Trajectories: Prosperity and Poverty Reduction in the Colombian Territory CO LO M B I A - P OVE R T Y A N D E Q U I T Y A S S E S S M E N T © 2024, International Bank for Reconstruction and Develop- ment / World Bank 1818 H Street N.W. Washington D.C. 20433, United States of America Telephone: (202) 473-1000 Internet: www.worldbank.org In Spanish: www.bancomundial.org Email: feedback@worldbank.org Rights Reserved This volume is a product of the staff of the International Bank for Reconstruction and Development / The World Bank. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of the World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this publication. Rights and Permissions. The International Bank for Reconstruction and Development / The World Bank encourages the dissemination of its work and will normally grant permission to reproduce portions of this work promptly, provided the sources are acknowledged. Concept and Cover Design Carlos Reyes, Reyes Work Studio Interior Design and Typesetting Kilka Diseño Gráfico Acknowledgments This report was prepared by a core team led by María Eugenia Dávalos (TTL, ECLPV) under the guidance of Carlos Rodriguez Castelan (Practice Manager, ELCPV) and Rafael Muñoz (Pro- gram Leader) and the overall direction of Mark Thomas (Coun- try Director for Colombia) and Peter Siegenthaler (Country Manager for Colombia). Team members who contributed with substantial inputs to the development of chapters include Alfredo Bateman, Ana Arjona, Andres García Suaza, Angela Vega Landaeta, Ellin Ivarsson, Fernando Giuliano, Gisela García, Horacio Alvarez, Ivonne Astrid Moreno, Jeremy Veillard, Juan Manuel Monroy, Juan Sebastian Vallejo, Julieth Parra, Karina Acosta, Marcela Portocarrero, Phoebe Ishak, Sandra Segovia, Sarah Elizabeth Moore, Silvia Alejandra Otero, and Tomas Martin. Desiree Gonzalez and Dahiana Merizalde provided administrative and logistics support throughout the prepara- tion of this work; Maria Clara Ucros and Jairo Bedoya guided the dissemination and communication strategy; and Carlos Reyes and Patricia Carley provided design and editing services, respectively. The team is grateful for the dialogue with the Nation- al Planning Department, the Centro de Estudios Económicos Regionales (CEER) of the Central Bank of Colombia, and the Geographic Institute Agustín Codazzi (IGAC) in the prepara- tion of this work and the comments and suggestions received. The team is also grateful for comments from academics from various regions of the country consulted throughout the pro- cess and for the contributions from the peer reviewers: Trang Van Nguyen, Maria Eugenia Genoni, and Maria Laura Sanchez Puerta, as well as many other colleagues who took the time to provide valuable suggestions, including Leonardo Cañon, Sam- uel Freije, Leonardo Iacovone, Jacobus Joost De Hoop, Daniel Mahler, Hugo Ñopo, and Paula Rossiasco, among others. TABLE OF CONTENTS List of Acronyms...............................................................7 Movement of people is not always in search of higher incomes ................................................................... 51 Executive Summary.........................................................9 Distances: some places remain isolated, and Chapter 1. climate-related hazards are bound to amplify Introduction..................................................................... 18 distances .......................................................................................54 The promise of cities: Cities can offer better Chapter 2. access to services but not always higher A Falling Yet Unequal Poverty Trend .....................22 incomes..........................................................................................56 Poverty has recently declined, propelled by a labor market recovery ............................................... 23 Chapter 5. Not every group has equal opportunities to Charting the Trajectory: a Policy Agenda............. 58 escape poverty........................................................................... 25 Policy Area 1: Investing in increasing access to productive assets for the poor and poor areas.....61 Chapter 3. Policy Area 2: Improving the institutional Increasing Equity in People’s Access to framework......................................................................................61 Assets: a Catalyst for Higher Social Mobility........32 Opportunities to accumulate assets are Annexes ............................................................................ 71 unevenly distributed across groups............................ 33 Annex A. Monetary Poverty Methodology ............ 72 Lower access to productive assets leads to Annex B. Profiles of Poverty in Colombia, fewer economic opportunities........................................41 2005, 2018, and 2023 (the latter Institutions are not fully equipped to not comparable)....................................................................... 73 eliminate gaps in opportunities, including Annex C. Poverty Profiles: Conditional territorial gaps............................................................................43 Regressions, 2005, 2018, and 2023 (the latter As a result of unequal access to assets and not comparable)....................................................................... 75 opportunities, social mobility is low, leading Annex D. Correlates of Municipal Poverty to a vicious cycle of inequities ........................................46 Changes, 2008−18, 2018−21................................................77 Annex E. The State Discontinuity Index: Chapter 4. Methodology...............................................................................79 The Need to Boost the Dynamics Annex F. Additional Resources on that Can Reduce Spatial Gaps ................................. 50 Inequalities between Population Groups.............. 80 References....................................................................... 83 List of figures Figure ES1. Total and Extreme Poverty Rates (%) ....10 Figure 14. Poverty Rate at the Municipal Level Figure ES2. Regional Relative Poverty Rates 2008 vs. 2018 (%)...................................................30 in OECD Countries, regions with Figure 15. Divergence in Poverty Rates between maximum and minimum rates................... 11 2008 and 2018 (municipal poverty in Figure ES3. Correlation between Poverty Rates 2008 vs. changes in 2008−18)......................30 and Percentage of People in Formal Figure 16. Correlation between per Capita Jobs ............................................................................... 12 Income in 2021 and Population Figure ES4. Correlation between per Capita Density in 1500......................................................30 Income in 2021 and Population Figure 17. Unmet Basic Needs Index 1973 vs. Density in 1500....................................................... 13 2018, by department ........................................30 Figure ES5. Divergence in Poverty Rates Figure 18. Percentage of Population Aged between 2008 and 2018 (municipal 18−64 Years of Age with Completed poverty in 2008 vs. changes in 2008−18).13 Secondary Education ......................................34 Figure ES6. Relative Educational Mobility, Figure 19. Learning Poverty vs. Poverty Rate ..........34 Departments of Colombia and Figure 20. People with Access to Preventive Selected OECD Countries, most Health Care (%)...................................................... 35 recent data ...............................................................14 Figure 21. Avoidable Infant Mortality Rate, Figure ES7. Principles and Policy Areas..........................16 Bogotá frontier ..................................................... 35 Figure 1. The Assets-Based Approach.........................19 Figure 22. State Discontinuity Index .............................36 Figure 2. Aging Index and Poverty at the Figure 23. State Discontinuity Index Department Level ..............................................20 Comparable between Countries Figure 3. Total and Extreme Poverty Rates (%) ....24 (1 more discontinuity)................................36 Figure 4. Poverty Rate (US$6.85 per day), circa Figure 24. Results on Standardized Tests (Saber 2022...............................................................................24 11- reading) 2014−19............................................. 37 Figure 5. Decomposition of National Total Figure 25. Average Score (0-6) on Perceptions Poverty Changes by Income Source of Rights and Freedoms ................................ 37 (in percentage points)...................................... 25 Figure 26. Proportion of Tree Cover, 2008..................38 Figure 6. Decomposition of National Figure 27. Annual Growth of Deforestation (%), Poverty Changes between Growth, 2008−18.......................................................................38 Redistribution, and Line (inflation).......... 25 Figure 28. Percentage of People in Households Figure 7. Poverty Rates (%) by Group, 2023 ............26 with their Own Home and Deed Figure 8. Regional Relative Poverty Rates in Registered in the Name of a OECD Countries................................................... 27 Household Member .........................................39 Figure 9. Between-Region Inequality Figure 29. Municipalities with Extremely High Decomposition, as % of the Land Inequality (Gini coefficient>0.8) ..39 maximum possible between-group Figure 30. Percentage of People in Households inequality (ELMO decomposition)...........28 with at Least One Financial Product ... 40 Figure 10. Subnational Poverty Rates Figure 31. Distribution of Bank Correspondents (minimum, maximum, and per Thousand Inhabitants 2023............... 40 difference) for a Sample of LAC Figure 32. Employed Population with Formal Countries...................................................................28 Jobs (%).......................................................................43 Figure 11. Poverty Rate per Department, 2023.......28 Figure 33. Correlation between Poverty Rates Figure 12. Concentration of Poor Population by and Percentage of People in Formal Region and Department (%)........................28 Jobs ..............................................................................43 Figure 13. Income Growth Incidence Curves, by Figure 34. Poverty Rate and Value Added per region and income decile .............................29 Worker, by type of microbusiness ...........43 Figure 35. Value Added per Worker vs. Figure 46. Internal Migration vs. Poverty at Department’s Poverty Rate .........................43 Origin and Destination ................................... 53 Figure 36. Management Component, Municipal Figure 47. Decomposition of Labor Income Performance Index 2021.................................45 Gap Relative to Main Cities Figure 37. 2021 MDM Scores for Municipalities (Bogotá, Medellín, Cali, Manizales, above and below the Poverty Average .45 Bucaramanga), percentage points ........ 53 Figure 38. Labor Income Inequality of Figure 48. Accessibility Categories Based on Opportunity Index, % explained by Travel Time, by municipality (darker each circumstance 2022.................................47 color higher travel distances)............ 55 Figure 39. Absolute vs. Relative Mobility in Figure 49. Travel Time (minutes) vs. Education, 1980 cohort .................................. 48 Multidimensional Poverty Rate, 2018..... 55 Figure 40. Multidimensional Poverty vs. Relative Figure 50. Percentage of the Population with Mobility, 1980 cohort ........................................49 Access to the Internet ..................................... 55 Figure 41. Relative Educational Mobility, Figure 51. Internet Penetration, 2021................................ 55 Departments of Colombia and Figure 52. Accessibility Categories under a Selected OECD Countries, most Severe Flood Event, compared to the recent data ..............................................................49 Figure 48 baseline .............................................56 Figure 42. Number of People Living in Figure 53. Number of Municipalities under Conditions of Internal Displacement 100,000 Inhabitants, by poverty rate (millions, 2022)....................................................... 52 and population size .......................................... 57 Figure 43. Multidimensional Poverty Rate, Figure 54. Number of Municipalities per Households Affected by Armed Accessibility Category and Conflict and Forced Displacement (%).52 Population Size .................................................... 57 Figure 44. Reasons for Migrating in the Past Figure 55. Monetary Poverty Clusters at the Year, 2016−21............................................................ 53 Municipal Level, 2008−18................................59 Figure 45. Characteristics of Internal Migrants, by reason 2016−21................................................ 53 List of boxes List of tables Box 1. Colonial Legacy and Regional Table 1. Policy Examples to Reduce Inequalities .............................................................. 31 Poverty and Promote Equity across Box 2. Land and Poverty: Some Insights Colombia...................................................................67 from the Literature............................................ 40 List of Acronyms CONPES Consejo Nacional de Política Económica y Social (National Council for Economic and Social Policy of Colombia) DANE Departamento Administrativo Nacional de Estadística (National Administrative Department of Statistics) DNP Departamento Nacional de Planeación (National Planning Department) ECP Encuesta de Cultura Política (Political Culture Survey) ECV Encuesta National de Calidad de Vida (National Quality of Life Survey) EMICRON Encuesta de Micronegocios (Microbusinesses Survey) GEIH Gran Encuesta Integrada de Hogares (Large Integrated Household Survey) HDI Human Development Index IO Inequality of Opportunity Index LAC Latin America and the Caribbean MDM Medición de Desempeño Municipal (Municipal Performance Measure) NARP Negros, Afro-descendientes, Raizales y Palenqueros OECD Organisation for Economic Co-operation and Development PPP Purchasing Power Parity PDET Programa de Desarrollo con Enfoque Territorial (Program with a Territorial Focus) PTA Programa Todos a Aprender (Everyone to Learn Program) Trajectories: Prosperity and Poverty Reduction in the Colombian Territory Colombia - Poverty and Equity Assessment Imagine two girls born on the same day in Colombia. One tions, typically cities. Even when people do not migrate within is born with parents with a high level of education and in a the country, large distances and low connectivity to economic prosperous city. The other is born with poor parents with lit- centers can limit access to services and jobs in many regions. tle schooling, and in a municipality far from the capital of her Although economic growth may be an uneven process, ac- department, which is already a department with high poverty. cess to opportunities should not be. Investing in people’s as- The life trajectory of both girls could be very different. There is sets and access to quality services, particularly in their access a high probability that the conditions in which each one was to quality education and health, but also by reducing distances born will impact their chances of obtaining quality education, to opportunities through improved physical and digital connec- a good job and living a life free of poverty. tivity, can help increase their economic opportunities and thus Colombia has made significant strides in reducing poverty, their incomes, and also help people to leverage the benefits of even in the aftermath of the COVID-19 pandemic. However, not migrating to other areas of the country, if they wish to move. every group in society has had equal opportunities to escape Realizing these goals also requires strengthening institutions poverty, and there are pronounced disparities in welfare across at all levels of government—their technical and fiscal capacity the territory. These gaps are largely driven by the unequal ca- and their ability to coordinate and associate. This would allow pacity of the state to provide public services in the various de- achieving economies of scale in service provision, better serve partments and municipalities, resulting in lower opportunities the poor across the entire country and promote local develop- for many to accumulate essential assets, such as a quality ed- ment. Policy design should account for the heterogeneity of ucation, health, land, and housing. Moreover, this lower access needs and capacities across the territory, the importance of co- to assets frequently leads to fewer economic opportunities, ordinating across relevant actors, and the value of delivering including lower quality jobs and very low social mobility, con- an integrated package of services to simultaneously address ditions that are even more pronounced in such departments as various barriers to reducing poverty. La Guajira, Vaupés, and Guainía. The disadvantages present in Closing the gaps in opportunities across population certain parts of the territory often overlap with higher needs for groups and territories is not a short-term endeavor, but there certain population groups, such as indigenous peoples. are policy options within reach, and the country can point to The lower access to opportunities further penalizes many many success stories in this effort. We must invest in leveling individuals and families who are forced to move within the opportunities for all – so that these two girls can count on the country to escape conflict situations or motivated by a lack of tools to prosper – and chart a trajectory to poverty reduction access to quality basic services in their places of origin, and thus and equity. cannot leverage better economic opportunities in their destina- EXECUTIVE SUMMARY It is widely accepted that inequalities derived from circumstances beyond a person’s control, such as ethnicity or place of birth, are unfair. Yet, patterns of poverty across the globe are usually tightly linked to these very attributes, including in Colombia. The opportunities to have a better education, access to quality health care, or improved water sources vary greatly depending on whether you are an Afro- Colombian living in Chocó, an indigenous person living in Vaupés, or a Venezuelan migrant living in Bogotá. Indeed, the socioeconomic profiles of the poor in Colombia persistently reveal that not everyone has the same opportunities to escape poverty and prosper. This report, “Trajectories: Prosperity and Poverty Reduction opportunity for higher economic growth potential. Specifically, in the Colombian Territory,” explores the factors behind the the report makes four main contributions. First, it brings new lower capacity of many to lead productive lives in Colombia, evidence on regional inequalities to the public debate through particularly their lower accumulation of assets. It examines new data and measures that go beyond what has been tradi- how this capacity is persistently different across municipali- tionally produced and that zoom in to a more granular geo- ties and departments, as well as across gender, ethnicity, and graphical level: the municipality. This provides new insights migratory status. The report’s focus on assets puts the lens on into the challenges at hand and their heterogeneity across the poor people’s capacity to generate income and escape poverty. country. The report prepares and presents, for instance, munic- It therefore adopts an assets-based approach, which recognizes ipal data on incomes and poverty and their evolution, allowing that the capacity of an individual or household to generate an assessment of patterns and drivers of progress. Second, its market income depends on the assets they possess—their analytical anchoring in the assets-based approach puts the fo- human, physical, financial, social, and natural capital—and cus on people and their ability to lead productive lives. Third, on how they subsequently use those assets in markets to gen- through this focus on assets, it provides a multisectoral view of erate income. the challenges to addressing territorial inequalities in access to Addressing territorial inequalities has been recognized opportunities and accordingly offers a multidimensional view as a policy priority in Colombia, and this report brings new of the policy discussion it proposes. Finally, the report suggests evidence to the public debate. With nearly two out of five policy principles and examples to inform the broader debate on future workers living in poorer areas, a failure to invest in the reducing poverty and inequality across Colombia. productive capacity of these younger generations is a missed CO LO M B I A 9 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T A Falling Yet Unequal Poverty Trend Colombia’s stable economic growth over the past two de- contraction of the middle class, and a recovery that is still un- cades has been accompanied by a reduction in poverty and derway. The pandemic caused a severe economic contraction inequality. There was a significant reduction in poverty from of 7.2 percent in 2020, with a significant loss of jobs, especially 49.7 percent in 2002 to 28.2 percent in 2019 (Figure ES1), with in informal sectors. Government transfers helped mitigate the a notable expansion of the middle class and a decrease in in- impact on poverty, but labor market weaknesses, linked to low come inequality. Most of the poor are in urban areas: out of the productivity and high informality, were intensified. Since 2021, 16.8 million poor, 11.9 million live in urban areas, and seven de- poverty has resumed a downward trend and the middle class partments account for over half of the country’s impoverished has grown slightly, though inflation, particularly in food prices, population. has posed a challenge to poverty reduction efforts. The COVID-19 pandemic reversed over a decade of prog- ress, with a sharp increase in poverty and extreme poverty, a Figure ES1. Total and Extreme Poverty Rates (%) Total and extreme poverty as well as inequality have declined. Poverty and extreme poverty rates 50 0.75 45 0.7 40 0.65 Gini coefficient 35 30 0.6 25 0.55 20 0.5 15 0.45 10 5 0.4 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 Poverty rate Extreme poverty Gini Source: Data from the National Administrative Department of Statistics (Departamento Administrativo Nacional de Estadística [DANE]). Note: Poverty and extreme poverty lines were updated in 2019 and there is a methodological interruption from 2021 onwards. Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 10 Increasing Equity in People’s Access to Assets: a Catalyst for Higher Social Mobility However, not every group has the same chance to escape pov- ties have lower access to a quality education, health care, water, erty. Poverty rates are significantly different across groups and and financial and other assets. For example, learning poverty territories. For instance, Colombia stands out due to its regional (the percentage of children unable to read and comprehend a gaps in welfare (Figure ES2) when compared to OECD countries. simple text by the age of 10) is as high as 100 percent in certain Moreover, indigenous populations have poverty rates almost municipalities, and in some areas, 90 percent of infant deaths double those of non-ethnic groups, rates that have shown a are deemed avoidable but for the proper care. Similarly, higher much slower reduction: since 2021, the reduction was 1.9 per- avoidable mortality rates (mortality from preventable or treat- centage points for indigenous populations compared to 7.5 for able causes) are observed in the Pacific, Amazonia, La Guajira, non-ethnic groups. and Amazonia-Orinoquia regions, with some municipalities These gaps in poverty rates are underpinned by vastly experiencing between 53 and 90 percent of avoidable deaths. different access to and quality of assets across groups and The persistence of territorial inequalities is partly a territories. Granular municipal-level data on poverty, learn- consequence of systemic disparities in the state provision of ing, avoidable mortality, road connectivity, and state presence, public services and goods. A new measure of “state discontinu- among other measures, show that people in poorer municipali- ity” illustrates the variation in the provision of services across Figure ES2. Regional Relative Poverty Rates in OECD Countries, regions with maximum and minimum rates Colombia and Mexico have the largest regional poverty gaps among OECD countries. COL MEX ISR USA CHL ITA LTU ESP BEL PRT DEU CAN AUS CHE POL IRL AUT SVK GRC GBR SWE FRA HUN CZE FIN DNK 0 10 20 30 40 50 60 70 Poverty rate Minimum region Maximum region Source: OECD (2022). Note: Relative poverty rates are measured as the share of people with an income below the relative poverty line (60 percent of the national median income). CO LO M B I A 11 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T regions. Some areas are service deserts, and the poorest de- the evidence shows an association between the past presence partments historically and to date, such as Vichada, Guainía, of illegal armed groups and lower human capital, including Chocó, La Guajira, and Amazonas, have higher heterogeneity in lower performance in reading and mathematics in standard- state presence within their territories. The heterogeneous pres- ized test scores at the end of high school. Municipalities with ence of the state across the territory has been accompanied by a past presence of illegal armed groups also perceive a lower a long-persistent armed conflict that has helped to exacerbate enjoyment of rights and freedoms across several dimensions, the challenge. As a result, economic and educational opportu- including voting, political engagement, and freedom of move- nities are lower in the localities affected by conflict and require ment, when compared to regions untouched by conflict. additional policy efforts to level the playing field. Specifically, Figure ES3. Correlation between Poverty Rates and Percentage of People in Formal Jobs The quality of jobs is lower in poorer places. 80 70 Poverty rate (%) 60 50 40 30 20 10 0 0 20 40 60 80 Formal employment rate (%) Source: Estimates using the Great Integrated Household Survey (Gran Encuesta Integrada de Hogares [GEIH]) 2023. This limited access to assets translates into lower eco- Territorial gaps have been quite persistent. It is partic- nomic opportunities. Access to quality jobs is unequal, and ularly troubling that the more prosperous regions in Colom- millions of families rely on income from small and generally bia as far back as the 1500s are still the ones that are better off low-productivity businesses. Departments with higher poverty today (Figure ES4). The 1973 Census captured deprivations in rates tend to have a lower proportion of formal employment, basic needs that today, 50 years later, remain concentrated in below 20 percent for such departments as Nariño, Sucre, La the same areas. During the period of poverty reduction in the Guajira, and Cauca (Figure ES3). Many people work in micro- past decade and a half, gaps in municipal poverty have even business, but not all of these microbusinesses arise from a busi- widened, as indicated by various poverty measures, such as ness opportunity in the market; instead, roughly half emerge the poverty headcount ratio, extreme poverty, multidimen- because the entrepreneurs have no other income alternative. sional poverty, and the unsatisfied basic needs index. That is, This latter group are also more likely to be from poor house- municipalities that had higher initial poverty levels generally holds (46 compared to 30 percent for opportunity enterprises), experienced a slower rate of decrease compared to their coun- and their enterprises have lower productivity. Opportunities are terparts—in other words, a divergence in poverty rates (Figure limited even among people with similar levels of education and ES5). At the same time, there is evidence of economic output other endowments, as the use and returns to those assets are and income convergence over the same period. not the same across the country. For example, only 27 percent of working-age women in Chocó participate in the labor market. Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 12 Figure ES4. Correlation between per Capita Income Figure ES5. Divergence in Poverty Rates between in 2021 and Population Density in 1500 2008 and 2018 (municipal poverty in 2008 vs. Places that had higher economic activity in 1500 changes in 2008−18) still have higher incomes today. Poorer municipalities reduced poverty at a slower rate than better-off ones. 14 Bogota .1 13.8 AntioquiaValle del Per capita income 2021 (log) Monetary poverty growth 13.6 Cauca Risaralda .05 (annual), 2008-2018 Caldas 13.4 Atlantico Santander Cundinamarca Quindio Meta San Andres y Tolima 13.2 Providencia 0 Bolivar Norte de 13 Huila Casanare Magdalena Caqueta Cesar Cauca Santander Choco Cordoba 12.8 Arauca Sucre -.05 Guaviare 12.6 Amazonas 12.4 Guainia La Guajira -.1 Vichada 12.2 Vaupes 12 -.15 0 5 10 15 6.5 7 7.5 8 8.5 9 Precolonial population density (1500) Municipality Linear prediction Monetary poverty 2008, log-scale Source: Updated per capita income for 2021, sourced from Maloney Source: Convergence background analysis for this report by Acosta, and Valencia-Caicedo (2016). Davalos, and Segovia (forthcoming). Yet, some areas have made significant progress in reduc- services and assets (such as access to infrastructure services and ing poverty. The evidence shows that faster poverty reduction better health and educational outcomes) and have improved has taken place in municipalities that have increased access to their fiscal performance. CO LO M B I A 13 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Figure ES6. Relative Educational Mobility, Departments of Colombia and Selected OECD Countries, most recent data Some departments in Colombia fare even worse in relative mobility. 0.80 0.60 0.40 0.20 0.00 Narino Huila Guania Cauca La Guajira Vaupes Caquetá Vichada Caldas Chocó Putumayo Tolima Boyacá Córdoba Colombia Santander Bolivar Guaviare Casanare Amazonas Antioquia Sucre N.Santander Meta Magdalena Risaralda Cundinamarca Cesar Quindio Bogotá Valle del Cauca Atlántico Arauca OECD average San Andres Source: Dávalos and Monroy (forthcoming), using the Global Database of Mobility and the National Quality of Life Survey (Encuesta Nacional de Calidad de Vida [ECV]). Resolving gaps in the productive capacity of the poor As a result of the persistently lower productive capacity and poorer regions calls for solid institutional capacity at of certain groups and regions, social mobility is notably low all levels of government. Yet currently, institutions are not in Colombia compared to global averages, with poorer areas fully equipped for this task. Institutions in poorer municipal- showing more limitations. New estimates reveal that at least ities are weaker, affecting both their management and fiscal 30 percent of labor income and 24 percent of household in- capacities. With subnational governments in some cases op- come inequality of the mean log deviation inequality measure erating in silos and without formal mechanisms to establish (and more than 44% of the Gini income inequality) is deter- continuous dialogue with the central government, there are mined by circumstances at birth, such as parents’ socioeco- missed opportunities to leverage economies of scale to deliver nomic background, sex, ethnicity, and place of birth. Along the services, particularly for people in smaller municipalities, or to same lines, educational mobility is also limited, that is, there is capitalize on the spillover benefits they can derive from cities. a tight connection between an individual’s education position Existing policies in various dimensions inadvertently amplify in society and that of his or her parents. Although mobility is these territorial gaps. For instance, school and teacher quality globally low in Colombia, it is even lower in some areas of the are vastly different in poorer areas than in richer ones, yet there country (Figure ES6). The chances of attaining education higher is no adaptive curriculum or policy on textbooks and school re- than one’s parents vary across regions; a child with low-educat- sources for subnational areas. ed parents living in Guainía or Vaupés has a 24 percent proba- bility of completing secondary education, whereas for a child living in Bogotá, it is 65 percent. Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 14 The Need to Boost the Dynamics that Can Reduce Spatial Gaps The forces that might be expected to close gaps across the ter- likely to amplify travel times (distances) in places with already ritory, such as internal migration, the pull of larger cities, and low accessibility. Digital technologies can help increase con- the shortening of distances, cannot be relied on to solve the nectivity, improve access to information and services, reduce problem. Internal migration in Colombia, for example, includes trade costs, and open economic opportunities. However, there millions of forcibly displaced people who have relocated to es- are large spatial disparities in access to these technologies and cape conflict rather than to search for better economic oppor- missed opportunities; in fact, Colombia has one of the highest tunities. Among those who move in search of better economic shares of urban workers with jobs suitable for telecommuting opportunities, most are highly educated men. The promise of but without internet access at home. cities or agglomerations to provide income opportunities for Given these dynamics, there is a strong case for invest- larger populations often fails to materialize due to limited con- ing in people’s assets and for reducing distances to new op- nectivity and the high cost of living. Furthermore, poorer mu- portunities through better physical and digital connectivity. nicipalities face longer travel distances and thus higher travel This will enable more Colombians to move and improve their costs to economic centers: 21 percent of municipalities, hosting returns and thus their incomes and will also help to ease con- around 4 million people, have low accessibility or are isolated. gestion, promote commuting, and equalize returns across the Without additional investments in resilient infrastructure, territory. climate-related disasters, such as floods and landslides, are Charting the Trajectory: a Policy Agenda How can new trajectories be charted to reduce poverty and delivery of services. At the same time, advancing differen- inequality of opportunities? It is to be expected that economic tiated policies (between territories or population groups) growth is an unbalanced process, and each region will continue should not entail atomizing policies to such an extent that to have different productive strengths. Yet, efforts should be they become unfeasible to monitor and implement. made to ensure that opportunities for escaping poverty and ■ Coordinated policies. Institutional capacity, as well as vertical leading a more productive life are not unbalanced as well. and horizontal coordination across levels of government How to think spatially about development challenges? It and other societal institutions, can maximize policy im- calls for incorporating three interrelated principles into policy pact. Effective coordination can merge local governments’ design: on-the-ground insights with the overarching goals and perspectives of the regional and national governments, ■ Differentiated policies. A differentiated approach recogniz- contributing also to leveraging economies of scale and es the uniqueness of regional and group needs and chal- resources. lenges, signaling that policies should not be uniform. The ■ Integrated policies. It is more impactful to deliver packages concept of “development clubs” is increasingly being im- of policies and interventions that consider the multiple plemented in other countries. Municipalities are grouped dimensions of poverty and development challenges (e.g., in a way that considers common characteristics or paths to better education, connectivity, health services, and access enable them to capture the spillover effects of cities and to water and sanitation) than to rely on a single develop- to capitalize on the potential of economies of scale in the ment axis. CO LO M B I A 15 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Figure ES7. Principles and Policy Areas { Di erentiated policies Investing in Access to Productive Assets Coordinated policies Improving the Institutional Framework Integrated policies Source: Authors’ elaboration. There are policy options within reach to accelerate the access to specialized care for patients with chronic con- trajectory toward reducing poverty and inequality in Colom- ditions in underserved territories. bia. This report proposes policy examples in two areas that □ In education, introducing a national science-based in their design and implementation should incorporate the textbook and educational resource policy for foun- above-mentioned principles of differentiation, coordination, dational learning areas, including a professional and integration (Figure ES7). development program for teachers, can contribute to equalizing learning. It could also help to increase Policy Area 1: Investing in Access to Productive Assets the supply of preschool (with positive implica- tions for closing gender gaps from increased access ■ Boosting human capital is essential, as it is arguably the to childcare) and upper secondary education in most important asset that poor people can count on to rural areas. lead a productive life. Yet, some particularly poor and re- mote areas of Colombia have low or no access to health There are successful interventions in Colombia in pro- services, and there are significant gaps in the availabili- moting human capital accumulation. For example, the ty of health personnel, medicines, and financing. School All to Learn Program (Programa Todos a Aprender) has and teacher quality also vary widely across the territory, had positive outcomes in terms of student learning, translating into large gaps in learning, with transfers that school attendance, increased within-grade promo- do not compensate for these needs. Policy options to close tion, and overall educational quality. In health, the these gaps include: Estrategia Hospital Padrino exemplifies how to offer tailored support to areas of greatest need in health ser- □ In health, it is important to carry out reforms to scale vice provision, with impactful results on, among other up the implementation of family medicine and multi- measures, maternal mortality. disciplinary primary health care models, prioritizing rural, remote, and peri-urban areas with low access to ■ Promoting access to land requires advancing existing quality primary health care, and to introduce incen- multipurpose cadaster initiatives to record the rights, re- tives for health care professionals to work in rural and strictions, and responsibilities of land and its owners. An remote areas to better meet the needs of underserved additional benefit is that it would allow the introduction populations. The nexus with digital technologies is of a property tax and strengthen local fiscal performance, critical, as investing in telemedicine can help improve which are key to reducing inequality in the territories. Municipalities like Galapa saw an increase of 153 percent Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 16 in property taxes through capacity strengthening and the level, particularly where multi-sectoral and integrated updating of cadastral records. For housing, one of the key interventions are required. This institutional setup would sources of intra-urban vulnerability, the targeting of na- lead the agenda and dialogue between the subnational tional housing policies and programs should be improved and central governments. The absence of this kind of for- to facilitate home ownership among households with the mal coordination mechanism in Colombia may have lim- lowest income levels. ited the effectiveness of past efforts to work across levels ■ Promoting the sustainable use of natural capital to re- of government and across sectors. duce poverty and aligning this goal with the climate ■ Improving the institutional framework also calls for agenda require the scaling up of incentive programs. strengthening tools to think spatially in policy mak- These can include, for example, payments for environmen- ing. This involves establishing typologies to differentiate tal services, production-conservation-restoration agree- national policies and regulations (so-called “develop- ments, and natural resource management models, such ment clubs”) instead of the predominance of uniform as community forestry and naturebased tourism, in both approaches. Thinking spatially requires looking beyond coastal-marine and inland territories.1 the municipal tier to achieve economies of scale and to ■ Enhancing connectivity to promote local development enhance relationships at the subnational level that can and access to services and to improve returns on people’s lead to economic opportunities and better access to ser- assets calls for expanding connections from residential vices for lagging places. areas in peri-urban spaces and smaller municipalities to ■ Strengthening the fiscal and management capacity of the tertiary and secondary road network to improve over- subnational governments, particularly among weaker all access to jobs, housing, markets, and services, which entities, needs to be a focus. Lessons learned from past tend to be concentrated in larger urban areas. efforts suggest that the emphasis needs to be on policy ■ More accurately targeting excluded and disadvantaged sustainability, that is, on considering a permanent infra- groups requires, among other policies, continued efforts structure of shared capacities and capacity building at the to monitor progress and challenges (a differentiated ap- subnational level to service a variety of entities. Strength- proach to statistics) and the design of policies and pro- ening capacities includes developing support and train- grams that address the specific barriers to opportunities. ing tools for territorial entities with differential criteria in terms of competencies, capacities, and resources; creating These and other measures aim to empower the poor, incentives to take advantage of associative schemes be- stimulate economic growth,2 and capitalize on the benefits of tween municipalities; and improving management and migration and agglomerations by improving opportunities in revenue collection abilities. origin and destination areas. Closing the gaps in opportunities across groups will not be Policy Area 2: Improving the Institutional Framework easy, as centuries of unequal development cannot be quickly reversed. Yet, it is possible, and the country can point to many ■ Closing territorial gaps involves setting up an institution- success stories. This report is presented as a contribution to this al arrangement for coordinating policy priorities across policy dialogue. the interventions of line ministries at the subnational 1 Consistent with World Bank (2023c). 2 Consistent with recommendations in World Bank (forthcoming). CO LO M B I A 17 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Chapter 1. Introduction Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 18 It is all too well known that Colombia is one of the most unequal countries in the world.3 Its 2022 Gini coefficient of 0.548 is the highest among 15 countries in the Latin America region,4 and is also high by OECD and world standards.5 Moreover, four out of five Colombians believe that incomes are distributed unfairly in the country.6 Today, there is an increased focus on understanding how opportunities in life are different for some population groups than for others. For example, people in Chocó face monetary poverty at rates nearly three times higher than people in Bogotá or Cundinamarca. Poverty rates are also 1.9 and 1.4 times higher for indigenous peoples and Afro-Colombians, respectively, than for the average Colombian. Such “inter-group” inequality, often referred to as “horizontal inequality,” stems not from individual effort but from entrenched disparities in opportunity, shaping a person’s potential to escape poverty. This perspective on poverty and inequality has gained more traction within policy discussions, signaling a paradigm shift toward addressing systemic inequities in opportunities. Figure 1. The Assets-Based Approach emphasizes territorial inequalities in line with the government priority of regional convergence and with efforts to advance the decentralization and territorial development agendas in the Financial past decades.7 Yet, to shed light on the pervasiveness of these capital “horizontal inequalities,” it goes beyond the territorial aspect to also explore inequalities between population groups, focusing ASSET ACCUMULATION X on women, indigenous people, Afro-Colombians, Venezuelan Human Social capital INTENSITY OF USE capital migrants, and forcibly displaced populations. X RETURNS The report adopts an assets-based approach to explore = the drivers of inequalities between groups and territories.8 INCOME GENERATION CAPACITY This approach is centered on the elements that determine the + taxes, transfers, and capacity of an individual or household to generate market in- subsidies come, considering the assets they possess: their human, phys- Physical Natural capital capital ical, financial, social, and natural capital (Figure 1). The extent to which individuals can accumulate those assets and use them in markets to generate income determines their productive ca- pacity. This Poverty and Equity Assessment aims to explore the Why is it important to address and close gaps in opportuni- trends and factors driving poverty and inequality in Colom- ties between groups and territories? First, this endeavor is a de- bia, particularly how opportunities are different for people velopment goal aligned with the pursuit of social justice. Second, living in certain regions and belonging to certain population inequalities in opportunities reflect a suboptimal allocation and groups. With new data, analysis, and measures, the report utilization of resources that can undermine economic growth. 3 Among many other documents, World Bank (2021) discusses income inequality. 4 Countries with recent available data in the World Bank Latin America and the Caribbean (LAC) Equity Lab. 5 The official national income Gini coefficient was 0.546 in 2023. 6 Latinobarometer 2023, https://www.latinobarometro.org/lat.jsp. 7 This includes such efforts as the Misión del Sistema de Ciudades, the Misión para la Transformación del Campo, the Misión de Crecimiento Verde and the Misión de Decentralización, and important advances, including the Planes de Ordenamiento Territorial (POTs) modernos (territorial development plans), support for a multipurpose cadaster, and instruments for designing development programs with a territorial focus (programas de desarrollo con enfoque territorial [PDETs]). 8 Attanasio and Székely (1999); Carter and Barrett (2006); and Bussolo and Lopez-Calva (2014). CO LO M B I A 19 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T In a country that will reach the end of its demographic dividend poverty levels—a failure to invest in the productive capacity of in 2050, and with a large share of future workers living in poorer these younger generations is a missed opportunity for higher eco- areas (Figure 2)—38 percent of Colombia’s future working-age nomic growth potential. Moreover, unequal opportunities within population currently live in a department with above-average the country can lead to social tension. Figure 2. Aging Index and Poverty at the Department Level The younger population, the labor force of the future, is concentrated in the poorer departments. 120 QUI 100 CAL RIS Ageing index, 2021 80 BOG TOL BOY ANT SANT VAL SAP NAR 60 CUN CAU ATL COR MET 40 MAG CES GUAV ARA CHO 20 AMZ LGJ VIC GUAI VAU 0 0.2 0.4 0.6 0.8 1 Poverty rate 2021 Source: UNFPA population projections and aging index, and poverty data from the 2021 Great Integrated Household Survey (Gran Encuesta Integrada de Hogares [GEIH]) and poverty map. Note: The aging index is calculated as the percentage of the population aged 60 years and over compared to the population aged 14 years or younger. Addressing territorial inequalities has been recognized regional convergence as one of its five proposed transforma- as a policy priority in Colombia. Changes to the Colombian tions. Important advances have been made in recognizing and Constitution in 1991 included an emphasis on fiscal and admin- trying to address gaps over the years.10 istrative decentralization aimed at reducing regional dispari- This report brings new evidence on regional inequalities ties. Several programs have been implemented over the years, in Colombia to the public debate. First, it puts forward new including the most recent Decentralization Mission, aimed data that allow a more in-depth assessment of the needs of at proposing evidence-based initiatives to reorient the de- certain regions and groups. This includes, for example, granular centralization model. In addition, the National Development municipal-level income and poverty estimates11 and measures Plan 2018–2022 included a focus on decentralization,9 and the of road accessibility and avoidable mortality rates. Second, it most recent National Development Plan 2022−2026 included employs innovative methods, such as the state discontinuity 9 A chapter on Pacto por la descentralización: conectar territorios, gobiernos y poblaciones, with the clear vision of “empowering regions and connecting rural areas to foster equitable development”. 10 For example, as described in World Bank (2021), “”these efforts to adopt a territorial approach have led to important advances, such as putting forward the planes de ordenamiento territorial (POTs) modernos (territorial development plans), support for a multipurpose cadaster, instruments for designing programas de desarrollo con enfoque territorial (PDETs), or development programs with a territorial focus, and a more recent effort to move the decentralization agenda forward.” 11 Acosta, Davalos, and Segovia (2024). Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 20 index, which assesses differences in access to services across pose policy principles and policy options to guide the broader the territory. Third, it carries out new analysis, such as the discussion on reducing poverty and inequality in Colombia. impact of climate-related disasters on accessibility. Fourth, it The report complements other analyses exploring aspects of explores inter-group inequalities from the perspectives of an poverty and inequality in the country, including World Bank ability to accumulate assets and social mobility, which puts the (2021), which details the distributional impact of the fiscal sys- focus on differences in the productive capacity of the poor and tem and climate change, among other factors, as well as World the persistence of those differences. Finally, it draws from the Bank (2019), which carried out a gender diagnostic. global and local literature on territorial development to pro- CO LO M B I A 21 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Chapter 2. A Falling Yet Unequal Poverty Trend Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 22 Poverty has recently declined, propelled by a labor market recovery Over the past two decades, Colombia has experienced stable decreased from 0.575 in 2001 to 0.513 in 2019 (similar to the but moderate economic growth. The economy grew at an av- Latin American average decline from 0.565 to 0.510).14 erage of 3.7 percent per year between 2000 and 2023, a rate In 2020, as the COVID-19 shock hit, 3.6 million people be- that was relatively mild compared to the country’s aspirational came poor in Colombia. Among this group, 2.8 million fell into peers, such as Chile, Poland, or Malaysia.12 As a result, by 2023, extreme poverty. During that year, the economy contracted by Colombia’s GDP per capita had declined to 40−60 percent of 7.2 percent, official total poverty increased by 6.8 percentage the GDP per capita of the three peer countries from 70−80 per- points, and extreme poverty rose by 5.5 percentage points, wip- cent in 1990. Despite this, Colombia’s growth rates have shown ing out over a decade of progress in lifting people out of pover- remarkable stability since 2000; in fact, it was the only coun- ty. The middle class (as measured nationally) was also hit hard, try in Latin America that avoided a recession, except for the decreasing from 30.1 to 25.4 percent of the population, while COVID-19−induced recession in 2020. The accumulation of fac- the vulnerable population decreased from 32 to 30.4 percent. tors of production (physical capital and labor), rather than total The official Gini coefficient, which had already seen an increas- factor productivity, has guided Colombia’s economic growth, ing trend in the two years before the crisis, increased further, and recent growth has been characterized by the dominance going from 0.526 to 0.544. Although government emergency of the non-tradable sectors.13 transfers mitigated roughly a quarter of the negative impact This long trend of economic growth was accompanied of the pandemic on poverty (see the decomposition of pov- by sustained poverty reduction in Colombia. Poverty declined erty changes in Figure 5), existing labor market weaknesses, from 67 percent in 2001 to 34.8 percent in 2019 (at US$6.85 per including low productivity and high informality jobs, were ex- day/2017 purchasing power parity [PPP]), showing a relatively acerbated by the crisis.15 Economic activity and jobs contracted high reduction compared to the Latin American average. This the most in sectors with a large proportion of informal labor reduction led to an expansion of the non-poor yet vulnerable (such as commerce, manufacturing, and hospitality), and 2.5 populations (those with incomes between US$6.85 and US$14 million jobs were lost over the year. per day/2017 PPP), especially the middle class (US$14−$81 per After the pandemic-induced peak, monetary poverty in day/2017 PPP), which almost tripled in number, rising from 11.8 Colombia began to rebound, yet 16 million people still live in percent in 2001 to 31.2 percent in 2019 (yet still small in Colom- poverty in Colombia.16 The post-pandemic economic recovery bia compared to most Latin American countries). During this led to a decline in total poverty incidence from 39.7 percent in period, income inequality measured by the Gini coefficient also 2021 (under the new poverty methodology)17 to 33 percent in 2023, or 2.9 fewer million people living in monetary poverty 12 As aspirational peer countries, World Bank (forthcoming) uses those with similar socioeconomic and structural conditions to Colombia in 1990 that have demonstrated better economic growth. 13 World Bank (forthcoming). Such sectors as recreation and culture, financial activities, and communication had average annual growth rates of between 5 and 7 percent between 2006 and 2023, while agriculture and manufacturing grew 2.3−2.4 percent per year. Consequently, services (other than the public sector) and commerce contributed the most to economic growth between 2006 and 2015. 14 The middle-class averages 38.2 percent of the total population in Latin America and the Caribbean (LAC), ranging from 21.4 percent in Honduras to 69.4 percent in Uruguay. Data are from the World Bank Equity Lab. 15 World Bank (forthcoming). 16 According to official estimates. 17 A new poverty series was launched by the National Administrative Department of Statistics (Departamento Administrativo Nacional de Estadística [DANE]) in 2022 with an updated sampling frame from the 2018 Census, and thus far allowing comparisons for 2021 and 2023. Methodological changes included new sample stratification among municipalities and changes to sample size and distribution, among others. Estimates are expected to be revised backwards by DANE to allow for a longer time series. CO LO M B I A 23 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T (Figure 3). The most significant improvements were observed panded: 32.4 percent of Colombians were considered middle in urban areas.18 In tandem, extreme poverty decreased from class in 2023, compared to 29.9 percent in 2022 (national mid- 13.7 to 11.4 percent over the same period, although at a slower dle-class measure). Yet only around 11 percent of those in rural pace, and remained unchanged between 2021 and 2022. Multi- areas are part of the middle class, and over a half of the middle dimensional poverty also fell, driven partly by an improvement class resides in Bogotá (25.4 percent), Antioquia (17.4), and in school attendance.19 At the same time, the middle class ex- Valle (10.5).20 Figure 3. Total and Extreme Poverty Rates (%) Figure 4. Poverty Rate (US$6.85 per day), circa 2022 Total and extreme poverty, and inequality have Poverty in Colombia is high compared to other declined. countries in the region. 50 0,65 56,0 60,0 52,4 45 Poverty rate ($6.85) 50,0 36,1 0,6 34,8 40 40,0 32,2 29,9 27,5 23,5 30,0 21,8 21,5 35 19,9 Gini coefficient Poverty rate 16,9 13,9 20,0 14,1 0,55 10,9 30 6,4 4,7 10,0 25 0,0 0,5 Guatemala Honduras Nicaragua Colombia Peru Ecuador El Salvador Brazil Mexico Dominican Republic Paraguay Bolivia Costa Rica Panama Argentina Uruguay Chile 20 15 0,45 10 5 0,4 2020 2022 2023 2014 2018 2016 2019 2017 2021 2012 2013 2015 Poverty 2022 LAC-18 Poverty Extreme poverty Gini Source: Data from the National Administrative Department of Source: World Bank harmonized data. Statistics (Departamento Administrativo Nacional de Estadística [DANE]). Note: There is a methodological interruption from 2021 onwards. These recent improvements in monetary poverty were age points.21 Counter to the labor market recovery, high infla- driven by the post-pandemic labor market recovery but were tion undermined the reduction in poverty, as shown in Figure hindered by a simultaneous rise in the cost of living. The em- 6. In 2022, inflation reached a peak—particularly food inflation ployment rate reached its pre-pandemic levels by 2023, driving (27.8 percent in 2022)—explained by a disruption in global val- recent declines in poverty (Figure 5). However, the recovery did ue chains, increases in energy prices, and loose fiscal and mon- not narrow pre-existing gender gaps: gender employment and etary conditions. This inflation undermined poverty reduction labor force participation gaps persisted at roughly 24 percent- by 9 percentage points between 2021 and 2022. 18 Poverty in urban areas decreased by 17 percent between 2021 and 2023, from 37 to 30.6 percent, while in rural areas it decreased by 15 percent, from 48.5 to 41.2 percent. 19 Multidimensional poverty decreased from 16 to 12.1 percent during the same period, with greater improvement in urban areas (declining from 11.5 to 8.3 percent) than in rural areas (31.1 to 25.1 percent). The main dimensions driving this improvement were school attendance (deprivation reduced by 58 percent, from 5.5 to 2.3 percent) and health insurance coverage (deprivation reduced by 34 percent, from 10.1 to 6.7 percent). 20 Data from DANE. 21 In 2023, the employment rate for males was 70.4 percent, reaching its pre-pandemic level of 70.7 percent (2019); for females, the rates were 45.9 percent in 2023 and 45.7 percent in 2029. Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 24 Figure 5. Decomposition of National Total Poverty Figure 6. Decomposition of National Poverty Changes by Income Source (in percentage points) Changes between Growth, Redistribution, and Line The labor market was the main driver of the (inflation) reduction in poverty post-pandemic. The rise in the cost of living (affecting the poverty line) undermined the recovery. Contribution to poverty changes 8 15 Change per component 6 10 4 5 2 0 0 -5 -2 -4 -10 -6 -15 -8 2012-2019 2019-2021 2021-2023 -20 2012-2019 2019-2021 2021-2023 Labor income Share occupied Transfers Others Growth Redistribution Poverty line Source: Data from GEIH. Source: Data from GEIH. Not every group has equal opportunities to escape poverty Escaping poverty is not within everyone’s grasp. Numerous being poor increases for certain population groups, including publications have pointed to Colombia’s very high income women, indigenous populations, Afro-descendants, and inter- inequality.22 This inequality is also reflected in differences be- national migrants. tween groups. Poverty rates are persistently higher among in- Poverty disparities across territories are also high. Al- digenous populations, households with children, international though economic growth is expected to be an unbalanced migrants (mostly the Venezuelan population), and victims of process, regional inequalities in economic output in Colombia displacement (Figure 7 and Annex B). Indigenous populations are more than two times higher than in OECD countries (World have poverty rates almost double those of non-ethnic groups Bank 2021).23 These disparities are also observed on the welfare and have shown a much slower reduction: since 2021, the re- side. Compared to other OECD countries, Colombia and Mexico duction was 1.9 percentage points for indigenous populations stand out for having the widest poverty gaps between regions compared to 7.5 for non-ethnic groups. International (mostly (Figure 8). There is more than a 50-percentage-point difference Venezuelan) migrants, however, have faced a larger reduction between the best and the worst performing departments in in the past three years (a decline of 10 percentage points, from terms of poverty compared to an average gap of 18 percentage 58 to 48 percent), although they still face higher poverty than points in OECD countries. Within the Latin America and the the national average. Annex C shows that the probability of Caribbean (LAC) region, Colombia has reached 17 percent of 22 Including World Bank (2021). 23 Bogotá, the richest area in the country, exhibits a GDP per capita comparable to high-income countries like Chile or Uruguay, while regions such as Vichada are more comparable to the lower quartile of low-income countries. Bogotá alone contributes to 25 percent of Colombia’s GDP and, together with Antioquia, Valle, Santander and Cundinamarca, accounts for 60 percent of the total GDP—while representing 48 percent of the total population. The sectoral composition of the economy also varies among regions. Such activities as commerce, finance, construction, and manufacturing are more predominant in richer departments, whereas the economy of poorer regions relies more on the agriculture and public administration sectors. Other departments, such as Meta and Casanare, have a predominance of oil-related activities and a small proportion of the population (World Bank, forthcoming). CO LO M B I A 25 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T the hypothetical maximum segregation possible between de- (Figure 11), evident also in other development measures. For in- partments, surpassed only by Panama (Figures 9 and 10).24 This stance, while Bogotá has a Human Development Index (HDI)25 reflects significant gaps in living conditions across the country of 0.786, Guainía’s has deteriorated, reaching a low of 0.538.26 Figure 7. Poverty Rates (%) by Group, 2023 Poverty is significantly higher for some population groups, such as indigenous people, households with children, and migrants. 80 71 70 60 60 55 Poverty rate 50 48 43 45 41 40 33 34 32 32 31 31 30 28 28 20 15 10 0 Female Male Rural Urban Indigenous NARP None None 1 2 3+ children No migrant* Internal* International* Victim of discplacement** Total Sex Área Ethnicity Number of children Migrants Source: Data from GEIH; data for victims of forced displacement are from the National Quality of Life Survey (Encuesta National de Calidad de Vida [ECV]) 2022. Notes: The Afro-descendant group NARP includes Black, Afro-Colombian, Raizal, and Palenquero populations (Negra, Afro-colombiana, Raizal y Palenquera). The ethnicity group “none” includes those who do not identify as indigenous or Afro descendants. *Migrants correspond to people who moved in the last five years. ***Victims of displacement refers to people in households with victims of forced displacement. Most of the poor are in urban areas. Out of the 16.8 million is an overlap between groups facing deprivations; for instance, poor, 11.9 million live in urban areas (in a country where 76 per- municipalities with high levels of unmet basic needs also have cent of the population was living in urban areas in 2023), and a high number of indigenous people.27 seven departments account for over half of the country’s im- Poverty reduction in Colombia has been uneven across poverished population (Figure 12). The Caribbean region alone regions and municipalities. Figure 13 shows income growth comprises a third of the poorest individuals. Importantly, there incidence curves by period and region. The periods capture a 24 The ELMO method for between-group decomposition computes between-group inequality as a percentage of the maximum potential segregation between groups considering the number of groups and their relative sizes, instead of using total inequality as the denominator as in traditional between-group decomposition (Elbers et al. 2007). Therefore, the ELMO decomposition assesses how close the groups are to complete income segregation or maximum social injustice. 25 The UNDP reports that “The Human Development Index (HDI) is a summary measure of average achievement in key dimensions of human development: a long and healthy life, being knowledgeable and having a decent standard of living. The HDI is the geometric mean of normalized indices for each of the three dimensions.” 26 UNDP (2024a). The evolution of the HDI between 2010 and 2022 has shown different patterns across the country. A total of 14 departments demonstrated positive progress in all index components (health, education, income), while 17 departments showed limited progress. Two departments, Guainía and Vaupés, experienced a decline in the HDI. This implied a widening of human development gaps, as the difference in HDI between Bogotá and Guainía increased from 0.17 to 0.25 points. 27 World Bank (2021). Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 26 pre-pandemic phase of poverty reduction (2012−19), the time cantly from 42 to 27 percent during that period, driven by an of the COVID-19 shock (2019−21), and the post-pandemic time average economic growth of 3.5 percent per year, a closer look frame (2021−23). The curves show that income growth for the at the municipal level reveals a more complex picture. Despite poorest has been higher than for other deciles at the national the overall poverty reduction, not all municipalities experi- level in recent years, yet heterogeneous; noteworthy are the enced the same level of improvement: 91 municipalities, rep- cases of the Caribbean and Pacific regions. To abstract from the resenting 8.2 percent of the total (depicted in dark red in Figure uniqueness of the COVID-19 shock, a focus has been placed on 14), experienced an increase in poverty levels. Additionally, 440 the 2008−18 period of poverty reduction, using new municipal- municipalities, or 40 percent (highlighted in pink), advanced at ity-level poverty data.28 While national poverty declined signifi- a slower pace than the national average. Figure 8. Regional Relative Poverty Rates in OECD Countries Colombia and Mexico have the largest regional poverty gaps among OECD countries. COL MEX ISR USA CHL ITA LTU ESP BEL PRT DEU CAN AUS CHE POL IRL AUT SVK GRC GBR SWE FRA HUN CZE FIN DNK 0 10 20 30 40 50 60 70 Poverty rate Minimum region National average Maximum region Source: OECD (2022). Note: Relative poverty rates are measured as the share of people with an income below the relative poverty line (60 percent of the national median income). The poorer municipalities are not narrowing the gap in analysis finds evidence of GDP and income convergence in the poverty rates with the better-off ones. Using a period of pov- 2008−18 period,29 yet poverty divergence (Figure 15). Condi- erty reduction in the past decade and a half, a municipal-level tional on initial levels of poverty, and in the context of overall 28 See Acosta, Davalos and Segovia (2024) on the methodology for the creation of poverty maps. 29 Acosta, Dávalos and Segovia (forthcoming). Past literature shows that GDP convergence studies for Colombia have not produced conclusive results; instead, they have been sensitive to the methodology, the time frame, or other factors as surveyed by Galvis-Aponte et al. (2017). In general, there is evidence of a lack of historical convergence in GDP per capita between regions. During the 1980s, there was convergence between departments with similar structural characteristics (geography, location, relevance of mining activities), but this trend stopped in the 1990s, and the gaps have widened since 2010 (World Bank, forthcoming). In the past two decades, there has been a subtle tendency to shift from some level of convergence to divergence (Acosta and Bonet-Morón 2022; World Bank, forthcoming). Although there is evidence of convergence in social indicators, such as the HDI or school enrollment, this has not been true for poverty indicators, as the multidimensional poverty index has shown greater improvements in the wealthiest municipalities (Acosta and Bonet-Morón 2022). CO LO M B I A 27 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T economic growth and poverty reduction during this period, basic needs. Municipalities that had higher initial poverty lev- divergence is observed for various poverty measures: total, els generally experienced a slower rate of decrease compared extreme, and multidimensional poverty, as well as unsatisfied to their counterparts. Figure 9. Between-Region Inequality Figure 10. Subnational Poverty Rates (minimum, Decomposition, as % of the maximum possible maximum, and difference) for a Sample of LAC between-group inequality (ELMO decomposition) Countries The magnitude of territorial inequalities is high. Subnational differences in poverty rates are high. HTI PAN 20% PER 18% 18% 17% COL 16% 16% 14% 13% MEX 14% 12% 12% BRA 12% 10% 10% SLV 10% 9% 8% PRY 6% 5% BOL 4% 4% CRI 2% CHL 0% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Poverty rate PAN COL URY HND CRI ARG PER MEX BRA PRY SLV BOL Minimum region Maximum region Source: Estimates using harmonized data from the Socio-Economic Source: Estimates using the Global Subnational Poverty Atlas (GSAP). Database for Latin America and the Caribbean (SEDLAC). Note: Data are for 2022, except for Bolivia and Panama (2021) and Honduras (2020). Figure 11. Poverty Rate per Department, 2023 Figure 12. Concentration of Poor Population by Sixty-eight percent are poor in Chocó compared Region and Department (%) to 21 percent in Cundinamarca. Over half of the poor population are concentrated in the Caribe and Central regions. 80 Bogotá D.C. 11% 70 68 Valle Bolívar 60 7% Bogotá Pacific 6% 12% Poverty rate 50 17% Atlántico 40 5% 30 21 Eastern Córdoba 20 Caribbean 33% 17% 6% 10 0 Chocó La Guajira Sucre Cesar Córdoba Magdalena Bolívar Cauca Tolima Huila Norte de Santander Caquetá Nariño Boyacá Santander Atlántico Quindío Risaralda Valle del Cauca Antioquia Meta Bogotá. D.C. Caldas Cundinamarca 4% Central 22% Antioquia 11% Santander Source: GEIH 2023. Source: GEIH 2023. Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 28 Figure 13. Income Growth Incidence Curves, by region and income decile Income growth post-pandemic benefited the poorest but was not the same across regions. 30 30 National Bogotá 25 Annual growth rate of income Annual growth rate of income 25 20 20 15 15 10 10 5 5 0 0 -5 -5 -10 -10 -15 -15 -20 -20 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Decile Decile 2012-2019 2019-2021 2021-2023 2012-2019 2019-2021 2021-2023 30 30 Central Region Caribbean 25 Annual growth rate of income 25 Annual growth rate of income 20 20 15 15 10 10 5 5 0 0 -5 -5 -10 -10 -15 -15 -20 -20 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Decile Decile 2012-2019 2019-2021 2021-2023 2012-2019 2019-2021 2021-2023 30 30 Eastern Pacific 25 25 Annual growth rate of income Annual growth rate of income 20 20 15 15 10 10 5 5 0 0 -5 -5 -10 -10 -15 -15 -20 -20 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Decile Decile 2012-2019 2019-2021 2021-2023 2012-2019 2019-2021 2021-2023 Source: Estimates using GEIH 2012, 2019, 2021, 2023. Spatial inequalities reflect historical patterns. The areas 16). Also, regions with higher levels of unsatisfied basic needs that were the most prosperous in the 1500s, measured by their in the 1973 Census are broadly the same ones with the highest population density, remain the most prosperous today (Figure deprivations today. Such departments as Chocó, Guainía, and CO LO M B I A 29 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T La Guajira did not see the poverty reductions experienced in tractive institutions, and a low state presence have been docu- other places of the country (Figure 17). Geographic factors, ex- mented as drivers of the persistent inequities over time (Box 1). Figure 14. Poverty Rate at the Municipal Level 2008 Figure 15. Divergence in Poverty Rates between vs. 2018 (%) 2008 and 2018 (municipal poverty in 2008 vs. While some municipalities reduced poverty at a changes in 2008−18) fast pace, others did so more slowly or not at all. Poorer municipalities reduced poverty at a slower rate than better-off ones. Source: Estimates using poverty map data from Acosta, Davalos, and Source: Convergence background analysis for this report by Acosta, Segovia (2024). Davalos, and Segovia (forthcoming). Figure 16. Correlation between per Capita Income Figure 17. Unmet Basic Needs Index 1973 vs. 2018, in 2021 and Population Density in 1500 by department Places that had higher economic activity in 1500 Spatial gaps in access to services are also still have higher incomes today. persistent over time. 14 13,8 Bogota 80 Per capita income 2021 (log) Antioquia Valle del 13,6 Cauca Risaralda 70 VAU VIC CHO CaldasCundinamar 13,4 Atlantico Santander Quindio 60 GUAI Meta San Andres ca Tolima LGJ 13,2 50 UBN 2018 y Bolivar Providencia 13 Casanare Caqueta Huila Norte de 40 Cesar Magdalena Cauca Santander AMZ COR 12,8 Arauca Cordoba Choco Sucre Guaviare 30 ARA SUC MAG BOL 12,6 CES CAQ Amazonas 20 NDS NAR PUT CAU 12,4 Guainia La Guajira SAPCAL HUI TOL MET CAS Vichada 10 ATL RIS ANT SANT BOY 12,2 QUI BOG VAL CUN Vaupes 0 12 50 60 70 80 90 100 0 5 10 15 UBN 1973 Precolonial population density (1500) Source: Updated per capita income for 2021, sourced from Maloney Source: Population Censuses of 1973 and 2018, DANE. and Valencia-Caicedo (2016). Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 30 Box 1. Colonial Legacy and Regional Inequalities Geographical factors, such as natural resources or health conditions, determined early settlement patterns during the colonial period (Bonet and Meisel-Roca 2006), productive specialization, and economic activity distribution (Gallup, Gaviria, and Lora 2003). However, the colonial legacy seems to be one of the main drivers of the persistent regional inequalities. During the colonial period, colonizers established very different institutions across regions, ranging from highly extractive ones in mining regions with a high presence of slave labor to somewhat more egalitarian structures elsewhere (Bonet and Meisel-Roca 2006). Those institutions persisted during the post-independence period, and this stability accounts for the persistence of regional differences. Regions with a history of slavery and extractive institutions, such as the Pacific and Caribbean coasts, continued to have poor economic institutions. This led to a lack of incentives for wealth creation, limited provision of public goods, and inadequate protection of property rights. In contrast, wealthier regions, such as the Eastern Cordillera, Valle del Cauca, and certain cities in the Caribbean, benefited from better public policies and provision of public goods (Fergusson et al. 2017). The department of Chocó is an example of the extractive colonial legacy, with a development model based on slave labor that kept people in precarious conditions. The institutions in this department focused on extraction without reinvestment, limiting its socioeconomic progress (Robledo-Caicedo 2019). State presence has historically been concentrated in the prosperous regions, leading to better public policies and provision of public goods in these areas. This trend persisted throughout the 20th century, despite the country’s economic growth, modernization, and urbanization (Fergusson et al. 2017). Import substitution policies in the second half of the 20th century may have intensified economic inequalities by favoring industrial sectors focused in the three main cities of the country, imposing higher costs on regions dependent on industrial inputs due to protectionism (Hahn-De-Castro and Meisel-Roca 2018). Despite historical patterns, the evidence shows that in- social spending within their budgets, also saw a faster reduc- creased access to services and assets has contributed to re- tion in poverty: an increase of 1 percentage point in the share ducing municipal poverty. Regression analyses using poverty of municipal current income corresponding to own resources data at the municipality level between 2008 and 2018 (a period was associated with a reduction in the poverty rate of 0.026 of poverty reduction) find that municipalities that experienced percentage points. progress in access to infrastructure services and better health Moreover, an improved fiscal performance has signifi- and educational outcomes, and that were better connected to cantly contributed to the municipal convergence of GDP economic centers, experienced a faster reduction in poverty per capita and household incomes, as well as to such devel- (see Annex D). For example, a 1 percentage point increase in opment indicators as the monetary and multidimensional aqueduct coverage was related to a 0.055 percentage point de- poverty rates.30 This is echoed by Ramírez, Díaz, and Bedoya crease in the poverty rate. Similarly, every percentage increase (2017), who showed that the variable that most explained the in the number of people with higher education was associated changes in poverty indicators at the municipal level in Colom- with a 1.1 percentage point reduction in poverty incidence. Ad- bia between 1993 and 2005 was the municipalities’ own reve- ditionally, a unitary decrease in the number of infant deaths nues—and in particular, the property tax. These authors also per 1,000 inhabitants was linked to a 0.3 percentage point found that areas with better administrative capacities and a decrease in poverty, but only for larger municipalities (those more robust economic base collected more of their own reve- with over 30,000 inhabitants). In the same period, municipal- nues and had greater possibilities of reducing poverty. ities with a higher share of own resources, and with increased 30 Acosta, Dávalos, and Segovia (forthcoming). CO LO M B I A 31 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Chapter 3. Increasing Equity in People’s Access to Assets: a Catalyst for Higher Social Mobility Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 32 Opportunities to accumulate assets are unevenly distributed across groups Human capital, the most important asset that people, particu- unable to read and comprehend a simple text by the age of 10, av- larly poor people, can count on to lead a productive life, greatly erages 51.4 percent in Colombia, above Chile (27.2 percent), Brazil differs spatially in Colombia. Higher human capital, referring (46.9), and Mexico (47.6) but lower than Argentina (59.1), Ecuador mainly to health and education, can lead to higher productivi- (65.9), or Panama (78.5).35 Within Colombia, it can be as high as ty, incomes, and overall economic growth. In Colombia, these 100 percent in certain municipalities,36 and, on average, worse in returns are not spatially homogeneous; for example, returns to public schools.37 Learning poverty is higher in the poorest munic- higher education are greater for a person who completed sec- ipalities (Figure 19), and children living in poorer places have a ondary education in a city such as Medellin instead of Arauca, lower chance of acquiring quality education, with implications and having a university degree has a higher wage premium.31 for their ability to escape poverty and their overall life prospects. Access to education in Colombia has improved substan- Population group and area of residence also affect tially over time, but regional differences remain significant. health outcomes, reflected in the fact that mortality from In 2002, people aged 15 years and over attained, on average, preventable or treatable causes is very high in some areas of 7.5 years of education, and nearly 75 percent had only primary the country. Access to health insurance in Colombia is nearly education.32 Today, 56 percent have completed the secondary universal (though some groups, namely Venezuelan migrants, level.33 Yet the regional differences are enormous, as over 70 are lagging),38 but the poor, rural populations, ethnic groups, percent of people in Bogotá have secondary education com- and migrants underutilize preventive health services (Figure pared to only 28 percent in Guainía or 14 percent in Vichada. 20). This report also puts forward a measure of avoidable mor- People living in poverty and in rural areas, indigenous and tality: the share of excess mortality compared to a frontier or Afro-descendant communities, and those affected by forced gold standard country or region.39 Compared to the gold stan- displacement have lower educational attainment (Figure 18). dard, Colombia presents an unfavorable picture: 74 percent of Venezuelan migrants have higher educational attainment than mortality is avoidable, and the rates are more severe for males, other at-risk groups, a portable asset that they counted on be- infants under 1 year old, and people aged 60 and over. Colom- fore migrating to Colombia.34 bia, along with Mexico and Brazil, is among the LAC countries Beyond educational attainment, quality education is not with higher levels of avoidable mortality. In contrast, countries accessible to all, as half of 10-year-olds in Colombia cannot prop- such as Chile and Uruguay have levels closer to those of the erly read. Learning poverty, meaning the percentage of children gold standard. Looking within Colombia and using Bogotá as 31 UNDP (2024b). 32 Estimates based on the Continuous Household Survey (Encuesta Continua de Hogares [ECH]) for 2002, https://microdatos.dane.gov.co/catalog/75. 33 Estimates based on ECV (2023). 34 Further details on other inequities in educational outcomes, such as enrollment and dropout rates, can be found in World Bank (2021). 35 World Bank, “World Development Indicators,” https://datatopics.worldbank.org/world-development-indicators. 36 A total of 144 out of 1,116 municipalities have learning poverty rates above 80 percent. The municipality of La Guadalupe-Guainía has a learning poverty rate of 100 percent, and Puerto Alegría and Puerto Nariño in Amazonas and Puerto Colombia and Cacahual in Guainía have learning poverty rates above 95 percent. 37 Learning poverty is 60 percent in public schools and 30 percent in private schools (Demombynes 2023). 38 Health insurance coverage is 95.9 percent in the country, but was 51 percent for international migrants in 2023. Some departments are also far below from this average: Vichada (77.5 percent) or Arauca (88.1 percent). 39 Avoidable mortality is defined as the difference between the mortality levels in each year (at the municipal tier) and the unavoidable or frontier mortality rates, which are the lowest mortality levels that can be obtained in each age group with past and current technologies and knowledge. The authors have adapted the Healthy Longevity Initiative methodology to construct municipal-level measures. The frontier used in this exercise is that for women in Bogotá and Japan in 2035 (Vega-Landaeta, Garcia, and Veillard, forthcoming). CO LO M B I A 33 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T a frontier or reference for the country’s municipalities, a very mortality rates are observed in the Pacific, Amazonia, La Gua- heterogeneous picture emerges: even within the country, 45 jira, and Amazonia-Orinoquia regions, with municipalities percent of mortality is deemed avoidable. Higher avoidable experiencing between 53 and 90 percent of avoidable deaths. Figure 18. Percentage of Population Aged 18−64 Figure 19. Learning Poverty vs. Poverty Rate Years of Age with Completed Secondary Education Higher monetary poverty is correlated with Educational attainment is lower among poor, higher learning poverty. rural, and indigenous groups. Poor Non-poor 100 Male Female 80 Learning poverty Rural Urban 60 NARP Indigenous None No-mig 40 Forced Internal mig. displac. International mig. 20 20% 30% 40% 50% 60% 70% 80% 0 Percentage of population aged 18 - 64 years with 0 20 40 60 80 100 completed secondary education and over Poverty rate, 2021 Source: Estimates using ECV 2023; data from people in households Source: Demombynes (2023), based on 2018 census data, 2017 Saber 5, with victims of forced displacement are from the ECV 2022. and poverty map 2021 data from Acosta, Davalos, and Segovia (2024). Avoidable mortality rates for children under 1 year of age without adequate access to basic services and opportunities. show significant territorial disparities. Proven health care in- This continues to be the case. Such heterogeneity is measured terventions before and after childbirth have a significant im- here by the State Discontinuity Index,40 which ranges from 0 to pact on preventing mortality for children under 1 year old and 1, with higher values depicting higher discontinuity (see Annex have been scaled up successfully around the world. Yet, roughly E for methodological details). For Colombia, this measure is con- 13,000 infants die from avoidable causes every year in Colom- structed using data on services that the state provides: access bia, and in some municipalities as many as 90 percent of deaths to improved water, education (learning poverty), and health are avoidable. In some of the poorest areas, such as La Guajira outcomes (avoidable infant mortality). In line with the litera- and the Pacific and Orinoquía regions, avoidable mortality is ture, the poorest departments historically and to date, such as particularly high (Figure 21). Communicable diseases, as well Vichada, Guainía, Chocó, La Guajira, and Amazonas, have high- as maternal, perinatal, and nutritional conditions, are the main er heterogeneity within their territories (Figure 22). This lack of causes of avoidable deaths among children under 1 year old. equitable access to services across poorer territories exacerbates These differences in human capital outcomes partly re- inequalities and weakens social cohesion.41 In comparison to oth- flect an unequal presence of the state in the territory. As Box 1 er countries in the region, such as Brazil and Ecuador, Colombia describes, state provision of public services has historically been has a greater discontinuity in the state’s presence (Figure 23). 42 concentrated in prosperous regions, leaving excluded groups The levels of state discontinuity tend to be higher in rural areas.43 40 Ceriani, Lopez-Calva, and Restrepo-Oyola (forthcoming). 41 Ibid. 42 Regional comparisons use a discontinuity index constructed by three dimensions: service delivery percentage of households with piped water in the region and percentage of households with electricity); education (percentage of illiterate people); and health (infant mortality rate). 43 Ceriani, Lopez-Calva, and Restrepo-Oyola (forthcoming). Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 34 Figure 20. People with Access to Preventive Health Figure 21. Avoidable Infant Mortality Rate, Bogotá Care (%) frontier The poor, rural, indigenous and male Mortality among infants is severe: some populations, as well as foreign migrants, have municipalities have up to 90% avoidability. lower access to health care. Poor Non-poor Male Female Rural Urban NARP Indigenous None Forced International displac. Non-mig mig. Internal mig. 20% 30% 40% 50% 60% 70% Percentage of people with access to preventive care Source: Estimates using ECV 2023; data for people in households with Source: Vega-Landaeta, Garcia, and Veillard (forthcoming). victims of forced displacement are from 2022. Notes: The total avoidable deaths in less than one year are roughly 13,000, 134 (with international frontier), and 7,000 (with Bogotá as a frontier). Areas that experienced armed conflict also suffer from school, Figure 24), higher illiteracy rates (7 percent lower), low- low state presence and thus a lower accumulation of human er sewerage coverage (20 percent lower), and lower access to capital.44 Whether the presence of armed groups resulted in state-provided services, compared to areas without armed con- lower access to services and educational attainment, or wheth- flict. Additionally, municipalities where illegal armed groups er armed groups established themselves in places with a lower stayed longer tend to also have lower educational attainment, state presence, cannot be fully ascertained. Nevertheless, using and this relationship has not decreased over time. This signals a novel dataset, the evidence shows an association between additional and more severe challenges in the accumulation of past presence of illegal armed groups and lower human cap- assets in these areas that intertwine with the dynamics of con- ital: lower learning (specifically, lower performance in reading flict they experienced. and mathematics in standardized test scores at the end of high 44 Arjona, Moore, and Otero-Bahamón (2024). CO LO M B I A 35 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Figure 22. State Discontinuity Index Figure 23. State Discontinuity Index Comparable The poorest regions have higher heterogeneity between Countries (1 more discontinuity) in the presence of the state within their Using an index comparable across countries, territories. Colombia is in the middle of the regional distribution. 45 Honduras 0,22 Bolivia 0,21 El Salvador 0,19 Peru 0,18 Colombia 0,17 Brazil 0,12 Ecuador 0,11 Guatemala 0,07 Panama 0,06 Dominican Republic 0,04 Costa Rica 0,01 Source: Ceriani, Lopez-Calva, and Restrepo-Oyola (forthcoming). Source: Ceriani, Lopez-Calva, and Restrepo-Oyola (forthcoming). Note: The map displays discontinuity indexes based on the percent of Note: The indicator of comparable discontinuity between countries households without access to an improved water source, the percent was constructed from the percent of households without access to of 10-year-old children who cannot read and understand simple text, piped water, the percent of illiterate people, and the infant mortality and the rate of preventable mortality at age 0 (per 1,000). rate (per 1,000). Social capital, which is instrumental to economic oppor- is very similar in those two regions and even lower in the Pacific tunities, also varies across groups and regions. Social capital region. Moreover, only 15 percent of the population aged 18 and refers to an asset that fosters personal relationships, civic in- over belongs to a group or organization.47 Turning to a different volvement, and social networks, and it can be important for measure, municipalities with a past presence of illegal armed an individual’s access to support and opportunities. Fifty-six groups perceive a lower enjoyment of rights and freedoms48 percent of adults in Colombia count on support from a trust- across several dimensions, including voting, political engage- ed network.46 Yet, this percentage varies greatly by region: the ment, and freedom of movement, when compared to regions highest is 74 percent in Bogotá and the lowest is 46 percent in untouched by conflict, as depicted in Figure 25. This may affect the Central region. When accounting for network size, the per- their perceived ability to build networks. centage of people with a trusted network of six or more people 45 Ceriani, Lopez-Calva, and Restrepo-Oyola (forthcoming) also include a specification for the index that includes a security dimension (the number of days in the year without episodes of violence) and an additional variable for the public service dimension (electricity coverage). 46 DANE, Political Culture Survey (Encuesta de Cultura Política [ECP]), 2023, https://www.dane.gov.co/files/operaciones/ECP/pres-ECP-2023.pdf. 47 Ibid. Groups can include communal action boards, religious, cultural, or sporting organizations, work cooperatives, ethnic organizations, educational organizations, residential complex boards, community-based organizations, or farmers organizations. 48 Arjona, Moore, and Otero-Bahamón (2024). Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 36 Figure 24. Results on Standardized Tests (Saber 11- Figure 25. Average Score (0-6) on Perceptions of reading) 2014−19 Rights and Freedoms Municipalities with a past presence of armed Past presence of armed groups is correlated groups also have lower learning outcomes. with lower perceptions of rights and freedoms. Freedom to: Protest Create a civic organization Become a member of a political party Run for a position of popular election Vote Choose a partner Religion Speech Mobility 0 1 2 3 4 5 6 Score Armed conflict Non-armed conflict Source: Arjona, Moore, and Otero-Bahamón (2024). Source: Arjona, Moore, and Otero-Bahamón (2024). Natural capital, mainly forests, is concentrated in the tenure and use (see Box 2), as illegally acquired lands—from poorest areas, and its use is often unsustainable and has not baldíos, collective territories, or protected areas—are mainly led to poverty reduction. Colombia is rich in forests, particu- transformed into pasture, meaning that cattle accounts for a larly in poorer areas in the Pacific and the Orinoquia/Amazonia significant percentage of deforestation growth.52 regions (Figure 26). They cover nearly 72 percent of the coun- Poverty reduction is not consistently linked to deforesta- try’s surface, making Colombia 10th in the world in terms of tion growth. In other words, in addition to being counter to Co- forest coverage.49 However, the country has witnessed signif- lombia’s climate objectives, the unsustainable use of natural icant deforestation, with a loss of almost 5 million hectares in capital has not been fully to the benefit of poorer populations. the past two decades50 (6 percent of its forest coverage in 2000, Instead, there is a positive correlation between deforestation equivalent to 30 times the size of Bogotá). Deforestation has and such factors as illicit crops, palm oil cultivation, livestock, been larger in some municipalities in the Caribbean, Norte economic growth, and agricultural credit. From 2018 to 2021 in de Santander, Antioquia, and Amazonia/Orinoquia regions particular, municipal government performance emerged as a (Figure 27) and has accelerated in those areas since the 2016 significant factor in combating deforestation.53 The main chal- Peace Agreement,51 though the overall pace has subsided in lenge is thus to ensure the sustainable use of natural capital in the past year. In areas such as the Amazonia, deforestation is a way that both protects it and benefits poor populations living closely linked to proximity to roads and historical issues of land in resource-rich areas. 49 Data from Global Forest Watch, https://www.globalforestwatch.org/dashboards/global/. 50 Ibid. 51 Prem, Saavedra, and Vargas (2019) found that areas controlled by the FARC before the permanent ceasefire experienced a larger increase in deforestation. 52 World Bank (2023c). Most deforestation in the area occurs within 2.3 kilometers of a road and even just outside national natural parks and indigenous territories. Despite the legal protection of the Amazonia for environmental conservation, the same laws also create a system of subtractions allowing lands to be removed from the forest for rural development (Cheston et al. 2023). 53 Prem, Saavedra, and Vargas (2019) also found that the increase in deforestation after the Peace Agreement is attenuated in municipalities with higher state capacity. CO LO M B I A 37 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Figure 26. Proportion of Tree Cover, 2008 Figure 27. Annual Growth of Deforestation (%), Natural capital wealth is higher in the Pacific 2008−18 and Amazonia/Orinoquia regions (which tend to There is a higher growth of deforestation in the be poorer areas). Caribbean, Norte de Santander, Antioquia, and Orinoquia/Amazonia regions. Source: Estimates using data from Global Forest Watch, consistent with official data from Colombia’s Institute of Hydrology, Meteorology and Environmental Studies (IDEAM) (over 90 percent correlation between datasets). Physical assets are also unequally distributed. Hous- This high concentration and lack of access to land have reduced ing ownership is lower among the poor and other population opportunities for rural populations and intensified tensions groups54 (Figure 28), and rural land ownership is extremely un- and conflicts, as described in Box 2. equal. The latter is particularly the case in certain regions, such With a high share of the poor living in urban areas, in- as the Pacific area, Putumayo, and Meta, as well as some mu- adequate housing and infrastructure are the main sources of nicipalities in the central region, as nearly 300 municipalities intra-urban inequality. Fully 4.4 out of 5.1 million households have Gini coefficients of rural land property of 0.8 and above with housing vulnerability are in urban areas.56 One in every (IGAC 2023) (Figure 29). The top 1 percent of rural landowners six urban households reports a qualitative housing deficit,57 with larger land extensions owns 25 percent of total private with substantial regional gaps: in Vichada, eight out of 10 ur- rural land, reaching 80 percent in the Pacific region.55 Interna- ban households are in a qualitative housing deficit, whereas tional comparisons show a much higher land concentration in in Boyacá or Quindío, the figure is roughly eight out of every Colombia than in other countries in the region (Oxfam 2017). 100 households. 54 Approximately 26 percent of indigenous people live in collectively owned dwellings. The migrant population includes foreign migrants and internal migrants who have lived in another country or changed municipalities in the past five years. The displaced population are individuals living in households with at least one person who is a victim of conflict or displacement. 55 IGAC (2023). 56 Household vulnerability is measured by overcrowding and qualitative housing deficit (World Bank 2021). 57 DANE (2023). Qualitative housing deficit refers to households living in dwellings that require improvements or adjustments to meet adequate living conditions. Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 38 The lack of adequate dwelling materials increases the percent due to improvements in physical and social vulnera- risk of climate-related disasters, despite improvements over bility.58 Nevertheless, departments in the Caribbean, such as time. Between 2005 and 2018, the percentage of municipalities Bolivar or Cordoba, still have over a half of their municipalities with low and very low risk of disasters increased from 13 to 43 classified as having high risk of climate shocks. Figure 28. Percentage of People in Households Figure 29. Municipalities with Extremely High Land with their Own Home and Deed Registered in the Inequality (Gini coefficient>0.8) Name of a Household Member Gini of the Distribution of Rural Land Lower housing ownership among the poor, Property 2023 (excluding indigenous rural and ethnic groups, migrants, and reservations and collective displaced people. territories) Poor Non-poor Male Female Rural Urban NARP Indigenous Non-ethnic group Foreign mig. Forced displ. Non-migrant Internal mig. 0% 10% 20% 30% 40% Percentage of population Source: Estimates using ECV 2022. Source: IGAC (2023). Financial inclusion is also quite different across groups Banking infrastructure is almost non-existent or very limited and territories. Financial inclusion improved substantially in in municipalities in these as well as in the Pacific and La Guaji- the past year, as the number of adults with at least one finan- ra regions. In fact, 22.6 percent of municipalities do not have a cial product increased from 14 to 36 million between 2006 and banking office, and 10 municipalities do not even have a bank 2023 (Banca de las Oportunidades 2023). However, disadvan- correspondent (located primarily in Amazonas, Guainía, or taged groups still face barriers to accessing financial products, Vaupés). Moreover, even though correspondents are the main as less than 40 percent of the poor, rural, or indigenous popu- access point to financial services, particularly where banking lations live in a household with at least one financial product offices are non-existent, some people live too remotely; the compared to 63 percent of the total population (Figure 30). median distance to the nearest correspondent is less than 500 Regional gaps are also prevalent (Figure 31); one out of every meters in 27 departments but reaches 11.2 kilometers in Ama- three people in Cauca or Chocó live in a household with at least zonas (Banca de las Oportunidades 2023). one financial product compared to four out of five in Bogotá. 58 Alonso-Beltrán et al. (2024). Risk refers to potential loss of or damage to life, health, or assets, and it is measured as a function of hazard (probability of occurrence of a phenomenon), exposure (people, infrastructure, or any asset located in hazard-prone areas), and vulnerability (socioeconomic and physical factors determining susceptibility to being impacted by a hazard). CO LO M B I A 39 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Figure 30. Percentage of People in Households Figure 31. Distribution of Bank Correspondents per with at Least One Financial Product Thousand Inhabitants 2023 There is lower financial inclusion among the There is an unequal distribution of banking poor, rural and ethnic groups, and international infrastructure migrants. Poor Non-poor Male Female Rural Urban NARP Indigenous Non- ethnic group Non-migrant Internat. mig. Internal mig. 0% 20% 40% 60% 80% Percentage of population in a household with at least one financial product Source: Estimates using GEIH 2023. Source: Banca de las Oportunidades (2023). Note: Financial products considered: checking account, savings account, time deposit, mortgage, auto loan, investment loan, credit card, and others. Box 2. Land and Poverty: Some Insights from the Literature In Colombia, the limited access to land, a high concentration of land and informal land tenure, and a lack of support for agricultural productivity in rural areas restrict income generation (Ibañez 2016; Arteaga et al. 2017). In rural areas, only two out of every five people can access land, and land informality is 52.7 percent (Neva and Prada 2019). The high informality results from complicated and costly processes, outdated cadastral information, and administrative barriers (Arteaga et al. 2017). In addition, land ownership inequality is very high, as shown by a Gini coefficient of 0.89 (IGAC 2023), due to historical dynamics involving occupation, violence, armed conflict, money laundering by drug traffickers, and policies promoting large-scale agricultural activity and land accumulation (Ibánez and Muñoz-Mora 2011; Misión para la Transformación del Campo 2014). The high concentration of land ownership limits land access and leads to the migration of the rural poor to less productive areas, contributing to environmental damage and land underutilization (World Bank 2004). Only 38.6 percent of rural land is used for agriculture, with extensive cattle ranching dominating over agricultural production—79.7 compared to 20.1 percent, respectively (Ibáñez 2016). The increased land inequality is also associated with the number of displaced people, which in turn could lead to a further concentration of rural property (World Bank 2004). Disagreements over access to land and the use of natural resources have also led to socio-environmental conflicts marked by scarcity and ecosystem changes (Rincón et al. 2016). Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 40 Land reform initiatives beginning in 1936 have been ineffective in deconcentrating land ownership or reducing land conflicts (Ibáñez and Muñoz-Mora 2011). The most consistent land policy has been the adjudication of wastelands, but this has been without complementary programs for productivity gains (Arteaga et al. 2017). In fact, participation in a land delivery program may lead to a decline in agricultural productivity. Rincón-Barajas (2023) shows that agricultural productivity dropped by 50 percent in the 2.5 years following land delivery, recovering only after the fifth year, though the delivery of credit or technical assistance can mitigate the drop in productivity. The last attempt for rural reform stemmed from the 2016 Peace Agreement with the FARC, which included a Comprehensive Rural Reform program aiming to increase peasant families’ access to land. This included the creation of a Land Fund (with land acquired or purchased by the state and the annulment of illegally acquired and unduly occupied vacant lands) and instruments to improve land formalization (Arteaga et al. 2017). Lower access to productive assets leads to fewer economic opportunities Populations and territories that have limited access to assets women not in education, employment, or training (NEET) is subsequently have lower economic opportunities, meaning four times higher than that of men in Colombia, with 92 per- that their productive capacity is thwarted by their inability cent of those young women dedicated to household tasks to accumulate assets. Once people count on productive as- (World Bank 2019). Moreover, the gender gap in labor force sets, their ability to generate income depends on how they use participation has remained relatively stable over the past two those assets in markets and the returns they receive. Most of decades, with 52.5 percent of women participating in the la- the poor’s income is generated in labor markets (73 percent in bor market compared to 76.6 percent of men in 2023 (a gap of Colombia and 64 percent among the monetary poor), where 24 percentage points, slightly down from the 28 percentage they use their education, skills, and health to receive wages, or points in 2007). In poorer regions, such as Chocó, only 27 per- where they employ other assets, such as access to finance and cent of working-age women participate in the labor market, property, to set up a business. However, if the ability to accumu- compared to 55 percent of men in the same department or 62 late assets is unequal, as has been the case so far, it is likely that percent of women in Bogotá. economic opportunities will be so as well. Other studies elab- In terms of returns, regional gaps also exist. Poorer de- orate on key labor market challenges in Colombia that have partments, such as La Guajira, Chocó, Córdoba, and Sucre, face poverty and equity implications: high informality, low produc- the largest gaps in labor income compared to the main eco- tivity, and the impact of the future of work. These studies also nomic centers. Using a Oaxaca-Blinder decomposition of labor propose policies to propel economic growth, productivity, and income gaps of departments relative to the main cities (Bogotá, formalization (World Bank, forthcoming; World Bank 2021). Medellín, Cali, Manizales, and Bucaramanga), it can be seen Yet, even if the access to assets were the same, the re- that while differences in endowments account for nearly half of gional differences in use and returns hinder the possibilities these income gaps, people’s endowments have higher returns of leveraging them. In terms of assets use, a clear illustration is in leading regions or cities.59 This suggests labor mobility bar- that of gender gaps in labor force participation. Colombia has riers (mobility of people is further discussed in Chapter 4) and made progress in eliminating health and education differences local factors that hinder economic opportunities. Unsurprising- in recent decades, but women still face challenges to engaging ly, these departments consistently rank among the lowest in in labor markets (World Bank 2019). The proportion of young 59 Evidence from a decomposition of Oaxaca household labor income per capita was implemented using variables, such as: sex of the head, age and age squared, household size, whether there are children in the household, educational level of the head, and type of employment (right). CO LO M B I A 41 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T terms of competitiveness, infrastructure, educational attain- es arise from a business opportunity in the market (opportuni- ment, innovation, and market efficiency.60 ty); instead, roughly half emerge because the entrepreneurs As a result, access to quality jobs is unequal. The labor have no other income alternative (necessity). These necessity market in Colombia is of poor quality, not only because of the microbusinesses are more often by entrepreneurs from poor high levels of informality and low average real hourly earnings, households (46 compared to 30 percent for opportunity enter- but also because of other dimensions regarding stability, work- prises), and they have lower productivity (Figure 34). Spatially, ing conditions, and job satisfaction, among others.61 Formal the productivity of microbusinesses65 is lower in poorer places: jobs have higher levels of quality of work. However, the same productivity in certain departments, such as Antioquia, Risar- population groups that have a lower capacity to accumulate alda, or Bogotá, is more than double that for Cauca or Córdoba assets have a lower likelihood of getting a formal job: poor, (Figure 35). About two out of three microbusinesses in Colom- rural, and indigenous groups, Afro-Colombians, and interna- bia are classified as weak in the Microbusinesses Robustness tional migrants tend to have access to jobs of poorer quality Index,66 and they also show lower productivity and a higher (Figure 32). Along the same lines, departments with higher percentage of owners in a situation of productive exclusion.67 poverty rates tend to have a lower proportion of formal em- According to the type of municipality, the percentage of weak ployment (Figure 33), below 20 percent for such departments microbusinesses increases from 54.6 percent for big cities and as Nariño, Sucre, La Guajira, and Cauca (compared to 67 per- agglomerations to over 80 percent for predominantly rural mu- cent in Bogotá).62 Informality is persistently high and strongly nicipalities. Females also have lower access to other sources of concentrated, particularly in municipalities in the south of the income, as 37 percent of microbusinesses have a female own- Caribbean region and South-Occident areas of the country, in er.68 Furthermore, the percentage of female-owned microbusi- contrast to the Andean zone where there is a cluster of low in- nesses classified as weak is 65 percent, while for male-owned it formality.63 is 68 percent, yet the value added per worker in female-owned Millions of families rely on income from small and gen- microbusinesses is 60 percent of that for male-owned en- erally low-productivity businesses rather than salaried jobs. terprises, arguably related to the higher burden of domestic Microbusinesses, those with nine or fewer employees, consti- tasks.69 Barriers to market participation are also observed in tute 99 percent of the business fabric of the country and 66 areas where armed groups imposed rules that made it more percent of total employment.64 Not all of these microbusiness- difficult to enter the market with a small business.70 60 Chocó ranks 29th out of 33 departments, according to the Departmental Competitiveness Index (DCI), La Guajira ranks 26th, Sucre 23rd, and Cordoba 22nd. The new departments (Vichada, Guainía, Vichada, and others) are listed below and are not included in the Oaxaca decomposition of Figure 47. 61 Gómez-Salcedo et al. (2017) construct a multidimensional quality of work index based on six dimensions: working conditions, income, social security, labor stability, employment perception, and underemployment. The authors not only find general low levels of quality of work for Colombia, but also lower values for lower educated workers and workers in firms with lower size, and no differences by gender. 62 Formal jobs refer to salaried workers who contribute to social security or are self-employed in the formal sector (registered companies). 63 Acosta et al. (2024). 64 Data from 2019. The universe of businesses considered in the document does not include the agriculture, finance, or government sectors, or domestic services (Fernández and Sarango 2023). 65 Measured as value added per work. 66 According to the Microbusinesses Robustness Index (IMICRO), which measures if these small economic units accumulate a set related to accountability, use of information and communications technologies, and formalization, which are closely linked to performance. IMICRO results correspond to 2021 (Angulo et al. 2023). 67 Inclusión SAS and UN (2023). Measured by an index with (i) monetary poverty, (ii) social protection and labor inclusion, and (iii) education and human capital. 68 Angulo et al. (2023). 69 Ibid. 70 Arjona, Moore, and Otero-Bahamón (2024). Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 42 Figure 32. Employed Population with Formal Jobs Figure 33. Correlation between Poverty Rates and (%) Percentage of People in Formal Jobs Fewer poor and rural people, migrants, and The quality of jobs is lower in poorer places. members of ethnic groups have formal jobs. Poor Non-poor 80 70 Male Female Poverty rate (%) 60 50 Rural Urban 40 NARP 30 Indigenous Non-ethnic gr 20 Non-migrant 10 International Internal mig. migrant 0 0 20 40 60 80 10% 20% 30% 40% 50% 60% Formal employment rate (%) Formal employment (% of working population) Source: Estimates using GEIH 2023. Source: Estimates using GEIH 2023. Figure 34. Poverty Rate and Value Added per Figure 35. Value Added per Worker vs. Worker, by type of microbusiness Department’s Poverty Rate Poverty and productivity are different for Small firm productivity is lower in the poorer “opportunity” and “necessity” microbusinesses. places. 46% 70% 50% 544.523 560.000 per worker (COP) 40% 540.000 60% Value added Poverty rate 30% 520.000 30% 500.000 50% Poverty rate (%) 20% 475.418 480.000 10% 40% 460.000 0% 440.000 30% Opportunity Necessity Opportunity Necessity 20% 10% Poverty rate Value added per 0% 300 400 500 600 700 800 worker (COP) Value added per worker (thousands of pesos) Source: Estimates using GEIH 2021 and the 2021 Microbusinesses Survey (Encuesta de Micronegocios [Emicron]). Institutions are not fully equipped to eliminate gaps in opportunities, including territorial gaps Recognition of the role of institutions in regional develop- growth, employment, and productivity. They also foster in- ment is growing. Institutions shape the capacity of people to novation, support a vibrant entrepreneurial ecosystem, and accumulate assets and the way that economic actors engage enhance the economic outcomes of decentralization efforts. within a territory, which in turn affect the region’s economic The literature posits that lagging and peripheral regions stand CO LO M B I A 43 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T to gain more from improvements in institutional quality. Con- of institutional capacity and fiscal management at the sub- versely, ineffective institutions and corruption can impede their national level is needed to reduce territorial inequality and capacity to develop.71 improve coordination in planning, investment, and service The institutional setup in Colombia aims to enable ser- delivery toward poverty reduction and overall regional devel- vice delivery at the subnational level but is challenged by opment. This opens up a host of possibilities through regional overlapping competencies across government tiers. There interventions that pool resources and capacities. are 32 departments and over 1,000 municipalities, each with Subnational governments also grapple with revenue political, administrative, and fiscal autonomy. These subna- generation and fiscal autonomy and management. Public tional governments navigate a complex legislative landscape investment resources are significantly concentrated in the with over 200 competencies in public service provision, many central government, despite the constitutional regime favor- of them central to building the productive capacity of the poor. ing decentralization. The central government in Colombia Yet, there is diluted accountability in the provision of services, represents almost 70 percent of total public investment, much as subnational governments share—without a clear definition higher than other countries in the LAC region, such as Brazil of roles—the provision of public services across multiple sec- or Mexico (25 percent), or Perú or Argentina (36 percent).74 In tors, including education, health, water, sanitation, housing, part, this concentration of resources reflects a process of re- and transport.72 Moreover, poorer municipalities have lower centralization.75 Performance measures indicate weaknesses in municipal performance (Figure 36). The problems of institu- domestic resource mobilization, public investment, and service tional quality are compounded in areas affected by armed con- delivery. The Measure of Municipal Performance (Medición de flict; for example, in areas where armed actors were present Desempeño Municipal [MDM]) produced by the National Plan- in the past, people turn to state institutions less for dispute ning Department (Departamento Nacional de Planeación [DNP]) resolution.73 underscores these issues, especially in the poorest municipali- To address these coordination challenges, the Colombian ties, which exhibit significant struggles with revenue mobiliza- government has enacted laws to promote coordination and tion (Figure 37).76 The MDM’s management component for 2021 collaboration among subnational governments and with the shows the largest deficits for poorer municipalities, especially central government. These include the Regions Law (Law 1962) resource mobilization. The score for municipalities with pov- and Law 1955, both of 2019, which support regional associations erty rates below the average is twice as high as that for munici- and investment projects. The government has also introduced palities with higher poverty rates, indicating the challenges of Contratos Plan and territorial pacts to enhance inter-institution- poorer municipalities in collecting their own resources to invest al coordination and support strategic investments for regional in the well-being of their populations. On fiscal management, development. Despite these measures, further strengthening the DNP’s Fiscal Performance Index highlights the risk to fiscal 71 Rodríguez-Pose et al. (2024). 72 The Organic Law of Territorial Planning (Ley Orgánica de Ordenamiento Territorial, Law 1454 of 2011) establishes the competencies of the municipal and regional governments; however, it does not provide specific guidance on the public services that need to be provided by each level of government. 73 Arjona, Moore, and Otero-Bahamón (2024). 74 Eguino et al. (2020). 75 Lopez-Murcia (2022). The 1991 Constitution provided that transfers to the municipalities would be linked to the current revenues of the nation (CRN) and should reach 46.5 percent of these. However, Legislative Act 01 of 2001, which was a response to the economic crisis of the late 1990s, decoupled the growth of transfers from the CRN. As a result, the distributed resources under the Sistema General de Participaciones (SGP) went from representing 42 percent of the CRN in 2002 to less than 22 percent in 2023. Along the same lines, the average amounts of Sistema General de Regalías (SGR) projects have decreased by approximately 34 percent, from 4.3 billion in 2012 to 2.8 billion in 2023 (data from DNP). 76 The MDM has two components: management, which measures local administrations’ actions and results toward the improvement of population welfare; and outcomes, which measures social welfare progress in terms of education, health, public services, and security. The MDM methodology also includes a classification according to the initial capacities of municipalities: G1 high level, G2 high-medium level; G3 medium level, G4 low-medium level, and G5 low level (Angulo et al. 2018). Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 44 sustainability of 749 municipalities (68 percent) in 202277 and ipalities: the poverty rate is 26 percent higher in municipalities emphasizes the need for improved debt management, revenue declared to be in fiscal deterioration (index <40) than in those generation, and capital formation. Similarly, the index points deemed to be solvent (index between 70 and 80).78 to low performance in critical areas among the poorest munic- Figure 36. Management Component, Municipal Figure 37. 2021 MDM Scores for Municipalities Performance Index 2021 above and below the Poverty Average Poorer municipalities with lower initial Poorer municipalities face higher challenges in capacities have lower municipal performance. resource mobilization. 90 Total score Municipal performance index Outcomes Management component 80 Security - 70 Public services 60 Health 50 Education 40 Total score Management 30 Territorial planning 20 Open government Execution of resources 10 Resource mobilization 0 0 0,2 0,4 0,6 0,8 1 0 20 40 60 80 100 Poverty 2021 Score Cities G1 G2 G3 G4 G5 Poverty below average Poverty above average Source: DNP. Source: DNP; 2021 poverty map data from Acosta, Davalos, and Note: Dots are color-coded based on the MDM initial capacities Segovia (2024). classification according to four dimensions: economic activity, geography, demographics, and resources (Angulo et al. 2018). G1 corresponds to high initial capacities; G2, high-medium; G3, medium; G4, low-medium; and G5, low capacities. In a context of vastly different realities within Colombia, graphic, socioeconomic, and environmental factors. Similarly, public policies remain, to a large extent, uniform. For instance, the OECD goes beyond the municipal level to create a typology 88 percent of municipalities are classified by the government for regions that are larger than municipalities but smaller that under category 6 (categories from 1 to 6, depending on popu- departments (territorial level 2 regions). Analysis of econom- lation and annual incomes, Law 617 of 2000), which does not ic and social dynamics at this regional level could allow for allow for sufficient differentiation across municipal entities. To greater differentiation toward inter-subnational government address this issue, the ongoing Decentralization Mission79 has coordination.80 explored other municipal categorizations, considering demo- 77 DNP, “Resultados Índice Desempeño Fiscal 2022,” https://colaboracion.dnp.gov.co/CDT/Desarrollo%20Territorial/Desempeno_Fiscal/PPT%20IDF%202022.pdf. The Fiscal Performance Index measures financial management of local entities using two dimensions: (i) results (dependence on transfers, investment, indebtedness, surplus, or deficit); and (ii) territorial financial management (revenue programing and execution capacity, investment execution capacity, compliance with limits of Law 617 of 2000, and bonuses for own efforts or cadastral updating). 78 Only two municipalities were classified in 2022 as “Sustainable” with a score above 80. 79 The Decentralization Mission is an initiative led by DNP to prepare technical studies and inputs on the current state competence distribution between the national and territorial governments and to propose initiatives to reorient the decentralization model. 80 Bateman et al. (2018). CO LO M B I A 45 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Education policies need to recognize the heterogeneity ever, wealthier areas contribute more of their own resources to of regions and subnational governments in outcomes and education, far exceeding central government transfers.82 capacity, so that service delivery to the poorest areas is more Families in poor areas are disproportionately exposed effective. The education system is not fully prepared to tackle to higher health risks, malnutrition, and a greater burden of the disparate learning outcomes across the territory. School infectious diseases, non-communicable diseases, and disabil- and teacher quality are vastly different in poorer areas than ity. Nearly 5 million Colombians, mostly living in remote and in richer ones, yet there is no adaptive curriculum or policy on rural areas, have little or no access to health services. When textbooks and school resources for subnational areas. Hence, health care is accessible to the poorest, it is often of poor qual- the burden of curriculum development and educational re- ity, fragmented, primarily curative rather than preventive, and source production falls on teachers and school administrators. not tailored to the needs of the population. This situation is The best prepared teachers tend to select the best schools and further exacerbated by gaps in the availability of health per- areas, leaving the most vulnerable schools with the least pre- sonnel, medicines, and financing in areas with the highest con- pared or temporary teachers, often leading to poor teaching centration of vulnerable populations. The low availability and practices in the most impoverished places. Furthermore, ed- poor quality of primary health care services in remote and rural ucational institutions lack the resources to cover the costs of areas is a driver of poor health outcomes, in addition to social educational materials, as schools with poorer outcomes sys- determinants of health and other factors. Also, there are few tematically receive less funding from the national government. financial incentives for insurers and health care providers to Territorial entities with high levels of learning poverty receive ensure quality health services to the poorest and underserved only marginally higher transfers per capita for quality.81 How- areas.83 As a result of unequal access to assets and opportunities, social mobility is low, leading to a vicious cycle of inequities Social mobility is a dimension of progress. Inequality can be create skilled jobs, and thus economic growth, even more so in accepted to a certain extent, especially when income inequal- a middle-income country like Colombia (World Bank 2024b). ity is paired with upward mobility. Parents aspire to a better Conversely, as the recent World Development Report 2024 life for their children, and those children hope to climb up the states, “Social immobility, in the aggregate, holds back the energies socioeconomic ladder rather than be destined to the rung they that drive the forces of creation” (World Bank 2024b). were born into. But in reality, not everyone enjoys the same Economic advantages or disadvantages often persist opportunities to lead the life of their choosing, as the ability through multiple generations. There are circumstances out- to accumulate assets is often dictated by factors beyond their side a person’s control that influence one’s trajectory in life control. Moreover, social mobility is central to more equitable and capture the inequality of opportunities in a society, such and inclusive growth. Higher mobility helps boost talent and as place of birth, ethnicity, race, sex, and family background. 81 The Colombian education system is decentralized, allowing territorial entities certified in education (ETCs) to autonomously handle administrative functions. ETCs include departments, municipalities with over 100,000 inhabitants, and smaller municipalities meeting institutional capacity and resource requirements, among others. The central government transfers an allocation per student of the ETCs and provides additional funds, including quality allocations based on equity, performance, and dispersion criteria. 82 For an expanded discussion, see World Bank (2021). 83 Ibid. Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 46 As previously discussed, these circumstances matter in peo- There are several measures aimed at capturing these ple’s opportunities to accumulate productive assets, generate inequalities in opportunity and lack of social mobility. The income, and dictate a life path beyond such factors as effort or Inequality of Economic Opportunity Index (IO) represents the luck. The World Bank (2024b) argues that social immobility is (lower bound) share of income inequality that can be attribut- linked to three factors: (i) networks, which preserve social and ed to differences in individuals’ predetermined circumstances. economic groups; (ii) neighborhoods, which shape access to In addition, two measures of mobility across generations are opportunities (as described throughout this report) and can commonly used: absolute mobility, capturing the extent to limit movement of people and agglomerations; and (iii) norms, which a generation of children is better off than their parents; which limit some groups’ (including women’s) opportunities to and relative mobility, the extent to which an individual’s wel- accumulate assets and lead the life of their choosing. fare position in society is independent of the position of his or her parents. IO and relative mobility are measures that capture Figure 38. Labor Income Inequality of Opportunity societal unfairness and the persistence of inequalities.84 The Index, % explained by each circumstance 2022 positive relationship between relative mobility and equality One-third of inequality of opportunity is related of opportunities supports the fact that circumstances beyond to place of birth. a person’s control are crucial in determining intergenerational persistence.85 2% 6% Gender In Colombia, at least 30 percent of labor income and 24 11% percent of household income inequality of the mean log de- Presence of conflict in the municipality at birth viation inequality measure (and more than 44% of the Gini 17% 2% Area of birth (urban vs. rural) income inequality) is determined by circumstances at birth.86 9% A large share of this inequality in opportunities is linked to peo- Population density of municipality of birth ple’s socioeconomic background—captured by their parents’ Department of birth educational level (Figure 38)—and to place of birth (which de- 54% Ethnicity termines nearly a third of the inequalities in opportunities to earn income over a lifetime). Consistently, Colombia stands out Highest education of parents in the region as a country where place of birth is an important factor when explaining inequality of opportunities.87 In Latin Source: Estimates based on ECV 2022, for the population aged 20−84 America generally, more than 20 percent of current income years. The estimates correspond to ex ante inequality (lower bound), inequality is attributed to inherited factors, ranging from 16 following Ferreira and Gignoux (2011). This index is estimated as the percent in Argentina to 32 percent in Guatemala.88 ratio of inequality explained by observable circumstances at birth to total inequality. Inequality is measured with the mean log deviation index. 84 Narayan et al. (2018). 85 CAF (2022) found that in the past decades there has been an increase in the probability of children from low-complexity occupation families moving to medium- complexity occupations, but not to high-complexity jobs. Moreover, there has been no change in the probability of children from parents in moderate-level occupations obtaining high-complexity jobs. See CAF (2022) and Brunori, Ferreira, and Neidhöfer (2023). 86 Estimates are based on ECV 2019, 2021, and 2022 using an extended pool of circumstances that include the size of the born municipality, the presence of conflict in that municipality, and imputation methods for parental education level. More detail in Davalos and Monroy (forthcoming). 87 Brunori, Ferreira, and Neidhöfer (2023). 88 Ibid. CO LO M B I A 47 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Figure 39. Absolute vs. Relative Mobility in Education, 1980 cohort Colombia stands out around the world as a country with high absolute mobility yet very low relative mobility. Source: Dávalos and Monroy (forthcoming), using the Global Database of Mobility. Colombia has one of the lowest rates of relative mobility department level (focusing on attainment rather than account- in the world. Global patterns show that despite very high ab- ing for the differences in learning previously discussed) shows solute mobility in education, many countries in LAC have low an even lower standing than the national average and one that relative mobility.89 Colombia is no exception. Absolute mobility is far below the OECD average (Figures 40 and 41). To illustrate, is high: given the expansion of access to education, 78 percent for a person born in Guainía or Vaupés with parents who have a of people born in the 1980 cohort have more years of school- low level of education, the likelihood of graduating from high ing than both of their parents.90 At the same time, the country school is less than 24 percent; for someone born in Bogotá, the has very low relative mobility in both income and education same probability is 65 percent. In such departments as La Gua- compared to countries around the world (World Bank 2021). jira, Vaupés, and Guainía, low relative mobility coexists with It is an unusual case globally in its high absolute mobility but high poverty rates. Afro-descendants, indigenous populations, extremely low levels of relative mobility (Figure 39). Worrying- and people living in rural or smaller cities fare worse in social ly, income mobility tends to lag behind educational mobility.91 mobility.92 Encouragingly, however, relative mobility has been Beyond Colombia’s poor global standing in relative social on the rise since the 1950s cohort, and gaps between regions mobility, some regions within the country fare even worse. are narrowing.93 Constructing measures of relative educational mobility at the 89 World Bank, “Movilidad Intergeneracional. Monitoreo de Pobreza y Desigualdad en América Latina y el Caribe” (Washington, DC: World Bank, 2017), https:// documents1.worldbank.org/curated/en/517511508162369015/pdf/120437-REVISED-SPANISH-Intergenerational-mobility-in-LAC-October-16-2017-ESP.pdf. 90 Estimates calculated by pooling ECV 2019, 2021, and 2022 following the methodology in Van der Weide et al. (2023). Years of Schooling (YOS) variable reported for absolute mobility. 91 Narayan et al. (2018). 92 Estimates for this report suggest that indigenous people are less likely to reach educational achievements independent of their parents’ education. For instance, their relative mobility is just 75 percent of the level of non-indigenous and non-NARP (as a group of reference). 93 Absolute mobility had been increasing until the 1980s cohort, with a higher pace in places that are considerably poor, such as the Amazon, the Oriental, and even the Caribbean regions. For instance, for individuals living in departments with high poverty and born in the cohort of 1940, the absolute mobility measured as the Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 48 Figure 40. Multidimensional Poverty vs. Relative Figure 41. Relative Educational Mobility, Mobility, 1980 cohort Departments of Colombia and Selected OECD Relative mobility is particularly low in some Countries, most recent data poorer areas, such as La Guajira. Some departments in Colombia fare even worse in relative mobility. Narino Guania La Guajira Caquetá Caldas Putumayo Boyacá Colombia Bolivar Casanare Antioquia N.Santander Magdalena Cundinamarca Quindio Valle del Cauca Arauca San Andres 0,00 0,20 0,40 0,60 0,80 Source: Dávalos and Monroy (forthcoming), with estimates based on Source: Dávalos and Monroy (forthcoming), using the Global ECV 2022. Database of Mobility and the ECV survey. share of adults reaching higher educational levels than their parents, conditional on parents not having the maximum years of schooling, was below 45 percent. For the cohort of 1980, the share reached 73 percent. On the other side, the gap between regions in terms of relative mobility has shrunk along cohorts since the 1950s, though the gap remains high among regions, with the Pacific region lagging far behind. CO LO M B I A 49 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Chapter 4. The Need to Boost the Dynamics that Can Reduce Spatial Gaps Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 50 Three interrelated forces influence the widening or narrowing of territorial gaps: spatial mobility, distance, and agglomerations. Mobility, refers to workers’ ease or difficulty in moving across the territory in search of better economic opportunities. Distance involves the efficient flow of goods, services, capital, information, and ideas between places, including physical distance. Agglomerations refers to the concentration of economic activity and people in specific locations, which tends to increase productivity and economic growth.94 Shedding light on how these dynamics have operated in Colombia can help guide policy making. However, as described in this chapter, relying on these dynamics is not sufficient. Investing in people’s assets and reducing distances to new opportunities (through better physical and digital connectivity) can enable individuals to move and improve their returns and thus their incomes, and can also help to ease congestion, promote commuting, and equalize returns across the territory. Movement of people is not always in search of higher incomes The movement of people to areas with better economic op- wages. Forced displacement has impacted almost 9 percent of portunities can equalize living standards across regions by the population, making Colombia the fourth most internally improving individuals’ earning potential. Promoting the ac- displaced country in the world (Figure 42). This phenomenon cumulation of assets that allow individuals—if they so wish— disproportionately affects ethnic groups, impacting 28 percent to move in search of better opportunities can help to reduce of Afro-Colombian and indigenous populations, despite their disparities. This involves strengthening the accumulation of representing only 18 percent of the population.95 Households portable assets (such as health, education, and credit) and affected by conflict, especially those that have experienced non-portable assets (e.g., land formality) that enable greater forced displacement, face worse living conditions than the economic opportunities. However, not all migrations result in national average, with poverty rates nearly twice as high (Fig- these expected economic benefits. For example, when migra- ure 43). tion is driven by the absence of basic services, such as water or Some people do move in search of better opportunities, health care, or in response to crises like climate disasters or con- and these are mostly men and better-educated people. Still flicts, the anticipated advantages may not come to fruition. Fur- among those who have moved (9.7 percent in the past five thermore, migration opportunities are not evenly distributed; years), only one-third relocated due to employment oppor- if only those who are better-off can relocate, it could intensify tunities (Figure 44). Similarly, the impoverished population spatial polarization (Grover, Lall, and Maloney 2022). tends to migrate to accompany other household members Not everyone who moves within Colombia does so rather than to look for better jobs. Economic migrants gen- in search of better economic opportunities. With a de- erally tend to be male and better educated and also have a cades-long conflict, millions of people have moved away from higher chance of attaining salaried employment and escaping their homes to escape violence rather than to search for higher poverty (Figure 45). 94 Grover, Lall, and Maloney (2022). 95 Ibáñez and Querubín (2004). Forced displacement data up to 2003. CO LO M B I A 51 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Figure 42. Number of People Living in Conditions of Figure 43. Multidimensional Poverty Rate, Internal Displacement (millions, 2022) Households Affected by Armed Conflict and Forced Colombia has among the highest number of Displacement (%) internally displaced people. Forcibly displaced households have higher poverty rates. Number of people (millions) 8,0 6,9 Multidimensional poverty rate 25 7,0 5,9 21,4 5,7 20,5 6,0 4,8 4,5 20 5,0 3,9 3,9 3,6 3,6 3,4 4,0 15 12,9 3,0 2,0 10 1,0 0,0 5 Ethiopia Colombia Afghanistan Ukraine Nigeria Somalia Yemen Dem. Rep. Congo Sudan Syria 0 Total Households with Households with at least one at least one victim of armed victim of forced conflict displacement Source: Internal Displacement Monitoring Centre. Source: ECV 2022. Migration within Colombia often originates from the impoverished areas. Other market and institutional obstacles, poorest departments, yet asset disparities can impede the full such as land informality, may deter mobility, since households realization of migration’s potential advantages. As poverty are reluctant to abandon non-portable assets: as much as 52.7 intensifies in places of origin, the propensity to migrate increas- percent of rural land is informally held (Neva and Prada 2020), es (Figure 46). Conversely, better-off places are net receivers of with higher rates in poorer regions.96 Additionally, cultural and migrants. For numerous households, migration presents an emotional attachments, though challenging to quantify, may opportunity for higher incomes: 49.5 percent of the income dis- also serve as a migration deterrent, particularly in deeply root- parity between departments and leading cities is attributable ed communities. to variances in the returns on individual endowments, such as Equipping people with the assets they need to move, human capital (see Figure 47). Although migration barriers are if they so desire, is the main avenue to promote mobility in not formally institutionalized as in some nations, they none- Colombia. It is critical that the country invest in improving the theless exist and may disproportionately affect the poorest quality of education and job-relevant skills, particularly among households and those in the most isolated areas. These barriers the poorest and excluded groups, as this is central to its devel- may include travel expenses, skill mismatches between labor opment and growth potential and represents a step toward demand and supply, and diminished social capital in urban ar- reducing gaps between groups and territories. Advancing the eas, which restricts access to information. Enhancing people’s process of land formalization also provides households with a endowments, through improved education and skills, for in- (more) productive asset while additionally removing potential stance, could help bridge income gaps, especially in the most barriers to moving. 96 Informality of land tenure is 41.3 percent for municipalities with poverty rates below the national average and 57.3 percent for those with poverty rates above the national average. Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 52 Figure 44. Reasons for Migrating in the Past Year, Figure 45. Characteristics of Internal Migrants, by 2016−21 reason 2016−21 Fewer than a third of households migrate for Among those who migrate for work, there is a economic reasons. higher share of better-educated men. Accompanying other 53,6 Male, % 71% members of the household 41,7 Urban, % 56% 21,0 Work Poor, % 28% 34,1 4,7 High school or +, % 51% Shaping a new home 5,6 Employed, % 84% 3,9 Formal employment, % Education 32% 3,8 Salaried, % 69% 3,3 Health 3,2 Self-employed, % 27% Other reasons 13,6 Entrepreneur, % 2% 11,6 0% 20% 40% 60% 80% 0 20 40 60 Percentage of internal migrants Percentage of internal migrants Poor Non-poor Other reasons Job reason Source: Migration module of GEIH 2016−21. The surveys were pooled to Source: Migration module of GEIH 2016−21. The surveys were pooled to obtain more consistent information about the reasons for migration. obtain more consistent information about the reasons for migration. Figure 46. Internal Migration vs. Poverty at Origin Figure 47. Decomposition of Labor Income Gap97 and Destination Relative to Main Cities (Bogotá, Medellín, Cali, There is a higher propensity to migrate, as poverty Manizales, Bucaramanga), percentage points rates are higher at origin and lower at destination. Migration will not yield the expected returns to close income gaps without adequate endowments. 1 La Guajira Sucre 0,9 Probability of migrating from Córdoba the department, 2018-2022 Chocó 0,8 Cauca Norte de Santander 0,7 Magdalena Bolívar 0,6 Nariño Cesar 0,5 Santander Caquetá 0,4 Tolima Boyacá Income gap 0,3 Huila Antioquia Endowments 0,2 Caldas Valle del Cauca Returns 0,1 Atlántico Risaralda 0 Meta 20 25 30 35 40 45 50 55 60 65 70 75 80 Cundinamarca Quindío Poverty rate 2018 Poverty at origin 95% CI -50 50 150 250 350 450 Poverty at destination 95% CI Per capita labor income gap relative to cities Source: GEIH 2022; poverty map 2018 data from Acosta, Davalos, and Source: GEIH 2022. A decomposition of Oaxaca household labor income Segovia (2024). per capita was implemented using variables, such as: sex of the head, Note: The probability of migrating from the department in the previous age and age squared, household size, whether there are children in the five years is estimated using as covariates poverty at origin and household, educational level of the head, and type of employment. destination, sex, age, schooling, whether head of household or spouse, household size, and whether identified as peasant, indigenous, or NARP. 97 The gap is calculated with respect to Bogotá and the metropolitan areas of Medellin, Manizales, Cali, and Bucaramanga, which had the highest per capita income in 2022. The income gap is calculated by dividing the per capita labor income of the five metropolitan areas by that of the department, and then subtracting 1. In the departments of Antioquia, Quindío, Valle del Cauca, and Santander, the department is being compared without the respective metropolitan area to the five metropolitan areas. CO LO M B I A 53 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Distances: some places remain isolated, and climate-related hazards are bound to amplify distances Distances can put services, markets, and opportunities out of low connectivity may foster illegal economies, as 43 percent of reach for many. Digital and physical distance can make lagging municipalities with coca crops are isolated or have low acces- and remote regions less attractive for firms and limit people’s sibility (compared to 17 percent of the municipalities without access to more and better services and opportunities. coca crops).100 In Colombia, some parts of the country are currently dis- Digital technologies can help reduce some distances. connected from economic centers98 due to a lack of roads and Digital connectivity can increase access to information and other transport infrastructure. The condition of roads, partic- services, reduce trade costs, and open economic opportuni- ularly in rural areas, is poor; a fifth of the secondary roads and ties. However, there are large spatial disparities in access to one-third of the tertiary roads are inadequate and unpaved these technologies (Figure 51), with lower access in areas with (World Bank 2023a). Using census, satellite, and administrative higher travel times. Such disparities are also present between data, this study constructs categories of accessibility for each population groups (Figure 50). While 80 percent of the popu- municipality, capturing the travel time to economic nodes.99 lation in Valle and Bogotá have access to the internet, only 10 It shows that 21 percent of municipalities, hosting roughly 4 percent in Vichada y Vaupes do. Internet access is lower in the million people, have low accessibility or are isolated (Figure poorest places, and even when it is available the poorer groups 48). Distances are significantly greater in the Amazon, Orinoco, have fewer means to connect. A recent report on digital tech- Pacific, and southern Caribbean regions, where poverty rates nologies proposed that a differential regulatory framework be tend to be higher. There is a positive correlation between trav- developed to deliver connectivity to rural and remote areas in el time to economic centers and municipal multidimensional Colombia (World Bank 2023b). Even within urban areas, and poverty, capturing the lower access to services and economic particularly in the presence of high intra-urban costs, there are opportunities in the most remote areas (Figure 49). But beyond opportunities to be seized; among LAC countries, Colombia has this, limited connectivity affects regions’ economic potential. one of the highest shares of urban workers with jobs suitable For example, 40 percent of primary crops in Colombia are pro- for telecommuting but without internet access at home (Ian- duced in areas with medium-low or low accessibility. Moreover, chovichina 2024). 98 Economic centers are defined using various criteria: first, 29 department capitals (San Andres, Mitú, Agua de Dios, and Inírida) following DNP’s connectivity methodology, and 21 additional municipalities selected by: nighttime lights (among the 25 percent brightest outlier cities), categorization of high development municipalities (C and G1), and location (nearby especially isolated areas). 99 Low accessibility refers to a travel time of more than six hours to the nearest developed city (of five considered), while isolated are municipalities with no path to the five most developed cities or whose nearest road network is more than 2 kilometers away. The estimates of accessibility rely on various data sources: Customized Open Street Maps network (498.239 km road network and 19.120 km water network), 2018 DANE’s urban census area from the Marco Geoestadístico Nacional, and the 2022 composite of NASA’s black marble products (Background analysis for this report on road accessibility). 100 Background analysis for this report. Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 54 Figure 48. Accessibility Categories Based on Travel Figure 49. Travel Time (minutes) vs. Time, by municipality (darker color higher Multidimensional Poverty Rate, 2018 travel distances) The higher the travel time to economic centers, Some municipalities are quite isolated or remote the higher the multidimensional poverty rate. from economic centers. 100 90 Multidimensional poverty 2018 80 70 60 50 40 30 20 10 0 0 200 400 600 Travel time (minutes) 1. Low Acc 2. Mid Low 3. Mid High 4. High Acc Acc Acc Source: Background analysis for this report, using Open Street map, Source: Background analysis for this report, using 2018 2018 census, and 2022 NASA nighttime lights data. multidimensional poverty census-based data, Open Street map, 2018 census, and 2022 NASA nighttime lights data. Figure 50. Percentage of the Population with Figure 51. Internet Penetration, 2021 Access to the Internet Peripheral and poor parts of the country have Poor and ethnic groups, migrants, and victims of lower access to digital technologies. displacement have lower access to the internet. Poor Non-poor Male Female Rural Urban NARP Non- ethnic Indigenous gr Non-migrant Forced Internal displ. migr. International mig. 20% 30% 40% 50% 60% 70% 80% Percentage of population Source: ECV 2023. Data for forced displaced people are from 2022. Source: Data from the Ministry of Information and Communications Technologies (MinTIC). CO LO M B I A 55 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Climate-related disasters are increasing in frequency Figure 52. Accessibility Categories under a Severe and will impact physical distances. The frequency of and dam- Flood Event, compared to the Figure 48 baseline age from natural disasters have increased in Colombia since Climate-related disasters may increase 1998, with Antioquia, Putumayo, Santander, and Caldas the distances. most severely affected departments.101 Overlaying connectivity data with event data on flooding and landslides shows that the (poorer) periphery is at risk of becoming increasingly isolated due to adverse events like floods (Figure 52) and landslides. These are places with higher exposure, higher vulnerability to climate-related shocks, and higher poverty (World Bank 2023c). Climate change will also have a substantial negative effect on wages, particularly impacting poorer and rural workers more than wealthier urban ones (World Bank 2021). Source: Background analyses for this report, combining the accessibility data with data from the World Bank’s Climate Change Knowledge Portal, extreme event data on flooding and landslides; Coupled Model Intercomparison Project No. 5 (CMIP5). Note: The return period of 50 years corresponds to the probability of exceedance of 1/50, equivalent to 2 percent for any given year. The promise of cities: Cities can offer better access to services but not always higher incomes Agglomerations have the potential to provide better jobs promise does not always hold. Cities are often costly, congest- and entrepreneurship opportunities and more affordable ed, and polluted, costs that frequently undermine the benefits access to basic services. The economies of scale from increased that density and proximity can offer. Emerging research points densities facilitate the provision of public services at a lower to the role of investing in secondary cities as drivers of pover- cost, and thus the accumulation of assets such as human cap- ty reduction.104 Labor market discrimination can also limit the ital. Agglomerations also promise higher productivity and ability of certain groups to reap the benefits that agglomera- innovation, from the concurrence and flow of ideas to lower tions can provide.105 For instance, nearly half of the Venezuelan knowledge-sharing costs.102 Agglomerations have thus long migrant population in Colombia, mostly living in cities, has had been considered a path to poverty reduction.103 However, the difficulties in finding a job, one-third of whom report that it 101 Ayala-García and Ospino-Ramos (2023). 102 Grover, Lall, and Maloney (2022); and Rodríguez-Pose et al. (2024). 103 Glaeser (2012). 104 Christiaensen and Kanbur (2016). 105 Grover, Lall, and Maloney (2022). Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 56 was due to discrimination linked to their nationality.106 Similar- would benefit from the access to opportunities that larger cities can ly, there is a correlation between implicit bias by job counselors offer. Moreover, 234 municipalities are isolated or have low access in public employment services with lower referral rates for Af- to economic centers (Figure 54). The analysis finds a high positive ro-descendants to job openings.107 correlation between lower density, higher travel times, and lower In Colombia, limited accessibility hinders the benefits of education, signaling that many municipalities are simultaneously agglomerations. A significant number of very small municipalities more isolated and less populated, with a less-educated population. have poverty rates higher than the national average (Figure 53). As previously noted, distances and barriers to mobility (including Due to the small scale of these municipalities, their inhabitants low education and skills) can limit people’s access to cities. Figure 53. Number of Municipalities under 100,000 Figure 54. Number of Municipalities per Inhabitants, by poverty rate and population size Accessibility Category and Population Size Hundreds of municipalities are very small and Many small municipalities are disconnected have high poverty. from economic centers. Number of municipalities per Number of municipalities Poverty>39,3% 450 300 400 accessibility category 250 Poverty<39,3% 350 200 300 150 250 200 100 150 50 100 0 50 1-5,000 5,001-10,000 10,001-15,000 15,001-20,000 20,001-25,000 25,001-30,000 30,001-35,000 35,001-40,000 40,001-45,000 45,001-50,000 50,001-55,000 55,001-60,000 60,001-65,000 65,001-70,000 70,001-75,000 75,001-80,000 80,001-85,000 85,001-90,000 90,001-95,000 95,001-100,000 0 0-10 10-20 20-30 30-50 50-100 100-500 +500 Population size (thousands) Isolated Low Acc Mid Low Acc Mid High Acc High Acc Population size Source: Population projections and poverty map 2021 data from Source: Population projections and accessibility index for this report. Acosta, Davalos, and Segovia (2024). What is more, Colombian cities are not always prepared (World Bank 2021). Also, moves could be driven by the search to be engines of prosperity. Findings for Colombia indicate the for proximity to markets in the absence of adequate transpor- presence of “sterile agglomerations,”108 where urban growth is tation infrastructure. Bogotá, for one, is the most congested not driven by structural transformation or the typical advan- city in the world, and there is evidence of underinvestment in tages of agglomeration economies, and instead the benefits roads connecting the capital to cities in the north, northwest, are offset by the high costs.109 Increased densities in cities may and southwest (Ianchovichina 2024). This search for proximity thus be fueled by other elements that in turn could raise over- can happen even within cities at the neighborhood level in a all costs and make cities more congested. For example, the in- context where housing and inadequate infrastructure are the flux of people to cities might be driven by the search for basic primary sources of intra-urban inequality in Colombia (World services that are unavailable in localities of origin. In fact, Co- Bank 2021). The income Gini coefficient within the largest cities lombian cities seem to help reduce multidimensional poverty in Colombia ranged from 0.550 in 2002 to 0.510 in 2021. (higher services) but not monetary poverty (higher incomes) 106 DANE, “Pulso de la Migracion,” March−April 2023. 107 Duryea et al. (2024). 108 Grover and Maloney (2022). 109 Grover and Maloney (2022); and Ianchovichina (2024). CO LO M B I A 57 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Chapter 5. Charting the Trajectory: a Policy Agenda Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 58 This chapter presents policy options for increasing the access to assets and economic opportunities of disadvantaged population groups and territories in Colombia. It aims to contribute to the policy dialogue, particularly as the Colombian government has placed a high priority on inclusion and regional convergence.110 The chapter uses a territorial lens as a key contribution to the policy discourse that is also relevant to inter-group inequalities. For one, there is an overlap between groups facing deprivations (particularly ethnic groups) and a concentration in certain territories. Moreover, subnational governments can play a role in providing many of the services that all population groups, particularly the most disadvantaged, need. There is a vast array of literature that expands on this diagnostic on inequalities between population groups and the policies to address them (see Annex F). The policy options presented here draw from existing Designing policies with a spatial lens remains an un- analyses as well as from new insights.111 They are not meant tapped opportunity in Colombia. Globally, increased attention to be comprehensive, nor do they include the economy-wide has been put on designing and implementing policies that are policies necessary to propel economic growth, productivity, and better tailored to local contexts and endowments (place-sen- formalization. The latter are addressed in other reports, includ- sitive strategies), and that balance policies and priorities de- ing World Bank (2021), which also includes a discussion of so- signed at the central level with the duties, institutions, and cial assistance transfers, the forthcoming World Bank Country priorities at the local and regional levels.113 Although political Economic Memorandum, and an upcoming World Bank report jurisdictions must be taken into account, using a spatial lens on regional economic growth in Colombia.112 The latter will dive means that understanding regional challenges and opportu- deeper into, for instance, how to improve the fiscal transfer sys- nities beyond departmental or municipal borders should come tem to subnational governments. first. Once this regional context is well understood, policy mak- ers can design roles and responsibilities for different levels of Figure 55. Monetary Poverty Clusters at the government and the relevant departments and municipalities. Municipal Level, 2008−18 Patterns of convergence vary by cluster/club. Thinking spatially about development challenges calls for incorporating three main principles114 in the policy design. First, differentiated policies recognize the uniqueness of re- gional needs and challenges and move away from uniform approaches. Throughout the report, the different realities that exist within the country have been described, as well as the challenges derived from uniform policies. The notion of “development clubs” is increasingly being used to implement development strategies that are attuned to the particularities of different places. This approach could guide the complexity Source: Acosta, Davalos, and Segovia (forthcoming). level of policy interventions to ensure that they are adequate 110 Prioritized in the National Development Plan (2022−2026). 111 Including World Bank (2019, 2021, and 2023c). 112 The report will focus on regional growth convergence trends, structural transformation and its link to deforestation incentives across Colombia, the opportunities set off by trade nearshoring tendencies, and a review of the fiscal transfer system. 113 For a full review of the literature on territorial development strategies, see Rodríguez-Pose et al. (2024). 114 Ibid. CO LO M B I A 59 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T for each place.115 Such a concept not only moves away from uni- Second, coordinated policies emphasize that effective form policies, but also avoids excessively granular strategies policy execution requires coordination at various levels of (e.g., solely at the municipality scale or even below it) that fail government. This coordination ensures that strategies meet to fully seize the spillover effects of agglomerations on higher the broad development club needs while also addressing the incomes, or the potential of economies of scale in the delivery unique socioeconomic structures within each club, including of services to the population, particularly in less affluent or those of ethnic minorities. More specifically, coordination be- smaller places. In other words, atomized policies are not the tween levels of government can merge the on-the-ground in- answer either. sights of local administrations with the overarching goals and There are numerous ways to define and measure a de- perspectives of the regional and national governments. Horizon- velopment club, depending on the policy objective. Develop- tal coordination between departments and especially between ment clubs may cluster together municipalities that are alike municipalities should also be fostered. Strengthening the ability across several dimensions, such as in their poverty levels, eco- to build alliances with actors beyond government (e.g., civil soci- nomic composition, and challenges to growth. A development ety, academia, and the private sector) can help address territorial club could be constructed as a dynamic measure, incorporat- challenges in a comprehensive manner, providing valuable in- ing information from a region’s trajectory. The Development sights, additional resources, and innovative solutions. Trap Index is one such dynamic measure.116 Poverty and income Third, integrated policies go beyond purely sectoral per- convergence clusters can also provide additional insights; for spectives to comprehensively address development challeng- example, Figure 55 shows that the rate of convergence in mone- es and avoid a fragmentation of interventions. Rather than tary poverty is markedly slower in Cluster 1 compared to Cluster relying on a single development lever, integrated strategies 7.117 Giles Alvarez et al. (2024) also propose department clusters use a combination of policies that can vary by development using economic and social indicators. club. Territories and population groups face multiple barriers Lastly, it can be argued that the principle of policy dif- to higher living standards, and instead of isolated single-sec- ferentiation can be applied not only to territories, but also to tor interventions, addressing several deficiencies at the same population groups. In particular, this differentiation can inform time could have a greater impact on and success in closing op- the design of policies and programs that address the specific portunity gaps, for example, complementary interventions in demands and needs of disadvantaged groups (such as wom- infrastructure services (such as water and sanitation), health, en, indigenous groups, Afro-descendants, LGBTQI+ individu- and connectivity. This integrated approach is key to generat- als, displaced populations, conflict victims, etc.), empowering ing a sustainable and lasting impact on reducing poverty and them to have a stronger voice in decision-making processes. inequality and can be informed by measures such as multidi- At the same time, policies that are too granular or that aim at mensional poverty. It is therefore important that such sectors a level of differentiation that makes them unfeasible to imple- as education, sanitation, and health, which have important ment might be counterproductive. weight in subnational spending and are central to poverty re- duction and equity, act in an integrated manner.118 115 It is used in European Union (EU) countries, as described in Rodríguez-Pose et al. (2024). 116 Rodríguez-Pose et al. (2024) note that the index “combines three dimensions of economic dynamism (GDP per capita, productivity, and employment) and assesses regional performance on these three dimensions in relation to (i) the past performance of the region itself (normally comparing performance in the past five years relative to the previous five); and (ii) the country the region belongs to.” For the case of Colombia, a country with high informality, estimates for this report show that it makes a difference whether employment or formal employment trajectories are used. 117 Cluster 1 is predominantly comprised of municipalities situated in the least developed regions, specifically the Amazon, Caribbean, and Pacific areas, with 80 percent of its municipalities falling within these underdeveloped zones. In contrast, Cluster 7 includes municipalities located in the more developed Andean Trapeze region. Consequently, it is anticipated that differentiated policy strategies would be necessary to address the distinct conditions of these areas. It is important to note, however, that region-specific policy strategies may not be fully adequate, as municipalities within the same region may belong to different clusters with varying characteristics. 118 Zapata (2022). Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 60 This chapter presents examples in two policy areas that, place-based principles outlined above. The policy options are in their design and implementation, should incorporate the detailed below in Table 1. Policy Area 1: Investing in increasing access to productive assets for the poor and poor areas This focus on assets puts the lens on poor people’s capaci- and in rural areas can contribute to closing gender gaps, as ty to generate income and escape poverty. Policy options to women, and particularly rural women, are at a disadvantage in increase human capital include introducing core curriculum this regard (World Bank 2019). Overall, these investments can guidelines for core competencies, which would benefit places enhance the benefits from the movement of people by strength- with higher deficiencies in learning; advancing existing mul- ening their portable assets while reducing push factors for mi- tipurpose cadaster initiatives for increased access to land as a gration that are driven, for example, by the absence of basic physical asset and higher resource mobilization; promoting services in places of origin or by a response to shocks. initiatives to protect natural capital in line with the climate Policy options in this area are also consistent with in- agenda while using it sustainably to benefit rural populations creasing economic growth. The Country Economic Memoran- (consistent with recommendations for halting deforestation dum (World Bank, forthcoming) recommends, for example, in World Bank 2023c); and establishing higher connectivity to improving the quality and coverage of infrastructure, especially improve the overall access to services of remote and isolated in transportation; enhancing learning, and expanding financial places. Investments in productive assets among the poorest inclusion. Policy Area 2: Improving the institutional framework Improving the institutional framework is central to tackling at the subnational level, particularly in those cases in which spatial inequalities and inequalities between population multi-sectoral and integrated interventions are required. groups. As mentioned, institutions can shape the capacity of Moreover, this body could lead the agenda and the dialogue people to accumulate assets and local job creation. Accordingly, between the subnational and central governments and address the recent report of the Decentralization Mission119 (DNP 2024) transversal challenges for the inclusion of specific groups. Such puts its emphasis on constitutional and legislative proposals an institutional design exists in Perú, where the Secretariat of to reform the institutional framework on decentralization in Decentralization holds the authority to shape and oversee the Colombia, including the territorial financing system, compe- country’s territorial development and decentralization efforts tencies, and governance. and leads the coordination across national, regional, and local Complementary to these proposals, it is also necessary governments. to significantly strengthen coordination mechanisms at all Yet, there is currently no formal mechanism of coor- levels of government. This could be achieved through an in- dination in Colombia, and the dialogue between levels of stitutional arrangement aimed at coordinating policy priori- government often happens in informal or ad hoc ways. This ties around territorial development and territorial equity, from has limited the effectiveness of past efforts aimed at working both top-down and bottom-up perspectives. This institutional across levels of government and across sectors. In terms of its setup could facilitate the inclusion of the principles of differ- implementation, the DNP could be formally mandated with entiation, collaboration, and integration in policy making, as this role; failing that, a center of government body could be it would provide coordination for line ministries’ interventions created that reports directly to the president’s office to facili- 119 This is an initiative led by the DNP to prepare technical studies and inputs on the current state competence distribution between the national and territorial governments and to propose initiatives to reorient the decentralization model (Bateman et al. 2018). CO LO M B I A 61 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T tate coordination across all line ministries. This proposal could permanent infrastructure of shared capacities and capaci- embed or complement that of the Decentralization Mission, ty building at the subnational level (for revenue collection which proposes the creation of a permanent consultation and or investment project design, for example), rather than a technical advisory body between the national government, the focus on training the team in place at a certain moment territorial entities, and social actors.120 in time. This is also true for ethnic institutions, key actors Even in the absence of constitutional or legislative re- for the development of ethnic minorities. In line with this, forms, progress can be made in the shorter term. First, im- DNP (2024) also proposes adopting regional bodies, or proving the institutional framework calls for strengthening bodies of a technical and autonomous advisory nature, tools to think spatially in policy making. As noted, this calls for decentralization. Promoting capacity building may for establishing typologies or classifications to differentiate also include developing support and training tools for ter- national policies and regulations. It is not necessarily a matter ritorial entities with differential criteria in terms of com- of having a single one, as they will depend on the purpose of petencies, capacities, and resources. Finally, improving the regulation, policy, or intervention. Nevertheless, thinking capacities in the management of a multipurpose cadas- spatially requires looking beyond the municipal tier to achieve ter and the collection of property taxes can help improve economies of scale and to enhance relationships at the subna- fiscal management, particularly if the focus is on the main tional level that can lead to economic opportunities for lagging revenue sources (property tax and industry and commerce places, particularly as local investments in assets yield returns. tax [Impuesto de Industria y Comercio, ICA]). Even within the current normative space, there is room to im- ■ Associative schemes: Creating incentives to take advan- prove. For example, methodological mandates for the design tage of associative schemes between municipalities can of adequate typologies for distinctive policy objectives can be help seize economies of scale in management and service incorporated as part of the process of construction of, consul- delivery.121 Territorial pacts, such as the figure of the con- tation on, and approval of policy documents (National Council tract plan at the supra-municipal and supra-departmental for Economic and Social Policy of Colombia/Consejo Nacional de level, can promote collaboration and dialogue between Política Económica y Social [CONPES]). Functional subregional- actors in search of win-win scenarios. This cooperation ization exercises, such as those carried out by the DNP, can be could not only improve the efficiency and potential impact relevant inputs to guide this differentiation (Bateman et al. of policy implementation (e.g., facilitating regional infra- 2019). The Decentralization Mission’s final report (DNP 2024) structure or productive interventions) but also ensures includes analysis and proposals on new categories of territorial that resources are used more effectively and in a coordi- entities. nated manner.122 Second, the policy focus needs to also be on strength- ■ Monitoring and evaluation: Strengthening the data and ening the fiscal and management capacity of subnational mechanisms for monitoring and evaluation can improve governments, particularly among weaker entities, for better accountability for development results.123 With such a high service delivery and local development. In particular: persistence of spatial and between-group inequalities across decades, and numerous previous policy efforts to ■ Capacity building: Lessons learned from past efforts that address them, understanding what works and what does have not always succeeded suggest that the focus needs not (and why) is necessary to correct course as needed. to be on fostering sustainability in strategies of technical Yet, this has to be implemented in a way that, once more, assistance and capacity building. This means thinking of a accounts for differences in competencies, capacities, and 120 See Proposal 1 in DNP (2024). 121 See also Proposal 9 in DNP (2024). 122 DNP, GIZ, and RIMISP (2020). 123 Zapata (2022). Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 62 resources across municipalities and does not overburden the evidence in Colombia on the drivers of medium-term mon- the lower-capacity entities. etary poverty reduction (see Chapter 2) shows that increased Why the focus on these policy areas? Although economic access to services and assets and improved fiscal and institu- growth has lifted millions of families out of poverty in Colom- tional capacity in subnational governments are associated with bia, it has not managed to close the gaps between groups and cases of success. territories that have sometimes endured for decades or even Designing and implementing policies in these areas to centuries. It may do so in time, but simply waiting is not an narrow gaps between groups and territories can be done, and option. Accelerating the path to closing those gaps calls for in- there are many examples of good practices in Colombia. Some vestments in redistribution, namely, “in the form of services and of these are outlined below, aligned with the policy areas in institutions that the poor need for economic mobility.”124 In addition, Table 1. 1. Investing in increasing access to productive assets for the poor ■ Health. The Colombia: Estrategia Hospital Padrino exempli- ular focus on marginalized and vulnerable communities fies how to offer tailored support to areas of greatest need (differentiated policies). Launched in 2012 by the Ministry of in health service provision (differentiated policies). Through National Education, the program is structured to provide public-private alliances (coordinated policies), it enables an extensive support and resources to schools, teachers, and accredited hospital capable of providing highly complex students. Key components of the PTA include continuous care to accompany and technically support a hospital with professional development for teachers, the provision of lower capacity that is operating at the primary- or second- textbooks for children and teacher’s guides in language ary-care level. Scaling up telemedicine is a key component and mathematics, and on-site instructional coaching. of the strategy, through which specialized clinicians pro- The program also includes the provision of standard- vide second opinions and staff training. The results have ized formative assessments. Educational resources and been impactful. In the Hospital San José de Buga (Valle training are evidence based and subject to evaluations, del Cauca), maternal deaths were reduced to zero, and ex- aiming to improve early-grade literacy, numeracy, and treme maternal morbidity was reduced by 50 percent be- socio-emotional learning. By targeting underprivileged tween 2020 and 2023. The program has been extended to areas, the PTA seeks to ensure that students from low-in- over 100 public institutions in Valle del Cauca, Cauca, and come backgrounds have access to improved educational Nariño, providing care to more than 430 patients through opportunities. Over the years, and through continuous im- telemedicine and incorporating more than 2,000 health provements in its delivery mechanisms, the PTA has had professionals (Herrera et al. 2022). Beyond this example, positive outcomes in terms of student learning, school there are several innovative models of primary health care attendance, increased within-grade promotion, and over- in Colombia, internalizing the principles of differentiated all educational quality.125 The program has played a key care models, comprehensive care through a multi-sectoral role in narrowing the achievement gap between urban perspective, and coordinated policies between national, and rural schools, particularly in municipalities associ- local, and other stakeholders (Bonilla Torres et al. 2023). ated with the country’s development programs with a ■ Education. The All to Learn Program (Programa Todos a territorial focus (programas de desarrollo con enfoque terri- Aprender [PTA]) is a comprehensive, government-led ini- torial [PDET]). Through its comprehensive approach and tiative designed to improve foundational learning and to targeted interventions, the PTA has been instrumental in reduce learning poverty in public schools, with a partic- fostering a more equitable and effective education system 124 Page and Pande (2018). 125 World Bank and DNP (2022); and Universidad de los Andes (2017). CO LO M B I A 63 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T in Colombia, though recent changes to the program in the ment scheme that made it possible to comprehensively past couple of years might limit these promising results. address different dimensions of urban development ■ Connectivity. (integrated policies), significantly improving infrastruc- □ Ongoing efforts to improve connectivity may contrib- ture and social services in the prioritized areas. It was ute to closing gaps, but these interventions will need applied in the most vulnerable areas (differentiated to be carefully monitored and evaluated to assess their policies) and included interventions for better physical results. On rural connectivity, the government has a ro- infrastructure (including roads) and better pedestrian bust rural agenda that includes investments in roads, mobility. Community participation was an important rivers, and airports. Specifically, the Airports for Special factor in its success (coordinated policies). Air Services Strategy aims to improve airport infrastruc- ■ Inclusion. ture in the country’s most remote areas (differentiated □ The system and the blocks of care in the city of Bogotá128 policies), prioritizing the improvement of six airfields use the existing social infrastructure to provide com- in the municipalities of Chocó, Nariño, Vichada, and prehensive and complementary services (integrated pol- Guainía. This initiative seeks to improve connectivity icies) aimed at women caregivers (differentiated policies). in these regions and boost their economy, competi- It has allowed more than 400,000 women caregivers to tiveness, and tourism. Similarly, the government has receive comprehensive interventions that ranged from an ambitious railway agenda, especially the ongoing the delivery of basic services that free up care time to development of suburban rail systems, which aims to the provision of entrepreneurship, employability, and improve connectivity between major cities and their other training services that enabled more than 100,000 surrounding municipalities. One example is the on- women to get formal work in the city of Bogotá. going construction of the commuter “RegioTram de □ In the past seven years, Colombia has dealt with an Occidente” in the Bogotá-Sabana region. The latter is influx of close to 3 million Venezuelan refugees and designed to link Bogotá with its bordering municipali- migrants. The government developed two national ties, helping to address critical issues, such as road con- public policies (CONPES 3950129 and CONPES 4100130) gestion, connectivity, accessibility, and regional urban that present a coordinated, integrated, and targeted (dif- development. The national government is financing ferentiated) roadmap for integration with an emphasis this project along with the Cundinamarca government on long-term development. Results of these policies (coordinated policies).126 include the regularization of over 2 million Venezue- □ The Integral Urban Project in the city of Medellín127 im- lans,131 which has fostered the inclusion of over 1.5 mil- plemented an innovative coordination and manage- lion of them into the health social security system,132 126 Conpes 3902 2017. 127 See EDU, Medellín Mayor’s Office, “Medellín, Urban Transformation Model - Integral Urban Project - PUI in the Northeast Zone. Housing Consolidation in the Juan Bobo Ravine,” 2014. 128 The experience has been documented and has received considerable recognition. See, for example: https://oecd-opsi.org/innovations/bogota-care-blocks/; https:// www.brookings.edu/wp-content/uploads/2021/12/City-playbook_Bogota.pdf; https://www.bbc.com/mundo/articles/cqeq38je3qro; https://repositorio.cepal.org/ server/api/core/bitstreams/fc7eb717-d508-44ff-a05e-ffe4bd08bed4/content; and https://www.sdp.gov.co/transparencia/info-especifica-entidad/publicaciones/ estudios/evaluacion-del-sistema-distrital-de-cuidado-implementado-la-secretaria-distrital-de-la-mujer-marco. 129 DNP, “Documento Conpes 3950. Estrategia para la Atención de la Migración desde Venezuela,” https://colaboracion.dnp.gov.co/CDT/Conpes/ Econ%C3%B3micos/3950.pdf. 130 DNP, “Documento Conpes 4100. Estrategia para la Integración de la Población Migrante Venezolana como Factor de Desarrollo para el País,” https://colaboracion. dnp.gov.co/CDT/Conpes/Econ%C3%B3micos/4100.pdf. 131 Migración Colombia, “Estatuto Temporal de Protección,” September 2024, https://www.migracioncolombia.gov.co/etpv/etpv. 132 Ministry of Health, “National Migration and Health Observatory,” September 2024, https://www.sispro.gov.co/observatorios/onmigracionysalud/Paginas/ Observatorio-Nacional-de-Migracion-y-Salud.aspx. Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 64 among other benefits, such as access to other social tion. The District Public Policy for reception, inclusion, services, the formal employment sector, and financial and development for “new Bogotans” (international services. Similarly, in 2022, the city of Bogotá created a migrant population) (Decree 600 of 2023) established designated office to coordinate all of the city’s actions a 10-year roadmap to enhance the migrant population’s and policies regarding migrants in order to ensure a dif- access to goods and services. ferentiated, cohesive, and holistic approach to migra- 2. Improving the fiscal and management capacity of subnational governments for better service delivery and local development ■ Associativity among subnational governments. □ In turn, the Association of Municipalities of Sabana □ Experiences of supra-municipal associativity (coordi- Centro135 (Asocentro) operates in a very different con- nated policies) show its importance and potential, par- text, with a group of municipalities that are part of ticularly in the realization of economies of scale, the the Bogotá agglomeration, involving a high presence strategic formulation of the territory, and joint projects of formal economic activity, proximity, and functional or project management. The Association of Municipal- interdependence with the main national market and ities of Catatumbo133 operates in a rural area that was relatively low rates of poverty. In the case of Asocen- affected by the conflict, relatively disconnected from tro, strategic planning at the supra-municipal level the main national markets, and experiencing a high in- has allowed progress in regional projects promoting cidence of poverty. The association has obtained co-fi- innovation and income generation, the development nancing for the regional projects with the participation of territorial planning instruments or strategic projects, of municipal governments, the government of Norte de such as the management and use of solid waste, and Santander, and the national government. It also creat- attention to local needs by taking advantage of econo- ed the Provincial Center for Agribusiness Management; mies of scale, among others. facilitated the financing of the rural electrification pro- ■ Alliances. In “The Guajira Mission,” private companies gram, benefiting 16,000 rural families with access to (Grupo Aval and Promigas), the country’s Administrative energy service; and approved and financed 50 projects Department for Social Prosperity, and local communities of producer associations to strengthen agricultural and worked together (coordinated policies) to meet unmet basic livestock systems, improving the incomes of peasant needs in terms of food security, water supply, and energy families (integrated policies). Associativity has also made solutions (integrated policies). The alignment of objectives it possible to advance the strategic planning of the sub- and recognition of the potentialities and limitations of region as a whole (differentiated policies), which facili- each actor (differentiated policies) made it possible to iden- tates coordination with other levels of government and tify and produce concrete results.136 For their success, these the merging of efforts and resources.134 types of projects require incorporating the on-the-ground 133 Comprising 15 municipalities: Ábrego, Bucarasica, Cáchira, Convención, El Carmen, El Tarra, Hacarí, La Esperanza, La Playa de Belén, Ocaña, Río de Oro, San Calixto, Sardinata, Teorama, and Tibú. For a systematization of the case, see Cañizales et al. (2021). 134 For example, the construction of technical documents, such as the 2032 Regional Agenda, and the construction of a portfolio of strategic projects are highlighted. This has also been useful for the communication and visibility processes of the association and its municipalities and has facilitated dialogue not only with the national government but also with private sector actors, which has allowed, among other outcomes, the involvement of rural producers in the creation of more than 100 producer associations and 35 rural women’s associations. 135 ​ ee https://www.asocentro.gov.co/. Comprising 11 municipalities: Cajicá, Chía, Cogua, Cota, Gachancipá, Nemocón, Sopó, Tabio, Tenjo, Tocancipá, and Zipaquirá. S 136 See https://misionlaguajira.com. With this project, for example, communities, such as Parenska 1, Parenska 2, or Grasamana, have access to drinking water for the first time with interventions that include the construction of a water treatment plant, the recovery of the functionality of several wells and mills, and the drilling and commissioning of new wells. CO LO M B I A 65 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T insights of communities and subnational governments. dating of cadastral records have enabled municipalities Similarly, such tools as the implementation of “Works to increase property tax collection. As mentioned above, for Taxes”137 have shown the potential of this alignment a municipality’s ability to raise its own resources has been of interests, efforts, efficiencies, and resources between a driver of poverty reduction. Galapa saw an increase of the public and private sectors (coordinated policies). In 2023, 153 percent in property taxes and a larger share of proper- 91 projects were approved for a total value of US$667,226 ty taxes in tax revenues, increasing from 16 to 25 percent. million. This tool has allowed the growing participation Similarly, Pereira (department capital) increased property of private companies, with 75 firms involved in 2023, of taxes by 34 percent, as well as the share of property taxes which 38 were new participants. The transport sector has in tax revenues (from 39 to 41 percent).139 Key factors of been the largest recipient of resources, concentrating 45 success included the institutional will of the municipal percent of the total invested since 2018. However, other governments and the potential for improvement, given sectors, such as education, housing, and energy, have the low revenue collection in past years. also seen a significant number of projects approved (FIP ■ Strengthening tools to think spatially. A good example 2024). Despite the successes, however, there are areas for can be drawn from Antioquia. The Department of An- improvement,138 and the policies need to be carefully mon- tioquia adopted a differentiated approach (subregional- itored and evaluated. ization) to respond to the heterogeneity of realities and ■ Resource mobilization. There are instances of success potentials within the department, overcoming the frag- in the higher accumulation of physical capital (land for- mentation of efforts that a municipal perspective may malization) while increasing the capacity for resource bring. It uses subregional and regional categories and mobilization and management at the subnational level. typologies to understand territorial differences without Supplementary training in tax management and the up- having to fragment the efforts at the municipal level. 137 The mechanism allows companies to finance infrastructure and development projects in areas affected by poverty and conflict in exchange for a discount on their income tax. This scheme seeks to accelerate the execution of priority works, such as roads, schools, hospitals, and aqueducts, by allowing companies to invest directly in these communities. The government supervises and approves projects, ensuring that they meet regional development objectives and contribute to the well-being of the local population. 138 For example, a strategic reorientation to ensure that projects contribute more effectively to territorial transformation and the closing of social and economic gaps (FIP 2024). In the same vein, it is desirable to make progress in strengthening the technical and administrative capacities of subnational governments to manage and promote public-private partnership projects, thus facilitating greater participation of the private sector in the development of local and regional infrastructures. 139 https://www.galapa-atlantico.gov.co/Paginas/Home.aspx; and https://www.pereira.gov.co/publicaciones/7510/pereira-cuenta-hoy-con-20-mil-nuevos- predios-y-ocupa-el-primer-lugar-en-gestion-catastral-a-nivel-nacional/#:~:text=La%20actualizaci%C3%B3n%20del%20catastro%20aplicada,204%20mil%20 en%20la%20actualidad. Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 66 Table 1. Policy Examples to Reduce Poverty and Promote Equity across Colombia What does the Area of Main responsible Policy examples diagnostic say? intervention entities Investing in increasing access to productive assets for the poor and poor areas • Scale up the implementation of family medicine and multidisciplinary primary health care models, prioritizing rural, remote, and peri-urban areas with low access to quality primary health care. Low or no access to health services • Introduce incentives for health care professionals to work in in some areas rural and remote areas to better meet the needs of underserved (particularly poor populations. Ministry of Health and and remote), and Health • Invest in telemedicine and digital transformation more broadly Social Protection gaps in availability to improve access to specialized care for patients with chronic of health personnel, conditions in underserved territories. medicines, and financing • Strengthen local governance, regularly reporting data on health inequalities at the municipal level and requiring municipalities to develop plans to reduce health inequalities and assess their progress annually. Share best practices and scale up innovative programs. • Introduce a national science-based textbook and educational resource policy for foundational learning areas, such as language and mathematics, including a professional development program for teachers. • Link the Todos a Aprender Program (which provides pedagogical support to teachers) to the guidelines of the basic curriculum. • Introduce a diagnostic national assessment system with school- level representation for at least 1st and 2nd grades, with minimum proficiency levels clearly stated by grade and a clear mechanism to communicate the results at the school level. Very different school and teacher quality • Introduce/expand programs that teach students at the right level, across the territory with a focus on developing basic math and reading skills, based and thus gaps on student needs rather than age or grade, through, for example, Ministry of National Education tutoring, such as Compass and Let’s All Learn to Read. in learning, with Education transfers that do • Complement in-person learning with Adaptive CAL, where not compensate for connectivity is available. these needs • Orient funding toward a per-student allocation of service delivery transfers that reflects the operational cost of providing good quality education at all levels and equity concerns. • Seek economies of scale in the provision of education. For example, pool the resources of public tertiary education institutions that do not cover operating costs to avoid fragmentation and link them to results and institutional improvement plans. • Increase the supply of preschool and upper secondary education in rural areas (with positive implications for closing gender gaps from increased access to childcare). CO LO M B I A 67 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T What does the Area of Main responsible Policy examples diagnostic say? intervention entities Geographic Institute Agustín Codazzi (IGAC) Superintendence of Notaries and Registry (SNR) High land • Advance existing multipurpose cadaster initiatives to record the National Planning informality, limiting rights, restrictions, and responsibilities of land and its owners. Department the use of land as a This would allow the introduction of a property tax and strengthen Ministry of Environment productive asset and Physical capital: local fiscal performance, which is key to reducing inequality in the and Sustainable the fiscal capacity Land territories. Local finance can be strengthened by (i) improving the capacities of municipal government in managing a multipurpose Development, Unit of of subnational National Natural Parks cadaster and collecting property taxes, and (ii) deepening the governments operational linkages between cadaster and fiscal management.140 Agencia Nacional de Tierras (ANT) Subnational governments (department and municipal levels) • Ensure the continuity of national housing policies and programs and improve their targeting to facilitate households with the lowest Ministry of Housing, City income levels to access house ownership. and Territory Housing is one of the main sources • Strengthen the technical and fiscal capacity of subnational Fondo Nacional del of intra-urban governments to develop and implement housing policies and Ahorro vulnerability; programs tailored to local and regional household needs and Physical capital: economic means. Subnational secretaries some departments of housing and public Housing still with high • Enhance the capacity of middle- and low-size private sector housing enterprises vulnerability of developers to increase housing production in the least developed housing to climate areas where housing stock is notably scarce. Private and other shocks entities: Camacol, • Scale up housing improvement policies and programs and bolster Asobancaria, Family the implementation capacity of both private and public entities to Compensation Funds accelerate the reduction of the qualitative housing deficit. • Scale up incentive programs, for example, payments for environmental services, productionconservation-restoration agreements, and natural resource management models, such as Ministry of Environment community forestry and naturebased tourism, in both coastal- and Sustainable Sustainable use marine and inland territories. High deforestation Development of natural capital • Complementary policies are needed, such as an accelerated Regional Autonomous implementation of the multipurpose cadaster and land property Corporations regularization and integral rural reform, among others (see World Bank 2023c). (See above under Land policies and responsible entities.) Ministry of Transport National Roads Many municipalities • Expand connectivity from residential areas in peri-urban spaces Institute (INVIAS) are isolated or have and smaller municipalities to the tertiary and secondary road Connectivity low connectivity to network to improve overall access to jobs, housing, markets, and Subnational economic centers services, which tend to be concentrated in larger urban areas. governments (department and municipal levels) 140 Since Law 14 (1983), the government has taken the initiative to strengthen local finances by establishing a national system to link cadaster and property tax. It is important to consider the environmental priorities in territorial planning as outlined in the National Development Plan 2022-2023, related to environmental protection, natural resource management, prevention of threats and disaster risk, climate change management and food sovereignty (Article 32, Law 2294 of 2023). Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 68 What does the Area of Main responsible Policy examples diagnostic say? intervention entities • Monitor progress on human capital accumulation, social inclusion, and anti-discrimination measures, with a differentiated approach (i.e., survey and administrative data by group). • Advance design policies and programs that meet their specific Inequalities between Better targeting demands and needs, which contributes to reversing the processes groups (across of excluded and of structural discrimination. Ministry of Equality and assets and income disadvantaged • Integrate the protection of the cultural rights of ethnic minorities Equity and poverty) persist groups into investment design. • Strengthen their voice and participation in decision-making spaces, supporting the technical, financial, and organizational capacities of indigenous and Afro-descendant peoples, through their representative associations (e.g., community development). Improving the institutional framework • Adjust existing institutional arrangements (e.g., related to the Limited coordination Institutional DNP) or create new ones so that a center of government body across levels of arrangement has the formal authority and mandate to coordinate territorial President’s office government and for territorial development and interventions across sectors (multisectoral sectors development perspective) and across levels of government. • Establish definitions of the relevant typologies in the process of building policies (it can be part of the process of building and approving CONPES documents) and laws. This includes Strengthening establishing a territorial scale in the definition of clubs or Policies are typologies. A subregional scale may be more appropriate or tools to think often uniform, relevant in some cases. spatially: National Planning not recognizing development • Design policies to focus on groupings of municipalities that Department heterogeneities in clubs for policy recognize the importance of functional relationships and territorial needs and capacities differentiation interdependencies, which can be fundamental for the identification and management of “interjurisdictional facts” (Subregional Facts, Metropolitan Facts, and Regional Facts - Decree 1033 of 2021; see also Bateman et al. 2019). Ministry of Finance and Public Credit National Planning Weaknesses in • Improve municipal capacities to raise own resources, particularly in Department domestic resource the management of a multipurpose cadaster and the collection of Geographic Institute mobilization and property taxes, focusing on the main revenue sources (property tax Agustín Codazzi (IGAC) Local fiscal management, and industry and commerce tax [ICA]). performance particularly Ministry of Housing, City among the poorest • Improve subnational governments’ ability to manage resources and Territory municipalities through building management and quality spending capacity. Subnational governments (department and municipal levels) CO LO M B I A 69 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T What does the Area of Main responsible Policy examples diagnostic say? intervention entities • Build policy and management capacity through inter-municipal cooperation arrangements that consist of pooling resources and sharing technical capacity for policy and project design and implementation and sharing administrative services related to tax administration, procurement, human resources, and core public financial management functions. Focus heavily on the sustainability of the capacity-building effort. • Provide support and training tools for territorial entities to Low technical plan, design, and manage infrastructure and essential services, Ministry of Finance and capacity, particularly Institutional with differential criteria in terms of competencies, capacities, Public Credit technical and resources. Focus on the weakest entities. Exercises in among the poorest National Planning capacity methodologies and training include the Territorial Planning System municipalities Department (SisPT) on the formulation and monitoring of subnational planning instruments. • Strengthen the technical capacity of weak subnational governments to plan, design, and manage key infrastructure and services. • Translate legal reforms into concrete instruments that can guide subnational governments in the implementation of territorial laws/ decrees. • Improve the information systems and data management of National Planning Room to improve subnational governments to feed territorial planning and decision Department data and making. Monitoring and Statistics Office (DANE) monitoring, and • Strengthen monitoring and evaluation capacity, including with evaluation little information the design and implementation of compliance units, to identify Administrative about what works which interventions work and which do not, and correct course as Department of Public needed. Service National Planning • Coordinate the joint interests and efficient use of resources in the Department Partnerships event of initiatives from the national government and/or other between subnational governments. Ministry of Finance and Scope for further subnational Public Credit coordination to and national • Implement a top-down/bottom-up approach for municipalities align priorities and governments (or development clubs) to align their priorities with the territorial Departmental implementation (vertical) commitments of the National Development Plan in a differentiated governments way with the territorial scale (regional and subregional). Experiences of implementation and design of the plan contracts (in Administrative and their different versions) are relevant for this purpose. Planning Regions in Colombia (RAP) Little coordination National Planning between Department subnational Associativity • Establish regulatory, institutional, and financial incentives to governments, among promote associativity. Ministry of Interior limiting subnational opportunities governments • Implement functional territorial pacts, using the plan contract at Subnational to leverage joint (horizontal) the supra-municipal and supra-departmental levels. governments capacities and (department and economies of scale municipal levels) • Identify and map actors and capacities (organized civil society, National Planning academia, private sector, etc.). Department Missed opportunities • Align interests and possible complementarities, and prioritize Ministry of Interior to strengthen Alliances beyond topics and projects of joint importance, when relevant. Identify collaboration with government Subnational roles, ways to merge and coordinate resources and capacities, and other actors governments governance schemes for prioritized projects. (department and • Monitor and evaluate schemes. municipal levels) Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 70 Annexes CO LO M B I A 71 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Annex A. Monetary Poverty Methodology The data for the official monetary poverty methodology are of Bogotá as a reference. The use of the National Household based on the Great Integrated Household Survey (Gran En- Budget Survey for the construction of the monetary poverty cuesta Integrada de Hogares [GEIH]). As well as providing infor- and extreme monetary poverty lines implied a methodologi- mation on the labor market, this survey contains questions on cal update in 2019. household income that allow the per capita income of the ex- A new poverty series was launched by the National Ad- penditure unit to be constructed. In addition, the information ministrative Department of Statistics (Departamento Adminis- on pensions and institutional transfers in the GEIH is enriched trativo Nacional de Estadística [DANE]) in 2022 with an updated with the administrative records of the Integrated Social Secu- sampling frame from the 2018 Census, and thus far allowing rity Contribution and the different existing institutional social comparisons between 2021 and 2023. Methodological varia- assistance transfers. tions included new sample stratification among municipalities Relevant to this report’s analysis, the methodology em- and changes to sample size and distribution, among others. ploys a Spatial Price Deflator (Deflactor Espacial de Precios [DEP]). Estimates are expected to be revised backwards by DANE to For the construction of the DEP, a basket of goods is formed allow for a longer time series. The official per capita poverty that represents the common expenses in all domains, and line in 2023 was $435,375 Colombian pesos, and the per capita from the quantities acquired and their unit values, an index is extreme poverty line was $218,846.  constructed to deflate geographically. The deflator was calcu- For more information on the methodology for monetary lated using information from the National Household Budget poverty estimation, see DANE’s website on monetary poverty Survey (2016−2017) and, in accordance with the breadth of the here. domains of analysis for monetary poverty, it was carried out for the 24 urban domains and one rural domain, using the city Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 72 Annex B. Profiles of Poverty in Colombia, 2005, 2018, and 2023 (the latter not comparable)   Poverty Extreme Poverty   2005 2018 2023* 2005 2018 2023* National 45.0 27.0 33.0 13.8 7.2 11.4 Sex       Female 45.3 27.6 34.1 14.0 7.5 11.9 Male 44.6 26.4 31.9 13.6 7.0 10.9 Area       Rural 56.4 36.1 41.2 16.7 15.4 19.8 Urban 41.1 24.4 30.6 9.0 4.9 8.9 Ethnicity       Indigenous   52.3 60.1 27.8 37.1 NARP   37.1 42.9 12.2 17.2 Non-ethnic group   25.0 30.9 5.9 9.7 Age group       0−17 years 57.8 39.2 48.2 19.8 11.5 18.1 18−64 years 37.8 22.0 27.8 10.3 5.4 8.9 65+ years 35.6 18.8 24.3 10.5 5.1 8.9 Number of children and adolescents       None 16.5 9.7 15.2 3.8 2.7 5.1 One 28.7 17.8 27.9 6.0 3.8 8.1 Two 44.7 32.0 44.5 10.5 7.0 13.9 Three or more 72.5 57.3 70.8 27.5 18.6 30.5 Migration status in the past five years       Non-migrant   25.4 31.8 6.8 10.9 Internal migrant   25.7 28.4 6.6 9.9 International migrant   39.6 47.8 11.4 15.9 People in households with victims of   54.7 20.5 forced displacement** Region       Bogotá 26.6 12.4 23.7 4.7 2.5 5.5 Caribbean 55.9 38.1 46.3 16.1 9.6 18.2 CO LO M B I A 73 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T   Poverty Extreme Poverty   2005 2018 2023* 2005 2018 2023* National 45.0 27.0 33.0 13.8 7.2 11.4 Central 44.2 23.9 29.2 13.9 5.9 9.2 Eastern 48.1 25.3 28.1 16.1 6.1 9.1 Pacific 46.5 33.1 34.9 16.9 12.0 13.7 Department       Antioquia 41.8 21.2 26.4 13.9 5.6 8.2 Atlántico 48.8 24.2 29.2 9.7 2.4 6.5 Bolívar 51.9 36.2 46.6 12.3 7.0 15.9 Boyacá 58.3 26.6 31.1 27.4 5.4 12.0 Caldas 43.8 22.1 22.3 12.1 5.7 6.0 Caquetá 53.7 40.1 36.8 20.7 8.8 11.0 Cauca 56.1 50.5 45.3 27.4 22.9 18.6 Cesar 56.2 42.9 52.2 14.9 12.5 22.7 Córdoba 63.7 44.2 50.4 28.1 11.1 20.5 Cundinamarca 44.6 16.4 21.5 15.1 4.3 8.2 Chocó 74.7 61.1 67.7 44.2 34.5 43.5 Huila 56.1 37.0 37.3 21.0 9.2 11.9 La Guajira 57.8 53.7 65.3 22.2 26.7 40.6 Magdalena 57.5 46.6 47.6 14.7 14.7 17.1 Meta 38.3 25.4 26.4 11.0 7.3 8.4 Nariño 58.7 41.4 36.1 23.8 12.7 10.5 Norte de Santander 59.4 41.7 37.0 19.4 10.2 9.5 Quindío 45.6 24.1 27.2 10.4 4.4 7.8 Risaralda 35.5 17.7 27.1 8.4 3.1 6.9 Santander 41.4 20.1 29.6 11.9 3.9 8.3 Sucre 63.8 41.0 54.5 23.0 7.5 19.9 Tolima 50.7 31.0 38.1 15.9 9.2 15.8 Valle del Cauca 36.6 20.4 26.6 8.4 5.1 9.1 Source: ECH 2005; GEIH 2018 and 2023. Notes: *2005 and 2018 poverty rates are not comparable to 2023 poverty rates. Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 74 Annex C. Poverty Profiles: Conditional Regressions, 2005, 2018, and 2023 (the latter not comparable) Outcome: Probability of being in extreme Covariates Outcome: Probability of being poor poverty 2005 2018 2023 2005 2018 2023 -0.052*** -0.023*** -0.020*** -0.110*** -0.050*** -0.045*** Urban area (vs. rural) (0.002) (0.002) (0.002) (0.002) (0.001) (0.001) -0.007*** -0.010*** -0.015*** -0.008*** -0.007*** -0.010*** Male (0.002) (0.002) (0.002) (0.001) (0.001) (0.001) 0.118*** 0.150*** 0.081*** 0.122*** Indigenous population (vs. Non- ethnic groups) (0.005) (0.005) (0.003) (0.003) 0.019*** 0.059*** 0.014*** 0.036*** NARP (vs. Non-ethnic groups) (0.003) (0.003) (0.002) (0.002) 0.126*** 0.081*** 0.127*** 0.024*** 0.010*** 0.029*** One child or adolescent (vs. Childless households) (0.003) (0.002) (0.002) (0.002) (0.001) (0.001) 0.279*** 0.212*** 0.278*** 0.067*** 0.037*** 0.076*** Two children or adolescents (vs. Childless households) (0.003) (0.002) (0.003) (0.002) (0.001) (0.002) Three or more children or 0.532*** 0.426*** 0.515*** 0.203*** 0.117*** 0.192*** adolescents (vs. Childless households) (0.003) (0.003) (0.003) (0.002) (0.002) (0.003) 0.031*** 0.022*** 0.020*** 0.023*** 0.014*** 0.017*** Age group: 0−17 years old (vs. 18−64 years old) (0.002) (0.002) (0.002) (0.002) (0.001) (0.001) 0.086*** 0.060*** 0.066*** 0.048*** 0.023*** 0.041*** Age group: 65+ years old (vs. 18−64 years old) (0.004) (0.003) (0.003) (0.004) (0.002) (0.002) -0.009*** -0.026*** -0.004** -0.004* Internal migrants (vs. No migrants) (0.003) (0.003) (0.002) (0.002) 0.091*** 0.112*** 0.041*** 0.043*** International migrants (vs. No migrants) (0.005) (0.005) (0.003) (0.004) 0.171*** 0.157*** 0.108*** 0.051*** 0.027*** 0.062*** Caribbean region (vs. Bogotá) (0.003) (0.003) (0.003) (0.002) (0.002) (0.002) 0.121*** 0.095*** 0.019*** 0.057*** 0.020*** 0.019*** Central region (vs. Bogotá) (0.003) (0.003) (0.003) (0.002) (0.002) (0.002) 0.143*** 0.097*** -0.004 0.066*** 0.018*** 0.015*** Eastern region (vs. Bogotá) (0.004) (0.003) (0.003) (0.003) (0.002) (0.002) CO LO M B I A 75 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Outcome: Probability of being in extreme Covariates Outcome: Probability of being poor poverty 2005 2018 2023 2005 2018 2023 0.128*** 0.160*** 0.026*** 0.073*** 0.056*** 0.028*** Pacific region (vs. Bogotá) (0.004) (0.003) (0.004) (0.003) (0.002) (0.003) Observations 602,988 708,823 767,528 602,988 708,823 767,528 Standard errors in parenthesis *** p<0.01, ** p<0.05, * p<0.1 Source: ECH 2005, GEIH 2018, and 2023. Note: 2005 and 2018 poverty rates are not comparable to 2023 poverty rates. Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 76 Annex D. Correlates of Municipal Poverty Changes, 2008−18, 2018−21 Outcome: Total Change in Poverty Outcome: Total Change in Poverty Rate 2008−18 Rate 2018−21 Fewer than Fewer than 30,000+ 30,000+ Full sample 30,000 Full sample 30,000 inhabitants inhabitants inhabitants inhabitants Poverty rate, initial year -0.699*** -0.724*** -0.552*** -0.323*** -0.328*** -0.228*** Education: Change in the percentage of people aged -1.090*** -1.133*** -0.786*** 20 years and over with higher education 2005−18 Education: Percentage of people aged 20 years and over -0.001 -0.001 0.002* with higher education, 2018 Education: Learning poverty 0.093*** 0.080*** 0.113 0.001*** 0.001*** 0.002*** index Health: Change in infant mortality rate per 1,000 -0.021 -0.000 -0.304*** 0.000 0.000 -0.000 inhabitants (2008−18/2018−20) Labor market: Annual change in the number of social 4.756 3.434 5.229 security contributors per 1,000 inhabitants (2009−16) Labor market: Number of social security contributors per 1,000 -0.000 -0.000 -0.000 Inhabitants, 2016 Natural capital: Cumulative -0.000 -0.000 -0.000 -0.000 -0.000 -0.000 deforestation 2001−21 Financial capital: Change in the number of banking correspondents -0.000 -0.002 -0.000 0.000 -0.000 0.000 per 1,000 inhabitants (2011−18/2018/2021) Physical capital: Informality in 4.844*** 4.251*** 5.683* 0.038*** 0.041*** 0.021 the land tenancy index, 2019 Physical capital: Change in -0.055*** -0.059*** -0.047 coverage of aqueduct, 2005−18 Physical capital: Coverage of -0.000 -0.000 0.001*** aqueduct, 2018 Physical capital: Change in qualitative housing deficit, -0.024* -0.028* -0.004 2005−18 Physical capital: Qualitative -0.000 -0.000 0.000 housing deficit, 2018 Security: Change in homicides per 1,000 inhabitants 0.023 -0.018 -0.250 -0.009 -0.007 0.000 (2008−18/2018−21) CO LO M B I A 77 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Outcome: Total Change in Poverty Outcome: Total Change in Poverty Rate 2008−18 Rate 2018−21 Fewer than Fewer than 30,000+ 30,000+ Full sample 30,000 Full sample 30,000 inhabitants inhabitants inhabitants inhabitants Institutional: Municipal performance (2016, 2021) - -0.018 -0.023 0.048 -0.000 0.000 -0.001** Management component Institutional: Change in the percentage of revenues corresponding to own resources -0.026* -0.029 -0.043 -0.000 -0.000 -0.001* (2008−18/2018−21) Institutional: Change in the percentage of incomes -0.027* -0.033** -0.038 -0.000 -0.000 -0.001* corresponding to transfers (2008−18/2018−21) Accessibility: Log of travel time 0.199 0.424* 0.053 0.001 0.000 0.002 Accessibility: Internet -0.240*** -0.282** -0.111 -0.001** -0.001* -0.002 penetration (2018/2021) Agglomerations: Change in population density -4.512** -5.674** 2.423 0.093 0.119 -0.077 (2008−18/2018−21) Constant 22.881*** 25.891*** 2.846 0.059** 0.072** -0.053 Observations 1,027 827 200 1,030 829 201 R-squared 0.716 0.744 0.703 0.624 0.649 0.654 Standard errors in parentheses Department fixed effects *** p<0.01, ** p<0.05, * p<0.1 Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 78 Annex E. The State Discontinuity Index: Methodology Following Ceriani, Lopez-Calva, and Restrepo-Oyola (forthcom- State Density: The density of the state is defined at the ing), the construction of the State Discontinuity Index relies on local level as the overall extent of state effectiveness in all two measures: state effectiveness and state density. policy domains. If every dimension counts equally in defining State Effectiveness: The effectiveness of the state is de- state density, the state’s density boils down to the unweighted fined at the local level as the extent of the reach of the state’s arithmetic mean of the levels of efficiency in the different pol- intervention in a specific policy domain. Note that the specific icy areas. The paper includes a discussion on the elasticity of definition depends on the characteristics of the policy domain substitution among the different policy domains. considered and on data availability. For some areas of interven- State Discontinuity: The authors then summarize the tion, it can be expressed as the percentage of the population information on the state’s density at the subnational level in successfully reached by a service, for instance, the share of the an overall index of discontinuity of the state. If the state’s den- population living in an area connected to electricity, water, sity is the same in all regions, it means that the state has the or the sewerage system; the share of the literate population; same level of reach throughout the territory, and no region is the share of immunized children; or the share of households neglected at the expense of others. On the other hand, if a state reached by the radio or a mobile signal. For other domains of is denser in some regions than in others, this may be ultimately state intervention, it can be expressed as the frequency of a ser- symptomatic of the political will to exclude specific groups of vice (e.g., how often in the week solid waste is collected) or as the population. The discontinuity of the state is defined as the the number of service providers per inhabitant (e.g., number unequal distribution of the state’s density on the territory and of doctors per inhabitant). Other definitions of state effective- measured as the average relative density differences among ness may include the number of hours in a day a given service regions, normalized by the average density in the country. is accessible (e.g., electricity); a measure of the capillarity of Therefore, if all regions have the same density, D = 0; if the state the network of streets and railways in the region; and so forth. is absent in all regions except one, discontinuity is maximum Nevertheless, since the different domains of state intervention and D = 1. need to be subsequently aggregated, the authors introduce the concept of normalized effectiveness. CO LO M B I A 79 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Annex F. Additional Resources on Inequalities between Population Groups Gender inequalities ■ UNDP, “Ruta para el Desarrollo Territorial con Enfoque de ■ CAF, “Estudio Experimental de Género,” 2024, https://scio- Género,” 2023, https://www.undp.org/es/colombia/publi- teca.caf.com/handle/123456789/2212   caciones/ruta-desarrollo-territorial-enfoque-genero   ■ World Economic Forum, “Global Gender Gap Report 2023,” ■ UNDP, “Informe sobre Desarrollo Humano 2019” (with a 2023, https://www.weforum.org/publications/global-gen- gender focus in chapter 5), 2020, https://www.undp.org/ der-gap-report-2023/  es/colombia/publicaciones/informe-sobre-desarrollo-hu- ■ World Bank, “Colombia Gender Assessment,” mano-2019   2019, https://documents1.worldbank.org/curated/ ■ DANE and UN Women, “Mujeres y Hombres: Brechas de en/665381560750095549/pdf/Gender-Equality-in-Colom- Género en Colombia,” 2022, https://colombia.unwomen. bia-Country-Gender-Assessment.pdf org/sites/default/files/2022-11/MyH%20BrechasColom- bia-NOV5-17Nov%20%284%29.pdf   LGBTQI+ population ■ Pulso Social, “Población LGBT y Bienestar Subjetivo,” 2022, cas/2023-May-NotaEstadistica-RegistroVoluntarioLGBTI. https://www.dane.gov.co/files/investigaciones/boletines/ pdf pulso-social/presentacion_EPS_poblacion_lgbt-ene2022- ■ Andres Ham, Angela Guarin, and Juanita Ruiz, “How Ac- may2022.pdf curately are Household Surveys Measuring the Size and ■ DANE, “Mercado Laboral de la Población LGBT: Año Móvil Inequalities for the LGBT Population in Bogota, Colombia? Marzo 2023 - Febrero 2024,” 2024, https://www.dane.gov. Evidence from a List Experiment,” Inter-American Devel- co/files/operaciones/GEIH/bol-GEIHMLLGBT-mar2023- opment Bank, 2023, https://publications.iadb.org/en/ feb2024.pdf publications/english/viewer/How-Accurately-are-House- ■ DANE, “Registro Voluntario para la Visibilidad de la Di- hold-Surveys-Measuring-the-Size-and-Inequali- versidad Sexual y de Género en Colombia” (includes data ties-for-the-LGBT-Population-in-Bogota-Colombia-Evi- on poverty, employment, education, etc.), 2023, https:// dence-from-a-List-Experiment.pdf www.dane.gov.co/files/investigaciones/notas-estadisti- Indigenous and Afro-descendent populations ■ World Bank, “Latinoamérica Indígena en el Siglo XXI: Pri- entes en Colombia a partier del CNPV 2018,” 2022, https:// mera Década,” 2015, https://documents1.worldbank.org/ colombia.unfpa.org/es/publications/condiciones-de-vi- curated/en/541651467999959129/pdf/Latinoam%C3%A9r- da-y-pobreza-multidimensional-poblaciones-indige- i ca - i n d % C 3 % A D g e n a - e n - e l - s i g l o -X X I- pr i m e - nas-y-afro ra-d%C3%A9cada.pdf ■ Contraloría General de la República, “Brechas y Desigual- ■ DANE and UNFPA, “Condiciones de Vida y Pobreza Mul- dades del Sector Social en Nuestro País,” 2023, https:// tidimensional de Poblaciones Indígenas y Afrodescendi- www.economiacolombiana.co/economia/brechas- Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 80 y-desigualdades-del-sector-social-en-nuestro-pais- core/bitstreams/1a94f5e8-aed0-44ed-bcc7-8802eb56f87c/ 2752#:~:text=Por%20sexo%20y%20etnia%2C%20 content   la,centros%20poblados%20y%20rural%20disperso ■ Comisión Interamericana de Derechos Humanos, “La Situ- ■ CEPAL, “Pueblos Indígenas y Afrodescendientes de Améri- ación de las Personas Afrodescendientes en Las Américas,” ca Latina y el Caribe: Información Sociodemográfica para 2011, https://www.oas.org/es/cidh/afrodescendientes/ Políticas y Programas,” 2006, https://repositorio.cepal.org/ docs/pdf/afros_2011_esp.pdf   server/api/core/bitstreams/3aa42ba0-c591-445a-9fb4- ■ World Health Organization, “La Salud de 8bdc5e332177/content  la Población Afrodescendiente en América ■ CEPAL and UNFPA, “Afrodescendientes y la Matriz de la Latina,” 2021, https://iris.paho.org/bitstream/han- Desigualdad Socia en América Latina: Retos para la In- dle/10665.2/54503/9789275323847_spa.pdf?se - clusión,” 2020, https://repositorio.cepal.org/server/api/ quence=1&isAllowed=y  International migrants  ■ World Bank and UNHCR, “Venezuelans in Chile, Colom- the Shadows: Impacts of Amnesties in the Lives of Refu- bia, Ecuador and Peru: A Development Opportunity,” gees,” World Bank, 2022, https://ideas.repec.org/p/wbk/ 2023, https://www.jointdatacenter.org/wp-content/up- wbrwps/9928.html loads/2024/03/Venezuelan-Migrants-and-Refugees-FI- ■ Dany Bahar, Ana Maria Ibáñez, and Sandra V. Rozo, “Give NAL.pdf Me Your Tired and Your Poor: Impact of a Large-Scale Am- ■ Luz Karíne Ardila Vargas, María Eugenia Davalos, and nesty Program for Undocumented Refugees,” CESifo, 2020, Andrés Felipe García-Suaza, “Barreras a la Integración de https://ideas.repec.org/p/ces/ceswps/_8601.html los Migrantes y Refugiados Venezolanos en Colombia: ■ Dejusticia, “Migración y Trabajo Decente: Retos para el Sur Lecciones de Política de la Encuesta Pulso de Migración,” Global,” 2022, https://www.dejusticia.org/publication/mi- World Bank, 2023, https://documents1.worldbank. gracion-y-trabajo-decente-retos-para-el-sur-global/  org/curated/en/099061623144519678/pdf/P1757800d- ■ Fundación Juntos se Puede, “Análisis de la Vulnerabilidad 2ed540660bfe10d3809943ceee.pdf?_gl=1*jfudtb*_gcl_ y Brechas de Integración de la Población Migrante en Co- au*ODc3OTMxMTcxLjE3MjM2NTUzMjg lombia,” https://www.juntossepuede.co/wp-content/up- ■ DNP, “Índice de Integración” (IMI and IMIM), https://2022. loads/2021/04/Brechas-de-vulnerabilidad-3.pdf   dnp.gov.co/DNPN/observatorio-de-migracion/Paginas/ ■ International Organization for Migration, “Estudio sobre indice-de-Integracion-IMI.aspx Migrantes en Situaciónes de Vulnerabilidad: Vías y Barre- ■ Maria Jose Urbina Florez, Sandra Viviana Rozo Villarraga, ras para la Protección, Regularización y la Integración de Andrés Moya, and Ana Maria Ibáñez, “Least Protected, Víctimas de Trata y Sobrevivientes de Violencia Basada en Most Affected: Impacts of Migration Regularization Pro- Género (VBG),” 2023, https://respuestavenezolanos.iom. grams on Pandemic Resilience,” World Bank, 2023, https:// int/sites/g/files/tmzbdl526/files/documents/2023-03/Mi- ideas.repec.org/p/wbk/wbrwps/10291.html grantes-vulnerables.pdf  ■ Ana María Ibáñez, Andrés Moya, María Adelaida Orte- ga, Sandra V. Rozo, and Maria Jose Urbina, “Life out of CO LO M B I A 81 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Internally displaced populations ■ Ana Maria Ibáñez Londoño, “El Desplazamiento Forzoso en ■ Jennifer Granados Jiménez, “Las Migraciones Internas y su Colombia: Un Camino sin Retorno Hacia la Pobreza,” Uni- Relación con el Desarrollo en Colombia: Una Aproximación versidad de los Andes, 2008, https://economia.uniandes. desde Algunos Estudios no Clasificados como Migración edu.co/sites/default/files/publicaciones/libros/El-despla- Interna de los Últimos 30 Años,” Pontificia Universidad zamiento-forzoso-en-Colombia.pdf Javeriana, 2010, https://repository.javeriana.edu.co/bit- ■ Ana María Ibáñez, Andrés Moya, and Andrea Velásquez, stream/handle/10554/664/eam27.pdf?sequence=1&isAl- “Promover la Recuperación y la Resiliencia de la Población lowed=y  Desplazada: Lecciones de Colombia,” Inter-American De- ■ CEPAL, “El Impacto del Desplazamiento Forzoso en Co- velopment Bank, 2023, https://ideas.repec.org/p/idb/ lombia: Condiciones Socioeconómicas de la Población De- brikps/12683.html splazada, Vinculación a los Mercados Laborales y Políticas ■ Jorge Rodríguez and Gustavo Busso, “Migración Interna Públicas,” 2008, https://repositorio.cepal.org/items/9e8f- y Desarrollo en América Latina entre 1980 y 2005: Un Es- 3f9c-dce9-4352-b937-9617f1b3222a   tudio Comparativo con Perspectiva Regional Basado en ■ UNHCR, “Tendencias Globales: Desplazamien- Siete Países,” CEPAL, 2009, https://repositorio.cepal.org/ to Forzado en 2019,” 2020, https://www.acnur.org/ server/api/core/bitstreams/19202393-20f9-4514-b42f- media/tendencias-globales-de-desplazamiento-for- 401b71f31e2b/content   zado-en-2019#_ga=2.82678686.94517531.1592487358- 1913390598.1589481009 Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 82 References Acosta, Karina, and Jaime Bonet-Morón. 2022. “Convergencia Angela Nieto, Dora Rey, and Fabio Sánchez. 2017. “Fondo Regional en Colombia en el Siglo XXI.” Documentos de de Tierras del Acuerdo Agrario de La Habana: Estima- Trabajo sobre Economía Regional y Urbana 308, Centro ciones y Propuestas Alternativas.” Documentos CEDE 41, de Estudios Económicos Regionales, Cartagena. Centro de Estudios sobre Desarrollo Económico, Univer- Acosta, Karina, María Davalos, and Sandra Segovia. 2024. sidad de los Andes, Bogotá. https://repositorio.uniandes. “Monetary Poverty Mapping in Colombia.” Background edu.co/entities/publication/af73bb00-4fbb-4fe5-a049- paper, World Bank, Washington, DC. Unpublished. dd17c4b7f364. ———. Forthcoming. “Municipality-Level Convergence in Co- Attanasio, Orazio, and Miguel Székely. 1999. “An Asset-Based lombia.” Background paper, World Bank, Washington, DC. Approach to the Analysis of Poverty in Latin America.” Unpublished. Working Paper R-376, Latin American Research Network, Acosta, Karina, Juliana Jaramillo-Echeverri, Daniel Lasso, and Inter-American Development Bank, Washington, DC. Alejandro Sarasti-Sierra. 2024. “Informalidad Municipal Ayala-García, Jhorland. 2017. “Movilidad Social.” In Estudios So- en Colombia.” Documentos de Trabajo Sobre Economía ciales del Pacífico Colombiano, edited by Galvis-Aponte and Regional y Urbana 327, Centro de Estudios Económicos Luis Armando, 103−137. Banco de la República de Colom- Regionales, Cartagena. bia: Bogotá. Alonso-Beltrán, Sofía, Denisse Cruz-Rubí, Alejandrina Montes Ayala-García, Jhorland, and Keisy Ospino-Ramos. 2023. “Desas- León, and Juan Soriano-Escobedo. 2024. “Housing Vulner- tres Naturales en Colombia: un Análisis Regional.” Docu- ability and Flood Hazard.” Background Paper for the Re- mentos de Trabajo Sobre Economía Regional y Urbana 317, gional Study on Climate and Poverty in Latin America and Centro de Estudios Económicos Regionales, Cartagena. the Caribbean, supported by the Climate Support Facility Banca de las Oportunidades. 2023. “Reporte de Inclusión Finan- Whole-of-Economy Program, World Bank, Washington, ciera 2023.” Banca de las Oportunidades, Bogotá. DC. Unpublished. Bateman, Alfredo, Angela María Penagos, Juan Mauri- Angulo, Roberto, Dalma Ariza, Alfredo Bateman, Natalie Gó- cio Ramírez, Tomás Martin, Yadira Díaz, and Santi- mez, Jorge Iván González, Javier Pérez, Juan Mauricio ago Satizábal. 2019. “Lineamientos Conceptuales y Ramírez, et al. 2018. “Medición del Desempeño Municipal: Metodológicos para la Definición de una Subregional- Hacia una Gestión Orientada a Resultados.” Documentos ización Funcional en Colombia.” Documento de Trabajo, CEDE 38, Centro de Estudios sobre Desarrollo Económico, Centro Latinoamericano para el Desarrollo Rural (RIMISP), Universidad de los Andes, Bogotá. http://economia.uni- Departamento Nacional de Planeación, Bogotá. andes.edu.co. Bateman, Alfredo, Ángela María Penagos, Juan Mauricio Angulo, Roberto, Francisco Espinosa, Dalma Ariza, and Car- Ramírez, Tomás Martin, Yadira Díaz, Santiago Satizábal, los Felipe Reyes. 2023. “El Crecimiento y la Productividad and Ailén Espinoza. 2018. “Tipología de Subregiones como Política Social.” Documentos de Antecedentes del Funcionales para Colombia a partir de la OCDE: Metod- Informe Nacional de Desarrollo Humano 05, Inclusión ología y Resultados.” Centro Latinoamericano para el SAS, United Nations Development Programme, Bogotá. Desarrollo Rural (RIMISP), Departamento Nacional de Arjona, Ana M., Sarah Moore, and Silvia Otero-Bahamón. 2024. Planeación, Bogotá. https://rimisp.org/wp-content/files_ “Subnational Inequality after Armed Conflict in Colombia: mf/1550500505InformeTipologiaSubregionesFunciona- Mapping a Research Agenda.” Background paper, World lesparaColombia1.pdf. Bank, Washington, DC. Unpublished. Bonet, Jaime, and Adolfo Meisel-Roca. 2006. “El Legado Co- Arteaga, Julian, Carolina Castro Osorio, Diana Cuéllar, Ana lonial como Determinante del Ingreso per Cápita De- María Ibáñez, Roció Londoño, Manuel Murcia, Javier Nevá, partamental.” Documentos de Trabajo Sobre Economía CO LO M B I A 83 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Regional y Urbana 75, Centro de Estudios Económicos Re- dash.harvard.edu/bitstream/handle/1/37376059/2023-02- gionales, Cartagena. cid-wp-430-colombia-amazonia-policy-report-en.pdf?se- Bonilla Torres, Janet, Luis Gabriel Bernal Pulido, Óscar Alberto quence=1. Bernal Acevedo, and Manuela Villar Uribe. 2023. “Apren- Christiaensen, Luc, and Ravi Kanbur. 2016. “Secondary Towns dizajes Innovadores en Atención Primaria en Salud en and Poverty Reduction: Refocusing the Urbanization Colombia.” World Bank, Washington, DC. Agenda.” Policy Research Working Paper 7895, World Brunori, Paolo, Paul Hufe, and Daniel Mahler. 2023. “The Roots Bank, Washington, DC. of Inequality: Estimating Inequality of Opportunity from DANE (Departamento Administrativo Nacional de Estadísti- Regression Trees and Forests.” The Scandinavian Journal ca). 2023. “Encuesta Nacional de Calidad de Vida (ECV).” of Economics 125 (4): 900−932. https://doi.org/10.1111/ DANE, Bogotá. sjoe.12530. Dávalos, María, and J. Monroy. Forthcoming. “Inequality of Op- Brunori, Paolo, Francisco H.G. Ferreira, and Guido Neidhöfer. portunities and Intergenerational Mobility in Colombia.” 2023. “Inequality of Opportunity and Intergeneration- Background paper, World Bank, Washington, DC. Unpub- al Persistence in Latin America.” WIDER Working Paper lished. 2023/39, World Institute for Development Economic Re- Demombynes, Gabriel. 2023. “Learning Poverty at the Local search (UNU-WIDER), United Nations University, Helsinki. Level in Colombia.” Policy Research Working Paper 10498, Bussolo, Maurizio, and Luis F. Lopez-Calva. 2014. Shared Pros- World Bank, Washington, DC. perity: Paving the Way in Europe and Central Asia. Europe and DNP (Departamento Nacional de Planeación). 2024. “Informe Central Asia Studies. Washington, DC: World Bank. Final Misión De Descentralización.” DNP, Bogotá. CAF (Development Bank of Latin America). 2022. “Inherited In- DNP, GIZ (German Development Cooperation), and RIMISP equalities: The Role of Skills, Employment, and Wealth in (Centro Latinoamericano para el Desarrollo Rural). 2020. the Opportunities of New Generations.” CAF, Buenos Aires. “La Asociatividad Territorial como Mecanismo Estratégico Cañizales, Emiro, Filipe Teles, Sara Moreno Pires, Valentina de Concurrencia de Esfuerzos y Recursos.” In Reflexiones so- Ochoa, and Tomás Martín. 2021. “10 Strategic Recom- bre Ordenamiento y Desarrollo Territorial en Colombia, 785−96. mendations for Municipal Associativity: A Case Study of Bogotá: DNP. the Association of Municipalities of Catatumbo Province of Duryea, Suzanne, Jaime Millán-Quijano, Judith Morrison, Ocaña and South of Cesar.” SKL International, Stockholm. and Yanira Oviedo Gil. 2024. “Measuring Racial Bias in Carter, Michael, and Christopher Barrett. 2006. “The Econom- Employment Services in Colombia,” Working Paper IDB- ics of Poverty Traps and Persistent Poverty: An Asset-Based WP-01594, Inter-American Development Bank, Washing- Approach.” Journal of Development Studies 42 (2): 178–99. ton, DC. Ceriani, Lidia, Luis Felipe Lopez-Calva, and Samuel Re- Eguino, Huáscar, Marcelo Capello, Gabriel Farfán Mares, Juan strepo-Oyola. Forthcoming. “A Note on State Disconti- Carlos Aguilar, and Hugo Menéndez. 2020. “¿Cuán Efici- nuity.” Background paper, World Bank, Washington, DC. ente es la Gestión de La Inversión Pública Subnacional? Unpublished. Situación de los Países Federales de América Latina.” In- Cheston, Timothy, Patricio Goldstein, Timothy Freeman, Ale- ter-American Development Bank, Washington, DC. jandro Rueda-Sanz, Ricardo Hausmann, Shreyas Gadgin Elbers, Chris, Peter Lanjouw, Johan A. Mistiaen, and Berk Özler. Matha, Sebastián Bustos, Eduardo Lora, Sarah Bui, and 2007. “Reinterpreting between-Group Inequality.” The Jour- Nidhi Rao. 2023. “Seeing the Forest for More than the nal of Economic Inequality 6: 231−45. Trees: A Policy Strategy to Curb Deforestation and Ad- Fergusson, Leopoldo, Carlos Molina, James Robinson, and Juan vance Shared Prosperity in the Colombian Amazon.” CID F. Vargas. 2017. “The Long Shadow of the Past: Political Faculty Working Paper 430, Center for International De- Economy of Regional Inequality in Colombia.” Documen- velopment, Harvard University, Cambridge, MA. https:// Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 84 tos CEDE 22, Centro de Estudios sobre Desarrollo Económi- Hahn-De-Castro, Lucas Wilfried, and Adolfo Meisel-Roca. 2018. co, Universidad de los Andes, Bogotá. “La Desigualdad Económica entre las Regiones de Colom- Fernández, Cristina, and Alexander Sarango. 2023. “Informali- bia, 1926-2016.” Cuadernos de Historia Económica 47, Ban- dad, Productividad e Inequidad. Un Análisis desde el Pun- co de la República, Bogotá. to de Vista de las Firmas y los Trabajadores.” Documentos Herrera, Cristian A., Jeremy Veillard, Nicole Feune de Colombi, de Antecedentes del Informe Nacional de Desarrollo Hu- Sven Neelsen, Geoff Anderson, and Katherine Ward. 2022. mano 2024, United Nations Development Programme, “Construyendo Sistemas de Salud Resilientes en América Bogotá. Latina y el Caribe: Lecciones Aprendidas de la Pandemia Ferreira, Francisco H. G., and Jérémie Gignoux. 2011. “The Mea- de COVID-19.” Washington, DC: World Bank. surement of Inequality of Opportunity: Theory and an Ap- Ianchovichina, Elena. 2024. The Evolving Geography of Productivi- plication to Latin America.” Review of Income and Wealth 57 ty and Employment: Ideas for Inclusive Growth through a Terri- (4): 622−57. torial Lens in Latin America and the Caribbean. Latin American FIP (Fundación Ideas para la Paz). 2024. “Obras por Impuestos: and Caribbean Studies. Washington, DC: World Bank. Medida para Transformar las Zonas más Afectadas por el Ibáñez, Ana María, and Pablo Querubín. 2004. “Acceso a Tier- Conflicto Armado en Colombia ¿Mantenerlo? ¿Para qué?” ras y Desplazamiento Forzado en Colombia.” Documento Análisis de Coyuntura, FIP, Bogotá. https://multimedia. CEDE 23, Centro de Estudios sobre Desarrollo Económico, ideaspaz.org/media/website/FIP_ObrasxImpuestos.pdf. Universidad de los Andes, Bogotá. https://ideas.repec. Gallup, John Luke, Alejandro Gaviria, and Eduardo Lora. 2003. Is org/p/col/000089/001940.html. Geography Destiny? Lessons from Latin America. Washington, Ibáñez, Ana María. 2016. “El Proceso de Paz con las FARC: ¿Una DC: Inter-American Development Bank. Oportunidad para Reducir la Pobreza Rural y Aumentar la Galvis-Aponte, Luis Armando, Wendy Galvis-Larios, and Lucas Productividad Agropecuaria?” Revista de Ingeniería: 8−13. Wilfried Hahn-De-Castro. 2017. “Una Revisión de los Estu- Ibáñez, Ana María, and Juan Carlos Muñoz-Mora. 2011. “The dios de Convergencia Regional en Colombia.” Documentos Persistence of Land Concentration in Colombia: What de Trabajo sobre Economía Regional y Urbana 264, Centro Happened between 2000 and 2010?” In Distributive Justice de Estudios Económicos Regionales, Cartagena. in Transition, edited by Morten Bergsmo and Cesar Rodri- Giles Alvarez, Laura, Mónica Hernández Florez, Cristhian Lar- guez-Garavito, 279−307. Torkel Opsahl Academic Publish- rahondo, Juan Carlos Muñoz-Mora, Germán David Angulo, er. and Laura María Quintero. 2024. “Desigualdades Territori- IGAC (Instituto Geográfico Agustín Codazzi). 2023. Frag- ales en Colombia: Realidades y Perspectivas.” Inter-Ameri- mentación y Distribución de la Propiedad Rural en Colombia. can Development Bank, Washington, DC. Bogotá: IGAC. Glaeser, Edward. 2012. Triumph of the City. London: Pan Books. Inclusión SAS, and UN (United Nations). 2023. “Inclusión Pro- Gómez-Salcedo, Mónica Sofía, Luis Armando Galvis-Aponte, ductiva en Colombia: Mediciones y Marco de Política.” and Vicente Royuela. 2017. “Quality of Work Life in Colom- United Nations Colombia, Bogotá. bia: A Multidimensional Fuzzy Indicator.” Social Indicators López-Murcia, Julián D. 2022. Recentralisation in Colombia. Lon- Research 130 (3): 911–36. don: Palgrave Macmillan. Grover, Arti Goswami, and William F. Maloney. 2022. “Proximi- Maloney, William F., and Felipe Valencia-Caicedo. 2016. “The ty without Productivity: Measuring Agglomeration Effects Persistence of (Subnational) Fortune.” The Economic Journal with Plant-level Output and Price Data.” Policy Research 126 (598): 2363–401. Working Paper 9977, World Bank, Washington, DC. Misión para la Transformación del Campo. 2014. “Diagnóstico Grover, Arti, Somik V. Lall, and William F. Maloney. 2022. Place, de las Condiciones Sociales del Campo Colombiano.” De- Productivity, and Prosperity: Revisiting Spatially Targeted Poli- partamento Nacional de Planeación, Bogotá. cies for Regional Development. Washington, DC: World Bank. CO LO M B I A 85 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Narayan, Ambar, Roy Van der Weide, Alexandru Cojocaru, Rincón-Barajas, Jorge A. 2023. “Productivity Dynamics and Christoph Lakner, Silvia Redaelli, Daniel Gerszon Mahler, State Support after a Land Titling Program: Evidence from Rakesh Gupta Nichanametla Ramasubbaiah, and Stefan Colombia.” Land Use Policy 131. Hubert Thewissen. 2018. Fair Progress? Economic Mobility Robledo-Caicedo, Jilmar. 2019. “¿A Dónde se Fue la Fortuna? across Generations around the World. Washington, DC: World Historia Económica y Social del Chocó, Colombia.” Cuad- Bank. ernos de Historia Económica 52, Banco de la Republica de Neidhöfer, Guido, Joaquín Serrano, and Leonardo Gasparini. Colombia, Bogotá. 2018. “Educational Inequality and Intergenerational Mo- Rodríguez-Pose, Andrés, Federico Bartalucci, Nancy Loza- bility in Latin America: A New Database.” Journal of Devel- no-Gracia, and María Eugenia Davalos. 2024. “Overcoming opment Economics 134: 329−49. https://doi.org/10.1016/j. Left-Behindedness: Moving beyond the Efficiency versus jdeveco.2018.05.016. Equity Debate in Territorial Development.” Policy Research Neva, Nelson, and Ricardo Prada. 2020. “Indicador de Infor- Working Paper 10734, World Bank, Washington, DC. malidad en la Tenencia de la Tierra en Colombia Vigencia UNDP (United Nations Development Programme). 2024a. 2019.” Índice de Informalidad, Unidad de Planificación “Colombia at Crossroads: Navigating Regional Divides and Rural Agropecuaria, Bogotá. Embracing Opportunities.” National Human Development OECD (Organisation for Economic Co-operation and Develop- Report, Executive Summary, UNDP, New York. ment). 2022. OECD Regions and Cities at a Glance 2022. Paris: ———. 2024b. Educación: Motor de Igualdad, Crecimiento y De- OECD. https://doi.org/10.1787/14108660-en. sarrollo Humano. Cuaderno 4, Informe sobre Desarrollo Oxfam. 2017. “Radiografía de la Desigualdad: Lo que nos Dice Humano para Colombia. Bogotá: UNDP Colombia. el Último Censo Agropecuario sobre la Distribución de la Universidad de los Andes. 2017. “Evaluación de Impacto Todos Tierra en Colombia.” Oxfam en Colombia, Bogotá. a Aprender: Resultados 2014 2016; Convenio 163 de 2014, Page, Lucy, and Rohini Pande. 2018. “Ending Global Poverty: Convenio 834 de 2015, y Convenio 1109 de 2016.” Universi- Why Money Isn’t Enough.” Journal of Economic Perspectives dad de los Andes, Bogotá. 32 (4): 173–200. Van der Weide, Roy, Christoph Lakner, Daniel Gerszon Mahler, Prem, Mounu, Santiago Saavedra, and Juan F. Vargas. 2019. Ambar Narayan, and Rakesh Gupta. 2023. “Intergener- “End-of-Conflict Deforestation: Evidence from Colom- ational Mobility around the World: A New Database.” bia’s Peace Agreement.” HiCN Working Paper 288, HiCN Journal of Development Economics 166 (2). https://doi. Households in Conflict Network, Institute of Development org/10.1016/j.jdeveco.2023.103167. Studies, University of Sussex, Brighton, UK. Vega-Landaeta, Angela, Gisela Garcia, and Jeremy Veillard. Ramírez, Juan Mauricio, Yadira Díaz, and Juan Guillermo Be- Forthcoming. “Mortalidad Evitable a Escala Municipal en doya. 2017. “Property Tax Revenues and Multidimensional Colombia.” Background paper, World Bank, Washington, Poverty Reduction in Colombia: A Spatial Approach.” World DC. Unpublished. Development 94: 406−421. World Bank. 2004. “Colombia: Una Política de Tierras en Tran- Rincón, Alexander, Diana Lara, Mario Pérez and Johnny Harold sición.” Documentos CEDE 29, Centro de Estudios sobre Rojas. 2016. “Aproximación a los Conflictos por los Recur- Desarrollo Económico, Universidad de los Andes, Bogotá. sos Naturales.” In Biodiversidad 2015. Estado y Tendencias de https://repositorio.uniandes.edu.co/server/api/core/bit- la Biodiversidad Continental de Colombia, edited by María streams/58af55bd-2ae5-49c4-80d4-d2b874e6800d/con- Fernanda Gómez, Luz Adriana Moreno, German Ignacio tent. Andrade, and Cristina Rueda-Uribe. Bogotá: Instituto Al- ———. 2019. “Colombia Gender Assessment.” World Bank, exander Von Humboldt. Washington, DC. ———. 2021. “Building an Equitable Society in Colombia.” World Bank, Washington, DC. Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 86 ———. 2023a. “Adaptive Social Protection in Colombia: What ———. 2024b. World Development Report 2024: The Middle-In- is the Distributional Impact of Shocks and How Can the come Trap. Washington, DC: World Bank. Social Protection System Better Respond? An Application ———. Forthcoming. “Colombia Country Economic Memoran- of the Social Protection Stress Test Tool.” World Bank, dum: Growing for Prosperity.” World Bank, Washington, Washington, DC. DC. ———. 2023b. “Cierre de Brecha Digital en el Departamento World Bank, and DNP (Departamento Nacional de Planeación). Amazonas.” World Bank, Washington, DC. 2022. “Evaluación Cuasiexperimental de Impacto del ———. 2023c. “Colombia: Country Climate and Development Programa Todos a Aprender en el Marco de la Pandemia Report.” World Bank, Washington, DC. COVID-19.” World Bank, Washington, DC ———. 2024a. “Effects of Climate Change and Possible Ex- Zapata, Juan G. 2022. “¿Qué Hacer en Desarrollo Territorial?” treme Events on Colombian Accessibility.” Background In ¿Qué Hacer en Políticas Públicas?, edited by Luis. F. Mejía, paper, World Bank, Washington, DC. 69−72. Bogotá: Fedesarrollo. CO LO M B I A 87 P OVE R T Y A N D E Q U I T Y A S S E S S M E N T Trajectories: Prosperity and Poverty Reduction in the Colombian Territory 88