percent to obtain licenses or permits.314 charges ever brought against them, or convicted for minor crimes (although detention is meant for suspects of grave 202. Corruption and impunity affect the poorest the most crimes).308 through higher out-of-pocket expenses. Between 2007 and 2010, the average cost of a bribe for Mexican house- holds increased by nearly 20 percent (from Mex$138 to Control of corruption Mex$165).315 Furthermore, while on average, Mexican 200. Mexico performs weakly in control of corruption rank- households spent 14 percent of their income on bribes, for ings. Transparency International’s Corruption Perceptions low-income households the expense represented 33 per- Index, placed Mexico in 123rd out of 176 countries, scor- cent of their income.316 The poor and vulnerable are also ing last among OECD nations. Furthermore, Mexico’s rela- particularly susceptible to clientelism—the exchange of tive place in performance in control of corruption rankings political support for typically short-term benefits—given 301 PGR (2017). The complete list of opportunity areas refers to: 1) improving the early stages of criminal procedure (whereby the system presents backlog partly related to unprecedent- ed increases in the investigation files, and a lack of use of available tools); 2) lack of delegation of immediate attention units, such as toward (3) alternative mechanisms to resolve controversies; 4) human capital-related issues, including lack of appropriate skills and capacities; 5) normativity, including multiple layers of normativity leading to duplication and obstruction of the system, and incentives to continue operating under outdated rules; 6) dispersion and lack of coordination between subsystems, which often operate in silos; 7) high turnover in the position of attorney general; 8) corruption. 302 Starting in 2008, the transition to the New Criminal Justice System (Nuevo Sistema de Justicia Penal, NSJP) was due to be complete in all states by June 2016. The reform engages all aspect of the criminal justice system, from police to prisons; also affecting legal procedures in noncriminal cases (Ingram, 2018). 303 43 percent of judges surveyed indicate that prosecutorial police regularly engage in illegal arrests. 304 Cortés et al. (2017). See also Bergman et al. (2014) who find that about a third of prison inmates in Mexico City and the State of Mexico were solicited for bribes by the police; while 50 percent reported being struck or beaten physically to make a deposition or plead guilty (Bergman et al. 2014). 305 Azaola and Bergman (2009). 306 Bergman et al. (2014). 307 Zepeda (2005). 308 Zepeda (2005). Two thirds of all individuals convicted in 2002 received sentences associated with minor crimes. About 15 percent of criminal cases did not reach the verdict phase (due to reasons such as lack of evidence), and 14 percent were acquittals. Overall, 25 percent of people charged with a crime are acquitted by the courts and released. Moreover, five percent of individuals charged with misdemeanors—who have the right to be released on bail—remain in pretrial detention as they are too poor to pay for bail. SYSTEMATIC 309 As measured by the Transparency International Corruption Perception Index 2016. At 23, Mexico lies far below the second lowest OECD country (Turkey, ranked 75) and far below its COUNTRY Latin America and Caribbean peers: Brazil, Chile and Uruguay. DIAGNOSTIC 310 Schwab (2017). MEXICO 311 Perceptions of corruption range from a low of 69 percent of the population who believes that corrupt actions are frequent or very frequent in the government in Querétaro to 95 percent in Mexico City (IMCO 2016). 312 IMCO (2015). 313 IMCO (2015). 106 314 INEGI (2017a). 315 Transparencia Mexicana (2010). 316 Transparencia Mexicana (2010). their liquidity constraints and higher time preference for organizations affects the ability to conduct business, par- 5. UNLOCKING MEXICO’S the present.317 ticularly for small and medium enterprises, less able to hire INCLUSIVE private security or get special treatment from authorities. GROWTH POTENTIAL 203. Full implementation of transparency and accountabili- The rate of extortion has grown monotonically every year ty reforms can lead to better control of corruption. The Free- since 2002.323 Crime and violence have been reported to dom of Information Law, passed in 2002, was praised for cost Mexico about $40 billion each year.324 A recent esti- its thoroughness. Yet, progress in terms of results has been mate, taking into account the indirect costs of violence— slow, arguably because of the weakness in the judicial in the form of productivity shortfalls, foregone earnings system and institutional development at the subnational and distorted expenditure—and a multiplier effect, in level.318 According to the IDAIM index, by 2015, only the addition to direct costs, puts the total economic loss from federal government and one state had reached the opti- violence at about US$249 billion in 2017 alone.325 Accord- mal level of transparency established, while 14 states had ing to some estimates, drug-related crime has negatively reached only moderate performance, and 17 were rated affected income growth across Mexican municipalities deficient. Even though Mexico now ranks first among 133 between 2005 and 2010.326 countries on the Global Right to Information rating, 66 per- cent of citizens perceive transparency to be insufficient. In 206. The recent spike in drug-related homicides has been most cases, lack of effectiveness has been associated with associated with inequality and poverty. A recent study a gradual recapture of new institutions by political inter- shows that an increase of one point in the Gini coefficient ests or the bureaucracy, or because they were conceived translates into an increase of more than 10 drug-related as partial improvements without effective enforcement homicides per 100,000 inhabitants between 2006 and mechanisms. Public procurement and discretionary 2010.327 This result is found in the period where Mexico awarding for contracts at the local level appear to be at launched its war on drugs, when the proliferation of gangs particular risk of capture. Most recently, and as a result of lowered the marginal cost of criminal behavior, and the extensive social pressure, Congress has passed Constitu- expected pay-off of criminal activities increased. Evidence tional reforms and legislation to create the National Anti- also supports the hypothesis that crime increases the like- corruption System (NAS). Sustained mobilization from civil lihood of observing chronic poverty at the municipality society and the thorough prosecution of corruption cases, level in Mexico.328 Even violence that does not entail a loss can help ensure that the anticorruption system follows a in infrastructure, it is shown, has long-run effects on pov- more effective than the transparency law.319 erty incidence. The impact of crime on chronic poverty is nonlinear, and largest on the poorest municipalities. Crime and violence 207. Gender-based violence is often prevalent in the inter- 204. The lack of institutional effectiveness may also be re- section between poverty and inequality, and crime and vio- flected in the rise in crime and violence. Increased crime and lence. Femicide in Mexico has increased significantly since violence can affect the cost of doing business, presenting 2007. As many as seven women are killed every day and 63 a threat to private property. The World Economic Forum’s percent of them have experienced gender-based violence global competitiveness index (2017–18) ranks Mexico at some point in their lives. This increase has coincided 131st among 137 countries in terms of the business cost of with the expansion of drug cartels, and the rise in overall crime and violence and 131st both in terms of organized homicide rates in the country. In fact, femicides tend to be crime and the reliability of police services.320 Insecurity and higher in states with high presence of drug cartels such as crime are consistently rated as the top problem by Mexi- Guerrero and Chihuahua.329 In addition to the direct loss of can citizens, as well as the biggest obstacle cited by firms human lives, gender-based violence is associated with a to operating in Mexico.321 loss of productivity. 205. Crime and violence carry high social and economic 208. Increased crime and violence contribute to a worse allo- costs. The country has experienced increases in crime and cation of productive assets, including notably through their violence over the last decades.322 Extortion from criminal effect on labor market outcomes, with implications for equity 317 However, recent qualitative evidence from villages in Oaxaca suggest that some programs have become less susceptible to electoral effects; Progresa-Prospera appears to be more as an entitlement than a political favor shifting with the waves of elections (Díaz-Cayeros et al. 2016). 318 IMCO (2015). 319 “Governance constraints on Growth in Mexico” background paper to the SCD. Experience shows that institutional reforms, such as anticorruption agencies and transparency legisla- tion, tend to be effective only when supported by an engaged civil society (Heilbrunn, 2004; Kocaoglu and Figari 2006). 320 Schwab (2017). 321 INEGI (2014); World Bank Enterprise Survey (2010). 322 After declining under the administration of presidents Zedillo (1994–2000) and Fox (2000–06), the number of intentional homicides rose sharply after 2007, the first year in office of President Calderón (2006–12). Throughout the Calderón administration, INEGI (2014) reported 121,669 homicides. Following a three-year decline in 2012–14, homicides began to rise again in 2015. According to the National Security System (SNSP), the number of intentional homicide victims was 17,324 in 2015, 22,571 in 2016, and 27,734 in 2017. Over the first SYSTEMATIC five years of the Peña Nieto administration, it is estimated that approximately 116,000 people were murdered (Calderón et al. 2018). About a third to half of all homicides in Mexico are COUNTRY associated with organized crime (bearing signs such as the use of high-caliber automatic weapons, torture, dismemberment, and explicit messaging) (Calderón et al. 2018). DIAGNOSTIC 323 Ingram (2018). MEXICO 324 Jaitman et al. (2017). 325 IEP (2018). 326 Enamorado, López-Calva, and Rodríguez-Castelán (2014). 107 327 Enamorado et al. (2016). 328 Martínez-Cruz and Rodríguez-Castelán (2016). 329 SEGOB, INMUJERES, ONU MUJERES (2016). 5. UNLOCKING and growth. Crime and violence are affecting the accumu- percent of the revenue that it could theoretically collect if MEXICO’S INCLUSIVE lation and use of human capital, diverting away resources VAT was applied at the standard rate to all goods and ser- GROWTH from their highest valued use, via an unskilled young labor vices. By contrast, this VAT revenue ratio is 42.6 percent in POTENTIAL force. Recent evidence has found a high correlation be- Colombia, 55.1 percent in Peru and 64.4 percent in Chile. tween the share of NEETs and murder rates between 2008 and 2013, a period when the murder rates in Mexico tri- 211. Despite progress in streamlining tax expenditures, they pled.330 The correlation is especially large in states bordering remain high, affecting the availability of public resources. As the United States, which were particularly impacted both by discussed in Chapter 4, Mexico’s tax-to-GDP ratio is rela- organized crime and the economic crisis of 2008–09 due to tively low compared with regional and OECD peers. Tax ex- the destruction of job opportunities. By limiting the ability of penditures, including exemptions, deductions, deferrals, youth to acquire and use human capital, crime and violence and preferential rates applied to specific activities or types constitute an exclusionary force to productive activities. of taxpayers, are a key factor for low tax collections. SHCP The opportunity cost, and feedback loop, of this foregone estimates that tax expenditures cost the government 3.7 human capital in the long term, both at the individual and percent of GDP in foregone revenue in 2017 alone.332 Tax societal level, poses further concerns. Early labor experienc- expenditures are incurred, inter alia, through exemptions es away from formal employment such as unemployment to wage income and pension income, a negative income or an informal sector job, can have scarring effects over tax designed to encourage formal employment, exemp- employability and wages in adult life. For instance, evidence tions and zero-rating in the VAT regime and, since 2017, suggests that while 85 percent of high school graduates in through a discount to the excise tax on fuel. Mexico are able to stay in the formal sector once they are in, only 17.1 percent of high school graduates are able to find a 212. Tax expenditures not only reduce public resources, formal job after having worked in the informal sector.331 they can also have broader distributional and economic effects. Though often justified by worthwhile objectives, such as supporting the poor, creating employment or 5.1.3. Resource allocation and encouraging the growth of specific sectors or regions, tax institutional policy coordination expenditures reduce the resources available for budgetary 209. Inefficiencies in public resource allocation may result expenditures. Mexico’s tax expenditures in indirect taxes from tax and expenditure policy issues. The structure of the (including the zero-rated VAT on food and medicine) may tax system and tax expenditures contribute to low tax col- make the tax system more progressive – but this effect lections, limiting the overall amount of public resources. In- comes at the expense of the redistributive capacity of efficiencies may also be linked to labor and social insurance public expenditure.333 In countries with a progressive in- revenues or expenditure regulations and policies (and their come tax system, many income tax expenditures can act enforcement), special subsidies, the distribution of resources as “upside-down subsidies”, benefiting the richer income to subnational governments through the intergovernmen- groups (in the highest tax bracket) more than the poor- tal transfers system, as well as technical and allocative effi- er (many of which are not taxable).334 On the other hand, ciency across spending categories in the budget affecting tax credits designed to incentivize formal employment public service delivery (including investments). This section and targeted to low-wage and part-time earners could discusses some potential sources of inefficiencies based on increase the demand for low-skilled formal labor, under existing work of the World Bank and local researchers. certain conditions.335 In Mexico, the employment subsidy reduces the tax wedge for low-salary workers, offsetting some of the negative effects of regressive social security Tax structure and tax expenditures contributions. The effect on formal employment, however, 210. The tax structure is tilted toward direct taxes, partly re- has not yet been evaluated. Overall, by introducing differ- flecting significant tax expenditures in indirect taxes. Income ential treatment of taxpayers, tax expenditures make tax tax revenue makes up nearly 42 percent of Mexico’s total systems more complex and less productive. tax revenue, well above the averages for both the Latin America and Caribbean region and the OECD. By contrast, Public spending: rigidities, inefficiencies, and Mexico derives less than 40 percent of its revenue from in- distributional issues direct taxes, whereas these taxes account for over half of revenues in Latin America and Caribbean countries. Mex- 213. A large portion of public spending is rigid, limiting fu- ico’s tax structure reflects differences in its revenue-gen- ture flexibility and policy choices. By some measures, rigid erating capacity and has economic and distributional spending represents 80 percent of total spending. This implications. For example, due to broad exemptions and reduces the capacity of authorities to react to economic zero-rating in the VAT regime, Mexico collects only 31.5 shocks and changing country priorities through fiscal pol- SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 330 De Hoyos et al. (2015). 331 Calderón ( 2015). 332 SHCP (2017). 333 Scott (2014). 108 334 OECD (2010). 335 OECD (2015, 2016). However, the probability of positive effects is higher under certain conditions, including when the minimum wage is a binding constraint to formal employment (Pages, 2017). 5. UNLOCKING Figure 90. The role of taxes and transfers on redistribution in OECD countries, latest available MEXICO’S year INCLUSIVE GROWTH 45 Inequality reduction (%) POTENTIAL 40 35 30 25 20 15 10 5 0 -5 EST SWE ESP SVK POL DEU NOR PRT GRC CZE FRA AUT LUX DNK BEL FIN SVN IRL OEC MEX CHL TUR KOR JPN CHE NZL USA ISR LVA CAN AUS GBR ITA ISL HUN NLD Taxes and transfers Taxes Transfers Source: OECD 2017. icy. It also reduces the fiscal space for investments over (and their families) are affiliated to the Mexican Institute of time. For example, during the needed fiscal consolidation Social Security (IMSS). Federal government employees are of the last years, capital investment was squeezed by low- affiliated to the Institute for Social Security and Services for er public investment, given the inflexibility of other spend- State Employees (ISSSTE). The poor and unemployed are en- ing categories. Reducing fiscal space for investments to rolled in Seguro Popular. There are more than 15 million peo- historically low levels may affect medium-term growth. ple with two or more health services.336 The subsystems are fundamentally disconnected, offering different plans and 214. Public spending inefficiencies have been reduced in unable to take advantage of economies of scale, which re- recent years, but much work remains ahead. Reducing inef- sults in unequal access to services and significant beneficia- ficiencies is particularly important as social spending pres- ry overlaps and inconsistencies across insurance schemes. sures are likely to continue to grow. Mexico has a reasonable platform for providing social protection, and social assis- 216. Overall, Mexico’s system of social provisioning can be tance programs are generally well-targeted. However, Mexi- characterized as ‘dualistic’.337 Individuals in formal jobs have co’s large number of social assistance programs has resulted better access to non-salary benefits—pensions, health, in some degree of duplication, overlaps and fragmentation, housing—than informal workers. Though social programs reducing the effectiveness and efficiency of the system. for the latter have increased —notably regarding Seguro There are at least 5,491 social development programs across Popular, as discussed—the dualistic structure of social government levels. This complexity is heightened by the programs for formal and informal workers and unequal lack of a robust social programs census with unique identifi- access remain. There are indications that current spending cation number, as well as the absence of a fully operating in- in pensions in Mexico may be reinforcing inequality. Of all tegrated social information system to date. Federal and state pension spending in 2016, 95.3 percent is destined for 5.1 social programs often have overlapping beneficiaries. For million contributive pensioners—40 percent of whom are example, 23 state-level cash transfer programs focus on the less than 65 years of age, and the majority are men—and elderly alone. Many of these programs overlap with the re- 4.7 percent goes to the 5.4 million noncontributive pen- cently expanded federal noncontributory pension program sioners through the Programa de Adultos Mayores—where (Pensión para Adultos Mayores). As highlighted in section 60 percent are women over 65 years.338 4.1, there are a number of other areas where expenditure ef- ficiencies can be achieved including in public procurement, 217. Mexico’s overall fiscal system (taxes and transfers) the wage bill and several public sector programs. shows limited redistributive capacity and ranks at the bottom of OECD countries. Fiscal incidence analysis suggests that 215. One of the main obstacles to enhancing spending effi- some social programs and targeted direct cash transfers in- ciency and equity is the strong fragmentation in the health deed have contributed to progressivity in the fiscal system, sector. The national health system is composed of a series though their potential is hindered by the tax system, and of clearly differentiated institutions according to the seg- the relatively large share of resources allocated to low-re- ment of the population they serve, which is determined by distribution instruments—such as subsidies to contributo- the beneficiary’s employment status. Private sector workers ry social security systems, energy, and public tertiary edu- SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 109 336 Ministry of Health (2016). 337 Levy (2008). 338 Macias Sánchez (2017); Viesti (2015). 5. UNLOCKING MEXICO’S Box 11. Fragmentation and duality in Mexico’s health system INCLUSIVE GROWTH Widespread poverty, a fragmented health system, and an inadequate and inequitable distribution of financial, physical POTENTIAL and human resources inhibit access to quality health services, despite efforts to achieve universal health coverage. The impact of the health budget tends to be weakened by allocative and technical inefficiencies. High administrative costs (over 9 percent of total health spending in 2014, one of the highest in the OECD) limit the health sector’s invest- ment budget, resulting in a low rate of gross fixed capital formation—just 0.1 of GDP percent in 2015. This, in turn, reduc- es the availability of beds and infrastructure and contributes to an inadequate number of nurses, physicians, and health professionals.a The lack of coherent prioritization criteria within and across subsystems is also a cause of inefficiencies and inequities. As discussed in prior sections, out-of-pocket (OOP) spending remains high—likely reflecting dissatis- faction with the quality or accessibility of services provided by health insurance—and disproportionately impacts the poorest households. Even though health access has expanded, insurance coverage is unequal and often insufficient. E.g., an estimated 13.4 percent of the population lacks health insurance, though Seguro Popular still covers adverse health events.b And, notwithstanding the expansion in insurance coverage, income levels are closely correlated with health outcomes: e.g. infant mortality rates are 20 times higher in the poorer municipalities than in the least marginalized ones. The fragmentation of Mexico’s healthcare system appears as the critical obstacle to accessing care. Several disconnect- ed sub-systemsc provide healthcare, where employment status determines insurance coverage and access to provid- er networks.d Types of benefits and quality of care provided varies substantially, as do the level and composition of financing. Multiple public sources of funds finance care for the same person, which translates into an inefficient use of resources and high administrative costs for the system as a whole.e Vertically integrated insurance programs limit patient choice and weaken efficiency incentives. It also inhibits access to care, as patients can only use healthcare pro- viders within their insurer’s network, regardless of geographic location, service availability or medical condition, and is a key determinant of the high administrative costs. Fragmentation issues are compounded by differences in resource allocation and socioeconomic conditions across states.f Furthermore, the use of parallel, inconsistent data-collection and information systems complicates sectoral monitoring, oversight, and policymaking. The absence of system-wide planning produces an inefficient allocation of facilities and resources. Insurers continue to maintain their own pharmaceutical and medical-device supply chains and develop their own standards of care, which can vary substantially. Efforts to harmonize the activities of insurance providers have been only partially effective. Savings have been achieved through the consolidated purchasing of pharmaceuticals and the role of the General Health Council (Consejo General de Salud), established to coordinate the adoption of new pharmaceuticals and medical devices for six insurers.g Yet, the council’s effectiveness has been limited because, in practice, insurers still conduct internal reviews of medical products and make independent decisions regarding the definition of their prioritization criteria, benefit packages and use of medical devices and pharmaceutical—often based on budgetary constraints. The system has not been able to incorporate funds specialized in strategic purchasing, to shift away from historical budgets, partly because of coordination issues.h To achieve a separation of provider and purchaser function, necessary for the effectiveness of the system, purchasers must be inde- pendent from local political pressure, and have the ability to make contracting choices.i Despite being one of the objectives of the 2003 reform, there appear to be no clear incentives for the states or providers to improve health care performance.j Reallocating the public health budget to increase pro-poor expenditures, narrow or close gaps in coverage, and remove obstacles to access could boost the efficiency of health spending. Harmonizing benefit packages across insurance sys- tems could reduce disparities in the quality and comprehensiveness of care. Consolidating the affiliation processes across insurance systems could close the coverage gap for individuals moving between systems. The authorities could also strengthen, integrate, and expand the various databases used by different elements of the health sector. a. Mexico has 2.2 practicing doctors, 2.6 practicing nurses, and 1.6 beds per 1,000 people, well below the OECD averages of 3.3, 9.1, and 4.8, respectively. OECD (2016). b. ENSANUT, 2016. c. IMSS (Instituto Mexicano del Seguro Social) covers private-sector workers and their families. ISSSTE (Instituto de Seguridad y Servicios Sociales de los Trabajadores del Estado) covers federal government employees. Self-employed, non-salaried, and informal workers are covered by one of several federal programs managed by the Secretariat of Health, including Seguro Popular and the conditional cash-transfer program IMSS-Prospera. d. OECD (2016). e. López et al. (2015). f. The geographic distribution of public health resources does not reflect the needs of the population, as budgeting does not follow a fixed schedule, and high transaction costs reduce the bargaining power of regional providers and undermine service quality. Financial management capacity varies across states and is especially low in the poorest regions, contributing to the overall inefficiency of the sector (Knaul et al. 2012). h. González-Block et al. (2015). i. Lakin (2010). j. Kurowski and Villar-Uribe (2012). cation.339 When compared with peers in terms of the distrib- of 42.9 (compared with Brazil, where the Gini falls from 57.4 utive capacity of its fiscal system, Mexico (3 percent) ranks to 43.8 post-fiscal income, plus in-kind transfers).340 in the bottom among OECD countries (25 percent OECD average) (Figure 90). Following the Commitment to Equity 218. Regional disparities in per capita spending are substan- SYSTEMATIC (CEQ) Methodology, the fiscal system reduces inequality in tial and linked to the level and quality of public goods and COUNTRY DIAGNOSTIC Mexico less than in Argentina, Brazil, and Uruguay: from a services. Per capita accumulated spending of state and MEXICO Gini coefficient of market income of 51.1 to a final income local governments in the Northeast states is more than 20 110 339 Scott (2013). 340 Lustig et al. (2013). 5. UNLOCKING Map 6. Government expenditure per capita, accumulated 10 years, 2006–16 MEXICO’S INCLUSIVE 112 GROWTH POTENTIAL 170 145 182 153 135 155 148 171 133 163 127 132 109 111 130 183 123 128 216 1115 186 130 136 151 145 High expenditures Low expenditures Source: World Bank calculations based on data of the National Institute of Statistics and Geography. Map 7. Share of paved roads in total road network, 2015 0.2 0.3 0.4 0.6 0.3 0.7 0.3 0.4 0.5 0.5 0.3 0.5 0.5 0.3 0.4 0.6 0.5 0.5 0.7 0.6 0.6 0.4 0.3 0.3 More than 40% Less than 40% Source: World Bank calculations based on data of the National Institute of Statistics and Geography. percent higher than in the Southwest. Public spending in has followed an asymmetric fiscal decentralization, with a Mexico City is 35 percent higher than in Michoacán and deeper decentralization of spending and service delivery Guerrero (Map 6). A similar pattern is reflected in infrastruc- responsibilities than tax assignments, which have remained ture investment spending, with more developed regions highly concentrated at the federal level (Figure 92). In Mexi- in the Northeast investing more than lagged states in the co, the share of total spending of subnational governments South. Disparities in per capita spending are reflected in in- (around 50 percent) is much higher than their share of total frastructure services. The percentage of houses with access tax revenues (around 10 percent), resulting in the largest to running water in Guerrero and Chiapas is 46 and 50 per- vertical fiscal gap among OECD countries (Figure 91). The cent, respectively, while in Mexico City and Nuevo León it is reliance of subnational governments on transfers from the 90 and 96 percent, respectively. Differences in the quality federal government has negatively affected accountability of the states’ road network, measured as the percentage of and efficiency of subnational service delivery. In addition, paved roads, also show similar regional patterns (Map 7). the dependence on intergovernmental transfers does not create incentives to enhance subnational governments’ ef- 219. While significantly improved through the reforms of forts in collecting own revenues, which account for only 1 2007 and 2015, the intergovernmental fiscal framework con- percent of GDP. Mexico’s property tax (Predial) revenues SYSTEMATIC COUNTRY tinues to show significant vertical imbalances that affect pub- are low compared to those of other regional and OECD DIAGNOSTIC lic service provision and accountability at the local level. Fiscal countries. Revenue collected from these taxes amount to MEXICO federalism arrangements in Mexico are characterized by a 0.3 percent of GDP, below the 0.5 percent level observed large vertical imbalance that has dampened the efficien- in Argentina, Colombia, and Chile and far below the OECD cy of public spending. Over the last two decades, Mexico average of 1.9 percent. 111 5. UNLOCKING MEXICO’S Figure 91. Vertical fiscal gap, OECD and selected countries, 2016 INCLUSIVE GROWTH 80.0 POTENTIAL 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 ISL DEU CAN CHE AUT FRA ISR SWE USA ESP FIN PRT SVK SVN CZE IRL EST ITA LUX POL DNK BEL ARG NOR HUN KOR GBR NEP NLD MEX PER Source: OECD 2016. Note: Vertical fiscal gap is defined as the percent of total decentralized expenditures that are financed with transfers different to local taxation. Figure 92. Revenue and expenditure decentralization, OECD and selected federations 80 Percentage of General Government Spending CAN DNK 60 ChE SWE MEX KOr ESP uSA PEr FiN 40 bEL DEu NLD NOr POL NEPAL iTA AuT Gbr EST CZE iSL 20 huN FrA SvK SvN irL PrT iSr LuX 0 0 10 20 30 40 50 60 70 80 Percentage of General Government revenues Source: World Bank Mexico Public Expenditure Review 2016. 220. The intergovernmental transfer system has a weak effect of Participaciones, however, is still limited as it does equalization power to reduce horizontal disparities in terms not give higher per capita transfers to less developed of fiscal capacity, and the higher needs for services across regions with lower fiscal capacity or higher expenditure states. The spatial concentration of tax bases associated needs.342 Currently, Participaciones still maintain a strong with socioeconomic regional disparities results in hori- devolutionary nature as evidenced by the direct relation zontal fiscal imbalances and fosters further regional in- between per capita transfers and per capita GDP (Figure equalities in service delivery.341 This is not compensated 93). The amount of Participaciones per capita received by for by the transfer system. Participaciones (unconditional Mexico City, and the oil-producing states Campeche, is federal transfers to states and municipalities) are the main two times higher than that received by Chiapas, Guerrero revenue-sharing transfer mechanism intended to reduce and Oaxaca. the vertical and horizontal imbalances in Mexico. In 2007, the government changed the assignment of tax bases and 221. Recognizing these original limitations of the “Aport- adjusted the distribution criteria for Participaciones. The aciones” transfers, the distribution formulas for several positive changes to the distribution formula for Participa- sectors were improved. The distribution criteria for the ciones favored regional fiscal equalization and introduced Earmarked Transfer Fund for the Education Payroll and incentives to improve SNG tax collection. The equalization Operating Expenses (Fondo de Aportaciones para la Nómi- SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 341 State and municipal tax collections per inhabitant in the northeast and center north regions is 4.5 times the collection of state and local governments in the southwest. Mexico City collects 15 times more state and municipal revenues per capita than Chiapas, Oaxaca and Tlaxcala and more than 10 times than in Guerrero and Zacatecas, the states with the lowest subnational tax collections. 112 342 Since 2007, transfers have been distributed using a formula based on population, state GDP, and the state’s own tax collection effort, introduced gradually. As population becomes a dominant factor in the distribution formula, it is expected that Participaciones will favor some regional redistribution: taxes that are largely collected in the most developed regions of the country will be distributed on an equal per capita transfer basis. 5. UNLOCKING Figure 93. Participaciones and GDP, by state Figure 94. Aportaciones and GDP, by state MEXICO’S 500 500 INCLUSIVE GROWTH 450 450 POTENTIAL 400 400 350 350 Per capita GDP Per capita GDP 300 300 250 250 200 200 150 150 100 100 50 50 0 0 3.0 5.0 7.0 9.0 11.0 3.0 5.0 7.0 9.0 11.0 Participaciones per capita Aportaciones per capita Source: World Bank calculations. Data correspond to 2016. Source: World Bank calculations. Data correspond to 2016. na Educativa y Gasto Operativo, FONE)343 were modified in 222. While some Aportaciones have clear equalizing effects, 2007 to introduce some demand-side considerations. The the overall system of aportaciones does little to correct hor- new FONE distribution formula included the number of izontal fiscal imbalances (Figure 94). The Earmarked Trans- students enrolled in basic education, as well as indicators fer Fund for Social Infrastructure (Fondo de Aportaciones of education quality. A compensatory component was para la Infraestructura Social, FAIS) finances investment in also added to the formula to provide additional funds to social infrastructure by state and local governments via states with per-student transfers below the national av- the Social Infrastructure Fund of the States (Fondo para la erage, and the new formula included a factor designed Infraestructura Social de Estados, FISE) and the Social Infra- to encourage SNGs to invest their own resources in edu- structure Fund of Municipalities (Fondo para la Infraestruc- cation. In 2014, under pressure from states that had had tura Social de Municipios, FISM). FISM represents about their transfers reduced by the new formula and recog- 88 percent of FAIS. FAIS distributes 2.5 percent of federal nizing that the reforms had failed to eliminate perverse taxes (equivalent to 0.3 percent of GDP) to the states incentives, the government modified the FONE distribu- through a formula based on poverty indicators and unmet tion formula to include both a supply-side component basic needs. State governments then distribute funds to (the teacher payroll and school operating costs) and a municipalities according to a similar formula. However, demand-side component (the number of students in while FAIS clearly favors less-developed states, it does each state). Much like FONE prior to the 2007 reforms, the not necessarily favor less-developed municipalities, as a distribution formula for the Earmarked Transfer Fund for relatively poor municipality in a relatively wealthy state Health Services (Fondo de Aportaciones para los Servicios could receive less than a relatively wealthy municipality de Salud, FASSA) is driven by supply-side factors, includ- in a relatively poor state. Moreover, the loose definition ing the number of decentralized federal health workers of “investment in social infrastructure” and the fragmen- and the operating costs of federal health facilities in tation of resources reduce the efficiency of FAIS transfers, each state. While FASSA also incorporates equalization and a World Bank evaluation of FAIS found that it has had criteria, the distribution of resources is largely defined a limited impact on monetary and nonmonetary poverty by historical budget allocations, which in turn reflect indicators.345 In addition, the Earmarked Transfer Fund for the supply-side conditions that prevailed when health Strengthening Federative Entities (Fondo de Aportaciones services were decentralized. Because wealthier regions para el Fortalecimiento de Entidades Federativas, FAFEF) tend to have larger numbers of federal health workers, distributes 1.4 percent of federal taxes (or 0.2 percent of more-qualified health workers, and better health facil- GDP) to the states according to the inverse of their aver- ities, FASSA may have exacerbated regional disparities age per capita economic output. However, because FAFEF in health services.344 However, FASSA’s equalizing effects resources finance SNG debt obligations, pension liabilities, should be assessed within the context of other sources and institutional and technical capacity-building, the fund of health financing that interact with FASSA, such as the has no direct impact on regional fiscal disparities. universal health program, Seguro Popular. Seguro Popular distributes an amount equal to 0.4 percent of GDP among 223. The discretionary use of federal transfers (e.g., Ramo states based on the size of their uninsured populations. 23) may help undermine the credibility of the subnational The establishment of Seguro Popular and its rapid expan- fiscal-discipline framework. Registered under Ramo 23, SYSTEMATIC sion is gradually correcting the unequal and inertial re- extraordinary federal transfers managed through agree- COUNTRY gional distribution of FASSA transfers. ments between the federal and state governments ac- DIAGNOSTIC MEXICO 113 343 FONE was previously known as the Earmarked Transfer Fund for Basic Education (Fondo de Aportaciones para la Educación Básica, FAEB). 344 For more details on Seguro Popular, see Policy Note X on the health sector. 345 World Bank, (2017). El Efecto del Fondo de aportaciones para la Infraestructura Social FAIS en el Desarrollo Regional de México. Mimeo. Washington DC: The World Bank. 5. UNLOCKING count for only 6.7 percent of transfers, but they create an in Latin America and the Caribbean, recent research finds MEXICO’S INCLUSIVE ad hoc system, with limited efficiency and transparency in that among the top 15 fragmented Latin America and Ca- GROWTH the intergovernmental transfer system. They also intensify ribbean cities, three are in Mexico.346 There is also recent POTENTIAL budgetary unpredictability and may foster soft budget evidence that fragmentation of urban areas in Mexico constraints. affects the productivity of cities. The administrative frag- mentation of urban areas negatively affects the economic productivity for a sample of OECD countries (including Institutional coordination and investment planning failures Mexico) by between 5 to 8 percent.347 However, that im- pact is reduced by about 40-60 percent when a metro- 224. Lack of coordination between the public and private politan governance body is present. Earlier research had sector, and between government levels led to negative also suggested a positive association between income effects on the process of urbanization, although this is growth and the presence of an overarching decision-mak- changing. Lack of coordination between developers and ing mechanism, such as multijurisdictional, multipurpose authorities to guide and regulate affordable housing proj- regional governments.348 Improved market access can also ects and investments has led to poorly planned housing help raise city productivity. Mexico’s road investment over developments with low-quality services, often located far recent decades are associated with local job growth and from job opportunities. Over the past 30 years, the built- output, and with increasing specialization among manu- up areas of Mexican cities expanded sevenfold and the ur- facturing firms.349 banized areas of the 11 biggest metropolitan cities, nine- fold. This horizontal expansion has been driven mainly by 226. Limited coordination among public entities (including large single-use housing developments on the outskirts of among levels of government) and between public and pri- cities. The supply of low-cost housing increased by about vate sector, constraint the strategic planning of investments one million units each year between 2006 and 2011. and contribute to suboptimal outcomes. The planning and However, as housing developers sought to produce more prioritization of investments across the country is an area housing units while keeping land costs low, they increas- where improved coordination could make a big difference. ingly built on rural tracts far from city centers. These areas Strategic investments to support growth and inclusion were subsequently rezoned as urban land on a plot-by- could be better set, building long-term pipelines of proj- plot basis. But limited attention to the overall functionality ects and bringing private sector resources, under a stra- and accessibility of new developments on the outskirts of tegic plan that could go beyond administration periods. cities to basic services, including education, health, water, This would also help to the maximization of private sector transportation, energy, among other. These also resulted financing of infrastructure in the country, while guarding in higher costs for these services for poorer households for fiscal risks. Problems of coordinated policies can be ob- and a large disconnect from most job opportunities. New served among federal entities as well as between federal approaches to public housing inside urban areas, support- and sub-national entities (where there is the highest need ed by government agencies, are helping, though they will for investments). This also applies to the system of public only gradually change this trend. procurement that could be more coordinated and stream- lined to save resources. 225. The uncoordinated urban growth of Mexican cities has implications for employment dynamics, administrative frag- 227. Ensuring coordination, transparency and accountabili- mentation, and productivity. Urban growth has widened ty is critical for the efficient functioning of PPPs. Mexico has the distance between jobs and housing, undermining a Law on Public-Private Partnerships (Ley de Asociaciones cities’ ability to match skills to jobs. Between 2000 and Público Privadas) that was approved in 2012 and applies to 2010, population density dynamics within Mexican cities the federal, state and municipal level. The country scores shifted. Most cities experienced a significant drop in the well on international assessments of the institutional and number of people living in central areas, accompanied legal framework for PPPs.350 However, there is no agency by increasing population densities in urban peripheries. at the federal or subnational level that has exclusive re- Eighteen of Mexico’s largest cities lost more than 20 per- sponsibility for PPPs, which makes long-term planning cent of their central city population during the period. difficult.351 Furthermore, a recent stakeholder survey352 The lack of mixed-use development and diversified em- identified the lack of standardized contracts and risk ma- ployment subcenters also affected the cities’ ability to sort trices, and - despite the existence of a multi-year National economic activities in space. This kind of growth has led Infrastructure Plan - the lack of a long-term vision of PPPs to sprawl and high levels of administrative fragmentation. as areas for improvement. This long-term vision also needs Using the urban extents to assess fragmentation of cities to take into account the fiscal implication/risks of PPPs. In SYSTEMATIC COUNTRY DIAGNOSTIC 346 Mexico City, as the most fragmented city in Latin America and the Caribbean, spans over 76 administrative units; Puebla, in third place, spans over 38 units, while Oaxaca de Juárez MEXICO and Joinville in Brazil occupy the 14th and 15th place with 20 administrative units (Duque et al. 2017). 347 Ahrend et al. (2014). 348 For example, Foster (1993), and Nelson and Foster (1999). 349 Blankespoor et al. (2017). 114 350 See, for example, World Bank (2018). 351 Economist Intelligence Unit (2017). 352 World Bank (2018), Policy Note. 5. UNLOCKING Box 12. A tale of two Mexican states MEXICO’S INCLUSIVE The success of productivity-enhancing policies often depends on deeper underlying determinants related to the process GROWTH POTENTIAL in which state and nonstate actors interact to design and implement policies—and in the institutions they establish to help coordinate expectations, improve cooperation, and enable commitment.a Regional disparities can thus be traced back to differences in the capacity of actors and groups to influence the allocation of resources and the design of policies. Focusing on these interactions between influential groups, Kahn (2017) explains the divergence in economic perfor- mance in the Mexican states of Querétaro and Puebla. Located in the central region of the country, both states under- went a process of structural transformation in the 1960s, as industry decentralized from the capital; reaching similar levels of economic development by the 1980s. Over the last decades, however, Querétaro has enjoyed sustained eco- nomic growth and increasing productivity, while growth has remained volatile in Puebla, following stagnation in the 1980s. Additionally, whereas the economy in Querétaro has moved toward higher value added activities—including a leading aerospace sector—in Puebla the economy remains highly concentrated, specifically around auto assembly, which represents over 26 percent of the state’s GDP in 2008. In contrast, the top four industries in Querétaro together accounted for less than 25 percent of state GPD. Kahn argues that politicians and policymakers in the two states implemented different development strategies—and institutions—reflecting differences in the organization and characteristics of actors, particularly chambers of business. In Querétaro, the government strategy followed a model of cooperation and coordination with private sector actors, while in Puebla, where the business sector was viewed more as a rival for influence, the dynamic between public and private sector actors was more insular, tending toward cooptation and capture. The nature of alliances between private sector, labor and the government differed in both states. Querétaro established institutions, such as the tripartite commission, that facilitated cooperation between government, the private sector, and labor elites, in a de facto ‘developmental coalition’ on issues ranging from labor bargaining agreements to taxation and investment promotion strategies. Conversely, in Puebla, the association between business owners and the govern- ment tended to be politicized—subject to the electoral cycle—and strained, with actors from both sectors alternatively confronting or coopting certain factions, engaging in short-term alliances and exchange of favors. Attempts to estab- lish a tripartite consultative body were unsuccessful for the most part. Another factor refers to the way in which modern large companies engaged with local business groups in both states. Major national and multinational enterprises arrived in both states during the 1960s. In Querétaro, the top executives of these firms integrated into the local business chambers, bringing along a culture of entrepreneurship, quality and in- novation into the local associations. In addition, their participation increased the bargaining power of these chambers in negotiations, forming a worthy counterpart to the government. In Puebla, on the other hand, the new large firms sel- dom participated in business associations, which remained largely under the control of a few influential families, with a strong, conservative political identity. The lack of a variety of large firms in Puebla, especially multinational enterprises (with the exception of a single large automotive manufacturer) meant that local business associations remained largely undeveloped. In contrast, the diversity of economic activity in Querétaro implied that companies had to organize to engage with the government, providing incentives for cooperation. The differences in business chambers between states also affected their stance toward the reforms of the time. Business associations in Querétaro tended to be more supportive of free trade, foreign investment and productivity-enhancing policies, while in Puebla associations had a more protectionist outlook, calling for policies to ameliorate the effects of reforms. The elite agreements that took place thus were structurally different. Broad public-private cooperation in Querétaro allowed implementing coordinated, longer-term strategies toward investment, infrastructure, and human capital; and to manage labor relations, including in the aftermath of crises, with the help of informal norms of conciliation and compromise. Conversely, the relationship between government and business in Puebla was less cooperative—where attempts at economy-wide coordination often resulted in partisan fighting-- and were more narrowly concentrated on a handful of firms. Source: Based on Kahn 2017. a. World Development Report 2017 (World Bank 2017). the face of pressing near-term budget constraints, PPPs publishing these assessments alongside the cost-ben- can seem attracted for delivering infrastructure up-front efit analyses already made available on the Obra Pública while postponing payment by spreading it over the length Abierta database. In addition, systematic reporting on SYSTEMATIC of a long-term concession. At present, the authorities are longer-term budget commitments, terms in the contracts COUNTRY DIAGNOSTIC obliged under the PPP Law to prepare a comparative as- that could impact future payments or revenues, and other MEXICO sessment of financing a project via PPP vis-a-vis financing contingent liabilities should be part of the periodic bud- it through public procurement. To improve transparency get documentation and accountability on the totality of and accountability, consideration should be given to the public-sector PPP portfolio. 115 5. UNLOCKING MEXICO’S Figure 95. Prioritization process INCLUSIVE GROWTH POTENTIAL Filter 1: research and Filter 3: Empirical policy work review benchmarking Structural impediments Priority policy Diagnostic and other areas constraints Filter 2: Consulation Filter 4: Country with Wb teams and Knowledge stakeholders 228. Differences in institutional effectiveness across states rounds of consultations with international scholars in aca- may have contributed to the large disparities in growth demia and experts on Mexico, practitioners, government rates. Case studies show that the process through which authorities in various sectors, private sector representa- actors and groups interact to influence the allocation of tives, civil society, development partners, the World Bank, resources and the design of policies is an important factor and other external experts (Figure 95).354 Additionally, two that contributes to this divergence.353 For example, the broad rounds of consultations with stakeholders were divergence in economic performance between the states held in-country, in October 2017 and April 2018, including of Querétaro and Puebla may at least be partly linked to field visits to several states.355 the interaction between policymakers, chambers of busi- ness and politicians, which led to different institutional 230. After applying these two initial filters for prioritization, arrangements and business development in each state. twelve structural impediments were selected to conduct the In this context, limited local capacity for planning may second stage of the prioritization process. These twelve struc- exacerbate the challenges of coordination. Municipalities tural impediments are organized into four categories: (1) often lack planning capacity, or a strategic vision that con- product and factor market issues, (2) rule of law, (3) resource siders a territorial planning approach, and instead focus allocation and policy coordination, and (4) other structural on sectoral, frequently siloed, programs (Box 12). constraints to growth inclusion and sustainability. The first three categories are described in section 5.1. The fourth cat- egory consists of other relevant structural impediments to 5.2 Prioritization process: sustainable inclusive growth that, given their nature, could From structural not be classified into any of the other categories discussed impediments to priority in chapters 2–4. Table 8 presents the list of these 12 structur- al impediments organized into four categories. policy areas 231. To identify the highest priorities among the structur- 229. The SCD applied two first filters to identify the main al impediments over the medium term, the SCD used the structural impediments (discussed in section 5.1) and other results of a data-driven benchmarking exercise as a third key constraints (that also arise from chapters 2,3, and 4) that filter. This benchmarking exercise is designed to measure are holding back faster growth and inclusion with sustain- Mexico’s performance in comparison to the world and ability in Mexico. The first filter, used a research and policy selected structural peer groups of countries; the com- review of the extensive work done on Mexico by local and parators include (1) the world, (2) the Latin America and international scholars and practitioners, analytical work Caribbean region, (3) upper-middle-income countries, (4) that has been conducted by the World Bank and other OECD countries, and (5) selected (structural) peers.356 The international organizations, and research papers and pol- indicators of this benchmarking exercise were selected for icy reviews published in peer-reviewed journals and aca- each of the four categories and policy areas of structural demic and policy outlets. The second filter relied on several impediments discussed in the previous section (5.1) and, 353 World Bank (2017). 354 Teams also provided written inputs, focusing on (i) the most important analytical pieces and sources of their sector, (ii) the key development challenges for sustainable, inclusive growth, and (iii) the main challenges in their sectors. Consulted Global Practice teams included: Agriculture; Education; Energy Extractives; Environmental and Natural Resources; Finance, Competitiveness and Innovation; Governance; Health, Nutrition and Population; International Finance Corporation; Macroeconomic, Trade and Investment; Poverty and Equity; Social Development; Social Protection and Labor; Transport and ICT’s; Urban and Disaster Risk Management; Water; Gender CCSA; Jobs CCSA; and Climate Change CCSA. 355 During the first round of external consultations, 51 experts were asked to provide feedback during meetings organized by type of entity. In particular, 19 experts from the private sector (including industries such as banking, pharmaceutical and energy), 11 experts from the academia, 6 experts from think tanks, 8 participants from NGOs and other development partners, and 7 experts from the government provided feedback. Consultations in Tuxtla Gutiérrez, Chiapas, included the participation from local secretaries from the economic and SYSTEMATIC labor ministries, other government officials from the Ministry of Finance and Tourism, private sector representatives (agricultural, tourism, manufacturing and construction) and local COUNTRY business organizations (COPARMEX, CNIC). In Querétaro, consultations focused around the aerospace and automotive industrial clusters and several local Ministries including: DIAGNOSTIC Ministry of Sustainable Development, Ministry of Education, Ministry of Labor, Ministry of Social Development, Ministry of Infrastructure, Ministry of Finance, and the State Water MEXICO Commission. Full list of participants in Annex 4. 356 Peer countries were selected based on similar characteristics and under three groups: regional, aspirational and structural peers. Within the region, peer countries were selected based on their per capita GDP and population. Regional peers include: Chile, Peru, Colombia, Brazil, Argentina, and Uruguay. Worldwide, aspirational peers were defined as countries that had similar populations and per capita GDP in the 1990s but then experienced high economic growth, reaching levels 20 percent higher than Mexico. Aspirational peers include: 116 Korea, Malaysia, Poland and Chile. Finally, structural peers were selected for sharing unique elements with Mexico: upper-middle income countries, with a comparable GDP per capita (average between 2010-16), proximity to least to one large economy, similar performance in the logistic performance index, a similar percentage of manufactured exports and similar market size. These countries were Turkey, Thailand, Romania, and Argentina (for more details, see annex 1). Table 8. Main structural impediments to achieve sustainable inclusive growth in Mexico 5. UNLOCKING MEXICO’S INCLUSIVE Policy areas Structural impediments GROWTH POTENTIAL Concentration (and market power) in critical input markets and barriers to entry at the local level Product and factor market issues Access to finance Labor market rigidities and informality Access to justice Rule of law institutions Control of corruption Crime and violence Tax structure and tax expenditures Resource allocation and Public spending: rigidities, inefficiencies and distributional issues institutional policy coordination Institutional coordination and investment planning shortcomings Investment in infrastructure Structural constraints to growth, Quality and utilization of human capital inclusion and sustainability Management of natural capital Table 9. Summary table of selected indicators (distance to top performers by structural impediment) Income Peers OECD Region World Group 1. Concentration (and market power) in critical input markets and ● 71% ● 88% ● 50% ● 42% ● 55% barriers to entry at the local level 2. Access to finance ● 83% ● 89% ● 74% ● 68% ● 72% 3. Labor market imperfectios and informality ● 55% ● 72% ● 47% ● 45% ● 61% 4. Access to justice ● 78% ● 83% ● 62% ● 53% ● 66% 5. Control of corruption ● 70% ● 91% ● 64% ● 65% ● 73% 6. Crime and violence ● 90% ● 100% ● 84% ● 73% ● 88% 7. Tax structure and tax expenditures ● 52% ● 72% ● 63% ● 58% ● 59% 8. Public spending: rigidities, inefficiencies and distributional issues ● 72% ● 93% ● 59% ● 56% ● 50% 9. Institucional coordination and investment planing failures ● 85% ● 90% ● 72% ● 84% ● 75% 10. Investment in infrastructure ● 74% ● 93% ● 60% ● 60% ● 67% 11. Quality and utilization of human capital ● 73% ● 91% ● 62% ● 59% ● 62% 12. Management of natural capital ● 73% ● 62% ● 72% ● 67% ● 69% for other key constraints to growth and inclusion identi- 232. For each structural impediment or constraint, compa- fied in chapters above that are not tackled directly in sec- rable indicators worldwide were selected to measure Mexico tion 5.1. The benchmarking exercise did not include other against different benchmark groups. An initial inventory of sectors/policy areas that may be important for the country 80 indicators was constructed and shared for internal con- but do not represent binding constraints at the moment sultations, to reach a final list of 178 indicators that were according to the diagnostic and the process undertaken organized under each one of the 12 impediments (the full under filters 1 and 2; and does not include either items list of indicators included under each structural impedi- that might be critical, but knowledge gaps impeded fur- ment is available in annex 6).357 The analytical exercise then SYSTEMATIC ther analysis. calculates the (normalized) distance in percentage terms COUNTRY DIAGNOSTIC MEXICO 357 After consultations with teams a total pool of more than 450 indicators were assembled. Based on the consultations with World Bank experts, a pool of indicators related to each of 117 the structural impediments was gathered, allowing to compare Mexico’s position vis-à-vis the defined benchmark groups. The sources of the selected indicators include the Global Fin- dex (Global Financial Inclusion Database), World Bank, Washington, DC, https://globalfindex.worldbank.org/; Doing Business 2018; Global Competitiveness Report 2017–18 (Schwab 2017); World Justice Project rule of law Index; CPI unhabitat; human development data; Yale’s environmental performance index 2017; Transparency International 2017. 5. UNLOCKING between Mexico and the best performer for each indica- gaps in the labor market. Assessing the segmentation MEXICO’S INCLUSIVE tor. Then, the analysis calculates an unweighted average of of female labor force participation and its implications GROWTH the indicators grouped within each of the twelve structur- for productivity would allow a deeper understanding POTENTIAL al impediments to obtain a mean distance to the frontier of the barriers that women face in the labor market. for each category. These mean distances are calculated for •• Corruption is closely associated with resource misal- the latest year available and for all benchmark groups. A location and appears to disproportionately affect the larger mean distance for any given structural impediment lowest income groups. Gaining a better understanding implies a larger performance gap between Mexico and the of the economic and distributive costs of corruption, top performer of a specific benchmark group (Table 9).358 using more rigorous methodologies, can help generate better evidence to design and support policies towards 233. Additionally, the SCD used the World Bank country and strengthening the rule of law. global experts’ knowledge to calibrate the prioritization of •• Assessing the distributive impacts of the current policy areas as the fourth filter. For example, while the quan- degree of access to the judiciary system in Mexico. titative benchmarking exercise suggests that labor market There are no rigorous estimates of the welfare loss imperfections and informality is only of moderate priority due to justice inaccessibility, though those at the low when measured relative to comparators, the qualitative end of the income bracket often suffer from limited evidence and recommendations by experts consulted sug- access to justice, where this unequal access to the ju- gests that this is a high-priority area for Mexico. Similar case diciary system in Mexico appears to be perpetuating for public spending issues which may appear to be moder- inequality. ate but when compared with OECD countries the distance •• Carrying out profile analysis of crime victims and the to the frontier is quite significant. Experts also suggested geographic distribution of crime, including within-cities significant territorial and horizontal inequalities of public violence. Future analytical efforts should also look deep- spending as high priority. Table 10 highlights the results of er into gender-based violence, the impact of crime on the final prioritization results following the application of productivity, and inequality in the penitentiary system. this fourth filter. It is important to point out that this list of •• Analyzing internal migration trends, the profile of mi- priorities does not mean that other issues, excluded from grants, and the push-pull factors of migration within the list, are not important. Rather, the aim is to provide a Mexico. This is particularly important given the rele- sense of priorities and policy directions, based on existing vance of internal mobility to reduce misallocation, as knowledge and the filters discussed in this section derived well as in relation to risk diversification and crisis re- from the diagnostic undertaken for this report. sponse. •• Gaining a more thorough understanding of the degree 234. Moreover, the priority policy areas do not include of inefficient spending due to duplicity and overlaps of strictly macroeconomic stability policies, given the excellent programs, and the fragmentation of the social protec- track record of Mexico in this area. However, the diagnos- tion system. The need for a more precise understand- tic assumes and sets prudent and sustainable fiscal and ing is clear considering that there are at least 5,491 so- monetary policies as a critical pre-condition for growth, cial development programs across government levels inclusion, and sustainability. in Mexico. •• Carrying out additional analyses on groundwater (or lack thereof ) and its impacts on health outcomes. 5.3 Knowledge gaps and data Current empirical research on groundwater and its limitations359 relationship with health outcomes is scarce, but in- creasingly required considering the growing stress on 235. A number of knowledge gaps have been identified in the Mexico’s water resources. process of preparation of this SCD, including the following: •• Analyzing the impact of electricity subsidies on groundwater extraction/exploitation. It is particularly •• Identifying and analyzing the structural obstacles that important to investigate and understand the degree to contribute to the low productivity in the agricultural which these subsidies might create unintended long- sector and how to overcome them. Currently, the ag- term environmental consequences. ricultural sector—particularly in southern states—con- •• Conducting a rigorous assessment of the consequenc- tributes little to aggregate productivity in Mexico. es of tourism and environmental degradation; partic- •• Conducting additional quantitative work to understand ularly the growth in beach tourism and the increasing the potential economic gains of closing the gender forest cover loss in years to come. SYSTEMATIC 358 The distance is defined as a range between 0 and 100; where 0 denotes that Mexico is part of the top 5 percent best achievers, and 100 implies that Mexico is part of the bottom COUNTRY 5 percent worst performance. Any value between 0 and 100 indicates Mexico’s position somewhere between the best and weakest achievers. The 5th and 95th percentile were DIAGNOSTIC used to define the threshold for the top and bottom performance, with the objective to reduce possible distortions caused by outliers. The performance of Mexico for each indicator MEXICO and the benchmarking group was classified as high (red), medium-high (yellow), medium-low (grey) and low (green). High indicated that the difference between Mexico and the best performer in the group was 75 percent or more of the gap between the top and lowest performers. Medium-high indicated that Mexico’s difference with the top performer was between 50 and 75 percent of the gap with the lowest performer. Medium-low indicated that the difference was between 25 and 50 percent, and Low indicated that the difference was less than 25 percent of the gap with the lowest performer. The overall score of each impediment is equal to the unweighted average of Mexico’s performance in the selected indicators 118 for each benchmark group. In general, Mexico has on average a medium performance in comparison the world and upper-middle income countries but presents low performance in comparison of the selected peers and OECD countries. 359 More detail on data sources and data limitations in Annex 5. Table 10. Priority policy areas 5. UNLOCKING MEXICO’S INCLUSIVE Policy areas Structural impediments GROWTH POTENTIAL 1. Concentration (and market power) in critical input markets and barriers to entry at the local level •• Promote (and strengthen) regulation and supervision to curb concentration (and market power) in critical input markets and complete implementation of product market reforms. •• Address regulatory failures particularly at the local level (reducing the costs of firm establishment and operation), reduce protection for incumbents, remove preferential treatment for politically-connected firms, improve business environment (including protecting SMEs from crime and extortion, removing informal fees). •• Strengthen SME capabilities to link with large companies that work within GVCs. 2. Access to finance •• Incentivize lending to MSMEs, including by establishing the legal framework for instruments such as Asset-Based Lending. Product and factor markets •• Strengthen mechanisms to foster financial inclusion, including through new technologies (e.g., fintech) and with emphasis in rural areas. •• Promote the development of the domestic capital market (expanding to a broader set of firms) by fostering competition. •• Reform state banks to provide them more developmental oriented goals. 3. Labor market rigidities and informality •• Reduce the costs of formalization for firms and workers, gradually severing the link between payroll taxes and social insurance programs, cutting the costs for hiring and firing, and reducing the length of legal procedures in labor courts. •• Strengthen programs targeted at firms to improve formalization and entrepreneurship. •• Strengthen the relationship between the education system and the private sector to equip workers with the skills demanded by employers. 4. Access to justice •• Increase access to justice for vulnerable populations. •• Improve litigation times of most frequent cases, such as debt cases, wrongful dismissals, non-violent drug-related offences. •• Accelerate the implementation of reforms to enhance commercial justice. •• Improve contract enforcement and enforcement of property rights. 5. Control of corruption Rule of law institutions •• Implement aggressive legislation to fight corruption (e.g., public-private contracts, public procurement, reduce cash transactions between citizens and public servants). •• Fully implement the National Anti-Corruption System and extend reforms to the subnational level. •• Fully apply OECD’s anti-corruption convention. 6. Crime and violence •• Implement programs to promote social cohesion and support youth-at-risk, including youth employment programs. •• Strengthen and hold accountable institutions in charge of providing public safety and preventing crime. 7. Tax structure and tax expenditures Gradually increase revenue mobilization through: base broadening, tax rates where needed, and tapping undertaxed bases (digital economy and subnational), while considering distributional impacts. Adjust the tax structure and reduce tax expenditures to gradually increase the share of indirect taxes (while reducing payroll taxes). Reduce collection gaps, through the modernization of tax administration tools and stronger voluntary compliance measures. 8. Public spending: rigidities, inefficiencies and distributional issues Resource allocation and •• Reduce public spending inefficiencies to create fiscal space for infrastructure (e.g., room for efficiencies could be found in institutional policy public procurement, wage bill, consolidation of public sector programs; reducing fragmentation in the health system). coordination •• Reduce existing spending rigidities (pensions, wages, other entitlements). •• Reduce dependency of payroll taxes for social insurance programs. •• Reform the current pension systems to ensure sustainability and promote adequacy and equity. •• Explore ways to reduce overlaps and expand reach of the social protection system to the poorest and most marginalized. •• Reduce vertical and horizontal fiscal gaps in subnational governments. •• Strengthen equalization capacity of the intergovernmental transfer system (while applying stronger incentives for fiscal effort and curtailing ad hoc transfers) to reduce regional inequalities in service delivery and outcomes. •• Build larger fiscal buffers to use during difficult times. SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 119 5. UNLOCKING MEXICO’S Policy areas Structural impediments INCLUSIVE GROWTH 9. Institutional coordination and investment planning shortcomings POTENTIAL Resource •• Enhance coordination between the public and private sector, especially at the subnational level, with emphasis on productive allocation and investment. institutional policy •• Enhance public investment management process, starting with long term strategic planning (beyond a government period), the coordination development a solid pipeline of projects with an enhanced feasibility analysis process and cross-institutional coordination. •• Promote integrated multisector urban planning and service provision. 10. Investment in infrastructure •• Raise investment in infrastructure, including through private sector participation. •• Further strengthen and streamline the PPP framework, while managing fiscal risks. •• Invest in transport, logistics and trade facilitation to strengthen the Pacific side export corridors (including gulf to pacific corridors), as well as to achieve higher efficiencies domestically. •• Reduce vulnerability of existing infrastructure and integrate concept of resilience into new investments. •• Expand access to services in lagged regions. •• Invest in water infrastructure modernization and electricity transmission capacity. 11. Quality and utilization of human capital Other structural constraints to •• -Support youth in making effective school-to-work transitions. growth, inclusion •• Ensuring basic learning with a particular focus on closing gaps in attainment outcomes. and sustainability •• Promote universal health care reform based on a standard benefit package that promotes horizontal and vertical integration of services. •• Strengthen the primary health care system with a focus on prevention and promotion. •• Eliminate barriers that hinder participation of women in the labor market focusing on access to quality childcare and promoting gender-neutral flexible work arrangements. 12. Management of natural capital •• Adopt long-term planning and prioritization of investments in water security. •• Strengthen effectiveness of current support to agriculture, forestry and other productive, resource-based sectors by focusing on long-term productivity and competitiveness. •• Build resilience to deal with climate change and extreme events and foster climate-smart growth. SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 120 References REFERENCES Aburto, J.M., H. Beltrán-Sánchez, V.M. Garcia-Guerrero and V. Antoni, F. 2016. The rise and rise of Mexican Fintech. Tech- Canudas-Romo. 2016. “Homicides in Mexico Reversed Life Ex- crunch. Retrieved on March 6, 2018 from: https://techcrunch. pectancy Gains For Men. com/2016/07/28/the-rise-and-rise-of-mexican-fintech/ Agencia EFE. 2017. Aparicio, G., 2014. “Does Formality Improve Firm Performance? Aguilar, A. and Vicarelli, M., 2011. El Nino and Mexican children: Evidence from a Quasi-Experiment in Mexico.” Boston Univer- medium-term effects of early-life weather shocks on cognitive sity Working Paper. Boston, MA: Boston University and health outcomes.  Cambridge, United States: Harvard Uni- Appendini, K., 2014. “Reconstructing the Maize Market in Rural versity, Department of Economics. Manuscript. Mexico”. Journal of Agrarian Change, Vol. 14 No. 1, January Ahmad, M.E., Garcia-Escribano, M.M., Anaya, J.A.G., Revilla, E., 2014, pp. 1–25. Brosio, M.G. and Lockwood, M.B., 2007. Why Focuson Spending Araujo, Jorge Thompson, Ekaterina Vostroknutova, Konstantin M. Needs Factors? the Political Economy of Fiscal Transfer Reforms in Wacker & Mateo Clavijo (eds.) (2016), Understanding the In- Mexico (No. 7-252). International Monetary Fund. come and Efficiency Gap in Latin America and the Caribbean. Ahmed, S.A., Diffenbaugh, N.S. and Hertel, T.W., 2009. Climate Directions in Development series. World Bank. volatility deepens poverty vulnerability in developing coun- Arias, J., Azuara, O., Bernal, P., Heckman, J.J. and Villarreal, C., tries. Environmental research letters, 4(3), p.034004. 2010.  Policies to promote growth and economic efficiency in Ahrend, R., et al., 2014. “What Makes Cities More Productive? Mexico (No. w16554). National Bureau of Economic Research. Evidence on the Role of Urban Governance from Five OECD Arrossi S, Matos E, Zengarini N, Roth B, Sankaranayananan R, Par- Countries”, OECD Regional Development Working Papers, No. kin M., 2007. The socio-economic impact of cervical cancer on 2014/05, OECD Publishing, Paris. patients and their families in Argentina, and its influence on Alatorre-Troncoso, A., 2014. “Mexico’s national payments for radiotherapy compliance. Results from a cross-sectional study. ecosystem services programme: In the wrong place at the Gynecol Oncol.;105 (2):335-40. right time”. Master of Science Thesis, Imperial College Lon- Aterido, Reyes; Hallward-Driemeier, Mary; Pages, Carmen., 2011. don. Does Expanding Health Insurance beyond Formal-Sector Albo M., A. and Morales E., F. J., 2012. “Penetration of Credit in Workers Encourage Informality? Measuring the Impact of Mexico: Evolution and Comparison with Some Latin American Mexico’s Seguro Popular, Policy Research working paper, WPS Countries.” Mexico Economic Watch. BBVA Research. 5785. World Bank. Alcaraz, C., D. Chiquiar, and A. Salcedo. “Informality and segmen- Atkin, David, Benjamin Faber & Marco Gonzalez-Navarro. 2016. tation in the Mexican labor market”. Working Paper No. 2015- Retail Globalization and Household Welfare: Evidence from 25. Banco de México, 2015. Mexico. NBER Working Paper No. 21176. Alesina, Alberto, and Dani Rodrik., 1994. “Distributive Politics and Attanasio, O. P., Meghir, C., and Santiago, A. 2012. Education choic- Economic Growth’’, Quarterly Journal of Economics 109 (1994), es in Mexico: Using a structural model and a randomized ex- 465–490. periment to evaluate Progresa. The Review of Economic Studies, Alix-Garcia, J.M., K.R. Sims, V.H. Orozco-Olvera, L. E. Costica, J.D. 79(1):37–66. Fernández Medina, and S. Romo Monroy. 2018. “Payments for Atuesta, L. H., Ibarra-Olivo, J. E. Lozano-Gracía, N., and Deichman, environmental services supported social capital while increas- U. 2018. Access to employment and property values in Mexi- ing land management”, Proceedings of the National Academy of co. Regional Science and Urban Economics. Forthcoming. Sciences, 115(27), 7016-7021. Azaola, E. and Bergman M. 2009. “Encuesta a población en re- Alix-Garcia, J. M., Sims, K. R., Yanez-Pagans, P., & Shapiro, E. N. clusión.” Mexico City: CIDE. (2012). Two-dimensional evaluation: the environmental and Bacha, E.L. and Bonelli, R., 2015. Accounting for Brazil’s Growth socioeconomic impacts of Mexico’s Payment for Hydrological Experience–1940-2002. Services program. Balmori de la Miyar José Roberto, 2016. “The Economic Conse- ALLVP Research, 2016. Top Latam VC Investors in Fintech. Re- quences of the Mexican Drug War,” Peace Economics, Peace trieved on April 01, 2018 from https://docs.google.com/ Science, and Public Policy, De Gruyter, vol. 22(3), pages 213- spreadsheets/d/1QDNBw9Zz-QVrGfU4yh76A5U6Ug- 246. DA75z2gExgQcKNwAA/edit#gid=1152930736 Banxico. N.d. Costo anual Total (CAT). Retrieved from: http://www. Angelucci, M., and O. Attanasio., 2009. “Oportunidades: Program banxico.org.mx/ley-de-transparencia/consultas-frecuentes/ SYSTEMATIC Effects on Consumption, Low Participation, and Methodolog- costo-anual-total.html COUNTRY ical Issues.” Economic Development and Cultural Change 57 (3): Bargain, O. and Kwenda, P., 2011. Earnings Structures, Informal DIAGNOSTIC 479–506 Employment, and Self-Employment: New Evidence from Bra- MEXICO Angelucci, M., D. Karlan, and J. Zinman., 2015. “Microcredit Im- zil, Mexico, and South Africa. Review of Income and Wealth, pacts: Evidence from a Randomized Microcredit Program 57: S100–S122. Placement Experiment by Compartamos Banco”, American Economic Journal: Applied Economics, 7(1): 151-82. Barraza-Lloréns, Mariana; Giota Panopoulou; Beatriz Yadira Díaz., 2013. Income-related inequalities and inequities in health and 121 REFERENCES health care utilization in Mexico, 2000 – 2006. Rev Panam Sa- Carrera Chávez, Benjamín, Carrillo Carrera, Judith. 2016. Pobreza lud Publica; 33(2):122–30. rural y Procampo: Un análisis comparativo de dos municipios Barron, M. A. and F. Rello. 2000. “The Impact of the Tomato Agroin- de Chihuahua. Ciencia Ergo Sum, 23. dustry on the Rural Poor in Mexico.” Agricultural Economics 23 Casar, María Amparo. 2015. “México: Anatomía de la Corrupción”. (3): 289-297. CIDE-IMCO, México, 2015. Bartelsman, E., J. Haltiwanger, and S. Scarpetta. 2013. “Cross-Coun- Castellanos, S., Garrido, D. and Mendoza, A. 2008. Importancia de try Differences in Productivity: The Role of Allocation and Se- la Ley para la Transparencia y Ordenamiento de los Servicios lection.” American Economic Review 103 (1): 305-34. Financieros para Propiciar el Uso de Sistemas y Medios de Beade, A, and F. Garcia. (2009, January 12). Brújula de compra Pago Eficientes en México. El Trimestre Económico, vol. LXXV de Profeco. https://www.profeco.gob.mx/encuesta/brujula/ (1), núm. 297, pp. 224-251. bruj_2009/bol108_entre_empeno.asp CEFP. 2007. “Nota Informativa: Antecedentes, evolución y distri- Bergman, M, Fondevila, G., Vilalta, C., and Azaola, E. 2014. “Delito y bución del Fondo de Aportaciones para Educación Básica y Cárcel en México, deterioro social y desempeño institucional.” Normal (FAEB)”. H. Cámara de Diputados, Centro de Estudios Mexico City: CIDE. de las Finanzas Públicas, notacefp/047/2007. Available at: Bergoeing, Raphael, Patrick J. Kehoe, Timothy J. Kehoe, and Rai- http://www.cefp.gob.mx/notas/2007/notacefp0472007.pdf mundo Soto, 2002. A Decade Lost and Found: Mexico and CEPF. 2017. “Criterios que se utilizan para la asignación presu- Chile in the 1980s. Review of Economic Dynamics 5, 166-205. puestal a entidades federativas y municipios en la Ley de Blankespoor, B., T. Bougna, R. Garduno-Rivera, and H. Selod. 2017. Coordinación Fiscal”. H. Cámara de Diputados, Centro de Es- “Roads and the Geography of Economic Activities in Mexico.” Pol- tudios de las Finanzas Públicas, eecefp/004/2017. Available at: icy Research Working Paper 8226, World Bank, Washington, DC. http://www.cefp.gob.mx/publicaciones/documento/2017/ Bloom, N.A., R. Lemos, R. Sadun, D. Scur, J. Van Reenen (2016), “In- eecefp0042017.pdf ternational Data on Measuring Management Practices”, Amer- Chávez, Juan Carlos; Fonseca, Felipe and Gómez-Zaldívar, Man- ican Economic Review, Vol. 106, Issue 5, Pages 152-156. uel. 2017. “Resoluciones de disputas comerciales y desem- Bolio, E., Remes, J., Lajous, T., Manyka, J., Rosse, M., Ramirez, E. peño económico regional en México,” Ensayos de Revista de 2014. “A tale of two Mexicos: Growth and prosperity in a two- Economía, 36 (1): 79-93. speed economy,” McKinsey Global Institute, London. Chávez, H. (2014, September 11). Alimentos son hasta 630% Bonilla-Cacin, M. and N. Aguilera. 2013. The Mexican Social Pro- más caros por intermediarios. El Financiero. Retrieved from tection System in Health. UNICO Studies Series No.1. World www.elfinanciero.com.mx/economia/alimentos-son-has- Bank, Washington D.C. ta-630-mas-caros-por-intermediarios Boone, J., 2008. A new way to measure competition. The Economic Chiquiar, D. and M. Ramos-Francia. 2009. “Competitiveness and Journal, 118(531), pp.1245-1261. Growth of the Mexican economy,” Banco de Mexico, Working Bosch, Mariano and Raymundo M. Campos-Vázquez. 2014. The Papers, No. 2009-11, November 2009. Trade-Offs of Welfare Policies in Labor Markets with Informal CIEP 2016. Gasto Público para una Educación de Calidad. Jobs: The Case of the “Seguro Popular” Program in Mexico. CONAFOR. 2014. Programa Anual de Trabajo. Gerencia de Servi- American Economic Journal: Economic Policy. Vol. 6, No. 4, pp. cios Ambientales del Bosque. Mexico. 71-99. CONDUSEF, 2017. Comparativo de Tasas de interés. INFONACOT us- Bruhn, M. 2011. “License to sell: the effect of business registration ing data from CONDUSEF (2017). Retrieved from: http://www. reform on entrepreneurial activity in Mexico” Review of Eco- fonacot.gob.mx/TransparenciaRendicionCuentas/transparen- nomics and Statistics, 93(1): 382-86. ciafocalizada/Documents/ComparativoJunio2017.pdf Bruhn, M. 2013. “A Tale of Two Species: Revisiting the Effect of CONEVAL. 2012. Avances y Retos de la Política de Desarrollo Social Registration Reform on Informal Business Owners in Mexico” en México 2012, México. Journal of Development Economics 103: 275-83. Conference Board (2017), Total Economy Database (November 2017). Bruhn, M. and I. Love. 2014. “The Real Impact of Improved Access https://www.conference-board.org/data/economydatabase/ to Finance: Evidence from Mexico.” Journal of Finance, 69:3, Córdova Guzmán, Jassel & Aldo F. Ponce. 2017. “Los tipos de cor- 1347-1376. rupción y la satisfacción con los servicios públicos. Evidencia del Busso, M., Fazio, M. V., & Algazi, S. Levy. 2012. (In) formal and (un) caso mexicano. Types of corruption and satisfaction with public productive: the productivity costs of excessive informality in services. Mexican case evidence”. Región y sociedad, no. 70. Mexico. Cortés, N. G., O. Rodríguez Ferreira and D. A. Shirk. 2017. Perspec- Bustos, P. B. Caprettini and J. Ponticelli. 2016. “Agricultural Pro- tives on Mexico’s Criminal Justice System: What Do Its Opera- ductivity and Structural Transformation. Evidence from Brazil” tors Think? 2016 Justiciabarómetro. San Diego, CA: Justice in American Economic Review, 106 (6), 1320-1365. Mexico. Cadena, Kiyomi. 2012. The Impact of Retail-Owned Banks on Council on Foreign Relations (2018), Criminal Violence in Mex- Employment and Entrepreneurship: Evidence from Mexico, ico, March 30. https://www.cfr.org/interactives/global-con- Master’s Thesis, Faculty of the Graduate School of Arts and flict-tracker#!/conflict/criminal-violence-in-mexico Sciences of Georgetown University. Cuberes D. and M. Teignier, “Aggregate Effects of Gender Gaps in Calderón, L. 2018. An Analysis of Mayoral Assassinations in Mex- the Labor Market: A Quantitative Estimate,” Journal of Human ico, 2000-17. Justice in Mexico. Working Paper Series, Volume Capital 10, no. 1 (Spring 2016): 1-32. 15, Number 1. De Hoyos, Rafael, Carlos Gutiérrez, and Vicente Vargas. 2015. “The Calderón, L., Rodríguez Ferreira, O., and Shirk, D.A. 2018. “Drug Harmful Interaction between Economic Crisis, Violence, and Violence in Mexico. Data and Analysis through 2017.” Justice Ninis in Mexico.” Washington, DC: World Bank. Background in Mexico. Department of Political Science & International Re- paper for the “Out of School, Out of Work” lations. University of San Diego. San Diego. De la Fuente, A. and Olivera, M. 2013. Poverty Impact of Climate Campos-Vázquez, Raymundo M. and Melissa A. Knox. 2012. Social Change in Mexico. World Bank Policy Research Paper # 6461. Protection Programs and Employment: the Case of Mexico’s Deichmann, Uwe; Fay, Marianne; Koo, Jun; Lall, Somik V. 2002. Seguro Popular Program. Economía mexicana nueva época, Economic Structure, Productivity, and Infrastructure Quality SYSTEMATIC vol. XXII, núm. 2, p.p. 403-448. in Southern Mexico. Policy Research Working Paper; No. 2900. COUNTRY Campos-Vázquez, Raymundo, Luis López-Calva, and Nora Lustig. World Bank, Washington, DC. DIAGNOSTIC “Declining wages for college-educated workers in Mexico: do Diaz-Cayeros, A. 2016. Fiscal Federalism and Redistribution in MEXICO younger or older cohorts hurt the most?.” (2016). Mexico. Cardozo, D., Freije, S., and López-Acevedo, G. 2018. Determinants Diaz-Cayeros, A., Estévez, F., & Magaloni, B. 2016. The political 122 of Female Employment in Mexico during the Last Decade. Un- logic of poverty relief: Electoral strategies and social policy in published manuscript. Mexico. Cambridge University Press. DOF. 2013. Reglas de operación del Programa Nacional Forestal. Retrieved from: https://www.reuters.com/article/us-mexi- REFERENCES Diario Oficial de la Federación. Mexico City: City: Comisión co-fintech/mexico-financial-technology-law-passes-final-hur- Nacional Forestal. dle-in-congress-idUSKCN1GD6KX Dollar, D., Kleineberg, T. and Kraay, A., 2016. Growth still is good for Esquivel, G. and Hernández Trillo, F., 2009. How Can Reforms Helo the poor. European Economic Review, 81, pp.68-85. Deliver Growth.  Growing Pains in Latin America: An Economic Dornbusch, Rudiger, and Alejandro Werner. 1994. “Mexico: Stabili- Growth Framework as Applied to Brazil, Colombia, Costa Rica, zation, Reform, and No Growth.” Brookings Papers on Econom- Mexico, and Peru, p.192. ic Activity, 1: 253-97. Fajnzylber, P., Maloney, W.F. and Montes-Rojas, G.V., 2011. Does Dougherty, Sean M. 2013. Legal reform, contract enforcement formality improve micro-firm performance? Evidence from and firm size in Mexico. OECD Economics Department Work- the Brazilian SIMPLES program.  Journal of Development Eco- ing papers, No. 1042. nomics, 94(2), pp.262-276. Dougherty, S. and O. Escobar. 2013. “The Determinants of Infor- Ferreyra, M.M, and Roberts, M. (Eds.) 2018. Raising the Bar for Pro- mality in Mexico’s States”, OECD Economics Department Work- ductive Cities in Latin America and the Caribbean. Washing- ing Papers, No. 1043, OECD Publishing, Paris. ton, DC: World Bank. Duarte, M. and D. Restuccia. 2010. “The Role of the Structural Fareed, F., M. Gabriel, P. Lenain, J. Reynaud. 2017. “Financial Inclu- Transformation in Aggregate Productivity.” Quarterly Journal sion and Women Entrepreneurship: Evidence from Mexico”, of Economics 125 (1): 129–73. OECD Economics Department Working Papers, No. 1411, Duque, J. C., Lozano-Gracía, N., Patino, J., and Restrepo, P. 2017. OECD Publishing, Paris. “Institutional Fragmentation and Metropolitan Coordination Finnovista. 2017. “Fintech Radar Mexico 2017”. in Latin American Cities: What Consequences for Productivity Firpo, S. and Pieri, R. 2017. “Structural Change, Productivity and Growth?” Background paper, World Bank, Washington, Growth, and Trade Policy in Brazil”, in Mc Millan, M., D. Rodrik, DC. and C. Sepulveda (Eds.) 2017. Structural Change, Fundamen- Duval and Furceri, 2016. “The Effects of Product Market Reforms: tals, and Growth: A Framework and Case Studies. Washington, The Role of Macroeconomic Conditions”, paper prepared for DC: International Food Policy Research Institute. the Seventeenth Annual Jacques Polak Research Conference, Flores-Quiroga, A. 2018, February 8. Mexico in full swing. Petro- available under https://www.imf.org/external/np/res/semi- leum Economist. Outlook 2018. Retrieved March 29, 2018, nars/2016/arc/index.htm from: http://www.petroleum-economistF.com Dyer, D. 2018. What was missing in Mexico’s energy reform, and Foster, K. A. 1993. “Exploring the Links between Political Structure how does that affect the rule of law? in Rios, V. and D. Wood and Metropolitan Growth.” Political Geography 12 (6): 523–47. (Eds.). 2018. The Missing Reform: Strengthening the Rule Fox, J. and Haight, L. 2010. ‘Mexican agricultural policy: Multiple of Law in Mexico. Woodrow Wilson International Center for goals and conflicting interests’, in Fox, J. and Haight, L. (Eds.) Scholars. Washington DC. 2010. Subsidizing Inequality: Mexican Corn Policy since NAFTA. Dyson, P.A., Anthony, D., Fenton, B., Matthews, D.R. and Stevens, Woodrow Wilson International Center for Scholars. Mexico. D.E. 2013. High rates of child hypertension associated with Freyermuth-Enciso, Graciela, Mónica Carrasco-Gómez, Martín obesity: a community survey in China, India and Mexico, Pedi- Romero-Martínez. 2016. The Role of Health Inequality in the atrics and International Child Health, 34:1, 43-49. Maternal Health Services Provided by Public Institutions in Eakin,H., Amy M. Lerner, David Manuel-Navarrete, Bertha Mexico. Health, 08, 206-218. Hernández Aguilar, Alejandra Martínez-Canedo, Beth Tell- Fried, B. J. & Lagunes, P. & Venkataramani, A. 2010. “Corruption and man, Lakshmi Charli-Joseph, Rafael Fernández Álvarez, Luis Inequality at the Crossroad: A Multimethod Study of Bribery Bojórquez-Tapia., 2016. Adapting to risk and perpetuating and Discrimination in Latin America.” Latin American Research poverty: Household’s strategies for managing flood risk and Review, vol. 45 no. 1, pp. 76-97. water scarcity in Mexico City, Environmental Science & Policy, Fuentes Castro, Hugo Javier et al. 2013. “Estudio de Evasión Global Volume 66, Pages 324-333. de Impuestos”. Instituto Tecnológico y de Estudios Superiores Economist, The (2017, May 11). “Why murder in Mexico is rising de Monterrey, ITESM. Campus Ciudad de México. México D.F. again”. Retrieved from: https://www.economist.com/news/ Gobierno de la República. 2014. “Programa Institucional del Ban- americas/21721973-gangs-get-smaller-and-diversify-why- co del Ahorro Nacional y Servicios Financieros 2014-18”, Lo- murder-mexico-rising-again gros 2014. Economist Intelligence Unit (2017). The 2017 Infrascope - Evalu- Gollin, D., D. Lagakos, and M. Waugh. 2014. “The Agricultural Pro- ating the environment for public-private partnerships in Latin ductivity Gap.” Quarterly Journal of Economics 129 (2): 939-93. America and the Caribbean. González Block MA (Coord.) (2015), Diagnóstico de los retos al Enamorado, Ted, Luis F. López-Calva, and Carlos Rodríguez-Cas- Sistema Nacional de Salud Universal. México, Cofemedir-Fun- telán. 2014. “Crime and Growth Convergence: Evidence from salud. Mexico.” Economics Letters 125 (1): 9-13. González Block MA & Martínez González G., (Coords) (2015) Hacia Enamorado, Ted, Luis F. López-Calva, Carlos Rodríguez-Castelán, la cobertura universal de la protección financiera de la salud Hernán Winkler. 2016. Income inequality and violent crime: en Mexico. Ciudad de México, AMIS. Evidence from Mexico’s drug war, Journal of Development González-Block Miguel Ángel, Figueroa Alejandro, García-Téllez Economics, Volume 120, Pages 128-143. Ignacio, Alarcón José. 2016. Asignación financiera en el Siste- Encuesta de Cohesión Social para la Prevención de la Violencia ma de Protección Social en Salud de México: retos para la y la Delincuencia (ECOPRED), 2014. INEGI. Available from: compra estratégica. Salud Pública Mex., 58 (5): 522-532. http://www.beta. inegi.org.mx/proyectos/enchogares/espe- González Calvillo, 2018. “Shifting Into 2018: A Snapshot of Mexi- ciales/ecopred/2014/ co’s Energy Sector.” Energy, Oil, Gas & Power Generation Prac- Enríquez, S., B. Larsen, and E. Sanchez-Triana. 2018. “Guidance tice, Client Alert, January. Retrieved March 31 from: http:// Note 8. Local Environmental Externalities due to Energy Price gcsc.com.mx Subsidies: A Focus on Air Pollution and Health.” In Flochel. T. Greve, Hannes & Ana K. Negrete-García. 2017. Impact of energy and S. Gooptu. Energy Subsidy Reform Assessment Frame- price changes on microenterprise. work (ESRAF). Guidance Notes toward Evidence-Based Energy González-Pier, E., Barraza-Lloréns, M., Beyeler, N., Jamison, D., SYSTEMATIC Subsidy Reforms. The World Bank. Washington D.C. Knaul, F., Lozano, R., Yamey, G. and Sepúlveda, J., 2016. Mex- COUNTRY DIAGNOSTIC Entrepreneur Magazine. 2017. México se convertirá en el hub de ico’s path towards the Sustainable Development Goal for MEXICO fintech en América Latina. https://www.entrepreneur.com/ health: an assessment of the feasibility of reducing premature article/306044 mortality by 40% by 2030. The Lancet Global Health, 4(10), pp.e 123 Espejo, S. 2018, March 1. Mexico financial technology law pass- 714-e725.performance: Evidence from Mexico. Unpublished es final hurdle in Congress. Technology News. Reuters Staff. manuscript. REFERENCES Government of Mexico (GoM). 2012. Ley General de Cambio IEP (Institute for Economics and Peace). 2015. Peace and Corrup- Climático. Amended version published in the Official Gazette tion Report. on January 19, 2018. http://www.diputados.gob.mx/Leyes- IEP. 2017. Índice de Paz México 2017: Mapa de evolución y facto- Biblio/pdf/LGCC_190118.pdf res que impulsan la paz. IEP Informe 48. ____, 2013. Estrategia Nacional de Cambio Climático. Visión 10 IFC.2018. Policy Note on PPPs in México. Unpublished ppt slides. - 20 -40. http://www.semarnat.gob.mx/archivosanteriores/ IMCO. 2015. “La corrupción en México: transamos y no avanza- informacionambiental/Documents/06_otras/ENCC.pdf mos”. IMCO. Ciudad de México. ____, 2014. Programa Especial de Cambio Climático 2014-2018 IMF. 2016. Corruption: Costs and Mitigating Strategies, Staff Dis- (PECC). Published on April 28, 2014. http://www.semarnat. cussion Note. gob.mx/sites/default/files/documentos/transparencia/pro- ___, 2016, April World Economic Outlook, Chapter 3. grama_especial_de_cambio_climatico_2014-2018.pdf ___, 2017. Mexico Selected Issues and Analytical Notes. IMF Coun- ____, 2015. Mexico Intended Nationally Determined Contribu- try Report No. 17/347. International Monetary Fund: Washing- tion. Submitted to the United Nations Framework Convention ton, D.C. on Climate Change on March 30, 2015. http://www4.unfccc. ___, 2017a. “IMF Fiscal Monitor: Achieving More with Less,” Inter- int/submissions/INDC/Published%20Documents/Mexico/1/ national Monetary Fund, Washington, D.C. MEXICO%20INDC%2003.30.2015.pdf ___, 2017, Selected Issues. ____, 2016. Mexico’s Mid-Century Climate Change Strategy. INCDMN. 2010. Instituto Nacional de Ciencias Médicas y Nutrición https://unfccc.int/files/focus/long-term_strategies/applica- Salvador Zubirán. Evaluación de efectos del Sistema de Pro- tion/pdf/mexico_mcs_final_cop22nov16_red.pdf tección Social en Salud (SPSS), 2008. México: Secretaría de Grogger, Jeffrey, Tamara Arnold, Ana Sofía León, Alejandro Ome. Salud, 2010. 2015. “Heterogeneity in the effect of public health insurance INECC. 2017. Estudio de la Calidad de los Combustibles Fósiles on catastrophic out-of-pocket health expenditures: the case y su Impacto en el Medio Ambiente. Instituto Nacional de of Mexico”, Health Policy and Planning, Volume 30, Issue 5, Pages Ecología y Cambio Climático. Available at: https://www.gob. 593–599. mx/cms/uploads/attachment/file/312845/Informe_Final_Cal- Guerrero, Isabel, Luis Felipe López-Calva, and Michael Walton. idad_Combustibles_2018.04.01.pdf 2009. “The Inequality Trap and Its Links to Low Growth in Mex- INEGI, 2010. Censo de Población y Vivienda 2010. Aguascalientes: ico.” In No Growth with? out Equity? Inequality, Interests, and INEGI. Competition in Mexico, ed. Santiago Levy and Michael Walton, _____, 2016. National Survey on the Quality and Impact of Gov- 111-56. Houndmills, U.K. and New York: Palgrave Macmillan; ernment. Washington, D.C.: World Bank. _____, 2016. Actualización del Marco Censal Agropecuario 2016. Haber, Stephen H. 2005. “Why Institutions Matter: Banking and _____, 2017. Censos económicos. Technical report, Instituto Na- Economic Growth in Mexico.” Unpublished. cional de Geografía y Estadística. Haight, L. 2010. Large payments of grain marketing subsidies _____, 2017b. Encuesta Nacional sobre Disponibilidad y Uso de go to transnational corporations. In Fox, J. and Haight, L. Tecnologías de la Información en los Hogares (ENDUTIH). (Eds.) 2010. Subsidizing Inequality: Mexican Corn Policy since [National Survey on Availability and Use of Information and NAFTA. Woodrow Wilson International Center for Scholars. Communication Technologies in Households] Mexico. Ingram, M. 2018. On the Justice System in Rios, V. and D. Wood Hanson, Gordon H. “Why isn’t Mexico rich?.”  Journal of Economic (Eds.). 2018. The Missing Reform: Strengthening the Rule Literature 48.4 (2010): 987-1004. of Law in Mexico. Woodrow Wilson International Center for Heinle, K., Molzahn, C., and Shirk David, A. 2015. “Drug Violence Scholars. Washington DC. in Mexico. Data and Analysis through 2014.” Justice in Mexico. IPD, 2014. “Las mermas y pérdidas de alimentos a lo largo de las Department of Political Science & International Relations. Uni- cadenas de valor agroalimentarias de México”. Investigaciones versity of San Diego. San Diego. del IPD. Instituto para la Planeación del Desarrollo. Hernández-Trillo, Fausto & Brenda Jarillo Rabling, 2006. “Is Lo- ITA, 2017. Mexico – Agribusiness. Retrieved from: https://www. cal Beautiful? Decentralization in Mexico in the Presence export.gov/article?id=Mexico-Agribusiness of Elite Capture,” Working papers DTE 360, CIDE, División de Jaitman, L., Caprirolo, D., Granguillhome Ochoa, R., Keefer, P., Economía. Leggett, T., Lewis, J.A., Mejía-Guerra, J.A., Mello, M., Sutton, H., Hernández-Trillo, Fausto. 2007. “Problemas de financiamiento en Torre, I. 2017. “The costs of crime and violence: new evidence gobiernos locales: ¿cuál es el papel de la Banca de Desarrollo?,” and insights in Latin America and the Caribbean”, Inter-Ameri- Working papers DTE 398, CIDE, División de Economía. can Development Bank. Washington, DC. Hernández Trillo, Fausto. 2016. “El objetivo versus los instrumen- Juan López, Mercedes, Adolfo Martínez Valle & Nelly Aguilera. tos de la política pública: El caso de una reforma verde para la 2015. Reforming the Mexican Health System to Achieve Ef- igualdad en México”. Estudios del cambio climático en Améri- fective Health Care Coverage. Health Systems & Reform Vol. ca Latina. Comisión Económica para América Latina y el Caribe 1, Iss. 3. (CEPAL). Kahn, Theodore A. 2017. Elusive Alliances: Local Governments, Herrendorf, B., R. Rogerson, and A. Valentinyi. 2013. “Growth and Transnational Capital, and Regional Development in Post-re- Structural Transformation.” Working Paper No. 18996, National form Mexico. Doctoral Dissertation, SAIS, Johns Hopkins Uni- Bureau of Economic Research (NBER), Cambridge, MA. versity. Hevia de la Jara, F. 2010. Stakeholder oversight of ASERCA’s farm Kaplan, David S. Eduardo Piedra, Enrique Seira. 2011. “Entry regu- subsidy programs”, in Fox, J. and Haight, L. (Eds.) 2010. Subsi- lation and business start-ups: Evidence from Mexico”, Journal dizing Inequality: Mexican Corn Policy since NAFTA. Woodrow of Public Economics, Volume 95, Issues 11–12, Pages 1501- Wilson International Center for Scholars. Mexico. 1515. IACHR, 2015. Situation of Human Rights in Mexico, Organization Kaplan, David and Sadka, Joyce. 2010. “The Plaintiff’s Role in En- of American States. forcing a Court Ruling: Evidence from a Labor Court in Mexico,” Iacovone, L., Jiménez, F., Sánchez-Bella, P., and Yuki, K. 2018. Build- Inter-American Development Bank Paper Series. SYSTEMATIC ing Shock Absorbers: The role of local business regulations in Kehoe, T.J. and MEZA, F., 2011. Catch-up growth followed by stag- COUNTRY response to a large external shock. Unpublished manuscript. nation: Mexico, 1950-2010.  Latin american journal of econom- DIAGNOSTIC IATP, 2010. NAFTA: Fueling Market Concentration in Agriculture. ics, 48(2), pp.227-268. MEXICO Institute for Agriculture and Trade Policy. Retrieved from: Kim, Y., and Zangerling, B., eds. 2016. Mexico Urbanization Re- https://www.iatp.org/files/451_2_107275.pdf view: Managing Spatial Growth for Productive and Livable 124 IEP. 2018. Mexico Peace Index 2018: Mapping the Evolution of Cities in Mexico. Directions in Development. Washington, DC: Peace and its Drivers. World Bank. King G, Gakidou E, Imai K, et al. 2009. Public policy for the poor? A Mc Millan, M., D. Rodrik, and C. Sepulveda (Eds.) 2017. Structural REFERENCES randomized assessment of the Mexican universal health insur- Change, Fundamentals, and Growth: A Framework and Case Stud- ance programme. The Lancet; 373: 1447-145. ies. Washington, DC: International Food Policy Research Institute. Knaul, F., E. González-Pier, O. Gómez-Dantes. 2012. The Quest for McKenzie, D. and Woodruff, C. 2008. Experimental evidence on Universal Health Coverage: Achieving Social Protection for All returns to capital and access to finance in Mexico. The World in Mexico. The Lancet Vol. 380, pp.1259-79. Bank Economic Review, 22(3):457-482. Kumler, Todd, Eric Verhoogen, Judith A. Frías. 2013. Enlisting Em- Medina C, and Barquera S. 2016. Actividad física. Encuesta Nacio- ployees in Improving Payroll-Tax Compliance: Evidence from nal de Salud y Nutrición 2016. Mexico, NBER Working Paper No. 19385 Meléndez Aguilar, Kristobal. 2016. Distribución del Gasto Educa- Kurowski, C., and Luis Ortiz. 2012. “Mexico’s Social Protection Sys- tivo Federalizado. Centro de Investigación Económica y Pre- tem in Health and the supuestaria, A.C. Financial Protection of Citizens without Social Security.” World Meza Gonzáles and Ramírez García. 2012. “Antecedentes sobre in- Bank, Washington, DC. seguridad pública y migración internacional”. In “México ante Kurowski, C., and Manuela Villar-Uribe. 2012. “Mexico’s Social Pro- los recientes desafíos de la migración internacional “, coord. tection in Health and the Telesforo Ramírez García and Manuel Ángel Castillo, Consejo Transformation of State Health Systems.” World Bank, Washington, Nacional de Población (CONAPO), Ciudad de México, 2012. DC. Montes de Oca, M. and C. Muñoz-Piña. 2016. “Fuel Subsidy Reform Laeven, Luc; Woodruff, Christopher. 2004. The Quality of the Legal in Mexico: Environmental and Social Benefits of the Gasoline System, Firm Ownership, and Firm Size. Policy Research Work- Subsidy Removal and the Way Forward for Optimal Pricing.” ing Paper; No.3246. World Bank, Washington, D.C. Report commissioned by the World Bank. Washington. D.C. Lakin, Jason M. 2010. “The End of Insurance? Mexico’s Seguro Pop- Montes Mendoza, Pablo, 2015. “¿Castigar o proteger?: las reac- ular, 2001–2007”. Journal of Health Politics, Policy and Law, Vol. ciones de los partidos políticos ante un escándalo.” Centro de 35, No. 3, June 2010. Investigación y Docencia Económicas. Lambsdorff, Johann G. 2003. How corruption affects productivity, Moreno-Brid, J.C., N. Pérez Benítez & H. Juan Villarreal Páez. 2017. Kyklos 56, pp. 457-474. Austerity in Mexico: Economic Impacts and Unpleasant Choic- Porta, R.L. and Shleifer, A., 2008.  The unofficial economy and eco- es Ahead, World Economic Review 7: 56-68, 201. nomic development (No. w14520). National Bureau of Econom- Mukherjee, Anit. 2016. “Fiscal Devolution and Health Financing ic Research. Reform: Lessons for India from Brazil, China and Mexico.” CGD La Porta, R. and Shleifer, A., 2014. Informality and develop- Policy Paper 97. Washington, DC: Center for Global Develop- ment. Journal of Economic Perspectives, 28(3), pp.109-26 ment. Levy, Dan, Ricardo Hausmann, Miguel Ángel Santos, Luis Espino- Muñoz-Piña, C. n.d. “El Impuesto a los Combustibles Fósiles por za, and Miguel Flores. 2015. “Why is Chiapas Poor?” HKS Facul- Contenido de Carbono en México.” Presentation made to the ty Research Working Paper Series RWP16-049. Mexican Senate. http://www.senado.gob.mx/comisiones/ Levy, Santiago.2008., Good Intentions, Bad Outcomes: Social Pol- cambio_climatico/reu/docs/SHCP.pdf icy, Informality and Economic Growth in Mexico, Brookings Myant, M. and L. Brandhuber. 2016. “Uses and abuses of the Institution Press. OECD’s Employment Protection Legislation index in research Levy, Santiago. 2009. “Social Security Reform in Mexico: For and EU policy making”. European Trade Union Institute, Brus- Whom?” in S. Levy & M. Walton (Eds.) No growth without eq- sels. uity?: Inequality, interests and competition in Mexico. World Negrete García, Ana Karen. 2017. Microenterprise Performance Bank. and Economic Development: Evidence from Mexico. Doctoral Levy, Santiago & Michael Walton (Eds.) 2009. No growth with- Dissertation, Faculty of Economic Sciences, University of Goet- out equity?: Inequality, interests and competition in Mexico. tingen. World Bank, Washington DC. Nelson, A. C., and Foster, K. A. 1999. “Metropolitan Governance Licetti, et al. 2016a. Structure and Income Growth.” Journal of Urban Affairs 21: López, José Joaquín. 2017. Financial frictions and productivity: 309–24. Evidence from Mexico. The Quarterly Review of Economics and Ochoa Sosa, R., Hernández Espinoza, A., Garfias Royo, M. and Mo- Finance, 66 (2017) 294–301. rillón Gálvez, D. Life cycle energy and costs of sprawling and López-Martín, Bernabe. 2016. Informal Sector Misallocation, No compact neighborhoods. 2016-09, Working Papers, Banco de México. OECD., 2008. Open Innovation in Global Networks, OECD Publish- Luna, B. et al., 2012. Perspectivas de desarrollo de la industria se- ing, Paris. millera de maíz en México. Revista Fitotec, 35(1), pp.1-7. _____., 2010. Tax Expenditures in OECD countries. Lustig, Nora. 1998. Mexico: The Remaking of an Economy, Second _____.,2015. OECD Economic Surveys: Mexico 2015, OECD Pub- edition. Washington, D.C.: Brookings Institution. lishing, Paris. Lustig, Nora, Carola Pessino and John Scott. 2013. “The Impact _____.,2015. OECD Employment Outlook 2015, OECD, Paris. of Taxes and Social Spending on Inequality and Poverty in _____., 2015. OECD Indicators of Employment Protection Legisla- Argentina, Bolivia, Brazil, Mexico, Peru and Uruguay: An Over- tion. Available at: http://www.oecd.org/els/emp/oecdindica- view,” CEQ Working Paper No. 13, April. torsofemploymentprotection.htm Macías Sánchez, Alejandra. 2017. El gasto en pensiones refuerza _____.,2015. The Innovation Imperative: Contributing to Produc- la desigualdad Centro de Investigación Económica y Presu- tivity, Growth and Well-Being, OECD Publishing, Paris. puestaria, A.C. _____.,2016. OECD Reviews of Health Systems: Mexico 2016, Mancera Corcuera, Carlos. 2008. Gasto Público y Educación. Doc- OECD Publishing, Paris. umento de apoyo del Informe sobre Desarrollo Humano Méx- _____.,2016. Better Life Index, Country Reports. OECD, Paris. ico 2011. Programa de las Naciones Unidas para el Desarrollo, _____.,2017. Agricultural Policy Monitoring and Evaluation 2017 México. Part Ii: Developments in Agricultural Policy and Support by Mardero, S., B. Schmook, J. O. López-Martínez, L. Cicero, C. Radel Country. OECD, Paris. and Z. Christman. 2018. “The Uneven Influence of Climate _____.,2017b. Economic Surveys Mexico January 2017 Overview. SYSTEMATIC Trends and Agricultural Policies on Maize Production in the OECD Publishing, Paris. COUNTRY DIAGNOSTIC Yucatan Peninsula, Mexico”, Land, 7, 80. _____.,2017c. OECD Telecommunication and Broadcasting Re- MEXICO Martínez-Cruz, Adán L.; Rodríguez Castelán, Carlos. 2016. Crime and view of Mexico 2017, OEDC Publishing, Paris. Persistent Punishment: A Long-Run Perspective on the Links be- _____.,2018. Getting it Right: Prioridades estratégi- 125 tween Violence and Chronic Poverty in Mexico. Policy Research cas para México, Éditions OCDE, Paris. http://dx.doi. Working Paper; No. 7706. World Bank, Washington, DC. org/10.1787/9789264292871-es REFERENCES _____., 2018a. Air and GHG emissions (indicator). doi: Ruiz, Claudia. 2010. From pawn shops to banks: the impact of for- 10.1787/93d10cf7-en (Accessed on 07 February 2018). mal credit on informal households. Policy Research working Pagés, C., 2010. The age of productivity. In  The Age of Productivi- paper; WPS 6634. Washington, DC: World Bank. ty (pp. 1-21). Palgrave Macmillan, New York. Sabet, Daniel. M. 2013. Corruption or insecurity? Understanding Pagés, C. (2017) “Do payroll tax cuts boost formal jobs in devel- dissatisfaction with Mexico’s police. Latin American Politics oping countries?” IZA World of Labor 2017: 345 doi: 10.15185/ and Society 55 (1): 22-45. izawol.345 San Martín Reyna, Juan Manuel, Héctor Enrique Ángeles Sánchez, Pan American Health Organization. 2012. Mexico. Health in the Carlos Alberto Juárez Alonso, and Jaime Diaz Martín del Cam- Americas, 2012 Edition: Country Volume. Retrieved from: po. 2017, Evasión Global 2017 Cholula, Puebla. Available at: https://www.paho.org/salud-en-las-americas-2012/index. http://www.sat.gob.mx/administracion_sat/estudios_eva- php?option=com_docman&view=download&catego- sion_fiscal/Documents/Evasion_global2017.pdf ry_slug=hia-2012-country-chapters-22&alias=137-mexico- Sánchez Bayardo, Luis Fernando; Iacovone, Leonardo; González, 137&Itemid=231&lang=en Álvaro S. 2018. Micro-level analysis of Mexican retail markets Persson, Torsten, and Guido Tabellini. 1994. ‘‘Is Inequality Harmful and their response to changes in market structure and com- for Growth?’’ American Economic Review 84 (1994), 600–621. petition policies. Policy Research working paper; WPS 8294. PGR, 2017. Procuraduría General de la República. Oficio. Núm. Washington, D.C.: World Bank Group. PGR/ 018 /2017. Ciudad de México. Sánchez Triana, E., J. Ruitenbeek, S. Enriquez, and K. Siegmann Prieto, A. 2017. “Follow-up: Political Backing for the Emissions (eds.). Forthcoming. Green and Inclusive Growth in the Yucat- Trading Scheme Simulation in Mexico.” JiQ Magazine 23, No. an Peninsula. Washington, D.C.: The World Bank. 1. http://cdn.jin.ngo/images/jin/publications/JIQ_Magazine_ Scott, J., 2010. Agricultural subsidies in Mexico: Who gets Apr2017.pdf what.  Subsidizing Inequality: Mexican Corn Policy since NAFTA. ProMexico. 2016. The Mexican Automotive Industry - Current Sit- Woodrow Wilson International Center for Scholars, Washington, uation, Challenges and Opportunities. DC, pp.67-118. ProMexico. 2017. The Aerospace Industry in Mexico – Current Sit- Scott, J., 2014. Redistributive impact and efficiency of Mexico’s uation and Growth Potential. In: Negocios, March/April 2017. fiscal system. Public Finance Review, 42(3), pp.368-390. Ramones, F. and Prudencio, D. 2014. Los efectos del Fondo de Secretaría de Comunicaciones y Transporte, 2016, Estadística Aportaciones para la Infraestructura Social sobre la pobreza Básica del Autotransporte Federal 2016. http://www.sct.gob. en México. Región y sociedad, 26(60), pp.63-88. mx/fileadmin/DireccionesGrales/DGAF/EST_BASICA/EST_BA- Ramos-Francia, M. and García-Verdú, S. 2017. Bank Credit Alloca- SICA_2016/Estad%C3%ADstica_B%C3%A1sica_del_Auto- tion and Sectorial Concentration in Mexico: Some Empirical transporte_Federal_2016.pdf Evidence. Working Papers 2017-14, Banco de México. Secretaría de Hacienda y Crédito Público, 2017, Presupuesto de Rico García-Amado, L., Ruiz Pérez, M., Reyes Escutia, F., Barrasa gastos Fiscales 2017, México D.F. http://www.sat.gob.mx/ García, S., & Contreras Mejía, E. 2011. Efficiency of Payments administracion_sat/estudios_evasion_fiscal/Documents/Eva- for Environmental Services: Equity and additionality in a case sion_global2017.pdf study from a Biosphere Reserve in Chiapas, Mexico. Ecological Secretaría de Medio Ambiente y Recursos Naturales, 2017, Trans- Economics, 70, 2361-2368. porte limpio ha logrado mitigar 7 millones de toneladas CO2, Rigaud, K.K., de Sherbinin, A, Jones, B., Bergmann, J., Clement, V., Comunicado de Prensa 122/17. Available at: https://www. Ober, K., Schewe, J., Adamo, S., McCusker, B., Heuser, S., Midg- gob.mx/semarnat/prensa/transporte-limpio-ha-logrado-miti- ley, A. 2018. Groundswell: Preparing for Internal Climate Migra- gar-7-millones-de-toneladas-co2?idiom=es tion. World Bank, Washington, DC. Secretaría de Medio Ambiente y Recursos Naturales (SEMAR- Ríos, Viridiana, and Duncan Wood (Eds.). 2018. The Missing Re- NAT), 2017b. Acciones de México para establecer un merca- form: Strengthening the Rule of Law in Mexico. Woodrow Wil- do de carbono. https://www.gob.mx/semarnat/articulos/ son International Center for Scholars. Washington DC. acciones-de-mexico-para-establecer-un-mercado-de-car- Rimisp-Centro Latinoamericano para el Desarrollo Rural. (2018) bono?idiom=es Informe Latinoamericano de Pobreza y Desigualdad 2017. Secretaría de Medio Ambiente y Recursos Naturales (SEMARNAT), Santiago, Chile: IDRC - FIDA. Partnership for Market Readiness (PMR), MexiCO2, and Envi- Rodríguez Castelan, Carlos. 2017. Conditionality as targeting? ronmental Defense Fund (EDF), 2018. Ejercicio de Simulación Participation and distributional effects of conditional cash del Mercado de Carbono en México. Reporte de resultados de la transfers. Policy Research working paper; no. WPS 7940. Wash- primera fase. http://www.mexico2.com.mx/uploadsmexico/ ington, D.C.: World Bank Group. file/Reporte%20General_Final_SIM_1.pdf Rodríguez-Castelan, Carlos and Lourdes Rodríguez-Chamussy. Secretaría de Salud, 2010, Rendición de cuentas en salud 2009. 2015. More Market Concentration, More Poverty? Evidence Available at: http://www.dged.salud.gob.mx/ contenidos/ from the Retail Sector in Mexico. Working Paper. dedss/descargas/rcs/RCS_09_6_Dic.pdf Rodríguez-Castelan, Carlos, Kiyomi Cadena, Laura Moreno Herre- SEGOB, INMUJERES, ONU MUJERES (2016). La Violencia Feminici- ra. 2017b. The impact of federal transfers for social infrastruc- da En México, Aproximaciones Y Tendencias 1985-2014”. ture on regional development in Mexico, 2005-2014. Working Seuring, T., Serneels, P.M. and Suhrcke, M., 2016. The impact of di- paper. abetes on labor market outcomes in Mexico: a panel data and Rodríguez-Castelan, Carlos, Luis-Felipe López-Calva, Oscar Barri- biomarker analysis. ga-Cabanillas. 2017a. The effect of local industry competition Shapiro-Garza, E. (2013). Contesting the market-based nature of on economic productivity. Evidence from the Mexican manu- Mexico’s national payments for ecosystem services program: facturing industry. Four sites of articulation and hybridization. Geoforum, 46, Rodríguez-Oreggia, E., De La Fuente, A., De La Torre, R. and More- 5-15. no, H.A. 2013. “Natural Disasters, Human Development and Silva Forné, Carlos. 2011. Policía y encuentros con la ciudadanía Poverty at the Municipal Level in Mexico”. The Journal of Devel- en Ciudad Nezahualcóyotl. Ciudad de México: Universidad opment Studies, 49:3, 442-455. Nacional Autónoma de México. SYSTEMATIC Rodríguez-Oreggia, Eduardo, Moreno Moreno, Héctor, & Mulato Silva Méndez, Jorge Luis. 2016, Reimbursable Advisory Services COUNTRY Martínez, David. 2008. Has product market competition re- for Designing Technical Standards for the Implementation DIAGNOSTIC shaped returns to schooling in Mexico?. EconoQuantum, 5(1), of Oral Commercial Lawsuits (P155528), Executive Summary, MEXICO 61-77. Governance Global Practice, The World Bank, 19 pp. Rull J.A., Aguilar-Salinas C.A., Rojas R., Ríos-Torres J.M., Gó- Silva Méndez, Jorge Luis and Caballero, José Antonio. 2012. Evalu- 126 mez-Pérez F.J., Olaiz G. 2005. Epidemiology of type 2 diabetes ación Rápida y Plan de Acción para el Tribunal Superior de in Mexico. Arch Med Res, 36(3):188-96. Justicia del Distrito Federal, The World Bank, 124 pp. Sims, Katharine and Jennifer Alix-Garcia (2017). Parks versus PES: USDA. 2011. Market Concentration in Selected Agricultural and REFERENCES Evaluating direct and incentive-based land conservation in Food Subsectors. GAIN Report Number: MX1042. US Foreign Mexico. Journal of Environmental Economics and Management, Agricultural Service. Retrieved from: https://gain.fas.usda. Volume 86, November 2017, Pages 8-28. gov/Recent%20GAIN%20Publications/Market%20Con- Skelton, Edward C., 2008. Reaching Mexico’s Unbanked, Economic centration%20in%20Selected%20Agricultural%20and%20 Letter, Federal Reserve Bank of Dallas. 3(7). Food%20Subsectors_Mexico_Mexico_5-25-2011.pdf Smith, Heidi Jane M. 2016. “El papel de las reglas fiscales en el Valentín Garrido, J.M. et al. 2016. Evaluación del programa PRO- control del endeudamiento: la debilidad institucional de los AGRO productivo en comunidades rurales de la sierra norte municipios mexicanos”, Revista Iberoamericana de Estudios de Puebla, Revista Mexicana de Ciencias Agrícolas. Vol.7 no.2. Municipales, Núm. 14, p.p. 115-149. Velázquez Guadarrama, César. 2008. Gasto federal descentral- Smith, Heidi Jane M. & Keith D. Revell. 2016. Micro-Incentives and izado y de estados y municipios. Documento de apoyo del Municipal Behavior: Political Informe sobre Desarrollo Humano México 2011. Programa de Decentralization and Fiscal Federalism in Argentina and Mexico. las Naciones Unidas para el Desarrollo, México. World Development Vol. 77, pp. 231–248. Viesti, Gianfranco. 2015. “Regional Development Scan: Mexico”, Solís, P. 2010. La desigualdad de oportunidades y las brechas de Collection Studies No.13, Eurosocial Programme for Social escolaridad. Arnaut, Alberto y Silvia Giorguli (coords.), Edu- Cohesion in Latin America. cación. Los grandes problemas de México. Weiss, Eli; and Rosenblatt, David. 2010. Regional Economic Growth Solís Ávila, J.C. and Sheinbaum Pardo, C. 2016. Consumo de en- in Mexico: Recent Evolution and the Role of Governance. Poli- ergía y emisiones de Co2 del autransporte en México y esce- cy Research working paper; WPS 5369. World Bank narios de mitigación. Rev. Int. Contam. Ambie. 32 (1) 7-23. Winkler, Hernan. 2015. The Effect of Income Inequality on Taxa- Soto Zazueta, Irvin Mikhail, & Cortez, Willy W. 2015. The Impact of tion and Public Expenditures: Evidence from Mexican Munici- Political Alternation on Corruption in Mexico. Revista de cien- palities 1990-2010. Unpublished manuscript. cia política (Santiago), 35(2), 371-392. WJP, 2018. The Rule of Law Index 2017-2018. The World Justice Statista. 2017. FinTech Market Report. https://www.statista.com/ Project. Washington, D.C. outlook/295/116/fintech/mexico#. World Bank, 2009. Mexico: Agriculture and Rural Development Sumner, Daniel A. and Joseph V. Balagtas (2007). “Economic Anal- Public Expenditure Review, Washington, DC: World Bank, ysis of the Ingreso Objetivo Program in Mexico,” Inter-Ameri- Sustainable Development Department, Latin American and can Development Bank. August 13, unpublished document. Caribbean Region. Retrieved from: http://documents.world- Transparencia Mexicana. 2010. Índice Nacional de Corrupción y bank.org/curated/en/500971468299092123/Mexico-Agricul- Buen Gobierno (INCBG). ture-and-rural-development-public-expenditure-review Transparency International. 2017. “People and Corruption: Latin _, 2010. Enterprise Surveys, The World Bank Group. America and the Caribbean”, Global Corruption Barometer. _, 2013. Financial Sector Assessment Program: Mexico. Capital Trejo, G. and Ley, S. (2015, February 1). Municipios bajo fuego Market Development. Technical Note 83138. (1995-2014). Nexos. Retrieved from: https://www.nexos.com. _, 2015. Environmental health in Mexico: An economic assess- mx/?p=24024. ment of health effects and their costs. Unpublished manuscript. Tuirán Gutiérrez, A. 2018. ¿Cuál es el nivel de corrupción de Méxi- _, 2016. Agricultural Services for Food Security and Compet- co respecto a otras naciones del mundo? Una medición de los itiveness, Project Information Document/ Integrated Safe- niveles de corrupción en México, de acuerdo a los resultados guards Data Sheet, Project ID: P160570. Available at: http://doc- de índices internacionales. Unpublished manuscript. uments.worldbank.org/curated/en/691511479941010871/ U.S. Department of Agriculture. 2015. Foreign Agricultural Ser- pdf/ITM00184-P160570-11-23-2016-1479941008301.pdf vice. Global Agricultural Trade System (GATS). _, 2016b. Mexico: Public Expenditure Review. Report No. UNCTAD, 2014. Mexico’s Agricultural Development: Perspectives AUS10694 Washington DC: The World Bank. and Outlook. Geneva, Swizterland _,2017.  World Development Report 2017: Governance and UNCTAD, 2017. World Investment Report 2017. the Law. Washington, DC: World Bank. © World Bank. https:// UNDP. 2011. Informe sobre Desarrollo Humano México 2011, openknowledge.worldbank.org/handle/10986/25880 Equidad del gasto público: derechos sociales universales con , 2017. Mexico Manual Subnational Pro-Comp Regulation. subsidios focalizados, México. , 2018. Infrastructure Public-Private Partnerships. Ungor, M. 2011. “Productivity Growth and Labor Reallocation: , 2018. Mexico Policy Notes (forthcoming). Washington, DC: Latin America versus East Asia” Manuscript, Central Bank of World Bank. Turkey, Istanbul. , Forthcoming. Ensuring Gender Mainstreaming in the Design United Nations, Department of Economic and Social Affairs, Popu- and Implementation of REDD Related Activities. A behavioral lation Division (2018). World Urbanization Prospects: The 2018 approach to natural resource management. Washington, DC. Revision , Forthcoming. Background Paper for the Flagship on “High Universidad de las Américas Puebla. 2015. Índice Global de Im- Growth Enterprises in Developing Countries”. punidad. World Economic Forum. 2017. Global Competitiveness Report Urzúa, Carlos M. 2008. “Evaluación de los efectos distributivos y 2017-18. Available at: https://widgets.weforum.org/glob- espaciales de las empresas con poder de mercado en México”. al-competitiveness-report-2017/ Report for the Federal Competency Commission (Comisión Zepeda Lecuona, G. 2005. Myths of Pretrial Detention in Mexico. Federal de Competencia) and OECD. México. Open Society Justice Initiative. Monterrey, Mexico. SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 127 Annex 1. Identification of comparators/peer countries This SCD benchmarks Mexico vis-à-vis i) OECD countries; ii) regional peers; iii) a set of countries which started in the 1990s at a similar position than Mexico, but have had a successful path over the past decades and reached higher levels than Mexico (“aspirational peers”); and, iv) a set of structurally similar countries (“structural peers”). •• Regional peers: Chile, Peru, Colombia, Brazil, Argentina, Uruguay. •• Aspirational peers: Chile, Poland, South Korea, Malaysia. •• Structural peers: Turkey, Thailand, Romania, Argentina. Aspirational countries are countries anywhere in the world that starting from similar GDP per capita levels to Mexico in the 1990s, had a successful path in economic growth (i.e. increasing GDP per capita) and productive inclusion (i.e. expand- ing the rate of female labor force participation). To increase comparability with Mexico’s case, we have limited the set of countries to those having a total population that is within the same of tier in the world ranking. Detailed definitions of data-driven ‘aspirational peers’ Description Countries Intersection Korea, Rep. Countries with Malaysia Aspirational peers •• Successful path in terms of growth in GDP per capita, and Poland Definition: countries with Korea, Rep. •• Total population in the comparable range. successful path BOTH in GDP Chile Malaysia per capita and female labor force participation over the past decade Countries with Poland SYSTEMATIC and starting from similar initial COUNTRY conditions to those in Mexico (in •• Successful path in expanding female labor Chile Chile DIAGNOSTIC the 1990s) force participation rate, and MEXICO Spain •• Comparable population size, and •• Comparable GDP per capita. 129 ANNEXES Definition of the successful path: •• GDP per capita (aspirational): countries that had similar levels of GDP per capita to Mexico in 1990 (average of 1990, 1991 and 1992) and increased reaching levels 20% or higher than Mexico’s in 2015. •• Female labor force participation rate (aspirational): countries that had similar levels to Mexico -- in 1990 (average of 1990,1991 and 1992) and increased FLFP rate to reach levels at least 20% higher than Mexico’s in 2015. Structural peers are countries anywhere in the world that meet the following criteria in terms of characteristics of its econ- omy: Detailed definitions of data-driven ‘structural peers’ Description Countries Countries with: •• Classified as upper-middle income. Turkey •• Comparable level of GDP per capita. Thailand Structural peers •• Near at least to one big market •• Comparable score in the logistics performance index. Romania •• Comparable percentage of manufactures exports (% of Argentina merchandise exports). •• Similar market size. Definition of countries with similar levels to Mexico’s in each indicator: •• Comparable countries in the GDP per capita (Constant US 2010), logistic performance index, manufactures exports and market size, are defined as the group of countries for which the simple average between 2010 and the last available year (circa 2016) is within the same percentile as that of Mexico or ten percentiles above/below. We excluded countries in this range, for which the average was higher or lower than 50% of Mexico. •• Distance to a big market is defined as the linear distance between the frontier of each country and the frontier of any of top ten countries with highest GDP per capita between 2010 and 2016 (big market). Countries with a distance within 650 km in this indicator are categorized in the set of ‘near to a one big market’. Data Sources: Indicator Data source GDP per capita World Development Indicators 2017 Female labor force participation rate World Development Indicators 2017 Manufactures exports (% of merchandise exports) World Development Indicators 2017 WBG - Logistics Performance Index TCdata 360 at World Bank Market size index World Economic Forum Global Competitiveness Index SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 130 Annex 2. Official ANNEXES poverty methodology in Mexico The official multidimensional poverty measurement combines income poverty with six indicators of social deprivation. In 2008, Mexico adopted an official multidimensional poverty measurement that combines income-based monetary poverty with non-monetary dimensions of wellbeing, called social deprivations. Based on social rights defined in the Constitution Law, these are: education, health, food, social security, quality and space of the dwelling, and basic services in the dwelling. According to Mexico’s methodology, an individual is considered poor if living below the wellbeing line and with at least one social deprivation. An individual is considered extreme poor if living below the minimum wellbeing line and three or more social deprivations. The monetary component of poverty uses current income per adult equivalent which includes; labor income, public and private transfers, and capital rents. It excludes dwelling imputed rent, and temporal transfers. The well-being line is different in urban and rural areas, defined as localities with a population above/below 2,500 inhabitants. Official poverty estimates are released to the public every two years based on data from the household survey. CONEVAL (Consejo Nacional de Evaluación de la Política de Desarrollo Social) is the designated office to calculate the official poverty rate based on the household survey (Módulo de Condiciones Socioeconómicas de la Encuesta Nacional de Ingresos y Gastos de los Hogares, MCS-ENIGH), produced by INEGI (Instituto Nacional de Estadística y Geografía). Since 2008, poverty estimates are also available at the state level biannually, and municipal poverty estimates are produced every five years. Given recent improvements to the household survey, values of income and expenditures produced in 2016 are not directly comparable with the historical series. Comparisons are only possible using a statistical model developed by CONEVAL and INEGI. Moreover, monetary poverty rates using international poverty lines from the 2016 data should not be compared with pre-2016 numbers. Cross-country comparisons across the Latin American region are based on the data harmonization effort known as the So- cio-economic Database for Latin America and the Caribbean (SEDLAC) – a joint effort of the World Bank and CEDLAS from the National University of La Plata (Argentina). SEDLAC includes 18 countries and more than 300 household surveys since the 80s. Since income-based welfare aggregate is widely used in the region for official poverty estimates, income-based microdata is used for global and regional poverty monitoring. It is important to highlight that the SEDLAC harmonized data set differs in the calculation of the income aggregate from the official national estimation. For the SEDLAC per capita income variable: labor income includes salaries for wage earners, income for self-employed (including self-consumption) and other labor incomes, non-labor income includes: public and private transfers and others and imputed rent is included. In addition, income is calculated over the previous month, while the official calculation uses a six-month average. SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 131 Annex 3. Jobs ANNEXES diagnostic1 This annex provides complementary description on current conditions in labor supply in Mexico and the related prevailing economic returns in wages and labor incomes. It presents stylized facts on the main characteristics of the workforce and the quality of the jobs that labor market participants are obtaining. A diagnostic of the current supply of skills and the returns to human capital is also presented, together with an analysis of the evolution of real wages in the most recent decade.2 Characterizing Mexico’s workforce There is a clear gender gap in labor force participation: while 8 men in 10 participate in the labor market, in the case of women only 4 in 10 do so. Participation in the labor market3 is clearly dominated by men. This pattern persists over time, despite the fact that in the last 13 years, there has been a slight increase in female participation. According to Rendón (2003), the increase in the participation of women in the labor market is explained as much by the decrease in the female mortality rate as by the reduction in fertility, which has resulted in greater prospects of entering the labor market.4 Likewise, Arteco- na and Cunningham (2002) and Bruhn and Love (2011) show that both the process of trade liberalization and the increase in access to the financial system have had positive effects on the employment and wages of Mexican women. Participation in the labor market has increased over time, but gender disparities remain throughout the life cycle. Participation in the labor force presents a clear pattern by gender and age. Participation has shifted toward the 25–34 and older age ranges. In the case of men, between 2005 and 2017, the proportion of 40-year-olds and above who participated in the labor market has increased consistently, while the opposite effect is observed among younger men, particularly, those 15 to 24 years of age. Although the highest participation rates are among the 30–34 age-group, this rate has been declining in the period under study. In the case of women, although participation shows a behavior similar to that of men in favor of greater labor participation in older ages, even the 30–34 age-group exhibits the highest participation. These results indicate a delay in entering the labor market in the first years, which would be partly offset by a longer stay at later ages. A greater demand for qualified personnel may be in part responsible for the observed evolution: young people facing incentives to delay their entry into the labor market in favor of a higher educational level. In turn, the longer time engaged in labor activity shows signs of weakness in retirement coverage. A significant driver of female LFP is education; yet, in Mexico, there is still an important proportion of women with tertiary education who remain outside the labor market. To better understand the drivers of labor force participation and their heterogeneity by gender, a logistic probability regression on sociodemographic characteristics is estimated. Results show that higher educational attainment is associated with higher female LFP in Mexico as in many other countries. However, the proportion of women with higher education levels who are not participating in the labor market is high in comparison with peer countries. Married women and, particularly, women with young children are more likely to remain out of the labor force. SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 1 The authors of this annex are Roy Nuñez and Lourdes Rodríguez-Chamussy. 132 2 The current version of this note focuses on stylized facts based on data of the National Labor Survey (ENOE). 3 Participation in the labor market is defined as the quotient between the economically active population with respect to the population of working age. 4 Rendón (2003). Empleo, Segregación y salarios por género. La situación del trabajo en México, 129-150. Among the 30 million people outside the labor force, there is a significant proportion of youth not in employment, education, ANNEXES or training (NEET). The phenomenon in Mexico is largely related to the low participation of women in the labor market. Young women are nearly four times more likely than young men to be NEETs in Mexico. Motherhood is a key driver of NEET status and overall low participation among women (in the extensive and intensive margins).5 Figure A.3.1. Share of youth not in employment, education or training (NEET) by gender, selected LAC countries 35 25 30 20 in female vesrsus male rates 25 Percentaje points difference 15 20 % of NEET 15 10 10 5 5 0 0 Chile Peru Uruguay Mexico Argentina Colombia Costa Rica Brazil Male Female Gender gap Source: ILOSTAT, 2017. Among the many determinants of labor force participation, environmental conditions and health may exert an important in- fluence. Recent findings in the literature point to a short-run relationship between pollution and work hours. Moderate effects of pollution on health seem to have an important impact on work. The estimations of the effect of a large refinery closing in Mexico City show that a 19.7 percent decline in pollution led to an increase by 1.3 hours in the hours worked per week, a 3.5 percent increase from the baseline.6 Migration dynamics and the labor market Among Mexicans, 7 in 10 who migrate abroad report doing so for work-related reasons. According to the Encuesta Nacional de Dinámica Demográfica 2014, the main reason to migrate abroad is to search for jobs or because jobs have already been found in a foreign country (68 percent), followed by those who report family reunion (14.4 percent) or study motivations (12.4 percent). Although this information is provided by a relative in the country (and not by the migrant), it shows the clear relationship between migration and work.7 Between 2005 and 2010, the foreign population in Mexico went from 0.5 percent to 0.9 percent of the total population. How- ever, between 2010 and 2015, the immigration numbers seem to have remained stable. Although these figures are small compared with the entire population, they do not allow the number to be identified of those who are in Mexico in transit toward the United States or those who, in their migratory process toward the north, have ended up becoming established in Mexico. Where are the jobs and who gets the jobs? Employment is concentrated in nonagricultural activities and in the private sector. The nonagricultural sector has the bulk of the employment. Men participate more in agricultural work, at around 19 percent, in contrast to 7 percent among women. Over the 13 years, there was a reduction in the number of workers in agricultural activities, with mostly men leaving the sector. This result is a reflection of the considerable wage gap between agricultural and nonagricultural sectors as well as the greater demand for labor in sectors such as trade, services, and construction in response to growth and development in urban areas. In terms of participation in the public or private sector, about 95 percent of the working population is in the private sector. SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 133 5 OECD (2017) 6 Hanna and Oliva (2015) 7 ANNEXES Wage employees and the self-employed account for almost 90 percent of the total number of workers with a slight increase in the proportion of wage employment taking place in the last 13 years. The remaining 10 percent is made up of those who work as employers or unpaid workers. However, there are important gender differences. The participation of men as em- ployers is considerably greater (5.6 percent for men versus 2.4 percent for women in 2017), and a higher participation of women in unpaid work (6.8 percent for women versus 3.5 percent for men in 2017). According to Bruhn and Love (2011), the greater access of men and women to the low-income financial system in Mexico has had a positive impact on employ- ment, but in a differentiated way. Thus, while, for women, it has increased the opportunities for salaried employment, it has encouraged the creation among men of new businesses, mostly of an informal nature. Figure A.3.2: Unpaid workers Age group (%) (a) 50.0 20.0 40.0 15.0 30.0 20.0 10.0 10.0 5.0 0.0 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-65 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-65 Age group Age group 2005 2010 2015 2017 2005 2010 2015 2017 Educational level (%) (b) 15.5 17 21 14.1 20 25.2 24.7 23.5 34.3 37.9 39.7 34.5 41.3 36.4 35.8 43.4 27.4 24.7 30.8 31.2 23.7 25.8 27.9 23.7 24.2 20.8 15.1 15.8 12.1 9.7 7.6 15 2005 2010 2015 2017 2005 2010 2015 2017 Males Female Incomplete primary Complete primary Complete secondary Midle and high education Note: ENOE 2005-2017. We only consider the second quarter of each year. Figure (a): left-hand graph shows males whereas the right-hand graph shows females. Figure (b): left-hand graph shows males whereas the right-hand graph shows females. In the case of women, the gain in expected work experience as age increases does not translate into a reduction in participation in unpaid work. In contrast, for men, the proportion of unpaid workers decreases considerably as they grow older. This could be SYSTEMATIC explained by a lower accumulation of work experience among women (for example, the exit from the labor market at the COUNTRY time of pregnancy or maternity) as well as by the intrahousehold distribution of activities and compensation (for example, DIAGNOSTIC MEXICO the unpaid work of small family businesses, mostly informal). Figure A.3.2b shows how unpaid workers are distributed ac- cording to educational level and gender. It is interesting to note that, for both men and women, the highest proportion of unpaid workers have full primary or full secondary education. However, there was an increase in the proportion of workers 134 with medium and higher education in this condition during the period under study. There are clear disparities in participation in formal employment, with the highest levels among men, in urban areas, and in ANNEXES the northern part of the country. Nuevo León, Coahuila, and Chihuahua have the highest rates of formal employment in the country, while Chiapas and Oaxaca have the lowest levels. Figure A.3.3: Employment by states and regions (percentage) Participation rates (%) Formal – informal workers (%) (a) (b) Chiapas Colima Aguascalientes Michoacan CDMX Baja California Quintana Roo Sonora B. C. Sur Colima B. C. Sur Campeche Nayarit Yucatan Coahuila Durango Nayarit Colima Puebla Baja California Chiapas Yucatan Quintana Roo Chihuahua Nuevo Leon Sinaloa CDMX Coahuila Jalisco Durango Guanajuato B. C. Sur Tamaulipas Guerrero Zacatecas Coahuila Hidalgo Guanajuato Puebla Jalisco Tlaxcala Tlaxacala Mexico Campeche Chihuahua Michoacan Jalisco Nuevo Leon Morelos Tamaulipas Aguscalientes Nayarit Hidalgo Campeche Nuevo Leon Sonora Durango Oaxaca Sinaloa Oaxaca Puebla Mexico Guerrero Queretaro Chihuahua San Luis Potosi Quintana Roo Baja Calornia Guanajuato San Luis Potosi Aguascalientes Michoacan Sinaloa Tabasco Morelos Sonora Oaxaca Mexico Tabasco San Luis Potosí Zacatecas Tamaulipas Veracruz Hidalgo Tlaxcala Guerrero Tabasco Veracruz CDMX Queretaro Yucatan Morelos Veracruz Zacatecas Queretaro Chiapas 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 0 20 40 60 Formal informal Northem North-central Central North-central Northem Central North-central Northem Central North-central Northem 82.7 56.5 43.5 55.6 44.4 49.9 38.8 61.2 38.7 61.3 82.3 48.5 Southem 37.3 62.7 38.9 61.1 81.3 46.8 Southem Southem 23.9 76.1 27.0 73.0 Southem Central 80.8 40.9 0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 0 10 20 30 40 50 Formal informal Note: ENOE 2017. We only consider the second quarter in the graph. The left-hand graph shows males whereas the right-hand graph shows females. Formal work is defined as work with access to social security. Regions were considered according to Chiquiar, D., et al. (2017). Northern region includes: Baja California, Chihuahua, Coahuila, Nuevo León, Sonora and Tamau- lipas. North-central region: Aguascalientes, Baja California Sur, Colima, Durango, Jalisco, Michoacán, Nayarit, San Luis Potosí, Sinaloa and Zacatecas. Central region: Mexico City (CDMX), Mexico State, Guanajuato, Hidalgo, Morelos, Puebla, Querétaro and Tlaxcala. Southern region: Campeche, Chiapas, Guerrero, Oaxaca, Quintana Roo, Tabasco, Veracruz and Yucatán. Almost the totality of formal employment in the southern region is in the public sector. About 95 percent of the country’s work is generated by the private sector. However, as we move toward the southern region of the country, the share of pub- lic employment increases both in the case of men and women (6.7 percent and 4.7 percent, respectively) and represents almost the totality of formal employment. The agricultural sector concentrates a higher share of labor in the south and north-central regions of the country. Men’s par- SYSTEMATIC ticipation in agricultural activities is 15 percent on average, with the southern zone of the country having the highest rate COUNTRY (27.3 percent) and the northern zone with the lowest rate (8.1 percent). In the case of women, participation in agricultural DIAGNOSTIC activities does not exceed 3 percent in any region. MEXICO 135 Wage jobs are more prevalent in the north, and self-employment in the south. In the case of men, it is clear that salaried employment increases as we go towards the northern region of the country, while the opposite holds true for self-em- ANNEXES ployment, which represents 28.3 percent of total employment among men in the southern region. Likewise, unpaid work is considerably higher in the southern part of the country (7 percent) compared with the rest of the regions. In the case of women, the structure is similar to men’s. However, the proportion of women employers is minimum in all regions com- pared with the rates presented by their peers of the opposite sex. The labor market is dominated by manufacturing jobs, although the construction and manufacturing sectors concentrate the highest proportions of male labor, while manufacturing, services, and retail concentrate female labor. For men, employment by industry is led by construction and manufacturing activities (81 percent and 71 percent, respectively), while those with lower employment rates are agriculture (65 percent) and commerce (62 percent). In the case of women, manufacturing and services are the sectors that absorb the largest amount of labor (73 percent and 68 percent, respectively), while the ac- tivities with less female labor are construction and agriculture (58 percent and 48 percent, respectively). The employment rate according to formality reveals that the sector that generates the greatest amount of formal work is manufacturing. Thus, 62 percent of men who work in this sector do it formally, while the corresponding rate is 55 percent among women. In the case of men, after manufacturing, the sectors that generate more formal employment are services and commerce, while 70 percent of women who work in the construction sector do so formally. Informal employment is concentrated in activities linked to agriculture in both sexes (91 percent men and 82 percent women). Inclusion in the labor market: Youth and women Female population out of the labor market but willing to work duplicates that of men. Figure A.3.4 presents information about the working-age population that is currently not working, but is available for work. This category includes those people who, for some reason, have become discouraged in their job search. Although both for women and men there was a slight increase between 2005 and 2010 and a stabilization for the rest of the period, the proportion of women in this condition is more than double that of men. Characteristics of the pool of available workers differ by gender, with men showing higher education levels on average. Also, social norms affecting the work search affect men and women differently, and this may explain in part the asymmetry (see Hicks, Santacreu-Vasut, and Shoham 2015). There is a greater proportion of discouraged workers in the central zone of the country, showing some correlation with urban- ization. Both in the case of men and women, the proportion of people outside the labor market who are willing to work is concentrated in the central region of the country (42 percent in the case of men and 40 percent in the women´s case). Next in importance are the north-central and southern parts of the country, while the lowest percentage is the northern region. Figure A.3.4: Available population by sex (millions) by education level (a) (%)(b) 4 4 3.5 3.6 14.6 16.3 3.5 25.8 25.4 22.2 23.9 32.5 35 28.7 34.6 3 3 2.7 35.3 37.6 36.4 39.2 39 38.7 27 27.4 2 2 22.9 23.1 1.4 25.1 22.3 1.4 1.4 20.3 18.5 29.7 1.0 16.1 21.7 17.1 14.5 13 8.2 7.8 1 1 2005 2010 2015 2017 2005 2010 2015 2017 MALES FEMALE Incomplete primary Complete primary 2005 2010 2015 2017 2005 2010 2015 2017 Complete secondary Midle and high education Note: ENOE 2005-2017. We only consider the second quarter of each year. The left-hand graph shows males whereas the right-hand graph shows females. SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 136 Wages, returns to skills/characteristics and labor cost ANNEXES There has been wage compression to the bottom over the past decade in Mexico. Between 2005 and 2017, the proportion of those earning up to 1 minimum wage remained close to 12 percent, while the proportion of those earning between 2 and 3 minimum wages increased. At the same time, the percentage of men who earned 5 of more minimum wages was con- siderably reduced in that period. The dynamics for women are similar however, the proportion earning up to 1 minimum wage is almost double relative to men. Although there is an increase in the number of women who earn up to 3 minimum wages, this advance is lower relative to that registered by men. An important proportion of low-wage earners are in rural areas. Both for men and women, there is a slim proportion of people in the rural area making more than two minimum wages (Figure A.3.5). Figure A.3.5: Workers and minimum wage ranges Workers (%) (a) 100 100 9.0 7.8 9.3 7.9 5.7 4.9 14.3 12.2 13.6 12.2 18.3 16.9 16.4 20.6 80 80 24.5 23.2 22.4 21.6 19.9 21.3 Workens (%) Workens (%) 60 60 29.6 29.6 24.8 28.0 35.6 37.4 32.9 40 40 30.8 29.1 31.5 24.3 25.1 20 20 23.1 21.5 22.7 23.9 12.1 11.5 11.7 12.7 0 0 2005 2010 2015 2017 2005 2010 2015 2017 until 1 MW until 2 MW until 3 MW until 5 MW More than 5 MW by urban-rural area (%) (b) until 1 MW 55.0 45.0 until 1 MW 73.5 26.5 until 2 MW 73.1 26.9 until 2 MW 84.7 15.3 until 3 MW 81.9 18.1 until 3 MW 90.7 9.3 until 5 MW 88.7 11.3 until 5 MW 93.6 6.4 more than 5 MW 93.1 6.9 more than 5 MW 95.8 4.2 0 20 40 60 80 100 0 20 40 60 80 100 Workers (%) Workers (%) urban rural Note: ENOE 2005-2017. We only consider the second quarter of each year. The left-hand graph shows males whereas the right-hand graph shows females. The rural area is defined as localities of less than 2,500 inhabitants. SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO Low-paid jobs are mostly informal; however, even among higher-paying jobs, the rate of informality is high, particularly for men. As one moves toward a better paid category of jobs, the percentage of workers with informal jobs is reduced in favor 137 ANNEXES of formal employment. However, there is a high proportion of those who earn more than 5 minimum wages and are in a situation of informality, both in the case of men and women (41 percent and 27 percent, respectively). There is a positive correlation between better paid jobs and workers education, much stronger among women than among men. The data for 2017 show that the participation of workers with secondary and higher education increases in better paid jobs. The correlation is much higher in the case of women, with nearly 85 percent among those earning more than 5 times the minimum wage having tertiary education, compared with 71 percent among men. Low-paid jobs are related to agricultural activities more prevalent in the southern region of the country. In contrast, for better paid jobs, the distribution of workers becomes more homogeneous in both genders between the northern and central regions of the country, with a clear lag in the southern zone (Figure A.3.6). Figure A.3.6: Workers and minimum wage ranges by industry (%) (a) until 1 MW until 1 MW until 2 MW until 2 MW until 3 MW until 3 MW until 5 MW until 5 MW more than 5 MW more than 5 MW 0 20 40 60 80 100 0 20 40 60 80 100 Workers (%) Workers (%) Construction Manufact. Wholesale and retail Services Other Agriculture Non-specified by regions (%) (b) until 1 MW 9.3 16.5 34.3 39.9 until 1 MW 11.1 20.5 38.8 29.6 until 2 MW 14.0 19.5 42.2 24.3 until 2 MW 17.9 22.8 40.1 19.1 until 3 MW 22.3 24.3 36.5 16.9 until 3 MW 25.5 23.1 37.6 13.8 until 5 MW 23.2 26.2 34.7 15.9 until 5 MW 21.3 24.1 37.4 17.2 more than 5 MW 28.2 23.3 33.2 15.2 more than 5 MW 25.6 21.8 37.2 15.5 0 20 40 60 80 100 0 20 40 60 80 100 Workers (%) Workers (%) Northern North-central Central Southern Note: ENOE 2017. We only consider the second quarter of each year. The left-hand graph shows males whereas the right-hand graph shows females. SYSTEMATIC COUNTRY On average, real labor income has been decreasing in Mexico over the past decade, and the trend is consistent across occu- DIAGNOSTIC pation, gender, and age-group. Real labor income showed a declining trend between 2005 and 2017. The trend is more MEXICO pronounced in the case of employers. In terms of gender, both women and men experienced declining labor income, with the stronger gradient among men. In particular, the labor income among women over 45 years of age showed a slight 138 increase during the period, offsetting losses experienced by younger cohorts. Returns to tertiary education in Mexico are high, but skill premium is decreasing. According to the OECD, the earnings of a ANNEXES person with tertiary education in Mexico must increase 22 percent to break even the cost of the investment. But in fact, earnings rise by almost 100 percent with respect to those without tertiary education. About 70 percent of the men and 85 percent of the women that earn more than 5 minimum wages, have middle school or higher education. Although the skill premium is still high, it is starting to subside. Over the last decade, hourly median wages among the bottom 40 percent have been stagnant at $1.4 USD (2011 PPP) while and the median for the top 60 has contracted about 15 percent. The declining real labor income is mostly an urban phenomenon. In the span of 12 years, wages in urban areas showed a decreasing trend, while wages in rural areas of the country remained stable around their 2005 levels. A marked reduction in real wages between 2007 and 2017 correlated with a strong loss among those in the commerce sector. References Artecona, R. and W.Cunningham, 2002. “Effects of trade liberalization on the gender wage gap in Mexico”. Policy Research Report on Gender and Development Working Paper Series No. 21. Washington, D.C., World Bank. Bruhn, M. and I. Love, 2011. “Gender differences in the impact of banking services: evidence from Mexico”, Small Business Economics, Springer, vol. 37(4), pages 493-512, November. Hanna, R. and Oliva, P. (2015) “The effect of pollution on labor supply: evidence from a natural experiment in Mexico City”, Journal of Public Economics, 122 (2015), pp. 65-79 Hicks, D.L., E. Santacreu-Vasut, and A. Shoham, 2015. “Does mother tongue make for women’s work? Linguistics, household labor, and gender identity”, Journal of Economic Behavior & Organization, vol. 110, Feburary 2015, pages 19-44. OECD, 2017. Building an Inclusive Mexico: Policies and Good Governance for Gender Equality, OECD Publishing, Paris. Rendón (2003). Empleo, Segregación y salarios por género. La situación del trabajo en México, 129-150. SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 139 Annex 4. List of ANNEXES stakeholders that participated in consultations Consultations List of stakeholders consulted Mexico City, October 23 to 27, 2017 Individual Position Entity Private Sector Carlos Marmolejo VP Operations CONSAR Alejandra Vargas Director of Inclusion CONSAR Alejandro Murphy CEO Citla Energy Gustavo Fernández CEO Grupo Torre Médica Iñigo Gutiérrez CEO Laboratorios Polanco Cecilia Sayeg General Manager Consejo Ejecutivo de Empresas Globales César Fragozo Head of the Sectoral Development Unit ProMéxico Luis E. González Coordinator ProMéxico Jordi Valis CEO Suez -BU LATAM - México Timothee Rossignol DPN ICA Juan Pablo Cruz y Corro Sánchez Treasurer FIRA Rebeka Azaola Country Manager Afluenta Diego Armenta Research Director AMEXCAP Alexis Milo Caraza Chief Economist HSBC César A. Vargas IO IFC Manuel Suárez Fernández Analyst Jr. IFC Otto Fritz Analyst Jr. IFC SYSTEMATIC COUNTRY Paulina Nuñez Analyst IFC DIAGNOSTIC MEXICO Violeta Velázquez SIO IFC 140 ANNEXES Individual Position Entity Academia Rosario Cárdenas Research Professor UAM Raymundo Campos Research Professor COLMEX Heidi Smith Research Professor Universidad Iberoamericana Isidro Soloaga Research Professor Universidad Iberoamericana Irving Rosales Research Professor Universidad Iberoamericana Fernanda Márquez Research Professor CIDE John Scott Research Professor CIDE Gerardo Esquivel Research Professor COLMEX Juan Belasteguigoitia Research Professor ITAM Guillermo Cejudo Research Professor CIDE Fernando Nieto Research Professor COLMEX Think-Tanks Rodolfo de la Torre Director of Social Development CEEY Roberto Vélez Grajales Executive Director CEEY Miguel Szekely Managing Director CEES Luis Foncerrada Managing Director CEESP Héctor Villareal Managing Director CIEP Erandi Maria López Program Assistant WeConnect México NGOs and International Development Partners Lucía Baltazar Inv. Analyst IFC María Enriqueta Cepeda Director INCIDE Social Leonor Calderón Director SEGIB Patricia Fernández Subdirector CNDH Roberto Martínez Director OCDE – México Carlos Cabrera Coordinator Fesmex Diego Vázquez Research Manager Oxfam Esperanza Delgado Director of International Relations MexFam Government Managing Director of Payment Systems and Corporate Lorenza Martínez Trigueros Bank of Mexico Services Carlos Márquez Unit Chief of International Relations SHCP Luis Madrazo Lajous Chief Economist SHCP Víctor Hugo Gómez Ayala Head of Investor Relations Office SHCP Alejandra Palacios Prieto President COFECE SYSTEMATIC COUNTRY Amparo Martínez Arroyo Managing Director INEEC-SEMARNAT DIAGNOSTIC MEXICO Miguel Gerardo Breceda General Coordinator of Green Growth INEEC- SEMARNAT 141 Gonzalo Hernández Secretary CONEVAL ANNEXES Consultations List of stakeholders consulted Mexico City, April 9 - 13, 2018 Individual Position Entity SD Meeting Centro de Investigación para el Desarrollo Carlos Zedillo Velasco Director Sostenible (CIDS) Instituto del Fondo Nacional de la Vivienda para Luis Jeremías Diez-Canedo Jaime General Coordinator of Biological Resources los Trabajadores (Infonavit) Instituto Nacional de la Infraestructura Física Héctor Humberto Gutiérrez de la Garza General Director Educativa (INIFED) Deputy Director of Concessions Assignments and Sandra Hernández Alvarado Secretaría de Comunicaciones y Transportes Railway Statistics Banco Nacional de Obras y Servicios Públicos, Carlos Mier y Terán Ordiales Director of Mass Transportation S.N.C. Felipe Arreguin Cortés General Director Instituto Mexicano de Tecnología del Agua (IMTA) Mario López Hydrology coordinator Instituto Mexicano de Tecnología del Agua (IMTA) Oficina de Resiliencia de la Ciudad de México de la Arnoldo Matus Kramer Director Secretaría del Medio Ambiente (SEDEMA) Marcelino Madrigal Commissioner Comisión Reguladora de Energía (CRE) Jorge David Fernández Medina General Coordinator of Planning and Information Comisión Nacional Forestal Griselda Medina Laguna Deputy Manager of Project Management Comisión Nacional del Agua Government and International Development Partners Carlos Márquez Padilla Unit Chief of International Relations SHCP Alejandrina Salcedo Chief Economist SHCP Managing Director of Payment Systems and Lorenza Martínez Trigueros Banco de México Corporate Services Othón M. Moreno González Policy Manager of Payment Systems Banco de México David Kaplan Senior Economist IDB Government (Tuxtla Gutiérrez, Chiapas) Ovidio Cortázar Ramos Secretary Secretary of Economy Ovidio Cortázar Ramos Secretary Secretary of Economy Andrés Montesinos Ramírez General Director of Investments Secretary of Economy Undersecretary of Promotion and Development of Carlos Alberto Salazar Estrada Secretary of Economy Competitiveness Director of Investment Promotion and Strategic Miguel Ángel Vázquez Castañeda Secretary of Economy Projects Carlos Montezuma Trujillo Director of Projects and Management Secretary of Economy SYSTEMATIC COUNTRY DIAGNOSTIC Director of Entrepreneurship and Boosting Manuel Alejandro Gallegos Cancino Secretary of Economy MEXICO Competitiveness Esperanza Zepeda Coutiño Undersecretary of Commerce Secretary of Economy 142 ANNEXES Individual Position Entity Subsecretaría de Desarrollo Industrial y Atracción Gabriel José Beltrán Rodríguez Secretary de Inversiones Director of Impulse to the Quality of Products and Laura Lorena Ponce Rocha Services Ernesto Gutiérrez Coello Technical Coordinator Carlos Eduardo Suárez Argüello Undersecretary of Tourism Development Secretary of Tourism Óscar Gerardo Ochoa Gallegos Secretary Secretary of Labor Rosa L. Trujillo V. Director of Employment Support Secretary of Labor Alejandro Bante Martínez Adviser Secretary of Labor Saul O. Santiago Adviser Secretary of Labor Private Sector and NGOs (Tuxtla Gutiérrez, Chiapas) José Antonio Toriello President COPARMEX Miguel Ángel Muñóz President Marcas Chiapas Eduardo González Castañón Manager Fondo Chiapas Gustavo González General Director Fondos Chiapas Juan José Zepeda President Comisión Estatal de los Derechos Humanos Government (Querétaro) Marco Antonio del Prete Secretary SEDESU Guillermo Lozano Director of Industrial Promotion SEDESU Private Sector (Querétaro) Miguel Ángel Castillero Director Grupo Brose Jose Ant. Velázquez General Director Aeroclúster Querétaro Ana Laura Gómez Medina Manager CANACINTRA Academia (Querétaro) Federico Pérez Fuentes Planning Director Universidad Aeronáutica Querétaro SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 143 Annex 5. Data sources ANNEXES Country: MEXICO Last reviewed: June, 2018 Section 1: General Information about the Statistical System Legal status of NSO Autonomous Public Agency Statistical Legislation (latest) Law, 2008 NSDS/Statistical masterplan National Statistic and Geography Program 2013-2018 Section 2: Micro data Data Representativeness Accessibility Optional Latest Second Type of Census/Survey (national, regional, (open access/ Disaggregation (Year) Latest (Year) urban/rural) with permission/ (Y/N) no access)   Sex Regional Censuses •• National •• State Censo de Población y Vivienda 2010  10 years •• Municipality No Access Y Y •• Locality •• AGEB •• National •• State Conteo de Población y Vivienda 2005 10 years •• Municipality No access Y Y •• Locality •• AGEB •• National •• State Encuesta Intercensal 2015 TBD No access Y Y •• Municipality •• Locality Actualización Marco Censal Agropecuario 2016 TBD No access •• Rural areas Censo Agrícola, Ganadero y Forestal 2007 1991 •• Urban areas No access with livestock activities SYSTEMATIC Censo Ejidal 2007  Random •• Ejido  No access     COUNTRY DIAGNOSTIC MEXICO 144 ANNEXES Section 2: Micro data Data Representativeness Accessibility Optional Latest Second Type of Census/Survey (national, regional, (open access/ Disaggregation (Year) Latest (Year) urban/rural) with permission/ (Y/N) no access)   Sex Regional •• National •• State Censo Económico 2014 5 years No access Y Y •• Municipaliy •• Locality •• National Censo Nacional de Procuración de Justicia 2017 Annual Open access •• State •• National Censo Nacional de Impartición de Justicia 2017 Anual Open access •• State Censo Nacional de Transparencia, Acceso a la •• National 2016 Anual Open access Información y Protección de Datos Personales •• State Censo Nacional de Gobierno, Seguridad Pública y •• National 2016 Annual Open access Y Y Sistema Penitenciario Estatales •• State Censo Nacional de Gobiernos Municipales y 2015 2 years •• Municipality Open access Delegacionales Surveys Household Surveys Encuesta Nacional de Ingresos y Gasto de los 2016 2 years •• Urban/rural Open access Y N Hogares (ENIGH) Módulo de Condiciones Socioeconómicas ENIGH •• Urban/rural 2016 2 years Open access Y Y (MCS-ENIGH)  •• State Encuesta Nacional de Calidad e Impacto 2015 2 years •• Urban Open access N Y Gubernamental (ENCIG)  •• Urban/rural Encuesta Nacional de Gasto (ENGASTO)  2013 Random 2012 Open access Y Y •• State •• Urban/rural Encuesta Nacional de Hogares (ENH)  2017 Annual Open access Y Y •• State •• State Encuesta Nacional de Ocupación y Empleo (ENOE) 2018-QI Quarterly •• 32 Cities Open access Y Y •• Locality size Encuesta Nacional de Seguridad Pública (ENSU)  2018-QI Quarterly •• Urban areas Open access Y Y Encuesta Nacional de Victimización y Percepción •• Urban/rural 2017 Annual Open access Y Y sobre Seguridad Pública (ENVIPE)  •• State Encuesta Sobre Confianza del Consumidor (ENCO)  2018-Abril Monthly •• Urban areas Open access Y N Encuesta Nacional sobre Disponibilidad y Uso de •• State Tecnologías de la Información en los Hogares 2017 Annual Open access Y Y (ENDUTIH)  •• Cities Encuesta Nacional de Inclusión Financiera (ENIF)  2015 3 years •• Urban/rural Open access Y N Encuesta Nacional de la Dinámica Demográfica 2014 Random 2009 •• Locality size Open access Y Y (ENADID)  SYSTEMATIC Encuesta Nacional sobre la Dinámica de las •• Urban/rural COUNTRY 2016 Random 2011 Open access Y Y DIAGNOSTIC Relaciones en los Hogares (ENDIREH)  •• State MEXICO Encuesta Nacional sobre Uso del Tiempo (ENUT) 2014 Random 2009 •• Urban/rural Open access 145 ANNEXES Section 2: Micro data Data Representativeness Accessibility Optional Latest Second Type of Census/Survey (national, regional, (open access/ Disaggregation (Year) Latest (Year) urban/rural) with permission/ (Y/N) no access)   Sex Regional Encuesta sobre la Percepción Pública de la Ciencia 2015 2 years •• Urban Open access y la Tecnología (ENPECYT)  Encuesta Nacional de Educación, Capacitación y •• Urban/rural 2009 2 years Open access Y Y Empleo (ENECE-ENOE)  •• State Encuesta Nacional de Empleo y Seguridad Social 2013 4 years •• State Open access Y Y (ENESS-ENOE)  Encuesta Nacional de Micronegocios (ENAMIN)  2012 2 years •• National Open access Y N •• Urban/rural Módulo de Trabajo Infantil (MTI-ENOE)  2015 2 years Open access Y Y •• State •• Urban areas Módulo Trayectorias Laborales (MOTRAL)  2015 3 years Open access Y Y •• 32 cities Evaluación Nacional del Logro Académico en  2014  Annual •• State N Y Centros Escolares (ENLACE) Encuesta Nacional de Salud y Nutrición (ENSANUT) 2016  6 years  •• State  Open access Y Y World Health Survey 2003 •• National Open access Y N •• National 2005-2006 Mexican Family Life Survey (MXFLS) 2012-2009 •• Urban/Rural Open access Y Y 2002 •• Regional Business/Establishment Survey •• National Encuesta Anual de Empresas Constructoras (EAEC)  2015 Annual No access •• State  Encuesta Anual de la Industria Manufacturera  2015  Annual •• National  No access     (EAIM)  Encuesta Mensual de la Industria Manufacturera 2017-may Monthy •• National  No access (EMIM)  Encuesta Nacional de Empresas Constructoras •• National 2017-may Monthly No access (ENEC)  •• State  Encuesta Anual de Servicios Privados No 2015 Annual •• National  No access Financieros (EASPNF)  Encuesta Anual de Transportes (EAT)  2015 Annual •• National No access •• National Encuesta Anual del Comercio (EAC) 2015 Annual No access •• State Encuesta Mensual de Servicios (EMS)  2017-may Monthly •• National No access Encuesta Mensual sobre Empresas Comerciales •• National 2017-may Monthly No access (EMEC)  •• State Encuesta Mensual de Opinión Empresarial (EMOE)  2017-may Monthly •• National No access •• National Encuesta Nacional de Calidad Regulatoria e •• State Impacto Gubernamental en Empresas (ENCRIGE) 2016 TBD No access SYSTEMATIC 2016  •• Some COUNTRY municipalities DIAGNOSTIC MEXICO Encuesta Nacional de Victimización de Empresas •• National 2016 2 years With permission (ENVE) •• State 146 ANNEXES Section 2: Micro data Data Representativeness Accessibility Optional Latest Second Type of Census/Survey (national, regional, (open access/ Disaggregation (Year) Latest (Year) urban/rural) with permission/ (Y/N) no access)   Sex Regional Encuesta Nacional sobre Productividad y •• National Competitividad de las Micro, Pequeñas y Medianas 2015 TBD No access Empresas (ENAPROCE)  •• State Encuesta sobre Investigación y Desarrollo •• National 2014 2 years No access Tecnológico (ESIDET) 2014  •• State Encuesta sobre Tecnologías de la Información y las 2013 TBD •• National No access Comunicaciones (ENTIC 2013)  •• National (for 34 products) Encuesta Nacional Agropecuaria (ENA) 2014 2012 No access •• State (relevant products) Enterprise Survey 2010 2006 •• National Open access Other Sources Global Financial Inclusion (Global Findex) 2017 2014, 2011 •• National Open access Y N •• Mexico City and Doing Business, Measuring Business Regulations 2018 Annual Open access N N Monterrey Doing Business, Measuring Business Regulations 2016 Random •• 32 States Open access N Y - Subnational The World Justice Project. Rule of Law Index 2018 Annual •• National Open access N N 2017-2018 UN-Habitat. Citi Prosperity Initiative 2016 NA •• 153 cities Open access Y Y Human Development Index HDI 2015 Annual •• National Open access Y N Environmental Performance Index-EPI YALE 2018 2 years •• National Open access N N Corruption perception index. Transparency 2017 Annual •• National Open access N N International Economic Complexity Ranking – ECI MIT 2016 Annual •• National Open access N N SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 147 ANNEXES Section 3: Macro Data Does the country subscribe to the IMF SDDS SDDS or participate in the eGDDS? Periodicity Timeliness If eGDDS - eGDDS Data Category   SDDS Country SDDS Country National accounts: Gross Domestic Product by Production and Q Q 1Q 53D Expenditure at Current and Constant Prices Consumer price index M F 1M 10D Central government operations M M 1M 30D Balance of payments Q Q 1Q 10W External debt Q Q 1Q 1Q Merchandise trade M M 8W 25D Production index M M 6W NLT 42D Employment Q M 1Q NLT 25D Unemployment Q M 1Q NLT 25D Producer Price Index M M 1M NLT 10D WHO Global Health Expenditure Mexico has National Health Account data following the SHA 2011 norm Section 4: Compliance with WBG’s Core Data Standards Compliant WBG Standard Actual yearly interval or % (Y/N) Household survey of income or consumption  One every 3 years Y 2 years PPP price survey One every three years Y 3 years 80% of births registered 95% (2015) CRVS Y 60% of deaths registered with cause of death 99% (2015) Section 5: Statistical Capacity Indicators Method 90 Source Data 100 Periodicity 90 Overall 93.3 Section 6: Data Openness Indicators Open Data Barometer Score 73 score/position 11 of 114 SYSTEMATIC Open Data Index Score  65 score/position 11 of 94 COUNTRY DIAGNOSTIC MEXICO 148 Annex 6. ANNEXES Benchmarking and prioritization exercise A prioritization of the impediments is supported by an analytical analysis benchmarking Mexico’s performance against selected peers, OECD countries, upper middle-income countries, the region and the World in dimensions corresponding to the identified impediments (See Annex 1, identification of peers). A set of pre-selected indicators was used to assess the position of Mexico in relation of these impediments. The original set of indicators was complemented with those suggested by experts and using additional sources of information when the data were available for a large set of countries worldwide. The comparison is based on the calculation of the distance between Mexico and the best performer in each specific indi- cator according to the following methodology: Where v Best Performance and v Low Performance correspond to the highest and lowest value respectively in the percentiles 95th and 5th and v mex denotes the value for Mexico. Information of the latest year available is used for each indicator, therefore the corresponding period of time may vary depending on the specific source of information and its latest avail- ability. The percentile 5th and 95th are used to define the umbral for the top and bottom performers with the objective of reducing possible distortions caused by outliers. The resulted distance is defined as a range between 0 and 100; where 0 denotes that Mexico is part of the top 5 percent best achievers, and 100 implies that Mexico is part of the bottom 5 percent worst performance. The compiled dataset was validated with the teams within the World Bank, who were requested to contribute with advice regarding the selected indicators for each impediment and to complete with additional indicators for their sector or area. Among the consulted sources, selected indicators come from the Global Findex Database 20178, Doing Business 20189, The Global Competitiveness Report 2017-201810, The World Justice Project Rule of Law Index11, CPI unhabitat12, Human Devel- opment Data13, Yale’s Environmental Performance Index 201714, Transparency International 201715. Finally, 178 indicators paired with one of the sub impediments.  he performance of Mexico for each indicator and the benchmarking group was classified as high (red), medium-high T (yellow), medium-low (grey) and low (green). High indicated that the difference between Mexico and the best performance SYSTEMATIC COUNTRY 8 World Bank. The global Findex Database 2017 https://globalfindex.worldbank.org/ DIAGNOSTIC 9 World Bank. Doing Business, Measuring Business Regulations, 2018 http://www.doingbusiness.org/ MEXICO 10 World Economic Forum. https://www.weforum.org/reports/the-global-competitiveness-report-2017-2018 11 The World Justice Project. Rule of Law Index 2017-2018 https://worldjusticeproject.org/our-work/wjp-rule-law-index 12 UN-Habitat. Citi Prosperity Initiative. http://cpi.unhabitat.org/download-raw-data 149 13 United Nations Development Programme. Human Development Reports. http://hdr.undp.org/en/data 14 Yale’s Environmental Performance Index. https://epi.envirocenter.yale.edu/. 15 Transparency International. https://www.transparency.org/ ANNEXES in the group is 75 percent or more the gap between the top and lowest performers. Medium-high indicated that Mexico difference with the top performer is between 50 and 75 percent the gap with the lowest performer. Medium-low indicated that the difference is between 25 and 50 percent the gap, and Low indicated that the difference between Mexico and the top performance is less than 25 percent the gap with the lowest performer. Overall score in main impediments In each sub-impediment, the results were aggregated using an un weighted average of the distance calculated for the subset of indicators. For the three main categories of impediments, an average of the corresponding sub impediments is calculated in the same fashion. Distance between Mexico and best performer reference in each reference group Indicator Peers OECD Income Group Region World Product and factor market issues ● 70% ● 83% ● 57% ● 52% ● 63% The role of public sector institutions ● 79% ● 91% ● 70% ● 64% ● 75% Allocation of public resources ● 69% ● 85% ● 64% ● 65% ● 60% Structural constraints to growth, inclusion and sustainability ● 73% ● 82% ● 65% ● 62% ● 66% Distance between Mexico and best performer reference in each reference group Impediment/Indicator Peers OECD Income Group Region World 1. Concentration (and market power) in critical input markets ● 71% ● 88% ● 50% ● 42% ● 55% and barriers to entry at the local level 2. Acces to finance ● 83% ● 89% ● 74% ● 68% ● 72 3. Labor markets and informality ● 55% ● 72% ● 47% ● 45% ● 61% 4. Access to justice ● 78% ● 83% ● 62% ● 53% ● 66% 5. Corruption ● 70% ● 91% ● 64% ● 65% ● 73% 6. Crime and violence ● 90% ● 100 % ● 84% ● 73% ● 88% 7. Taxs structure and tax expenditures ● 52% ● 72% ● 63% ● 58% ● 59% 8. Public spending: rigidities, inefficiencies and distributional ● 70% ● 93% ● 57% ● 53% ● 46% issues 9. Coordination inefficiencies and investment planning ● 85% ● 90% ● 72% ● 84% ● 75% 10 Investment infraestructure ● 74% ● 93% ● 60% ● 60% ● 67% 11. Quality and utilization of human capital ● 73% ● 91% ● 62% ● 59% ● 62% 12. Management of natural capital ● 73% ● 62% ● 72% ● 67% ● 69% SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 150 I. Product and factor market issues ANNEXES Distance between Mexico and best performer reference in each reference group Impediment/Indicator Peers OECD Income Group Region World 1.Concentration (and market power) in critical input markets and ● 71% ● 88% ● 50% ● 42% ● 55% barriers to entry at the local level 2. Access to finance ● 83% ● 89% ● 74% ● 68% ● 72% 3. Labor markets and informality ● 55% ● 72% ● 47% ● 45% ● 61% 1. Concentration (and market power) in critical input markets and barriers to entry at the local level Distance between Mexico and best performer reference in each reference group Income Value World Impediment/Indicator Peers OECD Region World Year Group Mex P95 Availability of latest technologies ● 49% ● 91% ● 24% ● 22% ● 42% 5 2017 6 Effectiveness of antimonopoly policy ● 53% ● 91% ● 35% ● 21% ● 59% 4 2017 5 Extent of market dominance ● 83% ● 91% ● 56% ● 34% ● 65% 3 2017 5 Getting electricity (DB16-18 methodology) ● 100% ● 84% ● 33% ● 29% ● 31% 69 2018 93 Individula using Internet % ● 70% ● 100% ● 48% ● 23% ● 41% 60 2017 94 Individula using Internet (% of population) ● 70% ● 100% ● 30% ● 37% ● 40% 60 2016 95 Intesity of local competition ● 49% ● 64% ● 29% ● 12% ● 38% 5 2017 6 Internet access in schools ● 89% ● 89% ● 68% ● 56% ● 60% 4 2017 6 Internet bandwidth ● 97% ● 100% ● 88% ● 84% ● 92% 38 2017 449 Mobile broadband subscriptions ● 80% ● 91% ● 59% ● 50% ● 59% 59 2017 129 Mobile telephone subscriptions ● 100% ● 100% ● 96% ● 82% ● 71% 88 2017 167 No days to start a business (reversed) ● 6% ● 32% ● 5% ● 2% ● 9% 8 2017 1 No procedures to star a business (reversed) ● 50% ● 100% ● 50% ● 33% ● 50% 8 2017 3 Production process sophistication, 1-7 (best) ● 48% ● 86% ● 18% ● 5% ● 49% 4 2017 6 Secure Internet servers (per 1 million people) ● 100% ● 100% ● 82% ● 94% ● 98% 41 2016 2075 Used the internet to buy something online in the past year ● 96% ● 100% ● 86% ● 79% ● 92% 7 2017 69 (% age 15+) SYSTEMATIC COUNTRY DIAGNOSTIC Used the internet to pay bills or to buy something online in MEXICO ● 91% ● 100% ● 81% ● 69% ● 87% 13 2017 80 the past year (% age 15+) 151 ANNEXES 2. Access to finance Distance between Mexico and best performer reference in each reference group Income Value World Impediment/Indicator Peers OECD Region World Year Group Mex P95 Account (% age 15+) ● 100% ● 100% ● 86% ● 87% ● 80% 37 2017 99 Affordability of financial services, 1-7 (best) ● 53% ● 82% ● 59% ● 52% ● 55% 4 2017 5 Coming up with emergency funds: possible (% age 15+) ● 100% ● 100% ● 100% ● 100% ● 100% 27 2017 81 Deposit in the past year (% with a financial institucion ● 100% ● 100% ● 78% ● 65% ● 70% 59 2017 95 account, age 15+) Domestic credit to private sector (% of GDP) ● 84% ● 100% ● 84% ● 77% ● 82% 35 2016 144 Getting credit (db15-18methodology) ● 10% ● 6% ● 5% ● 6% ● 10% 90 2018 100 Gross domestic savings (% of GDP) ● 72% ● 83% ● 61% ● 48% ● 44% 21 2016 46 No account because financial institucions are too far away (% without a financial institution account, age 15+) ● 82% ● 100% ● 86% ● 76% ● 80% 29 2017 5 (reversed) No account because of lack of necessary documentation ● 100% ● 100% ● 77% ● 72% ● 61% 19 2017 2 (% age 15+) (reversed) Paid utility bills: ussing mobile phone (% paying utility ● 91% ● 96% ● 89% ● 81% ● 89% 5 2017 41 bills, age 15+) Paid utility bills: ussing account (% paying utility bills, ● 100% ● 100% ● 91% ● 100% ● 91% 12 2017 91 age 15+) Real interest rate (%) ● 79% ● 84% ● 60% ● 56% ● 92% 0 2016 21 Received private sector wages: into a financial institucion ● 89% ● 100% ● 65% ● 54% ● 66% 41 2017 96 account (%wave recipients, age 15+) Received wages: into an account (% wage recipients, age ● 100% ● 100% ● 69% ● 59% ● 70% 45 2017 97 15+) Received wages: through a mobile phone (% wage ● 89% ● 88% ● 94% ● 84% ● 94% 2 2017 26 recipients, age 15+) SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 152 3. Labor markets and informality ANNEXES Distance between Mexico and best performer reference in each reference group Income Value World Impediment/Indicator Peers OECD Region World Year Group Mex P95 Cooperation in labor employer relations, 1-7 (best) ● 50% ● 73% ● 42% ● 40% ● 60% 4 2017 6 Country capacity to attract talent, 1-7 (best) ● 47% ● 66% ● 39% ● 35% ● 56% 4 2017 5 Country capacity to retain talent, 1-7 (best) ● 52% ● 69% ● 38% ● 42% ● 55% 3 2017 5 Fundamental labor rights are effectively guaranteed ● 56% ● 92% ● 44% ● 88% ● 65% 1 2017 1 Hiring and firing practices, 1-7 (best) ● 47% ● 67% ● 44% ● 28% ● 68% 3 2017 5 Labor Market Efficiency ● 66% ● 95% ● 65% ● 47% ● 78% 4 2017 5 Pay and productivity, 1-7 (best) ● 80% ● 90% ● 61% ● 36% ● 61% 4 2017 5 Redundancy costs, weeks of salary ● 44% ● 20% ● 41% ● 45% ● 44% 22 2017 37 II. The role of public sector institutions Distance between Mexico and best performer reference in each reference group Impediment/Indicator Peers OECD Income Group Region World Rule of law and unequal application of the law ● 78% ● 83% ● 62% ● 53% ● 66% Corruption ● 70% ● 91% ● 64% ● 65% ● 73% Crime and violence ● 90% ● 100% ● 84% ● 73% ● 88% 4. Access to justice Distance between Mexico and best performer reference in each reference group Income Value World Impediment/Indicator Peers OECD Region World Year Group Mex P95 Alternative dispute resolution mechanisms are ● 82% ● 100% ● 70% ● 74% ● 69% 1 2017 1 accessible, impartial and effective Civil justice is effectively enforced ● 100% ● 100% ● 91% ● 77% ● 90% 0 2017 1 Civil justice is free of discrimination ● 98% ● 100% ● 100% ● 100% ● 96% 0 2017 1 Civil justice is not subject to unreasonable delay ● 94% ● 100% ● 93% ● 71% ● 90% 0 2017 1 Complaint mechanisms ● 56% ● 74% ● 38% ● 51% ● 53% 1 2017 1 Enforcing contracts (DB17-18 methodology) ● 38% ● 55% ● 27% ● 6% ● 24% 65 2018 76 Intellectual property protection ● 66% ● 98% ● 40% ● 27% ● 66% 4 2017 6 Judicial independence (WEF) ● 100% ● 100% ● 66% ● 62% ● 84% 3 2017 6 Legal rights index, 0-10 (best) ● 20% ● 11% ● 18% ● 10% ● 10% 10 2017 11 SYSTEMATIC COUNTRY People can access and afford civil justice ● 100% ● 100% ● 100% ● 100% ● 94% 0 2017 1 DIAGNOSTIC MEXICO Property rights (WEF) ● 77% ● 97% ● 61% ● 38% ● 70% 4 2017 6 Strength of investor protection ● 100% ● 63% ● 43% ● 21% ● 39% 6 2017 8 153 ANNEXES 5. Corruption Distance between Mexico and best performer reference in each reference group Income Value World Impediment/Indicator Peers OECD Region World Year Group Mex P95 Civil justice is free ofimproper government influence ● 39% ● 79% ● 43% ● 46% ● 56% 1 2017 1 Corruption Perceptions Index 2017-Transparency ● 100% ● 100% ● 73% ● 85% ● 82% 29 2017 82 International Criminal system is free ofimproper government influence ● 60% ● 100% ● 57% ● 56% ● 67% 0 2017 1 Due process is respected in administrative proceedings ● 69% ● 93% ● 62% ● 62% ● 74% 0 2017 1 Government officials are sanctioned for misconduct ● 100% ● 100% ● 100% ● 89% ● 100% 0 2017 1 Government officials in the exejutive branch do not use ● 100% ● 100% ● 100% ● 100% ● 95% 0 2017 1 public office for private gain Government officials in the judicele branch do not use ● 100% ● 100% ● 92% ● 79% ● 87% 0 2017 1 public office for private gain Government officials in the legislative branch do not use ● 78% ● 100% ● 78% ● 79% ● 88% 0 2017 1 public office for private gain Government officials in the police and the military do not ● 100% ● 100% ● 100% ● 92% ● 92% 0 2017 1 use public office for private gain Goverment powers are effectively limited by independent ● 81% ● 100% ● 76% ● 78% ● 89% 0 2017 1 audating and review Goverment powers are effectively limited by the judiciary ● 51% ● 94% ● 59% ● 59% ● 69% 0 2017 1 Goverment powers are effectively limited by the ● 53% ● 66% ● 40% ● 46% ● 52% 1 2017 1 legislature Goverment powers are subject to non-governmental ● 45% ● 94% ● 45% ● 80% ● 53% 1 2017 1 checks Goverment regulations are applied and enforced without ● 100% ● 100% ● 91% ● 100% ● 91% 0 2017 1 improper influence Goverment regulations are effectively enforced ● 67% ● 100% ● 50% ● 44% ● 66% 0 2017 1 Public trust in politicians ● 87% ● 100% ● 92% ● 87% ● 97% 2 2017 6 Publicized laws and government data ● 9% ● 63% ● 6% ● 8% ● 28% 1 2017 1 Right to information ● 35% ● 72% ● 19% ●22% ● 35% 1 2017 1 Transparency of government policymaking ● 51% ● 67% ● 36% ● 30% ● 59% 4 2017 6 SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 154 6. Crime and violence ANNEXES Distance between Mexico and best performer reference in each reference group Income Value World Impediment/Indicator Peers OECD Region World Year Group Mex P95 Correctional system es effective in reducing criminal behavior ● 99% ● 100% ● 91% ● 79% ● 94% 0 2017 1 Crime is effectively controlled ● 100% ● 100% ● 100% ● 81% ● 100% 0 2017 1 Criminal adjudication system is timely and effective ● 98% ● 100% ● 97% ● 71% ● 99% 0 2017 1 Criminal investigation system is effective ● 82% ● 100% ● 81% ● 74% ● 90% 0 2017 1 Criminal investigation is impartial ● 80% ● 100% ● 82% ● 80% ● 87% 0 2017 1 Intentional homicides (per 100,000 people) ● 60% ● 100% ● 36% ● 21% ● 51% 16 2017 1 Organized crime ● 100% ● 100% ● 100% ● 80% ● 100% 3 2017 6 People do not resort violence to redress personal grievances ● 96% ● 100% ● 80% ● 78% ● 87% 0 2017 1 Reability of police services ● 100% ● 100% ● 100% ● 89% ● 100% 2 2017 6 The right to life and security the person is effectively ● 82% ● 100% ● 74% ● 80% ● 70% 0 2017 1 guaranteed III. Allocation of public resources Distance between Mexico and best performer reference in each reference group Impediment/Indicator Peers OECD Income Group Region World 7. tax structure tax expenditures ● 52% ● 72% ● 63% ● 58% ● 59% 8. Public spending: rigidities, inefficiencies and distributional ● 70% ● 93% ● 57% ● 53% ● 46% issues 9. Coordination inefficiencies and investment planning ● 85% ● 90% ● 72% ● 84% ● 75% SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 155 ANNEXES 7. Tax structure and tax expenditures Distance between Mexico and best performer reference in each reference group Income Value World Impediment/Indicator Peers OECD Region World Year Group Mex P95 Direct taxes as % of GDP/Indirect taxes as a % of GDP ● 35% ● 67% ● 53% ● 11% ● 55% 1 2016 2 VAT tax rate ● 50% 16 2018 5 27 Tax revenue (% of GDP) ● 74% ● 83% ● 73% ● 89% ● 63% 14 2016 28 Taxes on good and services (%of GDP) (indirect taxes) ● 83% ● 64% ● 78% ● 87% ● 67% 0 2016 0 Taxes on income, profits and capital gains (% of GDP) (direct ● 15% ● 73% ● 46% ● 46% ● 50% 0 2016 0 taxes) Subnational government taxes as percentage of total ● 95% 7 2015 76 subnational revenue 8. Public spending: rigidities, inefficiencies and distributional issues Distance between Mexico and best performer reference in each reference group Income Value World Impediment/Indicator Peers OECD Region World Year Group Mex P95 Equitable Seconadary School Enroliment girls ration in main cities ● 100% ● 100% ● 100% ● 100% ● 77% 1 2015 1 General government expediture as a percentage of GDP ● 82% 25 2015 58 Government expediture aper student, primary (% GDP per capital) ● 83% ● 98% ● 69% ● 79% ● 61% 15 2014 29.64 Government expediture aper student, secondary (% GDP per ● 71% ● 100% ● 65% ● 52% ● 66% 16 2014 36.46 capital) Improved sanitations facilities, rural (% of rural population with ● 53% ● 100% ● 43% ● 59% ● 27% 1490 2015 2000 access) Improved sanitations source, rural (% of rural population with ● 63% ● 100% ● 24% ● 16% ● 14% 1842 2015 2000 access) Improved sanitations source, urban (% of rural population with ● 33% ● 100% ● 18% ● 8% ● 10% 1944 2015 2000 access) Inequality-adjusted education index ● 98% ● 100% ● 67% ● 50% ● 50% 1 2015 1 Inequality-adjusted income index ● 78% ● 100% ● 52% ● 32% ● 53% 1 2015 1 Inequality-adjusted life expectancy index ● 78% ● 100% ● 24% ● 31% ● 27% 1 2015 1 Overall loss in HDI die to inequality (%) (reversed) ● 83% ● 100% ● 54% ● 52% ● 52% 23 2015 7 People practing open defecation, rural (% of rural population) ● 27% ● 100% ● 19% ● 16% ● 8% 6 2015 0 SYSTEMATIC COUNTRY People using at least basic drinking water services, rural (%% of DIAGNOSTIC ● 23% ● 100% ● 15% ● 16% ● 9% 94 2015 100 rural population) MEXICO People using safely managed drinking water services (%of ● 100% ● 100% ● 100% ● 100% ● 78% 43 2015 100 156 population) ANNEXES Income Value World Impediment/Indicator Peers OECD Region World Year Group Mex P95 Primary education enrollment net % ● 34% ● 70% ● 37% ● 31% ● 19% 95 2017 100 Received a public sector pension in the past year (% age 15+) ● 95% ● 100% ● 95% ● 83% ● 88% 5 2017 28 Rural poverty headcount ratio at national poverty lines (% of ● 81% ● 100% ● 100% ● 77% ● 81% 62 2014 5 rural population) (reversed) Secondary education enrollment gross % ● 75% ● 100% ● 56% ● 57% ● 42% 91 2017 130 Subnational government expediture as a percentage of GDP ● 68% 13 2015 35 Subnational government expediture as a percentage of general ● 35% 52 2015 76 government expenditure Subnational government revenue in property income as ● 100% 0 2015 9 percentage of totral subnational revenue Subnational government revenue in social contributions as ● 84% 1 2015 7 percentage of totral subnational revenue Subnational government revenue in tariffs & fees income as ● 100% 0 2015 27 percentage of totral subnational revenue Subnational government taxes as percentage of general ● 91% 7 2015 56 government revenue in taxes Tertiary education enrollment gross % ● 94% ● 100% ● 83% ● 88% ● 70% 30 2017 89 9. Coordination inefficiencies and investment planning Distance between Mexico and best performer reference in each reference group Income Value World Impediment/Indicator Peers OECD Region World Year Group Mex P95 Equitable Seconadary School Enroliment girls ration in main ● 100% ● 100% ● 100% ● 100% ● 77% 1 2015 1 cities DTF-Paying taxes (DB17-18 methodology) ● 37% ● 100% ● 32% ● 20% ● 33% 67 2018 91 Fiscal and Financial Management (InCiSe Index): economic evaluation, medium term budgeting, and performance ● 36% 1 2017 1 budgeting Human Rwaources (InCiSe Index): meritocratic recruitment, and ● 87% 0 2017 1 retaining talent Inclusiveness (InCiSe Index): proportion of women in civil ● 86% 0 2017 1 service, and ethnic/religious minority representation Integrity (InCiSe Index): corruption, adherence to rules, work ethics, impartiality, serving citizens, and processes to prevent ● 100% 0 2017 1 conflict Participation anbd Accountability Voter Turnout in main cities ● 88% ● 100% ● 42% ● 100% ● 64% 51 2015 93 Public private partnerships investment in energy (% of GDP) ● 100% ● 100% ● 90% ● 100% ● 97% 0 2016 0.02 SYSTEMATIC COUNTRY Public private partnerships investment in transport (% of GDP) ● 100% ● 100% ● 96% ● 100% ● 96% 0 2016 0.04 DIAGNOSTIC MEXICO Public private partnerships investment in water and sanitation ● 100% ● 100% ● 100% ● 100% ● 86% 0 2015 0 (% of GDP) 157 ANNEXES IV. Structural constraints to growth, inclusion and sustainability Distance between Mexico and best performer reference in each reference group Impediment/Indicator Peers OECD Income Group Region World Investment in infrastructure ● 74% ● 93% ● 60% ● 60% ● 67% Quality and utilization of human capital ● 73% ● 91% ● 62% ● 59% ● 62% Management of natural capital ● 73% ● 62 ● 72% ● 67% ● 69% 10. Investment in infrastructure Distance between Mexico and best performer reference in each reference group Income Value World Impediment/Indicador Peers OECD Region World Year Group Mex P95 Quality of air transport infraestruture 4.1 ● 77% ● 97% ● 43% ● 42% ● 52% 4 2017 6 Quality of electricty supply ● 45% ● 100% ● 26% ● 28% ● 40% 5 2017 7 Quality of overall infraestructure ● 62% ● 100% ● 37% ● 26% ● 54% 4 2017 6 Quality of port infraestructure ● 46% ● 68% ● 39% ● 42% ● 44% 4 2017 6 Quality of roads ● 41% ● 81% ● 34% ● 26% ● 46% 4 2017 6 Foreing direct invesment, net (% of GDP) ● 52% ● 81% ● 14% ● 10% ● 23% -0.02 2016 0.02 Foreing direct invesment, net inflows (% of GDP) ● 96% ● 100% ● 100% ● 100% ● 100% 0 2016 0 Investment in energy with private particpation (% of GDP) ● 100% ● 100% ● 85% ● 100% ● 100% 0 2016 0.02 Investment in telecoms with private particpation (% of GDP) ● 97% ● 100% ● 85% ● 88% ● 93% 0 2014 0.02 Investment in transport with private particpation (% of GDP) ● 100% ● 100% ● 96% ● 100% ● 96% 0 2016 0.04 Investment in water and sanitation with private particpation ● 100% ● 100% ● 100% ● 100% ● 86% 0 2015 0 (% of GDP) SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 158 11. Quality and utilization of human capital ANNEXES Distance between Mexico and best performer reference in each reference group Income Value World Impediment/Indicator Peers OECD Region World Year Group Mex P95 Share of youth not in education, employment or training total ● 57% ● 93% ● 33% ● 25% ● 46% 19 2017 4 (% of youth population) Women in labor force, ratio to men ● 61% ● 100% ● 50% ● 85% ● 61% 1 2017 1 Pupil-teacher ratio in primary education (headcount basis) ● 100% ● 100% ● 83% ● 91% ● 44% 27 2016 10 Pupil-teacher ratio in secundary education (headcount basis) ● 37% ● 79% ● 45% ● 40% ● 34% 16 2016 8 Quality of management schools ● 53% ● 93% ● 54% ● 51% ● 58% 4 2017 6 Quality of math and science education ● 89% ● 100% ● 87% ● 77% ● 90% 3 2017 6 Quality of primary education ● 83% ● 100% ● 83% ● 71% ● 84% 3 2017 6 Quality of scientific research institutions ● 45% ● 83% ● 23% ● 16% ● 46% 4 2017 6 Quality of the education system ● 84% ● 100% ● 76% ● 64% ● 79% 3 2017 5 Availability of scientists and engineers ● 62% ● 73% ● 40% ● 26% ● 46% 4 2017 5 Educational attainment, at least completed short-cycle ● 93% ● 99% ● 78% ● 71% ● 68% 15 2015 46 tertiary, population 25+, female (%) (cumulative) Educational attainment, at least completed short-cycle ● 80% ● 88% ● 73% ● 50% ● 64% 17 2015 44 tertiary, population 25+, male (%) (cumulative) Educational attainment, at least completed short-cycle ● 86% ● 99% ● 76% ● 66% ● 65% 16 2015 43 tertiary, population 25+, total (%) (cumulative) Educational attainment, at least Massters or equivalent, ● 96% ● 98% ● 87% ● 83% ● 89% 2 2015 16 population 25+, female (%) (cumulative) Educational attainment, at least Massters or equivalent, ● 92% ● 97% ● 83% ● 76% ● 86% 2 2015 16 population 25+, male (%) (cumulative) Educational attainment, at least Massters or equivalent, ● 94% ● 98% ● 85% ● 80% ● 87% 2 2015 15 population 25+, total (%) (cumulative) GDP per person employed (constant 2011 PPP $) ● 65% ● 100% ● 53% ● 40% ● 66% 38M 2017 106210 Labor force with advanced education (% of total workin-age ● 70% ● 93% ● 58% ● 81% ● 48% 71 2017 88 population with advanced education) Unemployment with advanced education, female (% of female ● 34% ● 38% ● 18% ● 36% ● 20% 4 2017 1 labor force with advanced education) SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 159 ANNEXES 12. Management of natural capital Distance between Mexico and best performer reference in each reference group Icome Value World Impediment/Indicator Peers OECD Region World Year Group Mex P95 Methane emissions (kt of CO2 equivalent) ● 79% ● 7% ● 76% ● 0% ● 0% 116705 2012 113564 Nitrous oxide emissions (thousand and metricntons of CO2 ● 83% ● 20% ● 33% ● 18% ● 18% 43436 2012 53101 equivalent) Rtotal greenhouse gas emissions (kt of CO2 equivalkent) ● 80% ● 56% ● 76% ● 0% ● 15% 663425 2012 780551 Adjusted savings: energy depletion (% of GNI) ● 39% ● 0% ● 89% ● 46% ● 85% 1 2016 9 Adjusted savings: mineral depletion (% of GNI) ● 89% ● 67% ● 86% ● 93% ● 92% 1 2016 7 Adjusted savings: natural resources depletion (% of GNI) ● 61% ● 55% ● 88% ● 76% ● 88% 2 2016 17 Adjusted savings: netforest depletion (% of GNI) ● 82% ● 0% ● 82% ● 95% ● 99% 0 2016 11 Alternative and nuclear energy (% of total energy use) ● 71% ● 92% ● 81% ● 86% ● 86% 5 2015 38 CO2 emissions (metric tons per capital) ● 79% ● 100% ● 74 ● 91% ● 77% 4 2014 16 CO2 intesity (kg per kg of oil equivalent energy use) ● 63% ● 77% ● 54% ● 63% ● 69% 3 2014 1 Electricity production from oil, gas and coal sources (% of ● 85% ● 94% ● 81% ● 82% ● 81% 81 2015 1 total) Energy use (kg of oil equivalent per capita) ● 83% ● 100% ● 77% ● 93% ● 82% 1488 2015 7604 Energy use (kg of oil equivalent) per $1,000 GDP (constant ● 67% ● 70% ● 78% ● 82% ● 84% 89 2015 298 2011 PPP) Enviroment Performsmnr Index 2018 ● 36% ● 92% ● 31% ● 39% ● 44% 60 2018 79 GDP per unit of energy use (ppp $ per kg of oil equivalent) ● 60% ● 65% ● 51% ● 51% ● 54% 11 2015 21 Level water stress: freshwater withdrawal as a propotion of ● 44% ● 44% ● 28% ● 50% ● 17% 26 2015 0 available freshwater resources Marine protected a reas (%of territoprial waters) ● 90% ● 95% ● 78% ● 82% ● 93% 2 2016 33 PM2.5 air pollution, population exposed to levels exceeding ● 100% ● 100% ● 100% ● 99% ● 100% 100 2016 0 WHO guideline value (% of total) Water productivity, total (constant 2010 us$ GDP per cubic ● 78% ● 100% ● 90% ● 90% ● 96% 14 2015 288 meter of total freshwater withdrawal) SYSTEMATIC COUNTRY DIAGNOSTIC MEXICO 160