INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Ashw n Re h Seb st n V v n M r Eu en Pere o Ju n C rlos Muñoz Mor 1 © 2020 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpre- tations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publi- cations, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Cover and section images under Standard license - AdobeStock. INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR EXECUTIVE SUMMARY B y the end of August 2020, five years since the This report is intended to reach a broad audience of policy intensification of the Venezuelan humanitarian crisis, makers, program administrators, development professionals, 5.2 million Venezuelans had fled their country, in an and academics in Colombia and in the broader development exodus whose scale and pace closely mirror those of the community, and aims to assesses the integration of Syrian refugee crisis – where by 2015, four years into the Venezuelan migrants into Colombian agri-food labor forced displacement crisis, 4.8 million people had escaped markets through a combination of original micro-level data Syria. The recent crisis in Venezuela, resulting in the forced analysis and in-depth semi-structured field interviews displacement of millions of migrants, continues to place a with Venezuelan migrants, producers’ associations, and burden on the financial, economic, and social systems of Colombian institutions. The main contributions of the study neighboring Colombia, as the host economy that has received are three-fold. the largest share of migrants. This represents a reversal of the migratory trends of the recent past, which saw millions First, the report offers a detailed overview of Venezuelan of Colombians move and settle in oil-rich Venezuela, migration into Colombia, spatially and over time, beginning in the 1960’s through Colombia’s 50-year armed enriching with new, and more detailed, insights conflict. Today, many of those Colombian nationals are the currently available information on migrants’ returning to their country of origin, accompanied by many employment outcomes and on their comparison to those more Venezuelans who are fleeing deprivation, threat, and of the local Colombian population. The exercise confirms insecurity in their country: the suddenness and quickness the exponential increase in legal visitors from Venezuela at which these forcibly displaced migrants have been through formal entry ports starting from the year 2017, with crossing over from Venezuela to Colombia is unprecedented 1.3 million entries in just the first quarter of 2018: among in the region. The Venezuelan migration crisis threatens these visitors, most are prime-aged men between the ages to pose serious challenges to Colombia’s finances, welfare of 18 and 39. In the wake of the migration crisis, 90 percent schemes, and social capital, which could be even further of Venezuelan visitors entered Colombia on tourist visas: aggravated as the country is confronted with other major the number of self-reported “tourists” increased 2.5 times developments, such as its own peace transition after between 2016 and 2017, and by 2018 it almost doubled decades of internal conflict or the recent unfolding of the again. Since visitors on tourist visas are not allowed to work COVID-19 pandemic.1 in Colombia, these numbers suggest that many of entrants are probably joining the informal labor force to provide In the immediate aftermath of the surge in the number for their livelihoods. Given that official statistics represent of Venezuelan migrants, the focus of the Colombian only a share of all migrants crossing into Colombia, as government was to register all migrants and provide they do not account for undocumented migration, it is relief through health and welfare systems. Nevertheless, expected that labor informality among migrants may be as time has continued to elapse, the focus has shifted to even more pervasive. Mapping traveler hotspots shows mechanisms for accommodating the wave of migrants, their clustering in frontier regions located at the Venezuelan developing strategies for their successful, sustainable and border, and in already highly populated urban centers such strategic integration into Colombia’s economy. While this as Bogotá, Medellín, and Cali. Stronger social networks and could be seen as a challenge for the economy at present, migrant communities, as well as ease of access via major it need not be viewed that way if timely measures are transportation routes, can explain why migrants chose to implemented through a longer-term, territorial development locate in urban centers. lens. Venezuelan migrants can represent a valuable addition to the Colombian labor force, bringing new skills and filling To obtain more detailed insights on migrants’ labor-market labor gaps in specific territories. This is especially true for outcomes, the report exploits several rounds of the national Colombia’s agriculture and food systems, which face a labor force survey, analyzing data on more than 2 million growing demand both locally and internationally, but at individuals in total, which allow to capture nuances in the same time are confronted with low productivity and migrant profiles that would not be available in the aggregate labor shortages. statistics obtainable from administrative data sources. The analysis shows that, prior to the surge in Venezuelan migration, migrants and the local population in Colombia 1  Other than representing a tragic health crisis, the COVID-19 pandemic faced similar employment rates, around 60 percent. What will have dramatic socio-economic implications that will affect and to a large is remarkable, however, is that the employment rate of extent are already affecting migrant integration as well. This report, which Venezuelan migrants was actually at its highest levels during was largely finalized prior to the onset of the disease, does however not seek to the height of the migration crisis. Contrary to widespread address the issues related to the pandemic. expectation, the intensification of the migration crisis 3 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR did not substantially affect migrants’ likelihood to find Although migrants have concentrated in urban areas taking employment in Colombia. The complication lies in the quality up sub-optimal food service jobs, they could potentially of jobs migrants are finding, and the related employment be better off working in different sub-sectors and in rural conditions: the jobs that Venezuelans are finding tend to areas: migrants that are employed in rural areas on average be of lower quality if compared to the local population, in encounter better-quality jobs in terms of wages, working terms of wages, working hours, and informality rates. After hours, and informality. Given still high unemployment rates the surge in migration in 2017, Venezuelan migrants earn 22.7 among Venezuelan migrants, the poor labor conditions of percent less than the average wage, work 3.2 more hours per those working in urban areas and the pressures exerted on week than local Colombians, and are 15 percentage points urban systems by sudden and dramatic population increases, more likely than locals to work without a formal contract. the integration of migrants into rural-agroindustrial These results are even more striking considering that a large clusters may represent an important opportunity. In this portion of Venezuelan migrants entering Colombia during sense, the analysis shows encouraging signs that, after the migratory crisis were typically highly educated, with 2017, Venezuelan migrants have become more likely to be a higher share of individuals with at least high school or employed in agriculture and food processing in areas with a tertiary education than the working-age local population. higher concentration of agro-industrial firms. The first part of the report therefore deduces that the average Venezuelan migrant is not really facing better The third and final contribution of the report is to employment prospects in Colombia in the aftermath of the identify lessons learned for the inclusion of Venezuelan crisis, but rather that they are more willing to take up a migrants in the agri-food sector in Colombia. The job – any job, to survive. recommendations flow from the findings of the data analysis, and also draw on two case-studies of successful A second contribution of the report is to provide labor market inclusion programs for Venezuelan migrants evidence that the agri-food sector in Colombia has and vulnerable populations in the Colombian coffee and a yet unfulfilled potential to support a smoother cut flowers value chains. The fact that areas with a stronger inclusion of Venezuelan migrants in the labor force. The presence of agro-industry see more Venezuelans employed study characterizes agri-food job markets in Colombia at in agriculture and food processing shows that migrants are different stages of the food chain (primary production, not necessarily averse to taking up jobs in these occupations, agri-food manufacturing, logistics and retailing, food if the opportunity arises and the right conditions are in services), and identifies promising opportunities for the place. The case studies shed light on the fact that vulnerable agriculture sector in Colombia to expand and absorb new populations, including migrants, are more likely to take up workers, due to the co-existence of unmet labor demand in agri-food jobs if specific active labor market mechanisms primary activities, on one side, and the growing importance are activated, including information campaigns, systematic of downstream activities in the food value chain, on job advertisements, and mobility incentives for facilitating the other. migrant integration into agriculture-linked employment. Similarly, they are more likely to remain in those jobs if Migrants, in fact, are already 5 percentage points more employment is supported by a benefit package including likely to be employed in agri-food activities than the health, transportation services, education, parental support, local population, and the likelihood has even grown by an and accommodation. additional 3 percentage points after 2017. Nevertheless, unpacking this observation along the value chain reveals The report concludes with a look at the path ahead, that this happens almost exclusively in food service through practical ideas and operationalization principles occupations and in urban areas. Crucially, migrants are not for delivering a strategy that includes both supply and more likely to work in primary agriculture even in areas that demand driven integration of migrants in labor markets, are predominantly agricultural, despite documented labor featuring agriculture and food systems more prominently. shortages in some key primary occupations such as coffee These principles highlight the need for a coordinated, picking. Furthermore, it is remarkable that migrants are not programmatic strategy that involves inter-institutional more systematically represented in better-remunerated and cooperation, incentives for agri-business, and a multi- more dynamic value-addition activities downstream in the faceted approach to address the specific needs of migrants in food chain (such as food processing or formal food services), agriculture employment. given that their skill set is on average well aligned with that of food processing workers. In line with these observations, The proposed operationalization principles for migrant job quality in the agri-food sector is still significantly worse integration into agriculture and food systems are as follows: for migrants than for the local population: on average, Venezuelans are hired more informally, work 5 more hours » Strengthen “labor pull” actions for agri-food jobs. A per week, and receive 25 percent lower wages than their targeted agri-food sector strategy for job search and job Colombian counterparts. placement of Venezuelan migrants in rural areas, where migrant to local ratios are low and labor shortages are 4 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR observed in the sector, can benefit migrants. This could and guidance on hiring migrants, as well as streamlined include job fairs, specialized training for highly demanded hiring processes could provide the impetus for employers skills in agri-food jobs, and placement throughout in the agri-food sector to process migrant workers as value chains from farm, to food manufacturing, to employees. Third, incentives for agribusiness for hiring warehouses, to transporters and to distributers. Skills Venezuelan migrants including tax breaks and appealing matching algorithms could be created to match higher to corporate social responsibility programs with the skilled workers with nonroutine, cognitive agri-food promotion of social responsibility certifications can sector jobs. Providing young migrants with access to promote the use of migrants. technology-based jobs with opportunities to leverage innovation could provide a rural, agri-food based pull, » Streamline the institutional approach. The public higher quality income opportunities, and alleviate sector, institutional approach to Venezuelan migrants migrant linked pressures in urban centers. could be strengthened by broadening the scope of the CONPES 3950 strategy to include participation of the » Make the agri-food sector and the rural environment Ministry of Agriculture and Rural Development more attractive for migrants. Investing in strengthening centrally through policies and programs. This includes services and social protection accessible to migrants, and specific provisions for the inclusion of migrants in advertising smaller rural economies among migrants, agriculture and food systems, especially in rural areas. could provide an incentive for moving to these areas. The report recommends an ambitious initiative, creating Benefits like health insurance, social protection schemes, targeted employer-bases PEP schemes specifically transportation, accommodation, study benefits, parental directed at jobs in the agri-food sector. help, permanency, and security for migrants who may have otherwise relied on strong social networks in larger Operationalizing the above proposed principles requires urban areas could tip migrant preference towards rural a spatial territorial development lens, and local context areas. Local authorities could also provide relocation specific customization that takes into account the services to support migrant transitions to these areas. geography of migrants’ distribution across the country, Options include (i) creating rural hubs for migrants, local to migrant population ratios, attitudes to migrant to ensure migrants are not scattered, have dedicated integration and local labor demand. Any measures should service delivery help, and feel more integrated into be considered as longer-term solutions that give migrants communities; (ii) replicating, for the migrants’ case, the the opportunity to assimilate to the socio-cultural context structure of social programs currently targeting other of local communities. This effort would require national and vulnerable groups. sub-national institutional participation and cooperation, to ensure that locals and migrants are all winners and » Smooth labor demand frictions by channeling inclusion strategies are mutually beneficial opportunities information to agribusiness employers and for stimulation of local rural economies, job creation and providing business incentives. First, information improved welfare. asymmetries could prevent Colombian employers with high labor demand and labor shortages, but little supply side information on the attributes of migrant workers from hiring migrants. Providing agribusiness with list and availability of migrant worker pool would smooth demand frictions. Second, clear information 5 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Box ES1 COVID-19 and Venezuelan displacement in Colombia The COVID-19 health emergency is creating further challenges to an already complex situation, exacerbating the vulnerability of Venezuelan refugees and migrants and at the same time hampering the ability of host countries to cater to the needs of the displaced population – by slowing local economies, causing job frictions especially in informal markets, and exacerbating social tensions and anti-migration sentiments. This report was finalized before the contagion spread to Colombia, so the empirical analysis does not quantify the effect of the pandemic on the migrants’ integration. Nonetheless, evidence is emerging on the deep challenges that Venezuelan refugees and migrants in the country are increasingly facing due to the COVID-19 outbreak and response. Precarious livelihoods and informality are making Venezuelans even more vulnerable to the current context, in particular in terms of job loss, access to food, and personal safety. The economic contraction has especially impacted those working in the informal sector, in the service sector (especially hospitality), and in construction – all sectors where Venezuela migrants and refugees are disproportionately represented. Many workers have lost their jobs, are on unpaid leave, or have seen a drastic reduction in their working hours, which is dramatically increasing unemployment and underemployment, and pushing many migrant and refugee families in poverty. Furthermore, the invisibility generated by high informality compounded by social distancing and mobility restrictions creates serious information gaps that stifle the ability of existing public programs to reach refugees and migrants.2 Over half of Venezuelan households interviewed for a Rapid Needs Assessment carried out in April 2020 by the national coordination platform (Grupo Interagencial sobre Flujos Migratorios Mixtos, GIFMM) in Colombia reported problems complying with quarantine measures, mainly due to the need to generate income to cover their basic needs. Roughly 95 percent of interviewees ranked food among their main concerns in the context of COVID-19, 53 percent expressed concerns with housing, 45 percent with finding employment and 26 percent with receiving medical attention.3 The difficulty in attending to their basic needs has been pushing an increasing number of Venezuelans to cross the border back into their home country, despite unchanged socio-economic challenges and threats to their safety and health. At the end of April 2020, around 12,000 Venezuelans had crossed the border back into their home country, where they face confinement and worrying conditions in improvised camps lacking basic hygiene and safety standards. The health emergency is profoundly affecting the entire economy with changes that will likely be long-lasting, including in the agri-food sector, and it is highlighting even more painfully the need for inclusive solutions to the Venezuelan displacement crisis, now and in future. The findings of this report are likely to be exacerbated by COVID-19 and the recommended operationalization principles will be essential to prevent an undoing of Colombia’s huge efforts to accommodate Venezuelan migrants. Mechanisms to smooth refugees’ and migrants’ integration in rural areas and in Colombia’s agriculture and food systems, such as the ones proposed in this report, can decongest urban spaces, ease pressure on local economies, and provide essential livelihood support through the creation of more and better jobs. 2  Coordination Platform for Refugees and Migrants in Venezuela. 2020. COVID19: Venezuelan Refugee And Migrant Response Plan 2020. May 2020 Revision. R4V.info 3  https://r4v.info/en/documents/details/76031 6 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR ACKNOWLEDGEMENTS T his report was prepared by Ashwini Rekha Sebastian Desarrollo Rural Sostenible), Andrés Silva, Asmed Orozco, and Viviana Maria Eugenia Perego (Economists, Erika Donoso, Ruth Mary Ibarra (Directorate of Value Chains, Agriculture and Food Global Practice of the World Ministry of Agriculture and Rural Development), Milton Bank) and Juan Carlos Muñoz Mora (Professor, EAFIT Mora (Ministry of Labor, Registro Único de Trabajadores University, Medellín). The team benefited from the guidance Extranjeros de Colombia), Claudia Rangel (Servicio Nacional and support of Elena Ianchovichina (Lead Economist, Office de Aprendizaje), Humberto Oliveira (Inter-American Institute of the Chief Economist for Latin America and Caribbean), for the Cooperation in Agriculture), Marcos Rodriguez (FAO Preeti Ahuja (Practice Manager, Agriculture and Food Colombia), Rocio Castañeda (UN Refugee Agency), Miguel Global Practice of the World Bank), and Luz Berania Díaz Pinedo and Angela Betancourth (3er planeta Networking), Ríos (Senior Agribusiness Specialist, Agriculture and Food César Eduardo Tamayo Tobón (EAFIT University, Medellín), Global Practice of the World Bank). The team acknowledges Marcela Eslava (Los Andes Univesity, Bogotá), Marcela valuable contributions and excellent research assistance Meléndez (Econestudio), Maria Teresa Palacios Sanabria, from Vileydy Adriana Gonzalez Mejia (EAFIT University, Maria Lucia Torres Villarreal, Paola Marcela Iregui Parra Medellín). Background research and case study analysis for (Universidad del Rosario, Bogotá), Felipe Campo Dominguez Chapter VI on existing integration policies for Venezuelan (CUSO International), Roberto Velez (National Federation of migrants in the agriculture sector in Colombia was Coffee Growers), Kateryn Mejía (Association of Colombian performed by Lorena Martínez Gómez and Maria del Mar Flower Exporters), Mónica Greiff (Bogotá Chamber of Polo (Food and Agriculture Organization). Commerce), Pablo Prieto (SunShine Bouquets), Antonio Gómez (Unión de Trabajadores Venezolanos y Colombo- Within the World Bank, William Wiseman (Program Leader, Venezolanos), Oscar Romero Navas (Bancamia), the Grupo LCC1), Dorte Verner (Lead Agriculture Economist, Agriculture Interagencial sobre Flujos Migratorios Mixtos, the delegates and Food Global Practice), Gabriel Lara Ibarra (Senior of the Coffee Committees of the city of Pereira in Risaralda, Economist, Poverty and Equity Global Practice) served as Santa Rosa de Cabal, Belén de Umbría, Sanctuary, in Norte peer reviewers and contributed valuable comments on the de Santander and Chinacota. report. The team is also grateful to Michael Morris, Ulrich Zachau, Taimur Samad, Andrea Merrick, Paula Rossiasco, Administrative and logistical support was provided by Marcelo Jorge Fabre, Leonardo Iacovone (World Bank), Dany Mario Mendez, Aida Romero, Jacqueline Larrabure, and Bahar (Brookings Institution), Ana María Ibañez Londono Sergio Alejandro Perez Baron Rivero of the World Bank. (Inter-American Development Bank), Valerie Mueller Graphic design services were provided by Jaime Sosa. (Arizona State University and International Food Policy Research Institute), Jean-Francois Maystadt (Institute of Funding was provided by the Office of the Chief Economist Development Policy (IOB) at University of Antwerp), and for Latin America and Caribbean of the World Bank and by Carlo Devillanova (Bocconi University, Milan) for advice and the Agriculture and Food Global Practice of the World Bank. brainstorming at various stages of the report. The team recognizes with gratitude the insights gathered from Juan Camilo Gonzalez Garzon, Winston Andrés Martínez, Eduardo Llaña Sanchez, Leydi Andrea Martinez Gutiérrez (Migración Colombia), Rafael Isidro Parra-Peña Somoza, Juan Carlos López, Nohora Forero Ramirez, Jorge Armando Rueda Gallardo, Ana Maria Paredes Chaux (Departamento Nacional de Planeación, Dirección de 7 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR CONTENTS EXECUTIVE SUMMARY............................................................................................................................................................................................ 3 ACKNOWLEDGEMENTS........................................................................................................................................................................................... 7 I. INTRODUCTION................................................................................................................................................................................................ 12 II. VENEZUELAN MIGRANTS IN COLOMBIA............................................................................................................................................. 16 A. CHARACTERIZING VENEZUELAN MIGRATION........................................................................................................................................ 16 B. CHARACTERIZING VENEZUELAN MIGRANTS: PROFILES..................................................................................................................... 19 III. COLOMBIA’S AGRI-FOOD JOBS................................................................................................................................................................ 32 A. CHARACTERIZING AGRI-FOOD JOBS ALONG THE VALUE CHAIN.................................................................................................... 35 I. PRIMARY PRODUCTION............................................................................................................................................................................. 35 II. FOOD PROCESSING..................................................................................................................................................................................... 38 III. FOOD DISTRIBUTION, RETAIL, AND FOOD SERVICES....................................................................................................................... 38 IV. ARE VENEZUELAN MIGRANTS INTEGRATING INTO COLOMBIA’S AGRI-FOOD SYSTEM?............................................ 42 A. ADDITIONAL EVIDENCE: THE INTERACTION WITH THE POST-CONFLICT PROCESS.................................................................. 49 V. INTEGRATING VENEZUELAN MIGRANTS INTO COLOMBIA’S AGRI-FOOD SYSTEM: TWO CASE STUDIES.................................................................................................................................................................................................. 53 A. THE FLORICULTURE SECTOR: CASE STUDIES FROM CUNDINAMARCA.......................................................................................... 53 I. ASOCOLFLORES: PROGRAM FOR LABOR MARKET INCLUSION OF VULNERABLE POPULATIONS...................................... 54 II. SUNSHINE BOUQUET: PROGRAM FOR LABOR MARKET INCLUSION OF VENEZUELAN MIGRANTS................................ 55 B. THE COFFEE SECTOR: CASE STUDIES FROM RISARALDA AND NORTE DE SANTANDER.......................................................... 56 I. HARVEST PLAN 2019 (PLAN COSECHA 2019): LABOR INCLUSION PROGRAM IN RISARALDA............................................. 56 VI. CONCLUSION..................................................................................................................................................................................................... 60 LOOKING AHEAD: OPERATIONALIZATION PRINCIPLES.................................................................................................................. 61 VII. BIBLIOGRAPHY................................................................................................................................................................................................. 63 VIII. ANNEX.................................................................................................................................................................................................................. 65 A. DATA SOURCES................................................................................................................................................................................................. 65 I. HOW MANY VENEZUELAN MIGRANTS ARE IN COLOMBIA?......................................................................................................... 65 II. AGRI-FOOD LABOR MARKETS................................................................................................................................................................. 65 B. REGRESSION ANALYSIS – TECHNICAL DETAILS..................................................................................................................................... 67 C. ADDITIONAL RESULTS.................................................................................................................................................................................... 68 D. ADDITIONAL EVIDENCE ON COLOMBIA’S AGRI-FOOD SECTOR....................................................................................................... 92 E. VULNERABLE POPULATIONS AND POST-CONFLICT INTEGRATION PATHS................................................................................... 98 I. RETURNEES.................................................................................................................................................................................................... 98 II. DEMOBILIZED EX-COMBATANTS............................................................................................................................................................ 100 F. INTEGRATING VENEZUELAN MIGRANTS INTO COLOMBIA’S AGRI-FOOD SYSTEM: TWO CASE STUDIES.......................... 102 I. IN-DEPTH SEMI STRUCTURED INTERVIEWS: RESULTS FOR THE FLORICULTURE SECTOR................................................... 102 II. IN-DEPTH SEMI STRUCTURED INTERVIEWS: RESULTS FOR THE COFFEE SECTOR................................................................. 105 8 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR List of Tables Table 1. Locals vs. Venezuelan migrants – Descriptive statistics............................................................................................................. 23 Table 2. Locals vs. Venezuelan migrants: Labor market characteristics of the employed population...................................... 27 Table 3. Labor outcomes of Venezuelan migrants in most recent migrant surge............................................................................. 29 Table 4. Labor outcomes for migrants in the most recent surge – Urban vs. rural areas.............................................................. 29 Table 5. Migrants’ employment in the most recent surge – Heterogeneity analysis...................................................................... 30 Table 6. Sector of employment – employed population............................................................................................................................ 32 Table 7. Worker profiles: primary agriculture, year 2018.............................................................................................................................. 35 Table 8. Worker profiles: food processing, year 2018.................................................................................................................................... 40 Table 9. Worker profiles: food services and retail, year 2018.................................................................................................................... 40 Table 10. Agri-food sector outcomes for migrants in recent migration surge.................................................................................... 44 Table 11. Agri-food employment for migrants in agriculture-intense locations................................................................................. 45 Table 12. Agri-food employment for migrants in food-processing-intense locations..................................................................... 46 Table 13. Agri-food employment for migrants in food-retail-intense locations................................................................................ 47 Table 14. Characteristics of migrants and local Colombians employed in the agri-food sector in 2018.................................. 48 Table 15. Labor conditions for migrants employed in agri-food sector................................................................................................ 48 Table 16. Labor market outcomes for migrants in recent migration surge, accounting for conflict-related internal migration........................................................................................................................................................................................................ 50 Table 17. Working conditions in agri-food for migrants in recent migration surge, accounting for conflict-related internal migration........................................................................................................................................................................ 51 Table 18. Sunshine Bouquet’s hiring process for migrants through employment postings in Cucuta....................................... 55 List of Figures Figure 1. Temporal shift in formal Venezuelan visitors entering Colombia, by gender 2012-1 – 2019-2................................... 16 Figure 2. Main reason for entry declared at Colombia’s immigration port by Venezuelan travelers........................................ 17 Figure 3. Main destination of Venezuelan Migrants, by number of migrants, 2012 Q1 - 2019 Q1............................................... 18 Figure 4. Age and gender disaggregation of Venezuelan visitors, before and after the 2017 surge in migration................. 20 Figure 5. Main Occupations* of the Venezuelan visitors (per 10,000 people, aged 18+)............................................................... 20 Figure 6. Venezuelan Migration of recent and longer-term migrants from GEIH, 2014 - 2019 (trimester).............................. 21 Figure 7. Main Reason for Migration in the last 12 months....................................................................................................................... 21 9 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Figure 8. Number Special Permits of Permanence (PEP) issued for Venezuelan’s migrants......................................................... 26 Figure 9. Spatial distribution of migrants with legal working permits (PEP)..................................................................................... 26 Figure 10. Annual growth of sectoral value added, Colombia, 2009-2018 average (%)................................................................. 33 Figure 11. Annual growth of agriculture value added, Latin American countries, 2009-2018 average (%).............................. 34 Figure 12. Average annual growth rate of agricultural TFP (%): Latin American countries, 2005-2015..................................... 34 Figure 13. Average annual growth rate of agricultural labor productivity (%): Latin American countries, 2005-2015........ 34 Figure 14. Land used in agri-food activities, by department.................................................................................................................... 36 Figure 15. Spatial location of cattle stock in Colombia, 2018.................................................................................................................... 36 Figure 16. Permanent workers in agriculture, 2014, by department....................................................................................................... 36 Figure 17. Workers in on-farm food processing, 2014, by department................................................................................................... 39 Figure 18. Location of food markets in Colombia, 2019.............................................................................................................................. 39 Figure 19. Agri-food activity and migration hotspots................................................................................................................................. 43 Figure 20. Recent internal migration due to armed conflict, 2014-2019 (by trimester).................................................................. 49 Figure 21. Location map for production of Cut Flowers............................................................................................................................. 54 List of Boxes Box ES1. COVID-19 and Venezuelan displacement in Colombia............................................................................................................... 6 Box 1. Integrating Venezuelan Migrants in Colombia: Regularization Programs and integration strategies........................... 25 Box 2. Agriculture in Colombia: a macro perspective................................................................................................................................... 33 Box 3. Coffee pickers in Colombia: a scarce resource.................................................................................................................................... 37 Box 4. An unsuccessful example: the labor inclusion program for Colombian returnees in Norte de Santander................. 58 Box 5. Testimonial statement................................................................................................................................................................................. 102 Box 6. Testimonial statement................................................................................................................................................................................ 103 Box 7. Testimonial statement................................................................................................................................................................................. 103 Box 8. Testimonial statement................................................................................................................................................................................ 104 Box 9. Testimonial statement................................................................................................................................................................................ 105 Box 10. Testimonial statement.............................................................................................................................................................................. 105 10 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR I INTRODUCTION 11 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR I INTRODUCTION 1.  At the end of August 2020, the number of Venezuelans in the short run. For example, it exerts pressure on social living abroad reached 5.2 million, up from 3.3 million services such as healthcare and education and contribute in December 2018 (Coordination Platform, 2019a and to the congestion of cities, in particular posing additional 2019b).4 The main destination of Venezuelan migrants is challenges as the country strives to preserve public health Colombia. The country has been hit by more than a third and food security in light of the recent COVID-19 pandemic. of this exodus: as of August 2020, the Colombian Statistical Colombian authorities, for example, have recently been Institute (DANE) counted around 1.8million Venezuelans stressing how the response to the COVID-19 emergency living in the country. This includes Venezuelan legal and aims to cover the local and migrant populations alike, but illegal migrants, and migrants transiting through Colombia that resources are insufficient to cover all existing demand.6 towards other destinations. Adding to the Venezuelans moving to Colombia, many Colombian citizens, who during 4.  It is also very likely that migration might continue to Colombia’s internal conflict and Venezuela’s boom years exert short-run pressure on Colombian local labor markets: crossed the Venezuelan border in search of a better life, estimates by the World Bank (2018) highlight, for instance, are now returning home: by the end of 2018, between the risk of an increase in informality and under-employment, 300,000 and 500,000 Colombians had been reported to as well as a reduction in real wages. A recent study on the have returned from Venezuela since the start of the crisis spillover effects of the Venezuelan crisis on Colombian labor (GRID, 2019). markets (Caruso et al., 2019) presents evidence of negative recent wage effects of Venezuelan immigration on local 2.  In the last five years, migration from Venezuela to workers, in particular in the informal sector, in urban areas Colombia has been directed mostly to Bogotá, the and for low-skilled jobs. frontier zones (Norte de Santander, Arauca, La Guajira), and the Atlantic Coast, with 80 percent of the migrants 5.  These impacts are expected to be particularly strong concentrated in these areas (World Bank, 2018).5 Many in Colombian departments bordering Venezuela, which factors make Colombia the safest option and most are not only strongly hit by the migration surge but are common destination for Venezuelan migrants (International also historically lagging regions from a socio-economic Organization for Migration, 2018). Among these, particularly perspective. An influx of migrants is likely to jeopardize relevant are cultural similarity, geographic proximity, and their already strained local economies and social protection social networks (the presence of family or friends): in a systems. These challenges are even more salient as Colombia recent survey of 15,000 Venezuelan migrants crossing the transits out of the internal conflict that plagued the Venezuela-Colombia border in a municipality in Norte de domestic scene for more than 50 years and strives to rebuild Santander (Bermudez et al., 2018), around 95 percent of the national cohesion and heal the civil and social wounds that interviewed claimed they would already have a place to the conflict left as its legacy. The COVID-19 pandemic, which stay once in Colombia. is imposing dramatic socio-economic impacts worldwide and is likely to hit hardest on the most vulnerable segments 3.  The sudden surge in migrants’ influx in Colombia has of the population, adds complexity to an already strained required changes to the country’s policies and regulations situation – by slowing local economies, causing job frictions (Universidad del Rosario and Konrad Adenauer Foundation, especially in informal markets, and exacerbating social 2018), and is imposing a burden on public expenditure: tensions and anti-migration sentiments.7 Even though this overall, the expected cost to the Colombian government of report, which was finalized before the contagion spread to hosting the migrant population is roughly 0.2 to 0.4 percent Colombia, does not attempt to quantify the compounded of GDP (Reina, Mesa and Ramírez, 2018; World Bank Group, effect of the pandemic on the migration crisis, the health 2018; OECD, 2019). Apart from imposing a high fiscal cost, emergency and the complications it triggers make it even the Venezuelan migration crisis poses serious challenges to more necessary to find viable solutions for the thousands Colombia’s welfare schemes and social capital, especially of Venezuelans currently in the country. 4  The refugee crisis in Syria, which began in 2011, caused more than 6.3 6  Cf. interviews in the news e.g. https://www.eltiempo.com/politica/gobierno/ million people to be displaced by 2017. The speed in the growth of the total entrevista-con-el-gerente-del-gobierno-para-la-frontera-con-venezuela-481306; number of Venezuelan migrants and refugees is as high as that of the Syrian https://elpais.com/internacional/2020-04-12/desandar-el-camino-en-medio-de-la- crisis during its first years (cf. Bahar and Dooley, 2019) . pandemia-el-drama-de-los-venezolanos-que-regresan-por-falta-de-recursos.html. 5  Data sources for the report: Border Management. RAMV (Administrative 7  Cf. news reports e.g. https://www.ntn24.com/america-latina/venezuela/ Registry of Venezuelan Migrants). Report to June 12, 2018, and Own estimates america-latina/venezuela/fedenaga-informa-que-han-reportado-intentos-de- based on official data from DANE (National Administrative Department of invasiones-en-fincas; https://migravenezuela.com/web/articulo/xenofobia-el-otro- Statistics) and Migration Colombia. contagio/1842. 12 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR 6.  International experience shows that migration can also 10.  This report seeks to understand the integration of be an important source of growth in the medium term. The migrant workers into Colombian labor markets, and into OECD, for example, highlights how the increase in labor agriculture and food systems in particular, to inform supply could help raise productivity and boost the growth decisions on how best to create mechanisms and incentives potential of the Colombian economy (OECD, 2019). Migrant towards mobility and job take-up in the sector. The main flows can generate growth in host economies by bringing in objectives of the report are three-fold. First, it is to provide investment capital and entrepreneurship opportunities, and greater detail than is currently available on migrants and traditional economic activities by supplementing demand their characteristics. Such profiling provides a clear vista of a in local economies (Kerr and Kerr, 2011). range of migrant attributes relative to local Colombians and gives a temporal picture of how these profiles have evolved 7. The extent to which these benefits materialize depends in the aftermath of the surge of Venezuelan migrants in however on the ability of domestic labor markets to Colombia. To do so, the report analyzes a rich set of detailed successfully integrate the influx of migrant workers. To micro-level data at a high level of spatial and temporal mitigate the impacts on civil security and labor markets, disaggregation. The study of migrant skills, gender and age active social integration programs can play an important role profiles is important for assessing the potential for their (FIP, 2018; World Bank Group, 2018). Mitigating vulnerabilities integration into labor markets, as well as understanding that can become poverty traps and prioritizing the rapid any possible local labor displacement effects.9 The study incorporation of migrants and returnees into the labor of the Venezuelan population in Colombia also allows to market is therefore a key element to turn the challenges identify an exponential surge in migrants’ numbers starting of Venezuelan migration into a growth opportunity in the first quarter of 2017 and lasting until 2019. Even for Colombia. though the observation is based on official migration data and may or may not be mirrored by a parallel increase in 8.  The Colombian government has been taking a series of irregular migration, this surge can be used as a quasi-natural measures aimed at facilitating the integration of Colombian experiment for understanding migrant integration into local returnees into the labor force and social protection system, labor markets. From the perspective of the migrant, some as well as at promoting migrants’ self-sufficiency and initial correlations are therefore established to understand mitigating impacts in the receiving areas. Moreover, it has how Venezuelan migrants have been faring in Colombia’s adapted its legal and institutional framework to stimulate labor markets after 2017, during the migrant surge the country’s response capacity in the face of the migratory into Colombia.10 crisis. An example in this sense is the 3-year strategy delineated by the CONPES8 plan no. 3950 for the years 2018- 11.  Second, an objective of the report is to characterize 2021, which is specifically intended at creating integrated the agri-food sector in Colombia and to assess whether solutions to the migration crisis, including measures to Colombia’s agriculture and food systems can provide quality facilitate migrants’ inclusion in the labor force (cf. Box 1). job opportunities to Venezuelan migrants, in particular as the sector faces a strong push towards modernization and 9.  Yet, de facto, access to employment for Venezuelan growth, on the one hand, and important labor shortages people remains limited: estimates from DANE, for instance, threatening its productivity and future performance, on the reveal that, of the 773,000 working-age Venezuelans other. Understanding the labor dynamics associated with who arrived in Colombia between September 2018 and the growth and transformation of agriculture and food September 2019, 19.2 percent have no job, almost twice systems, as well as the likely areas of unmet labor demand the unemployment rate among Colombians in the same that could be potentially filled by Venezuelan migrants, period (10.1 percent). Illegal migration, difficulties to obtain holds promise for a successful migrant integration strategy the necessary documents (e.g. passports, birth certificates, that can benefit the entire country. conversion of education titles), limited or incomplete knowledge on the part of employers of existing integration 12.  The third objective is to identify existing programs programs and related bureaucratic requirements, and and policies that could be used to provide employment practical challenges in opening bank accounts and opportunities for migrants in the agri-food sector. This is receiving payments, are among the factors that hinder highlighted through two case studies in the coffee and an efficient and productive inclusion of migrants into the flower subsectors, with additional in-the-field grounding Colombian workforce. 9  For example, evidence from the recent Syrian refugee crisis shows that in 8  CONPES (Consejo Nacional de Política Económica y Social) is Colombia’s the short term the large influx of informal labor availability led to employment highest national planning authority and serves as an advisory body to the losses among native informal workers and declining earnings opportunities Government in all aspects related to the economic and social development of (Del Carpio and Wagner, 2015; Cerioglu et al, 2017). the country. To this end, it coordinates and guides the agencies responsible for 10  Nonetheless, it is acknowledged that the analysis provided in the report economic and social management in the Government, through the study and establishes correlations and not causation, as the endogeneity of migrant approval of documents on the development of general policies. locales is not fully dealt with. 13 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR of the existing reality and potential for migrant labor services). Section 4 tests whether the agri-food sector integration along the coffee and flower value chains. Field in Colombia is working as a catalyst for the inclusion of interviews and information for this analysis was gathered Venezuelan migrants in the labor force, with a focus both from Venezuelan migrants, producer associations, and the on employment (overall as well as at various stages of the ministry of Agriculture in Colombia. agri-food value chain) and on job quality; the section also touches on the additional challenges presented by post- 13.  The rest of the report is organized as follows. Section conflict mitigation, studying whether Venezuelan migrants 2 provides a general overview of Venezuelan migration in face further difficulties integrating in areas with a higher Colombia, spatially mapping migrants and characterizing concentration of conflict-related vulnerable populations. their profiles with respect to migration reason, permanency Section 5 presents two successful labor market inclusion status, and basic socio-demographic characteristics. programs for Venezuelan migrants in Colombian agriculture This section also highlights the key characteristics of the and food systems, summarizing lessons learned that permit policy put in place by the Colombian government could be relevant for the broader inclusion of Venezuelan in the face of the crisis, and it characterizes the profiles of migrants in the agri-food sector and discussing potential Venezuelan migrants with respect to human capital and implementation challenges. Section 6 provides some current employment outcomes (including occupational suggestions on operationalization of the findings for categories, temporality and wages). The analysis devotes public policy makers and private agricultural and food a special emphasis to rural areas, skill mismatch, and system actors. outcomes of vulnerable groups such as women and youth. Section 3 characterizes agri-food job markets in Colombia, identifying challenges and opportunities specific to the different stages of the food chain (primary production, agri-food manufacturing, logistics and retailing, food 14 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR VENEZUELAN II MIGRANTS IN COLOMBIA 15 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR II VENEZUELAN MIGRANTS IN COLOMBIA A. CHARACTERIZING 16.  Figure 1 shows the number of formal Venezuelan visitors to Colombia from 2012 first quarter to 2019 first quarter, VENEZUELAN MIGRATION and very clearly visualizes the magnitude of the recent surge in Venezuelan migration. Prior to the first quarter of 14.  Quantifying and characterizing Venezuelan migrants in 2017, the average number of Venezuelan visitors entering Colombia is a challenge, first because there is no unified Colombia was less than 200,000 per trimester (around definition of a migrant. Even in terms of nationality, 500,000 per year). In the first quarter of 2017, this number migrants can be Venezuelan, dual nationality, former started to increase, and after the second quarter of 2017 it Colombians returning to Colombia after living in Venezuela began growing exponentially: during 2018, around 1,500,000 for years, or migrants coming from Venezuela with other Venezuelans entered Colombia. The gender composition of nationalities. More importantly, irregular migration poses Venezuelan travelers shows a pronounced male bias: during additional challenges to migrant accounting. In June 2019, the peak of Venezuelan migration in 2018, approximately according to Colombia’s Government, there were 1,408,055 six out of ten migrants were male. Venezuelan living in Colombia, of which only 53 percent were regular migrants. Of these, 66,297 had permits 17.  The data from Migración Colombia also allow an under three months, 78,390 had work permits and 597,583 exploration of the main reasons for their trip declared transitory special work permits. The remaining 47 percent by visitors upon entering Colombia from Venezuela over of migrants, however, were irregular: 240,276 overstayed the years (see Figure 2). In 2017, 2018, and the first half of their three-month permits, and 445,389 entered Colombia 2019, a total of 11,145, 70,719, and 27,069 travelers entered through irregular channels altogether (Obsevatorio, 2019). Colombia in order to benefit from the PEP (Permiso Especial By the very nature of the irregular migration phenomenon, de Permanencia), a special permit introduced in 2017 by the this information is very likely to underestimate substantially Colombian government to allow Venezuelan migrants to the real magnitude of migration. legally work in Colombia (cf. Box 1). These small numbers show that not many migrants are really aware of the 15.  Data from the Colombian Migration Authority program, or that in any case it does not represent the main (Migración Colombia) provide a first snapshot of the motive for their arrival from Venezuela. At the same time, profiles of all formally registered Venezuelan who visited however, it must be acknowledged that these numbers Colombia through all ports from 2012 to 2019. Keeping in only represent a subset of all Venezuelan travelers into mind that this information (i) is not restricted to migrants Colombia. only (visitors could have different motives for travelling to Colombia); and (ii) does not account for irregular migrants, 18.  As it appears, the predominant declared motivation these data provide a useful macro vista of all regular for entry into Colombia has in recent years always been entrants at Colombia’s borders. tourism. Even though this has consistently be the case since Number of people 450,000 Figure 1 400,000 Temporal shift in 350,000 formal Venezuelan visitors entering 300,000 Colombia, by gender 250,000 2012-1 – 2019-2 200,000 Data source: Colombian 150,000 National Migration Statistics, 2019 100,000 50,000 - Fem le M le Ye r nd qu rter 16 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR 2012, however, Figure 2A. shows that in 2017 and 2018 there number of migrants entering Colombia in transit to another was a further, more than proportional surge in the number country, which from almost negligible levels skyrocketed to of travelers declaring tourism as the main motive for their 300,000 individuals in the first half of 2019. This loosely visit. While 70 percent of Venezuelan visitors who entered reflects the changing geography of Venezuelan migration, Colombia self-identified as tourists between 2012 and 2017, with migrants now moving further away from their after 2017 the corresponding share reached 90 percent. country and spreading throughout the Latin America and The fact that the number and percentage of Venezuelan Caribbean region. “tourists” jumps so visibly in the wake of the migratory crisis is extremely relevant, suggesting that, very likely, the 20.  Figure 3A and 3B show the destination within Colombia generic category of tourism masks a number of other deeper declared by Venezuelans entering Colombia – which, reasons for entry into Colombia. Without a regular work or allowing for a certain degree of discretion in final location residency permit, tourism might for many migrants be the decisions, can be used to map the major hotspots of only way to legally cross the Colombian border. As visitors migrant settlement over time. The comparison of Figure 3A entering Colombia for tourism are by law not allowed to and 3B indicates a shift and a higher spatial concentration perform any kind of work, it is also likely that many of of migrants destinations, with cities and smaller areas near the self-reported tourists are in fact joining the informal the border receiving large migrant inflows. After 2017, the labor force (working without a contract) to provide for frontier regions located at the Venezuelan border were their livelihoods. receiving more, and a larger share, of migrants, as well as main urban centers such as Bogotá, Medellín, Cali, 19.  In terms of the broad compositional change of the Bucaramanga and Cúcuta. migrant pool, from Figure 2B. another development worth highlighting is the recent exponential increase in the Number 300 Figure 2 A of people (10,000) 250 Main reason for entry declared 200 at Colombia’s immigration port by 150 Venezuelan travelers 100 Entry type: Tourist Data source: Colombian 50 National Migration Statistics, 2019 0 Tour st 2012 2013 2014 2015 2016 2017 2018 2019 (Q1-Q2) Number 40 Figure 2 B of people (10,000) 35 Main reason for entry declared 30 at Colombia’s 25 immigration port by Resolut on Venezuelan travelers 1272-17 PEP 20 Entry Type: Bus ness Resident, Work, 15 Transit, Events, Events 10 Business, PEP Tr ns t Data source: Colombian 5 National Migration Wor Statistics, 2019 0 Res dent 2012 2013 2014 2015 2016 2017 2018 2019(Q1-Q2) 17 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR 3A | Pre 2017 Q1 Figure 3A / 3B Main destination of Venezuelan Migrants, by number of migrants, 2012 Q1 - 2019 Q1 Data source: Colombian National Migration Statistics, 2019 3B | Post 2017 Q1 (included) > 100 - > 1.000 - > 10.000 - People <=100 > 50.000 NA <=1.000 <=10.000 <=50.000 18 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR 21.  In order to characterize the travelers entering Colombia and how are they integrating into local labor markets? from Venezuela, Figure 4 compares the age and gender Detailed answers to these questions can be obtained from decomposition of Venezuelan visitors before and after the the Gran Encuesta Integrada de Hogares (GEIH), a household start of the migration crisis in the first quarter of 2017, on survey collected monthly in the Colombian territory.12 The a time span ranging between the first quarter of 2012 to GEIH database has a specific migration module, which the second of 2019. The pyramids show a swelling in the can be merged to the main modules to obtain a general proportion of visitors within the prime age groups (18-29 overview of Venezuelan migration into Colombia, as well as and 30-39-year-olds): before 2017, most Venezuelan visitors of migrants’ labor-market outcomes.13 Crucially, the survey were already male aged 30-39 and 18-29 (14.2 and 12.3 is by design also able to capture irregular migration and percent, respectively), and women aged 18-29 (12.1 percent). informal labor market participation. After 2017, however, the share of the 18-29 age group boomed (35 percent of all migrants), increasing by more 25.  Using the GEIH survey, it is possible to classify than 50 percent for both genders. Venezuelan migrants14 in two mutually exclusive ways. First, recent migrants can be identified as those individuals who 22.  Another relevant characterization of Venezuelan claimed that 12 months or less before the interview they travelers is by occupational category at time of entry, as were living in Venezuela. Second, longer-term migrants are declared by 18+ year old (Figure 5). Over time, the most those who were living in Colombia 12 months before the frequently reported occupations are housewife and interview, but in Venezuela up to five years before that. merchant. However, recent years have been witnessing a surge in a mishmash of professional workers, mostly highly 26.  Figure 6 plots the patterns of migration over time for skilled, as well as in adult students. Interestingly, although the two categories of migrants just described. Reassuringly, the related numbers are small,11 a detectable increase can these patterns are very similar to those obtained using also be observed in the number of low-skill agri-food official migration records in Figure 1. For instance, Figure 6 system workers. clearly identifies the sudden surge in migrants after 2017. Crucially, however, it is not only the visual, qualitative 23.  As a necessary caveat, although this preliminary migration trends that are similar between the two datasets, evidence suggests some change in the profiles of but the migration magnitudes themselves: for example, if migrants over time, it has to be pointed out that the skills the official national migration statistics count 1,408,055 disaggregation in Figure 5 cannot postulate strongly on the Venezuelan migrants living in Colombia by June 2019, shift in the skills mix of all migrants, as it just accounts for estimation from GEIH identifies 1,506,270 migrants15 in the regular (formal) migrants. This is all the more relevant as same period. migrants who entered Colombia as the Venezuelan crisis deepened have been reported to be more vulnerable and For individuals who migrated in the last 12 months, a poorer than in previous migratory waves (World Bank, follow-up question gathers information on the main 2018), and might for the same reason be also lower skilled. reason for migration. Figure 7 depicts the main responses If this were correlated with a higher likelihood of entering obtained from migrants who entered Colombia in recent Colombia through informal channels, the information in years. Here, it can be inferred that the recent migration Figure 5 might provide a very incomplete picture of the wave is a phenomenon involving entire families, and not real situation, and warrant a more in-depth analysis using single individuals moving by themselves: most migrants alternative, more comprehensive data sources. 12  A clarification: the GEIH survey is representative of the Colombian population as a whole. More information about the sampling can be found at B. CHARACTERIZING this link. Nonetheless, it is not representative of migrants per say, it would VENEZUELAN MIGRANTS: represent the average migrant in Colombia, yet we cannot say anything about the representativeness among sub-samples. PROFILES 13  The survey includes respondents over age 12, as it is the legally acceptable age to be employed in Colombia. 14  This study does not distinguish between Venezuelan nationals and 24.  Who are the Venezuelan migrants entering Colombia, how Colombian returnees whenever GEIH data is used. The migration module similar are they to the local population in the host country, captures country of birth but not nationality which makes identification very imprecise. The data reveals that among migrants, the share that identify as 11  The numbers of agri-food low-skilled workers is likely reported being born in Colombia is 67% from 2012 to 2016, and 28% since the start with measurement error in the data from Migración Colombia. Certain of the migration crisis in 2017. As such, it is very likely that the analytical categories like waiter for example appear to be severely underreported. This evidence presented, on migrant integration into labor markets, are robust and underrepresentation in official migration statistics is in great part due to the represent Venezuelan nationals only, as well. fact that many of these workers are likely to enter Colombia informally rather 15  This statistic is the number of migrants who would have stated living in than through official border posts. Venezuela under 5 years ago. 19 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Figure 4 Pre 2017, Q1 >70 1 1 60-69 33 3 Age and gender 50-59 69 57 disaggregation of 40-49 10 8 75 Venezuelan visitors, 30-39 14 2 11 before and after 18-29 12 3 12 1 the 2017 surge in 0-17 56 56 migration Data source: Colombian Post 2017, Q1 >70 05 07 National Migration 60-69 16 22 Statistics, 2019 50-59 36 45 40-49 75 65 30-39 14 1 11 3 18-29 18 2 17 3 M le 0-17 61 6 Fem le A e roup % of tr velers 80 Figure 5 A r -food s stem wor ers 70 Main Occupations* Pens oner 60 of the Venezuelan En neer nd Rel ted 50 visitors (per 10,000 people, aged 18+) Other low-level wor ers 40 Data source: Colombian Student 30 National Migration Housew fe Statistics, 2019 20 Other profess on ls 10 Commerc nte - Merch nt 0 2012 2013 2014 2015 2016 2017 2018 2019 (Q1 & Q2) Note: *The chart disaggregates between agri-food workers (Farmer, Beekeeper, Coffee grower, Butcher, Chef, Breeder, Livestock and Related, Cultivator, Dairy Farmer, Baker and Related, Fisherman, Cowboy, Waiter) and Other low-level workers (Driver, Builder, Domestic, Electrician, Stylists and Related, Mechanic and Related, Operators and Related, Seller and Related, Other workers). However, this occupation field is very imprecisely defined, and as such only provides a lower bound for number of food-system workers. after 2017, in fact, claim that the main reason for leaving 27. Table 1 provides a dynamic comparison of the Venezuela was because their family moved. Predictably, characteristics of recent Venezuelan migrants vis-à-vis local however, the second most frequently cited reason is finding Colombians. To investigate shifting characteristics of the a job, followed by facing a (non-conflict-related) threat migrant pool over time, we include both longer-term and in Venezuela. Here, it is important to notice the limited recent migrants. We do nonetheless abstract from migrants overlap with the responses identified at port of entry by living in Colombia more than five years prior because the the Colombian migration authorities, shown in Figure 2. This latter definition does not allow us to identify individuals observation confirms the concern expressed in Section II.A who moved during the recent Venezuelan crisis. On average, above, that the generic “tourism” motivation signaled at new migrants are younger than local Colombians. In 2014- border authorities could mask other reasons for migration. 2016, whereas 25 percent of Colombians were between 15- As the GEIH survey includes regular and irregular migrants 25 years old, 33 percent of recent Venezuelan migrants were alike, the discrepancy between Figure 1 and Figure 7 is also 15-25. While this is true over time, the increase in this share informative about the potential bias arising from focusing during the Venezuelan crisis peak is remarkable: recent the analysis on legal migration alone.16 migrants aged 15-25 were 36 percent of the total in 2017, and 39 percent in the first two quarters of 2019. The number of Venezuelan migrants, both short and longer-term, in the 26- 20 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR 35 age group is also significantly higher than for Colombian only had basic education, whereas a bigger share of local locals and increased especially for longer-term migrants (35 Colombians had tertiary education. It can be postulated percent in the first two quarters of 2019). that prior to the peak of the Venezuelan crisis those crossing the border to Colombia were the less educated, who 28.  Quite expectedly, local Colombians are much more enjoyed fewer opportunities in Venezuela. From 2017 to likely on average to be enrolled in school than Venezuelan 2019, on the other hand, the share of Venezuelans with high migrants, whose average enrollment rates are abysmal in school (10-11 grade) and superior or university education the crisis period (with a 3-percentage points difference increased dramatically. The combined proportion of high between recent and longer-term migrants). In educational school and tertiary educated recent Venezuelan migrants attainment, there are distinct differences in how Venezuelan outweighed the less educated strata and was higher than migrants compare to local Colombians over time. In 2014- the baseline 2014-2016 share for local Colombians. In 2016, prior to the migration crisis, migrants were less other words, more recent waves of Venezuelan migrants educated than local Colombians: a larger share of them have higher rates of education and literacy than previous 1000 Number of Figure 6 m r nts 900 (x1000) Venezuelan 800 700 Migration of recent 600 and longer-term 500 migrants from 400 GEIH, 2014 - 2019 300 (trimester) 200 Data source: GEIH, 2019 100 0 1511 1711 1411 1611 1811 1701 1501 1707 1401 1601 1507 1901 1703 1705 1801 1407 1503 1505 1607 1709 1807 1403 1405 1509 1603 1605 1903 1905 1803 1805 1409 1609 YYYY / MM 1809 Number of 500 Figure 7 m r nts 400 (x1000) Main Reason for 300 Migration in the last 12 months 16 200 100 Data source: GEIH, 2019 0 500 1 450 400 350 16  For these figures, data is 300 shown from 2016 onwards 250 because the way in which 200 150 the question was phrased 100 changed between 2014 and 50 0 2016. Therefore, it is difficult 500 to cleanly match the main Mov n lon 450 reason for migration variable w th f m l 400 pre and post 2016. 350 300 Wor 250 200 150 Thre t 100 (non- rmed 50 confl ct) 0 1809 1609 1803 1805 1603 1605 1905 1903 1807 1607 1709 1703 1705 1801 1601 1901 1707 1701 1811 1611 1711 YYYY / MM 21 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR migrants and the local Colombian population. This result is attend to the needs of Venezuelans migrants into Colombia. in line with recent anecdotal evidence that, at the height In addition to recent, emergency programs for immediate of the Venezuelan crisis, its residents leaving for Colombia attention to migrants arriving into urban centers (providing have been on average more educated and literate than the for example basic services such as housing, food, and local Colombian population – even without controlling for healthcare in centers or refugee camps), the government is the fact that recent Venezuelan migrants are on average also creating a number of regularization programs so that younger than Colombian locals. migrants can legally participate in formal labor markets and access social security and healthcare (see Box 1). 29.  As could be expected, even at baseline in 2014-2016, the proportion of local Colombians with access to social 32.  The finding that migrants appear to be doing relatively security in health17 (92 percent) was higher than for well on labor markets needs to be better understood. For longer-term migrants (71 percent) and for recent migrants example, it is important to assess whether migrant labor (46 percent). The fact that longer-term migrants enjoyed supply is acting as a complement or a substitute to local a better coverage than recent migrants is also intuitive, labor. New migrants could potentially fill labor shortages in considering their longer exposure to Colombian services. certain sectors, by bringing new skills to the host country, With the intensification of the Venezuelan crisis, however, or by otherwise accepting to downgrade their skills in migrants’ access to healthcare deteriorated consistently, occupations for which they are overqualified. Maystadt and resulting in a much lower coverage by 2019: among recent Verwimp (2014), for example, consider whether migrants migrants, only 11 percent had access healthcare services (a from Burundi and Rwanda in the ‘90s and early 2000s quarter of what it used to be in 2014-2016), while the share filled labor shortages in labor-intensive primary agriculture dropped to 42 percent for longer-term migrants. in Tanzania, and find that non-agricultural workers and self-employed farmers were well-positioned to benefit 30.  A broad view of labor market outcomes portrays a from the refugee inflow. On the other hand, migrants might less intuitive picture. Traditionally, the employment rate of well be more willing to accept worse salaries and working Venezuelan migrants appears to have been not too far from conditions, reducing average wages and exacerbating that of local Colombians (66 and 56 percent for longer-term disparities (Orrenius and Zavodny, 2012). and recent migrants respectively vs. 61 percent for locals in 2014-2016). What is remarkable, however, is that the 33.  To shed light on this question, Table 2 provides employment rate of Venezuelan migrants peaked in 2018, at summary statistics for Venezuelan migrants employed in the apex of the migration crisis, when it reached 67 and 64 Colombia,19 again comparing their profiles with those of percent for longer-term and recent migrants, respectively. employed Colombians. In terms of gender composition, the Similarly, the unemployment rate18 of longer-term migrants migrant labor force has over time featured a lower share at baseline was quite close to that of Colombian locals of women than the Colombian one. As the overall migrant (9.5 vs. 8.7, respectively). The rate increases substantially pool in Table 1 is roughly equally split between men and for recent migrants (21 percent), which is fully consistent women, Table 2 suggests that in recent times it has mostly with the fact that migrants who have been in the host been men who migrated to engage in labor markets, while country for less than one year might still be searching for women might have traditionally been more motivated to employment in Colombia. Mirroring the trends observed for migrate to accompany their families. The share of employed employment, the unemployment rates of migrants did not female migrants, however, has been consistently growing escalate during the Venezuelan crisis: in 2018, for example, during the Venezuelan crisis, and in 2019 it almost reached the unemployment rate among longer-term and recent the level observed at baseline among non-migrants (40 vs. migrants was 10 and 18 percent respectively. Countering the 43 percent, respectively). widespread perception of Venezuelan migrants incurring high shoe leather costs in search of jobs, the numbers in 34.  In line with the overall age composition of the migrant Table 1 seem to suggest that the intensification of the pool, employed Venezuelan migrants are younger than migration crisis did not substantially affect migrants’ local Colombians. The share of migrant workers in the age prospects to find employment in Colombia. groups 15-25 and 26-35 has been consistently higher than for Colombians, both before and after the start of the 31.  This could in part be explained by the Colombian migration crisis in 2017. While in 2014-2016 workers in the government’s efforts to develop different strategies to age groups 15-25 and 26-35 represented 18 and 26 percent of the employed Colombian population, for example, they accounted for 28 and 36 percent of the employed recent 17  This was defined as a positive answer to the question whether the Venezuelan migrants. The share of youngest workers respondent is affiliated, is a contributor or is a beneficiary of any entity of social security in health? 18  The unemployment rate is defined as defined as the share of people not in 19  Descriptive statistics for the unemployed and inactive populations are employment but looking for jobs over the total labor force. provided in Annex C. 22 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Table 1 Locals vs. Venezuelan migrants – Descriptive statistics 2014 – 2016 2017 2018 2019 (Q1 & Q2)   Locals Venezuelan Migrants   (born in 12 12 12 12 5 Years 5 Years 5 Years 5 Years Colombia) Months Months Months Months 1. General Characteristics  1.1 Gender Female (%) 51.70 45.73 45.80 48.11 49.93 48.27 49.51 48.28 52.75 1.2 Age (years)  15 – 25 (%) 24.54 26.13 33.43 28.04 35.48 31.94 38.29 31.90 38.96 26 – 35 (%) 22.08 32.12 31.73 30.09 32.66 32.78 31.36 34.79 30.35 36 – 45 (%) 19.15 20.62 18.32 23.62 16.81 18.73 17.09 17.91 16.39 46 – 55 (%) 17.71 12.69 10.00 11.06 9.26 10.48 7.69 9.40 8.86 56 – 70 (%) 16.52 8.45 6.52 7.20 5.78 6.06 5.57 6.00 5.45 2. Education 2.1 Literacy Rate (%) 95.75 95.43 96.44 96.79 97.19 97.98 98.58 97.90 98.71 2.2 Currently in school (%) 13.06 8.30 5.32 8.43 2.90 5.80 2.06 5.12 2.22 2.3 Highest Educational Level None (%) 4.24 4.40 3.59 3.21 2.79 1.98 1.46 2.13 1.29 Preschool (%) 0.01 0.00 0.00 0.00 0.00 0.01 0.00 0.02 0.00 Basic Primary (1-5) (%) 25.68 26.90 24.52 21.36 17.52 17.15 12.84 15.37 11.01 Basic Secondary (6-9) (%) 17.78 23.86 24.78 24.34 21.52 25.80 25.07 24.82 25.28 High School (10 -11) (%) 27.08 26.14 30.98 31.35 33.70 35.40 35.93 35.94 37.43 Superior or University (%) 25.21 18.70 16.13 19.74 24.47 19.65 24.69 21.72 24.99 3. Health and Social Security 3.1 Access to Healthcare  Yes (%) 91.51 72.28 46.20 65.79 25.24 51.27 16.71 42.05 11.34 3.2 Type of Social Security   Contributive (%) 51.28 37.34 29.89 38.80 38.55 38.74 37.98 41.33 42.16 Special (%) 3.83 0.50 0.31 0.60 0.17 0.29 0.02 0.20 0.12 Subsidiary (%) 44.83 62.08 69.69 60.60 61.28 60.86 61.97 58.47 56.73 4. Labor Market Employment Rate (%) 61.04 65.48 56.00 64.34 57.71 66.75 64.04 66.32 56.96 Unemployment Rate (%) 8.67 9.53 20.84 9.83 21.71 10.08 18.01 12.07 21.26 Note: The employment rate is the percentage of employed over the working age population. The unemployment rate is the percentage of individuals not in employment but looking for work, over the economically active population (employed+unemployed). As the two rates are computed over different totals (working-age vs. economically active) they do not sum to 100. 23 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR (15-25 age group) has however risen considerably after programs described in Section II.B, for example, Table 2 the inception of the crisis: in 2018, this group represented unequivocally shows that migrants are much more likely almost 35 percent of all employed recent migrants. than Colombians to be informal workers, and even more so after 2017: in 2018 and 2019, the share of recent migrants in 35.  The higher levels of education identified for migrants the informal labor force was as high as 77 percent. Migrants in the wake of the Venezuelan crisis (cf. Table 1) are also are also significantly more likely to be working in temporary reflected among the employed sub-group. Before the crisis, occupations. In recent periods, migrants have worked nearly average years of schooling was around 8.5 for employed 4 hours more per week than both local Colombians and migrants in comparison to 9.2 for employed Colombians, migrants in the pre-crisis baseline period. but this jumped to more than 10 in the crisis years. 38.  Interestingly, migrants today are also earning much 36.  In terms of distribution of workers across urban and rural lower wages, both if compared to Colombians, and to areas, Table 2 shows no substantial differences between migrants before the peak of the Venezuelan crisis. While in migrants and local Colombians before the start of the the 2014-2016 period local Colombians earned on average migration crisis: while 40 percent of Colombian locals were COP 825,000, in 2018 recent migrants earned just COP working in rural areas in 2014-2016, the share was 36-39 438,500, about half of what local Colombians earned. Even percent for recent and longer-term Venezuelan migrants in more conspicuous is that, while at baseline longer-term the same years. The wave of migration, however, shattered migrants were even earning COP 105,700 more than local this trend, and today three out of four recent migrants Colombians, by 2019 their average earnings had dropped by work in urban areas. almost 50 percent, to COP 566,900.20 37. As for job quality, it is apparent that migrants enjoy overall significantly worse working conditions. Even though Caruso et al. (2019) show that migration increased the hours worked by locals, which suggests that work pressures have increased on average for all workers, Table 20  Wages are calculated in 2010 real terms. 2 shows that migrants are comparatively worse-off than their local counterparts. Despite their higher education and presumably higher skills, both recent and longer-term migrants work longer hours for much less pay, suggesting they are willing to accept more tenuous labor market conditions. Despite the PEP permits and regularization 24 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Box 1 Integrating Venezuelan Migrants in Colombia: Regularization Programs and integration strategies A significant step in the integration of migrants into Colombia’s labor markets was the launch in 2017 of the Special Permit of Permanence (PEP), which seeks to facilitate the integration and regularization of Venezuelan migrants regardless of their migration status (regular and irregular migration) and grants its beneficiaries permission to legally work in Colombia. In parallel, the government created the Venezuelan Migrants Administrative Registry (RAMV, from its Spanish acronym) between May and June 2018, which targets irregular migrants in the country and grants them access to basic health, education, and other services. So far, there have been four rounds of PEP: » PEP I: Resolution 1272 of 2017 (July 31) » PEP II: Resolution 0361 of 2018 (February 8) » PEP III: Resolution 2033 of 2018 (August 2) Only for those who were registered in RAMV, known as PEP-RAMV: » PEP IV: Resolution 3317 of 2018 (December 2018) The Colombian government is now currently drafting a new PEP round focused on workers that have an employment contract, which will be open to companies that have registered their Venezuelan employees in the Single Registry of Foreign Workers in Colombia (RUTEC)21. In total, 596,077 migrants have received the PEP over the years (Figure 8). In keeping with the age decomposition of (formal) migrants shown in Figure, males and females between 18 and 29 years are the most prevalent age group of migrants with PEP permits (22 percent and 20 percent, respectively), followed by migrants between 30 and 39 years old (14 and 11 percent for males and females). Figure 9 maps the spatial distribution of Venezuelan migrants with PEP in Colombia. Predictably, Bogotá is the city with the largest number of Venezuelan migrants with PEP (169,000), followed by Medellín (44,030), Barranquilla (31,798) and Cúcuta (31,370). In general, there is an important concentration of PEPs around the frontier zone,22 as well as in the central area of the country where most economic activity takes place, and in agricultural zones like coffee regions. In 2018, the Colombian government developed the CONPES plan no. 3950, Colombia Response Strategy to the Venezuelan Population, to serve as a guide for the integration of the Venezuelan population in the various socio- economic systems of the country. Backed by a budget of approximately US$ 125 million, CONPES 3950 builds an action plan for the years 2018 – 2021 and presents a total of 68 strategic actions ranging from healthcare, education, childhood, to work, accommodation, security and coexistence. Accordingly, the CONPES provides from action by the Ministry of National Education, the Ministry of Health and Social Protection, the Ministry of Labor, the Ministry of Housing, City and Territory, the Ministry of Foreign Affairs, the Ministry of the Interior, the Ministry of Education, the Ministry of Defense, the Ministry of Mines and Energy, the Ministry of Commerce, Industry and Tourism, the Administrative Department of the Presidency of the Republic and the National Planning Department. Implementation is left to the various ministries and government institutions involved. 21  The Unique Registry of Foreign Workers in Colombia (RUTEC) is a registry in which Colombian companies must register their foreign workers. 22  Bahar, Dooley and Huang (2018) note how the presence of Venezuelans in these areas is already changing the demographic composition of small villages. 25 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Box 1 continued Integrating Venezuelan Migrants in Colombia: Regularization Programs and integration strategies In the strategies to improve the labor situation of the migrant population from Venezuela, the CONPES contemplates two lines of action: » Line of action 12: Reduce access barriers to the formal labor market. This line envisages actions from the National Apprenticeship Service (SENA) for the certification of apprenticeships and organization of job fairs, as well as from the Public Employment Service to facilitate access to employment. » Line of action 13: Support for entrepreneurship and business development. Under this line of action, the Ministry of Commerce, Industry and Tourism is required to support Venezuelan migrants in processes of entrepreneurship and business development. Figure 8 Number of 300,000 Number Special m r nts (x1000) 250,000 Permits of 200,000 Permanence (PEP) issued for 150,000 Venezuelan’s 100,000 migrants 50,000 Data source: Colombian National Migration 0 Statistics, 2019 PEP I PEP II PEP III PEP IV (Jul 2017) (Feb 2018) (Au 2018) (Dec 2018) PEP RAMV Figure 9 Spatial distribution People of migrants with <=100 legal working permits (PEP) > 100 - Data source: Colombian <=1.000 National Migration > 1.000 - Statistics, 2019 <=10.000 > 10.000 - <=50.000 > 50.000 NA 26 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Table 2 Locals vs. Venezuelan migrants: Labor market characteristics of the employed population   2014 – 2016 2017 2018 2019 (Q1 & Q2) Locals Venezuelan Migrants   (born in 12 12 12 Colombia) 5 Years 5 Years 5 Years 5 Years 12 Months Months Months Months 1. General Characteristics  1.1 Gender   Female (%) 42.51 35.19 31.62 37.56 38.40 37.54 38.66 36.39 40.24 1.2 Age (years)  15 – 25 (%) 17.73 20.56 28.03 20.78 29.82 25.72 34.80 25.32 33.96 26 – 35 (%) 25.62 35.62 35.47 33.58 37.04 35.81 34.50 39.11 35.71 36 – 45 (%) 23.00 22.50 21.81 26.99 19.21 21.21 18.80 20.29 18.76 46 – 55 (%) 20.26 14.50 10.21 11.87 9.59 11.60 8.13 10.22 7.87 56 - 70 ( %) 13.40 6.82 4.49 6.79 4.34 5.66 3.77 5.07 3.69 1.3 Education Avg. years of schooling 9.24 8.51 8.67 9.16 9.65 9.60 10.20 9.98 10.18 1.4 Rural  Rural Worker (%) 39.87 39.20 35.98 40.08 34.30 30.24 16.89 37.93 25.24 2. Job Quality Informal Worker (%) 58.88 71.12 74.40 66.75 73.53 70.84 77.25 69.54 76.98 Temporary Worker (%) 9.56 10.54 18.98 9.13 16.85 10.19 21.18 12.85 24.35 Avg. hours worked per week 45.70 47.03 45.75 47.28 48.44 48.63 49.58 49.77 48.37 Avg. wage (in 100 COP, at fixed 8249.78 9307.26 5549.56 6808.46 4882.08 6133.48 4384.50 5668.60 4468.65 2010 prices) Source: Authors’ calculations using GEIH, various years. 39.  To an extent, the differences in the job profiles social networks or local labor market dynamics. With this observed both between Venezuelan migrants and local caveat in mind, the exercise is nonetheless useful in better Colombians and for migrants over time could be the qualifying the experience of Venezuelan migrants trying to result of a number of factors, including the differences integrate in Colombian labor markets. in individual characteristics presented above in Table 1. To better understand how Venezuelan migrants are 40.  The first coefficient reported in Table 3 shows that integrating into Colombia’s labor markets vis-à-vis the local Venezuelan migrants are in general 2.3 percentage points population, accounting for observable socio-demographic more likely to be employed than other workers, controlling characteristics and for location and time fixed effects, Table for other individual characteristics and department-year- 3 presents the results of simple multivariate regression month fixed effects. The coefficient on the interaction analysis. Regressions allow to explore the effect of a certain between being a migrant and the post-2017 period, variable on an outcome of interest, conditional on a set moreover, shows that migrants are 4 percentage points of given covariates.23 Specifically, Table 3 sheds light on more likely to find employment after 2017. The crisis can be how the fact of being a migrant is correlated to changes associated with a statistically significant increase in labor in typical employment outcomes, with a special emphasis market participation of Venezuelan migrants over the pre- on differential outcomes during the migration crisis (post crisis level by 74 percent. The magnitude of impacts on the 2017). It is important to highlight that, even though the employed outcome are significant for both longer-term and estimation is based on a quasi-experimental framework (as recent migrants, although substantially bigger for the latter. the Venezuelan crisis causing the sudden surge in migration is plausibly exogenous to local conditions in Colombia), the 41.  At the same time though, migrants are also 5 percentage results presented in Table 3 are to be interpreted as simple points more likely than non-migrants to be unemployed, correlations, and do not establish causality. Migrants’ as the first coefficient in Column 3 shows. Although this choices, for example regarding location within Colombia, result may appear counter-intuitive at first, it is very telling. are likely to be linked to several confounding factors that By definition, an individual is unemployed if they are are not accounted for in the estimation, such as existing not working but are actively looking for a job.24 The fact 23  Cf. Appendix B for details. 24  In the GEIH data, this is defined over a standard 7-day recall period. 27 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR that migrants are more likely to be both employed and by the start of the migration crisis, as shown by the unemployed implies they are less likely to be inactive, i.e., insignificant coefficients on the interactions with the post- out of the labor force, as Column 5 confirms. In practice, 2017 period. Interestingly, however, female migrants are Venezuelan migrants are on average more motivated than instead less likely to be employed during the most recent the local population to pursue a job. During the migration surge of Venezuelan migrants into Colombia. This result crisis, the increase in migrants’ employment accounts for a could be interpreted by noting that, before the migration reduction in both their unemployment and inactivity. crisis, the likelihood that a woman migrant would have left her country to find employment was higher than during 42.  The rest of Table 3 confirms migrants’ willingness the crisis, when many migrants moved with the principal to endure less than ideal labor conditions and earn low motivation to be with their families. Absent further analysis, wages than the local population. This is even more true however, interpretations can only be tentative. during a forced displacement migratory crisis. Columns 7 and 8 in Table 3, for example, show that Venezuelan 45.  Similar results to the ones presented in Table 4 can migrants typically earn less than local Colombians in the be observed when longer-term and recent migrants are labor market, and that this disparity is greatly heightened separated in the estimation (Annex Table C2). Tables C3 in the crisis period. Longer-term migrants do not typically to C8 in the Annex, instead, show distinct and statistically earn less than local Colombians in labor markets, but both significant results for other outcomes – real wages, weekly longer-term and recent migrants are burdened with lower hours worked, informality for specific categories of workers. earnings post 2017. Furthermore, Venezuelan migrants work On wages, for example, there is a positive coefficient for about 1.6 to 1.8 hours more per week than local Colombians, low educated workers in the post-2017 period. The effect is and the difference increases during the crisis. Similarly, specific to longer-term migrants and mitigates the earnings migrants are 11 percentage points more likely to be informal reduction that they suffered during the migration crisis. As workers, and informality rises significantly by an additional for working hours, it looks like longer-term female migrants 5.5 percentage points for Venezuelan migrants during the work approximately one hour more in the post 2017 period, crisis. While temporality is also significantly higher among adding to the already higher hours of work incurred by Venezuelan migrants relative to local Colombians, there is migrants during the migration surge. no significant rise in this type of employment post 2017. 46.  This chapter, by comparing Venezuelan migrants 43.  Considering the dynamics of migration, and the and their Colombian peers in terms of their micro-level fact that migrants seem to be mostly clustering in cities, characteristics and labor market outcomes, allows a first Table 4 assesses whether there are any differences in the assessment of migrants’ integration in the labor force of employment outcomes of migrants in urban and rural areas. their host country, in particular as a result of the recent Predictably, migrants are more likely to be employed in migration crisis. Overall the analysis implies that Venezuelan cities and out of the labor force in rural areas, as expressed migrants are broadly likely to find employment, but that by the coefficients on the migrant-rural interaction term in this is often accompanied by poorer working conditions. Columns 1-3. However, working conditions seem to be better The next chapter delves deeper into an analysis of for migrants in rural areas: here, they earn comparatively integration of Venezuela’s forced migrants into Colombia’s higher wages, and are less likely to be hired informally labor markets, identifying opportunities and constraints for (Columns 3 and 5), even after the migration surge post 2017. migrants’ integration specifically in agri-food jobs. 44.  Table 5 presents the results of further heterogeneity analysis of employment outcomes, this time focusing on migrants’ individual characteristics such as gender, age, and education. Specifically, Table 5 explores whether the likelihood among migrants of being employed differs between (i) men and women; (ii) younger and older workers; and (iii) workers with at least secondary education and workers with lower education. Overall, the results in the table are quite compelling: among migrants, women, youth, and low-educated individuals are in general more likely to be employed than men, the elder, and better- educated workers. Most of these results are not affected 28 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Table 3 Labor outcomes of Venezuelan migrants in most recent migrant surge Real Wage (100 COP, Weekly Hours Temporary   Employed Unemployed Inactive Informal Worker at fixed 2010 prices) of Work Worker   (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) Venezuelan 0.023 0.049 -0.0712 -676.935 1.620 0.111 0.033 Migrant *** *** *** *** *** *** *** [0.007] [0.007] [0.005] [147.754] [0.345] [0.008] [0.009] Migrant* 0.040 -0.010 -0.030 -1057.827 1.599 0.055 0.001 Post 2017 *** ** *** *** *** *** [0.005] [0.005] [0.005] [219.618] [0.285] [0.008] [0.004] 0.087 -0.070 -1389.639 1.830 0.153 0.062 ST Migrant -0.017 *** *** *** *** *** *** [0.011] [0.008] [0.006] [149.617] [0.360] [0.012] [0.015] 0.058 0.015 -0.074 1.462 0.080 0.012 LT Migrant -153.670 *** ** *** *** *** ** [0.005] [0.007] [0.004] [265.339] [0.358] [0.007] [0.005] ST Migrant* 0.061 -0.018 -0.043 -1026.578 1.510 0.060 -0.001 Post 2017 *** ** *** *** *** *** [0.009] [0.006] [0.006] [150.208] [0.300] [0.013] [0.006] LT Migrant* 0.020 -0.020 -1116.627 1.674 0.054 0.000 0.004 Post 2017 *** *** *** *** *** [0.006] [0.007] [0.006] [301.445] [0.351] [0.011] [0.004] Individual YES YES YES YES YES YES YES YES YES YES YES YES YES YES controls Department *Year* YES YES YES YES YES YES YES YES YES YES YES YES YES YES Month FE Dep Var - 0.624 0.624 0.078 0.078 0.297 0.297 7538.482 7538.482 45.084 45.084 0.578 0.578 0.047 0.047 Mean Observations 1912237 1912237 1912237 1912237 1912237 1912237 1018859 1018859 1192979 1192979 1192979 1192979 1192979 1192979 R-squared 0.225 0.225 0.024 0.024 0.259 0.259 0.285 0.286 0.111 0.111 0.277 0.277 0.052 0.052 Source: Authors’ calculations using GEIH, various years. Table 4 Labor outcomes for migrants in the most recent surge – Urban vs. rural areas Real Wage (100 Weekly Informal Temporary   Employed Unemployed Inactive COP, at fixed Hours of Worker Worker   2010 prices) Work (1) (2) (3) (4) (5) (6) (7) Venezuelan Migrant 0.027*** 0.051*** -0.079*** -729.615*** 1.722*** 0.118*** 0.034*** [0.006] [0.007] [0.006] [162.969] [0.387] [0.008] [0.010] Migrant*Post 2017 0.036*** -0.013** -0.023*** -1007.181*** 1.496*** 0.048*** 0.000 [0.004] [0.005] [0.006] [233.208] [0.315] [0.008] [0.005] Migrant * Rural -0.053* -0.036** 0.088*** 628.189** -1.306 -0.090*** -0.014 [0.027] [0.014] [0.029] [240.566] [0.832] [0.015] [0.022] Migrant* Rural* Post 2017 -0.049 -0.006 0.054 744.912 2.182 -0.041 -0.070** [0.092] [0.031] [0.079] [504.431] [3.201] [0.047] [0.026] Individual controls YES YES YES YES YES YES YES Department*Year*Month FE YES YES YES YES YES YES YES Dep Var - Mean 0.624 0.078 0.297 7538.482 45.084 0.578 0.047 Observations 1912237 1912237 1912237 1018859 1192979 1192979 1192979 R-squared 0.225 0.024 0.259 0.285 0.111 0.277 0.052 Source: Authors’ calculations using GEIH, various years. 29 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Dependent Variable: Worker is employed Table 5   (1) (2) (3) Migrants’ Venezuelan Migrant 0.050*** -0.014* 0.021*** employment in the [0.011] [0.007] [0.007] most recent surge Migrant* Post 2017 0.051*** 0.031*** 0.040*** – Heterogeneity [0.008] [0.004] [0.005] analysis Migrant*Female -0.052*** Source: Authors’ [0.012] calculations using GEIH, Migrant *Female*Post 2017 -0.020* various years. [0.011] Migrant* Youth 0.124*** [0.015] Migrant*Youth*Post 2017 0.013 [0.012] Migrant*No Diploma 0.062* [0.033] Migrant* No Diploma* Post 2017 0.009 [0.033] Individual controls YES YES YES Department*Year*Month FE YES YES YES Dep Var – Mean 0.624 0.624 0.624 Observations 1912237 1912237 1912237 R-squared 0.225 0.226 0.225 30 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR COLOMBIA’S III AGRI-FOOD JOBS 31 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR III COLOMBIA’S AGRI-FOOD JOBS 47. The evidence presented in Chapter II shows that 48.  Agriculture and food systems encompass a broad range Venezuelan migrants’ likelihood to participate in labor of actors, processes, and value-adding activities involved markets is even higher than local Colombians’ themselves. in the production, aggregation, processing, distribution, At the same time however, the jobs that Venezuelans are and consumption of agricultural and food products. As finding appear to be of much lower quality if compared to such, not only do agriculture and food systems comprise the local population, with important heterogeneity across the agriculture and food products originating from primary urban and rural areas. A simple descriptive analysis of production, but also those activities related to the provision employment patterns by sector and over time (Table 6) also of inputs and services which support the flow of goods shows that a higher percentage of migrants is employed in through the different stages of the food value chain. The occupations that can be ascribed to the agriculture and food remainder of this section provides an account of Colombia’s sector, and that the share is increasing substantially over broad agri-food sector, focusing on the four major stages time. This begs the question: could Colombia’s agriculture along the food value chain. These include (i) primary, on-farm and food sector provide good employment for Venezuelan food production; (ii) food processing (the transformation of migrants? What are the characteristics of agri-food labor primary crops and livestock products); (iii) food distribution markets that could make them conducive (or not) to the (the logistical system including storage, transportation, and integration of the Venezuelan workforce into Colombian distribution of food); and (iv) food services (the final stage labor markets? This chapter offers an overview of recent of the food system, including restaurants). trends in agri-food performance in Colombia, as well as the characteristics and composition of the associated labor markets. Table 6 Sector of employment – employed population 2014 – 2016 2017 2018 2019 (Q1 & Q2)   Locals Venezuelan Migrants   (born in 12 12 12 Colombia) 5 Years 5 Years 12 Months 5 Years 5 Years Months Months Months Agri-food sector (%) 43.7 52.0 55.1 50.6 54.7 53.9 60.7 56.0 59.6 Primary agriculture (%) 61.36 54.82 44.90 44.47 23.90 32.52 26.76 28.63 21.48 Food processing (%) 11.64 10.91 8.00 12.93 12.57 15.97 14.06 16.22 12.58 Food services (%) 27.00 34.27 47.10 42.61 63.53 51.52 59.18 55.15 65.94 Non agri-food sector (%) 56.3 48.0 44.9 49.4 45.3 46.1 39.3 44.0 40.4 Manufacturing (%) 69.16 62.95 59.69 58.69 63.41 62.01 62.15 67.16 67.02 Services (%) 30.84 37.05 40.31 41.31 36.59 37.99 37.85 32.84 32.98 Source: Authors’ calculations using GEIH, various years. 32 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Box 2 Agriculture in Colombia: a macro perspective Taking into account only primary production, Colombia’s agriculture has been second only to services in terms of sectoral growth, with an average annual growth rate of 2.9 percent between 2009 and 2018 (Figure 10). This figure is ahead of what can be observed for a number of agriculture giants in Latin America, such as Brazil, Mexico, and Argentina (Figure 11). The agriculture sector is also well positioned on international markets: in 2018 the value of agriculture, food and beverage exports was 17.5 percent of the country’s total export value, and the country displays a positive agri-food trade balance. Colombia has been for decades a leading exporter of coffee, cut flowers, and bananas, which jointly make up for more than 60 percent of Colombia’s agri-food export, and there is great potential to further diversify the country’s export basket.25 Colombia’s agri-food sector has potentially much to gain from the combination of a rapidly expanding domestic market and a global growth in food demand originating from increasing global population and rising incomes. With a current population of 45.5 million26 projected to increase to 53.6 million by 2030,27 and a sustained urbanization trend, the nature of domestic demand for food and agriculture products in Colombia is evolving – not only because the number of mouths to feed is increasing, but also because urban consumers are increasingly demanding more diversified and nutritious foods, as well as more convenient food available for purchase through retail channels. An important window of opportunity is opening up for Colombia’s agricultural products in external markets as well: as the world population grows from 7 billion in 2010 to a projected 9.8 billion in 2050, global food demand is expected to increase by 50 percent in 2050 (WRI, 2018), and global income gains will direct much of this growth towards more diversified diets including better quality staple grains, fruits and vegetables, and animal-source. The extent to which this momentum can translate into a boost for agriculture in Colombia will depend on the ability of the sector to keep pace with demand. The recently declining contribution of agri-food to total export earnings (from 22.1 percent in 2016 to 129.4 percent in 2017 to 17.5 percent in 2018),28 or the fact that the rate of agricultural TFP growth and of agricultural labor productivity in recent years have been among the lowest in the Latin America region (Figure 12 and Figure 13), for example, may threaten the potential of agriculture in the face of favorable domestic and global developments. 4,5 Figure 10 4 Annual growth 3,5 of sectoral value 3 2,5 added, Colombia, 2 2009-2018 average 1,5 (%) 1 Data source: World 0,5 Development Indicators, 0 2019 Serv ces A r culture, Industr M nuf tur n Tot l forestr , ( nclud n GDP nd f sh n construct on) 25  Trade data from DANE, 2018. 26  2018 National Census. 27  UN-DESA population estimates. 28  Trade data from DANE, 2018. 33 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Box 2 continued Agriculture in Colombia: a macro perspective Figure 11 8 7 Annual growth 6 5 of agriculture 4 value added, Latin 3 American countries, 2 1 2009-2018 0 average (%) -1 -2 Data source: World Development Indicators, s r l co le r do do z l u c n m lv b 2019 r Ch u u R ex m du t m Br en Bo n lv u r ru te lo M on Ec st r P Co c U S Ar Co P H u N El G Figure 12 3 Average annual 2,5 growth rate of 2 agricultural TFP 1,5 (%): Latin American 1 countries, 2005-2015 0,5 Data source: USDA, 2018. 0 u l le r do z n el r b lv Pe Ch u u u t om Br en Bo ez u ru Ec r l n Co U r Ve P A Figure 13 70 60 Average annual 50 growth rate of 40 agricultural labor productivity (%): 30 Latin American 20 countries, 2005-2015 10 Source: Calculations from 00 USDA, 2018. r le l u do z n el r b lv Pe Ch u u u t om Br en Bo ez u ru Ec r l n Co U Ar Ve P Note: Labor productivity computed as output per worker. Gross agricultural production measured in constant 2004- 2006 international dollars; agricultural labor measured in persons aged 15+ economically active in agriculture. 34 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR A. CHARACTERIZING AGRI-FOOD young people are more likely to migrate out of rural areas, they are leaving behind an ageing and less skilled on-farm JOBS ALONG THE VALUE CHAIN labor supply. I. PRIMARY PRODUCTION 52.  The average age of agricultural producers is higher than 40, and their education level is quite low: on average, 49.  After a few years of declining agricultural share of total primary agriculture workers attended 5.5 years of schooling, employment in the early 2010s, agriculture employment and more than half of them only attained basic primary has slightly regained momentum in 2015, when both the education (Table 7). The universe of agricultural producers absolute number of workers employed in agriculture and in Colombia is also predominantly male, with women the agricultural employment share started growing again: accounting for only 20 percent of primary agriculture jobs. as of 2017, primary agriculture in Colombia employed 3.67 In terms of job quality, around 5 percent of workers is hired million workers and accounted for 16.5 percent of total as daily laborer, and another 4 percent as a seasonal worker. employment, up from respectively 3.5 million and 16 percent Only a third of primary agriculture workers is hired with a in 2015.29 regular contract. 50.  Overall, the Andean region is where most land is destined to agri-food uses (Figure 14) and most primary agricultural activities take place, with different departments specializing Table 7 Worker profiles: primary agriculture, year 2018 in the production of specific crops.30 Cattle ranching, which Socio-demographic in 2019 provided about 6 percent of all jobs in the country Women (%) 19.18 and about 19 percent of agricultural jobs (FEDEGAN, 2019), Average age (years) 43.57 is in turn prevalent in the Caribbean and Orinoquía regions, Rural area (%) 83.68 which host most of the country’s herd of cattle (Figure 15). Predictably, four out of five primary agriculture workers Education live in rural areas. Consistent with the spatial distribution Average years of education 5.49 of agricultural land, the bulk of these jobs are located in the No education (%) 11.74 Andean region (Figure 16).31 Preschool (%) 0.01 Basic primary (%) 51.8 51.  Because of structural transformation, process Basic secondary (%) 16.67 mechanization, mounting urbanization, and changing High school (%) 15.87 aspirations, many workers are increasingly leaving rural areas and moving to cities. If the process is to an extent a Superior/university (%) 3.88 physiological component of economic development along Job quality Colombia’s growth path, it is nonetheless leaving gaps in Average week working hours 40.63 the size and composition of the labor force in primary Average salary (COP) 565,626.53 agriculture. Many agricultural value chains (most notably Daily laborer (%) 5.16 coffee, cf. Box 2) have been experiencing significant labor Seasonal worker (%) 4.52 shortages at key stages of the productive process, which Permanent (%) 90.29 significantly raises production costs. At the same time, as Has contract (%) 33.27 Source: Authors’ calculations using GEIH, 2018. 29  Data from FAOStat. 30  Valle del Cauca, for example, has the highest production of large-scale cash crops, mostly driven by sugarcane production. Coffee, the most prevalent cash crop in Colombia, is instead a special case, as it is harvested across different regions of the country. Cereals are mainly produced in Tolima where most of the rice and maize are cropped, whereas Antioquia and Tolima are the main fruit-producing departments, with most of the land devoted to the production of avocadoes and bananas. Finally, Cundinamarca is the main producer of roots, tubers (potatoes in particular) and plantains as well as other vegetables, beans, and peas. Cf. Appendix D. 31  According to the latest Agricultural Census (2014), in particular, the departments with the highest number of individuals employed in agricultural activities are Nariño (512,500 workers), Boyacá (482,500), Cundinamarca (444,400), and Antioquia (396,700). 35 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Figure 14 Land used in Agri-food activities Land used in agri- (in 1000 ha) food activities, by department Data source: National 100 Agricultural Survey, 2017 200 300 Heads of cattle Figure 15 Spatial location <2,500 of cattle stock in Colombia, 2018 2,501 - 10,000 Data source: FEDEGAN, 2018 10,001 - 20,001 20,001 - 50,000 50,001 - 100,000 >100,000 Figure 16 Workers (x 1,000) Permanent workers <=10 in agriculture, 2014, by department > 10 - <=50 Source: Authors’ elaboration using data from National > 50 - <=100 Agricultural Census, 2014 Note: The sample is > 100 - <=500 composed of workers who reported having worked > 500 in the previous 30 days. 36 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Box 3 Coffee pickers in Colombia: a scarce resource Labor deficits in in the coffee region have been persistent since 2013 (García et al., 2016). In 2015, Colombia’s National Coffee Growers Federation (FNC) estimated that the country’s largest coffee regions needed 20 to 40 percent more pickers, a deficit that has been not offset in more recent years: in 2018, the sector still needed around 18 percent more than the existing pool of workers in the occupation, which means a shortage of 60,000-90,000 jobs. As a result, coffee picking accounts nowadays for around 50 percent of total coffee production costs (FNC, 2017). Anecdotal evidence highlights how the coffee industry is perceived by many workers, especially the youth, as antiquated and lacking high-quality employment. The remote location of many coffee farms, away from urban areas and mostly excluded from good connectivity networks, is another element that discourages labor supply in coffee production, as is the seasonality of the crop which keeps workers’ salaries (higher than the legal minimum during harvest time) below the minimum wage on an annual basis.Traditionally, many experienced pickers move along the coffee zone (especially to large farms in the North, Centre, and Coffee Region) during the year, following the cyclical behaviour of the crop in order to maximize income from this activity. The main destinations are the departments of Huila, Antioquia and Caldas, whereas the departments that register the highest out-migration rates are El Cauca, Tolima and Valle. Still, in the last few years the labor force deficit in coffee picking throughout the country has been in the order of the tens of thousands (FNC, 2017). A recent survey on 7,578 coffee pickers conducted in 2016 offers some insights on the demographic, socioeconomic, and labor market outcomes of the workers engaged in this activity, as well as on the aspirations and challenges that affect their decision whether to remain in the coffee industry. Based on the survey results, the labor force in coffee collection is predominantly male, while women represent only 14.2 percent of the total. Women are particularly active in small farms, where they generally contribute family labor. The educational level in the survey population is relatively low: 35 percent of the sample had not completed primary school at the time of the interview, and another 29 percent had only received primary education. Women have an average of 1.3 years more education than men, and generally reach a higher level of education. Although the low educational level of the collecting population does not seem to impact significantly the performance of manual labor, it has been found to generate limitations in terms of technology adoption, in the speed of implementation of new practices, and in the social organisation of work. In terms of work conditions, informality predominates in the coffee collection activity: only 1.5 percent of the interviewed pickers were hired as regular employees, whereas 50 percent had a piecework contract, 37.5 percent were daily hires, and 11 were unpaid family workers. As a result, the rate of enrolment in old-age protection or work insurance schemes is also very low (4 and 3 percent, respectively). Expectedly, the level of formality also influences workers’ earnings: those with a piecework contract have an average monthly income 1.6 times higher than the legal monthly minimum wage, whereas daily laborers only receive on average 0.6 times the minimum wage (and women are on average paid 10.7 percent less than men). There are, however, substantial differences in remuneration across regions of the country. Most workers also receive in-kind benefits such as meals on the job. The most influential determinants of pickers’ decision on where to work are payment level, humane treatment, and food quality; women also consider the proximity of the workplace to their place of residence. Conversely, the major discouraging factors pertain to low employment quality, especially in terms of informality in hiring, seasonality of income, and virtually no social protection. Indeed, the vast majority of collectors agree that they would remain in the work if they had a better income, affiliation to the social protection system and a settlement at the end of the work. 37 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR II. FOOD PROCESSING average higher-quality employment prospects: in 2018, approximately 58 percent of food processing jobs were 53.  The rapidly expanding agri-food manufacturing sector in the formal sector, more than 70 percent above the in Colombia displays important opportunities to increase corresponding figure for primary production. The average value addition in the agriculture and food sector by wage was also 70 percent higher than in primary agriculture. converting raw materials into high-value products. 54.  Among the country’s rural units of production, the 2014 Agricultural Census reports that 29.3 percent of agricultural III. FOOD DISTRIBUTION, RETAIL, AND FOOD units and and 17.3 percent of non-agricultural units perform SERVICES some kind of on-farm activities related to the processing of agricultural products,32 mostly in the Andean region 57. Food distribution mainly takes place at wholesale food (Figure 17).33 markets (centros de abasto) located in the main urban centers. There are in total 30 wholesale food markets in 55.  As for off-farm activities, the Colombian agro-industry Colombia, mostly spread throughout the North-West of the has recently experienced a sustained increase in market country (Figure 18).35 These markets work at both large and demand for value-added food products, and is undergoing small scale (in smaller cities), aggregating food in central substantial consolidation and structural change through markets and then re-distributing back to urban and rural mergers, acquisitions, and foreign investment (USDA, retail points like local markets, shops, and supermarkets. 2019). The food processing industry is ramping up its Apart from large agricultural trading, wholesale markets competitiveness and performance, updating its technological offer job opportunities to bulk loaders or small food platforms and investing in product and process innovation retailers, mostly at the informal level. to boost performance on domestic and international markets (ANDI, 2019). The National Superintendence for 58.  In the final stage of the food value chain, Enterprises (Supersociedades)34 reports 2,032 formally Supersociedades counted 1,128 formal enterprises engaged registered commercial agri-food processing firms in 2019, in food retail and food services in 2019, 57 percent of which with a very high degree of spatial concentration: almost located in Bogotá and the Antioquia department.36 The a third of all registered firms are concentrated in Bogotá, picture has to be complemented by a broad constellation while the departments of Antioquia, Valle del Cauca, and of informal establishments such and unregistered food Cundinamarca make up for another 40 percent of the total. retailers. Even though traditional retail remains the most common retail format in the country, especially in smaller, 56.  Compared to primary agriculture, food processing jobs remote towns still excluded from modern networks, are equally split between men and women, and they are, the Colombian urban food retail sector is undergoing a predictably, overwhelmingly located in non-rural areas “supermarket revolution” similar to many South American (Table 8). The sector also attracts a better-educated pool of countries. Over the last decade, large supermarkets and workers, three quarters of whom have at least secondary grocery chains (both domestic and international), as well education. Moreover, food processing seems to offer on as cheaper discount outlets, have been opening new stores at intense rates, and food e-commerce has recently been 32  The Census collects information on the following food processing activities: gaining momentum (Euromonitor, 2019). (i) extraction of oils; (ii) sugar manufacturing; (iii) rice milling; (iv) production of panelas and honeys; (v) processing and transformation of flora products 59.  On the food services side, the latest Annual Survey (tubers, fruits, flowers, leaves, bark and resins); (vi) meat products processing; of Services (EAS, 2017) reports that restaurants, bars, and (vii) milk processing; and (vii) production of food for human consumption. catering services account for around 5 percent of all formal Among these, the most prevalent on-farm food processing activity among employment in the formal services sector. The Colombia Units of Non-Agricultural Production (UNAPs) is the production of panelas Restaurant Association (ACODRES, 2019),37 estimates that and honeys (practiced by 11,980 UNAPs), followed by the production of food there are currently 44,000 restaurants in the country’s for human consumption (3,944 UNAPs). For Units of Agricultural Production formal sector, and that informal establishments could be (UAPs) performing non-agricultural activities, the most popular are elaboration almost as many. In particular, large cities have recently of panelas and honeys, and production of food for human consumption 33  The departments of Boyacá and Huila host the highest number of jobs 35  Data for 2019 from the Information System on Agricultural Prices (Sistema in on-farm food processing activities, with respectively 302,902 and 123,353 de Información de Precios Agrícolas, SIPSA), which follows the main food workers. markets in Colombia. 34  The types of enterprise required to register with Supersociedades are Sole 36  The picture cannot but be incomplete, as the food service scene is proprietorship, Collective Society, Limited Partnership, simple and by shares, constellated by a much broader net of informal establishments, such as Limited Liability Company, Limited Company. As it ignores the informal sector, unregistered food retailers and convenience stores, food trucks, and informal this information only constitutes a lower bound for the universe of industrial restaurants, cafés, and bakeries. agri-food activity in Colombia. 37  http://acodres.com.co/Bogotá/quienes-somos-2/. 38 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR been experiencing, on one side, the growth of the high-end processing, with a higher prevalence of workers with restaurant niche, and, on the other, a substantial dynamism just basic secondary education, but both educational in the fast food segment. As a result of these parallel attainment and years spent in school are still significantly tendencies, the food services industry has become a major higher than in primary agriculture. Job quality presents a employer, especially in larger urban centers: in Bogotá alone, mixed picture, with average wages and informality levels in the formal branch is estimated to generate around 600,000 between those observed for food processing and on-farm direct and indirect jobs (ACODRES, 2019). agriculture. Interestingly, tertiary food activities show the lowest percentage of permanent employees along the food 60.  Among those employed in food retail and services, value chain, with more than 10 percent of workers hired in almost 70 percent are women, and less than 20 percent daily occupations, as well as the highest number of average live in rural areas (Table 9). The overall education level of working hours per week. workers in this sub-sector is slightly lower than in food Figure 17 Figure 18 Workers in on-farm food processing, 2014, by department Location of food markets in Colombia, 2019 18 - 594 - 1683 - 6220 - # Food markets 1 2 3 4 594 1683 6220 302910 Data source: National Agricultural Census, 2014 Data source: SIPSA, 2019 Note: the sample is composed of workers who reported having worked in the previous 30 days. 39 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Table 8 Table 9 Worker profiles: food processing, year 2018 Worker profiles: food services and retail, year 2018 Socio-demographic Socio-demographic Women (%) 51.79 Women (%) 69.08 Average age (years) 40.02 Average age (years) 39.05 Rural area (%) 24.94 Rural area (%) 18.88 Education Education Average years of education 9.33 Average years of education 8.94 No education (%) 3.49 No education (%) 3.3 Preschool (%) 0.04 Preschool (%) 0 Basic primary (%) 22.06 Basic primary (%) 23.95 Basic secondary (%) 14.73 Basic secondary (%) 19.39 High school (%) 37.65 High school (%) 35.57 Superior/university (%) 21.99 Superior/university (%) 17.79 Job quality Job quality Average week working hours 41.93 Average week working hours 42.7 Average salary (COP) 962,096.92 Average salary (COP) 700,735.77 Daily laborer (%) 9.65 Daily laborer (%) 11.06 Seasonal worker (%) 0.28 Seasonal worker (%) 0.42 Permanent (%) 90.02 Permanent (%) 88.42 Has contract (%) 57.62 Has contract (%) 47.27 Data source: GEIH, 2018 Data source: GEIH, 2018. 40 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR ARE VENEZUELAN IV MIGRANTS INTEGRATING INTO COLOMBIA’S AGRI-FOOD SYSTEM? 41 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR IV ARE VENEZUELAN MIGRANTS INTEGRATING INTO COLOMBIA’S AGRI-FOOD SYSTEM? 61.  Chapter III highlights promising opportunities for regression analysis on the employed labor force38 in Table the agriculture sector in Colombia to expand and absorb 10 reveals a significant, positive correlation between new workers. Agriculture value added has been growing employment in the broad agri-food sector and migration more than most other economic sectors in the country, status, conditional on standard socio-demographic the agriculture trade balance is positive, and a rapidly characteristics such as gender, age, and education, and time increasing demand for food both on international and and district fixed effects. In column (1) of the table, the domestic markets can be a catalyst for further growth in coefficient in the first row means that being a Venezuelan the performance and relevance of the sector in the country. migrant is associated, controlling for covariates, to an At the same time, certain labor-intensive subsectors are increase in 5.2 percentage points in the likelihood of being experiencing a shortage of workers for specific primary employed in agri-food activities versus employment in other jobs, for example coffee pickers during peak season. As the sectors. Although the number may seem relatively small at country’s food systems modernize, moreover, more and first sight, it represents a 31 percent increase with respect more opportunities for jobs beyond primary production to the overall probability of employment in the agri-food are being created downstream along the food value sector for the entire working-age population across years chain, in processing and added-value activities, with on (found on the first row of the table’s second panel). average better wages and work conditions. At the same time, overlaying migration hotspots and agrifood activity 64.  Crucially, this seems to be the case in the key years of the disaggregated by subsector evidences opportunities for migration crisis, after year 2017. The coefficient in the second inclusion in specific territories (Figure 19). Taking stock from row of column (1) says the probability of being employed in this evidence, the current section seeks to respond to the agri-food as a Venezuelan migrant is 3.5 percentage points second main objective of this report, and to assess whether higher in or after 2017 than before. Venezuelan migrants Colombia’s agriculture and food systems are providing during the migration surge are therefore 8.7 percentage quality job opportunities to Venezuelan migrants. points more likely to be employed in the Colombian agri- food sector than the average person, a 52 percent increase 62.  To set the stage for the analysis, it has to be noted that, in the overall probability of employment in agri-food. at the institutional level, inclusion strategies for Venezuelan migrants could be strengthened by targeting the agri-food 65.  Interestingly, agri-food employment seems to be a sector more specifically and decisively. Notably, the main predominantly short-term strategy in general. In fact, strategy of the Colombian government for the integration although the sector employs both longer-term and recent of Venezuelan migrants, the CONPES 3950 (cf. Box 1), did not Venezuelan migrants, it does significantly more so for the contemplate budget nor action by the Ministry of Agriculture latter – almost four times as much (see the magnitudes and Rural Development (MADR). There was also no specific of the first two coefficients in column 2 of Table 10).39 In mention to initiatives for the integration of Venezuelan the aftermath of the migration crisis, however, differences migrants into agri-food labor markets in the Impact Plan between recent and longer-term migrants level up (see to mitigate the effects generated by the migration crisis the third and fourth coefficients in column 2 of Table 10), in Colombia’s border departments, announced by the a hint that agri-food markets may be playing a significant Colombian government in mid-2019, even though the plan role in absorbing Venezuelan labor, regardless of their date proposed measures relating to the promotion of investment of arrival. and employability in the agricultural sector targeting local producers affected by immigration. And the work inclusion 66.  When breaking down of the agri-food sector in primary, programs by the Department of Social Prosperity focused processing and services activities, nonetheless, a surprising on the Venezuelan population generally abstract from a pattern emerges (columns 3-8 of Table 10). In fact, despite dedicated coverage of the rural sector or agri-food workers. documented labor shortages in primary agriculture, Venezuelan migrants are consistently less likely to be 63.  In a context of relatively mild formal mechanisms employed in primary activities, and even less so after 2017 specifically designed to facilitate migrants’ inclusion in if one considers recent migrants. At the same time, there is agri-food jobs, are labor markets organically taking care of themselves? Is the agri-food sector able to attract migrant 38  Technical details are provided in Appendix section B. Qualitative results workers, absorbing part of the labor shock induced by stay the same when considering the dependent variable unconditional on being the migration wave into Colombia? Simple multivariate employed. 39  Equality of coefficients has been tested for through standard F-tests on linear restrictions. Cf. Appendix B for details. 42 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR (a) Primary agriculture Figure 19 Agri-food activity Land used in and migration Agri-food activities hotspots (in 1000 ha) Data source: National Agricultural Survey, 2017; Supersociedades, 2019; 100 GEIH, 2018. Panels (b) and (c) report the department 200 share of formally- registered food processing and food services firms 300 over the national total. #Migrants > 1.000 - <= 10.000 (b) Food processing > 10.000 - <= 50.000 > 50.000 % Food processing enterprises 10 20 30 40 (c) Food services % Food services enterprises 10 20 30 40 43 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR almost no relationship between employment in the food with the addition of two measures of agricultural intensity processing industry and the status of Venezuelan migrant, at the department level. Specifically, columns 1, 3, 5, and 7 with only a tiny employment probability increase just include the department’s share of agriculture land over its for recent migrants after 2017. The bulk of the correlation total land area, whereas columns 2, 4, 6, and 8 include a between agri-food employment and Venezuelan migration binary indicator equal to 1 if a department lies above the seems instead to stem almost exclusively from employment 75th percentile of the distribution of agricultural land share in food services. across departments in Colombia. In theory, if it were just a matter of saturated agricultural labor markets, one should 67. To an extent, this is understandable in the case of recent expect to find a positive and significant coefficient on the migrants, who tend to cluster into cities where they are more interaction between migrant and agriculture intensity in likely to find informal social networks and flexible labor the regression for employment in primary agriculture. As is markets (including in food shops and restaurants). The fact apparent from the table’s results, however, this is not the that even longer-term migrants, who presumably have had case in practice. In fact, not only is agriculture intensity the time to spread across the country or to obtain a work relevant for the employment prospects of Venezuelan permit through various regularization programs, are not migrants in the agri-food sector only after 2017 (columns likely to find employment in primary agriculture activities or 1-2), but it seems to matter only in the case of employment food processing however speaks of an unrealized potential in food services, rather than in primary agriculture or in the sector – which is even more remarkable in light of food processing. the agricultural labor shortages and agro-industrial growth highlighted in Chapter III. 69.  What happens in areas with a higher specialization in food processing and food services? Table 12 and Table 13 68.  One may still argue that migrants might overwhelmingly replicate the exercise in Table 10 using measures relating to locate and remain into areas where agricultural activities the department-level share of, respectively, food processing are less prevalent, and where agricultural labor markets and food retail and services enterprises over the total of are therefore already saturated by the local workforce registered firms, using data from Supersociedades. As – but that things would be different in more markedly stressed in Chapter III, this information does not account agricultural areas. To account for this possibility, the for the universe of informal or unregistered firms in both regressions in Table 11 replicate the specifications in Table 10 sectors and is thus a rather crude proxy of the intensity of Table 10 Agri-food sector outcomes for migrants in recent migration surge Employed in Agri-food Employed in Primary Employed in Food Employed in Food   sector Agriculture Processing Sector Services Sector   (1) (2) (3) (4) (5) (6) (7) (8) Venezuelan Migrant 0.052*** -0.020*** 0.001 0.071*** [0.007] [0.004] [0.002] [0.005] Migrant* Post 2017 0.035*** -0.007 0.005** 0.037*** [0.007] [0.005] [0.002] [0.007] ST Migrant 0.087*** -0.021*** 0.000 0.108*** [0.009] [0.003] [0.003] [0.008] LT Migrant 0.025*** -0.019*** 0.002 0.042*** [0.008] [0.006] [0.003] [0.006] ST Migrant*Post 2017 0.022* -0.011** 0.007* 0.026*** [0.012] [0.005] [0.004] [0.009] LT Migrant* Post 2017 0.047*** -0.005 0.004 0.047*** [0.008] [0.007] [0.003] [0.008] Individual controls YES YES YES YES YES YES YES YES Department*Year*Month FE YES YES YES YES YES YES YES YES Dep Var - Mean 0.168 0.168 0.074 0.074 0.029 0.029 0.065 0.065 Observations 1192979 1192979  1192979  1192979  1192979  1192979 1192979  1192979 R-squared 0.147 0.147 0.303 0.303 0.007 0.007 0.039 0.039 Source: Authors’ calculations using GEIH, various years. 44 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Table 11 Agri-food employment for migrants in agriculture-intense locations Employed in agri-food Employed in primary Employed in food Employed in food   sector agriculture processing services   (1) (2) (3) (4) (5) (6) (7) (8) Venezuelan Migrant 0.059*** 0.053*** -0.018** -0.020*** 0.002 0.001 0.076*** 0.072*** [0.014] [0.007] [0.007] [0.004] [0.003] [0.002] [0.009] [0.005] Migrant* Post 2017 0.025** 0.030*** -0.001 -0.005 0.006 0.005** 0.019* 0.030*** [0.009] [0.008] [0.007] [0.005] [0.004] [0.002] [0.010] [0.006] Migrant* Share of agriculture -0.090 -0.020 -0.006 -0.064 in department [0.124] [0.057] [0.041] [0.069] Migrant*share of ag*Post 2017 0.132** -0.075 -0.010 0.216** [0.052] [0.058] [0.048] [0.080] Migrant*Intensive agriculture -0.016 -0.004 0.000 -0.013 in department [0.015] [0.007] [0.007] [0.009] Migrant*Intense*Post 2017 0.045** -0.014 0.000 0.058*** [0.016] [0.009] [0.010] [0.019] Individual controls YES YES YES YES YES YES YES YES Department*Year*Month FE YES YES YES YES YES YES YES YES Dep Var – Mean 0.168 0.168 0.074 0.074 0.029 0.029 0.065 0.065 Observations 1192979 1192979 1192979 1192979 1192979 1192979 1192979 1192979 R-squared 0.147 0.147 0.303 0.303 0.007 0.007 0.039 0.039 Source: Authors’ calculations using GEIH, various years. secondary and tertiary food activities in every department. channels, the presence of larger registered enterprises Absent any readily available data on agri-food informal would not be able to capture these movements. activities and smaller establishments at the department level for the entire country, the Supersociedades measures 72.  Clearly, even though the regressions in Table 10-12 still offer useful evidence in this sense. control for basic covariates and fixed effects, it is quite likely that at least part of the picture just described might be the 70.  In Table 12, migrants are overall less likely to be result of a number of confounding factors. In this sense, employed in agri-food (overall, and in any of the sub- it is important to highlight that the analysis presented in sectors specifically) in areas with a higher presence of this section is reporting a set of correlations and is silent food processing firms. In the years of the migration crisis, on the causal effect of migration on employment. For however, their likelihood to find employment in these areas example, this section is showing that, conditional on certain increases both for food processing and, more surprisingly, explanatory variables, a Venezuelan migrant is more likely for primary agriculture. The latter result speaks interestingly to be employed in the agri-food sector than a non-migrant, of the linkages that food processing firms might have but there could be a host of reasons for why this may be with agricultural activity on the territory, although this the case, including unobserved individual heterogeneity – interpretation can only be speculative in the absence of a for instance and very possibly, if migrants were on average more formal test of the mechanism. more willing to endure harder working conditions which in turn might be more prevalent in the agri-food sector. For 71.  In Table 13, on the other hand, a higher concentration the sake of this report, nevertheless, it also matters that of food retail and services firms does not seem to matter this relationship exists, even though it may not be driven much for the agri-food employment prospects of migrants. solely by a person’s migrant status per se. If anything, there is a hint that migrants are overall less likely to be employed in food services in areas with a 73.  To gain more insight on relevant characteristics of the higher share of registered food retail and services firms. For agri-food labor force, Table 14 compares migrants and as counterintuitive as this result may seem at first sight, non-migrants on basic descriptive statistics, breaking down it might nonetheless be rationalized by stressing that the the labor force between agri-food and non-agri-food food service-intensity measure is constructed using only workers, and then further splitting the latter along primary, registered enterprises in the formal sector. If migrants went processing, and services activities for 2018 at the peak of into the service sector through predominantly informal the migrant surge. A gender disaggregation across different 45 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR sub-sectors points to a consistent rise in the share of women of youth participating in the non-agri-food sector is 10 and working in upstream activities in the agri-food sector. nearly 20 percentage points higher among longer-term and Nearly 70 percent of local Colombians working the food recent migrants, respectively, vis-à-vis the local Colombian services industry are women, relative to 58 and 60 percent population. This gap increases when looking at the agri- of longer-term and recent migrants respectively. Although food sector, and while 26.9 percent of local agri-food the shares of women working in food services is lower workers are youth, the share rises to 42 and 52 percent for among migrants compared to local Colombians, it is the longer-term and recent migrant agri-food workers. Among sub-sector with the highest participation of women, except recent migrants, the share of workers that are youth and for among recent migrants. As a reference, in non-agri food participate in food services (55) is higher than for food sectors female participation is 44, 35 and 34 percent for processing (51) and primary agriculture (48). However, for local Colombians, longer-term and recent migrants. Among longer-term migrants a larger share participates in food short term migrants’ 61 percent of workers are women in processing than in services or on farm. food processing. Women are also in the majority among longer-term migrants participating in food processing at 55 75.  Chapter II of this report has already established that percent. On the other hand, migrant Venezuelan women’s migrants typically work longer hours than local Colombians participation in primary agriculture is lower than women for lower pay. How does this translate by sector of local Colombians and generally very low. employment? From Table 14 it is evident that migrants on average work almost 5 hours more than the average 45 hours 74.  Youth participation in different segments of labor of work per week of local Colombians in the non-agri-food markets is at the same time an important indicator of sector. It is interesting that on average migrants and local current demographics in a population (or among sub- Colombians work longer hours in the non-agri-food sector populations like the migrant one), on the one hand, and of than in agri-food. Among local Colombians and longer-term the skills and employability of the younger population, on migrants, this holds true for work disaggregated by primary the other. In this case, given the increasingly high share of agriculture, food processing and food services, where work young migrants during the Venezuelan crisis, understanding in these sub-sectors is 2 to 5 hours less than in non-agrifood. the dynamics of labor market absorption of youth across On the other hand, recent migrants work an equivalent 50 productive sectors and sub-sectors is essential. The share Table 12 Agri-food employment for migrants in food-processing-intense locations Employed in agri- Employed in primary Employed in food Employed in food   food sector agriculture processing services   (1) (2) (3) (4) (5) (6) (7) (8) Venezuelan Migrant 0.077*** 0.060*** -0.006 -0.016*** 0.006* 0.003 0.076*** 0.073*** [0.016] [0.009] [0.004] [0.004] [0.003] [0.002] [0.014] [0.007] Migrant* Post 2017 0.018 0.027*** -0.024*** -0.013*** -0.003 0.002 0.046*** 0.038*** [0.012] [0.008] [0.007] [0.004] [0.004] [0.002] [0.011] [0.006] Migrant* Share of food process- -0.393** -0.224*** -0.079** -0.090 ing firms in department [0.171] [0.076] [0.032] [0.154] Migrant*share of food proc *Post 0.245 0.274*** 0.137** -0.166 2017 [0.167] [0.095] [0.049] [0.196] Migrant*Intensive food process- -0.025** -0.012* -0.007*** -0.006 ing in department Migrant*Intense*Post 2017 [0.010] [0.007] [0.002] [0.008] 0.024 0.021*** 0.012*** -0.008 [0.016] [0.007] [0.004] [0.017] Individual controls YES YES YES YES YES YES YES YES Department*Year*Month FE YES YES YES YES YES YES YES YES Dep Var – Mean 0.167 0.167 0.072 0.072 0.030 0.030 0.065 0.065 Observations 1107113 1107113 1107113 1107113 1107113 1107113 1107113 1107113 R-squared 0.143 0.143 0.301 0.301 0.007 0.007 0.038 0.038 Source: Authors’ calculations using GEIH, various years. 46 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Table 13 Agri-food employment for migrants in food-retail-intense locations Employed in agri- Employed in primary Employed in food Employed in food   food sector agriculture processing services   (1) (2) (3) (4) (5) (6) (7) (8) Venezuelan Migrant 0.058*** 0.058*** -0.017*** -0.017*** 0.002 0.003 0.074*** 0.072*** [0.009] [0.009] [0.004] [0.004] [0.002] [0.002] [0.007] [0.007] Migrant* Post 2017 0.035*** 0.029*** -0.009* -0.013*** 0.003 0.002 0.040*** 0.040*** [0.008] [0.008] [0.004] [0.004] [0.002] [0.002] [0.007] [0.006] Migrant* Share of food retailing -0.084 0.011 0.011 -0.106*** firms in department [0.060] [0.020] [0.021] [0.034] Migrant*share of food retail*Post -0.107 -0.167** -0.019 0.079 2017 [0.120] [0.067] [0.054] [0.117] Migrant*Intensive food retailing -0.015 -0.001 -0.004 -0.010 in department [0.012] [0.004] [0.002] [0.010] Migrant*Intense*Post 2017 0.032 0.004 0.006 0.022 [0.020] [0.008] [0.005] [0.020] Individual controls YES YES YES YES YES YES YES YES Department*Year*Month FE YES YES YES YES YES YES YES YES Dep Var – Mean 0.170 0.170 0.076 0.076 0.029 0.029 0.064 0.064 Observations 1086132 1086132 1086132 1086132 1086132 1086132 1086132 1086132 R-squared 0.154 0.154 0.314 0.314 0.007 0.007 0.038 0.038 Source: Authors’ calculations using GEIH, various years. hours per week in the non-agri-food and food services potential difficulty in integrating migrants into Colombia’s sectors, while working 42 and 44 hours in food processing upstream agri-food sector, despite documented labor and primary agriculture. Recent migrant respondents would shortages in primary agriculture and potential for inclusion have only been in Colombia for 12 months or less at the time in agriculture value-chains. of the survey, and the majority are located in urban centers, working in food services. This special group of migrants are 77.  Table 14 points at evident significant disparities in wage less `settled’, and one can postulate they are more likely and working hours between migrants and non-migrants to work longer hours to compensate for lower wages and employed in agri-food. To further investigate this point, greater recent needs. Table 15 takes a closer look at labor conditions for those employed in the agri-food sector, again via multivariate 76.  Wages in the agri-food sector are typically lower regression analysis. With the same caveat as above that the than non-agri-food, and migrants earn less than local reported coefficients identify correlations and not causal Colombians across the board. As mentioned in Section II, effects, the picture that is painted in Table 15 is strikingly this could signal willingness to work for lower wages, but simple: Venezuelan migrants employed in the agri-food also lower willingness to pay among employers for migrant sector are on average facing significantly worse work labor. Local Colombians in the non-agri-food sector earn conditions. Despite working longer hours, they receive a around COP 781,200 (2010 real value) in wages per week lower wage, and they are more likely to be hired informally whereas longer-term and recent migrants in non-agri-food or on temporary contracts. Results tend to hold for recent earn 21 and 44 percent less respectively. Quite expectedly, and longer-term migrants alike. Put differently, migrants recent migrants fare the worst both in the non-agri-food are willing to fill positions that are not as attractive to and in the agri-food sector. Longer-term migrants earn on Colombian workers. Vulnerability and economic need, average COP 613,300 in non-agri-food and COP 478,800 and in many cases the lack of a regular residency or work in agri-food. By comparison, recent migrants earn COP permit, may be making migrants more susceptible to be 438,500 in non-agri-food and COP 401,500 in the agri-food hired at worse working conditions. Consistently with this sector. There is a penalty levied on the value of a recent view, most results in Table 15 are even reinforced after 2017, migrants’ work. Further, in wage terms both longer-term with the intensification of the migratory crisis. and recent migrants fare better in food services than in primary agriculture and food processing. This points to a 47 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Table 14 Characteristics of migrants and local Colombians employed in the agri-food sector in 2018 Non Agri-food Primary Food Food   agri-food Overall agriculture Processing Services Local Colombians Avg. share female 44.00 34.60 19.19 51.66 70.14 Avg. age 35.00 40.41 41.36 38.72 38.64 Avg. share youth 25.50 26.86 25.47 28.02 30.32 Avg. hours worked per week 45.12 41.11 40.31 43.26 42.10 Avg. wage (in 100 COP, at fixed 2010 prices) 7811.76 4779.26 4151.37 7089.35 5171.59 Migrants 5 year Avg. share female 35.34 42.36 12.60 54.57 57.54 Avg. age 34.51 33.47 34.21 32.52 33.30 Avg. share youth 35.95 41.84 40.31 44.70 41.90 Avg. hours worked per week 49.53 45.57 44.35 41.17 47.76 Avg. wage (in 100 COP, at fixed 2010 prices) 6133.48 4788.10 4287.64 4374.92 5264.48 Migrants 12 month Avg. share female 34.19 47.98 15.50 61.34 59.90 Avg. age 32.03 31.06 33.17 31.21 30.04 Avg. share youth 44.92 52.30 48.32 50.98 54.46 Avg. hours worked per week 50.48 47.37 44.03 42.36 50.12 Avg. wage (in 100 COP, at fixed 2010 prices) 4384.50 4014.80 3308.19 3902.18 4320.92 Source: Authors’ calculations using GEIH, 2018. Table 15 Labor conditions for migrants employed in agri-food sector Dependent variable conditional on being employed in agri-food sector   Real Wage (in 100 COP, at fixed Weekly Hours of Work Informal Worker Temporary Worker   2010 prices) (1) (2) (3) (4) (5) (6) (7) (8) Venezuelan Migrant -696.407*** 2.357*** 0.080*** 0.028*** [202.199] [0.596] [0.016] [0.010] Migrant* Post 2017 -675.908*** 2.396*** 0.025* -0.004 [115.069] [0.614] [0.014] [0.006] ST Migrant -1206.444*** 3.262*** 0.097*** 0.041** [264.144] [0.802] [0.022] [0.016] LT Migrant -245.815 1.512** 0.064*** 0.017** [176.586] [0.547] [0.019] [0.008] ST Migrant*Post 2017 -732.916*** 1.422 0.047* 0.010 [162.390] [0.875] [0.023] [0.009] LT Migrant* Post 2017 -676.881*** 3.298*** 0.010 -0.013* [156.002] [0.650] [0.020] [0.007] Individual controls YES YES YES YES YES YES YES YES Department*Year*Month FE YES YES YES YES YES YES YES YES Dep Var - Mean 4864.072 4864.072 42.395 42.395 0.752 0.752 0.044 0.044 Observations 168089 168089 199897 199897 199897 199897 199897 199897 R-squared 0.167 0.167 0.157 0.157 0.197 0.197 0.066 0.067 Source: Authors’ calculations using GEIH, various years. 48 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR A. ADDITIONAL EVIDENCE: THE social protection systems, and social cohesion may become over-stretched by simultaneously attending these co INTERACTION WITH THE POST- existing needs. CONFLICT PROCESS 80.  Without attempting to provide an exhaustive account of the peace process in Colombia,41 this section seeks 78.  The sustained inflow of Venezuelan migrants is not to understand whether areas hosting conflict-related happening in a vacuum. Especially since the Havana peace vulnerable individuals might indeed be facing additional accords of 2016 with the Revolutionary Armed Forces of challenges in the integration of Venezuelan migrants. Colombia (FARC, from its Spanish acronym), Colombia has been in the delicate process of transitioning out of 81.  In order to address the question whether the integration its decades-long internal conflict. The conflict had been a of Venezuelan migrants might be more challenging in areas complex, cyclical phenomenon that permeated the entire already trying to tackle the labor inclusion of internally national territory and involved multiple actors with different displaced and demobilized Colombians, this section origins and strategies of violence, causing longer-termlasting replicates the empirical analysis performed in Chapters II social, economic, and political impacts and generating deep and III of this report, accounting for the presence of conflict- wounds within the country’s civic framework (Ibarra, 2018). related vulnerable populations. As the exact identification The peace process entails the double challenge of catering of the demobilized and former displaced population is not to the victims of forced displacement, as well as managing possible within the GEIH survey, the analysis follows the the reintegration of ex-combatants into the social and indirect identification approach proposed by Calderon and economic life of the country and is clearly a top national Ibañez (2016), which is based on available information on priority as the country regains its stability after the fragility the Colombian population that moves within the country. and violence that plagued the domestic scene for more In particular, this section will consider an internal migrant than 50 years.40 that reported, as their main reason for migrating, conflict, violence, or other related threats a reasonable proxy for 79.  For what concerns the management of the Venezuelan either an internally displaced person or an ex-combatant. crisis, this not only means a likely competition for Colombian social services and social protection systems resources – 82.  Figure 20 plots the number of individuals who reported but also, crucially for this study, potential local labor market moving in the last 12 months due to the armed conflict. challenges, to the extent that the areas most affected by Even though this method cannot tell the displaced and migration also receive significant numbers of demobilized demobilized apart, it reveals that, on average, these and internally displaced people. While the fair and effective movements have involved 13,562 individuals per year management of the post-conflict transition is clearly a throughout the 2014-2019 period: that is to say, every year separate question from the response to the migration crisis, 13,562 new people have relocated as a consequence of there is a non-negligible risk that Colombian resources, the conflict. 40  Annex C provides a brief characterization of the two types of conflict- 41  For excellent reviews, see for example research produced by the Kroc related vulnerable populations (victims of forced displacement and Institute for International Peace Studies of the University of Notre Dame. reintegrating ex-combatants), focusing on their geographic concentration and their integration into local labor markets. Figure 20 Recent internal 40 migration due to armed conflict, 2014- 30 2019 (by trimester) Source: Authors’ 20 elaboration, using data from GEIH, various years. 10 Intern l m r nts ( n 1000 people) 0 16 17 18 19 20 20 20 20 49 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Table 16 Labor market outcomes for migrants in recent migration surge, accounting for conflict-related internal migration Real Wage (100 Weekly Informal Temporary   Employed Unemployed Inactive COP, at fixed Hours of Worker Worker   2010 prices) Work (1) (2) (3) (4) (5) (6) (7) Venezuelan Migrant 0.023** 0.048*** -0.071*** -647.495*** 1.566*** 0.105*** 0.043*** [0.008] [0.008] [0.006] [181.464] [0.359] [0.007] [0.009] Migrant*Post 2017 0.040*** -0.010 -0.030*** -1031.656*** 1.385*** 0.063*** -0.013* [0.008] [0.007] [0.005] [226.556] [0.361] [0.012] [0.006] Migrant*High share of internal 0.000 0.002 -0.002 -126.892 0.221 0.025 -0.038*** migrants [0.017] [0.012] [0.011] [205.559] [0.792] [0.022] [0.010] Migrant*High share of internal 0.001 0.001 -0.001 -96.075 0.891 -0.032 0.058** migrants*Post 2017 [0.020] [0.014] [0.009] [276.245] [0.834] [0.032] [0.023] Individual controls YES YES YES YES YES YES YES Department*Year*Month FE YES YES YES YES YES YES YES Dep Var - Mean 0.625 0.078 0.297 7538.482 45.084 0.578 0.047 Observations 1909259 1909259 1909259 1018859 1192979 1192979 1192979 R-squared 0.224 0.024 0.259 0.285 0.111 0.277 0.052 Source: Authors’ calculations using GEIH, various years. 83.  With these numbers, Table 16 and Table 17 replicate the not to say that catering to Venezuelan migrants and to the regression analysis performed in previous sections of this actors affected by the internal conflict may not be imposing report, with the addition of an indicator for high share of competing claims on public resources, which can in turn call conflict-related internal migrants over total department for at times very delicate public spending decisions on the population.42 Contrary to widespread concerns, Venezuelan part of local authorities. Nor does this imply that there migrants in departments hosting a higher density of conflict- might not be other potentially serious concerns in terms related individuals do not seem to be facing additional issues of overall social cohesion and security. Further caveats in employment, either before or after the migration surge. arise from the fact that the internal migration indicator In fact, they are equally likely to be employed as elsewhere, does not differentiate between returnees and demobilized are not more likely to be unemployed, earn similar wages individuals, who most certainly rather face very different and work comparable hours, and are not more likely to be dynamics on the ground, and that it refers only to internal employed in informal occupations. The only outcome that migration occurred within the 12 months before the survey, seems to change significantly in the areas most hit by post- when it is very likely that many vulnerable populations conflict is the likelihood to work in temporary occupations, would have instead moved at a previous point in time. which decreases overall for migrants in areas with a larger Finally, as pointed out in previous sections of this report, stock of conflict-related individuals but rises substantially the estimates reported in Table 16 and Table 17 are simply in the same locations after 2017. informative about the (lack of) correlations between the variables analyzed, and do not by any means provide a full 84.  Interestingly, the situation is identical when agri-food assessment of statistically exogenous causality patterns. labor markets are considered (Table 16). The integration of Venezuelan migrants into a department’s agri-food value 85.  Even though coming from a simplified descriptive chains does not seem to be affected at all by a higher approach, however, the observation that the labor market concentration of conflict-related populations. Clearly, this is integration of Venezuelan migrants does not appear to be conflicting with the presence of conflict-related populations 42  This is constructed as a binary indicator equal to 1 if in a certain year a is very relevant. Rather than triggering a competition department lies above the 75th percentile of the distribution of share of conflict- among vulnerable groups, Venezuelan migrants seem to be related internal migrants over total department population across departments facing the same labor market conditions throughout the in Colombia for that year. Results stay mostly unchanged if only the fixed, country, regardless of the existence of potential pockets beginning-of-period internal migrant share is used rather than a time-varying of fragility. measure. 50 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Table 17 Working conditions in agri-food for migrants in recent migration surge, accounting for conflict-related internal migration Real Wage Employed Employed Employed Weekly Employed in (in 100 COP, Informal Temporary   in Primary in Food in Food Hours of Agri-food at fixed 2010 Worker Worker   Agriculture Processing Services Work prices) (1) (2) (3) (4) (5) (6) (7) (8) Venezuelan Migrant 0.050*** -0.019*** 0 0.069*** -649.863*** 2.148*** 0.074*** 0.034*** [0.005] [0.004] [0.002] [0.004] [179.384] [0.461] [0.018] [0.012] Migrant*Post 2017 0.035*** -0.007 0.003 0.038*** -735.396*** 2.312*** 0.036* -0.015* [0.008] [0.005] [0.003] [0.008] [153.236] [0.544] [0.021] [0.008] Migrant*High share of 0.006 -0.005 0.003 0.008 -192.133 0.846 0.025 -0.024 internal migrants [0.018] [0.009] [0.005] [0.017] [483.697] [1.666] [0.025] [0.015] Migrant*High share of 0.002 -0.001 0.008 -0.004 243.954 0.357 -0.045 0.046* internal migrants*Post 2017 [0.026] [0.009] [0.007] [0.024] [511.994] [1.905] [0.041] [0.023] Individual controls YES YES YES YES YES YES YES YES Department*- YES YES YES YES YES YES YES YES Year*Month FE Dep Var – Mean 0.168 0.074 0.029 0.065 4864.072 42.395 0.752 0.044 Observations 1192979 1192979 1192979 1192979 168089 199897 199897 199897 R-squared 0.147 0.303 0.007 0.039 0.167 0.157 0.197 0.066 Source: Authors’ calculations using GEIH, various years. 51 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR INTEGRATING VENEZUELAN V MIGRANTS INTO COLOMBIA’S AGRI-FOOD SYSTEM: TWO CASE STUDIES 52 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR V INTEGRATING VENEZUELAN MIGRANTS INTO COLOMBIA’S AGRI-FOOD SYSTEM: TWO CASE STUDIES 86.  The conclusion that can be derived from Chapters II, exported, representing 75 percent of national exported III, and IV is that the position of Venezuelan migrants on air freight (between 300 and 400 million dollars a year in Colombian labor markets is not free from contradictions: air freight). The trade balance in the sector saw growth despite being on average more likely to be employed of 219,219 tons in 2016, grew by almost 4.5 percent in than local Colombians, migrants tend to work in worse 2017 and 4.2 percent in 2018, with similar growth forecast working conditions: typically hired more informally or on for 201943. The sector generates considerable income in temporary occupations, they work longer hours, and they foreign currencies, with around 1.4 billion dollars in 2018, receive a lower wage. The situation is no different in the and contributes almost 17 percent of the income tax for agri-food sector, despite the growth potential of the sector, Colombian agriculture. and even though migrants seem to be more likely than local Colombians to be employed in agri-food activities 90.  Floriculture in Colombia is a large and growing sector compared to other occupations. with much potential. As a result, producer participation in the sector is growing exponentially: land area under flower 87. What can be done to provide good job opportunities production increased by 20 percent between 2015 and 2019, for migrants in the agri-food sector in Colombia? What are and the volume of production increased by approximately key factors in successful integration programs? In line with 10 percent with yields averaging 30 tons per hectare. the third objective of this report, to derive lessons for the agri-food inclusion of Venezuelan migrants in Colombia, 91.  The department of Cundinamarca is the largest this chapter tries to answer these questions. To this end, producer of flowers in the country (see Figure 21). Roses and the chapter presents cases studies of programs in Colombia carnations account for the largest production values and related to labor market inclusion of Venezuelan migrants in are planted mostly in the municipalities of Tabio, Madrid, two important agriculture and food systems value chains, Facatativá, and Zipaquirá amongst others. with an eye to deriving lessons to foster the integration of Venezuelan migrants through job generation in the broader 92.  Floriculture is the leading agriculture activity in agri-food sector. terms of generation of employment per hectare (14 jobs per hectare). The sector’s demand for workers is highly seasonal. The seasons for production of cut flowers has A. THE FLORICULTURE peaks of employment in line with significant holidays in which demand for flowers increases. It is a sector with SECTOR: CASE STUDIES FROM high formality in employability. It generates approximately 90,000 formal direct rural jobs plus 18,000 additional CUNDINAMARCA jobs at peak times.44 Approximately 65 percent of jobs in floriculture are carried out by women who are the head of 88.  This section provides are brief description of the the household, and account for 25 percent of female rural floriculture – cut flower – sector in Colombia, and employment in the country. subsequently describes two programs focused on labor inclusion of vulnerable populations: (i) the Asocolflores 93.  For years now, it has been difficult to hold onto program, focused on the vulnerable Colombian population, personnel in the floriculture sector, which is stigmatized as and (ii) the Sunshine Bouquet program, focused on the being unattractive because of the difficult and precarious Venezuelan migrant population: both programs provides employment conditions and low wages. Growth in useful insights on the instruments used for labor-market industrial activities, like construction, in the traditionally inclusion and on key lessons learned in their application. flower-growing municipalities in the Centre-North and Western areas of Cundinamarca, led to major abandonment 89.  The floriculture sector is one of the biggest contributors of employment in the sector. There is also a high level of to Colombian agriculture. The sector targets mostly labor turnover as the seasonal nature leads many workers international markets and is currently the country’s second to seek out alternative employment. largest agricultural export. Colombia is the world’s second biggest exporter of flowers, the leading exporter of 43 In monetary terms, growth was around 5 percent but is not significant carnations, and the leading supplier for the United States. because of the continuous fluctuations in the exchange market in the last According to figures from MADR’s Directorate of Agricultural period. and Forest Chains in 2018, 95 percent of production was 44  Data kindly shared by MADR, November 2019. 53 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Figure 21 West re Center re Location map for Dep rt ment Ant oqu Dep rt ment Cund n m rc production of Cut Product on surf ce 5 568 h s Flowers Product on surf ce 2 715 h s Source: Ministry of 33% of n t on l product on 66% of n t on l product on Agriculture and Rural Spec es Rose, cl vel, lstroe- Development Spec es Chr s nthemums, h dr n e s, others mer , others Western center re Dep rt ments R s r ld , C ld s, Qu nd o nd C uc V lle Product on surf ce 95 h s 1% of n t on l product on Spec es Trop c l flowers nd fol e 94.  The shortage of labor and high rotation has led to high 96.  Forging alliances with government entities was key for costs for companies in personnel selection and training the program being able to meet its objective of integrating processes, to the point where this has become one of vulnerable populations in the flower sector. These included the bottlenecks for the competitiveness of the sector. As first, an alliance with SENA in 2012, to provide technical cut-flower production requires a high concentration of training for workers in order to increase “capacidades labor (albeit seasonal), and in many municipalities labor instaladas” (productive potential) of workers, and to supply is in shortage, the floriculture sector provides a find mechanisms for labor matching between workers major opportunity for integration of Venezuelan migrants and companies for the sector. To find laborers for the into jobs in primary agriculture and along the cut-flower sector SENA worked in several ways with Asocolflores value chain. including: (i) mobile labor offices (buses) were sent to municipalities to advertise and recruit workers; (ii) direct calls for employment through the town halls and companies affiliated with Asocolflores; (iii) employment fairs for I. ASOCOLFLORES: PROGRAM FOR LABOR companies that needed to fill vacancies. Second, an alliance MARKET INCLUSION OF VULNERABLE was made with the National Agency to Overcome Extreme Poverty (ANSPE) in 2013, which allowed an increase in the POPULATIONS variety of vacant employment positions in the sector to 95.  Asocolflores is the largest labor union in the sector, encompass a population in conditions of vulnerability. This with the role of generating more and stable employment initiative impacted 1,200 families through access to formal and increasing the value of flower production and export employment. However, these alliances did not eliminate in Colombia. It started the Vulnerable Population Labor the issue of high labor turnover and did not generate the Inclusion Program (2011-2016) in 2011 to respond to a labor expected sustainability. shortage in the flower sector, for which the impetus was the crisis experienced by the sector in the early 2000’s. The 97. A third alliance was formed with the Colombian Agency revaluation of the Colombian peso between 2004 and 2008 for Reintegration in 2014, which directly linked companies in led to the loss of more than 22,000 jobs and the closure of the floriculture sector and the population of ex-combatants. dozens of floriculture companies. The principal objective of Although the strategy was designed to create employment the program was to generate decent, formal employment, opportunities for this vulnerable group, it did not generate through alliances with the public and private sectors, to much success in recruiting and maintaining ex combatants. strengthen the workforce/labor force in the countryside and Retrospective evidence from Asocoflores suggests that provide employment links for the vulnerable population. this group was not interested in working in rural areas and 54 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Table 18 Sunshine Bouquet’s hiring process for migrants through employment postings in Cucuta Venezuelan migrant participants in Migrants Migrants Migrants returning the the selection processes selected hired following year to work 2017 – Valentine’s Day (February) 1159 737 580 2017 – Mother’s Day (May) 647 406 380 180 2018 – – Valentine’s Day (February) 622 451 352 141 2018 – Mother’s Day (May) 604 540 361 128 2019 – Valentine’s Day (February) 635 500 352 100 TOTALS 3,667 2,634 2,025 549 Source: Sunshine Bouquet preferred to migrate to cities in search of employment high rotation remained. Both the positive results and opportunities or to receive aid to create their own company. the obstacles encountered in the implementation of the program have brought to light the following lessons and 98.  With the lessons learned from all these experiences, the recommendations: (i) focus on a potentially interested labor union designed a new strategy to attract personnel population that has a lower degree of vulnerability; (ii) to offset the high rotation and the lack of workforce in the improve the awareness of affiliated companies to improve sector. Employment fairs were created in 2015, designed non-income benefits offered to workers, and to invest to bring companies closer to the populations, both in more in specialized skills; (iii) strengthen the alliance with traditionally flower-growing municipalities and in the SENA and with public employment agencies; (iv) organize neighboring areas. In addition, alliances were drawn up with specialized fairs focused on the floriculture sector with public employment agencies45 to offer job positions via linked/related companies (suppliers and marketers). their platforms. 99.  In parallel with the labor union strategy of inter- institutional strengthening of Asocolflores, the affiliated II. SUNSHINE BOUQUET: PROGRAM FOR companies were given incentives to carry out a program to LABOR MARKET INCLUSION OF VENEZUELAN promote and implement good employment practices. The MIGRANTS objectives of the programs are to provide employees with better conditions, extra-salary benefits and opportunities 101.  Sunshine Bouquet is a leading company in the floriculture for promotion in the companies. According to a field sector employing 1100 individuals in the municipality of study conducted by the Universidad de Externado in the Tabio (Cundinamarca).47 The company typically requires companies affiliated with Asocolflores, the initiatives or 9 individuals per hectare for rose production and 10 good employment practices that were most welcomed were individuals per hectare for the production of other flower on access to health. Around 76 percent of employees agreed species. There is pronounced seasonality and labor demand that access to health insurance was an important benefit for peaks during special periods, like the first quarter of the choosing to work in a floriculture company. Similarly, other year and a recurring need for a workforce. services were also considered significant in the decision to contract with a floriculture company: transport services 102.  In 2017 Sunshine Bouquet started a program to employ (64 percent), study incentives (58 percent), employee fund Venezuelan migrants. The program’s objective was to break (52 percent), maternity aid (25 percent), life insurance (22 the stigma around working in the sector, offering improved percent), paternity aid (19 percent), and daycare and child quality of employment, with better salaries and benefits. services (11 percent).46 The hiring program offers dignified and safe conditions for migrants from Venezuela, who arrive under vulnerable 100.  Despite all efforts made to integrate vulnerable conditions. The hiring process is carried out through populations into the floriculture sector through the employment announcements or calls for employment in Asocoflores program, the shortage of workforce and which various allies participated. As shown in Table 18, from 45  Work carried out in conjunction with the “compensation funds” [cajas de 47  In Tabio alone the company has 65 hectares of land area under compensación], organizations in Colombia with a pronounced social character, greenhouses in strict conditions of quality, with controls on temperature and with the Program as an important ally, with the publication of offers of humidity. Sunshine has a total of 14,000 employees and 820 hectares under employment and skills for employees. greenhouses throughout the Cundinamarca department, distributed in the 46  These results are based on evidence from visits to a number of companies municipalities of Tabio, El Rosal, Madrid, Tocancipá, Chía, Nemocón, Tenjo, affiliated to Asocolflores. (Systematization recorded in the publication “Sector Facatativá, Chipaque, Soacha, Cucunubá and Bogotá. In Tabio, the company privado: pensando en grande. Contribuciones al ODS 8. Universidad de Externado. reaches a volume of production of around one million stems per day at peak Colombia” [Private sector: thinking big.] periods. 55 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR 2017 to date, 3,667 migrants have participated in selection 108.  Colombia is the third largest producer of coffee after processes and 2,025 have been hired. Brazil and Vietnam. In 2018, the coffee-producing area was 877,144 hectares, with national production reaching 103.  The majority of migrants were hired on a temporary figures close to 14 million bags, of which 12.7 million bags basis, and more than 26% have returned to work with were exported (Directorate of Agricultural and Forestry the company the following year. Sunshine Bouquet offers Value Chains of MADR). The coffee area extends across all the official minimum wage of 925,148 Colombian pesos departments of the country. In the last ten years, Colombia (274 USD) and at peak periods the workers can earn, with has increased coffee yield by 56 percent with a productivity extra hours, up to 1,450,000 Colombian pesos per month of 10 to 19 bags per hectare, in large part due to the Coffee (430 USD). The transfer from Cucuta to Bogota, food, and Renewal Program. The program has improved the average accommodation are provided as additional benefits to the age of the crop, as well as its density and resistance to rust. basic salary when hired under the employment relation program. The employment announcements last on average 109.  Smallholders are the cornerstone of Colombia’s coffee one week, during which the Sunshine Bouquet team (two production: 96.5 percent of coffee growers registered hiring assistants, two psychologists, two doctors and the in the National Federation of Coffee Growers (FNC) head of hiring) interview approximately 120 people a day. are small producers, with farm sizes between 3 and 5 hectares. However, these producers are battling against an 104.  Similar to the Asocolflores Program, Sunshine Bouquet increasingly larger global coffee footprint, and as markets has created a series of interinstitutional alliances which have flood prices decline. Coffee producers in Colombia typically enabled both the selection process and the subsequent skills make less than 10 percent in profits, on the price of coffee. and specific training process for the Venezuelan workers to This makes coffee production as a livelihood less and be improved. less attractive. More than 50 percent of production costs go towards labor, with 80-90 percent of the labor force 105.  The aforementioned aspects of Sunshine Bouquet’s dedicated to harvesting. program of Inclusion in the labor Market provide a good example for integration of migrants in an agriculture value chain. Beyond work in primary agriculture, the floriculture sector typically also requires workers to be integrated in I. HARVEST PLAN 2019 (PLAN COSECHA the quality control, packaging and other off-farm labor 2019): LABOR INCLUSION PROGRAM IN in processing the flowers. As such, this program sends a RISARALDA positive signal on the potential for creating jobs among a marginal population when labor shortages exist and there 110.  The department of Risaralda and the coffee axis has is less availability of the local population. Sunshine Bouquet traditionally been a zone of emigration, and shortages of intends to continue providing employment opportunities coffee pickers remains a problem for the sector in Colombia. for vulnerable Venezuelan refugees and migrants. In the The Harvest Plan is a program in which some departments first quarter of 2020 it plans to hire 1000 people. of the coffee axis participate, to find a solution to this shortage of labor in coffee harvesting. It is a program for 106.  Results from in-depth semi-structured interviews coffee producers, facilitated and supported by the FNC of actors along the cut flower value chain carried out in through its Departmental Committees. November 2019, that provides more details on the profiles of Venezuelan migrant workers and the labor conditions 111.  The objective of the 2019 Harvest Plan in Risaralda they face are available in Annex E1. was to make coffee cultivation and harvesting attractive through: (i) achieving sufficient labor in coffee collection for coffee growers; (ii) ensuring that the price of the collection B. THE COFFEE SECTOR: CASE is reasonable and does not affect the profitability of the coffee grower, and (iii) dignifying the work of the pickers by STUDIES FROM RISARALDA AND promoting strategies that contribute to their quality of life. NORTE DE SANTANDER 112.  Under the 2019 Harvest Plan, a survey was conducted on coffee pickers in the department. Based on its results, an 107.  This section presents the “Harvest Plan 2019” of the awareness campaign was implemented on the importance Department of Risaralda - an initiative supported by the of good treatment of coffee pickers, and the greater needs Departmental Committee of Coffee Growers of Risaralda of facilities and benefits for pickers. Further to this, to to attract labor to the coffee sector and provide income attract labor, a dissemination campaign has been launched generation opportunities to vulnerable populations. through the main media, radio and television. 56 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR 113.  The survey of coffee pickers, participating in the 2019 percent of collectors prefer to work on small farms relative to Harvest Plan, showed that 85 percent were men and just 15 7 percent who preferred large, more commercially oriented percent women. While the composition of laborers is highly farms.Of relevance to the employability of Venezuelan gendered, and male-biased, evidence shows that harvest migrants, in the visits and in-depth semi-structured quality is typically higher among female pickers, who collect interviews, it appeared that large farms were much more the best selected grain, with a lower percentage of damage. willing to hire Venezuelan coffee pickers. Bigger farms were FNC is aware of the need to establish strategies that attract more willing to hire Venezuelans as there was a higher labor and maintain female coffee pickers. Coffee pickers are also need, and these farms showed less suspicion over potential quite old, per the survey: 72 percent of pickers are over 40 theft and security issues. years old and of these, 51 percent are over 50 years old. These data highlight the need to implement initiatives and 116.  Based on lessons learned from the results of the mechanisms that make the coffee business an appealing Harvest Plan survey and prior actions taken in the program, option for the young. For example, the National Coffee several recommendations emerge for hiring more workers, Growers Federation is currently implementing youth particularly Venezuelan migrants, and improving coffee entrepreneurship initiatives,48 centered on innovation picker conditions. These include: (i) strengthen institutional and technology. alliances: the Migration Office was key in managing the arrival of people from Venezuela; (ii) strengthen agreements 114.  Coffee pickers had a high level of satisfaction with with the transport sector: bus terminals became the worker conditions and benefits received through the strategic site for picking up collectors; (iii) foster alliances Harvest Plan; especially for accommodation, food and with the media to promote the need for labor on coffee hygiene, and recreation areas. Connectivity and better farms: around 46 percent of respondents learned that labor communication were shown as priority issues for coffee was needed for the coffee harvest in Risaralda by radio, pickers, as about 60 percent were married or had family television or press; (iv) improve working conditions offered commitments. Among the surveyed workers of the Harvest to collectors, including making conditions more attractive Plan, 74 percent of pickers in Risaralda were covered by to women whose collection is of higher quality (fruit at subsidized social protection, 8 percent used another system, the point of maturation and less damage); (v) classify tasks and 18 percent had no social protection at all. in the harvest according to the worker’s profile: include a filter and selection process for workers for differentiation 115.  Of the most important factors cited when choosing of tasks taking into account age and gender; (vi) strengthen a coffee farm at which to work, 60 percent of workers strategies to overcome security problems, specifically, considered the accurate measurement of their picked thefts and drug use on farms. coffee, and the quality and calibration of the measuring weight as most important. Other concerns like quality of accommodation and food, treatment by employers, high production of the farm, and proximity of the farm to their home were lower in relevance. Results also reveal that 47 48  RENACER program is launched by Procafecol, company responsible for the development of Tabernacles Juan Valdez and ANDI Foundation. RENACER, seeks to make visible and strengthen the leadership of young farmers in different regions affected by violence in Colombia. The program has 19,531 young people aged 14 to 28, in 22 departments, of which 31% are women and 69% are men 57 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Box 4 An unsuccessful example: the labor inclusion program for Colombian returnees in Norte de Santander The labor insertion Plan of Colombian Returnees in Norte de Santander (2013-2014) is presented as an interesting case study of a labor inclusion instrument that did not work as intended. The department of Norte de Santander is composed of 40 municipalities, of which 36 grow coffee. There are 15,600 registered coffee growers and 23,092 hectares of coffee throughout the department. Approximately 24 million kilos of dry parchment coffee are produced every year in the department, for a value of about US$ 4.6 million. The department grows coffee all year round: the seasonality in coffee harvesting is less marked than in the coffee axis departments, implying that the laborers could have a more stable position and source of income, with tasks distributed between collection and processing. Between 2013 and 2014, the National Federation of Departments funded a program for the inclusion of Colombians returning from Venezuela in coffee jobs in Norte de Santander. With the support and financing of the government, the Coffee Committee of Norte de Santander identified jobs in coffee farms with formal hiring for a duration of at least 4 months. Through the program, 140 Colombian returnees from Venezuela were hired on coffee farms. The experience, however, highlighted the need of ensuring better training of the newly hired laborers, as well as of additional filters for their selection. From anecdotal evidence, often Colombian returnees had different a skill set from those of traditional agricultural laborers and coffee pickers, leading to demotivation and self-exclusion, and in some cases self-injury to collect disability/accident allowance. 58 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR VI CONCLUSION 59 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR VI CONCLUSION 117.  As the political and socio-economic crisis in Venezuela 120.  For as many challenges migration is causing today, in escalates, the forced displacement of millions of fact, they can be turned into an opportunity for economic Venezuelans who leave their country to escape violence, growth and social development, if appropriate and timely insecurity and threats, and to fulfil their basic human rights measures are taken. Apart from its humanitarian value is posing unprecedented challenges to receiving countries per se, the successful integration of Venezuelan migrants across Latin America and beyond. can represent a valuable addition to the Colombian labor force, bringing new skills and filling labor gaps where they 118.  This study on the labor inclusion of Venezuelan exist. This appears to be true for Colombian agriculture migrants in Colombia’s agri-food labor markets thus comes and food systems, which face a growing demand at the at a time in which Colombia is facing a migration inflow that domestic and international levels, on one side, but stagnant is unparalleled both in the country and in the Latin America productivity and labor shortages, on the other. Certain region. As of August 2020, the number of people who had agricultural value chains, like floriculture and coffee, in crossed the border between Venezuela and Colombia was Colombia are particularly in need of a labor influx to fill estimated at around 1.8 million. specific occupations, and the growth of food processing and integrated food retailing and service activities cannot 119.  The Venezuelan crisis is posing significant internal but benefit from a skilled labor pool to boost efficiency challenges to the Colombian government, with potential and innovation. repercussions on economic, institutional, social, and environmental aspects. The challenges have, in turn, been 121.  This report shows, however, that this potential is evolving over the last few years, and even months. In the to an important extent yet to be fulfilled. At the policy- aftermath of the migratory intensification, the immediate making level, for example, it is notable that the CONPES priority for the Colombian government had been migrants’ 3950 strategy for migrant inclusion designed by the registration in the country and the provision of primary Colombian government does not provide any strategic care and humanitarian relief. With the passing of time guidance or budget for the Ministry of Agriculture and and the deepening of the crisis in Venezuela, however, it Rural Development to facilitate inclusion of migrants in the has become clearer and clearer that the phenomenon will agriculture sector. Looking at individual workers’ outcomes, have implications on a medium-long term horizon, whose this study does find that Venezuelan migrants tend to find successful management will require going beyond immediate more employment in agri-food activities than in other social assistance. The recent COVID-19 pandemic, in turn, sectors: however, it also shows that they are mostly doing is exacerbating pressure on local economies and social so at substantially worse working conditions (lower wages, protection mechanisms, making it even more challenging higher working hours) and with a higher likelihood of to cater to the migrant population. Colombian authorities ending up in the informal labor force when compared to the are starting to report that some migrants are even trying local population. to return to Venezuela, despite unchanged threat and misery there, to escape discrimination and arbitrary job 122.  Importantly, migrants also tend to cluster in urban areas, loss in Colombia: at the end of April, 2020, some 12,000 where they find employment in suboptimal food-service Venezuelans had crossed the border back into their home occupations, even though those that are employed in rural country, where they face confinement and inhuman areas seem on average to encounter better-quality jobs in conditions in improvised camps lacking basic hygiene and terms of wages, working hours, and informality. Crucially, safety standards.49 As this report was completed before the migrants are not more likely to work in primary agriculture pandemic stroke, it does not provide an assessment of the even in areas that are predominantly agricultural, and that effect of the health emergency on Colombian labor markets as such should be better able to absorb agricultural laborers. and on the conditions of Venezuelan workers in Colombia: Considering the high unemployment rate registered however, the COVID-19 challenge is making it even more among migrants, this may represent an important missed painfully evident that the migration crisis needs inclusive opportunity for matching them into agricultural inclusion solutions now and in future. programs. Similarly, it is striking that Venezuelan migrants are not more represented in better-remunerated and more dynamic value-addition activities downstream in the food chain (such as food processing or formal food 49  Cf. https://efectococuyo.com/coronavirus/cerca-de-12-000-venezolanos-han- services), which would offer a better match for their skill regresado-al-pais-senala-migracion-colombia/; https://www.bbc.com/mundo/ set. For example, migrants’ average years of education are noticias-america-latina-52225368; https://elestimulo.com/venezolanos-huyen- well aligned with the education of the average worker in del-coronavirus-para-caer-en-refugios/; https://www.theguardian.com/global- Colombia’s food processing: yet, Venezuelan migrants are development/2020/apr/16/venezuelans-returning-squalid-quarantine-migrants. significantly less likely to be employed in the sector with 60 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR sector, including for example the organization of job respect to the rest of the population with similar socio- fairs aimed at opportunities in the rural context and demographic characteristics. in agri-food chains, and the provision of specialized training in specific, scarce skills highly demanded in agri- 123.  These patterns can in part be explained by the fact food jobs. Better skills matching to provide higher skilled that many migrants need to provide for their livelihoods workers with jobs linked to processing just off-the farm, regardless of their regularized status or work permit: in this or youth with access to opportunities for innovation and sense, the higher informality of urban food services jobs vis- technology-linked positions could also provide impetus à-vis other sectors might provide an immediate response, for drawing migrants to agri-food and rural centers. or be the only option, for their financial needs. Very likely though, other appealing features of urban centers for » Make the agri-food sector and the rural migrants are their ease of access via major transportation environment attractive for migrants. As the strong routes, as well as the existence of tighter social networks social networks of larger urban areas may be difficult and migrant communities within cities: both factors may be to replicate in lower-tier cities and rural areas, the acting as strong barriers to mobility towards other parts of strengthening of service provision and social protection the country, especially if coupled with a lack of information in these areas could also help in sustaining livelihoods on labor opportunities in different areas. and easing a potentially crucial barrier to mobility. Benefits for consideration include health insurance, social 124.  Being that the case, there could be room for facilitating protection schemes, transportation, accommodation, migrants’ integration into better quality jobs and less study benefits, parental help, permanency, and security. crowded zones of the country, for example through information campaigns, more systematic job advertisement, » One option that could be considered is replicating, and mobility incentives. The fact that migrants have for the migrants’ case, the structure of social recently become more likely to be employed in agriculture programs currently targeting other vulnerable and food processing in areas that see a higher presence groups. Particularly in rural areas, vulnerable of agro-industrial firms shows that they are not averse to populations often have similar socio-demographic these types of occupations, when opportunities arise. The characteristics to the Venezuelans in the country, two case studies presented in Chapter VI are successful which suggests that programs designed for the examples of how a multi-faceted approach targeting the former group could also work well for the latter. specific needs and constraints of migrants and vulnerable This would not mean dividing the same pie among populations can go a long way in building a much more a higher number of beneficiaries, but rather using appealing employment package. the same channels to achieve the respective inclusion goals. LOOKING AHEAD: » Smooth labor demand frictions by channeling information to employers and considering OPERATIONALIZATION business incentives. As regards the labor demand side, a needed step is to efficiently channel information PRINCIPLES to potential employers, highlighting the opportunity represented by a larger labor pool to fill labor shortages 125.  Attracting migrants into different agri-food jobs in certain value chains, and providing clear information will require a coordinated, programmatic strategy and guidance on the process for hiring migrants. In jointly addressing the demand and supply sides of the parallel, incentives could be established for the business labor equation. While ensuring that employers see the sector for hiring Venezuelan migrants, for example in the opportunity represented by incoming migrant workers form of tax incentives, or appealing to corporate social and are duly accompanied in their hiring process. It will be responsibility programs with the promotion of social paramount to make sure that migrants perceive they do responsibility certifications. Involving National Producer not stand to lose from seeking employment in agri-food Federations and Producer Organizations would ensure activities and to facilitate and incentivize their transition. higher effectiveness and buy-in on the producers’ side. 126.  The analysis and questions raised by this study lead to » Streamline the institutional approach. The agri- some practical ideas and operationalization principles for a food inclusion framework could be strengthened with migrant integration strategy that involves agriculture and a streamlined institutional approach that increased food systems more prominently. cooperation among different government entities. In particular, it would be key to ensure policy alignment » Strengthen “labor pull” actions for agri-food jobs. with the CONPES 3950 strategy, broadening its scope to Active labor market policies such as job search and job include specific provisions for the integration of migrants placement support could be targeted at the agri-food in rural areas and in agriculture and food systems, with 61 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR a more systematic involvement of the Ministry of Agriculture and Rural Development in the promotion of labor inclusion in the agri-food sector. A more ambitious initiative could be to envisage the creation of targeted employer-based PEP schemes specifically directed at jobs in the agri-food sector. 127.  As a closing remark, it is worth emphasizing how the specificity of local productive landscapes and labor markets, the geography of migrants’ distribution across the country, and local socio-cultural contexts demand a differentiated approach in the application, design and adoption of public measures for migrants’ integration. Operationalizing the proposed framework will require spatial development approaches, and customized solutions that make those approaches relevant to particular local circumstances and priorities. This effort will require the active participation and cooperation of local institutions, to ensure that migrants and local communities alike can gain from an inclusion strategy that is equitable, effective, and beneficial for all. 62 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR VII BIBLIOGRAPHY ANDI. (2019). Innovación en la Industria de Alimentos. Bogotá: Cámara de la Industria de Alimentos. ARN. (2019). ARN en cifras, Julio 2019. Agencia Colombiana para la Reintegración y la Normalización. Bahar, D., Dooley, M., and Huang, C. 2018. Integrating Venezuelans into the Colombian labor market: Mitigating costs and maximizing benefits. Policy brief. Global Economy and Development at Brookings. Brookings Institution. Bahar, D. and Dooley, M. 2019. Venezuela refugee crisis to become the largest and most underfunded in modern history. Brookings Institution. Bermúdez, Y., Mazuera-Arias, R., Albornoz-Arias, N., Morffe Peraza, M.A. (2018). Informe sobre la movilidad humana venezolana. Realidades y perspectivas de quienes emigran [9 de abril al 6 de mayo de 2018]. San Cristóbal: Venezuela: Servicio Jesuita a Refugiados (SJR). Venezuela. Caruso, G., Gómez Canon, C., and Mueller, V. (2019). Spillover effects of the Venezuelan crisis: migration impacts in Colombia. Oxford Economic Papers, 2019: 1-25. Ceritoglu, E., Yunculer, H. B. G., Torun, H., & Tumen, S. (2017). The impact of Syrian refugees on natives’ labor market outcomes in Turkey: Evidence from a quasi-experimental design. IZA Journal of Labor Policy, 6(1), 5. Coordination Platform for Refugees and Migrants from Venezuela. (2019a). R4V Latin America and the Caribbean, Venezuelan refugees and migrants in the region - December 2019. United Nations High Commission on Refugees (UNHCR), International Organization for Migration (IOM). Coordination Platform for Refugees and Migrants from Venezuela. (2019b). Regional Refugee and Migrant Response Plan for Refugees and Migrants from Venezuela, January - December 2019. United Nations High Commission on Refugees (UNHCR), International Organization for Migration (IOM). Del Carpio, X. V., & Wagner, M. (2015). The impact of Syrians refugees on the Turkish labor market. The World Bank. Departamento Nacional de Planeación. (2018). CONPES 3950: Estrategia para la Atención de la Migración desde Venezuela. Euromonitor International. (2019). Retailing in Colombia. Country Report. FEDEGAN. (2018). Cifras de referencia del sector ganadero colombiano. Federación Colombiana de Ganaderos. FEDEGAN. (2019). Cifras de coyuntura del sector ganadero colombiano. Federación Colombiana de Ganaderos. FNC. (2017). La recolección de café en Colombia: Una caracterización del mercado laboral. Ensayos sobre Economía Cafetera, 32: 35-65. García, C., Zárate, C., and Ochoa, G. (2016). Demanda y Oferta de mano de obra en la caficultura colombiana. CRECE. Manizales: CRECE. 63 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR GRID. (2019). Global Report on Internal Displacement 2019. Internal Displacement Observatory (IDMC) and Norwegian Refugee Council (NRC). Ibarra, S. (2018). Reparaciones colectivas, caso colombiano. Unpublished. Kerr, S. P. and Kerr, W. R. (2011). Economic Impacts of Immigration: A Survey Economic Impacts of Immigration: A Survey, Harvard Business School Working Paper. Maystadt, J. F., & Verwimp, P. (2014). Winners and losers among a refugee-hosting population. Economic Development and Cultural Change, 62(4), 769-809. OECD (2019) Nota de la OCDE sobre el shock migratorio desde Venezuela hacia Colombia y sus implicaciones fiscales, https://www.oecd.org/economy/surveys/Colombia-migration-shock- note-spanish-2019.pdf OECD. (2019). OECD Economic Survey of Colombia 2019. Forthcoming. Paris: Organization for Economic Co-operation and Development. Orrenius, P. M. and Zavodny, M. (2012) Economic Effects of Migration: Receiving States. Oxford University Press. doi: 10.1093/oxfordhb/9780195337228.013.0005. Reina, M, C. Mesa and T. Rámirez (2018). Elementos para una política pública frente a la crisis de Venezuela. Cuadernos Fedesarrollo 69. Universidad del Rosario and Konrad Adenauer Foundation. 2018. Retos y oportunidades de la movilización humana venezolana en la construcción de una política migratoria colombiana. Bogotá, Colombia. USDA. (2019). Colombia, Food Processing Ingredients: Colombian Market Continues Offering Opportunities to U.S. Exporters. USDA Foreign Agricultural Service, Global Agriculture Information Network. United States Department of Agriculture. World Bank. (2015). Revisión técnica del proceso de reintegración de excombatientes en Colombia. Unpublished. World Bank (2018). Migración desde Venezuela a Colombia. Impactos y estrategia de respuesta en el corto y mediano plazo. World Bank. (2020). Future Foodscapes: Re-imagining Agriculture in Latin America and the Caribbean. Washington, DC: The World Bank. WRI. (2018). Creating a sustainable Food future: A Menu of Solutions to Feed Nearly 10 Billion People by 2050. World Resources Report. Washington, DC: World Resources Institute. 64 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR VIII ANNEX A. DATA SOURCES and into Colombia during the last 5 years or 12 months, or else that were born in other countries. Importantly, the I. HOW MANY VENEZUELAN MIGRANTS ARE survey covers irregular and regular migrants alike, although IN COLOMBIA? it does not allow to tell the two categories apart. As the GEIH data are collected monthly, they allow for a very accurate Estimates on the number of Venezuelan migrants in view of the dynamics of migrants’ influx in Colombia and Colombia can be constructed based on three main sources their labor market participation. of information, each one with its advantages and limitations (see Table A1). A first data source is the official information collected at all official entry points at Colombia’s frontier by Colombian migration authorities. This source gathers basic II. AGRI-FOOD LABOR MARKETS information on education, occupation, main motivation for the trip and final destination of all travels who legally enter At present, updated information on the rural labor market in Colombia, but only covers regular migrants. Colombia is somewhat scarce. Very detailed information on labor supply comes from the National Agricultural Census A good way to account for irregular migration is to use the (CNA, Censo Nacional Agropecuario), which is conducted Administrative Registry of Venezuelan Migration (RAMV, by the national statistics department (DANE) for the Registro Administrativo de Migrantes Venezolanos), which agricultural sector. The CNA provides statistical information was compiled by the Colombian government between April on the agricultural sector, including socio-demographic 2018 and April 2019 to coordinate the access of irregular characteristics of producers, their production decisions, migrants to public services and their social integration inputs, and other factors. The CNA is representative at the through different regulation programs such as the Special national level and can be disaggregated at the municipal Permanency Permit (PEP). Despite being an invaluable and sub-municipal level (Vereda). The unit of analysis repository of socioeconomic information on irregular in the Census are the Units of Production, which may be migrants in Colombia, the data in the RAMV have only so far agricultural (UAPs) or non-agricultural (UNAPs). The year been collected for one year, and therefore is not informative 2014 CAN used in this report provides information on about the dynamic ebbs and flows of Venezuelan migrants around 2.4 million UPAs and 540,000 UNAPs. into Colombia. A second data source to gather information on agricultural Finally, the Monthly Labor Survey Gran Encuesta Intengrada labor in Colombia is the National Agricultural Survey (ENA, de Hogares (GEIH), the nationally representative survey of Encuesta Nacional Agropecuaria). The survey, conducted Colombian labor markets, is probably the richest source of in 2017, is another cross-section and therefore lacks in detailed information on the participation of Venezuelans dynamism (even though a new round of data has been in Colombia’s labor markets. In general, the survey, which collected in 2019 and in pending release), but still provides is a repeated cross-section, collects each month a rich a very rich picture of the actual labor supply in primary set of information on labor participation, income, access agriculture Colombia. It is, however, silent on other activities to education and social security, as well as receipt of any downstream in the agri-food value chain. government support. Crucially for this report, moreover, the GEIH questionnaire contains a migration module that allows The main source of information on agri-food labor market tracking of all persons that report having migrated within outcomes thus remains the GEIH Labor Survey introduced in Data Source Type of information Type of migration Time Coverage Table A1 Colombian Migration Census of all visitors to Daily Main data sources Regular Authorities Registry Colombia (January 2014 – June 2019) on migration in Colombia used in Regularization Programs Venezuelans migrants April 2018- April 2019 (RAMV-PEP). participating in these Irregular (cross section) this report and their programs characteristics Monthly Labor Market Monthly Nationally representative Survey (GEIH) Regular and Irregular (used in this report: 2014-I survey – 2019-II) 65 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR sub-section A.ii above. The GEIH is statistically representative fisheries (primary); (ii) Agri-food manufacturing industry; for the main metropolitan areas, departments, as well as and (iii) Food services and retail (see Table A2). Figure A1 all rural centres scattered in more than 443 municipalities provides an overview of the evolution of employment in of Colombia. In total, each year 232,000 households are these three sectors in recent years. interviewed. GEIH includes a specific question on the sector in which each employed person is working, which allows for an accurate investigation of sector-level employment Table A2 Agri-food sector based on ISIC code numbers. In order to identify employment at different Sector ISIC code stages of the agri-food value chain, the activities listed Primary food production 111 to 220 by GEIH respondents employed in agri-food value chains Food Processing 1511 to 1594 have been recoded based on the International Standard Food Retailers 5521 to 5529 Industrial Classification of All Economic Activities (ISIC) into three main productive sectors: (i) Agriculture, forestry and Source: based on ISIC code 3650 Figure A1 3600 Employed in the 3550 agri-food sector – 3500 3450 Colombia, 2014-2019 3400 Source: calculations based 3350 on GEIH – DANE. 3300 Note: time series are Pr m r Product on 3250 seasonally adjusted using No emplo ed 3200 the X-13 seats (ARIMA) ( n 1,000 people) 3150 correction. 1000 900 800 700 600 500 400 300 Food Process n 200 No emplo ed 100 ( n 1,000 people) 0 1600 1400 1200 1000 800 600 Food 400 Ret l n No emplo ed 200 ( n 1,000 people) 0 2014 2015 2016 2017 2018 66 + 2 2 × 2017 + 2017 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR + = + 1 1 ′ + + + 2 2 + 1 1 × 2017 + 2 2 × 2017 + 2017 B. REGRESSION ANALYSIS – In the analysis + accounting ′ + for + post-conflict internal migration, regressions = + + × an feature additional2017 + binary indicator 2017 TECHNICAL DETAILS int for high share of conflict-related internal migrants over total department+population, × defined as whether in a The regression analysis presented in Chapters II, IV, and V certain year a department lies above the 75th percentile of = + + + × × × 2 017 2017 + + ′2 017 relies on an OLS (Ordinary Least Squares) empirical strategy the distribution of share of conflict-related internal migrants on repeated cross-sections. over total department + population + across departments in + × Colombia for that year (results stay mostly unchanged if In the baseline specification, given a labor-market outcome only the fixed, beginning-of-period + × × internal 2017 migrant + ′ share y, the regression equation for individual i living at time t in is used rather than a time-varying measure): department d is + + = + + × 2017 + 2017 = + + × 2017 + 2017 + + × + ′ + + = + ++ × × 2 × + 2017 017 + 2 017 ′ where m is a binary indicator equal to 1 if individual i is + + + + × a Venezuelan migrant, post2017 is a binary indicator switching 1 to in years after 2017, 1 is a row vector of coefficients Here, μ and φ are additional of interest. + × × 2017 + ′ = + 1 1 + 2 2 + 1 1 × 2017 individual-level = + + × covariates, a θtdis 2 017 + 2017 department-year fixed + + effect and ε is an idiosyncratic error term. + × 2 017 + 2017 Equality of any two coefficients 2 = .0 0 : 1 − (e.g. β1 vs. β2, or γ1 vs. γ2) 2 2 + ′ + + 1 : 1 − 2 ≠ 0 F-tests on linear is tested through standard post-estimation + ′ + + When analysing longer-term and recent Venezuelan restrictions, where: migrants separately, the regression equation is 0 : 1 − 2 = 0 = + 1 1 + 2 2 + 1 1 × 2017 1 : 1 − 2 ≠ 0 = + + × 2017 + 2017 + 2 2 × 2017 + 2017 Individual covariates include gender, age (linear, squared, + × and dummy for being aged 25 or less), living in a rural area, + ′ + + + × × 2017 + ′ and highest educational attainment (primary at most, where m1 and m2 are binary indicators for being respectively secondary school, technical and vocational education, recent a = and + a longer-term ++ + × Venezuelan 2017 + 2017 migrant. bachelor degree, post-graduate degree). Standard errors are clustered at the department level. coefficients = + of + ′ + + The + interest × are 2017 coefficient +on β 2017 Venezuelan migrant (or β1 and β2 in the recent vs. longer-term When the outcome of interest is whether the individual + × specification) and γ (γ1 and γ ) on the interaction between is employed, the analysis is performed on the years 2016, = + + ×22017 + 2017 migrant and post-2017, + × toexplore whether migrants × 2017 + ′ 2017, 2018, and 2019, and on individuals aged between 15 = + different experienced 1 1 + + outcomes 2 + 2 + before 1 1 × and after × the 2017 recent and 70, included. When considering wage, hours worked, migratory surge of + 2017. + informality, and temporary work, the sample is further 2 + 2 × × 2 2017 + 017 + 017 2 ′ restricted to those who are employed. The heterogeneity analysis + ++ across different subgroups ′ + of the population is performed by adding interactions When instead the outcome is being employed in the agri- between migrant status and a dummy variable d for food sector (overall and disaggregated along the food value = + + × 2017 + 2017 relevant socioeconomic characteristics: chain), the analysis is performed on the years 2016, 2017, + + × 2018, and 2019, and on individuals aged between 15 and 70, = + + 0 : 1 ×− 2 = 2017 0 + 2017 included, who report being employed (qualitative results + : × − ×0 ≠ 2017 + ′ do not change when running the regressions on the sample 1 1 2 + × unconditional on employment). When considering wage, + + + × × 2017 + ′ hours worked, informality, and temporary work in the agri- food sector, the sample is further restricted to those who + + are employed in the overall agri-food sector. Coefficients π and τ now 0 : 1 − 2 = whether measure 0 and by how much Venezuelan migrants : 1 1from2group − ≠ 0 d (e.g. women, or youngsters) are facing different challenges or opportunities = + + × 2017 + 2017 vis-à-vis the rest of the population, respectively before and after the migratory +surge of 2017. + × + × × 2017 + ′ 67 + + INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR C. ADDITIONAL RESULTS Table C1 Venezuelan migrants vs. Locals: Characteristics of the unemployed and inactive population 2014 - 2016 2017 2018 2019 (Q1 & Q2) Locals Venezuelan Migrants (born in 12 12 12 12 Colombia) 5 Years 5 Years 5 Years 5 Years Months Months Months Months 1.1 Gender Female 51.40% 45.36% 47.18% 46.73% 49.64% 48.49% 49.32% 48.22% 50.86% 1.2 Age (years) 15 - 25 17.58% 19.62% 23.38% 20.94% 23.56% 24.20% 27.11% 24.89% 26.23% 26 - 35 16.82% 24.89% 21.93% 22.34% 21.69% 24.86% 22.21% 27.01% 20.43% 36 - 45 14.74% 15.26% 12.44% 17.51% 11.17% 14.25% 12.11% 13.95% 11.03% 46 - 55 13.75% 9.65% 7.01% 8.19% 6.15% 7.96% 5.44% 7.35% 5.96% 56 - 70 12.96% 6.02% 4.61% 5.36% 3.84% 4.60% 3.94% 4.68% 3.67% 1.3 Education Avg. years 7.57 6.96 6.65 7.19 7.32 7.66 7.95 7.93 7.76 of schooling 1.4 Rural Rural Worker 19.95% 19.98% 17.81% 18.79% 13.26% 0.86% 0.33% 15.86% 9.07% 1.5 Job search characteristics - Active Looked for work in the 2.08% 5.11% 11.07% 4.24% 9.34% 3.95% 9.11% 5.57% 10.24% last week In the last 4 weeks 4.80% 8.83% 17.15% 8.39% 18.17% 8.74% 15.88% 11.12% 17.94% looked for a job or to set up a business? What did you do to search for job Requested help from 2.11% 4.69% 9.68% 3.37% 9.92% 3.60% 8.32% 5.07% 9.59% friends, family, etc Sent CV to companies 2.01% 3.35% 5.82% 3.92% 7.12% 4.15% 6.77% 5.22% 6.79% Attended recruiting calls 0.03% 0.08% 0.00% 0.05% 0.01% 0.00% 0.09% 0.00% 0.00% Want to get a paid job 6.69% 9.30% 11.67% 10.65% 10.89% 8.82% 7.25% 7.32% 9.30% or set-up a business? 1.5 Job search characteristics - Inactive main reason for not getting a 2.02% 4.89% 5.33% 6.45% 6.39% 6.95% 6.43% 7.24% 5.64% job is no job available in the city or region/cannot find work in my profession Data source: GEIH, various years 68 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR   Dependent Variable: Worker is employed Table C2   (1) (2) (3) (4) Labor outcomes Recent Venezuelan Migrant 0.009 -0.072*** -0.019* -0.012 for migrants with [0.015] [0.011] [0.011] [0.011] interactions, Longer-term Venezuelan Migrant 0.084*** 0.031*** 0.056*** 0.063*** disaggregated by [0.008] [0.008] [0.005] [0.004] recent and longer- ST Migrant * Post 2017 0.069*** 0.047*** 0.060*** 0.056*** term status [0.012] [0.009] [0.009] [0.008] Employment LT Migrant* Post 2017 0.033*** 0.014** 0.021*** 0.015*** [0.006] [0.006] [0.006] [0.004] ST Migrant* Female -0.052*** [0.016] ST Migrant* Female* Post 2017 -0.013 [0.019] LT Migrant* Female -0.052*** [0.011] LT Migrant* Female* Post 2017 -0.024*** [0.008] ST Migrant* Youth 0.163*** [0.013] ST Migrant*Youth*Post 2017 0.022* [0.012] LT Migrant*Youth 0.100*** [0.021] LT Migrant*Youth*Post 2017 0.004 [0.018] ST Migrant * No Diploma 0.068* [0.038] ST Migrant * No Diploma*Post 2017 0.051 [0.039] LT Migrant * No Diploma 0.053 [0.037] LT Migrant * No Diploma*Post 2017 -0.015 [0.042] ST Migrant * Rural -0.072** [0.027] ST Migrant * Rural*Post 2017 0.017 [0.063] LT Migrant * Rural -0.054 [0.042] LT Migrant * Rural*Post 2017 -0.08 [0.128] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES  Dep Var - Mean 0.624 0.624 0.624 0.624 Observations 1912237   1912237   1912237  1912237   R-squared 0.225 0.226 0.225 0.225 69 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR   Dependent Variable: Worker is unemployed Table C3   (1) (2) (3) (4) Labor outcomes Venezuelan Migrant 0.035*** 0.055*** 0.049*** 0.051*** for migrants with [0.008] [0.008] [0.006] [0.007] interactions Migrant* Post 2017 -0.015** -0.011** -0.011** -0.013** [0.006] [0.005] [0.005] [0.005] Migrant*Female 0.026*** [0.006] Migrant *Female*Post 2017 0.009 [0.009] Migrant* Youth -0.021*** [0.007] Migrant*Youth*Post 2017 0.006 [0.005] Migrant*No Diploma -0.006 [0.012] Migrant* No Diploma* Post 2017 0.016 [0.016] Migrant * Rural -0.036** [0.014] Migrant* Rural* Post 2017 -0.006 [0.031] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES Dep Var - Mean 0.078 0.078 0.078 0.078 Observations 1912237 1912237 1912237 1912237 R-squared 0.024 0.024 0.024 0.024   Dependent Variable: Real wage Table C4   (1) (2) (3) (4) Venezuelan Migrant -932.892*** -887.963*** -685.545*** -729.615*** Labor outcomes [108.211] [196.171] [149.820] [162.969] for migrants with Migrant* Post 2017 -1040.774*** -1218.217*** -1072.066*** -1007.181*** interactions [169.275] [278.747] [219.434] [233.208] Migrant*Female 648.797*** [222.800] Migrant *Female*Post 2017 -49.017 [248.532] Migrant* Youth 897.053*** [271.600] Migrant*Youth*Post 2017 436.963 [267.304] Migrant*No Diploma 371.939* [184.518] Migrant* No Diploma* Post 2017 661.202*** [190.993] Migrant * Rural 628.189** [240.566] 70 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR   Dependent Variable: Real wage   (1) (2) (3) (4) Migrant* Rural* Post 2017 744.912 [504.431] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES Dep Var - Mean 7538.482 7538.482 7538.482 7538.482 Observations 1018859 1018859 1018859 1018859 R-squared 0.285 0.285 0.285 0.285   Dependent Variable: Weekly Hours of Work Table C5   (1) (2) (3) (4) Labor outcomes Venezuelan Migrant 1.673*** 0.598* 1.575*** 1.722*** [0.345] [0.296] [0.334] [0.387] for migrants with Migrant* Post 2017 1.286*** 1.582*** 1.584*** 1.496*** interactions [0.289] [0.310] [0.270] [0.315] Migrant*Female -0.134 [0.373] Migrant *Female*Post 2017 0.775 [0.512] Migrant* Youth 4.279*** [0.421] Migrant*Youth*Post 2017 -0.596 [0.592] Migrant*No Diploma 1.893 [1.330] Migrant* No Diploma* Post 2017 0.750 [1.457] Migrant * Rural -1.306 [0.832] Migrant* Rural* Post 2017 2.182 [3.201] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES Dep Var - Mean 45.084 45.084 45.084 45.084 Observations 1192979 1192979 1192979 1192979 R-squared 0.111 0.111 0.111 0.111 71 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Dependent Variable: Informal Worker Table C6     (1) (2) (3) (4) Labor outcomes Venezuelan Migrant 0.094*** 0.113*** 0.113*** 0.118*** for migrants with [0.011] [0.006] [0.008] [0.008] interactions Migrant* Post 2017 0.050*** 0.055*** 0.056*** 0.048*** [0.012] [0.009] [0.007] [0.008] Migrant*Female 0.042*** [0.014] Migrant *Female*Post 2017 0.012 [0.017] Migrant* Youth -0.007 [0.015] Migrant*Youth*Post 2017 0.003 [0.011] Migrant*No Diploma -0.055* [0.028] Migrant* No Diploma* Post 2017 -0.054 [0.033] Migrant * Rural -0.090*** [0.015] Migrant* Rural* Post 2017 -0.041 [0.047] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES Dep Var - Mean 0.578 0.578 0.578 0.578 Observations 1192979 1192979 1192979 1192979 R-squared 0.277 0.277 0.277 0.277 72 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR   Dependent Variable: Worker is unemployed Table C7   (1) (2) (3) (4) Labor outcomes Short-Term Venezuelan Migrant 0.080*** 0.097*** 0.087*** 0.090*** for migrants with [0.010] [0.010] [0.008] [0.009] interactions, Long-Term Venezuelan Migrant -0.003 0.022*** 0.015** 0.017** disaggregated by [0.008] [0.007] [0.007] [0.008] ST Migrant * Post 2017 -0.029*** -0.020** -0.019*** -0.021*** recent and longer- [0.010] [0.008] [0.007] [0.006] term status LT Migrant* Post 2017 0.001 -0.001 0.000 -0.001 [0.005] [0.006] [0.006] [0.008] ST Migrant* Female 0.013 [0.011] ST Migrant* Female* Post 2017 0.021 [0.015] LT Migrant* Female 0.037*** [0.005] LT Migrant* Female* Post 2017 -0.002 [0.008] ST Migrant* Youth -0.030** [0.012] ST Migrant*Youth*Post 2017 0.008 [0.012] LT Migrant*Youth -0.024*** [0.006] LT Migrant*Youth*Post 2017 0.008 [0.007] ST Migrant * No Diploma -0.027* [0.015] ST Migrant * No Diploma*Post 2017 0.033 [0.021] LT Migrant * No Diploma 0.015 [0.019] LT Migrant * No Diploma*Post 2017 0.003 [0.026] ST Migrant * Rural -0.051** [0.023] ST Migrant * Rural*Post 2017 -0.085*** [0.019] LT Migrant * Rural -0.015 [0.012] LT Migrant * Rural*Post 2017 0.041 [0.054] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES  Dep Var - Mean 0.078 0.078 0.078 0.078 Observations 1912237 1912237 1912237 1912237 R-squared 0.024 0.024 0.024 0.024 73 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR   Dependent Variable: Real wage Table C8   (1) (2) (3) (4) Labor outcomes Recent Venezuelan Migrant -1625.036*** -1731.495*** -1413.733*** -1420.799*** for migrants with [167.167] [180.438] [152.045] [161.138] interactions, Longer-term Venezuelan Migrant -426.774** -337.013 -150.020 -201.742 disaggregated by [161.851] [330.567] [268.923] [296.550] ST Migrant * Post 2017 -1086.468*** -1239.042*** -1024.042*** -998.630*** recent and longer- [178.825] [182.699] [148.657] [155.212] term status LT Migrant* Post 2017 -1054.765*** -1243.440*** -1143.749*** -1069.608*** [206.907] [365.808] [302.368] [331.392] ST Migrant* Female 594.361*** [153.603] ST Migrant* Female* Post 2017 133.705 [244.864] LT Migrant* Female 694.196* [380.267] LT Migrant* Female* Post 2017 -157.111 [360.213] ST Migrant* Youth 1199.187*** [201.822] ST Migrant*Youth*Post 2017 500.966** [216.110] LT Migrant*Youth 920.924** [376.396] LT Migrant*Youth*Post 2017 339.544 [341.183] ST Migrant * No Diploma 1075.709*** [277.868] ST Migrant * No Diploma*Post 2017 -86.195 [275.307] LT Migrant * No Diploma -147.248 [245.793] LT Migrant * No Diploma*Post 2017 1197.381*** [265.213] ST Migrant * Rural 492.986* [259.325] ST Migrant * Rural*Post 2017 2076.703*** [615.846] LT Migrant * Rural 485.449 [376.631] LT Migrant * Rural*Post 2017 145.649 [577.381] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES  Dep Var - Mean 7538.482 7538.482 7538.482 7538.482 Observations 1018859 1018859 1018859 1018859 R-squared 0.286 0.286 0.286 0.286 74 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Dependent Variable: Weekly Hours of Work Table C9     (1) (2) (3) (4) Labor outcomes Recent Venezuelan Migrant 1.352*** 0.499 1.799*** 1.962*** for migrants with [0.332] [0.328] [0.361] [0.378] interactions, Longer-term Venezuelan Migrant 1.917*** 0.665* 1.406*** 1.535*** disaggregated by [0.453] [0.351] [0.347] [0.419] recent and longer- ST Migrant * Post 2017 1.299*** 1.521*** 1.514*** 1.376*** [0.408] [0.361] [0.275] [0.315] term status LT Migrant* Post 2017 1.246*** 1.613*** 1.647*** 1.600*** [0.417] [0.386] [0.340] [0.399] ST Migrant* Female 1.210* [0.692] ST Migrant* Female* Post 2017 0.453 [0.936] LT Migrant* Female -1.132** [0.475] LT Migrant* Female* Post 2017 1.064** [0.482] ST Migrant* Youth 4.711*** [0.738] ST Migrant*Youth*Post 2017 -0.694 [0.817] LT Migrant*Youth 3.867*** [0.598] LT Migrant*Youth*Post 2017 -0.437 [0.883] ST Migrant * No Diploma 1.418 [2.095] ST Migrant * No Diploma*Post 2017 -0.172 [2.637] LT Migrant * No Diploma 2.230 [2.137] LT Migrant * No Diploma*Post 2017 1.349 [2.319] ST Migrant * Rural -2.182** [0.847] ST Migrant * Rural*Post 2017 4.025 [4.158] LT Migrant * Rural -0.805 [1.026] LT Migrant * Rural*Post 2017 1.112 [4.729] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES  Dep Var - Mean 45.084 45.084 45.084 45.084 Observations 1192979 1192979 1192979 1192979 R-squared 0.111 0.111 0.111 0.111 75 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Dependent Variable: Informal Worker Table C10     (1) (2) (3) (4) Labor outcomes Recent Venezuelan Migrant 0.135*** 0.157*** 0.155*** 0.159*** for migrants with [0.016] [0.007] [0.012] [0.012] interactions, Longer-term Venezuelan Migrant 0.064*** 0.083*** 0.081*** 0.087*** disaggregated by [0.011] [0.008] [0.007] [0.008] recent and longer- ST Migrant * Post 2017 0.060*** 0.061*** 0.061*** 0.054*** [0.019] [0.014] [0.012] [0.013] term status LT Migrant* Post 2017 0.047*** 0.053*** 0.056*** 0.048*** [0.014] [0.012] [0.012] [0.010] ST Migrant* Female 0.047** [0.017] ST Migrant* Female* Post 2017 -0.001 [0.025] LT Migrant* Female 0.040** [0.017] LT Migrant* Female* Post 2017 0.019 [0.018] ST Migrant* Youth -0.013 [0.028] ST Migrant*Youth*Post 2017 -0.001 [0.023] LT Migrant*Youth -0.016 [0.012] LT Migrant*Youth*Post 2017 0.006 [0.016] ST Migrant * No Diploma -0.075* [0.041] ST Migrant * No Diploma*Post 2017 -0.033 [0.042] LT Migrant * No Diploma -0.039 [0.041] LT Migrant * No Diploma*Post 2017 -0.071 [0.051] ST Migrant * Rural -0.101*** [0.021] ST Migrant * Rural*Post 2017 -0.114 [0.072] LT Migrant * Rural -0.074*** [0.023] LT Migrant * Rural*Post 2017 -0.003 [0.058] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES  Dep Var - Mean 0.578 0.578 0.578 0.578 Observations 1192979 1192979 1192979 1192979 R-squared 0.277 0.277 0.277 0.277 76 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Table C11 Agri-food employment for migrants in agriculture-intense locations, disaggregated by recent and longer- term status Employed in Employed in primary Employed in food   Employed in food services agri-food sector agriculture processing   (1) (2) (3) (4) (5) (6) (7) (8) ST Migrant 0.097*** 0.088*** -0.013*** -0.020*** -0.001 0.000 0.110*** 0.107*** [0.019] [0.010] [0.004] [0.004] [0.004] [0.003] [0.016] [0.008] LT migrant 0.029* 0.027*** -0.022* -0.019** 0.003 0.002 0.048*** 0.044*** [0.015] [0.009] [0.013] [0.007] [0.005] [0.003] [0.007] [0.006] ST Migrant* Post 2017 0.011 0.018 -0.013* -0.010* 0.011 0.008* 0.013 0.020** [0.017] [0.013] [0.007] [0.005] [0.006] [0.005] [0.013] [0.009] LT Migrant* Post 2017 0.041*** 0.043*** 0.007 -0.002 0.003 0.003 0.031** 0.041*** [0.011] [0.008] [0.012] [0.007] [0.006] [0.003] [0.013] [0.008] ST Migrant* Share of agriculture -0.122 -0.108** 0.015 -0.029 in department [0.168] [0.041] [0.051] [0.142] ST Migrant*share of ag*Post 2017 0.146 0.031 -0.039 0.153 [0.140] [0.063] [0.056] [0.136] LT Migrant* Share of agriculture -0.056 0.038 -0.020 -0.074 in department [0.144] [0.097] [0.053] [0.058] LT Migrant*share of ag*Post 2017 0.078 -0.142 0.010 0.210** [0.095] [0.088] [0.062] [0.091] ST Migrant*Intensive agriculture -0.001 -0.011 0.002 0.008 in department [0.017] [0.007] [0.011] [0.020] ST Migrant*Intense*Post 2017 0.031 -0.002 -0.006 0.038 [0.031] [0.014] [0.011] [0.032] ST Migrant*Intensive agriculture -0.021 0.001 -0.001 -0.021* in department [0.022] [0.011] [0.010] [0.011] ST Migrant*Intense*Post 2017 0.041* -0.022** 0.005 0.058*** [0.020] [0.010] [0.015] [0.014] Individual controls YES YES YES YES YES YES YES YES Department*Year*Month FE YES YES YES YES YES YES YES YES Dep Var - Mean 0.168 0.168 0.074 0.074 0.029 0.029 0.065 0.065 Observations 1192979 1192979 1192979 1192979 1192979 1192979 1192979 1192979 R-squared 0.147 0.147 0.303 0.303 0.007 0.007 0.039 0.039 77 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Table C12 Agri-food employment for migrants in food-processing-intense locations, disaggregated by recent and longer-term status Employed in Employed in primary Employed in food Employed in food   agri-food sector agriculture processing services   (1) (2) (3) (4) (5) (6) (7) (8) ST Migrant 0.113*** 0.096*** -0.014** -0.020*** 0.005 0.002 0.122*** 0.113*** [0.024] [0.012] [0.006] [0.004] [0.004] [0.003] [0.022] [0.011] LT migrant 0.046*** 0.031*** 0.001 -0.013** 0.006 0.004 0.039*** 0.040*** [0.014] [0.009] [0.008] [0.005] [0.006] [0.004] [0.013] [0.008] ST Migrant* Post 2017 -0.008 0.008 -0.019* -0.014** -0.003 0.002 0.015 0.020** [0.023] [0.012] [0.010] [0.006] [0.008] [0.004] [0.016] [0.009] LT Migrant* Post 2017 0.038** 0.043*** -0.027*** -0.013*** -0.003 0.001 0.069*** 0.054*** [0.014] [0.010] [0.008] [0.004] [0.008] [0.004] [0.015] [0.009] ST Migrant* Share of food -0.425 -0.106 -0.086** -0.233 processing firms in department [0.292] [0.106] [0.033] [0.238] ST Migrant*share of food 0.482 0.145 0.184* 0.153 proc*Post 2017 [0.341] [0.126] [0.104] [0.262] LT Migrant* Share of food -0.326* -0.307* -0.075 0.056 processing firms in department [0.175] [0.165] [0.061] [0.141] LT Migrant*share of food 0.096 0.355** 0.112 -0.371* proc*Post 2017 [0.146] [0.157] [0.085] [0.197] ST Migrant*Intensive food -0.03 -0.004 -0.008** -0.018 processing in department [0.019] [0.009] [0.004] [0.012] ST Migrant*Intense*Post 2017 0.051 0.014 0.020*** 0.017 [0.031] [0.010] [0.006] [0.024] LT Migrant*Intensive food -0.017 -0.018 -0.007 0.009 processing in department [0.017] [0.014] [0.004] [0.009] LT Migrant*Intense*Post 2017 0.005 0.025* 0.007 -0.027* [0.012] [0.014] [0.005] [0.014] Individual controls YES YES YES YES YES YES YES YES Department*Year*Month FE YES YES YES YES YES YES YES YES Dep Var - Mean 0.167 0.167 0.072 0.072 0.03 0.03 0.065 0.065 Observations 1107113 1107113 1107113 1107113 1107113 1107113 1107113 1107113 R-squared 0.143 0.143 0.301 0.301 0.007 0.007 0.039 0.039 78 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Table C13 Agri-food employment for migrants in food-retailing-intense locations, disaggregated by recent and longer-term status Employed in Employed in primary Employed in food Employed in food   agri-food sector agriculture processing services   (1) (2) (3) (4) (5) (6) (7) (8) ST Migrant 0.087*** 0.093*** -0.023*** -0.021*** 0 0.002 0.110*** 0.112*** [0.013] [0.012] [0.004] [0.004] [0.003] [0.003] [0.011] [0.011] LT migrant 0.038*** 0.030*** -0.012** -0.014** 0.004 0.003 0.046*** 0.040*** [0.009] [0.009] [0.005] [0.005] [0.004] [0.004] [0.008] [0.008] ST Migrant* Post 2017 0.032** 0.013 -0.006 -0.014** 0.005 0.003 0.033*** 0.025** [0.013] [0.013] [0.006] [0.006] [0.004] [0.004] [0.010] [0.010] LT Migrant* Post 2017 0.036*** 0.043*** -0.012*** -0.012*** 0.001 0.001 0.047*** 0.054*** [0.010] [0.010] [0.004] [0.004] [0.004] [0.004] [0.008] [0.008] ST Migrant* Share of food 0.099 0.066*** 0.057* -0.025 retailing firms in department [0.083] [0.021] [0.028] [0.061] ST Migrant*share of food -0.590*** -0.293*** -0.081 -0.215 retail*Post 2017 [0.145] [0.080] [0.050] [0.142] LT Migrant* Share of food -0.281*** -0.044* -0.038 -0.199*** retailing firms in department [0.052] [0.024] [0.024] [0.036] LT Migrant*share of food 0.338*** -0.049 0.043 0.343*** retail*Post 2017 [0.115] [0.060] [0.065] [0.096] ST Migrant*Intensive food -0.042** -0.01 -0.005 -0.027** retailing in department [0.016] [0.006] [0.005] [0.012] ST Migrant*Intense*Post 2017 0.058* 0.022 0.004 0.032 [0.028] [0.014] [0.007] [0.024] LT Migrant*Intensive food 0.009 0.003 -0.003 0.008 retailing in department [0.019] [0.005] [0.004] [0.015] LT Migrant*Intense*Post 2017 0.012 -0.008 0.009 0.011 [0.029] [0.005] [0.009] [0.023] Individual controls YES YES YES YES YES YES YES YES Department*Year*Month FE YES YES YES YES YES YES YES YES Dep Var - Mean 0.17 0.17 0.076 0.076 0.029 0.029 0.064 0.064 Observations 1086132 1086132 1086132 1086132 1086132 1086132 1086132 1086132 R-squared 0.154 0.154 0.314 0.314 0.007 0.007 0.039 0.039 79 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR   Dependent variable: Real Wage in Agri-food sector Table C14   (1) (2) (3) (4) Labor outcomes Venezuelan Migrant -399.210 -595.338*** of migrants in [375.154] [199.668] Migrant* Post 2017 -841.534*** -709.145*** agri-food sector in [121.706] [127.992] agriculture-intensive Migrant* Share of agriculture in department -3734.342 areas [3053.067] Migrant*share of ag*Post 2017 2275.777** [1063.620] Migrant*Intensive agriculture in department -1098.973*** [321.942] Migrant*Intense*Post 2017 662.310** [237.767] ST Migrant -666.309 -1049.555*** [458.267] [255.961] LT migrant -174.327 -192.738 [328.047] [178.160] ST Migrant* Post 2017 -1217.601*** -839.295*** [192.142] [172.560] LT Migrant* Post 2017 -630.864*** -680.340*** [199.252] [165.694] ST Migrant* Share of agriculture in department -6848.384* [3461.528] ST Migrant*share of ag*Post 2017 6273.085*** [1118.970] LT Migrant* Share of agriculture in department -870.288 [3289.225] LT Migrant*share of ag*Post 2017 -505.739 [2274.655] ST Migrant*Intensive agriculture in department -1761.102*** [323.566] ST Migrant*Intense*Post 2017 1490.434*** [213.261] ST Migrant*Intensive agriculture in department -549.030 [591.955] ST Migrant*Intense*Post 2017 165.877 [534.550] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES Dep Var - Mean 4864.072 4864.072 4864.072 4864.072 Observations 168089 168089 168089 168089 R-squared 0.167 0.167 0.167 0.167 80 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR   Dependent variable: Real Wage in Agrifood sector Table C15   (1) (2) (3) (4) Labor outcomes Venezuelan Migrant -1274.265** -847.871*** of migrants in [493.629] [263.221] Migrant* Post 2017 -672.299** -665.356*** agri-food sector in [262.801] [147.060] food-processing- Migrant* Share of food processing firms in department 9112.246 intensive areas [5833.179] Migrant*share of food proc*Post 2017 545.190 [3529.245] Migrant*Intensive food processing in department 493.437 [357.138] Migrant*Intense*Post 2017 18.309 [242.334] ST Migrant -1903.991*** -1397.519*** [662.367] [338.831] LT migrant -623.171 -313.384 [384.600] [239.276] ST Migrant* Post 2017 -809.673* -722.183*** [445.034] [219.705] LT Migrant* Post 2017 -673.577** -710.508*** [295.618] [209.011] ST Migrant* Share of food processing firms in depart- 11372.263 ment [7539.061] ST Migrant*share of food proc*Post 2017 1818.909 [5180.953] LT Migrant* Share of food processing firms in depart- 5779.990 ment [4582.493] LT Migrant*share of food proc*Post 2017 472.882 [3649.574] ST Migrant*Intensive food processing in department 700.326 [427.342] ST Migrant*Intense*Post 2017 -37.844 [326.557] LT Migrant*Intensive food processing in department 199.993 [338.016] LT Migrant*Intense*Post 2017 167.844 [276.747] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES Dep Var - Mean 4985.830 4985.830 4985.830 4985.830 Observations 156912 156912 156912 156912 R-squared 0.164 0.164 0.164 0.164 81 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR   Dependent variable: Real Wage in Agrifood sector Table C16   (1) (2) (3) (4) Labor outcomes Venezuelan Migrant -994.811*** -860.084*** of migrants in [332.593] [256.903] Migrant* Post 2017 -786.222*** -678.918*** agri-food sector [169.110] [142.232] in food-retailing- Migrant* Share of food retailing firms in department 7124.901 intensive areas [6562.230] Migrant*share of food retail*Post 2017 4849.241 [3927.994] Migrant*Intensive food retailing in department 190.928 [279.667] Migrant*Intense*Post 2017 21.341 [202.871] ST Migrant -1518.118*** -1421.926*** [400.702] [334.887] LT migrant -672.417 -320.939 [583.401] [230.693] ST Migrant* Post 2017 -1228.094*** -725.798*** [373.861] [216.303] LT Migrant* Post 2017 -571.160 -708.858*** [488.489] [202.348] ST Migrant* Share of food retailing firms in depart- 6461.550 ment [5350.593] ST Migrant*share of food retail*Post 2017 23151.098 [15620.334] LT Migrant* Share of food retailing firms in depart- 15409.810 ment [23367.099] LT Migrant*share of food retail*Post 2017 -5809.826 [22251.576] ST Migrant*Intensive food retailing in department 542.553 [379.428] ST Migrant*Intense*Post 2017 -219.988 [293.491] LT Migrant*Intensive food retailing in department -207.065 [246.195] LT Migrant*Intense*Post 2017 242.526 [269.661] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES Dep Var - Mean 4964.407 4964.407 4964.407 4964.407 Observations 154964 154964 154964 154964 R-squared 0.168 0.168 0.168 0.168 82 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Dependent variable: Weekly hours of work in agri-food Table C17   (1) (2) (3) (4) Labor outcomes Venezuelan Migrant 2.103*** 2.347*** of migrants in [0.611] [0.650] Migrant* Post 2017 2.405*** 2.298*** agri-food sector in [0.641] [0.650] agriculture-intensive Migrant* Share of agriculture in department 3.263 areas [5.314] Migrant*share of ag*Post 2017 -0.497 [6.860] Migrant*Intensive agriculture in department 0.130 [0.969] Migrant*Intense*Post 2017 0.607 [1.332] ST Migrant 2.948** 3.164*** [1.061] [0.851] LT migrant 1.306* 1.579** [0.640] [0.593] ST Migrant* Post 2017 1.866 1.443 [1.228] [0.894] LT Migrant* Post 2017 2.874*** 3.096*** [0.739] [0.687] ST Migrant* Share of agriculture in department 4.048 [10.825] ST Migrant*share of ag*Post 2017 -5.515 [13.620] LT Migrant* Share of agriculture in department 2.627 [7.184] LT Migrant*share of ag*Post 2017 4.934 [9.924] ST Migrant*Intensive agriculture in department 1.313 [2.335] ST Migrant*Intense*Post 2017 -0.881 [2.805] ST Migrant*Intensive agriculture in department -0.831 [1.198] ST Migrant*Intense*Post 2017 1.952 [1.921] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES Dep Var - Mean 42.395 42.395 42.395 42.395 Observations 199897 199897 199897 199897 R-squared 0.157 0.157 0.157 0.157 83 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR   Dependent variable: Weekly hours of work in agrifood Table C18   (1) (2) (3) (4) Labor outcomes Venezuelan Migrant 3.187** 2.704*** of migrants in [1.198] [0.758] Migrant* Post 2017 1.447 1.985** agri-food sector in [1.045] [0.789] food-processing- Migrant* Share of food processing firms in depart- -12.351 intensive areas ment [17.334] Migrant*share of food proc*Post 2017 13.111 [12.564] Migrant*Intensive food processing in department -0.964 [1.298] Migrant*Intense*Post 2017 0.928 [1.164] ST Migrant 4.296*** 3.509*** [1.474] [0.952] LT migrant 1.968 1.866** [1.235] [0.752] ST Migrant* Post 2017 0.426 1.042 [1.617] [1.013] LT Migrant* Post 2017 2.607** 2.939*** [1.039] [0.924] ST Migrant* Share of food processing firms in -15.519 department [23.860] ST Migrant*share of food proc*Post 2017 13.244 [24.211] LT Migrant* Share of food processing firms in -6.658 department [15.185] LT Migrant*share of food proc*Post 2017 9.627 [8.322] ST Migrant*Intensive food processing in department -0.588 [1.936] ST Migrant*Intense*Post 2017 0.730 [2.013] LT Migrant*Intensive food processing in department -0.965 [0.994] LT Migrant*Intense*Post 2017 0.783 [0.967] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES Dep Var - Mean 42.659 42.659 42.659 42.659 Observations 185233 185233 185233 185233 R-squared 0.155 0.155 0.155 0.155 84 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR   Dependent variable: Weekly hours of work in agrifood Table C19   (1) (2) (3) (4) Labor outcomes Venezuelan Migrant 2.886*** 2.630*** of migrants in [0.665] [0.737] Migrant* Post 2017 1.501* 2.105** agri-food sector [0.746] [0.766] in food-retailing- Migrant* Share of food retailing firms in depart- -14.874*** intensive areas ment [2.222] Migrant*share of food retail*Post 2017 29.235** [10.959] Migrant*Intensive food retailing in department -0.893 [0.967] Migrant*Intense*Post 2017 0.593 [1.861] ST Migrant 3.641*** 3.375*** [0.875] [0.940] LT migrant 2.477*** 1.866** [0.640] [0.716] ST Migrant* Post 2017 0.374 1.234 [1.207] [0.994] LT Migrant* Post 2017 2.110** 2.978*** [0.787] [0.887] ST Migrant* Share of food retailing firms in -12.669*** department [2.315] ST Migrant*share of food retail*Post 2017 37.050 [31.131] LT Migrant* Share of food retailing firms in -32.286*** department [5.702] LT Migrant*share of food retail*Post 2017 43.666*** [6.936] ST Migrant*Intensive food retailing in department -0.574 [1.532] ST Migrant*Intense*Post 2017 -0.125 [2.973] LT Migrant*Intensive food retailing in department -0.850 [0.806] LT Migrant*Intense*Post 2017 0.830 [1.364] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES Dep Var - Mean 42.713 42.713 42.713 42.713 Observations 184686 184686 184686 184686 R-squared 0.152 0.152 0.152 0.152 85 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Dependent variable: Informal worker in agri-food Table C20     (1) (2) (3) (4) Labor outcomes Venezuelan Migrant 0.065** 0.079*** of migrants in [0.024] [0.017] agri-food sector in Migrant* Post 2017 0.064*** 0.036** agriculture-intensive [0.020] [0.014] Migrant* Share of agriculture in depart- 0.191 areas ment [0.177] Migrant*share of ag*Post 2017 -0.464*** [0.157] Migrant*Intensive agriculture in depart- 0.011 ment [0.043] Migrant*Intense*Post 2017 -0.083** [0.034] ST Migrant 0.062 0.091*** [0.040] [0.022] LT migrant 0.069** 0.068*** [0.031] [0.021] ST Migrant* Post 2017 0.127*** 0.070*** [0.031] [0.023] LT Migrant* Post 2017 0.015 0.013 [0.032] [0.020] ST Migrant* Share of agriculture in 0.455 department [0.437] ST Migrant*share of ag*Post 2017 -0.945*** [0.287] LT Migrant* Share of agriculture in -0.066 department [0.280] LT Migrant*share of ag*Post 2017 -0.048 [0.345] ST Migrant*Intensive agriculture in 0.080 department [0.093] ST Migrant*Intense*Post 2017 -0.179** [0.067] ST Migrant*Intensive agriculture in -0.048 department [0.050] ST Migrant*Intense*Post 2017 -0.010 [0.062] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES Dep Var - Mean 0.752 0.752 0.752 0.752 Observations 199897 199897 199897 199897 R-squared 0.197 0.197 0.197 0.197 86 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Dependent variable: Informal worker in agrifood Table C21     (1) (2) (3) (4) Labor outcomes Venezuelan Migrant 0.117*** 0.096*** of migrants in [0.034] [0.017] agri-food sector in Migrant* Post 2017 0.026 0.026 food-processing- [0.032] [0.019] Migrant* Share of food processing firms in department -0.576 intensive areas [0.418] Migrant*share of food proc*Post 2017 -0.011 [0.361] Migrant*Intensive food processing in department -0.050* [0.028] Migrant*Intense*Post 2017 -0.001 [0.022] ST Migrant 0.128** 0.116*** [0.048] [0.021] LT migrant 0.103** 0.074** [0.042] [0.027] ST Migrant* Post 2017 0.084 0.056* [0.049] [0.029] LT Migrant* Post 2017 -0.019 0.006 [0.039] [0.029] ST Migrant* Share of food processing firms in depart- -0.507 ment [0.657] ST Migrant*share of food proc*Post 2017 -0.597 [0.597] LT Migrant* Share of food processing firms in depart- -0.587 ment [0.430] LT Migrant*share of food proc*Post 2017 0.449 [0.387] ST Migrant*Intensive food processing in department -0.071 [0.042] ST Migrant*Intense*Post 2017 -0.020 [0.040] LT Migrant*Intensive food processing in department -0.028 [0.032] LT Migrant*Intense*Post 2017 0.012 [0.032] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES Dep Var - Mean 0.744 0.744 0.744 0.744 Observations 185233 185233 185233 185233 R-squared 0.197 0.197 0.197 0.197 87 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Dependent variable: Informal worker in agrifood Table C22     (1) (2) (3) (4) Labor outcomes Venezuelan Migrant 0.103*** 0.097*** of migrants in [0.016] [0.017] agri-food sector Migrant* Post 2017 0.036 0.025 in food-retailing- [0.022] [0.018] Migrant* Share of food retailing firms in department -0.226** intensive areas [0.092] Migrant*share of food retail*Post 2017 -0.817 [0.567] Migrant*Intensive food retailing in department 0.009 [0.018] Migrant*Intense*Post 2017 -0.061* [0.034] ST Migrant 0.126*** 0.115*** [0.020] [0.021] LT migrant 0.078** 0.078*** [0.028] [0.026] ST Migrant* Post 2017 0.119*** 0.056* [0.042] [0.028] LT Migrant* Post 2017 0.005 0.001 [0.030] [0.029] ST Migrant* Share of food retailing firms in -0.289*** department [0.079] ST Migrant*share of food retail*Post 2017 -3.685** [1.428] LT Migrant* Share of food retailing firms in -0.040 department [0.399] LT Migrant*share of food retail*Post 2017 -0.133 [0.424] ST Migrant*Intensive food retailing in department 0.041 [0.031] ST Migrant*Intense*Post 2017 -0.153*** [0.050] LT Migrant*Intensive food retailing in department -0.006 [0.026] LT Migrant*Intense*Post 2017 0.005 [0.035] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES Dep Var - Mean 0.743 0.743 0.743 0.743 Observations 184686 184686 184686 184686 R-squared 0.197 0.197 0.198 0.197 88 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Dependent variable: Temporary worker in agri-food Table C23     (1) (2) (3) (4) Labor outcomes Venezuelan Migrant 0.029** 0.028** of migrants in [0.012] [0.011] agri-food sector in Migrant* Post 2017 -0.013 -0.008 agriculture-intensive [0.009] [0.005] Migrant* Share of agriculture in department -0.010 areas [0.105] Migrant*share of ag*Post 2017 0.108 [0.071] Migrant*Intensive agriculture in department 0.001 [0.027] Migrant*Intense*Post 2017 0.027*** [0.009] ST Migrant 0.029 0.037** [0.022] [0.017] LT migrant 0.030** 0.020** [0.011] [0.008] ST Migrant* Post 2017 0.020 0.010 [0.013] [0.010] LT Migrant* Post 2017 -0.038*** -0.020*** [0.013] [0.006] ST Migrant* Share of agriculture in depart- 0.151 ment [0.231] ST Migrant*share of ag*Post 2017 -0.142 [0.111] LT Migrant* Share of agriculture in depart- -0.167* ment [0.083] LT Migrant*share of ag*Post 2017 0.311** [0.136] ST Migrant*Intensive agriculture in depart- 0.049 ment [0.053] ST Migrant*Intense*Post 2017 -0.029 [0.023] ST Migrant*Intensive agriculture in depart- -0.041*** ment [0.012] ST Migrant*Intense*Post 2017 0.065** [0.025] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES Dep Var - Mean 0.044 0.044 0.044 0.044 Observations 199897 199897 199897 199897 R-squared 0.066 0.066 0.067 0.067 89 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Dependent variable: Temporary worker in agrifood Table C24     (1) (2) (3) (4) Labor outcomes Venezuelan Migrant 0.027* 0.032** of migrants in [0.014] [0.013] agri-food sector in Migrant* Post 2017 -0.006 -0.003 food-processing- [0.009] [0.006] intensive areas Migrant* Share of food processing firms in department 0.017 [0.119] Migrant*share of food proc*Post 2017 0.049 [0.172] Migrant*Intensive food processing in department -0.011 [0.016] Migrant*Intense*Post 2017 -0.003 [0.015] ST Migrant 0.031 0.041* [0.024] [0.021] LT migrant 0.021* 0.022** [0.011] [0.009] ST Migrant* Post 2017 -0.004 0.007 [0.014] [0.008] LT Migrant* Post 2017 -0.008 -0.010 [0.015] [0.010] ST Migrant* Share of food processing firms in depart- 0.163 ment [0.287] ST Migrant*share of food proc*Post 2017 0.239 [0.298] LT Migrant* Share of food processing firms in depart- -0.071 ment [0.114] LT Migrant*share of food proc*Post 2017 -0.088 [0.181] ST Migrant*Intensive food processing in department -0.001 [0.027] ST Migrant*Intense*Post 2017 0.009 [0.026] LT Migrant*Intensive food processing in department -0.015 [0.012] LT Migrant*Intense*Post 2017 -0.015 [0.013] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES Dep Var - Mean 0.041 0.041 0.041 0.041 Observations 185233 185233 185233 185233 R-squared 0.069 0.069 0.069 0.069 90 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Dependent variable: Temporary worker in agrifood Table C25     (1) (2) (3) (4) Labor outcomes Venezuelan Migrant 0.033** 0.031** of migrants in [0.012] [0.013] agri-food sector in Migrant* Post 2017 -0.003 -0.003 food-processing- [0.008] [0.006] intensive areas Migrant* Share of food retailing firms in department -0.172** [0.068] Migrant*share of food retail*Post 2017 -0.124 [0.219] Migrant*Intensive food retailing in department -0.022 [0.016] Migrant*Intense*Post 2017 -0.030** [0.011] ST Migrant 0.045** 0.041* [0.020] [0.020] LT migrant 0.020** 0.021** [0.009] [0.009] ST Migrant* Post 2017 0.006 0.006 [0.012] [0.008] LT Migrant* Post 2017 -0.008 -0.009 [0.012] [0.010] ST Migrant* Share of food retailing firms in depart- -0.204*** ment [0.071] ST Migrant*share of food retail*Post 2017 -0.121 [0.404] LT Migrant* Share of food retailing firms in depart- -0.082 ment [0.251] LT Migrant*share of food retail*Post 2017 -0.181 [0.331] ST Migrant*Intensive food retailing in department -0.011 [0.028] ST Migrant*Intense*Post 2017 -0.023 [0.018] LT Migrant*Intensive food retailing in department -0.026 [0.018] LT Migrant*Intense*Post 2017 -0.037* [0.021] Individual controls YES YES YES YES Department*Year*Month FE YES YES YES YES Dep Var - Mean 0.044 0.044 0.044 0.044 Observations 184686 184686 184686 184686 R-squared 0.070 0.070 0.070 0.070 91 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR D. ADDITIONAL EVIDENCE ON COLOMBIA’S AGRI-FOOD SECTOR Figure D1 3% 1% Agricultural Land L r e-sc le c sh crops, 2049067 h 4% uses 8% Cere ls, 1034065 h Data source: National Agricultural Survey, 2017 A ro-forestr , 77629 h 35% 8% F llow, 609829 h Roots, tubers, pl nt ns, 497867 h Fru t, 450239 h 10% Ve et bles, 254925 h Idle l nd, 191761 h Other crops, 37482 h 13% 18% Figure D2 Spatial location of agricultural production in Colombia, 2017 a. Large-scale cash crops b. Coffee Data source: National Agricultural Survey, 2017 92 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR c. Cereals d. Fruits e. Roots, tubers, plantains f. Vegetables 93 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Figure D3 250000 120,0% Ranch structure in 100,0% Colombia, 2018 200000 Source: FEDEGAN, 2018 80,0% 150000 60,0% 100000 40,0% 50000 20,0% 0 0,0% <10 11 - 25 26 - 50 51 - 100 101 - 250 251 - 500 501 - 1000 >1000 He ds of C ttle Number of f rms Cumul t ve frequenc 100% Figure D4 90% 21,9% National 36,0% 80% 41,3% 48,8% employment 70% 65,5% 60% 73,4% 67,9% composition of 84,9% 50% formality across 40% 78,1% sectors, 2018 64,0% 30% 58,7% 51,2% Source: Authors’ 20% 26,6% 34,5% 32,1% calculations based on 10% 15,1% GEIH, 2018. 0% Note: informal workers re n s s te ts rt ce ce n n to po ho n tu ls ur n v rv uc ul ns r are defined as employed M er ct nd u Se rc r lS st t Tr f es n nu A individuals who report Co ,r nc de M n working absent a formal Tr F contract. Form l Inform l A e roup - Ye rs Figure D5 >90 Age pyramid of 80-84 agriculture jobs, 2017 70-74 Data source: National 60-64 Agricultural Survey, 2017 50-54 40-44 30-34 20-24 10-14 0-4 Sh re 6,00% 4,00% 2,00% 0,00% 2,00% 4,00% 6,00% 8,00% 10,00% 12,00% (%) Fem le M le 94 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Figure D6 Number of UNAPs, by on-farm food El bor t on of p nel s nd hone s processing activity, 2014 Product on of food for hum n consumpt on Data source: National M l process n 19980 3944 Agricultural Census, 2014 Notes: Flora products Me t products process n refer to tubers, fruits, flowers, leaves, barks, and O l extr ct on resins. Process n nd tr nsform t on of flor products Su r m nuf ctur n 1126 R ce m ll n 2263 1748 1229 318 282 Variable UAPs UNAPs Total Oil extraction 968 261 1229 Sugar manufacturing 261 57 318 Rice milling 165 117 282 Elaboration of panelas and honeys 10923 1057 11980 Processing and transformation of flora products 921 205 1126 Slaughter of animals 1287 461 1748 Milk processing 1767 496 2263 Production of food for human consumption 3124 820 3944 Total 19416 3474 22890 95 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Figure D7 4 M nuf ctur n of st rches nd st rch products 4 Food processing 12 enterprises, by main M nuf ctur n of p st , noodles, cous-cous, nd f r n ceous products 8 economic activity, 24 M nuf ctur n of prep red me ls nd d shes 4 2019 24 Data source: Process n nd preserv t on of f sh, crust ce ns, nd molluscs 8 Supersociedades, 2019 36 M nuf ctur n of p nel 24 M nuf ctur n of coco , chocol te, nd su r confect oner 24 12 M nuf ctur n of su r 40 48 Process n nd preserv t on of fru t nd ve et bles 40 68 M nuf ctur n of prep red n m l feed 72 188 M nuf ctur n of b er products 32 Process n nd preserv t on of me t 200 52 M nuf ctur n of ve et ble nd n m l o ls nd f ts 180 80 216 M nuf ctur n of d r products 48 196 M nuf ctur n of r n m ll products, st rches, nd st rch products 88 184 M nuf ctur n of other food products, nes 116 0 50 100 150 200 250 Number enterpr ses Sm ll nd Med um Enterpr ses L r e Enterpr ses Note: The concept of SME is defined based on the particular type of users that can use the financial statements. SMEs are defined as entities that: (a) have no public obligation to render accounts because: (i) they do not issue, or are in the process of doing, debt or equity instruments in public markets; or (ii) they do not maintain assets in fiduciary quality for a large group of third parties such as banks, credit unions, insurance companies, brokers, investment funds and investment banks; (b) publish financial statements for general information purposes for external users, such as owners not involved in management, current or future creditors, and credit rating agencies. 96 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Figure D8 Volume of food sold by type of product in each market during 2019 Cereals Dairy products and eggs Fish Fruit Meat Processed food Tons sold > 1.000 • <=5.000 > 5.000 • <=10.000 > 10.000 • <=50.000 > 50.000 Roots, tubers, and plantains Vegetables Data source: SIPSA, 2019 97 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Figure D9 4 C fé w th food serv ce 8 Food services 24 Event C ter n enterprises, by main 20 Other rest ur nts economic activity, 16 36 2019 Other food serv ce ct v t es 8 32 Data source: Ret l s le of bever es n spec l sed stores 20 Supersociedades, 2019. Self-serv ce rest ur nts 60 4 Ret l s le of me t 96 8 Rest ur nts nd mob le food serv ce ct v t es 260 40 Ret l s le n non-spec l sed stores 408 w th m nl food nd bever es 60 0 50 100 150 200 250 300 350 400 450 Number enterpr ses Sm ll nd Med um Enterpr ses L r e Enterpr ses Note: The concept of SME is defined based on the particular type of users that can use the financial statements. SMEs are defined as entities that: (a) have no public obligation to render accounts because: (i) they do not issue, or are in the process of doing, debt or equity instruments in public markets; or (ii) they do not maintain assets in fiduciary quality for a large group of third parties such as banks, credit unions, insurance companies, brokers, investment funds and investment banks; (b) publish financial statements for general information purposes for external users, such as owners not involved in management, current or future creditors, and credit rating agencies. E. VULNERABLE POPULATIONS 1995 and 2018, more than half of whom (55.4 percent) had access to land before they were forced to leave. AND POST-CONFLICT The most important milestone in the path towards victims’ INTEGRATION PATHS reparation was Law 1448 of 2011 (Victims and Land Restitution Law), which enacts comprehensive care, assistance, and I. RETURNEES reparation measures for victims of the internal armed conflict, including the recognition of the Colombian State As of May 2019, Colombia counted 8.8 million victims of of the right to collective integral reparation of communities, the armed conflict between 1985 and 2018, of whom 7.8 organisations and peoples. Among the guiding principles million (89 percent) declared to be victims of forced for the integral reparation process,50 the Law contemplates displacement (i.e., had been forced to leave their home in land and housing restitution and acquisition of rural real order to preserve their lives). The Colombian government estate, as well as the design of special programs and has launched initiatives to promote returning home projects for the generation of rural and urban employment among displaced households under the three principles of in order to support the self-support of victims. In fact, willingness, dignity, and security, pledging to guarantee all without a proper socio-economic reintegration, households conditions for the reincorporation of returnee households who decided retuning home seem to persist in the poverty in their place of origin. Even though the return decision is trap. Indeed, recent empirical evidence shows that returnee not always straightforward, thousands of households are households are seriously lagging behind vis-à-vis local non- returning to their places of origin, and, often using their own displaced populations: they usually report poorer nutrition social networks, are managing a gradual reincorporation into status, lower subjective economic well-being, and fewer the social and economic life of their previous communities assets such as livestock (Fransen, Ruiz and Vargas-Silva, (Centro de Memoría Histórica, 2015). 2017; Verwimp and Muñoz-Mora, 2017). Figure E1 shows the number of displaced people from 1985 One of the initiatives that have been put in place by the until 2019. The biggest displacement peak was around 2001- national government to facilitate return is the Familias en 2002, when the paramilitary expansion throughout the su Tierra (FEST) program, which seeks to contribute to the Colombian territory was increasing, and around 800,000 people were forced away from their homes. Among these, 50  The five guiding principles are compensation, restitution, rehabilitation, around 2.8 million farmers were forced to migrate between satisfaction, and non-repetition guarantee (see Law 1448/2011 for details). 98 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR socio-economic stabilization, as well as the collective and If one looks at the resulting spatial distribution of FEST symbolic reparation, for the victims of forced displacement beneficiaries in Figure E2 and compares it with the in the process of return or rural relocation (FUPAD, 2017). distribution of Venezuelan migrants in Figure 3, it is obvious FEST supports former displaced households that wish to that there are some areas of heavy overlap, especially in return to their origin land, by offering conditional incentives northern regions such as La Guajira, Cesar, Bolívar, and in the components of food security and reduction of Antioquia. As the FEST program specifically targets rural basic housing deficiencies, as well as judicial and financial households and activities, this overlap is particularly resources in support of productive activities. Table C1 shows relevant for the purpose of this report, and points at a the number of returnee households under the program. In potential for rural labor markets frictions within the regions total, 111,886 former displaced persons have returned to that are simultaneously exposed to return migration and their original places. With 31.3 percent of the total, Antioquia Venezuelan flows. is the region hosting most of the returnee process, followed by Bolívar and Chocó with 10.6 and 6.5 percent of the returnee flow, respectively. Figure E1 Intern ll 900 d spl ced Number of 1,000 people 800 internally displaced 700 people in Colombia, 600 1985-2019 500 Source: Authors’ elaboration, using data 400 from Unidad de Victimas. 300 200 100 0 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017 2019 Figure E2 Distribution of 4484 - 35022 returnees under the Familias en su Tierra 1846 - 4484 program, 2019 Source: Authors’ 396 - 1846 elaboration, using data from Unidad de Víctimas. 31 - 396 No data 99 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Table E1 Cumulative number of returnees under the community levels, conducted through an integrated Familias en su Tierra program, 2019 approach composed of eight dimensions: security (protection to demobilized under risk, and monitoring recidivism), Department Number of returnees Percent National personal, productivity, education, family context, housing, Antioquia 35,022 31.30 health, and political participation. After the 2016 peace Bolivar 11,903 10.64 agreement with the FARC, around 13,202 ex-combatants Chocó 7,341 6.56 started a reintegration process, which adds to the numbers from previous processes involving other groups, such as the Cesar 7,277 6.50 disarmament process of more than 20,000 fighters from Putumayo 6,673 5.96 thirty-seven paramilitary groups between 2001 and 2019 Sucre 5,744 5.13 (ARN, 2019). Tolima 4,484 4.01 La Guajira 4,076 3.64 With more than 20 demobilization processes or peace Magdalena 4,020 3.59 negotiation with illegal actors over the last 30 years, Caldas 4,004 3.58 Colombia is one of the countries with most reintegration Cundinamarca 3,659 3.27 programs and processes targeted to ex-combatants and illegal actors worldwide (Centro Nacional de Memoría Meta 3,481 3.11 Histórica, 2015). Caquetá 2,714 2.43 Risaralda 1,846 1.65 In general, there are two paths to start a demobilization Córdova 1,835 1.64 and reintegration process in Colombia. On one side are the Norte de Santander 1,565 1.40 individual demobilized, ex-combats who voluntary decided Valle del Cauca 1,441 1.29 to leave an irregular group by escaping from it. In addition to Nariño 1,235 1.10 the reintegration process, these individuals are also enrolled into a protection program aimed at minimizing the risk of Cauca 1,160 1.04 retaliation by the illegal armed group that they have left. On Santander 693 0.62 the other side, collective demobilization involves combatants Atlántico 516 0.46 from a certain group through a collective agreement such as Guaviare 396 0.35 a peace deal (e.g. the paramilitary and FARC demobilizations Bogotá 366 0.33 started in 2003 and 2016, respectively). In either case, Huila 221 0.20 however, the path towards disarmament, demobilization Quindío 81 0.07 and reintegration (DDR) is designed as a homogeneous and inclusive process that applies to all ex-combatants, Boyacá 68 0.06 regardless of individual characteristics such as group of Arauca 34 0.03 origin, gender, or ethnicity. All ex-combatants are therefore Casanare 31 0.03 offered a broad range of services ranging from psychologic Total 111,886 100 attention, family and community reintegration, health and Data source: Unidad de Víctimas, 2019. security, to legal sponsorship and productive support (ACR, 2016). As demobilization starts at different point in time for different groups and for different people, the demobilized II. DEMOBILIZED EX-COMBATANTS population can be thus further subdivided based on the degree of completion of the reintegration process: a first In addition to tackling internal displacement, another group is composed of people who just formally entered a critical pillar of the post-conflict stabilization and recovery DDR program, but have yet to start the process; a second agenda is represented by disarmament, demobilization, encompasses those for whom the process is underway; and and reintegration. In particular, the reintegration of ex- a third pools those who have successfully completed the combatants into civilian life, i.e. the process by which DDR program. Table E2 lists the total number of individuals ex-combatants acquire civilian status and gain sustainable at each phase of the process as of February 2019. employment and income as peaceful, socially responsible and economically self-sufficient individuals, has been recognized as a challenging yet essential element to sustained peace (World Bank, 2015). The present national reintegration policy consists of a longer-term strategy involving simultaneous activities at the individual and 100 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Table E2 Population in a disarmament, demobilization Figure E3 Spatial distribution of demobilized people and reintegration program, 2019. in Colombia, 2000-2019 Phase of the reintegration process Number of people Entered 2,979 In process 6,927 Ended 24,208 Total 34,114 Out of the reintegration process Number of people Voluntarily left before completion 17,890 Never started 8,420 Data source: Colombian Reintegration Agency, 2019. Figure E3 shows the spatial location of the demobilized population in Colombia based on the information of the Reintegration Agency of the Colombian Government (ACR) between 2000 and 2019. Antioquia is by far the region with the highest inflow of demobilized population, which amounts to 12,324 people: this is 50 percent more than the next two major locations combined (Bogotá and Meta, with 5,814 and 3,408 people respectively). Again, however, # People a substantial number of ex-combatants seems to be > 100 - > 1.000 - > 10.000 - concentrated in the North of the country as well, which <=100 <=1.000 <=10.000 <=50.000 hosts between 1,600 and 2,200 in regions such as Cesar and Santander that are simultaneously exposed to high Source: Authors’ elaboration, using data from Colombian Reintegration migration rates form Venezuela. In general, the demobilized Agency, 2019. tend to be overwhelmingly males between 25 and 40 years old, although there is a substantial presence of prime-age Figure E4 Age and gender composition of Demobilized women as well (see Figure E4). People One key factor for the success of demobilization programs 4 Demob l sed is the effective reintegration of participants into an nd v du ls economically productive life. To this end, the Colombian (10,000 people) 35 DDR programs seek to provide ex-combatants with training to generate new skills that are valuable on the 3 labor market, on the one hand, and financial support and credit for productive activities, on the other. Overall, 25 entering the DDR program does not necessarily translate into employment: across departments, employment rates 2 average at around 74 percent (cf. Table E3), with significant rates of unemployment and inactivity throughout the 15 country. Most employment, moreover, seems to take place in the informal sector, as shown in the last column of Table 1 E3: overall, of the demobilized individuals who are currently employed in Colombia, only 31 percent work under a legal 05 contract, and in a number of departments the pattern is even more extreme. M le 0 Fem le 05 1 18-25 26-40 41-60 60 NA A e Source: Authors’ elaboration, using data from Colombian Reintegration Agency, 2019. 101 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Table E3 Employment outcomes of the demobilized population, by department F. INTEGRATING VENEZUELAN Department No. workers Employment Percent MIGRANTS INTO COLOMBIA’S rate (percent) formal jobs Amazonas 22 77.27 5.88 AGRI-FOOD SYSTEM: TWO CASE Antioquia 8,440 73.82 38.67 Arauca 200 65.50 16.03 STUDIES Atlántico 1,016 75.39 31.46 Bogotá 4,185 78.21 46.32 I. IN-DEPTH SEMI STRUCTURED INTERVIEWS: Bolívar 1019 73.80 25.13 RESULTS FOR THE FLORICULTURE SECTOR Boyacá 667 72.41 35.61 This sub-section presents the broad results of semi- Caldas 470 72.98 25.36 structured interviews conducted in the flower growing Caquetá 1,010 72.97 13.57 sector. The following key informants were interviewed: Casanare 751 78.83 29.39 MADR Agriculture and Forestry Value Chain Department Cauca 754 73.47 12.09 focal point for cut flower growing value chain; Mayor Cesar 2,347 76.10 17.53 of Tabio; Producers’ Association of cut-flower growing sector, Asocolflores (Director of Social Responsibility and Chocó 586 69.11 8.40 President); cut flower growing enterprise Sunshine Bouquet Córdoba 2,162 74.19 21.63 (two managers); and eight workers, four Venezuelans (three Cundinamarca 1,420 80.35 40.40 men and a woman) and four Colombians (three women and Guainía 47 82.98 10.26 a man). Guaviare 167 64.07 11.21 Huila 1,050 80.10 18.31 The Venezuelan workers at Sunshine Bouquet that were La Guajira 301 78.74 15.61 interviewed were Colombian-Venezuelan, they were from Magdalena 1,504 76.86 24.22 mixed families with Colombian parents or children. This facilitated the processing of obtaining PEP (special residence Meta 2,537 74.14 30.68 permit). Of the Venezuelans interviewed, half of them Nariño 478 59.83 16.43 come alone, one has the family in Cucuta and the woman Norte de Santander 954 70.96 21.86 interviewed lives in Tabio with her parents and daughter. Putumayo 451 71.40 14.29 Quindío 340 75.88 33.33 Risaralda 753 73.44 31.65 San Andrés 1 100.00 0.00 Box 5 Testimonial statement Santander 1,876 73.35 32.19 Sucre 672 59.67 16.96 Rosaura is a single mother of a 7-year-old Tolima 1,045 75.02 23.60 daughter. She comes from the state of Vargas in Venezuela where she worked as a civil servant. Valle del Cauca 2,182 68.15 36.11 Vaupés 63 63.49 17.50 “At the beginning I came to Colombia to buy Vichada 57 68.42 17.95 medicine for my dad and returned home. After National Total 39,883 73.56 30.62 the crisis of border closure and crime problems at the border, I decided to settle in Cucuta. I started Source: Authors’ elaboration, using data from Colombian Reintegration at Sunshine Bouquet 1 year and 7 months ago Agency, 2019. with the Mother's Day flower campaign. After the Notes: figures are presented for the individuals for whom employment season, they renewed my contract, so I brought my information is available. Totals also include individuals who voluntarily parents and my daughter”. Rosaura and her family left the DDR program before completion. live on the outskirts of Tabio and pay a rent of 600,000 Colombian pesos. She says “I would call on Colombian companies to trust Venezuelan workers. We are eager to work to help our families”. 102 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Profession/Occupation. Three of the four Venezuelans to continue studying (male and female 22 and 23 years who work at Sunshine Bouquet are degree-holders and all old). Two accessed the company through the employment had more qualified occupations in their country. Among announcement in Cucuta, another came from another part the studies mentioned are: human resources manager, of the country and a fourth came from Zipaquira, a nearby police officer in a position of trust close to the government municipality, with transport provided by the company. (woman), a graduate in Mathematics working as a teacher and a computer student. The latter was forced to leave his Quality of employment. Interviews and visits to the degree unfinished to migrate in search of work. None of Sunshine Bouquet facilities evidence the efforts of the the respondents had previous work experience in the field. companies to improve the quality and working conditions of However, they feel fortunate to have a better salary than employees. In the case of Sunshine Bouquet, the company in their country and to be able to help their families. Except has made significant investments in incentives, worker for one worker, the others came into contact with Sunshine training and labor rights to reduce staff turnover. As for Bouquet through the employment announcement in gender differences, the proportion in Sunshine Bouquet is Cucuta. Two of the four Venezuelan workers, after finishing now more balanced: 60 women- 40 men. Nowadays more the first seasonal contract, were hired indefinitely by the and more men are hired to the workforce, surely attracted company. One returned to Cucuta with his family, after his by better conditions and opportunities for professional second time working with the company (with better salary promotion. Staff turnover has been lower in Venezuelan conditions than the first time). Another would like to return workers hired, and many of them come back to apply for to Bogota in search of better opportunities. work at the next announcement. » Working day. Unlike other activities in the agricultural sector, the working day in flower-cutting companies is 8.00 hours (from 7.00 in the morning to 3.30 in Box 6 Testimonial statement the afternoon) with a break for lunch. However, the departure time depends on the production. Overtime is Ángel (25 years old) from Yaracuay says: paid additionally, but is not optional. “I worked in Bogota in food sales, with sporadic » Remuneration. The remuneration offered by Sunshine jobs in companies such as RAPI and UBER. The Bouquet varies between the minimum wage 828,116 remuneration of these jobs was not enough to pay Colombian pesos (USD 245) per month in agricultural the rent and I decided to try the flower sector on the activity up to 1,450,000 Colombian pesos per month recommendation of a friend and for the lower cost (USD 429) in high seasons. It also depends on whether the of housing and living in general outside Bogota”. hiring is direct or through allies. Overtime hours are paid at 4,400 Colombian pesos (1.30 USD) on a business day and at 6,800 Colombian pesos (2 USD) if it is a holiday. The Colombian workers interviewed think that the official minimum wage, complemented by overtime Box 7 Testimonial statement pay, is insufficient to cover their families’ expenses. “In Cucuta I carried bags at the border, when I According to DANE data51, the poverty line for populated saw the opportunity to access a contract job, I did and rural dispersed centers is 676,740 Colombian pesos not hesitate to move to Tabio. My main concern is per month (USD 200). The salary, about $ 10 a day, would the homologation of my degree, I am a graduate be above the national poverty line and also the $ 5.50 in Human Resources and in Venezuela I was the daily line defined by the World Bank for middle-income manager of a pharmaceutical network“. countries. » Formality vs. informality. The personnel working in the facilities of the companies are under formal contract. » Labor rights. Contributions to a pension fund and health The four Colombian workers interviewed had a lower level coverage for permanent workers, and occupational risk of education than the Venezuelans interviewed, ranging insurance in occupations with the highest occupational from unfinished primary school to full baccalaureate, and risk such as fumigation. They also promote a good work were of a younger age range (22, 23, 24 and 36 years) than environment with holiday celebrations such as Mother’s the Venezuelans. Two of the Colombian respondents have worked temporarily in the company and would like to leave 51  National Administrative Department of Statistics (DANE) 103 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Day, Father’s Day, workers’ birthdays. These celebrations migrants had been linked indefinitely to the company. “We and recreational spaces are used as integration strategies. have found a proactive attitude in them and their learning The company offers its workers incentives for their curve is faster than that of the local population. People are children to study, maternity and paternity assistance committed, wanting to work”, says Fredy Leiton, in charge and health work days. of organizational development at Sunshine Bouquet, Tabio, Cundinamarca. » Training. Training is ongoing for workers according to the different roles within the production chain. The workers receive training in the technical aspects of floriculture activity and also in “soft” skills such as Box 8 Testimonial statement teamwork, leadership and companionship. “The fact that people from Venezuela arrive in » Transportation services. Transportation services are Colombia is an advantage for our country to be available to workers with routes from the different more competitive, and to be able to generate more settlements and surrounding municipalities to the fields, employment and above all they come to solve a and back again. This benefit, in addition to having a problem of labor shortage for the agro industry. positive impact on the worker’s economy, provides In Sunshine Bouquet we have invested to be able security to get to work. Part of the workers’ transport to offer attractive conditions to the worker, decent and food is subsidized by the company52. housing, recreation areas, health insurance, pension and promotion opportunities in the company”, says » Safety, health, hygiene conditions, etc. Occupational Pablo Prieto, Production and Operations Manager safety is regulated by law and its compliance is integral of Sunshine Bouquet. to the companies in the sector. For the conditions of work in cut flower growing, the safety of the workers is fundamental and this is evidenced both in the study of the Universidad de Externado and in the Tabio visit to Sunshine Bouquet. All workers operate with the The Venezuelans interviewed do not consider that they are appropriate protective measures (boots, gloves, face displacing the local workforce, but rather covering jobs that and mouth masks, hats, etc.). Sunshine Bouquet made are difficult to fill. a considerable investment in facilities for temporary workers’ accommodation53 for high demand seasons, Integration and social cohesion: coexistence and integration with leisure and recreation areas. Workers have a opportunities. Two of the Venezuelan workers interviewed canteen/cafeteria where they can eat (subsidized 50% now live in the municipality of Tabio (previously staying at by the company). Once the season is over, workers are the Sunset Bouquet facilities), but on the outskirts where allowed to stay while they find housing, whether they rent is cheaper and closer to the installations54. Their children stay in the company or are looking for employment in are in school and have access to public health services. another sector. Sending money home. The most widely repeated Local employers’ reasons for hiring Venezuelan answer is that everyone, to a greater or lesser extent, sends migrants. It is evident that the problem of shortage of whatever they can, as frequently as they can, but cannot Colombian labor and high turnover, as well as the increase afford to do so on a regular basis. in the area devoted to flower-growing, are the main determinants for hiring Venezuelan labor. Local perceptions. The Colombian population perceives work in floriculture as marginal, low-wage and with few Since Sunshine Bouquet, the experience of hiring Venezuelans opportunities; and the Colombian workers interviewed see has been very positive. As of June 2019, 150 Venezuelan the work in cut flower growing as temporary and express their desire to continue their studies. Although they do not 52  Transport from Cucuta is paid for in full by the company. The cost of the blame Venezuelans, they have the perception that work is outward journey is paid to the transport company and the return in cash to scarce as a result of Venezuelan migration55. the worker so that he/she has the choice of deciding to return or make his/her own way. On exceptional occasions (worker illness or emergency), Sunshine 54  Rent for one room between 200 and 300,000 Colombian pesos. A two-room Bouquet pays for the worker to return to Cúcuta by plane. The company has apartment between 550-600,000 Colombian pesos agreed transport services for workers from nearby municipalities and several 55  According to CONPES 3950, the effect of Venezuelan migrants on regional buses a day for employees from other municipalities direct to the company’s unemployment has been especially significant in Arauca (60% of the total installations (Zipaquirá, La Naveta, Cajicá, amongst others). unemployed), Riohacha (48.6% of the unemployed) and Cucuta (23.3% of total 53  Approximately 30 million pesos were invested in container housing for 350 unemployed). workers. 104 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR tax incentives for the hiring of Venezuelan migrant labor, (ii) support or sponsorship through transportation, (iii) Box 9 Testimonial statement publicity for employment announcements; and (iv) support “The jobs are there, but you have to go looking in the training processes. for them” says Andrés (36 years old), a Venezuelan graduate in Mathematics who worked as a high It could be concluded in the analysis of this case that the school teacher in Venezuela. Now he works as floriculture sector finds in the population from Venezuela a quality assistant at the Tabio plant and has an alternative and winning strategy for both parties in expectations of being promoted in the company as filling vacancies difficult to fill with the local population, well as starting a new stage of his life in Colombia.. and in particular by the young; these jobs represent an opportunity for Venezuelans or people from Venezuela with urgent needs to generate income. The Mayor’s Office claims to have had little relationship with Sunshine Bouquet. Additionally, according to public II. IN-DEPTH SEMI STRUCTURED administration sources, although the companies in the INTERVIEWS: RESULTS FOR THE COFFEE sector are generally known for the export and formal SECTOR business vocation of all their processes, they are also known for their secrecy56. The youth of the municipality are looking Below are presented the overall results of the semi- for other job opportunities, and jobs in the cut flower sector structured interviews conducted to the actors involved in are still considered job vacancies that are difficult to occupy the coffee chain in both Bogota and the Departments of due to health risks (respiratory and bone problems due to Risaralda and Norte de Santander. The three coffee farms exposure to high temperature changes). visited in Risaralda were: (1) Santa Rosa de Cabal; (2) Belen de Umbria and (3) Santuario. In the Department of Norte de Santander, a coffee farm was visited in the municipality of Chinacota. Annex II presents the details of the interviews Box 10 Testimonial statement in the coffee sector conducted in the municipalities of Risaralda and Norte de Santander. The actors interviewed Tabio is a municipality of Cundinamarca with were the following: 28,000 inhabitants 29 kilometers from Bogotá. According to statements by the mayor of Tabio, At national level: “the added value for the municipality of the flower company’s installations is very little. On the contrary, » Directorate of Agricultural and Forestry Value Chains of the municipality has an ecological vocation, focused MADR (Responsible for the coffee chain). on ecological tourism with its natural hot springs. The huge expanse of greenhouses and plastics » Coffee producers association: Federación Nacional de diminishes the potential of the municipality in Cafeteros (FNC). other areas of development. Less than 5% of the population of Tabio works in the company” At departmental level: » Coffee producers association: Coffee Departmental Committees. Challenges and opportunities. The Venezuelan workers interviewed advocate measures by the Colombian » Employer, farm manager. government to allow them to access the labor market and become integrated in the system. They also propose that » Venezuelan workers (22 men and 2 women) companies should have access to potential workers’ judicial background and references to be sure they are hiring » Colombian workers (9 men and 1 woman) honest workers. Economic activity. According to FNC data, 96.5% of Instruments and incentives for integration into the coffee growers are small producers. At the national level labor market. The employer would see the following there is a demand of 60 to 70,000 pickers for the harvest. as good measures to be taken by the government: (i) The average size of the farms visited was larger in Risaralda (42.3 hectares) than in Norte de Santander. The farms 56  Emphasizing, however, the openness and availability encountered with the visited are in general large or belong to owners who have company visited, Sunshine Bouquet 105 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR several coffee farms with a demand for many workers. The Venezuelan population interviewed is generally In general, smaller producers and farms are less likely to younger than the Colombian population. The average age hire Venezuelans. of Venezuelan workers is 31.75 years of age, versus 46.1 of Colombian workers interviewed (Figure E5). The main activity carried out by migrant workers in coffee is picking. At the time of the visits, the harvest was already Age data at interviewed workers matches the general advanced, and the workforce was also used in other work trend that migration brings younger population strata related to the harvest, such as cleaning and fumigation. focused mainly on vacancies of difficult coverage such as coffee harvesting. A relevant presence of Venezuelan workers in the construction or repair of facilities has been observed Quality of employment. Contractual, economic within coffee farms. The coffee harvest is spread over conditions and facilities of the coffee farm. The time throughout the national territory. This has facilitated economic conditions, working hours, the punctuality of the the mobility of the migrant population across the whole remuneration and the quality of employment are the same for country. This means that workers move to farms in areas the Venezuelan and Colombian interviewees. The difference where the harvest is still at an earlier stage, thus filling the lies in the income generated according to the kilos of lack of work that the seasonality in coffee could cause. coffee collected. The main reason for hiring Venezuelan labor in coffee is the » Working day. Working day at coffee farms is between increase in production -in certain areas, with an increase in 8 to 10 hours, five or six days per week including the area planted with coffee- and the exodus of local labor Saturdays depends on labor charge. These conditions to cities in search of better opportunities. vary slightly from one farm to another. The wage is per day although the payment is weekly. Migrant profile. None of the workers interviewed had the PEP work permit and less than half had the TMF. They » Remuneration. In relation to the collector’s salary, the are therefore in an irregular situation to work. Usually, the main modality used is piecework payment, that is, fixed worker arrives to work on the farm in a group, with siblings, amount per kilo collected, which varies throughout the cousins or as a couple, but 75% of the respondents come season, between 450-650 Colombian pesos per kilo (0.13- without family burden. 0.19 USD per kilo). The price paid per kilo increases as the harvest is decreasing, in order to motivate the collector Characterization Professional / Occupational. Only since the collection becomes more complex and scarce. two out of the 24 interviewed workers were women, who arrived and were working in the coffee farms with their At the time of the visits the harvest was quite sentimental partners. In the previous months, during the advanced, and in Belen de Umbría 80% of the harvest full harvest period the presence of women in coffee farms had already been collected, so the income generated was higher. Although it is mainly a male activity, it has been is not entirely representative. In Belén de Umbría the stated that better quality is appreciated in the picking done wage was being paid per day for being a period of by women than by men. end of harvest. In periods of full harvest the wage 60% Figure E5 Average age of 50% workers in the coffee sector. 40% Source: Semi structured 30% interviews to coffee workers 20% 10% 0% < 15 15-30 30-45 >45 Venezuel n Colomb ns 106 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR can be doubled or tripled. Furthermore, the skills of Venezuelans in grain harvesting increase in time. » The Colombian health system has a universal coverage benefit, delegated by the government to borrowing Payment prices according to harvest time. In entities that subsequently charge the state for the Santa Rosa de Cabal the highest price was payed: number of people served. Assistance to people through 650 Colombian pesos per kilo picked (0.19 USD per kilo), the non-contributory system implies a higher cost than although at the same time it is where on average assistance through the contributory system. All this workers generated less income, (25,550 Colombian pesos translates into an increase for state health services. per day, working from Monday to Friday, (7.55 USD per day). In contrast, in Santuario, an area characterized by Reasons for employer (coffee growers) for hiring high production, 480 pesos colombianos per harvested Venezuelan labor. In both departments, coffee growers kilo (0.14 USD per kilo) were being paid, though average have in the coffee bean collection phase an urgent situation income was higher. of need for labor to be solved, either by rural migration to cities or abroad and by the aging of the local population. » Formality vs. informality. All contractual relationships Coffee farmers have found in Venezuelan labor an in the coffee production chain that have been observed opportunity to supply the labor shortage in the sector. operate in the informal sector, both for Colombian and Venezuelan workers, as coffee pickers or involved in Furthermore, in some cases the employer acknowledges other related tasks. The farm manager is the person the more proactive attitude of the Venezuelan to work who, apart from managing the workers of the collection and his greater flexibility versus the more demanding and other tasks is generally in turn responsible attitude of the local worker, who in some occasions feels for the “coffee beneficiary process”. This worker is his employment status threatened. usually assigned the minimum wage but without a formalized contract. In general, large farms have more openness to working with Venezuelans for harvesting. » In general, the wife of the farm manager is in charge of feeding the workers and cleaning and hygiene of Integration and social cohesion. Coexistence and accommodation. Her profit is from daily food paid by integration opportunities. In general, the Venezuelans workers and varies according to the size of the farm interviewed show satisfaction with the work of coffee and number of workers, which also varies with the peak picking. Despite not reaching a collection volume that harvest period. In a medium farm there can be up to 100 allows them better income, most have felt welcome and workers or down to 20 in the last periods of harvest. The supported to start learning (although not all). However, average food cost for the worker is 12,200 pesos (USD despite this compliance and the cases in which the workers 3.61) per worker per day. stay for months on the same farm, it is perceived that the rotation is very high and especially women, younger » Labor rights. Remuneration is paid on time by workers and those who are alone are continuously in search the employer. Workers have an unpaid rest day of farms where there is still a greater volume to harvest. on Sundays. The farm provides accommodation and food, but this is deducted from the salary. Among the Colombian workers, there are some who feel Transportation to the farm is also paid by the worker. more empathy with Venezuelans, while others think that The workers declare they have a good perception of the Venezuelans do not like to work in the fields. Others think lodging and feeding facilities in the coffee farms. that the price per kilo of coffee collected would be higher if they were not involved. » Security conditions. It is the local population that exposes the problem of insecurity and theft in Within the farms visited there is apparent harmony. neighboring farms, with Venezuelan workers often held However, it is identified that Venezuelan workers are ill- responsible for these conflicts and insecurity. perceived in neighboring farms. Thefts, micro-trafficking, and insecurity problems are the main conflicts for which » Health coverage. Most Colombian coffee pickers have Venezuelans are identified. health coverage through the public system (SISBEN57), this coverage is under a non-contributory system. In the Venezuelan workforce in coffee collection is for employers case of Venezuelans with a health problem, they have a solution to avoid losing the crop. At the same time, some been treated in emergency centers. employers in the sector also see an opportunity to have a more willing workforce with a more proactive attitude 57  Beneficiary Selection System for Social Programs (SISBEN) is a tool, than that of local workers. On the other hand, on the farms consisting of a set of rules, norms and procedures to obtain reliable and visited, interviewed coffee employers assert to paying updated socioeconomic information of specific groups in all departments, and applying the same conditions for Colombians and districts and municipalities of the country Venezuelans. 107 INTEGRATING VENEZUELAN MIGRANTS IN COLOMBIA’S AGRI-FOOD SECTOR Table E5 Average income and national and international poverty limit Límite de pobreza nacional e internacional Miembros en Al mes por USD día por Ingreso familiar Pesos día familia media persona trabajador DANE: Línea de pobreza a nivel nacional 1.029.732,00 4 257.433,00 8.581,10 2,50 DANE: Línea de pobreza para centros poblados y 676.740,00 4 169.185,00 5.639,50 1,64 rural disperso BM: Línea de pobreza extrema 1,90 BM: Línea de pobreza en países de nivel medio 5,50 (Colombia) Salario mínimo oficial en actividad agrícola en 828.116,00 27.603,87 8,03 Colombia Promedio de generación de ingresos en la recolección del café en las visitas realizadas Café en RISARALDA Salario medio recolector café Santa Rosa Cabal ( a fecha visita 21.11.19 ) 7,56 Salario medio recolector café Belén de Umbría ( a fecha visita 22.11.19 ) 9,84 Salario medio recolector café Santuario ( a fecha visita 23.11.19 ) 11,87 Café en NORTE DE SANTANDER Salario medio recolector café Chinacota ( a fecha visita 26.11.19 ) 8,72 Source: own elaboration Costs and benefits. Both in Risaralda and in Norte de The minimum shipment of remittances consulted in Santander, the coffee harvesting activity is today profitable Mekacambio is 50,000 Colombian pesos (15 USD). The for an affected population in Venezuela, where the official average delivery of those who claim to send money is minimum wage is insufficient to buy just a few products approximately 75,000 Colombian pesos (USD 22). This from the basic basket and where having relatives outside the question is perceived as personal to the interviewee, who country is the main survival mode58. seems in some cases reluctant to answer. Table E5 presents a brief comparative analysis of the average Instruments and incentives for labor insertion. income of Venezuelans working on harvesting at the farms Venezuelan workers interviewed aspire to be able to regulate visited with the poverty line at the national level by DANE themselves in order to find a better job opportunity. and at Regional level of Latin America by the World Bank. Lastly, based on the Venezuelan answers in the context of The farms visited were at different harvest times with coffee this study, we can confirm the high labor informality in the plants of different productivity. The wage in coffee activity coffee sector. Nevertheless, the fact that coffee is an export is, as described above, changing and depends on various product could open a path towards initiatives that promote factors. In spite of this, it can be said that the generation of social responsibility towards workers. income in coffee collection activity is above the income of the minimum wage established in Colombia for agricultural activity: 828,116 Colombian pesos (USD 245.30). Sending money home. There is no pattern for the periodicity of remittances or the amount sent by venezuelan workers. By having varying incomes they send money when they have a better working week. Only 8% (2 out of those interviewed) declare not to send money home because they have the family with them. Meanwhile, others declare they send money according to the needs required by their relatives in Venezuela. 58  The minimum integral salary as of October 14 in Venezuela is 300,000 bolivars. At an approximate exchange rate of 43,000 bolivars per US dollar, the minimum integral salary is USD 6.9. 108