JOBS SERIES Issue No. 9 DIAGNOSTIC BANGLADESH T h o m a s F ar o l e an d Yo o n y o un g C h o Over view © 2017 International Bank for Reconstruction and Development / The World Bank. 1818 H Street NW, Washington, DC 20433, USA. Telephone: 202-473-1000; Internet: www.worldbank.org. Some rights reserved This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the govern- ments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http://creativecommons.org/licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: Thomas Farole and Yoonyoung Cho. 2017. “Bangladesh Jobs Diagnostic: Overview.” World Bank, Washington, DC. License: Creative Commons Attribution CC BY 3.0 IGO. Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. ACKNOWLEDGMENTS Lead authors of the report were Thomas Farole and Yoonyoung Cho, together with Laurent Bossavie and Reyes Aterido. Adrian Scutaru, David De Padua, Monica Paganini, Upasana Khadka, and Muhammad Faisal Ali Baig provided excellent research and analytical support. Important contributions were also received from the fol- lowing World Bank Group colleagues at various points in the process: Simon Davies, Michael Engman, Zahid Hussain, Yue Li, Manjula Luthria, Dino Merotto, Mrinal Sircar, Rebekah Smith, Hosna Ferdous Sumi, and Michael Weber. Pallavi Mittal provided excellent GIS mapping and geospatial analysis, and invaluable support on design and copy editing came from Nita Congress. Special thanks to Iffath Sharif for helpful advice and guidance throughout the process, and to Syud Amer Ahmed for his careful reading and excellent suggestions on draft versions. The work was carried out in close collaboration with the Bangladesh Bureau of Statistics (BBS). The team would like to acknowledge the support of the BBS in providing access to the primary microdata sources used in the report, including the Labor Force Survey, the Economic Census, and the Survey of Manufacturing Industries. The authors would like to thank participants of seminars and consultations held in Dhaka between November 29 and December 7, 2016, and on March 1 and June 1, 2017; and in Washington, D.C., on December 15, 2016. In addition, valuable comments and suggestions were received from Nomaan Majid and Aurelio Parisotto of the International Labour Organization. The team worked with close guidance from David Robalino, Pablo Gottret, and Stefano Paternostro. Peer reviewers were Gladys Lopez-Acevedo, Sandeep Mahajan, Cem Mete, and Michael Wong. The work was carried out under the leadership of Michal Rutkowski and Qimiao Fan. The report, produced as part of the Let’s Work Partnership in Bangladesh, is made possible through a grant from the World Bank’s Jobs Umbrella Trust Fund, which is supported by the Department for International Devel- opment/UK AID, and the governments of Norway, Germany, Austria, the Austrian Development Agency, and the Swedish International Development Cooperation Agency. iii CONTENTS ACKNOWLEDGMENTS............................................................................................................................ iii FOREWORD............................................................................................................................................. v EXECUTIVE SUMMARY.......................................................................................................................... vi OVERVIEW: BANGLADESH JOBS DIAGNOSTIC....................................................................................... 1 The Big Picture ...............................................................................................................................................1 Labor Market Outcomes.................................................................................................................................5 Sources of Job creation ................................................................................................................................ 12 Policy Direction............................................................................................................................................. 15 iv FOREWORD Bangladesh has made remarkable progress toward poverty reduction and shared prosperity. As recently as 2000, around one in three Bangladeshis was in extreme poverty based on $1.90 a day poverty line; today, this has fallen to below 13 percent. As in most countries, the vast majority of poverty reduction in Bangladesh over the past decade has been the result of higher labor earnings, and positive labor market developments have been at the center of such progress. Many factors—such as large-scale expansion of employment in manu- facturing driven by the ready-made garment sector, rapid urbanization, and international labor mobility and remittances—have contributed to positive developments in the labor market, changing the lives of many. Despite this impressive progress, Bangladesh cannot rest on its laurels. There are still more than 20 million extreme poor in the country, with many workers engaged in precarious labor activities. Bangladesh needs to build on its success to continue labor market transformations conducive for sustainable growth and poverty reduction. In this context, the role of evidence-based job policies is more pertinent than ever, and this multi- sectoral Jobs Diagnostic is a timely exercise. It provides a careful and comprehensive examination of the labor market in Bangladesh, providing insights on the sources of job creation, job quality, and access to jobs. The report conducts a macro structural assessment that includes analyses of both firms (demand side) and workers (supply side), identifies key challenges to be overcome, and provides areas for policy consideration. Challenges in the Bangladesh labor market highlighted in the diagnostic include the slowing pace of job creation despite continued growth, stagnant quality of jobs, and increasingly difficult access to jobs faced by vulnerable groups. Most striking is the poor quality of jobs, with the large majority of workers engaged in informal, unpaid, or agricultural work as opposed to formal, wage jobs. This puts numerous workers in positions of significant vulnerability. Gender disparities in the quality of jobs remain acute, with one in three working women engaged in unpaid work versus just 5 percent of working men. Addressing these challenges will require not only improving worker skills and linkages to jobs, but—critically—attention to facilitating job creation by raising the productivity of microenterprises and small and medium enterprises, unblocking regula- tory and infrastructural constraints to firm growth, and supporting diversification of manufacturing beyond the ready-made garment sector. The diagnostic both underscores the importance of addressing the jobs challenge for Bangladesh to continue its success in growth and poverty reduction, and provides an analytical underpinning for designing and imple- menting a comprehensive national jobs strategy. Given the complex nature of Bangladesh’s jobs challenges, concerted efforts are required on multiple levels to address macro, sectoral, regional, and labor policies. Such policies will contribute to the availability of jobs—stable, safe, and well-paying jobs—which is ultimately how ordinary Bangladeshis will judge the country’s development progress. Michal Rutkowski Qimiao Fan Senior Director Country Director Social Protection and Jobs Global Practice Bangladesh, Bhutan, and Nepal v EXECUTIVE SUMMARY This Jobs Diagnostic presents the characteristics and constraints of the labor market in Bangladesh, identifies the objectives of the jobs agenda, and proposes a policy framework to progress toward them. This multisectoral diagnostic assesses the relationships between supply- and demand-side factors that interact to determine job creation, quality, and inclusion outcomes. Understanding the factors that influence jobs outcomes requires a holistic approach capturing issues such as access to markets, inputs, capital, technology, skills, and matching of supply and demand. Standard labor analysis tends to miss crucial aspects of the demand side of job creation, while growth diagnostics have no direct link to jobs. The Jobs Diagnostic thus intends to provide the compre- hensive evidence base to support the development of a national jobs strategy that focuses on policies to foster an environment for more, better, inclusive jobs in Bangladesh. Bangladesh has made impressive development progress, but is facing mounting challenges on jobs Bangladesh experienced sustained per capita income growth and poverty reduction between 2003 and 2016, accompanied by strong job creation, steady structural transformation, and robust productivity growth. Per capita real gross domestic product (GDP) doubled between 2000 and 2016, with the poverty headcount rate (based on $1.90 per day poverty line) declining from 33.7 percent to below 13 percent over this period. Much of the decline in poverty can be attributed to increased labor incomes resulting from positive labor market developments. Bangladesh’s demographic transition, with the working-age population growing faster than the total population, provided a favorable condition for growth. Moreover, employment growth among the working-age population was strong (2.4 percent annually in 2003–16) with wage employment growing by 5.7 percent annually—driven in particular by large-scale job creation in manufacturing, mostly in urban areas— and contributing to growth in female employment of more than double the rate of that for the working-age population as a whole, bringing millions of women into the labor force. This job creation and structural trans- formation was accompanied by robust growth in labor productivity (4.3 percent annually), making Bangladesh one of the top performers among economies with similar income levels. However, the pace of job creation has fallen in recent years, productivity growth has been relatively weak in manufacturing, and productivity levels remain low. Employment grew at a rapid annual rate of 3.1 percent in 2003–10, allowing for lower unemployment and higher labor force participation (LFP). But these trends were reversed in 2010–16, with job growth falling to just 1.8 percent annually. Moreover, value added per worker in the industrial sector has grown at only half the level for the economy overall. Productivity in manufacturing is particularly weak, with average annual productivity growth of just 1.6 percent in 2003–10 and 2.2 percent since 2010. Productivity levels in all sectors remain extremely low by international standards, despite recent progress. Overall worker productivity in Bangladesh, measured by value added per worker, is half that of the South Asian average, less than 40 percent of the lower-middle income country average, and less than a tenth of the global average. Key jobs challenges in Bangladesh include poor quality of jobs, uneven access to jobs, slowing job creation, and lack of dynamic enterprises The quality of jobs in Bangladesh has remained poor, with substantial shares of workers engaged in informal, unpaid, or subsistence-level agricultural work. Only 22 percent of male and 20 percent of female workers are wage employees, with large shares of female workers in unpaid work and male workers in day labor. vi Informality—even among wage employees—is commonplace, with less than 40 percent of wage employees having a written contract. Access to better quality jobs varies largely by education level and location, with better educated workers in urban areas much more likely to be employed as wage and salaried workers, and lower-skilled and rural workers more likely to be self-employed. While ongoing structural transformation and increasing urbanization have reduced the share of total employment in agriculture, it remains by far the largest source of employment, accounting for 42 percent of all jobs in 2016. It is also a sector where job quality remains particularly problematic, with household, unpaid work, and underemployment common. There are substantial gender gaps in the Bangladesh labor market, reflected in low female LFP and the low share of females in nonagricultural employment. In 2003, the LFP rate was 27.5 percent for females versus 90.0 percent for males. Female LFP has steadily increased, reaching around 37 percent in 2010 but leveling off since then. Jobs in the ready-made garment (RMG) sector in urban areas and agriculture in rural areas have absorbed an increasing number of women, contributing to the increase in female LFP over time. Despite this progress, the female LFP rate is still substantially lower than that for males (85 percent in 2016). Female LFP in Bangladesh remains below both the lower-middle income country average (39 percent) and the middle-income country average (48 percent). Moreover, progress on female LFP has reversed recently among urban females, with their LFP rate declining from 36 percent in 2010 to 32 percent in 2016. Women also have less access to better quality jobs than men, with 39 percent of working women in unpaid work, as opposed to 5 percent of working men. A larger share of working females is employed in agriculture than of men, and women have a much smaller presence in services. Labor market outcomes for youth reflect a growing challenge in light of demographic pressures and the slowdown in the pace of job creation. Even though the shares of youth (age 15–29) employed in nonagricul- ture and wage employment are rising with improved educational attainment, this age cohort tends to have higher unemployment rates than older workers (age 30–64). In recent years, youth unemployment rates have spiked, while those for older workers have remained stable; this suggests that youth may be disproportionately affected by the pressures associated with the recent slowdown in job creation. School-to-work transitions for youth continue to be a challenge, as better educated youth, with probably a higher reservation wage, enter a labor market where quality jobs are scarce. Moreover, there are few programs targeted to youth to promote their employment, such as skills development after formal education, job intermediation, or entrepreneurship. Sluggish enterprise growth presents a challenge for job creation, with a large share of employment accounted for by low-growth microenterprises. Private sector growth and job creation have been modest considering Bangladesh’s robust economic growth; this is partly explained by the large number of microenterprises. Despite the prominence of large-scale manufacturing firms particularly in the RMG sector, permanent microenterprises, along with household enterprises and temporary establishments, account for 98 percent of all economic units in the country and half of all jobs. In contrast, firms with more than 500 workers—mainly in the RMG sector— account for just 0.04 percent of firms but 15 percent of all jobs. Most microenterprises exist to offer subsistence earnings in the absence of formal wage jobs, and are not positioned for growth. Thus, Bangladeshi firms tend to be micro and old, with good firms failing to grow and bad firms failing to shut down. What is missing in the private sector is growth and job creation from small and medium enterprises. Structural and spatial transformations are shaping the future context for jobs Slowdown in the export-oriented RMG sector underscores the need for continued structural shifts in the Ban- gladeshi economy. The RMG sector has been the driver of structural transformation and the dominant source of employment. However, job creation in the RMG and textiles sectors combined declined from over 300,000 jobs in 2003–10 to just 60,000 jobs annually since 2010. While other manufacturing sectors are growing rapidly to meet increasing domestic demand, large-scale, quality job creation will require greater export orientation among the non-RMG sectors. Some sectors—such as footwear, leather products, and pharmaceuticals—have improved their export performance, but Bangladesh has failed to diversify its export basket significantly and has not emerged as a major global exporter in any new sector over the last decade. Services sector employment growth has been steady but modest; and higher-value, tradable services are still small scale. vii Sustaining rapid job creation will also require addressing constraints to spatial transformation, including con- gestion in megacities and stalled emergence of secondary cities. Bangladesh’s urbanization process has been dominated by large-scale job creation in Dhaka, which has experienced significant structural change over the past decade. This structural transformation is underpinned by ongoing internal migration, which benefits pri- marily higher-skilled workers moving into Dhaka for jobs in industry. While Dhaka Division is now home to around 30 percent of Bangladesh’s population, it accounts for 45 percent of all industry jobs and 37 percent of all service jobs. This rapid growth has resulted in severe congestion costs, with lack of access to land, transport bottlenecks, and housing shortages driving investment and jobs into the still poorly served urban periphery and contributing to both demand- and supply-side constraints to productive employment. Despite these con- straints, secondary cities are not emerging as favored locations for industrial investment, and access to quality jobs is severely limited outside Dhaka. Firms in secondary cities are constrained not only by land shortages, but also by shortfalls in critical physical infrastructure such as transport and electricity, as well as gaps in health care and education service delivery, all of which make it difficult to attract skilled workers. In this context, international migration has been and likely continues to be a channel for many workers to find better quality jobs, despite high migration costs and risks to worker safety. On average, 544,000 Bangladeshis a year have migrated abroad temporarily over the past decade. In 2016, outmigration rose to 750,000, with women representing a significant share (16 percent) for the first time. The push factors for this growth in out- migration include the domestic job creation slowdown and pressure from the rapidly growing working-age population. Pull factors include strong demand from major destinations (mostly from the Middle East) and continued large wage differentials between domestic and overseas markets. However, future demand for Ban- gladeshi migrant workers is uncertain due to volatilities in the highly concentrated set of receiving economies, largely associated with oil prices, and increasing competition from other countries. Excessive concentration in employment sectors for outmigrants—construction for males and domestic work for females—underscores the importance of skills diversification. Costs for temporary labor migration for Bangladeshis are among the world’s highest, and are often the source of heavy indebtedness and overstay of migrant workers, highlighting the need for improved support for a broad-based approach to migrant worker protection. This diagnostic provides a framework to link objectives to a comprehensive policy, strategy, and actions Three interlinked objectives emerge from the analysis as critical for the Bangladesh jobs agenda: increasing the pace of job creation, raising the quality of jobs, and connecting vulnerable workers to jobs. To deliver large-scale job creation that will absorb a growing labor force, Bangladesh must diversify its manufacturing and services sectors, with a focus on increasing exports and foreign direct investment; accelerate productivity growth; and facilitate urbanization, especially in secondary cities. Raising the quality of jobs will require policies that address barriers to firm and worker productivity, expand access to worker protection, and facilitate labor mobility. It will also require translating productivity gains into higher wages. The actions to connect vulnerable workers to jobs will have to address specific needs of the targeted population—for instance, reducing barriers to female LFP, facilitating school-to-labor market transitions for youth, and lowering the costs of international migration for lower-income and peripherally located workers. Reducing constraints to labor mobility within the country and enhancing spatial integration of the domestic economy will also be critical in connecting vulnerable workers to job opportunities in the private sector. Addressing this agenda will require a comprehensive strategy with coordinated policy actions on multiple levels. Policies related to the economy’s macro fundamentals and the investment climate will be necessary to promote more trade and investment, diversify the manufacturing sector, and expand high-productivity services. For labor markets, policy actions will need to promote the quality and relevance of worker education and skills, provide services to link workers to job opportunities in both domestic and international markets, and facilitate entre- preneurship. These actions will need to be delivered while expanding worker protection and social insurance. Finally, sectoral and regional policies will need to strengthen firm capabilities and domestic supply chains and encourage innovation, while supporting the development of secondary cities and facilitating urban-rural con- nectivity. viii OVERVIEW: BANGLADESH JOBS DIAGNOSTIC THE BIG PICTURE Bangladesh has experienced robust, sustained growth for more than a decade. Between 2003 and 2016, per capita real gross domestic product (GDP) increased by 87 percent, with annual GDP growth rates averaging 6.2 percent. Moreover, growth has been stable, with rates not falling below 5 percent in any year over this period. This steady growth has coincided with sharp poverty reduction, with the international poverty head- count rate (at $1.90/day purchasing power parity [PPP] at 2011 prices) falling from 33.7 percent to below 13 percent between 2000 and 2016. Consequently, the number of extreme poor declined from more than 44 million to around 21 million over the same period. Estimates show that growth in labor income made by far the largest contribution to poverty reduction. Economic growth, along with a favorable demographic transition, has translated into substantial progress in jobs outcomes (figure 1). Between 2003 and 2016, the Bangladesh economy generated more than 1.15 million net jobs per year on average, with total employment among the working-age (age 15–64) population growing 2.4 percent annually. The total level of employment growth was above that of the working-age population, leading to an increase in the employment ratio. Moreover, employment outside of the agricultural sector grew substantially faster (3.7 percent), and wage employment grew by 5.7 percent annually, driven in particular by large-scale job creation in manufacturing, mostly in urban areas. This growth contributed to an increase in female employment of 4.4 percent annually—more than twice the rate of growth of the working-age population as a whole—bringing millions of women into the labor force and increasing female labor force par- ticipation (LFP). Along with employment growth came strong income growth, with real wages among wage employees rising 4.9 percent annually over this period. Figure 1 Summary of key labor market developments: annual growth, 2003–16 Real wages (wage employed) 4.9% Wage employment 5.7% Female employment 4.4% Nonagricultural employment 3.7% Total employment (age 15–64) 2.4% Labor force (age 15–64) 2.3% Working-age population (age 15–64) 2.1% 0 1 2 3 4 5 6 Percent Source: Bangladesh Labor Force Survey 2003, 2016. Trade integration, through exports in the ready-made garments (RMG) sector, has been a critical catalyst of job-creating industrialization and urbanization. The export-oriented RMG sector, based mostly in megacities 1 such as Dhaka and Chittagong, has grown rapidly and contributed to large-scale job creation during the 2000s. Bangladesh exports—89 percent of which come from the RMG sector—grew by more than 16 percent annually between 2003 and 2016; this was double the global average rate and contributed to a tripling of the export share of GDP over the period (figure 2a). As a result, jobs in the RMG sector expanded from 600,000 in 2003 to over 3.2 million by 2010. Driven by RMG performance, the manufacturing sector accounted for more than one-quarter of net job creation between 2003 and 2016. A large inflow of remittances from migrants has also contributed to boosting domestic demand. Over half a million Bangladeshis go abroad for work each year. In 2015, the value of remittances from international migrants was Tk 1,200 billion (approximately $15 billion), putting Bangladesh among the top 10 largest remit- tance-receiving economies in the world (figure 2b). In the period 2003–15, remittances grew by 14 percent annually.1 Consumption and investment from these remittances—which were the equivalent of more than 13 percent of GDP in 2015—are seen to have played an important role in increasing domestic demand in both urban and rural areas. Figure 2 Export growth and remittances as a share of GDP and growth, 2003–15 a. Export growth b. Remittances as a share of GDP and growth 600 9 8.3 18 8 16 500 7 14 Share of GDP (%) 6.2 Growth (%) 6 5.2 12 Index (2003 = 100) 400 5 4.5 10 4.1 4 3.4 8 300 3 6 2 4 200 1 2 0 0 Bangladesh India Pakistan Vietnam South LMICs 100 Asia 2003 2005 2007 2009 2011 2013 2015 0 Bangladesh South Asia (excl. Bangladesh) World Share of GDP Annual growth Sources: Figure 2a—United Nations Commodity Trade Statistics Database via World Integrated Trade Solution (WITS); figure 2b—World Development Indica- tors Database. Note: LMICs = lower-middle-income countries. Robust and stable growth in domestic demand—through both consumption and investment—has likely played an important role in supporting growth, structural change, and job creation. Consumption grew steadily at an average of almost 5.5 percent annually between 2003 and 2015, and import penetration remained stable at around 20 percent of demand over this period. On top of consumption growth, Bangladesh has also experi- enced strong expansion of both public and private investment, with overall investment growing 8.6 percent annually between 2003 and 2015, and the pace of public investment doubling in 2010–15 from the previous period. Job growth has been complemented by robust overall labor productivity growth (figure 3). Between 2003 and 2016, value added per worker grew by 4.25 percent annually and accounted for three-quarters of per capita growth. The contribution of productivity to overall growth was much greater in more recent years (2010–16) than earlier (2003–10). Strong productivity growth was catalyzed by structural transformation, with workers shifting from agriculture to industry and services, along with steady within-sector productivity growth. Com- pared with a group of 26 peer countries with similar levels of per capita income, Bangladesh was among the top performers in terms of productivity gains both from structural change and from within-sector productivity growth. 1 17 percent annually in taka terms. 2 Figure 3 Annual contribution to overall growth and to productivity growth a. Decomposition of growth in per capita value added b. Productivity change decomposition by major sector and intersectoral shift 6 5 5 4 4 3 Percent Percent 3 2 2 1 1 0 0 2003–16 2003–10 2010–16 2003–16 2003–10 2010–16 −1 Demographic change Participation rate Agriculture Industry Services Intersectoral shift Employment rate Productivity Source: Bangladesh Labor Force Survey 2003, 2010, 2016. Despite overall labor productivity growth, productivity levels remain low, performance in the industrial sector has been weak, and total factor productivity has lagged behind. Value added per worker in the industrial sector has grown at only half the level as for the economy overall, with manufacturing productivity being particularly weak—just 1.6 percent annually between 2003 and 2010, increasing to 2.2 percent since 2010. Productivity levels in all sectors remain extremely low by international standards, despite recent progress. Value added per worker in Bangladesh is just half the average in the South Asia region, less than 40 percent of the average across lower-middle-income countries (LMICs), and less than 10 percent of the global average. Even the rela- tively strong productivity growth achieved over the last decade is slightly below the average in LMICs and trails far behind China (9.2 percent) and India (6.3 percent). In addition, growth in value added per worker in Bangla- desh has not been matched by total factor productivity performance. Total factor productivity growth averaged just 0.68 percent between 2003 and 2015, nearly one-sixth the pace of labor productivity growth—suggesting labor productivity growth was driven by capital deepening rather than efficiency gains. The pace of job creation has slowed markedly since 2010. Annual employment growth in the working-age population fell by one-third between the 2003–10 and 2010–16 periods—from 3.1 percent to 1.8 percent (figure 4). Thus, while 2003–10 was a period where job creation was sufficiently rapid to lower unemploy- ment and increase participation, these trends were reversed in 2010–16, despite accelerated growth in the latter period. All broad sectors of the economy experienced declining job growth, most notably agriculture and industry, while maintaining the pace of output growth. In agriculture, employment actually contracted by close to 2 percent between 2010 and 2016. A similar contraction was experienced in the construction sector. Employment growth in the RMG sector—which has experienced intensifying international price competition, higher regulatory scrutiny, and increasing wage pressures—was virtually halted in recent years, despite the sector maintaining steady output growth. This development partially reflects the substitution of technology for workers: between 2006 and 2012 (the latest year for which data are available), the average capital stock per worker in the RMG sector grew by 15 percent annually, and anecdotal evidence suggests investments in tech- nology have increased substantially since then. The pace of job creation will need to accelerate to address underemployment and pervasive gaps in job quality. The creation of around 1.4 million jobs annually for the working-age population would maintain current levels of participation and employment rates over the next decade. But the creation of substantially more and better quality jobs will be needed to address the major gaps in job quality that exist in the Bangladeshi labor market. While unemployment is low, at around 4.3 percent of the labor force in 2016, another 3.2 percent of workers are officially underemployed;2 and another 15 percent of the employed—more than 8.4 million workers—are 2 The underemployed are defined as those employed working less than 35 hours per week and willing and able to work more. 3 Figure 4 Annual growth in GDP per capita and employment Total employment (age 15–64) 3.1% 2003–10 Labor force (age 15–64) 2.9% Working-age population (age 15–64) 2.5% GDP per capita 4.7% 2010–16 Total employment (age 15–64) 1.8% Labor force (age 15–64) 2.1% Working-age population (age 15–64) 1.9% GDP per capita 5.2% 0 1 2 3 4 5 6 Percent Sources: Bangladesh Labor Force Survey 2003, 2010, 2016; and national accounts. unpaid family workers. In total, around 20 percent of the labor force is thus either unpaid or underemployed. Moreover, about 80 percent of all workers are self-employed, day laborers, or household workers. Favorable demographics, as evidenced by a rapidly growing ratio between the working and nonworking age popula- tions, will offer an important boost to support growth and quality job creation over the next decade. Whether Bangladesh is able to exploit this demographic opportunity will depend on whether the economy can create a large number of jobs, whether there will be sufficient LFP, and whether the productivity of workers can be increased further. Going beyond a steady state in the labor market and addressing major jobs gaps requires raising the growth rate toward 8 percent (figure 5a). Analysis of labor market scenarios over the next decade indicates that annual growth of around 5.6 percent would be sufficient to absorb the additional labor force and maintain a steady state outcome—maintaining the average employment, unemployment, and productivity levels of the 2003–16 period. However, given faster labor productivity growth (in line with more recent trends) and less demand for labor, higher GDP growth would naturally be required to deliver the same number of jobs. Reducing unemploy- ment and increasing LFP would require growth around the 8 percent target set out in the Bangladesh National Development Plan. Indeed, delivering on improved labor market outcomes and raising aggregate growth go hand-in-hand—both require increased productivity along with higher LFP. Improving job quality—Bangladesh’s top job priority—will require large-scale job creation in the formal sector, and extending quality gains into the informal sector. Under the job growth scenarios discussed above, Bangla- desh will need rapid growth in the formal sector to see significant gains in job quality. For example, reaching a target of 35 percent of workers in wage employment by 2025 (from 22.3 percent today)3 would require the creation of 1.4 million formal sector jobs for the working-age population each year over the next decade (figure 5b). This figure is 2.5 times the number of wage jobs created in the period 2003–16 and implies that the formal sector will need to grow several times faster than the overall economy to sustain a sufficient pace of quality job creation. 3 While data on the share of wage and salaried workers is not available in many countries, available relevant comparisons for Bangladesh include Indonesia: 38.7 percent (2015); Vietnam: 39.3 percent (2015); Peru: 47.1 percent (2015); and Sri Lanka: 56.4 percent (2014). Source: World Development Indicators Database. 4 Figure 5 Job creation projections for 2015–25 under alternative scenarios and formal sector job creation requirements a. Projected annual job creation and GDP growth b. Formal sector job growth requirements for various requirements under alternative scenarios wage employment share targets (using Scenario 2) 2.5 9 2.0 10 8.0% 8.7% 7.1% 8 2.0 7 7.3% 8 2.2 1.5 5.6% 1.7 million 6 5.7% million Percent 1.5 6 Millions Millions Percent 5 1.4 1.6 1.0 4 million 1.0 million 1.0 4 1.3 3 2.6% million million 0.5 0.5 2 2 1 0.4 million 0 0 0 0 Scenario 1: Scenario 2 Scenario3 Steady state: 30% 35% 40% Steady state 22.3% Annual job creation Annual GDP growth requirement Annual job creation Annual job growth rate Note: Figure 5a assumptions: Scenario 1: 2003–15 productivity, 2015 LFP and unemployment (3.5% productivity growth, 57% LFP, 5% unemployment); Scenario 2: 2010–15 productivity; 2003–15 trend LFP, 2015 employment (4.6% productivity growth, 59.3% LFP, 5% unemployment); Scenario 3: 2010–15 productivity, increased LFP, reduced unemployment (4.6% productivity growth, 62.5% LFP, 2.5% unemployment). Figure 5b assumptions: LFP figures are taken as a share of the 15+ population. LABOR MARKET OUTCOMES Several key labor market outcomes indicating both the quantity and quality of jobs improved robustly over the past decade (table 1). While overall LFP and employment rates changed only modestly over this period, there was a steady increase in female LFP. The increases in the share of wage and salaried employment, as well as of nonagricultural employment, are in line with the structural transformation pattern discussed previously and suggest progress in job quality. Improvements in job quality are also evidenced by a steady increase in the share of formal employment (those who have a written contract with employers) among wage employees. Table 1 Summary of key labor market outcomes: working-age population (%) Share of wage Share of formal Share of LFP rate LFP rate Employment employment in employment among nonagricultural Year (all) (females) ratio total employment wage employees employment 2003 58.6 27.5 56.1 15.1 — 49.2 2006 59.0 29.0 57.2 17.9 32.6 52.9 2010 60.0 37.0 58.1 18.5 33.8 53.7 2016 60.5 37.4 57.9 22.5 34.6 58.3 Sources: Bangladesh Labor Force Survey 2003, 2006, 2010, 2016. Note: — = not available. Despite these positive trends, the current status of the labor market suggests a significant underutilization of human resources, with 40 percent of the working-age population out of the labor force—largely driven by low female LFP (figure 6). Female LFP in 2016 remained below the average for LMICs (39 percent) and middle- income countries (48 percent). Growth in female LFP has stagnated in recent years and reversed among urban females. The LFP rate for urban females was only 32 percent in 2016 compared with more than 84 percent for men, and a large majority of inactive females are neither in employment nor education. The low level of female participation compared to male is often attributed to the social and gender norms associated with early marriage, household responsibilities, and mobility constraints, among others. The weakening of these binding constraints as well as increasing availability of work opportunities for women (e.g., agriculture work for rural 5 Figure 6 Snapshot of the Bangladesh labor market, 2016 Employed 39.5 million (97%) Males 40.8 million (69%) Unemployed 1.3 million (3%) Labor force 59.5 million (60%) Employed 17.5 million (93%) Working age Females (15–64) 18.7 million (31%) 98.4 million (62%) Unemployed 1.3 million (7%) Total population 158 million Males 4.7 million (53%) Nonworking age In school 59.7 million (38%) 8.7 million (21%) Females 4.1 million (47%) Out of the labor force 38.9 million (40%) Males 2.8 million (9%) Not in school 30.2 million (79%) Females 27.4 million (91%) Source: Bangladesh Labor Force Survey 2016. women; manufacturing work for urban women) is likely associated with an overall increase in female labor market activities over time. The quality of jobs in Bangladesh remains low, although structural transformation is bringing moderate prog- ress toward better jobs. Wage and salaried, nonagricultural, and formal jobs provide higher earnings and are accessible by those with higher educational attainment. Despite ongoing structural transformation, the agricultural sector—where household labor, unpaid work, and underemployment are prevalent—is still the largest single source of jobs (figure 7). Moreover, just 23 percent of males and 20 percent of females are wage employees, with large shares of individuals in unpaid work (females) and day labor (males). A large majority of workers are informal, with less than 35 percent of wage employees having a written contract. Large gender differences persist in type of employment and thus in access to quality jobs. A dramatically higher share of working females is employed in agriculture than men (63 percent versus 32 percent), while women’s presence in services is far less pronounced (21 percent versus 45 percent). Gender disparities in job quality are also apparent in the substantially higher share of female jobs that are unpaid: 39 percent versus 5 percent for males. The shares of wage employment are similar for men and women, but nonagricultural self-employ- ment—typically nonfarm business activities—is less common for women than men, which underscores the pertinent challenge of job quality for women. Access to better quality jobs—as proxied by wage employment—for women is largely driven by young urban females working in the manufacturing sector. Among urban women under age 25, almost half work in the 6 Figure 7 Distribution of employment sector and type by sex, 2016 a. Employment sector Males Females 1% 2% Agriculture 12% 11% 1% Manufacturing 1% 5% 32% Construction 2% 14% Commerce 1% Transport and communications Financial and business-oriented services 15% 63% Community and family-oriented services 18% Others 14% 8% b. Employment type Males Females 5% 20% 23% 21% Wage employee Self-employed, nonagriculture 39% Self-employed, agriculture Day labor 10% Unpaid work 19% 32% 6% 25% Source: Bangladesh Labor Force Survey 2016. manufacturing sector, and over 60 percent work as wage employees (figure 8). By contrast, across all age groups, most women in rural areas engage either in unpaid family work or agriculture. The dramatic differ- ences in labor market activities between younger and older women in urban areas are not echoed in rural areas. Similarly, while the share of formal workers among wage employees increased significantly from 40 percent in 2006 to 52 percent in 2016 for urban females, rural females (as well as males) experienced little progress toward formality. Youth is another group that has experienced positive labor market changes over time. Compared with older adults, youth have much higher educational attainment, with relatively little variation by sex and location. The share of individuals not in education, employment, or training (NEET) among boys (age 10–16) who may engage in informal labor has substantially declined over time, reflecting increases in schooling and reduction in child labor. Nonfarm activities in rural areas have increased for male youth, while increases in wage employment in urban areas and farm activities in rural areas are notable for female youth. Overall, decreases in agriculture and increases in wage employment are clearly observed among young workers over time; the trend is less clear for older cohorts. Despite these positive developments, youth appear to be bearing the brunt of the 2010–16 slowdown in job creation. The recent increase in unemployment rates among youth, particularly for females, indicates that 7 Figure 8 Distribution of female employment type across working age: urban versus rural a. Urban b. Rural 100 100 80 80 60 60 Percent Percent 40 40 20 20 0 0 15 25 35 45 55 15 25 35 45 55 Age Age Wage employee Self-employed, agriculture Self-employed, nonagriculture Day labor Unpaid work Source: Bangladesh Labor Force Survey 2016. labor market pressures resulting from the slowdown in job creation may have disproportionately affected these groups (figure 9). School-to-work transitions for youth continue to be a challenge, as the supply of better edu- cated youth, with a higher reservation wage, enter a labor market where quality jobs are increasingly scarce. Moreover, active labor market programs used widely in other parts of the world to link young job seekers to employment—such as opportunities for skills development after formal education, employment service and intermediation, wage subsidies, or entrepreneurship programs—are limited in Bangladesh. Figure 9 Trends in unemployment rate by sex, location, and age a. Youth (age 15–29) b. Older adults (age 30–64) 12 12 10 10 8 8 Percent Percent 6 6 4 4 2 2 0 0 2003 2006 2010 2016 2003 2006 2010 2016 Urban female Rural female Urban male Rural male Source: Bangladesh Labor Force Survey 2003, 2006, 2010, 2016. Microdeterminants of the status and sector of employment suggest uneven access to quality jobs by sex and region, as well as by age groups (figure 10). The likelihood of being employed as a wage and salaried worker (in contrast to being self-employed) shows that access to better quality jobs varies largely by education level and location, with the variation far more significant for females. Similarly, the likelihood of working in agriculture (as opposed to services) shows a mirrored pattern. It is noteworthy that access to better jobs is significantly limited outside Dhaka Division, and for those without a secondary and postsecondary education in urban areas. When education and other variables are controlled, youth are disadvantaged relative to older adults in accessing better jobs. There are large differences in wages across sectors, and formality delivers a significant earnings premium. In 2016—as in 2003—earnings were highest for financial and business-oriented services, public utilities, and com- munity and family-oriented services. However, wage growth was modestly equalizing, with higher wage sectors 8 Figure 10 Determinants of the likelihood of being employed in wage and agriculture jobs a. Determinants of wage employment b. Determinants of agricultural employment Sylhet Ranpur Rajshahi Khulna Chittagong Barisal Postsecondary Secondary Primary complete Primary incomplete Urban Married Youth −30 −20 −10 0 10 20 30 40 50 60 −50 −40 −30 −20 −10 0 10 20 30 Percentage point Percentage point Female Male Source: Bangladesh Labor Force Survey 2016. Note: Multinomial logit regressions—status (wage employment, self-employment, and day labor and unpaid work) and sector (agriculture, industry, service) of employment—were examined. The relative likelihoods (in percentage points) of working in wage employment compared to self-employment and in agriculture compared to services, are presented. experiencing a lower growth rate. Between 2003 and 2016, real weekly wages increased by 3.6 percent per year. While the annual growth rate of high-productivity services sectors such as financial and other business services was 4.3, that of the lowest-earning sectors in 2003 (including agriculture, manufacturing, construc- tion, and transport and communications) was around 5 percent on average (figure 11). Evidence indicates that formality raises earnings by over 10 percent, controlling for other factors such as education and region. Earnings are positively correlated with education, with the largest effect at the postsecondary level. An observed trend, however, is that the returns to education (whether measured at each level of education or for average Figure 11 Real weekly wage in 2003 and 2016 3,500 7 3,000 6 2,500 5 2,000 4 Percent Tk 1,500 3 1,000 2 500 1 0 0 Agriculture Mining Manufacturing Public Construction Commerce Transport Financial and Community Other utilities and com- business- and family- munications oriented oriented services services 2003 2016 Annual growth Source: Bangladesh Labor Force Survey 2003, 2016. 9 years of schooling in the Mincerian results) have been declining over time for both males and females, but more rapidly for women—probably due to the increased supply of educated individuals (figure 12a). Along with this, the strong selectivity of female workers in the earlier periods became muted (figure 12b), as female premium in the returns to education disappear. This muted selectivity is likely associated with the major expansion of educa- tion opportunities over the last generation, including the virtual elimination of the previously large gender gap. In comparing birth cohorts of 1960 and 1995, the average years of schooling increased from 4 to 8 for males and from 2 to almost 8 for females. The gender wage gap among wage employees has declined steadily over time (figure 13). A small gap remains with regard to monthly wages but disappears entirely when measuring hourly wages, indicating that the gender earnings gap derives from shorter hours of work for females compared to males. The reduced gender wage gap also implies that the appealing aspect of female labor (i.e., lower wage rate) compared to the male counterpart with equivalent educational attainment is now less pertinent and indicates a more competitive environment for female workers. With the recent slowdown in manufacturing job growth, this changing environment for women may in part explain the recent reversal in female LFP for urban females. Figure 12 Trends in returns to education and gender gap in the returns a. Returns to education by gender b. Male-female gap in returns to education 160 12 20 2 140 10 1 10 120 Percentage points Percentage points 0 0 100 8 −10 Percent Percent 80 −1 6 −20 60 −30 −2 40 4 −40 20 −3 2 −50 0 1999 2003 2006 2010 2016 1999 2003 2006 2010 2016 −60 1999 2003 2006 2010 2016 −4 −20 0 Males Females Primary Lower secondary Upper secondary Postsecondary Mincerian Source: Bangladesh Labor Force Survey 1999, 2003, 2006, 2010, 2016. Note: Wages and earnings are reported only for wage employees and day laborers; thus, a subpopulation of the labor force with earnings information is used in the analysis. Figure 13 Trends in gender wage gap a. Male-female gap based on monthly earnings b. Male-female gap based on hourly earnings 70 60 60 50 50 40 Percent 40 30 Percent 30 20 20 10 10 0 0 −10 1999 2003 2006 2010 2013 2016 −20 1999 2003 2006 2010 2013 2016 Baseline model Industry of employment controlled Source: Bangladesh Labor Force Survey 1999, 2003, 2006, 2010, 2013, 2016. Note: Demographic characteristics (age and education), urban, and localities, are all controlled for in the regression analysis. 10 In the context of the more challenging domestic labor market pertaining in recent years, a growing number of workers are seeking overseas employment. On average, 544,000 Bangladeshis migrated abroad temporarily each year over the past decade. Outmigration has increased rapidly in recent years, reaching almost 760,000 in 2016, with women now representing 16 percent of all outmigrants (figure 14). International migration remains highly concentrated in a few markets, largely in the Middle East. While demand for labor is increasing, most migrants are concentrated in low-skill activities (mainly construction for males and domestic work for females). Given continued increases in the working-age population in Bangladesh (with demographic transition) and slowing rates of job growth, along with steady demand from the Middle East and large wage differentials between destination and domestic markets, international migration is likely to continue to expand in the coming years barring unforeseen external circumstances. However, the volatility of key markets, which are reliant on oil-linked economies, poses substantial risks and the diversification of markets is thus a priority. Figure 14 Annual number of outmigrants over time a. Total b. Female 1,000 140 120 800 100 Thousands 600 Thousands 80 400 60 40 200 20 0 0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Source: Bangladesh Bureau of Manpower, Employment and Training. With a large number of workers going abroad, remittances play an increasingly important role in the Bangladesh economy. At the household level, remittances can be a major source of income, and contribute significantly to consumption, savings, and investment. For remittances to have a greater impact on longer-term poverty reduc- tion and a spillover effect on the economy, investment in productive activities as opposed to consumption or purchase of nonproductive assets would be required. At present, the Survey on Investment from Remittances indicates that relatively little is indeed used for productive activities, such as establishing businesses. Many challenges need to be addressed in order to fully utilize outward migration as a key channel for employment and poverty reduction. As in any labor market, market failures associated with skills mismatches, information gaps, and job search frictions are present in the overseas labor market. In addition, issues such as high migra- tion costs, worker safety, and risks from destination market concentration can further aggravate international migration challenges and hamper job matching efficiency. In particular, migration costs—mostly for visa fees paid to intermediaries—are substantially higher for Bangladeshi workers compared to other nationals. Such high costs are often the source of heavy indebtedness and overstay of migrant workers; they also prevent low-income populations from utilizing overseas employment as a poverty reduction strategy. Policy attention is needed to mitigate the risks associated with international migration and to reduce market failures. SOURCES OF JOB CREATION Bangladesh’s enterprise and employment landscape is highly segmented. Progress in the labor market out- comes discussed above is largely due to structural transformation and urbanization. Over the past 15 years, job creation has been rapid in both the services and—especially—the industrial sectors, with the structure of employment shifting away from agriculture, even in rural areas. One result of this transition is that jobs in 11 Bangladesh are highly segmented by sector, firm type, and sex, with jobs concentrated in microenterprises (especially for males, particularly in commerce) and in large-scale manufacturing firms (especially for females, particularly in RMG). Over the period 2003–13,4 services (and especially commerce-focused) microenterprises accounted for 90 percent of all new firm entries, but just 60 percent of employment; industry accounted for 9 percent of net new firm creation and 40 percent of jobs. Despite this large growth of the industrial sector, Bangladesh’s share of employment in the industrial sector trails well behind its peers, suggesting there may be substantial scope for structural transformation to drive quality job creation. In the midst of a slowdown in RMG job creation, continued diversification of the manufacturing and services sectors—along with expansion of these sectors into global markets—is likely to be critical to sustain large- scale, high-quality, and inclusive job creation. The RMG sector has been the driver of structural transformation and the dominant source of employment, particularly for women. However, employment growth in the RMG sector has slowed dramatically (figure 15). Job creation in the RMG and textiles sectors declined from more than 300,000 jobs a year between 2003 and 2010 to just 60,000 jobs annually since then. Moreover, the balance of employment is shifting away from female labor. Other manufacturing sectors—mainly those that are domes- tically oriented, such as food processing, furniture, and metals—are growing rapidly to meet the increasing domestic demand. But sustained, large-scale, quality job creation may require a greater export orientation from the non-RMG sectors. Even though some sectors (e.g., footwear, leather products, and pharmaceuticals) have performed well, Bangladesh has failed to achieve significant export diversification and did not emerge as a major global exporter in any new sector over the last decade. Services sector employment growth has been steady, but not remarkable, with higher-value, tradable services still very small in scale. Figure 15 Employment growth in nonagricultural sectors and in manufacturing subsectors a. Nonagricultural sectors b. Manufacturing subsectors RMG and 11.0% textiles 2003 0.6 1.3 2.4 million 1.5% mil. million Other 0.8% manufacturing 8.2% 2010 0.9 3.2 million 2.6 million mil. Construction 6.0% and utilities −0.6% 2016 3.1 million 1.4 3.8 million 3.1% million Services 2.7% −2 0 2 4 6 8 10 12 0 2 4 6 8 10 Percent Number (millions) 2003–10 2010–16 RMG Textiles Other manufacturing Source: Bangladesh Labor Force Survey 2003, 2010, 2016. The structural transformation led by the RMG sector and the emergence of domestic market-oriented man- ufacturing is closely associated with rapid spatial shifts of firms. Bangladesh’s urbanization process has been dominated by large-scale job creation in Dhaka, which has experienced significant structural change over the past decade (figure 16). This change is underpinned by ongoing, large-scale internal migration, which benefits primarily higher-skilled workers moving into Dhaka for jobs in industry. While Dhaka Division is now home to around 34 percent of Bangladesh’s population, it accounts for 45 percent of all jobs in industry and 39 percent of all services jobs. Dhaka has three times more industry jobs than the next highest division (Chittagong) and almost twice as many services jobs as the next highest division (Rajshahi). Returns to migration appear to be 4 Most of the analysis in this section is derived from the Economic Census, which is available for two periods: 2001–03 and 2013. 12 Figure 16 Employment by division over time 20 15 Millions 10 5 0 Barisal Chittagong Dhaka Khulna Rajshahi Rangpur 2003 2010 2016 Source: Bangladesh Labor Force Survey 2003, 2010, 2016. high, as rural-urban migrants are over four times more likely to be in formal employment (with a written con- tract) and earn significantly more than those who stay behind in rural areas.5 Despite increasing congestion in Bangladesh’s megacities, secondary cities have not emerged as alternative locations for industrial investment: rural areas have been the main drivers of job growth outside Dhaka. The strong growth in rural employment may be partially explained by the shift of large firms into the urban periphery of Dhaka into areas that are peri-urban but classified as rural (figure 17a)—a shift driven by rapidly rising urban congestion costs. Meanwhile, substantial firm and job growth is taking place in the northwest of the country (around Bogra and Rajshahi) and, secondarily, around Khulna (figure 17b). In these cases, job growth is mostly coming from agricultural and rural micro- and small enterprises for light manufacturing. While Bangladesh’s secondary cities are less congested than its megacities, they face similar problems with lack of access to land, along with severe shortfalls in the critical infrastructure supporting industrialization (most notably transport and electricity). Limited availability of social infrastructure (health and education) remains a major constraint to attracting skilled workers. As a result, the majority of firms and jobs outside of Dhaka and Chittagong are located not in the secondary cities or their near periphery but rather in the rural hinterland. Less than half of all firms in the catchment (defined as 50 kilometers) of the six largest secondary cities in Bangladesh are located within 30 kilometers from the city center; in contrast, in Dhaka and Chittagong, 70 percent of jobs are within 30 kilometers of the city center. The firm landscape highlights the segmentation between large-scale manufacturing in Dhaka and Chittagong, and the rest of the country, where microenterprises dominate. While very large firms are typical in RMG and other export-oriented manufacturing, microenterprises are still the norm in Bangladesh. Permanent microen- terprises, household enterprises, and temporary establishments account for 98 percent of all economic units in the country. However, firms with 10 or more employees account for more than 35 percent of all jobs; and firms employing more than 500 workers—a large share of which are in the RMG sector—account for just 0.04 per- cent of firms (less than 2,500) but for 15 percent of all jobs. These enterprise dynamics raise questions over the business environment in Bangladesh. In an economy with an efficient business environment, active entry of new firms, growth of competitive firms, and exit of uncompeti- tive firms should be observed. In Bangladesh, firms are relatively old; and the share of young firms declined by more than 4 percentage points over the past decade, while the share of young firms among microenterprises fell by 9 percentage points. This situation suggests either that barriers to firm entry exist or that firms are failing to exit—or both. While there are no explicit data on firm entry patterns, data from the Economic Census for 2003–13 indicate net growth in new enterprises of around 7 percent per year; this suggests that firm entry is 5 Direct comparison between migrants and stayees does not yield accurate estimates on returns because of differences other than migra- tion between the two groups (e.g., education, motivation, information). 13 Figure 17 Location of large employers in the Dhaka metropolitan area and firm density hotspot analysis, 2013 a. Location of firms with 500+ employee in Dhaka’s rural and urban b. Hotspot analysis of firm density in Bangladesh districts Source: Economic Census 2013. Note: The hotspot analysis in figure 17b assesses the probability that the spatial distribution of firm density is random by looking at whether there are significant differences in firm concentration between each grid location and its neighborhood (i.e., the set of spatially proximate grids). If the neighborhood is significantly denser (less dense) than the study area, that grid is a hotspot (cold spot). relatively robust. But once firms enter the market, they do not appear to grow or exit at rates that would be expected based on international norms. As a result, microenterprises are not only the largest share of firms in Bangladesh (as is normal in many countries), but they also account for the largest share of employment (which is unusual). What is missing in Bangladesh—as is revealed starkly in the comparison with Vietnam (figure 18)— is firm growth and job creation from small and midsize firms, particularly among the younger cohorts of firms. Lack of enterprise growth and of productivity gains suggests the existence of barriers preventing the manufac- turing sector from maximizing its potential for job growth. The lack of growth among firms is partly explained by the large number of microenterprises, most of which exist to offer a source of subsistence earnings in the absence of formal wage jobs, and so are not necessarily intended for growth. But the fact that even nonmicro- enterprises in Bangladesh do not grow suggests problems with lack of market demand and/or inefficiencies. In fact, a nonlinear correlation between firm size and productivity is also observed, which largely mirrors the trends in firm size and employment distribution. Small firms and the very largest show higher productivity, while there is a clear negative correlation between size and productivity for firms in the remainder of the dis- tribution. These findings may indicate the possible existence of factor distortions and/or problems of technical inefficiency. While further research is needed, anecdotal evidence and previous analysis in Bangladesh suggest a number of possible explanations, including barriers to efficient market competition, lack of access to invest- ment capital, lack of access to land, business regulatory constraints, management capacity, and worker skills. 14 Figure 18 Age and size distribution of enterprises and employment: Bangladesh and Vietnam, 2013 a. Bangladesh: total employment b. Bangladesh: number of firms 5.0 1.8 4.5 1.6 4.0 Number (millions) 3.5 1.4 Number (millions) 3.0 1.2 2.5 1.0 2.0 1.5 0.8 1.0 >10 0.6 0.5 6–10 0.4 0 Firm age >10 2–5 0.2 1–9 10–19 6–10 1 0 Firm age 20–99 2–5 100+ 1–9 Firm size 10–19 20–99 1 100+ Firm size c. Vietnam: total employment d. Vietnam: number of firms 4.0 120 3.5 3.0 100 Number (millions) Number (thousands) 2.5 80 2.0 60 1.5 1.0 40 >10 0.5 >10 20 6–10 0 6–10 Firm age 1–9 0 2–5 10–19 Firm age 1–9 2–5 10–19 20–99 1 100+ 1 20–99 Firm size 100+ Firm size Source: Economic Census 2001–03 and 2013; World Bank 2017 Vietnam Jobs Diagnostic. POLICY DIRECTION A comprehensive and well-coordinated policy effort is required, aimed at improving the macro environment and investment climate, regional and sectoral policies, and the labor market and skills. These areas of priority are in line with Bangladesh’s Vision 2021 and 7th Five-Year Plan, and could cover three interlinked objectives: ll Faster, diversified job creation: to deliver large-scale job creation that will absorb a growing labor force ll Improved job quality: to increase earnings, employment stability, working conditions, and formality ll Improved access to jobs for vulnerable groups: to ensure that labor markets deliver shared prosperity for all members of Bangladeshi society A basic framework of specific policy areas that may be relevant in addressing constraints is presented in table 2. Faster, diversified job creation Regulatory reform and revision of distortionary business policies is critical in accelerating structural transfor- mation and improving competitiveness, especially for non-RMG sectors and for small and medium enterprises. For the non-RMG sectors, policy reforms that level the playing field with currently favored sectors such as RMG are needed to enhance competitiveness and attract new investments; these could include measures such as standardizing the tax and subsidy regime, and extending special customs and financing facilities to all sectors. 15 Table 2 Jobs policy framework and initial recommendations on policy domains Policy domain Key challenges Potential policy actions Objective: Faster, diversified job creation Investment climate ŸŸ RMG sector has been key in driving large- ŸŸ Level the playing field to enhance and trade scale (female-intensive) job creation, but competitiveness of other sectors beyond now export basket is overly concentrated RMG, including: and job creation in the sector is slowing rapidly; other sectors not emerging quickly –– Opening access to customs and trade enough at sufficient scale in part due to facilities such as bonded warehouse distortionary policy environment regime to all sectors ŸŸ Microenterprises are the backbone of the –– Extending facilities for raw material private sector and increasingly critical for financing across sectors and implementing the rural economy; improving earnings expedited duty-drawback payments and growth in this sector faces constraints –– Standardizing tax and subsidy regime (e.g., both from regulatory barriers and access to export subsidies) to make sector agnostic finance and increase transparency ŸŸ Enhance access of industry to quality, competitive inputs through removing trade policy distortions, e.g., removal of para- tariffs and nontariff measures/nontariff barriers ŸŸ Introduce or enforce business and investment climate reforms, e.g.: –– Implementation of National One-Stop- Service for investment –– Reforms to court system to improve contract and insolvency enforcement—e.g., out-of-court methods –– Reforms to business licensing and permitting procedures, with a particular focus on micro- and household enterprises Urbanization— ŸŸ Lack of access to land for new investments ŸŸ Accelerate implementation of special infrastructure and and expansion in all parts of the country economic zone program and address amenities (especially Dhaka) constraints to accessing quality land through institutional reforms, including merging ŸŸ Insufficient infrastructure services and poor various industrial land institutions and connectivity in periphery of Dhaka, where freeing access/packaging of government- industrial sector is shifting due to high owned land congestion costs ŸŸ Develop and implement integrated urban ŸŸ Limited investment in secondary cities with (metropolitan) development plans (land, poor access/quality of key infrastructure infrastructure, transport) across the network services (power, water/sanitation, solid of large cities (city corporations) waste management) as well as limited social services ŸŸ Strengthen the capacity of city corporations to carry out development planning and implement public investment Objective: Improved job quality Firm-level quality and ŸŸ Low levels of technology, along with poor ŸŸ Reforms to the standards regime to promote productivity management practices and skills gaps adoption of improved standards and technologies (including liberalizing market ŸŸ Infrastructure/investment climate constraints for testing and certification) to firm productivity ŸŸ Lack of knowledge, finance, and incentive to invest in productivity-enhancing technology and practices 16 Policy domain Key challenges Potential policy actions Pensions ŸŸ Limited access to pension, as the current ŸŸ Develop comprehensive and coherent system consists of (1) a poverty targeted measures for old age support, including: social pension program (Old Age Allowance) under the Ministry of Social Welfare, and –– Reform options for government retirees (2) a mandatory pension for government pension retirees (Pension for Retired Government –– Reform options for the social pension Employees and their Families) under the Ministry of Finance –– Introduce voluntary, contributory savings and pension schemes –– Introduce options for mandatory pensions for the formal sector Objective: Improved access to jobs for vulnerable groups Youth and women ŸŸ Increasing unemployment rates among ŸŸ Develop a workforce development strategy labor market youth, particularly females transitions ŸŸ Reform technical training systems to enhance ŸŸ Limited counseling, job search assistance, employability or intermediation services, as well as inadequate training opportunities ŸŸ Review/reform labor laws to minimize bias against women and ensure stronger ŸŸ Entrenched occupational segregation and enforcement (family leave, maternity lack of mobility for women in employment leave, health and safety, anti-harassment legislation) ŸŸ Educational campaigns and training to promote females and youth into nontraditional employment ŸŸ Pilot programs for female-targeted employment services International ŸŸ Migration flows from Bangladesh highly ŸŸ Expand set of countries with which there migration concentrated in a small set of economies, are memorandums of understanding and creating risks over sustainability of growth bilateral labor agreements ŸŸ Exorbitant costs often imposed by ŸŸ Publicly publish text of agreements for intermediaries transparency, following international best practice ŸŸ Rights and conditions of work are not uniformly provided or protected in receiving ŸŸ Strengthen capacity to implement countries Overseas Employment and Migrants Act 2013, Expatriates’ Welfare and Overseas ŸŸ Irregular migration flows to high-income Employment Policies 2016 destinations are rising, with potential pressure in future for repatriation ŸŸ Develop policy on assisted voluntary return and reintegration Trade and investment climate reform is also important in removing restrictive distortions. Removal of para-tariffs and nontariff measures/nontariff barriers could help firms have easier access to higher-quality and more com- petitively priced inputs. In parallel, interventions that streamline business licensing and permitting procedures will benefit the entry and growth of all firms, and targeted efforts are needed to support micro- and household enterprises for job creation—especially in the nonagricultural rural economy. Similarly, reforms to improve con- tract and insolvency enforcement in the court system will be particularly critical in supporting small and medium enterprise expansion. Better planned and faster urbanization—key to the development of secondary cities and the sustainability of Dhaka—will need strategic and coordinated investments in amenities, infrastructure, and administrative capacity. At the national level, policies could expedite the implementation of special economic zones, and facil- itate access to industrial land more broadly, by freeing up (and packaging) substantial government-owned land and implementing institutional reforms such as consolidating the administrative structures that govern indus- trial land use. A national perspective is needed to coordinate interventions across the country’s large cities and their city corporations; and to develop and implement integrated urban (metropolitan) development plans that 17 address issues of infrastructure, land, and transport systems. In parallel, it would be useful to enact policies to empower and enable local government—including the various city corporations beyond those of Dhaka and Chittagong—to carry out development planning and implement public investment. Such measures will require strengthening city corporations’ technical and institutional capacity. Improved job quality Increasing the level of formality of the labor market will contribute substantially to enhancing the quality of jobs, but this will need to be supported by interventions boosting firm productivity. Productivity growth is already challenged by infrastructure and investment climate constraints, and these constraints are amplified by poor management practices and skills gaps in labor supply. Low and slow-growing firm productivity is explained by a combination of challenges, including financing constraints and lack of knowledge. An important set of interventions to encourage productivity growth would thus include reforms to the standards regime to promote adoption of improved standards and technologies. Developing a policy to enable workers in the informal sector to have access to benefits and social insurance would enhance job quality. For instance, many workers have no access to pensions, as the current pension system—consisting of the poverty targeted social pension (Old Age Allowance) and a mandatory pension for government retirees (the Pension for Retired Government Employees and their Families)—excludes the majority of workers. As Bangladesh continues its process of demographic transition, the importance of old age support will eventually become politically and socially pressing. Thus, the development of an overarching strategy or policy that expands coverage while remaining sustainable would be needed. This strategy could include reform of government retiree pensions, reform of social pensions, introduction of various schemes (e.g., voluntary, contributory savings), and possible mandatory pensions for the private formal sector. Improved access to jobs for vulnerable groups Facilitating access to jobs for women and youth must be a high policy priority in order to have a sustainable impact on poverty reduction. Despite rising unemployment rates among youth and females, there have been limited efforts for counseling, job search assistance, or intermediation services. The training opportunities and services that are available have proven to be insufficient and/or are not linked to market demand. As a pre- requisite to deepen interventions, Bangladesh may benefit from a workforce development strategy to identify strategic areas of investment to ensure linkage of jobs and workers, including curriculum reform. A measure likely to emerge from such a strategy—and that could even be taken in parallel—is reform of the technical training center system to reduce fragmentation and improve quality of services, so that the system fulfills its goal of enhancing employability and preparing today’s youth for a rapidly changing workplace environment. Supporting greater access to jobs and greater employment mobility for women will require a comprehensive approach to address the legal as well as cultural barriers that inhibit female LFP, reinforce entrenched occupa- tional segregation, and prevent upward mobility. Among the policy options to be considered are investments in educational campaigns to raise awareness and break down gender stereotyping in occupations, direct support for female training programs, and piloting women-only employment services. In addition, reforms to existing labor laws that institutionalize bias against women should be prioritized, along with more effective enforce- ment of laws that support women’s access to safe, quality jobs. Finally, given the growing importance of overseas employment for development, there are several interventions that could facilitate more and safer temporary migration of workers. At the government-to-government level, there is scope to expand the set of countries with which there are memorandums of understanding and bilat- eral labor agreements. These agreements would reduce migration costs as well as the vulnerability of migrant workers overseas. Vulnerability can be further reduced by developing policies to facilitate return and reinte- gration—such policies may be especially important for migrant workers concentrated in economies exposed to commodity price volatility or other macroeconomic shocks such as regional conflict. The Bangladesh gov- ernment has taken several important steps in recent years through legislation (e.g., the Overseas Employment and Migrants Act 2013 and Expatriates’ Welfare and Overseas Employment Policies 2016), and efforts could be made to ensure full implementation of such legislation. 18 Address: 17 76 G S t NW, Washington, DC 20006 Website: ht tp://w w w.worldbank .org/en/topic/jobsanddevelopment Twit ter : @W BG_ Jobs Blog: ht tp://blogs.worldbank .org/jobs