MANAGED LABOR MIGRATION IN AFGHANISTAN: DEMOGRAPHIC PROFILE, SHORT-TERM PROJECTION, AND SUPPLY OF MIGRATION IN AFGHANISTAN Daniel Garrote Sanchez Background Paper BGP 2c to the World Bank Project on “Afghanistan: Managed International Labor Mobility as Contribution to Economic Development and Growth” STANDARD DISCLAIMER: This volume is a product of the staff of The World Bank. The findings,interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this 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. COPYRIGHT STATEMENT: The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org Managed Labor Migration in Afghanistan: Demographic Profile, Short-term Projections, and Supply of Migration in Afghanistan Daniel Garrote Sanchez1 Background Paper BGP 2c to the World Bank Project on “Afghanistan: Managed International Labor Mobility as Contribution to Economic Development and Growth” Washington, D.C and Beirut January 2018 Abstract Afghanistan has one of the youngest populations in Asia, creating strong demographic pressure in the labor market. Around 400,000 youth will continue to enter the labor force annually during the next years. Given the significant slowdown the country’s economic activity, the pace at which jobs are generated is and will be insufficient. Through an analysis of labor supply and demand, this paper estimates substantial net emigration pressures for the years 2016–2030, on the order of around 200,000 people per year. The projected profile of future migrants is one of young men with some basic level of education and from middle-income households. In addition to this economic migration, other factors like increased insecurity, conflict, and natural disasters might further accentuate these dynamics. 1 The author is now a senior researcher at the Lebanese Center for Policy Studies. He was a labor economist and migration specialist at the World Bank’s International Labor Mobility Team, Social Protection and Jobs Global Practice, Washington, DC, when the initial draft of the paper was written. List of Contents Abbreviations and Acronyms.................................................................................................... iii 1. Introduction...................................................................................................................... 5 2. Afghanistan’s Demographic Structure.................................................................................... 7 3. Demographic Projections.....................................................................................................11 3.1 Projections of Labor Supply in Afghanistan.................................................................... 11 3.2 Projections of Labor Demand in Afghanistan.................................................................. 14 3.3 The Labor Intensity of Economic Growth....................................................................... 16 4. Projected Gaps Between Supply of and Demand for Labor: the Potential Pressures to Migrate....... 17 5. Profile of Economic Migration Pressures and Key Priorities Among Future Trends........................ 23 6. Other Projected Pressures on Migratation............................................................................. 27 6.1 Insecurity and Conflict-Induced Displacement of Refugees............................................... 27 6.2 Natural Disasters....................................................................................................... 27 6.3 Urbanization............................................................................................................. 28 7. Conclusions...................................................................................................................... 29 References.............................................................................................................................. 30 List of Boxes Box 1: Data limitations and assumptions behind the projected supply of migration.............................. 17 List of Figures Figure 1: Fertility, mortality, and migration rates and their impact on population growth, 1960–2014....... 8 Figure 2: Age pyramid in Afghanistan, 2015.................................................................................. 9 Figure 3: Projected changes in population, population structure, and labor force participation,1991–2030... 11 Figure 4: Trends in the female participation rate in Afghanistan, 1990–2020....................................... 13 Figure 5: Trends in Afghanistan’s economic activity, 2003/04–2014/15............................................... 15 Figure 6: Projections of annual flows of unmet labor supply, 2015–2030............................................ 19 Figure 7: Projections of annual flows of unmet supply of labor, 2016–2030......................................... 19 Figure 8: Projections of labor supply and net migration trends, 1991-2030......................................... 21 Figure 9: Labor market opportunities by age group........................................................................ 23 Figure 10: Education levels by age group...................................................................................... 24 Figure 11: Skill levels and returns to education by age group........................................................... 25 Figure 12: Estimated annual conflict-related civilian fatalities in Afghanistan, 2006–2015...................... 27 Figure 13: Number of natural disasters and people affected in Afghanistan, 1955–2015........................ 28 Acronyms and Abbreviations GDP Gross domestic product ILO International Labour Organization IMF International Monetary Fund UN DESA United Nations Department of Economic and Social Affairs iii 1 Introduction Afghanistan has faced high levels of migration flows for the last decades in a context of persistent conflict and insecurity, severe poverty, and lack of economic opportunities. Even though forced displacement was dominant during the 1980s and 1990s, the current nature of migration is very complex, with the usual mix of diverse economic and non-economic factors. The relevance of inflows of returnees during the last two decades adds another layer of complexity, with a debate on their impact on reconstruction efforts in Afghanistan. For example, Stigter and Monsutti (2005) cast doubt on the ability to integrate a large number of refugees who have spent a prolonged period of time abroad. Overall, this mix of factors and the high and volatile patterns of “inflows” and “outflows” complicate any prediction of future trends of migration in Afghanistan. The future potential supply of Afghan labor migration will largely depend on the political stability in the country as well as its relative economic and demographic developments compared to other countries, particular neighboring Iran and Pakistan, to where the costs of migrating are lower. The evidence clearly indicates that these flows will continue in the future and can be substantial. This paper analyzes the determinants of basic migration pressures and presents essential elements for the formulation of benchmark hypotheses concerning their future development. Section 2 depicts the current demographic structure in Afghanistan. Section 3 projects the future trends of both the supply of and demand for labor, highlighting underlying assumptions and potential associated risks. These projections produce different scenarios of migration pressure derived from economic conditions in the labor market (section 4). Migration pressures (i.e. the potential supply of net emigrants) vary among individuals depending on microeconomic characteristics that lead to different propensities to migrate. Therefore, section 5 analyzes the profile and, in particular, the skills of prospective migrants. Section 6 briefly explores other pressures to migrate beyond the projected economic factors, such as security concerns, natural disasters, and urbanization. Finally, Section 7 summarizes the main results of the paper. 5 1.Introduction The main findings are that:  he workforce is expected to increase significantly during the period • T 2016–2030, with an average flow of 400,000 young Afghans entering the labor market every year. This trend is shaped by the country’s rapid demographic growth and its relatively young age structure.  ubdued economic activity, with GDP growth projected to recover very • S slowly from 1.5 percent in 2015 to 5 percent in 2030, will be unable to generate the increase in jobs required to match labor supply growth.  s a result of labor mismatches, the expected migration pressure will be • A substantial in the next 15 years, averaging around 200,000 people per year. This projection includes economic migrants only.  t the microeconomic level, the projected profile of future migrants • A is one of young men with some education and from middle-income households. Young people are more mobile, but they face higher difficulties in finding employment opportunities despite improvements in their educational attainments. Regarding the relationship between income and migration, the lowest-income individuals have lower emigration rates even if they have the incentive to leave, due to the costs of migration. n addition to the strong economic migration projected as a result of • I labor market imbalances, increased insecurity, conflict, and natural disasters might exacerbate these trends in coming years. 6 2 Afghanistan’s Demographic Structure According to the United Nations Department of Economic and Social Affairs (UN DESA 2015), the total population in Afghanistan was 32.5 million in 2015. Afghanistan’s current demographic structure, similar to that of other least developed countries, is still at an early stage of demographic transition. Until the early 2000s, death rates declined rapidly in the country while birth rates were stable at a very high level (Figure 1). As a result, this period was characterized by rapid population growth. Since 1960, the population has increased more than threefold, growing at an average rate of 2.3 percent annually, making Afghanistan one of the fastest growing countries in the world. UN DESA figures show that, during the last decade, the country started a second demographic phase with declining natality. This is slowing the pace of total population growth, but the population is surging nevertheless. It should be noted that the latest available survey data -who haven’t been incorporated to UN DESA estimates- report that total fertility rates have actually increased slightly, raising from 5.1% in 2010 (AMS, 2010) to 5.3% in 2015 (DHS, 2015). This makes our estimates of the migration pressure a lower bound and increases the importance of managed migration as promising employment tool. While the long-term upward trend in the population dynamics of the country was determined by the evolution of fertility and mortality rates, net migration caused large swings. Since the early 2000s, the natural population increase (births minus deaths) has been quite stable, at around 800,000 extra people per year. However, in periods of more political stability and less conflict (as during the first years of the United States’ intervention) and due to repatriation programs promoted by neighboring countries like Iran and Pakistan, net “in- migration” occurred, increasing the population by more than 1 million people annually. 7 2. Afghanistan’s Demographic Structure Figure 1: Fertility, mortality, and migration rates and their impact on population growth, 1960–201 BIRTH AND DEATH RATES Per 1000 60 50 40 30 20 10 0 200 200 200 196 196 196 196 198 198 198 199 199 199 199 201 201 197 197 197 Birth rate Death rate CONTRIBUTION TO CHANGES Thousands IN POPULATION 1200 700 200 -300 -800 2000 2003 2006 2009 1982 1985 1988 2012 1967 1970 1973 1976 1979 1997 1964 1994 1961 1991 Population Change Net births Net Migration Source: UN DESA 2015. Falling mortality rates not only fueled the population boom but also caused a considerable increase in the number of very young people among the total population. The population pyramid in Afghanistan for 2015 has a wide base (Figure 2), with almost half of the total population 15 years old or younger (45 percent). In contrast, only 2.3 percent are 65 or older, yielding a total dependency ratio of close to 1:1. Based on these numbers, the average Afghan in 2015 was only 21 years old, making Afghanistan one of the youngest countries in Asia. 8 2. Afghanistan’s Demographic Structure Figure 2: Age pyramid in Afghanistan, 2015 POPULATION STRUCTURE (2015) Thousands 65+ 60-64 55-59 50-54 45-49 40-44 35-39 30-34 25-29 20-24 15-19 10-14 5-9 0-4 -4000 -3000 -2000 -1000 0 1000 2000 3000 4000 Male Female Source: Laborsta ILO 2016 Overall, the current demographics in Afghanistan are characterized by declining but still relatively rapid population growth, with a very young population, the so-called “youth bulge,” which means that a large number of new workers will enter the labor force over the next decade. Demographic pressures will therefore remain high and probably increase in coming decades. 9 10 3 Demographic Projections 3.1 Projections of Labor Supply in Afghanistan The Afghan population is projected to continue to grow dramatically in the next decades, doubling its size from 28 million in 2010 to 56 million in 2050 (UN DESA 2015). In line with past rapid population growth, the Afghan labor force also grew at a fast pace during the last decades. The International Labour Organization (ILO) estimates that the annual increase in the labor force reached 400,000 people in 2015, much higher than the 200,000 observed at the turn of the century (left panel of Figure 3). Projections for the next 15 years maintain an average net flow of 400,000 new labor force entrants every year. As the population increases, this represents a decreasing rate of growth, from the current 4 percent to around 2.6 percent in 2030 (right panel of Figure 3). Figure 3: Projected changes in population, population structure, and labor force participation, 1991–2030 ANNUAL CHANGES IN THE DEMOGRAPHIC STRUCTURE Thousands 1800 1600 1400 1200 1000 800 600 400 200 0 03 05 09 23 25 29 07 27 93 95 99 01 13 15 19 21 97 17 91 11 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 19 19 19 19 19 Population Pop + 15 Pop in the labor force CONTRIBUTIONS TO THE LABOR % ia FORCE GROWTH 12 10 8 6 4 2 0 -2 03 05 09 23 25 29 07 27 93 95 99 01 13 15 19 21 97 17 91 11 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 19 19 19 19 19 Population growth Age structure Participation Rate LFP growth Source: Own calculations based on Laborsta ILO, KILM ILO, and UN DESA data. 11 3. Demographic Projections To more formally assess the drivers behind this labor supply growth, the right panel of Figure 3 decomposes it into variations of total population, the age- structure, and labor force participation. Three main observations can be drawn: 1) Demographic growth has clearly been and is expected to continue to be the main driver of the forthcoming expansion in Afghanistan’s labor force. During the 1990s it contributed more than 100 percent to labor force growth, which reduced from close to 10 percent annually to 4 percent during the last decade. As previously seen, Afghanistan’s demographic growth was shaped by both short-term net migration swings and longer-term trends in the natural rate of growth. UN DESA projections of population growth forecast a mild deceleration of population growth from 2.8 percent in 2015 to 1.8 percent in 2030. 2) The forecasted changes in the population age-structure contribute positively to the growth of the Afghan labor force, due to the generalized decline in age dependency ratios (mainly accounted for by a falling share of young people). This contribution has increased over the years and is expected to compensate for the somewhat decreasing role of population growth to sustain the rapid increase in the labor force. The working-age population is currently increasing at a more rapid rate than that of the total population, given the time lag between the observed decline in fertility and its effect on the number of young people arriving on the labor market (typically about 20 years). This evolution will entail a favorable turn in the rate of demographic dependence (ratio of persons under 15 or over 65 years old and those of working age). While in 2015 there was one person of working age per dependent, the support ratio (the inverse of the dependency ratio) is expected to increase to 1.25 working-age people per dependent in 2020, and to 1.5 in 2050 (UN DESA 2015). Therefore, not only is population size increasing but Afghanistan’s demographic structure is also increasing the labor force. This means that the strain placed by the inactive on the active will ease, with possible positive effects on the country’s development capacity. However, the positive impact on development will manifest itself only if persons of working age who want to work actually succeed in finding jobs. Thus it is key to facilitate the entry of youth into the labor market. 3) Changes in labor force participation (the active population within the working-age group) have played a negligible role, and are projected to be the same in the next years. These projections are based on two main assumptions:  et migration converges to zero in the next decades, basically minimizing • N the role of migration in the future evolution of the labor force.  he overall activity rate is mostly constant, with a minor decrease in • T the case of men and a slow increase in female participation. The labor force projections of 400,000 extra people per year are subject to heavy uncertainty and the actual number could be even higher. Regarding migration, the large historical fluctuations in inflows and outflows of migrants suggest caution about this assumption. In particular, large returns of refugees 2 would put additional pressure on the already stretched labor market, a situation that could be exacerbated by the particular challenges of reintegrating them in the economy. 2 Promoted not only by the improvement of conditions in Afghanistan, but also by the active role of neighboring countries’ governments to reduce the size of the Afghan diaspora in their countries. 12 3. Demographic Projections An increase in the female participation rate could exert even more pressure on the labor market. The participation rate of women has remained very low by international standards, placing a serious burden on the economy. Since the early 2000s, the activity rate among women increased from an estimated 15 percent to 19 percent of the total working-age group (Figure 4), in line with data from the 2011–2012 National Risk and Vulnerability Assessment (NRVA) survey. For the period 2015–2020, ILO data forecast a mere 0.8 percentage point increase. However, if the participation rate follows the upward trend of the last 15 years, it could increase to 22 percent of the total population of working-age women. This would mean another 250,000 women entering the labor force during this five-year period, with the associated pressure of absorbing them in the productive economy. Figure 4: Trends in the female participation rate in Afghanistan, 1990–2020 FEMALE ACTIVITY RATE % Working Age Women 25 20 15 10 5 0 00 02 06 08 20 04 90 92 96 98 10 12 16 18 94 14 20 20 20 20 20 20 20 20 20 20 20 19 19 19 19 19 Source: Laborsta ILO. Overall, the labor market will be subject to heavy pressure for several years, especially as the expansion of the working-age population is coupled with rising labor force participation among young male and female workers. Against this background, labor market outcomes in Afghanistan will crucially depend on whether economic growth sufficiently stimulates labor demand commensurate with the forecasted labor force expansion. 13 3. Demographic Projections 3.2 Projections of Labor Demand in Afghanistan The demand for labor in the broad sense of the term (which refers to the availability of remunerated employment and the existence of possibilities of self-employment) is determined by the dynamism of the economy and its structure, in terms of the labor intensity of economic activities. As a result, both macro- and microeconomic factors play a role (ILO 1994). To project the future development of labor demand in Afghanistan, both GDP and the elasticity of employment to GDP must be evaluated. Since the turn of the century, the Afghan economy’s growth was rather dynamic. For the period 2003/04–2012/13, GDP grew on average at a buoyant 8.6 percent annually (Figure 5). Security improvements, coupled with a large influx of aid from donor countries, fueled this economic improvement. But Afghanistan also maintained macroeconomic stability and implemented important structural reforms during that time (IMF 2015). Aid flows helped develop key social and infrastructure projects such as roads, electricity, health, and education, which translated into better economic conditions in the country (World Bank 2013). As the sectoral decomposition shows, the two sectors that contributed the most to this vibrant economic activity were transport and communication and construction (left panel of Figure 5). This reflects the efforts of reconstruction after decades of war and instability. On the other hand, the agriculture sector showed large fluctuations, contributing to the vast majority of GDP volatility. As a result, while transport and construction increased their share in the total economy during the period 2002-2015 (from 10 percent to 25 percent and from 4 percent to 13 percent, respectively), agriculture’s share significantly decreased, from 44 percent to 25 percent of total GDP in 2015 (second Graph in Figure 5). 14 3. Demographic Projections Figure 5: Trends in Afghanistan’s economic activity, 2003/04–2014/15 REAL GDP GROWTH CONTRIBUTIONS % ia 15 12 9 6 3 0 -3 -6 03/04 04/05 05/06 06/07 07/08 08/09 09/10 10/11 11/12 12/13 13/14 14/15 Agric Manuf Constr Transp & com Govern Other serv Total SHARE OF SECTORS IN THE ECONOMY 100 90 80 70 60 50 40 30 20 10 0 02/03 03/04 04/05 05/06 06/07 07/08 08/09 09/10 10/11 11/12 12/13 13/14 14/15 Agric Manuf Constr Transp & com Govern Trade & other serv Source: Central Statistics Office. The high rate of growth seen in 2003–2012 is unlikely to be sustainable in the long term and masks important vulnerabilities. Overall, the combined flows of security- and aid-related resources were equivalent to 60 percent of GDP in 2009 (World Bank 2016a), reflecting the heavy reliance on external funding. In spite of the booming economic activity, poverty rates remained at 36 percent in 2012 (World Bank 2015). More recently, economic activity slowed down markedly, to an estimated 0.8 percent in 2015 (IMF, 2017). Behind this new scenario of slow growth are declining aid funds and increasing political and security uncertainties, proving that the country is still far from genuine and self-sustained economic growth. 3 The Central Statistics of Office starting compiling GDP data only in 2003 so no information exists about previous trends in economic activity. 15 The International Monetary Fund (2017) and World Bank (2016b) forecast subdued economic growth in the next years. In light of the increasing uncertainties, the World Bank forecast that only by 2030 would 5 percent annual growth be reached. These projections assume only gradual improvement in security conditions and continued although declining foreign aid support. From this it is estimated that total security and development aid flows will decline from an equivalent of 45 percent of GDP in 2013 to about 25 percent by 2018. In October 2016, international donors met in Brussels to ensure the continuation of international political and financial support to Afghanistan and committed USD 15.2 billion over the next four years. Based on the increased support from the international community, the IMF increased Afghanistan’s forecasted growth for the next years, reaching 5 percent by 2021. However, the risks to this central scenario are still tilted to the downside pointing to an overall threat to the sustainability of economic growth. 3.3 The Labor Intensity of Economic Growth In line with most other least developed countries, the elasticity of employment to economic growth in Afghanistan during the last decade was positive but lower than one (Valensisi and Gauci 2013). According to Laborsta ILO statistics, average employment growth during the boom years was around 4 percent per year. With an average 8.6 percent economic growth rate, the elasticity during that period was about 0.47. For the most recent period of 2010–2014, which encompasses the slowdown in economic activity, the elasticity of employment to GDP was higher (0.69) as the employment rate grew at a similar pace. Due to the overall inaccuracy of data, the Laborsta ILO employment estimates must be taken with a high degree of caution. The employment effect of growth depends on a whole range of factors, from the regional and sectoral composition of economic developments to other factors such as the evolution of asset inequality (in terms of land, human capital, financial resources, etc.). The sectoral component of growth is a relevant factor to understand labor elasticity. As such, the declining relevance of a labor-intensive sector such as agriculture might lower the elasticity, although other more dynamic sectors like construction are also labor-intensive. Projects funded by international aid have generally been labor-intensive (construction, infrastructures), benefiting 6–10 percent of the working population in recent years. Therefore, the projected aid fund slowdown could have negative consequences for employment prospects in the country. 16 4 Projected Gaps Between Supply of and Demand for Labor: the Potential Pressures to Migrate In the past, international migration provided a traditional solution and was the natural response to labor market disequilibria. The evolution and current situation of the Afghan labor market, with the rapid increase in the workforce, particularly among youth, and high unemployment and underemployment rates, suggests that the capacity for employment creation may not be enough to meet the needs of new job seekers in the near future, and that the number of potential migrants will rise. This section analyzes several scenarios for the future prospects of migration pressures that would affect the supply of Afghan migrants abroad. Three different scenarios of migration are presented: the first two (Scenario 1.1 and Scenario 1.2) are based on projections of the supply of and demand for labor in Afghanistan that lead to an excess supply of labor, while the third one (Scenario 2) is based on the joint dynamics of labor supply and migration flows, setting aside the demand for labor. Box 1: Data limitations and assumptions behind the projected supply of migration The present projections’ exercise is subject to strong limitations due to data scarcity. In order to mitigate these shortcomings, several scenarios and robustness exercises are developed. Even if the different scenarios point to fairly robust results, these have to be read with a significant degree of caution and more as broad trends on migration pressures rather than accurate estimates. Data limitations:  ack of a census data, which hinders the precision of population growth estimates -L that UN DESA provides. Furthermore, UN DESA population projections do not incorporate the latest survey data that shows higher fertility rates and population growth between 2010 and 2015. Therefore, the update of demographic trends according to the latest data would entail stronger migration pressures in the country during the coming years.  mployment data: time series data in Afghanistan is weak due to the limited -E frequency of household or labor force surveys, increasing uncertainty of ILO Laborsta estimates. Main assumptions for projections:  -I LO Laborsta estimates of employment supply (labor force): the projections assume convergence of net migration to zero. However, historical patterns show high fluctuations of in- and out-migration in the country. They are also built upon a relatively constant labor force participation rate (slight decrease for men and increase for women). Further inflows of returnees and increases in female labor participation in line with historical trends would put additional pressures on the baseline estimates of the migration supply. In particular, the return of refugees from Iran and Pakistan would accelerate the fertility rates given the very young age structure of Afghans in these countries.  abor demand estimates: They are based on historical co-movements between -L economic activity and employment. Besides data limitation, labor demand projections assume a continuation of historical patterns. Focusing only on employment, they also don’t take into consideration the high presence of underemployment and vulnerable employment creates additional pressures to migrate.  igration projections: They are based exclusively on demographic and economic -M pressures in the country. Thus, they don’t take into consideration other relevant factors that can have drive migration decisions in Afghanistan, mainly insecurity and conflict, natural disasters, and urbanization. 17 4. Projected Gaps Between Supply of and Demand for Labor: the Potential Pressures to Migrate SCENARIO 1: Migration pressure from excess supply based on labor demand projections of GDP growth and labor elasticities Under this scenario, the projected migration pressure in Afghanistan up to 2030 is assumed to be due to the excess supply of labor in the economy. To project it, the dynamics of labor supply forecast by the UN DESA and Laborsta ILO are used, and compared with the simulated labor demand that would emerge, using the economic growth forecasts made by the Word Bank in 2016 and the assumption that employment elasticities will follow historical patterns. As argued earlier, the employment effect of growth depends on different elements, such as the regional and sectoral composition of economic activity. As a consequence, applying historical employment elasticities to growth is just a simple approximation that implicitly assumes that the pattern of growth remains constant over time. Employment elasticities to growth are known to present notorious variability over the years, posing further concern for their use in future projections. Thus it is particularly important to carry out robustness checks and potentially use different elasticities observed throughout time. The dynamics of labor supply projected by UN DESA and Laborsta ILO statistics are based on the interplay of three conceptually distinct elements: the size of total population, its age-class structure, and the overall labor force participation rate, which in turn results from the weighted average of labor participation rates across age classes. SCENARIO 1.1: Projections of labor demand applying employment-to-GDP elasticity of 2003–2012 to the IMF GDP growth forecast (LOW ELASTICITY) This scenario assumes that the employment intensity of economic growth for the period 2016–2030 is the same as that observed during the booming decade of 2003–2012. During this period the elasticity stayed at a 0.47, which means that for each point of economic growth only half went to employment creation and the other half to increased productivity. SCENARIO 1.2: Projections of labor demand applying employment-to-GDP elasticity of 2010–2014 to the IMF GDP growth forecast (HIGH ELASTICITY) This scenario assumes that the employment intensity of economic growth for the period 2016–2030 is the same as that observed during the more recent moderation of the last five-year period (2010–2014). During this period, GDP growth fell and employment growth kept its pace, so elasticity surged to 0.69, which means that for each point of economic growth, 70 percent went to employment creation and the other 30 percent to increased productivity. Figure 6 shows the projected growth rates of supply of and demand for labor under the two scenarios above. On one hand, labor supply progressively slows its rate of growth from 4 percent to 2.8 percent over the next 15 years. In turn, the World Bank projects a slow and moderate recovery of economic activity, from 1.5 percent in 2015 to 5 percent in 2030. Even the more optimistic scenario of a higher labor intensity of economic growth would entail a subdued acceleration of labor demand from 1 percent to 3.5 percent annual growth, unable to match labor supply growth until 2027. In the more negative scenario of a lower elasticity of employment to growth, the growth rate of labor demand would not be higher than 2.4 percent in 2030, insufficient to provide enough jobs to cover the entire supply of labor even by that time. 18 4. Projected Gaps Between Supply of and Demand for Labor: the Potential Pressures to Migrate Figure 6: Projections of annual flows of unmet labor supply, 2015–2030 PROJECTIONS OF LABOR SUPPLY AND DEMAND GROWTH % ia 6 5 4 3 2 1 0 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 GDP growth Labor Force Labor Demand (low elasticity) Labor Demand (high elasticity) Source: Own calculations, based on UN DESA, Laborsta ILO, and World Bank data. In levels, the projections quantify the amount of unmet supply of labor that would need to migrate for economic reasons in the next 15 years. Among the 400,000 Afghans who will enter the labor market every year, the high-elasticity scenario predicts that only 30 percent will be absorbed in 2016, increasing slowly until demand is able to absorb them in 2030 (Figure 7). This could exert strong pressures to migrate, averaging around 150,000 people annually and 2.2 million in total between 2016 and 2030. Under the scenario of low intensity of labor, the economy’s absorptive capacity is even lower, about 22 percent in 2016 and up to 60 percent in 2030. As a result, migration pressure will average 250,000, adding to more than 3.7 million people during the next six-year period. The average of both scenarios points to an environment of high migration pressures in Afghanistan, at an expected 200,000 migrants per year for the next decade and a half. Figure 7: Projections of annual flows of unmet supply of labor, 2016–2030 PROJECTIONS OF MIGRATION PRESSURES Thousands 350 300 250 200 150 100 50 0 2 2 2 2 2 2 2 2 2 2 3 1 1 1 1 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Source: Own calculations, based on UN DESA, Laborsta ILO, and World Bank data. 19 4. Projected Gaps Between Supply of and Demand for Labor: the Potential Pressures to Migrate An important caveat to the above simulations is that in Afghanistan, excess labor supply was traditionally channeled into underemployment and vulnerable employment, not just into unemployment. According to the Afghan Living Conditions Survey (ALCS 2013/2014), while 22 percent of the active population was unemployed in 2013/14, another 18 percent worked fewer hours than they desired. This is because many Afghans cannot afford to be unemployed in an environment of high poverty incidence. The informal sector traditionally exerted a shock-absorbing effect. However, working conditions now are poorer, usually of a seasonal or part-time nature, providing insufficient income. Therefore, labor demand measured by employment rates is likely to be biased, since it does not take into account the widespread presence of informal or vulnerable employment, and overstates the economy’s capacity to accommodate the labor supply. This objection does not invalidate the main message of the present analysis, but rather can be expected to reinforce it, since underemployment and vulnerable employment makes it even harder to generate sufficient jobs to productively employ new entrants. On top of the 200,000 people expected to face migration pressure, a further number would be only partially (and potentially insufficiently) integrated into the labor market, increasing the need to migrate in search of economic opportunities. SCENARIO 2: Migration as a share of labor supply The caveats in the previous analysis and in particular the unreliable quality of employment statistics lead to caution in basing the projections on employment elasticities. However, using a very different set of elasticities does not change the overall picture of migration pressure, indicating the robustness of the results. To further test this, Scenario 2 uses only the trends in labor supply and net migration, ignoring the more problematic estimates of labor demand. Interestingly, the net flows of economic migration closely mimicked those of labor supply during the last decades (Figure 8). For example, the five-year period 2000–2005 saw a rapid increase in additional annual flows of labor supply as well as a surge in the net flows of economic migrants (not taking into consideration the net flows of refugees). As labor supply is affected by net flows of both refugees and economic migrants, an endogeneity problem arises in the correlation shown. During 2000–2005, large flows of refugee returnees were observed, driving the increased labor supply.4 On the contrary, the role of net economic migration flows was minimal. Even so, in the absence of economic migration, labor supply would have increased more, not less, so it does not invalidate this analysis. Actually, these dynamics show that the more refugees returned to the country, the greater the pressure on the local labor market, leading to more pressures to migrate and more economic migration. The overall underlying upward trend in the labor supply projected to continue during the next years, even without counting large inflows of refugees returning to Afghanistan, points to an increase in migration pressure of a magnitude similar to that of the two previous scenarios, at around 200,000–230,000 per year (Figure 8). In the recent context of high influx of refugee returnees since 2016, more migration pressures could be created in addition to the baseline projections. 4 See the companion paper “International Labor Mobility of Nationals: Experience and evidence for Afghanistan at macro level,” a background paper to the World Bank Project on “Afghanistan: Managed International Labor Mobility as Contribution to Economic Development and Growth.” 20 4. Projected Gaps Between Supply of and Demand for Labor: the Potential Pressures to Migrate Figure 8: Projections of labor supply and net migration trends, 1991-2030 LABOR SUPPLY AND NET MIGRATION PRESSURES Thousands 500 400 300 200 100 0 91-95 96-00 01-05 06-10 11-15 16-20 21-25 26-30 Net annual economic migrats Labor Supply Source: Own calculations, based on UN DESA and Laborsta ILO data. Admittedly, the above scenarios should be interpreted with some degree of caution, given their simplifying assumptions – many of which are actually necessary to obviate data limitations. Yet the overall reading of the findings is consistent: of the 400,000 new Afghan labor force entrants every year, labor demand will only be able to accommodate 200,000, putting high pressure on the other half to migrate in search of economic opportunities. 21 22 5 Profile of Economic Migration Pressures and Key Priorities Among Future Trends Beyond the sheer numbers of projected migration pressures, it is important to analyze the profile and skills set of those more likely to migrate. As examined in a companion paper on the microeconomic evidence of international migration in Afghanistan,5 current economic migrants are mostly young men in their 20s, sons or young brothers of household heads who leave the country to find economic opportunities to complement the household income.6 Young people are more mobile but face higher difficulties in finding employment opportunities, with both unemployment and underemployment rates above total population (Figure 9). As a result, this group of recent entrants in the labor force is particularly prone to migrating. Figure 9: Labor market opportunities by age group UNEMPLOYMENT RATE BY AGE 50% 40% 30% 20% 10% 0% 15-19 20-29 30-39 40-49 50-59 60-69 Total Male Female UNDEREMPLOYMENT BY AGE (% Employment) 35% 30% 25% 20% 15% 10% 5% 0% 15-19 20-29 30-39 40-49 50-59 60-69 Total Male Female Source: Afghanistan Living Conditions Survey 2013/2014. 5 See “International Labor Mobility of Nationals: Experience and evidence for Afghanistan at micro level,” a background paper to the World Bank Project on “Afghanistan: Managed International Labor Mobility as Contribution to Economic Development and Growth.” 6 For half of the households with family members who moved abroad, remittances are the main source of income. 23 5. Profile of Economic Migration Pressures and Key Priorities Among Future Trends Today’s young Afghans are more educated than those of previous generations, which shapes the type of migration pressures that will develop in the next years. As Figure 10 shows, literacy rates sharply increased in Afghanistan in the last decade: while Afghans aged 30 years old or above have a constant illiteracy rate of 80 percent, this rate has progressively fallen to just above 30 percent for those 10–15 years old. Accordingly, young Afghans increased the primary completion rate to 70 percent and the secondary rate to 50 percent, a dramatic improvement from the rates of older generations (only 10 percent with secondary school and 20 percent with primary education). Tertiary education rates did not improve as fast as those of lower education, but still increased threefold, from less than 5 percent among those 30 years old or more to 12–13 percent for those 21–23 years old. Therefore, the current levels of education among youth project a significant rise in the skill level of the new workforce in coming years. Figure 10: Education levels by age group MAXIMUM LEVEL OF EDUCATION BY AGE 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 No Education Primary Secondary Tertiary Source: Afghanistan Living Conditions Survey 2013/2014. In spite of their higher levels of education, youth not only have more difficulties finding gainful employment but are also less able to work in higher-skilled jobs compared to older generations (left panel of Figure 11). The percentage of the population in blue-collar manual jobs is fairly constant among different age groups (around 70–80 percent), compared to 20–30 percent in higher-skilled white-collar jobs. The exception is youth 15–20 years old, who have an even higher share of low-skilled jobs, perhaps related to their initial positions in apprenticeships or other jobs that have a learning curve and progress from lower to higher skill-level tasks. In any case, the higher skills of new entrants in the workforce do not seem to translate into higher occupational levels in the labor market. This finding is also supported by the lower returns to education observed among youth, which increases by cohort age (right panel of Figure 11): for those between 15–19 years old, returns to education are not significantly different from zero, while they are 2.5 percent among those 20–24 years old, and 3 percent for those in their 30s. These differences are statistically significant. Therefore, the overall picture is that labor migration pressures tend to be felt most by male youth who: (i) are increasingly educated but have difficulties in accessing jobs; (2) have higher levels of unemployment, underemployment, and low-skilled opportunities; and (3) have seen no improvement in their occupational skill levels despite their better education. 24 5. Profile of Economic Migration Pressures and Key Priorities Among Future Trends Figure 11: Skill levels and returns to education by age group SKILL-LEVEL BY AGE 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 White collar Blue collar RETURNS TO EDUCATION (MALES) BY AGE 5% 4% 3% 2% 1% 0% -1% 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-59 60-69 Source: Afghanistan Living Conditions Survey 2013/2014. Among the 400,000 Afghans entering the labor market every year, 30 percent (120,000) are projected to be illiterate, 10 percent will have basic literacy levels (40,000), 45–50 percent will have some secondary education, and another 10–15 percent will enter with some tertiary schooling. Combining these relative weights with the different propensities to migrate (whereby those with some education are more prone to migrate than the illiterate or those with higher education), a key focus of migration policy will be on youth with primary or some secondary school education, who will account for half of the yearly increase in labor supply. 25 26 6 Other Projected Pressures on Migration In addition to migration pressures emerging from excess labor supply due to demographics and insufficient demand, three other sources of pressure prevail: insecurity and conflict, natural disasters, and urbanization. 6.1 Insecurity and Conflict-Induced Displacement of Refugees Insecurity and conflict have been a constant threat in Afghanistan for more than three decades, and successive wars have resulted in one of the largest and most prolonged refugee crises. The flows of migrants, particularly refugees, have been very sensitive to these conditions, with some years characterized by more than 1 million inflows or outflows. Although security conditions improved drastically in 2002–2006 (leading to a massive return of refugees promoted by UNHCR), another surge in insecurity arose over the last 10 years, with more than a threefold increase in civilian casualties related to the conflict between pro-government and Taliban and other forces (Figure 12). This will be a persistently strong push factor in the upcoming years, as migrants move abroad in search of safety. Moving forward, the predicted increase in insecurity can lead to more migration pressure on top of the projected economic migration figures of previous scenarios. Figure 12: Estimated annual conflict-related civilian fatalities in Afghanistan, 2006–2015 4000 3500 3000 2500 2000 1500 1000 500 0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Source: United Nations Assistance Mission to Afghanistan. 6.2 Natural Disasters Afghanistan is a landlocked country with a high incidence of natural hazards such as earthquakes, floods, landslides, and droughts (ANDMA 2008). As Afghanistan’s economy is heavily dependent on the agriculture sector, a substantial segment of society will be affected by climate change challenges. Depending on their magnitude and on local communities’ coping mechanisms, disaster-related shocks may force people to leave their place of residence either temporarily or permanently (IOM 2014). According to EM-DAT, 30 natural disasters, mostly droughts, occurred between 2011–2015, causing 1,624 deaths and affecting more than 2.1 million people (Figure 13). Over the last 20 years, the number of deaths was 17,744 and the total number of affected people around 8.4 million. Not only is the magnitude of impacts elevating, but the rising trends are worrisome: before 1990 fewer than 5 natural disasters were recorded every five-year period, but their incidence dramatically increased to 30–50 events per five-year period, with an exponential rise in the number of affected people. 7. http://www.emdat.be/ 27 6. Other Projected Pressures on Migration Figure 13: Number of natural disasters and people affected in Afghanistan, 1955–2015 IMPACT OF NATURAL DISASTERS Thousands 3500 60 3000 50 2500 40 2000 30 1500 20 1000 500 10 0 0 -6 -9 -0 -0 -1 -7 -8 -6 -8 -9 -7 1 - 06 55 66 86 96 01 76 61 81 91 11 71 20 20 20 19 19 19 19 19 19 19 19 19 Affected people Number of disasters (rhs) Source: EM-DAT. In line with this increasing trend, several reports from the Intergovernmental Panel on Climate Change (IPCC) forecast a worsening impact of climate change, causing a rise in the frequency and severity of droughts, floods, or other disasters, and making food and water supply availability more unreliable. The frequency of extreme events and abrupt temperature variations will also intensify (Nyong 2006; Stern 2006). Assessing the impact of natural disasters on future migration patterns in Afghanistan is quite challenging. While it is generally accepted that the increased incidence of natural disaster shocks as well as progressive environmental degradation due to climate change will induce population movements, the size of these population movements and the form they will take (internally displaced versus internationally) are unknown as they depend highly on individuals’ ability to anticipate and cope with these shocks (Gubert and Nordman 2009). Natural disasters have important sectoral effects on the economic structure of the country. In particular, they disproportionately affect the agriculture sector, inducing an income and employment reduction in the rural population, for whom agriculture is the main source of income. This might accelerate the transition from a rural to a more urban economy and exert further migration pressures in the medium to longer term. 6.3 Urbanization According to UN DESA (2012), the number of Afghans living in urban areas rose very rapidly in the last decades, surging from 15.7 percent in 1980 to 23.2 percent in 2010. In coming years this trend is expected to not only continue but also to accelerate, reaching estimated 43.4 percent of the total population in 2050. The distribution of the population over rural and urban areas generally has relevant effects on both labor markets and migration pressure. Urbanization affects the employability prospects of individuals as well as their earnings. In Afghanistan, the rapid growth of main urban areas like Kabul has put pressure on the capacity to generate sufficient employment opportunities. At the same time, urbanization lowers the cost of accessing information and international migrants’ networks. Experience from Maghreb MENA countries suggests that both features incentivize urban population to migrate abroad (Gubert and Nordman 2009). 28 7 Conclusions This paper analyzes the current and future economic migration pressures in Afghanistan for the years 2015-2030 based on simple models of supply and demand in the labor market. The labor supply is estimated based on demographic projections on fertility and mortality rates and age structure, and assumptions of labor force participation rates. In turn, the demand for labor is projected based on GDP projections and different scenarios of employments elasticities to the level of economic activity. While the scarcity of data and strong assumptions behind the projections warrant a cautious interpretation of the results, they provide useful plausible scenarios of upcoming pressures derived from economic conditions in the labor market. The rapid growth and young structure of the population in Afghanistan are expected to translate into a net increase in the labor force of around 400,000 per year until 2030. Thus, the youth bulge will put a lot of pressures on the labor market that would need to create the same high amount of jobs to avoid an increase in labor market mismatches which are already high (in the form of high levels of unemployment, underemployment and other forms of vulnerable employment). The perspectives from the demand side are not very favorable with an only slow economic recovery that is still very dependent on international donors’ aid. As a result, job creation will only be able to accommodate half of the newcomers in the labor force, pointing to strong strains in the economy and significant pressures to migrate: on average, around 200,000 Afghans annually could be unable to find jobs in the coming years facing thus push factors to migrate to search for income-generating activities elsewhere. On top of these economic pressures, the increasing insecurity in the country, natural disasters and the ongoing process of urbanization can accelerate these trends. The characteristics of prospective migrants depend both on the evolution of skill levels in the country and the individual propensities to migrate. Based on a microeconomic analysis, the report shows that current economic migrants are mostly young men with some education and from middle-income countries. Youth, who are more mobile and face at the same time higher barriers to enter the labor market, are more likely to migrate and thus are the main priority to target in any possible managed migration scenario. In terms of skills, the rapid increase in education levels among the youth, in addition to the higher propensity to migrate among mid-educated/ middle-income individuals, points to a more educated pool migrant workforce(from mostly illiterate to mainly primary or some secondary school). An important aspect of migration to analyze with respect to the overall skills level in the economy will then be its potential education-enhancing incentive and multiplier effects in order to compensate the outflows of Afghans that, most likely, will have higher than average education levels. 29 References AMS. 2010. “Afghanistan Mortality Survey”, Afghan Public Health Institute (APHI) of the Ministry of Public Health (MoPH) and the Central Statistics Organization (CSO). Afghanistan National Disaster Management Authority (ANDMA). 2008. “Afghanistan Disaster Profile.” ANDMA, Kabul. DHS. 2015. “Demographic and Health Survey in Afghanistan”, Central Statistics Organization (CSO) and Ministry of Public Health (MoPH). Gubert, F., and C. Nordman. 2009. “Migration trends in North Africa: focus on Morocco, Tunisia and Algeria.” OECD Journal: General Papers, Vol. (4): 75–108. OECD, Paris. International Labour Organization (ILO). 1994. “Migration pressure: What is it? What can one do about it?” Working Paper No. 55. ILO, Geneva. International Monetary Fund (IMF). 2015. “ Islamic Republic of Afghanistan: 2015 Article IV Consultation and First Review Under the Staff-Monitored Program.” IMF Country Report No. 15/324. IMF, Washington D.C. International Monetary Fund (IMF). 2017. “World Economic Outlook Database”, April, 2017. International Organization for Migration (IOM). 2014. “Afghanistan Migration Profile.” IOM, Kabul. Nyong, A. 2006. “Africa’s Vulnerability to Climate Variability and Change.” OECD Global Forum on Sustainable Development on the Benefits of Climate Change Policies, Paris, 6–7 July. Stern, N. 2006. “The Economics of Climate Change.” The Stern Review. Cambridge University Press. Stigter, E., and A. Monsutti. 2005. “Transnational Networks: Recognizing a Regional Reality.” Afghanistan Research and Evaluation Unit (AREU), Kabul. UN Department of Economic and Social Affairs (UN DESA). 2015. “Trends in International Migrant Stock: Migrants by Age and Sex.” http://www.un.org/ en/development/desa/population/migration/data/estimates2/estimates15. shtml UN Department of Economic and Social Affairs (UN DESA). 2012. “World Urbanization Prospects: The 2011 Revision.” http://esa.un.org/unpd/wup/ Valensisi, G., and A. Gauci. 2013. “Graduated without Passing? The Employment Dimension and LDCs Prospects under the Istanbul Progamme of Action.” UNECA, Addis Ababa. 30 References World Bank. 2013. “Afghanistan in Transition: Looking Beyond 2014 -.” World Bank, Washington D.C. World Bank. 2016a. “Systematic Country Diagnostic.” February. World Bank, Washington D.C. World Bank. 2016b. “Global Economic Prospects.” June. World Bank, Washington D.C 31 38