Sri Lanka POVERTY ASSESSMENT Accelerating Economic Transformation Synthesis Report Sri Lanka POVERTY ASSESSMENT Accelerating Economic Transformation Synthesis Report © 2021 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. 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 construed or 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 The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Cover design: Wojciech Wolocznik, Cambridge, United Kingdom Interior design and typesetting: Piotr Ruczyński, London, United Kingdom Contents Acknowledgements   6 Background papers   7 Abbreviations   7 Executive Summary   8 1. Introduction    12 2. Recent Progress in Poverty Reduction and Shared Prosperity     15 Good progress was made in poverty reduction before COVID-19, and non-monetary well-being improved   16 Recent growth was inclusive but inequality is relatively high   18 COVID-19 led to significant welfare losses   19 3. Sri Lanka’s Poverty Reduction: What Worked, What Did Not, and Implications for the Post-COVID-19 Era    23 Context    24 What worked: Productivity growth went hand in hand with income growth   25 What did not work: Agricultural productivity was low and the performance of the social protection system weak   28 Implications for a post-COVID-19 era    29 4. Accelerating Economic Transformation for Poverty Reduction and Shared Prosperity    31 Priority 1. Increasing agricultural productivity through diversification   32 Priority 2. Addressing the constraints to accessing remunerative nonfarm jobs   36 Priority 3. Raising the quality of jobs, especially in the informal sector   38 Priority 4. Facilitating spatial transformation and strengthening inclusion   41 5. Conclusion and Policy Implications    46 References   51 Figures Figure 1 Poverty headcount rates    16 Figure 2 Real GDP growth    16 Figure 3 Change in per capita consumption between 2009/10 and 2012/13   18 Figure 4 Per capita consumption growth, bottom 40 percent vs. total population, 2009/10–2013 and 2012/13–2016   18 Figure 5 Change in per capita consumption between 2012/13 and 2016   19 Figure 6 Share of jobs lost by province   20 Figure 7 Average income loss across the income distribution   20 Figure 8 Occupations amenable to teleworking across the income distribution   21 Figure 9 Poverty rate by province, pre- and post-COVID-19   22 Figure 10 Net job creation by sector, 2013—2016   24 Figure 11 Decomposition of growth in per capita value added   25 Figure 12 Decomposition in growth per capita value added, Sri Lanka vs peer countries   26 Figure 13 Changes in poverty due to demographics and income sources   26 Figure 14 The demographic window of opportunity   27 Figure 15 Labor force participation rate by gender and education   27 Figure 16 Gini index in Sri Lanka vs peer countries   30 Figure 17 Simpson Index of Diversification and export orientation index by productivity quintile   32 Figure 18 District level variation in diversification, export orientation, productivity, share of paddy farmers and share of paddy production in national production   34 Figure 19 Crop mix by farmers’ gender   35 Figure 20 Income shares from different sources by income quintile   36 Figure 21 Income levels from different sources by income quintile   37 Figure 22 Tourist arrivals and gross tourist receipts, 2008–19   37 Figure 23 Employment in the tourism industry, 2008–19   38 Figure 24 Formal-informal wage gap   39 Figure 25 Distribution of log hourly earnings of formal and informal employees   39 Figure 26 Cumulative compensation payable as severance and gratuity upon dismissal   40 Figure 27 $3.20 poverty rate in 2016 by district   41 Figure 28 Number of $3.20 poor by district   42 Figure 29 Human Capital Index by province   43 Figure 30 Access to tap water by province   44 Table Table 1 Profiles of the poor and nonpoor   17 SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 6 Acknowledgements This poverty assessment report was written by Yeon Soo Kim (Senior Economist) and synthesizes a series of background papers prepared over the last two years. The core team for this work included Anna Luisa Paffhausen (Economist), Cristina Chiarella (Consultant), Emiko Fukase (Consultant) and Tiloka de Silva (Consultant). The background papers that form the basis of this report are listed below. The overall activity was carried out under the overall guidance of Faris H. Hadad-Zervos (Country Direc- tor for Sri Lanka, Nepal and Maldives), Idah Z. Pswarayi-Riddihough (former Country Director), Zoubida Allaoua (Regional Director, South Asia), Chiyo Kanda (Country Manager, Sri Lanka and Maldives), Tae Hyun Lee (Lead Country Economist), Andrew Dabalen (Practice Manager, Poverty and Equity) and Benu Bidani (former Practice Manager). The team received valuable inputs and guidance from a number of colleagues over the last two years (all World Bank staff or consultants unless otherwise indicated). These include peer reviewers Ambar Narayan (Lead Economist), Samuel Freije-Rodriguez (Lead Economist), Andrew Goodland (Lead Agri- culture Specialist), Nistha Sinha (Senior Economist), Kishan Abeygunawardana (Senior Economist), and Nisha Arunatilake (Director of Research, Institute of Policy Studies). Inputs were also received from Elizabeth Bulmer (Lead Economist), Seenithamby Manoharan (Senior Rural Development Specialist), Thomas Walker (Senior Economist), Shalika Subasinghe (Senior consultant), Hideki Higashi (Senior Economist), Ashini Samarasinghe (Transport consultant), Amila Dahanayake (Economist) and Hafiz Zainudeen (Associate Operations Officer, IFC). The reported benefited from consultations with the Min- istry of Labor and Ministry of Samurdhi. The team would like to thank the Government of Sri Lanka for its support and the Department of Census and Statistics (DCS) for sharing its data. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 7 Background papers “Informality, Job Quality, and Welfare in Sri Lanka” (2020). “Agricultural Productivity, Diversification, and Gender” (2021). “The Impact of COVID-19 on Livelihoods and Poverty in Sri Lanka” (2021). “Sri Lanka Poverty Update” (2021). “The Rural Nonfarm Sector and Livelihood Strategies in Sri Lanka” (2021). Abbreviations GDP gross domestic product HCI Human Capital Index HIES Household Income and Expenditure Survey PPP purchasing power parity SID Simpson Index of Diversification SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 8 Executive Summary Sri Lanka has an impressive track record of reducing poverty and sharing prosperity more broadly. Between 2012/13 and 2016 alone, the $3.20 poverty rate fell from 16.2 percent to 11 percent, continuing pro- gress from the previous decade. There was broad-based progress in welfare, with increased access to basic services such as electricity, improved housing conditions, greater asset ownership and decreased indebt- edness, particularly among poor households. However, while growth was inclusive with the average per capita consumption growth rate accelerating across the distribution, it was less pro-poor, as the growth rate of the bottom 40 of the consumption distribution remained below that of the total population. A dynamic decade that spurred post-war growth and continued with a process of economic transformation led to a productivity boost and labor reallocation from agriculture to industry and services. Sri Lanka witnessed a period of impressive dynamism in the last decade. Post-war infrastructure invest- ments significantly improved access, particularly between Colombo and the south. Tourism thrived with the number of visitors quadrupling between 2009 and 2017. Sri Lanka’s leading export industries, tea and garments, continued to perform well, while the coconut subsector benefited from a rise in global demand for coconut products. The “sharing economy” was introduced and popularized. These underly- ing currents likely helped workers become more productive and improve their earnings. The services sec- tor experienced a particularly strong boost in productivity. However, growth in output per worker came primarily from within-sector productivity gains and much less from reallocation effects. This implies that a large share of the labor that moved out of agriculture moved into industry and services sectors of low productivity. Moreover, within-sector productivity improvements have slowed down in recent years. The COVID-19 pandemic is expected to have resulted in a significant reversal of welfare gains. The growth momentum had already started to fade when the COVID-19 pandemic hit Sri Lanka and the world in early 2020. The country had just emerged from the aftermath of the Easter Sunday attacks in 2019 which curtailed tourist arrivals. Following a strict nationwide lockdown between mid-March and mid-April 2020, restrictions were gradually eased; but economic activities of firms and households con- tinued to be hampered, leading to widespread earnings losses. Sri Lanka’s GDP contracted by 3.6 percent in 2020. At the same time, poverty at $3.20 per day is expected to have increased to 11.7 percent in 2020, SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation Executive Summary 9 up from 9.2 percent in 2019. Government mitigation efforts helped absorb the labor market impact but were likely not sufficient. With vaccines expected to be rolled out, growth is likely to slowly return in 2021 and poverty could gradually fall again. Before COVID-19, poverty reduction was mainly driven by increased labor earnings from nonfarm sectors. New jobs were created in these sectors but there are concerns about their quality. Gains in nonfarm earnings were the most important factor behind recent poverty reduction. Jobs were created in construction, trade, manufacturing, and transport—sectors that tend to employ less-skilled workers. While most jobs of the new jobs are salaried, there is concern about the quality of these jobs, since many are informal, with precarious contractual arrangements, few benefits, and low remunera- tion. Informal jobs stand in stark contrast to public sector jobs that come with significantly better earn- ings and superior benefits. Outside of the public sector, formal jobs tend to be concentrated in the export sectors of tea and garments, where there have been concerns about low productivity. The impact of Sri Lanka’s demographic transition on growth and poverty reduction is already tangible. Sri Lanka has reached the advanced stage of demographic transition relatively early and the population is ag- ing. Its demographic window of opportunity opened in 1995 and is expected to close in 2025, sooner than for its regional peers. Demographic changes had a negative contribution to growth and a smaller contribution to poverty reduction compared to regional peers. Increasing labor force participation, especially among fe- males, and enhancing the productive capacity among workers could help counter these demographic trends. Progress in improving agricultural earnings was slow; and the performance of social protection continued to be weak—these trends held back further progress in economic transformation and poverty reduction. A number of factors led to the slow progress in agriculture, among them a reversal of favorable export price trends, falling output levels (particularly in 2016 when an extreme drought occurred), low growth in minimum wages, and a stagnating paddy sector, which engages many poor farmers and continues to exhibit low productivity. Weak targeting and delivery systems in the prevailing social protection scheme undermine the scheme’s ability to protect the poor against shocks and help them become more productive. The Samurdhi pro- gram’s expansion in 2015 led to large increases in benefit levels but had little impact on coverage. Less than half of the poor were beneficiaries of Samurdhi, and benefit amounts remain largely inadequate. Limited coverage and small benefits reduce the program’s ability to effectively reduce poverty and build resilience among the poor. Nevertheless, there have been recent efforts to build better delivery and SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation Executive Summary 10 targeting systems and refine graduation programs. Productive inclusion programs could help the poor- est households graduate from social assistance and develop sustainable sources of earnings. The main message of this poverty assessment is that to sustain further progress on poverty reduction and shared prosperity, Sri Lanka needs to accelerate its economic transformation to create better jobs for more people. The priorities for economic transformation are structured around the following four themes: (i) increasing agricultural productivity through diversification; (ii) addressing the constraints to accessing remuner- ative nonfarm jobs; (iii) raising the quality of jobs, especially in the informal sector; and (iv) facilitating spatial transformation and strengthening inclusion. The main conclusions and policy recommendations under each priority area are summarized below. Increasing agricultural productivity through diversification • Diversification is likely to benefit poorer farmers, many of whom are engaged in low-productivity paddy farming. Increasing the productivity of paddy and shifting toward a higher-value, export-oriented crop mix would help increase productivity and earnings. • Agricultural interventions would further benefit from a mix of programs besides the provision of fertilizer subsidies; these programs could incentivize farmers to adopt climate-smart technolo- gies, invest in better agro-logistics, or access value chains. • Promoting equitable access to farming resources for female farmers can help close the income gap. This includes facilitating access to land as well as other agricultural inputs. Addressing the constraints to accessing remunerative nonfarm jobs • Education emerged as an overall important determinant of participation in nonfarm activities. That is, greater educational attainment appears to drive households’ livelihoods choices toward more remunerative opportunities (including public sector employment), while lower education results in greater reliance on farm and elementary nonfarm activities. Education does not affect the probability of engaging in farm versus unskilled nonfarm activities, likely reflecting the skills barrier to high-paying nonfarm employment. This is relevant because diversification into low-return activities will likely not help increase incomes. • Tourism has abundant potential to support income growth in rural areas as it has job-creation potential for the less skilled and requires relatively little investment. Tourism has a long and diver- sified supply chain, as it includes many different inputs and output activities, such as small-scale agriculture, handicrafts, and transport and other services, all of which can contribute to poverty reduction. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation Executive Summary 11 Raising the quality of jobs, especially in the informal sector • Widespread informal employment is associated with inferior working conditions, limited job secu- rity, and heightened risk of poverty due to low earnings. Yet given the complex operating environ- ment for firms created by stringent labor regulations, high cost of compliance, and overlapping regulations, the benefits of formalization may be low if the constraints to accessing finance are not lifted. Reforms could focus on increasing productivity and creating jobs, by addressing the causes and consequences of informality, rather than targeting informality itself. This would also support a resilient recovery in a post-pandemic world. • Human capital can be further improved by closing the learning gaps and by investing in skills that can cater to the demands of the private sector, which can in turn help improve labor market out- comes. Education is also highlighted as a major factor that explains an overwhelming share of the wage gap between formal and informal workers. Facilitating spatial transformation and strengthening inclusion • Spatial transformation is intrinsically linked to economic transformation. The cornerstone of spa- tial transformation and inclusion lies in strengthening public service delivery, particularly in the areas of education, health care and water supply, where the biggest gaps remain. Lagging regions perform more poorly in these areas. • A strong social protection system can contribute to economic transformation. Improving the tar- geting performance of the programs and making the system more adaptive can help build resil- ience to shocks among the poor and vulnerable. 1. Introduction SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 1. Introduction 13 Sri Lanka witnessed a period of dynamism and structural changes over the last decade. Following decades of resilient growth and the end of the civil war in 2009, gross domestic product (GDP) grew at an average of 6.2 percent between 2010 and 2016. Strong performance reflected a peace dividend and significant post-war investments. The process of structural transformation, defined as the shift of eco- nomic activities from agriculture to industry and services, was at the center of the changing dynamics in the economy. Tourism thrived and became a leading foreign exchange earner as the number of visi- tors quadrupled between 2009 and 2017. Tea and garment exports, Sri Lanka’s other top tradables, con- tinued to expand. These trends helped establish Sri Lanka as a solid middle-income country with a GDP per capita of $3,852 in 2019. Growth led to job creation, with a marked shift in labor allocation away from agriculture. Against this backdrop, Sri Lanka achieved remarkable progress in reducing poverty and sharing prosperity, but the COVID-19 pandemic has resulted in significant welfare losses. According to the lat- est available data, the share of people living on less than $3.20 per day (in 2011 purchasing power parity [PPP]) fell from 16.2 percent to 11 percent between 2012/13 and 2016, continuing progress from the pre- vious decade during which poverty fell by more than half. 1 Extreme poverty measured at $1.90 per day dropped below 1 percent in 2016. However, there had already been signs of a slowdown in growth before the COVID-19 outbreak, and the pandemic dealt a significant shock to the economy; GDP was estimated to have contracted by 3.6 percent in 2020. Widespread losses in livelihoods and earnings occurred. In a context of high vulnerability, large shocks can lead to quick and significant deterioration in welfare. Projections suggest that the $3.20 poverty headcount could have increased to 11.7 percent in 2020, after having fallen to 9.2 percent in 2019. Various mitigation measures implemented by the government like- ly helped absorb the labor market impact and soften the impact on poverty but were insufficient to ful- ly offset the shock. This poverty assessment documents Sri Lanka’s latest trends in welfare and explores opportunities for and challenges to more decisive progress. It is now well acknowledged that economic growth alone does not guarantee poverty reduction. The link between growth and poverty reduction through jobs relies on the poor having access to remunerative jobs in growing sectors. In Sri Lanka, poverty reduc- tion between 2012/13 and 2016 mainly occurred through labor market improvements in nonfarm sec- tors. In a marked contrast to the preceding decade (2002-2012/13), agricultural incomes stagnated follow- ing a reversal of favorable commodity price trends, falling output and low productivity growth. Social assistance was not very effective at reducing poverty due to its weak targeting performance and inade- quate benefit levels, though the large expansion of the Samurdhi program in 2015 appears to have ben- efited some poor households in lagging regions. 1. The $3.20 poverty rate declined from 37.8 percent in 2002 to 16.2 percent in 2012/13. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 1. Introduction 14 The main message of this report is that accelerating economic transformation to provide more and bet- ter jobs will be a top priority for a resilient and sustainable recovery from the current COVID-19 cri- sis as well as for long-term poverty reduction and shared prosperity. The concept of economic trans- formation is fundamentally about moving people from low-productivity to high-productivity activities and about “changing the nature of jobs, . . . changing what people do, where they do it and how they do it” (World Bank, 2019a). This relates to how productive workers are and the impediments to becoming more productive. Structural changes and market integration are needed for people to move into high- er-productivity jobs and thereby raise their incomes and improve their welfare. In Sri Lanka, while labor reallocation from farm to nonfarm employment occurred, gains in labor productivity were mainly due to increases in within-sector productivity and much less due to reallocation effects. This suggests that workers who moved out of agriculture ended up moving into other low-productivity sectors. To meet the World Bank’s twin goals of poverty reduction and shared prosperity, productivity growth is particularly needed in the type of jobs that are accessible by the poor and bottom 40 percent of the consumption dis- tribution. This motivates the discussion of the four priorities for economic transformation: (i) increasing agricultural productivity through diversification; (ii) addressing the constraints to accessing remuner- ative nonfarm jobs; (iii) raising the quality of jobs, especially in the informal sector and (iv) facilitating spatial transformation and strengthening inclusion. These are discussed in further detail in section 3. The remainder of the report is structured as follows. Section 1 documents recent progress in poverty and shared prosperity, with a focus on the period starting in 2012/13. This section also presents an assess- ment of the distributional impact of the ongoing COVID-19 crisis using the latest available data. Section 2 examines broader demographic and sectoral changes and then summarizes what worked and did not work in promoting economic transformation and creating more and better jobs as the linkage to poverty reduction. Section 1 and section 2 draw mainly from two World Bank reports, “Sri Lanka Poverty Update” (World Bank 2021c) and “The COVID-19 Impact on Livelihoods and Poverty in Sri Lanka” (World Bank 2021d). Section 3 unpacks the broader trends and provides a more detailed exposition of the challenges to economic transformation for the four priorities listed above. This section draws mainly from the fol- lowing background papers: “Informality, Job Quality, and Welfare in Sri Lanka” (World Bank 2020b), “Agricultural Productivity, Diversification, and Gender” (World Bank 2021a) and “The Rural Nonfarm Sector and Livelihood Strategies in Sri Lanka” (World Bank 2021e). The report concludes in section 4 with a summary of the main messages and key policy recommendations. 2. Recent Progress in Poverty Reduction and Shared Prosperity SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 2. Recent Progress in Poverty Reduction and Shared Prosperity 16 Good progress was made in poverty reduction before COVID-19, and non-monetary well-being improved Poverty in Sri Lanka declined strongly between 2012/13 and 2016, continuing progress from the previ- ous decade. 2 The poverty rate based on the World Bank’s international poverty line for lower-middle-in- come countries (at $3.20 per person per day in 2011 PPP) dropped from 16.2 percent in 2012/13 to 11 percent in 2016. This was a further improvement from 2009/10, when the estimate stood at 19.9 percent (figure 1). Extreme poverty has been almost eliminated, with only 0.9 percent of the population living on less than $1.90 per person per day in 2016 (figure 1). The depth of poverty also decreased, with improvements observed among the poorest of the poor. 3 Projections indicate that the $3.20 poverty rate likely fell further to 9.1 percent in 2019. Consistent with relatively high economic growth (figure 2), poverty reduction through 2016 was mainly the result of an increase in average consumption, rather than a redistributive effect. 4 FIGURE 1 Poverty headcount rates FIGURE 2 Real GDP growth 40 10 35 8 30 6 Percent 25 4 Percent 2 20 0 15 -2 10 -4 5 0 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2002 2006/07 2009/10 2012/13 2016 2019 2020 $1.90 $3.20 Real GDP growth BAU scenario (Jan. 2020) Source: World Bank staff calculation using HIES, various years. Sources: Department of Census and Statistics (for real GDP growth); World Bank Note: The figure shows the $1.90 and $3.20 per day poverty rate 2020a (for BAU scenario) (in blue and yellow, respectively). Estimates for 2019 and 2020 are from Note: Figures for 2021 are based on projections. BAU = business as usual. microsimulation-based projections. Poverty reduction was strong in rural areas, where an overwhelming majority of the poor continue to live. The $3.20 poverty rate in rural areas declined from 17.6 percent to 11.5 percent between 2012/13 and 2016. Poverty declined in almost all districts, and the improvement was particularly steep in those where the poverty rate was initially high (above 30 percent) in 2012/13, such as Mannar, Mullaitivu, Batticaloa, and 2. The Household Income and Expenditure Survey (HIES) is the main source of data in this report. The HIES is conducted by the Department of Census and Statistics (DCS); surveys can be found on the DCS website at http://www.statistics.gov.lk/ page.asp?page=Income%20and%20Expenditure. New HIES data were collected in 2019 but have not been released yet. All esti- mates after 2016 are based on projections. 3. This is based on a decline in the poverty gap index, which measures the average shortfall of the total population from the poverty line. The index is expressed as a percentage of the poverty line and fell from 3.4 to 2.1 percent between 2012/13 and 2016. 4. This is based on a Datt-Ravallion decomposition, which quantifies the relative contribution of growth and redistribution to changes in poverty. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 2. Recent Progress in Poverty Reduction and Shared Prosperity 17 Moneragala. Significant improvements in poverty reduction were also observed in the heavily rural Uva, Southern, and North Central Provinces. The overall number of districts with poverty rates above 30 percent declined, from five districts in 2012/13 to just two districts in 2016—these are Kilinochchi and Mullaitivu, both located in the Northern Province. However, the pace of progress was much slower in the estate sec- tor where poverty remains at high levels—estimated at 25.4 percent in 2016. With over 90 percent of the poor residing in rural areas, poverty in Sri Lanka remains overwhelmingly rural, though an outdated sector classification based on administrative boundaries significantly underestimates urbanization. This makes it difficult to get a more accurate understanding of the true extent and nature of urban and rural poverty. 5 The poor are less educated and more likely to be working in agriculture than the nonpoor. Sri Lanka has traditionally had high human capital achievements, but the gap between the poor and the nonpoor remains large; for example, less than 20 percent of the nonpoor have less than primary education, compared to nearly a third of the poor. Differences in human capital attainment lead to differences in labor market outcomes: working-age adults in poor households have lower labor market attachment and are significantly more likely to be engaged in agricultural activities that are pre- dominantly of low productivity. Poor households TABLE 1 Profiles of the poor and nonpoor are also larger in size and have higher dependency   Total Nonpoor Poor ratios than others (table 1). There are no significant Demographic characteristics       differences in the gender of the household head by Household size (number of members) 4.5 4.3 5.4 poverty status, though differences based on self-re- Dependency ratio 0.7 0.7 0.9 ported headship may not reveal the full extent of Age of head 34 34.3 31.1 gender differences in welfare outcomes. Female head (%) 53 53 52 Head is married (%) 48 49 43 Agricultural households have higher poverty rates, Education but they do not make up the majority of the poor. Less than primary (%) 20.9 19.5 32.5 Households whose head is engaged in agricultur- Primary completed (%) 50.0 49.3 56.1 al wage work exhibit elevated poverty rates. While O-level completed (%) 15.3 16.1 8.4 the national $3.20 poverty rate is 11 percent, house- A-level completed (%) 11.1 12.1 2.7 holds have an average poverty rate of 11.9 percent Bachelor’s and above (%) 2.7 3.0 0.3 if the head is in agricultural self-employment and Economic activity 21.5 percent if in agricultural wage work (table 1). Working adults in household (%) 51.1 51.4 47.9 However, agricultural households as a group (based Adults in wage work (%) 62 61.2 68.7 on household head’s main activity) comprise only Adults in self-employment (%) 38 38.8 31.3 27.5 percent of the poor, while non-agricultural Adults working in agriculture (%) 23 21.4 35.5 households account for 43.2 percent. Households Adults working in non-agriculture (%) 77 78.6 64.5 with an inactive head account for 29.4 percent. Source: World Bank staff estimates based on HIES 2016. 5. See World Bank (2021c) for details. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 2. Recent Progress in Poverty Reduction and Shared Prosperity 18 Trends in non-monetary indicators of well-being are suggestive of broader progress in welfare. Access to electricity is now almost universal thanks to significant progress among the poor: for example, elec- tricity coverage in the Northern Province increased remarkably, from 67 percent to 93 percent; coverage also surpassed 97 percent in the Eastern, Uva and North Central Provinces after improving by more than 10 percentage points between 2012/13 and 2016. This is despite electricity prices not being set accord- ing to a cost-reflective price mechanism, though the implicit subsidies are overall progressive (World Bank 2020c). Ownership of assets—such as major durables (e.g., refrigerators, washing machines) and land—increased across the broader distribution. The share of mobile phone owners increased from 85 percent to 92 percent with an even larger increase among the poor. Finally, household income sources increasingly shifted toward nonfarm sources, and household debt levels improved significantly, espe- cially among the poor. Recent growth was inclusive but inequality is relatively high Growth was inclusive but less pro-poor. Per capita consumption growth of the bottom 40 percent record- ed an annualized 4.2 percent after 2012/13, but this was still below the national average of 4.7 percent. This result suggests that growth was inclusive but less pro-poor. Overall, consumption growth acceler- ated relative to the previous period of 2009/10–2012/13 (figure 3), when per capita consumption of the bottom 40 percent increased by only 1.5 percent, compared to a 3.8 percent increase for the total popu- lation over the same period (figure 4). FIGURE 3 Change in per capita consumption between FIGURE 4 Per capita consumption growth, bottom 2009/10 and 2012/13 40 percent vs. total population, 2009/10–2013 and 2012/13–2016 32 28 24 Growth rate (percent) 2009/10 – 2012/13 20 16 12 Growth rate in mean 2012/13 – 2016 8 4 0 1 2 3 4 5 0 0 10 20 30 40 50 60 70 80 90 100 Percent Percentiles of per capita household consumption Bottom 40 Total Population Source: World Bank staff estimates based on HIES 2009/10 and 2012/13. Source: Global Database of Shared Prosperity, 2020, https://datacatalog. Note: Figure shows a Growth Incidence Curve, i.e., the growth rate in per capita worldbank.org/dataset/global-database-shared-prosperity. consumption across percentiles of the same distribution. Calculations for 2009/10 do not include the Mannar, Kilinochchi, and Mullaitivu districts. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 2. Recent Progress in Poverty Reduction and Shared Prosperity 19 Inequality increased slightly between 2012/13 and FIGURE 5 Change in per capita consumption between 2016, after a marked deterioration in previous 2012/13 and 2016 years. The Gini index of inequality rose slightly, 32 28 from 38.7 to 39.3 between 2012/13 and 2016, after Growth rate, percent 24 a notable increase from 36.1 in 2009/10. The latter 20 Growth rate in mean happened as consumption grew more slowly among 16 households in the bottom of the distribution and 12 8 faster among those in the top between 2009/10 and 4 2012/13 (figure 3). In comparison, overall consump- 0 tion growth accelerated to an average of 18 percent 0 10 20 30 40 50 60 70 80 90 100 Percentiles of per capita household consumption between 2012/13 and 2016 and was also more bal- anced across the distribution. Increases were still Source: World Bank staff estimates based on HIES 2012/13 and 2016. Note: Figure shows a Growth Incidence Curve, i.e., the growth rate in per capita strongest for the top 10 percent (figure 5). consumption across percentiles of the same distribution. COVID-19 led to significant welfare losses The COVID-19 pandemic has dealt a significant shock to the Sri Lanka economy. Swift containment measures helped check the domestic transmission of COVID-19, but prevailing social-distancing meas- ures and mobility restrictions impacted almost all sectors of the economy, leading to widespread jobs and earnings losses. GDP contracted by 3.6 percent in 2020. Data at the sectoral level show that industries have been affected most severely but there are large variations across subsectors. For example, construc- tion and textile manufacturing suffered the largest shock as the sectors are more sensitive to demand shocks and the work requires physical presence. The overall impact was relatively small in the servic- es sector, but the aggregate number again masks significant heterogeneity across subsectors—trans- port, food and accommodation, and personal services experienced large output losses. Extended travel restrictions shut down most of the tourism industry through 2020, except for some domestic tourism. Meanwhile, agricultural production was largely undisrupted, partly due to government efforts to ramp up domestic production and promote import substitution, though challenges with transport and mar- keting were reported. The fishery sector suffered a significant shock. Weak external demand impacted export-oriented subsectors and their prevailing wages. Because actual data are still limited, the assessment of the distributional impact of the COVID-19 cri- sis is informed by a macro-micro simulation and complemented with findings from a rapid phone survey. New HIES data, which are used to produce poverty estimates, were collected in 2019 but have not been released yet. However, even the latest data would not provide insights into the impact of the pan- demic, since 2019 still predates the COVID-19 outbreak. For this reason, the distributional impact assess- ment relies on estimates from a simulation exercise using macroeconomic and household survey data to SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 2. Recent Progress in Poverty Reduction and Shared Prosperity 20 project the impact of macroeconomic shocks on household welfare. Preliminary results from a recent World Bank COVID-19 rapid phone survey are also referenced as appropriate. 6 The labor market shock has been unequal across the income distribution, with the poor losing the largest proportion of their earnings. Simulations suggest that job losses were concentrated in the high- ly urbanized Western Province. About 29 percent of jobs were located in the Western Province before COVID-19, but more than a third of COVID-19-related job losses are expected to have occurred there (fig- ure 6). Workers who continued to be engaged in the labor market are also likely to have suffered earnings losses. In fact, preliminary findings from the COVID-19 rapid phone survey show that among respondents engaged in the labor market prior to the pandemic, more than half suffered a labor market shock, pri- marily in the form of earnings losses (reported by more than 30 percent), while a more modest impact occurred through temporary absence and job losses. Estimates suggest that the poorest experienced the largest proportionate earnings shock while the richest suffered from smaller proportionate income losses (figure 7). The latter are more likely to be working in the services sector which suffered the small- est contraction. They also tend to have formal, secure jobs and better access to digital technology that allows them to conduct wage work or business operations remotely. FIGURE 6 Share of jobs lost by province FIGURE 7 Average income loss across the income distribution 40 8 Percentage loss in per capita income 35 30 7 25 6 Percent 20 5 15 4 10 3 5 2 0 1 WP CP SP NP EP NCP NWP UP Sab 0 0 10 20 30 40 50 60 70 80 90 100 Share of jobs before COVID-19 Share of jobs lost due to COVID-19 Income percentile Source: World Bank staff estimation using HIES 2016. Source: World Bank staff estimation using HIES 2016. Note: WP = Western Province; CP = Central Province; SP = Southern Province; NP = Northern Province; EP = Eastern Province; NCP = North Central Province; NWP = North Western Province; UP = Uva Province; Sab = Sabaragamuwa Province. 6. The World Bank conducted a rapid phone survey across eight South Asian countries. In Sri Lanka, the survey was imple- mented between September and December 2020, and primarily aimed to understand changes in the labor market among dif- ferent groups. Additional questions were included on households’ ability to meet basic needs, safety nets, and coping mech- anisms. The sample comprised over 5,000 respondents who were reached via random digit dialing. While mobile phone ownership is quite high in Sri Lanka, at 92 percent according to the 2016 Household Income and Expenditure Survey, the modality of the phone survey does not allow for stratification by geographic units. Efforts were made during implementation to ensure a balanced sample, and characteristics of the final phone survey sample are relatively aligned with those from the 2019 Labour Force Survey. Full survey results with more detailed analysis will become available in the coming months. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 2. Recent Progress in Poverty Reduction and Shared Prosperity 21 Only a small share of high-income earners work at jobs that can be done remotely and have access to digital equipment. About 27 percent of workers in Sri Lanka have jobs that could potentially be done remotely; this figure is based on the task content of occupations. While nearly half of workers in the top 20 percent could work remotely, the share is much lower in the lower end of the income distribution. FIGURE 8 Occupations amenable to teleworking However, whether the job is actually amenable to across the income distribution being performed from home also depends on access 60 to digital technology and equipment. Once owner- 50 ship of digital devices such as computers or tablets Percentage of workers 40 is taken into account, the share of potential tele- workers drops even further: very few workers in 30 the lower half of the income distribution can work 20 from home, and the share is still only about 30 per- 10 cent among the highest income earners. 7 0 0 10 20 30 40 50 60 70 80 90 100 With jobs and earnings lost, poverty increased sig- Percentiles of earning (poorer << richer) nificantly, and over 500,000 people are estimated Work from home to have fallen into poverty as a result of the cri- Work from home with digital access & ability sis. High vulnerability implies that many work- Source: World Bank 2021b. ers with low earnings and without a proper safety Note: WFH = work from home. net could quickly fall into poverty in the event of a negative shock. In Sri Lanka, the $3.20 poverty rate increased significantly, from 9.2 percent in 2019 to an estimated 11.7 percent in 2020, when the effects ofCOVID-19 were being felt. This more than revers- es progress since 2016, when the poverty rate was 11 percent, and thus implies significant welfare loss- es. Widespread informality and precarious employment arrangements suggest a high risk of displace- ment or earnings losses in the event of shocks (World Bank, 2020c). A formal unemployment insurance scheme to protect workers during spells of joblessness is not in place. Low earnings lead to a higher risk of poverty and allow workers little room to accumulate savings that they can resort to during times of crisis. Meanwhile, formal workers concentrated in export-oriented sectors such as tea and garments were not spared either, with disruptions to both supply and demand sides over the course of the pandemic. 8 The new poor—those who fell into poverty as a result of the pandemic—are more likely to be liv- ing in urban places, but the vast majority of the poor continue to live in rural areas with previously 7. The share of jobs that can be done remotely is estimated based on an established methodology. For details, see the special focus section of World Bank (2021b). 8. A tripartite agreement was reached between employers and trade unions, and facilitated by the government, which stipu- lates that employers pay 50 percent of wages during periods employees are required to stay home. This agreement could have helped reduce employment losses. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 2. Recent Progress in Poverty Reduction and Shared Prosperity 22 high levels of poverty. The new poor are more urban, slightly better educated, and less likely to be in agriculture than those who were poor before the crisis. Across provinces, the poverty rate increased the most in the Northern, Eastern, Sabaragamuwa, and Central Provinces, which were the poorest provinces pre-COVID-19 (figure 9). The districts of Kandy (in North Central Province) and Ratnapura FIGURE 9 Poverty rate by province, pre- and (in Sabaragamuwa Province), which had the larg- post-COVID-19 est number of old poor, also account for the largest 20 share of the new poor, followed by Gampaha and 18 16 Kalutara (both in Western Province). However, the 14 COVID-19 crisis did not fundamentally shift the pov- 12 Percent erty profile or the nature of poverty in Sri Lanka. 10 8 6 Mitigation measures in the form of cash and food 4 assistance likely helped absorb the pandemic-in- 2 duced labor market shock and soften the impact 0 WP CP SP NP EP NCP NWP UP Sab on poverty. To mitigate the impact of the pandem- 2019 2020 post-COVID-19 ic on the poor and vulnerable, the government Source: World Bank staff estimation using HIES 2016. implemented several welfare measures by scaling Note: WP = Western Province; CP = Central Province; SP = Southern Province; up existing schemes. More than 4.9 million fam- NP = Northern Province; EP = Eastern Province; NCP = North Central Province; NWP = North Western Province; UP = Uva Province; Sab = Sabaragamuwa ilies are reported to have received a payment of Rs Province. 10,000, administered through the Department of Samurdhi Development, during the first lockdown period. This is in addition to the individual allow- ances for elderly or disabled individuals or chronic kidney disease patients. Around 1.4 million fami- lies received relief payments of (or equivalent to) Rs 5,000 during the second COVID-19 wave. It is esti- mated that the $3.20 poverty rate fell further as a result, by another two percentage points. While the programs were very costly and had wide coverage, mitigation was likely less effective than it could have been due to weak targeting. 3. Sri Lanka’s Poverty Reduction: What Worked, What Did Not, and Implications for the Post-COVID-19 Era SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 3. Sri Lanka’s Poverty Reduction 24 Context Sri Lanka witnessed a period of impressive dynamism in the last decade. This dynamism spurred post- war growth that led to labor reallocation from agriculture to industry and services and that boosted pro- ductivity growth. While GDP growth rates decelerated somewhat from the initial high levels observed in the early 2010s, they were still relatively high through 2016. Among the factors enabling economic trans- formation, road infrastructure benefited from significant post-war investments. Improved connectiv- ity across the country increased market accessibility and thus contributed to enhanced agglomeration along the Kandy-Colombo-Galle corridor and associated growth drivers. In particular, the Southern Expressway, which opened in 2011, halved travel time from Colombo to the south where some of the most popular tourism destinations are located. Tourism thrived, with the number of visitors quadru- pling between 2009 and 2017. The “sharing economy” was introduced, and services such as Uber, PickMe, and Airbnb became popularized in urban areas. Sri Lanka’s leading export industries, tea and garments, continued to perform well. The coconut subsector experienced an export boom thanks to a rise in global demand for coconut products. Finally, Sri Lanka’s flagship social assistance program, Samurdhi, under- went a significant expansion in 2015 that effectively tripled the program budget. Structural transformation continued during this FIGURE 10 Net job creation by sector, 2013—2016 period and productivity grew strongly, helping Construction workers improve their earnings. Economic activ- Trade Manufacturing ities and employment shifted from agriculture to Transportation industry and services. Agriculture accounted for Household activities 24.8 percent of the employed in 2016, down from Education Public administration 28 percent in 2012/13. The share working in indus- Accommodation & food services try increased from 26.1 to 27.3 percent and the share Financial & insurance activities in services from 45.9 to 47.9 percent. Many jobs Other service activities Human health & social work were created in construction, trade, manufactur- Administrative & support service ing, and transport—sectors that tend to employ Information & communication poor, less-skilled workers. At the same time, the Other Professional, scientific & technical agriculture sector lost a large number of jobs (fig- Mining & quarrying ure 10). Income growth was strong during this peri- Agriculture od, with real earnings growing at an annualized -180 -160 -40 -20 0 20 40 60 80 7 percent between 2013 and 2017. Number of jobs (thousands) Source: Department of Census and Statistics, Labour Force Survey Annual Reports, 2013–16, http://www.statistics.gov.lk/LabourForce/ Applying an economic transformation lens makes StaticalInformation/AnnualReports. it possible to see what worked and what did not work to support more and better jobs and to generate the higher productivity that contributed to pov- erty reduction. Economic transformation involves shifting from low-productivity to high-productivity SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 3. Sri Lanka’s Poverty Reduction 25 activities, across and within sectors, and is typically associated with higher urbanization. On the other hand, growth generated by higher commodity prices could stimulate income growth in the short term, but such gains might not be sustainable in the longer term. To achieve poverty reduction and shared prosperity, job and productivity gains would be most needed in the sectors where the poor and bottom 40 percent are more likely to be engaged. In addition to creating an enabling environment, constraints on both the supply and demand side of labor would need to be considered to improve the quality of jobs. Strong social safety nets can support risk management and livelihood transitions. The section concludes with a discussion on the implications of these factors in a post-COVID-19 world. What worked: Productivity growth went hand in hand with income growth Gains in labor productivity led real GDP growth in the last decade. Real GDP per capita grew at an annual rate of 5.2 percent between 2002 and 2018. A large share of the growth was due to increases in labor pro- ductivity, which led to better jobs and higher wages. However, productivity growth has fallen in recent years, after enjoying a boost between 2002 and 2012. The contribution of demographic change to FIGURE 11 Decomposition of growth in per capita GDP growth was negative, reflecting the decline in value added the share of the working-age population (figure 11). 2002–18, Total = 5.2% 2012–18, Total = 3.7% 2002–18, Total = 6.1% Most of the productivity growth came from -1 0 1 2 3 4 5 6 7 increases in within-sector productivity, particular- Percentage yearly contribution to growth (percent) ly in the services; much less came from reallocation Within-sector productivity Static reallocation effects. Labor productivity increases when work- Dynamic reallocation Employment rate Participation rate Demographic change ers move from low- to high-productivity sectors or Sources: Based on World Bank Job Structure Tool and data from World when productivity levels within sectors improve. In Development Indicators database. Sri Lanka, reallocation yielded little in the way of productivity gains, as most of the movement occurred from agriculture toward sectors with low produc- tivity, such as trade. Instead, most of the productivity growth was due not to growth-promoting struc- tural change but to improvements in within-sector productivity (Diao, McMillan, and Rodrik, 2017). 9 In peer countries such as Vietnam and Bangladesh, productivity growth was comparably strong, but a signif- icant share was due to reallocation, and these countries also benefited from favorable demographic trends (figure 12). In any case, within-sector productivity growth in Sri Lanka has slowed down in recent years. 9. Diao, McMillan, and Rodrik (2017) examine the relationship between patterns of structural change and labor productivity growth within specific industries. They argue that growth accelerations were based on rapid within-sector labor productivity growth (in the case of Latin America) or growth-increasing structural change (in the case of Africa). This stands in contrast to the experience of East Asian countries, in which both components of labor productivity contributed to overall growth. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 3. Sri Lanka’s Poverty Reduction 26 FIGURE 12 Decomposition in growth per capita value Productivity growth went hand in hand with income added, Sri Lanka vs peer countries growth, particularly in nonfarm sectors, which led Malaysia poverty reduction in the last decade. Poverty reduc- Philippines tion in Sri Lanka between 2002 and 2012/13 was pre- Bangladesh Vietnam viously attributed to increases in returns to nonfarm Thailand wage work (Ceriani, Inchauste, and Olivieri 2015). The Sri Lanka change in poverty rate can be broken down into the -1 0 1 2 3 4 Annual change (percentage point) 5 6 following components that jointly determine income, Within-sector productivity Static reallocation consumption, and subsequently poverty: (i) share Dynamic reallocation Employment rate of adults in household; (ii) share of adults working; Participation rate Demographic change (iii) share of adults working in agriculture; (iv) share Source: Based on World Bank Job Structure Tool and data from World Development Indicators database. of adults working in nonagriculture; (v) income from agricultural activities; (vi) income from nonagricul- tural activities; (vii) nonlabor income, which includes benefits from the poverty alleviation program Samurdhi, remittances received, and all other nonlabor income; (viii) in-kind income; and (ix) the consumption-to-in- come ratio. 10 The breakdown of the relative contribution of these factors to poverty reduction is shown in figure 13. The shares of the individual components add up to 100 percent and correspond to the 5.2 percent- age point reduction in the $3.20 poverty rate between 2012/13 and 2016, from 16.2 percent to 11 percent. The main takeaway is that about two thirds of the reduction in poverty is accounted for by a higher share of adults working in nonfarm sectors and those workers obtaining higher labor earnings. FIGURE 13 Changes in poverty due to demographics and income sources Consumption-to-income ratio Percent adults in household Percent adults working in household Percent working in agriculture Agricultural income Percent working in nonagriculture Nonagricultural income Samurdhi income Remittances income Other nonlabor income In-kind income -80 -60 -40 -20 0 20 40 60 80 Percent Source: World Bank staff calculation using HIES 2012/13 and 2016. Note: The figure shows percentage contribution of different demographic factors and income sources to poverty reduction. A negative contribution means that the component helped reduce poverty. The sum of all bars amounts to 1. HH = household. 10. A Shapley-Shorrocks decomposition is employed to assess the role of changes in demographics, employment, public transfers, and remittances for poverty reduction. Poverty is a function of total household per capita consumption (which depends on household size and composition), household income, and a scaling factor that maps income to consumption. For details, see World Bank (2021c). SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 3. Sri Lanka’s Poverty Reduction 27 The impact of Sri Lanka’s demographic transition FIGURE 14 The demographic window of opportunity on growth and poverty reduction is already tan- Upper MICs 1995 2030 gible. Sri Lanka has reached the advanced stage of Lower MICs 2020 2060 Afghanistan 2045 2080 demographic transition relatively early and the popu- Bangladesh 2015 2045 lation is aging. Its demographic window of opportu- Bhutan 2015 2045 India 2015 2050 nity opened in 1995 and is expected to close in 2025, Maldives 2010 2040 sooner than for its regional peers (figure 14). 11 Among Nepal 2020 2050 peer countries, the contribution of demographic Pakistan 2035 2080 Sri Lanka 1995 2025 changes to growth was negative only in Sri Lanka, China 1990 2025 whereas it occupied a notable share in Bangladesh, Korea, Rep. 1990 2015 Cambodia 2025 2055 Vietnam, Malaysia and the Philippines (figure 12). Indonesia 2005 2045 In terms of poverty reduction, the combined con- Malaysia 2010 2045 tribution of demographics, measured by the share Philippines 2025 2060 Thailand 1995 2020 of adults and the share of adults in the household Vietnam 2005 2035 engaged in economic activities, was poverty-reduc- 1960 1980 2000 2020 2040 2060 2080 2100 ing but the magnitude was small (as shown by the Source: World Bank, forthcoming. relative height of the corresponding bars in figure 13). Note: MIC = middle-income country. Increasing labor force participation, especially among females, and enhancing the productive capac- ity among the workforce through better education and skills investments can help counter these demographic trends to sustain long-term growth and economic transformation. Human capital out- comes generally favor women, as they enjoy a longer life expectancy, more years of schooling, and bet- ter learning outcomes. The main constraint lies in their access to labor market opportunities: with the FIGURE 15 Labor force participation rate by gender exception of those with a degree and above, where and education there is actually a reverse gap, females have a sig- 100 80 nificantly lower labor force participation rate than 60 Percent men (figure 15). While women are not overrepre- 40 sented in the informal sector, they occupy a large 20 share of the workforce in leading export sectors 0 such as textile and tea production, where produc- Grade 5 Grade 1 G.C.E. 2 G.C.E. Degree & and below 6–10 (O level) (A level) above tivity is low. Most unpaid family workers in agri- Male Female culture are also women (Hirimuthugodage et al Source: DCS, 2019. 2014; World Bank 2021e). Note: GCE = General Certificate of Education. 11. A demographic window is opened when the share of the youth in total population falls below 30 percent and the share of the elderly remains below 15 percent; the country’s young bulge has reached working age, resulting in a labor supply effect that can lift economic growth and income levels; this effect is commonly referred to as the first demographic dividend. In addition, higher income levels can lead to a virtuous circle of higher savings and investment, an effect referred to as the sec- ond demographic dividend. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 3. Sri Lanka’s Poverty Reduction 28 The opportunity costs of being employed appear high for women compared to the average expected return. Labor force participation of women has been persistently low, at 34.5 percent in 2019 compared to 35.5 percent in 1996.  1 2 Social norms and the lack of child care and elderly care are some of the likely culprits behind this. In addition, ownership of labor-saving household durables remains broadly low: only 22 percent of Sri Lanka households own a washing machine, and almost half do not own a refriger- ator. Given the year-round warm climate in many parts of the country, the latter means that more time is spent in cooking, since prepared food cannot be easily stored. Moreover, nearly 70 percent of house- holds rely on firewood as their principal type of cooking fuel. All of these suggest a high burden on wom- en’s time. These challenges are compounded by the lack of good-quality jobs: the availability of formal, private salaried jobs is low, as contractual arrangements are precarious. Hence there may be a preference for public sector jobs, which usually come with superior remuneration and working conditions.  1 3 In fact, in recent years better-educated women were significantly more likely than men to have taken up a public sector job. At the same time, many less-skilled women still find themselves in a weak position in the labor market—for example, about 16 percent of female workers are engaged as contributing family workers. What did not work: Agricultural productivity was low and the performance of the social protection system weak Farmers have a higher propensity to be poor than nonfarmers, as productivity in agriculture is low. The contribution of the agriculture sector to growth has been declining but the sector remains an impor- tant source of livelihoods for the poor. While the production sector contributes a small share toward Sri Lanka’s total GDP (7.4 percent in 2019), the broader agriculture and food sector is significantly larger, at around 25 percent of GDP. Food and beverage manufacturing alone account for about 6 percent of GDP. Agricultural households do not make up the majority of poor, but they are more likely than others to be living in poverty, primarily because they engage in low-productivity, low-return economic activities. The slowdown in agricultural income growth marks a departure from trends in the previous decade, during which the sector benefited from favorable prices. Lack of improvements in the agriculture sec- tor meant that its contribution to poverty reduction was low between 2012/13 and 2016 (figure 13). This stands in contrast to previous analysis of poverty reduction between 2002 and 2012/13, which was driven 12. World Development Indicators. Labor force participation rate, female (% of female population ages 15+). Accessed February 10, 2021. 13. A preference for public sector jobs is widely reported, especially among better-educated youth. Public sector jobs are much better remunerated, with significantly higher earnings, better benefits, and better job security than most private sector jobs. This raises concerns in several areas: on the demand side, the private sector may struggle to attract highly skilled workers in the presence of queuing; and on the supply side, skills investments could skew toward catering to the demands of the public sector. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 3. Sri Lanka’s Poverty Reduction 29 by growth in labor income, attributed to an increase in returns to nonfarm wage workers followed by higher returns to self-employed farm workers (Ceriani, Inchauste, and Olivieri 2015). Higher commod- ity prices drove the boost in farm earnings in the preceding period, but this boost was not accompanied by broader productivity improvements. The slowdown in agricultural earnings growth was likely due to a number of adverse trends, including: (i) falling export prices; (ii) declining output levels, particularly in 2016 (likely a result of an extreme drought event in that year); (iii) stagnating agricultural minimum wages; and (iv) a low-productivity paddy sector, which engages many poor farmers. Weak targeting and delivery systems in the prevailing social protection scheme undermine its ability to protect the poor against shocks and eventually help them become more productive. Samurdhi is Sri Lanka’s flagship poverty alleviation program, but its cash assistance explains a relatively modest 20 per- cent of the overall reduction in poverty between 2012/13 and 2016 (figure 13). This result is mainly due to undercoverage and levels of leakage as high as 70 percent. The program expansion in 2015 focused on increasing benefits and had little impact on coverage. As a result, less than half of the poor are beneficiar- ies of Samurdhi, and benefit amounts are inadequate: among beneficiary households in the bottom 10 percent, benefits account for only around 12 percent of household monthly consumption. The COVID-19 crisis further exposed these weaknesses. Several welfare programs were implemented to mitigate the impact of the crisis on the poor and vulnerable, which helped cushion the labor market shock. But imple- mentation relied to a large extent on existing delivery systems, and despite wide coverage, the mitigat- ing impact on the poor was somewhat limited relative to the large resources expended on these efforts. Implications for a post-COVID-19 era The Sri Lanka economy is slowly recovering from the COVID-19 crisis and poverty is expected to start falling again in 2021. Following a sharp contraction in GDP of 16.9 percent in the second quarter of 2020, the economy started to gradually rebound in the third quarter. While the economy contracted by 3.6 percent in 2020, GDP is projected to grow at 3.4 percent in 2021. In line with this, the latest projection indicates that the $3.20 poverty rate could fall to 10.9 percent. While urban areas have been disproportionately affected by the crisis, the nature of poverty and the structural challenges impeding further progress are expected to remain broadly unchanged. The pro- file of the poor may have slightly shifted as a result of the COVID-19 crisis, but the nature of poverty and the constraints to the livelihoods of the poor have not changed fundamentally. Because of the pandemic, some sources of livelihoods for the rural poor may take time to recover: for example, while Sri Lanka has opened up for tourism again, it will be some time until the country reaches pre-crisis levels tour- ists. Thus, the lessons learned from the process of poverty reduction in the previous decades, as sum- marized above, are expected to remain broadly valid. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 3. Sri Lanka’s Poverty Reduction 30 Inequality is expected to widen, with potential consequences for the long term. In the short term, the pandemic will likely widen inequality through the impact on the labor market. As seen in figure 7, earn- ings losses were distributed unevenly across the distribution, with the richest households experiencing smaller reductions than those in the bottom of the distribution. Consistent with this, the Gini index, FIGURE 16 Gini index in Sri Lanka vs peer countries a commonly used aggregate measure of inequality, Malaysia is expected to slightly increase, from 39.3 to 39.8. Sri Lanka This is concerning given that Sri Lanka’s Gini index Bhutan was higher than its peers’ even before the pandem- Thailand Vietnam ic (figure 16). Pakistan Bangladesh Combined with pre-existing inequalities, the Maldives potential long-term impact of the COVID-19 cri- 0 5 10 15 20 25 30 35 40 45 sis on inequality through reduced social mobility Percent could be significant. An area of particular concern Source: World Bank PovcalNet. is access to education, as schools were closed for nearly a year during the pandemic. Widening disparities in educational outcomes due to lack of access to digital technology and e-learning content could leave long-lasting scars and exacerbate inequalities between urban and rural areas and different socioeconomic groups. The potential disparities are cor- roborated using computer ownership as an imperfect proxy for access to digital connectivity: according to HIES 2016, only 19.6 percent of rural households and 6.0 percent of estate households had a comput- er. Percentages were higher among urban households, though still low at around 34.8 percent. Access to education is generally considered an important determinant of social mobility, able to lift the poorest from the bottom of the ladder—and there is increasing evidence that that is the case in Sri Lanka, with more educated individuals being more productive and more likely to hold better jobs and earn more. 4. Accelerating Economic Transformation for Poverty Reduction and Shared Prosperity SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 4. Accelerating Economic Transformation 32 Accelerating economic transformation to provide more and better jobs will be a top priority for a resil- ient and sustainable recovery from the current crisis, as well as for long-term poverty reduction and shared prosperity. This transformation will entail stronger productivity growth within sectors as well as continued employment shifts to more productive sectors that are accessible by the poor and bottom 40 percent. This section is structured around four priorities for economic transformation: (i) increasing agricultural productivity through diversification; (ii) addressing the constraints to accessing remuner- ative nonfarm jobs; (iii) raising the quality of jobs, especially in the informal sector; and (iv) facilitating spatial transformation and strengthening inclusion. Priority 1. Increasing agricultural productivity through diversification Diversification is key to increasing productivity and earnings. Diversification is typically referred to as the next stage in the transition from a traditional subsistence-oriented agriculture to commercial- ly oriented, high-value-added agriculture. In Sri Lanka, most of the output is typically marketed, and pure subsistence farming is relatively rare. Diversification at this stage is likely to help many but not all farmers: as figure 17 shows, greater diversification is positively associated with productivity only up to FIGURE 17 Simpson Index of Diversification and a certain level (the middle quintile of 40—60 per- export orientation index by productivity quintile cent of the productivity distribution), after which 0.5 productivity actually increases with greater special- 0.4 ization. Among poor farmers, the lack of diversifi- 0.3 cation is partly due to widespread paddy farming, Indexes which contributes to low overall productivity lev- 0.2 els. Smallholder farms keep over 40 percent of their 0.1 rice production for their own consumption; 14 the 0 reasons for this are not immediately understood, 1 2 3 4 5 given relatively high levels of market integration, Productivity quintile the country’s small size, and frequent price controls XIND SID (that are beneficial for the consumer). Meanwhile, Source: World Bank staff calculation using HIES 2016. productivity monotonically increases with higher Note: X-axis shows quintiles by farm productivity. XIND = export orientation index; SID = Simpson Index of Diversification. export-orientation. Note: X-axis shows quintiles by farm productivity. Addressing the factors behind the low paddy productivity could contribute to food security, poverty reduction, and the sector’s overall productivity. The primary form of agriculture in Sri Lanka is crop 14. Estimate is from HIES 2016. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 4. Accelerating Economic Transformation 33 production, with about 46 percent of smallholder farmers engaged in rice cultivation. 15 The share of rice farmers remains high despite productivity levels that are significantly lower than for other crops and that have seen little growth in the past decade. Crop productivity is lowest for rice and cereals and higher for other seasonal food crops (e.g., vegetables), annual crops (e.g., yams, tobacco), and export crops (e.g., tea and rubber). Paddy farmers are more likely than other farmers to be poor and are the main beneficiar- ies of fertilizer subsidies which appear to help increase productivity. Fertilizer subsidies have been used primarily to meet food security objectives. However, they occupy a large share in the government’s agri- cultural budget, and more resources could be spent on programs that incentivize farmers to grow a high- er-value crop mix, adopt climate-smart technologies or access value chains. 16 Measures to help smallholder farmers diversify toward higher-value crops could help achieve higher agricultural productivity. Diversification toward higher-value crops and export crops is strongly asso- ciated with higher productivity, whereas paddy farming is not. This can be seen from figure 18 which shows the district-level variation in productivity (measured as the gross value of output per acre), diver- sification, export orientation, 17 share of paddy farmers in district, and share of paddy production out of total. Diversification is measured using the Simpson Index of Diversification (SID), which takes into account the share of land devoted to different crops. 18 A low index value indicates low diversification. Multivariate analysis further shows that the positive relationship between productivity and diversifi- cation holds after accounting for other factors that influence productivity, including household char- acteristics (such as household size to proxy own labor supply, household head’s education), agricultural inputs, access to land, access to finance, and mechanization. 19 Diversification strategies could be combined with value chain interventions for farmers and agribusi- nesses. For example, the strong export orientation in some areas shown in figure 18 is primarily asso- ciated with smallholder tea production, from which some female farmers have apparently benefited. However, the sector is facing difficulties, including falling productivity, labor shortages, and increased global competition from other tea-producing countries. The wages of estate workers remain at very low levels, often below Rs 20,000 per month. Sri Lanka’s diverse agro-climatic conditions hold much potential to generate higher earnings for farmers from the cultivation of a more diversified, higher-val- ue crop mix. Larger agribusinesses can attract private investment and create wage employment that can 15. Estimate using HIES 2016. 16. World Bank (2013) suggested that subsidies could distort market decisions by encouraging the cultivation of paddy and dis- incentivize movements to other types of agriculture that have more potential for value addition. 17. Export-orientation is expressed as the relative weight of domestic-oriented and export-oriented crops in a farmer’s portfo- lio. Domestic-oriented crops include paddy and most other food crops, whereas export crops include tea, rubber, and coconut, among others. 18. Livestock activities are excluded. 19. For details of the multivariate analysis, see World Bank (2021a). SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 4. Accelerating Economic Transformation 34 be accessed by low-skilled groups, including some youth and women. The boom in the coconut sector coincides with the rise in global demand for various coconut products and could be an example of suc- cessful integration into global value chains. Forging better linkages with the tourism sector also holds promise; the benefits to local development can be enhanced if food can be sourced locally and sustain- ably, and in a way that meets the demands for high-quality foods in the sector. FIGURE 18 District level variation in diversification, export orientation, productivity, share of paddy farmers and share of paddy production in national production Simpson Index of Diversification (SID) Export Orientation Index (XIND) Productivity (Rs./acre) 0.16 – 0.20 0.4 – 0.61 20,000 – 35,000 0.12 – 0.16 0.3 – 0.4 15,000 – 20,000 0.08 – 0.12 0.2 – 0.3 10,000 – 15,000 0.04 – 0.08 0.1 – 0.2 5,000 – 10,000 0 – 0.04 0 – 0.1 0 – 5,000 Share of Paddy Farmers in Total Farmers (%) Share of Paddy Production (%) 80 – 92 12 – 18 60 – 80 9 – 12 40 – 60 6–9 20 – 40 3–6 0 – 20 0–3 Source: World Bank staff estimation using HIES 2016. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 4. Accelerating Economic Transformation 35 Promoting equitable access to resources can help close the income gap between male and female farm- ers. Female farmers have higher productivity, but this is mainly attributed to their lower access to land and selection of a more profitable crop mix. Specifically, the inverse relationship between land area and productivity that is commonly found in the literature also applies to Sri Lanka—that is, farmers with less land exhibit higher average productivity. 20 As female farmers have less access to land (one acre on average, compared to two acres for male farmers), their productivity measured in output per acre is higher, but this advantage does not translate into higher total earnings because they cultivate less land. This situa- tion suggests that earnings gains can be narrowed by reducing disparities in access to land, inputs, and other resources. With regard to land ownership, besides barriers related to social norms, the land law can be discriminatory against women who opt to be governed by personal laws (Zainudeen 2016). The grant of state land in agricultural settlement schemes under the Land Development Ordinance of 1935 and its subsequent amendments continue to favor men over women because grants are generally made to the male head of the household (Ranaraja 2020). Another issue is that female farmers tend to select crop mix that is less diversified, but more heavily concentrated in export crops such as tea, whereas almost half of male farmers cultivate paddy, making them appear more diversified even though their crop mix is not productivity-enhancing (figure 19). This implies that diversification needs to be promoted selectively. FIGURE 19 Crop mix by farmers’ gender Bana Co Ban ffe Co na / ana e, P Chilies Chilies Others ffe Others ep e, Fruit / Fru Pe pe Ho pp r, B rti y s its er, dd cu ete 0.7% 0.5% 3.5% ltu 5.2% Pa Be 2.9% Ho l re 2.5% tel rtic 6.1 ultu .4% 0.3 7.3 re % % 20 % Coco 1.9 % nut s 12.2 % Onion dy 0.2% 45.4% Pad Male Coconut 16.1% Female 8.9% Vegetables Rubber 1.4% 3.9 0.7 % O .3% % % ther 14 1.8 Yam cerea Tea bb er s ls Ru .9% 30.6% .2% 0.6% Vegetables 8.4% r cer ms 0 3 eals Onion Ya Tea s Othe Source: World Bank staff estimations using HIES 2016. 20. Numerous studies have found this relationship to hold in a variety of contexts. However, the reason for this relationship is not entirely clear, and the literature has not reached a consensus so far: family labor, differences in land quality, and meas- urement error have all been explored as possible drivers. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 4. Accelerating Economic Transformation 36 Priority 2. Addressing the constraints to accessing remunerative nonfarm jobs Increasing access to well-paying nonfarm jobs, particularly in rural areas, will further help pover- ty reduction and shared prosperity. Livelihoods in rural areas have been increasingly shifting toward industry and services activities, and the growth in nonfarm jobs has occurred mainly in skilled and semi-skilled employment. 21 Nonfarm income growth was found to have driven recent poverty reduc- tion. This has important implications for the role of the rural nonfarm sector since over 90 percent of the poor live in rural areas and the majority of rural households earn their living from nonfarm sources. The nature of rural nonfarm activities is quite heterogeneous, while opportunities to diversify with- in the farm sector appear limited for the poor. The range of economic activities among rural nonfarm workers is diverse, with 17.5 percent engaged in trade-related activities, 11.4 percent in construction, 9.9 percent each in textiles/apparel and public administration, and 8 percent in transport. The distri- bution of women is slightly different from that of men, with the largest share of working women found in textile/apparel (21.2 percent), followed by trade (17.4 percent), and with a relatively high share in pub- lic administration (11.3 percent), education (12 percent), and health care (3.7 percent). This distribution occurs because women are more likely to be employed in the public sector than in the private sector. Poor households are much more likely to be engaged in wage employment in industries and sig- nificantly less likely to be in services than nonpoor households. Poor households tend to specialize in farm activities, and opportunities for off-farm wage employment are rather limited. In terms of FIGURE 20 Income shares from different sources income sources, households in the bottom quin- by income quintile tile have a high reliance on transfers, also reflect- 100 ing their lower participation in the labor market. 80 In comparison to poor households, households 60 Percent in the top quintile draw a much higher share and 40 amount of income from nonfarm employment, 20 particularly in the services sector, and from non- 0 Q1 (poorest) Q2 Q3 Q4 Q5 (richest) labor sources such as windfalls and transfers (fig- Farm self NF self services Transfers ure 20 shows income shares from various sources; Farm wage NF wage industry Windfall income NF self industry NF wage services figure 21 shows income levels from various sourc- es). Transfers in the top quintile are mainly from Source: World Bank staff calculation using HIES 2016. Note: X-axis displays household income quintiles. NF = nonfarm; pensions, whereas those in the bottom quintile are self = self-employment; wage = wage employment. 21. Skilled employment refers to jobs that are not elementary occupations as per the ISCO classification. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 4. Accelerating Economic Transformation 37 FIGURE 21 Income levels from different sources from various social assistance programs. Notably, by income quintile incomes from nonfarm self-employment (enter- 120 prises) are not necessarily higher than those from 100 wage employment, which is likely due to the prev- Rs (monthly, thousands) 80 alence of firms that are small in size and have low 60 productivity levels. 40 20 Education emerges as an overall important deter- 0 minant of participation in nonfarm activities; Q1 (poorest) Q2 Q3 Q4 Q5 (richest) moreover, there appears to be a skills barrier Farm self NF self services Transfers Farm wage NF wage industry Windfall income in moving from farm to better-paying skilled NF self industry NF wage services employment. Workers with higher education are Source: World Bank staff calculation using HIES 2016. more likely to work in skilled nonfarm sectors, Note: X-axis displays household income quintiles. NF = nonfarm; self = self-employment; wage = wage employment. while those with lower education are similarly likely to be engaged in farm work (in the form of self-employment or wage employment) or unskilled nonfarm employment. This likely reflects a skills barrier to high-paying nonfarm employment. This result echoes findings in the literature from other countries and highlights that not all nonfarm jobs are considered better alternatives to farming. Further, female workers are less likely to participate in nonfarm activities than male workers, while access to land and higher average agricultural productivity in the residing district are also associated with low- er nonfarm participation. 22 Among alternatives to farming in rural areas, the FIGURE 22 Tourist arrivals and gross tourist receipts, tourism sector experienced remarkable growth 2008–19 and became an important source of livelihoods in 2.5 5 the last decade. Tourist arrivals more than quadru- 2.0 4 Receipts (billion $) Arrivals (millions) pled after 2009 (figure 22) and employment in the 1.5 3 tourism industry exceeded 400,000 in 2019 (figure 1.0 2 23). To put this in perspective, the export-orient- ed textile/garment industry employs around half 0.5 1 a million workers. Riding this trend, the role of the 0 0 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 “sharing economy” has also become more promi- nent in Sri Lanka, as seen from the rising number Total tourist arrivals of Airbnb rental units available in key tourist des- Gross tourist receipts tinations and the wide availability of ride-sharing Source: Central Bank, Economic and Social Statistics of Sri Lanka, various years, https://www.cbsl.gov.lk/en/publications/other-publications/ services in major cities. In fact, tourism had become statistical-publications/economic-and-social-statistics-of-sri-lanka. 22. Details on this analysis can be found in World Bank (2021e). SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 4. Accelerating Economic Transformation 38 FIGURE 23 Employment in the tourism industry, the third-largest source of foreign exchange earn- 2008–19 ings before the COVID-19 outbreak and an impor- 450 tant source of livelihoods in rural areas. 400 Number (thousads) 350 300 The tourism sector has the potential to create 250 a large number of jobs, including for less-skilled 200 groups, and can further propel poverty reduction. 150 As a labor-intensive industry that requires relative- 100 50 ly less capital, tourism provides a range of different 0 employment opportunities across the skills spec- 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 trum, especially for the less-skilled groups in the Direct employment Estimated indirect employment labor force. Tourism-related activities could help Source: Central Bank, Economic and Social Statistics of Sri Lanka, various households complement their primary source of years, https://www.cbsl.gov.lk/en/publications/other-publications/ statistical-publications/economic-and-social-statistics-of-sri-lanka. livelihood, especially in rural areas. Tourism tends to thrive in places that have a warm climate, rich cultural heritage, inspiring landscapes, and abundant biodiversity—factors that are particularly appar- ent in Sri Lanka’s rural areas. Tourism has a long and diversified supply chain, as it includes many differ- ent inputs and output activities. Spending by tourists can benefit a wide range of sectors such as small- scale agriculture, handicrafts, and transport and other services, all of which can contribute to pover- ty reduction (Avilla and Kim 2019). Flexible, part-time jobs can be created in tourism, which would par- ticularly benefit women and help increase Sri Lanka’s persistently low female labor force participation. 23 Priority 3. Raising the quality of jobs, especially in the informal sector Informal employment is widespread and comes with low job security and inferior working condi- tions. About 70 percent of workers in Sri Lanka are employed informally as defined by their access to social security. Very few informal workers—less than 4 percent—have permanent contracts, even when working full-time, leaving them in a precarious and vulnerable position. Entitlement to benefits such as paid leave is extremely limited among informal workers. Time-related underemployment is less com- mon, but excessively long work hours pose a significant problem for both formal and informal workers: nearly 20 percent of workers work more than 60 hours per week. This practice is widespread despite regulations that prescribe otherwise. A particularly high incidence of long work hours is found among 23. The Ministry of Labour has initiated a discussion with stakeholders regarding the relaxation of restrictions to wom- en’s night time work, which have long been considered a constraint to women’s labor force participation, including in the tourism industry. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 4. Accelerating Economic Transformation 39 the self-employed. While employees working over 45 hours per week can receive overtime payment, it is not clear whether informal workers are entitled to the same. Informal workers are at higher risk of poverty, but even formal workers face relatively low earnings if they work in the private sector. There is a wide gap in average earnings between formal and informal workers, which has persisted over time (figure 24), and remains even when comparing workers of similar characteristics using multivariate analysis. An Oaxaca decomposition shows that education accounts for an overwhelming share of the wage gap between formal and informal workers, whether it is owing to differences in educational attainment or the returns to the same. The risk of extremely low or inad- equate pay is significantly higher for informal workers.  2 4 Open unemployment is not very common in Sri Lanka, and in the absence of a formal unemployment scheme, workers may be pushed into subsist- ence jobs with very low pay. As of 2017, 27.1 percent of informal workers earned less than the stipulated monthly minimum wage. Among formal workers, 5.8 percent fell under the same threshold. Moreover, averages mask great variation in wages among formal workers, particularly between the private sector and the public sector. The wage distribution of formal private sector workers is actually closer to that of informal workers than public sector workers (figure 25). FIGURE 24 Formal-informal wage gap FIGURE 25 Distribution of log hourly earnings of formal and informal employees 35 Monthly earnings (Rs, thousands) 1.5 30 25 1.0 20 Density 15 0.5 10 5 0 0 2 3 4 5 6 7 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 Log hourly earnings Formal Informal Formal, Private Formal, Public Informal Source: World Bank staff calculation using Department of Census and Source: World Bank staff calculation using Labour Force Survey 2017. Statistics, Labour Force Surveys, 2006–17, http://www.statistics.gov.lk/ Note: Outliers are excluded. LabourForce/StaticalInformation/AnnualReports. Note: Monthly earnings are in 2017 rupees. Stringent labor laws have encouraged informality. The high cost of compliance and complex regula- tions have been criticized for making it difficult and expensive to dismiss employees (figure 26). The level of compensation payable upon separation presents a heavy burden to employers: for example, a worker 24. Extremely low pay is defined using a cut-off of Rs 12,500 which corresponds to the national minimum wage plus a man- dated allowance. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 4. Accelerating Economic Transformation 40 with two years of service would be entitled to a separation payment equivalent to five months’ of salary. In fact, Sri Lanka has the second-highest redundancy cost in the world, at 234 weeks of salary, which is significantly higher than Singapore (12 weeks), Malaysia (96 weeks) or Vietnam (98 weeks) (World Bank 2020b). Multiple and overlapping types of coverage for workers create a complex operating environment for firms, making compliance difficult and costly. For example, regulations that govern paid leave and holidays differ depending on whether the workers are covered by the Shop and Office Employees Act (SOEA) or Wages Board Ordinance (WBO) or are employed in the public sector. As a result, the total num- ber of holidays can range from 72 days to 129 days per year. Application and coverage are not always clear. FIGURE 26 Cumulative compensation payable as severance and gratuity upon dismissal 70 60 Number of months’ salary 50 40 30 20 10 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 Employment tenure (years) Severance Gratuity Source: World Bank 2020b. Formalization does not necessarily ease other constraints such as access to credit, reducing the incen- tive to formalize. The prevalence of the informal sector can be explained by a trade-off between firms’ costs for labor and capital: that is, informal firms have higher capital costs and lower labor productivi- ty but can avoid certain labor costs associated with mandated taxes and benefits. While being informal likely precludes firms’ access to formal credit institutions, formalization itself does not guarantee eas- ier access because of high collateral requirements and other bureaucratic constraints. The Enterprise Survey for Sri Lanka (World Bank 2012) ranks access to finance as one of the most important challeng- es to conducting business, second only to “practices of the informal sector.” Commercial banks still rely on traditional lending processes, which have different approval stages and require substantial collater- al, rather than applying cash flow–based lending processes. The lengthy loan approval process, and the inability of many small and medium enterprises to provide collateral and reliable book records, make lending either unattractive or inaccessible to many creditworthy firms. This means that streamlining the business registration process or reducing registration costs will not suffice to decrease informality. Land tenancy was also cited as a major issue (de Mel, McKenzie, and Woodruff 2011). SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 4. Accelerating Economic Transformation 41 Rather than targeting informality itself, reforms could focus on increasing productivity and jobs growth by investing in human capital and building an adequate social protection system. While Sri Lanka performs well on social indicators, there are gaps in human capital achievements. Access to pre-primary education remains a challenge, and investments in socio-emotional skills are relatively low. Learning outcomes are lagging: Sri Lanka children are expected to complete 13.2 years of schooling, but their learning-adjusted years of schooling average only 8.5 years (World Bank 2019b). The skills of the workforce need to better suit the needs of the private sector. Employer surveys show that over 30 per- cent of first-time job seekers with secondary or technical education are perceived as ill-prepared for their jobs, mainly because they lack required skills or competencies. The workforce also lacks the digital skills needed to compete and become productive in a digitalized economy. Finally, in a context where social safety nets are weak, informal jobs are likely to offer a necessary survival strategy. A well-conceived mix of policies could aim to tackle the causes and consequences of informality together and thereby encour- age the creation of high-quality jobs. This effort could also support a resilient post-COVID-19 recovery, as issues related to disruptions in human capital attainment and challenges faced by new labor market entrants will need to be addressed. Priority 4. Facilitating spatial transformation and strengthening inclusion Spatial transformation is intrinsically linked to economic transformation, but spatial disparities are high across Sri Lanka. The most lagging provinces, as measured by the level of provincial GDP, are the Northern and Eastern Provinces, which com- bined contribute less than 10 percent to total GDP. FIGURE 27 $3.20 poverty rate in 2016 by district The $3.20 poverty rates are significantly higher in $3.20 poverty rate (percent) these regions, with Mullaitivu and Killinochchi in 37 the Northern Province recording poverty rates of over 30 percent. Meanwhile, districts in the North Western and Western Provinces recorded the lowest poverty rates (figure 27). The spatial dimensions of development are closely related to sectoral transfor- mation and the sources of livelihoods. Agriculture 3 accounts for more than 15 percent of provincial GDP in the Northern and Eastern Provinces, which is about double the national average. Smallholder farmers in these regions are more likely to engage in paddy farming and livestock activities than Source: World Bank staff calculation using HIES 2016. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 4. Accelerating Economic Transformation 42 farmers in other regions. Meanwhile, a large num- FIGURE 28 Number of $3.20 poor by district ber of the poor live in and around the highly rural Number of $3.20 poor 218,431 Highlands. The districts with the highest pover- ty rates are located in the Northern and Eastern Provinces, but the absolute number of poor is rel- atively small because of the sparse population in these provinces. On the other hand, the three dis- tricts of Ratnapura, Kandy, and Badulla combined 6,093 account for over a quarter of all the poor (figure 28). While poverty is predominantly rural in Sri Lanka, urban poverty is less well understood. With an urbanization rate of 18.6 percent in 2019, Sri Lanka would appear to be one of the least urbanized coun- Source: World Bank staff calculation using HIES 2016. tries in the world. 25 However, this figure rests on an outdated sector classification that relies on administrative boundaries to define urban, rural, and estate sectors and that significantly underestimates the current extent of urbanization. Evidence based on sat- ellite imagery shows that over time, people and economic activities have become increasingly concen- trated around a mass of urban agglomerations, particularly along the Kandy-Colombo-Galle corridor (Newhouse et al 2016). 26 In practice, this implies that some areas that are currently classified as rural are in fact peri-urban areas with patterns of economic activities and spatial production that resemble those in urban areas. The underestimation of urban poverty could risk overlooking some pockets of poverty that exist in large urban areas and could affect urban and rural planning more broadly. Market integration has been constrained by lower accessibility in poor regions, though connectivity— an important pillar of regional development—remains a wider issue in predominantly rural places. Road density, defined as the total length of roads per km2, is lowest in the Eastern Province (0.24), fol- lowed closely by the North Central (0.32), Northern (0.41), and Uva (0.42) Provinces. 27 Accessibility and road conditions generally decline with distance from the Western Province. Mobility is further con- strained by suboptimal road maintenance and weak public transport services. Release of surplus labor from rural areas and their shift into urban employment is a critical driver of economic transformation behind the scale and agglomeration in urban centers. 25. World Development Indicators database, “Urban Population (% of total population),” World Bank, Washington, DC (accessed February 10, 2021), https://data.worldbank.org/indicator/SP.URB.TOTL.IN.ZS. 26. Independent studies have suggested that the actual level of urbanization would be around 44 percent (for example, UN Habitat [2018]). 27. Data on road length by province are from National Transport Commission (2017). Data on land area by province are from Central Bank (2019). Roads refer mainly to paved roads and do not include rural roads. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 4. Accelerating Economic Transformation 43 Limited access to opportunities has contributed to out-migration, providing a survival strategy for some poor households, but migration prospects are uncertain in the pandemic context. While migrants originate from all parts of the country, an increasing share has been coming from the Eastern Province in recent years. This out-migration has helped reduce poverty and may be a survival strategy for some households. For example, female migrants look for opportunities overseas to finance the building of a house and to meet education and health needs of their children (ILO 2018). The migration of an adult male household member can come with an additional cost as women’s vulnerabilities are intensified by the breakdown of families. Over 40,000 migrants have reportedly returned since the onset of the pan- demic. As re-integration in the local labor market could be uncertain and most of the world is still fac- ing uncertain prospects for the recovery process, these households may be left vulnerable. Spatial transformation can be supported by better access to basic services, particularly in the areas of education, health care, and water supply, where large gaps remain. Growth can be unbalanced but development still inclusive – as per the conclusion of the World Development Report (World Bank 2009), which emphasizes equitable access to basic services as the cornerstone of spatially inclusive growth. However, there is considerable variation in subnational human capital achievements in Sri Lanka. The Southern Province had the highest Human Capital Index (HCI) value at 63.3 percent, followed by the Western Province at 61.8 percent. 28 The lowest scores were observed in the Northern (52.2 per- FIGURE 29 Human Capital Index by province cent) and Eastern (50.7) Provinces. The HCI val- HCI value ue is inversely correlated with the distance to the 63.3 Western Province (figure 29). On population health, there has been good progress on mortality-relat- ed outcomes, but malnutrition remains an acute issue, particularly in the estate sector: for exam- ple, 32 percent of estate children under the age of five were stunted in 2016. Sri Lanka has a universal 50.7 healthcare system, but public spending on health is low, around 1.5 percent of GDP in 2018. This is lower than Singapore (2.2 percent), Vietnam (2.7 percent), Malaysia (1.9 percent), and Thailand (2.9 percent). 29 Increased spending since the COVID-19 pandem- Source: World Bank staff illustration using subnational Human Capital Index ic has exerted an additional strain on the system. (HCI) data from World Bank 2019b. 28. The HCI aggregates information from five indicators: child survival, expected years of schooling, harmonized learning out- comes, adult survival and fraction not stunted. The index is expressed in percentage units. 29. Figures for public spending on health are from World Health Organization Global Health Expenditure Database “Domestic General Government Health Expenditure as % of GDP,” https://apps.who.int/nha/database/Home/Index/ en (accessed February 12, 2021). SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 4. Accelerating Economic Transformation 44 Water- and sanitation-related issues remain acute and are a broader challenge across the country. There are low levels of access to tap water; as of 2016, only 35 percent of the population and 21 percent of the poor had access. The urban-rural gap is particularly large, as almost 80 percent of the urban popula- tion have access, compared to only 27 percent of the rural population. Across provinces, access is lowest FIGURE 30 Access to tap water by province in the North Western Province (9 percent) and the Access to tap water (percent) Northern Province (11 percent), while it is highest 50.1 in the Western Province (50 percent), followed by the Southern (42 percent) and Eastern (43 percent) Provinces (figure 30). The most common source of drinking water is wells. Absolute water shortages remain a challenge for some households. Almost 11 percent of the poor did not have enough water 9.1 to drink in the previous year, and about 14 percent of them did not have enough water for bathing and washing. 30 Almost all Sri Lankans have access to adequate sanitation, but only 4 percent are con- nected to a drainage system. Relatedly, waste man- Source: World Bank staff estimation using HIES 2016. agement remains a significant issue, as almost 80 percent of Sri Lankans burn or dump their garbage. Addressing these gaps can support sustainable and green growth, while also providing opportunities for job creation. The estate sector remains marginalized, with high poverty, poor housing conditions, and less access to basic services than other sectors, and it requires more targeted interventions. Estate sector resi- dents remain one of the most marginalized group in Sri Lanka. The $3.20 per day poverty rate in the estate sector fell only gradually, from 28.3 percent in 2009/10 to 28 percent in 2012/13 and 25.4 percent in 2016. Estate sector households are behind in virtually all nonmonetary aspects of well-being: housing conditions are poor, with 62 percent living in row houses or line rooms; only 13 percent have tap water at home, and the quality of drinking water is poor, with high levels of contamination; access to education and health services is low—for example, only about half of women who gave birth received prenatal care 30. The slow progress in the water sector has been attributed to institutional complexity and low public expenditure. The National Water Supply and Drainage Board (NWSDB), a state-owned enterprise, is responsible for the provision of water sup- ply in most urban areas. Tariffs are not sufficient to recover costs. Water supply in rural areas is managed by around 4,500 community-based organizations, which suffer from sustainability issues and lack of technical capacity. Water is affordable due to low tariffs: households commit, on average, less than 1 percent of their total budget to water payments. About 20 per- cent of the urban population does not pay for water. Water subsidies accrue proportionately more among poor households as a share of their budget and are thus progressive, but richer households consume more water and benefit more from them in absolute terms (World Bank 2020c). Priorities would include increasing technical capacity and rationalizing the tariff structure to improve sustainability and support investments in the sector. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 4. Accelerating Economic Transformation 45 (compared to 69 percent among urban women). Child health outcomes continue to underperform, with higher infant mortality rates, malnutrition, and incidence of low birth weight among estate children. Despite high poverty rates, only 8 percent of estate sector residents were covered by Samurdhi. Given the overlapping and complex nature of the challenges and slow progress in the past, concerted efforts and targeted interventions could help make decisive progress. Finally, economic transformation and inclusion need to be supported by a strong social protection sys- tem. The targeting performance of social assistance programs is weak and leakages are high. Pandemic- related mitigation efforts relied mostly on existing delivery systems and further exposed some of the weaknesses of the existing system. Adaptive social protection systems can help build resilience to shocks by investing in the capacity to prepare, cope, and adapt. There have been recent efforts by the govern- ment to build better delivery and targeting systems and refine graduation programs. Productive inclu- sion programs can help the poorest households graduate from social assistance and develop sustainable sources of earnings. Further, while poverty among the elderly is currently low, likely due to high levels of cohabitation with adult children, aging trends and high levels of informality will intensify concerns about the lack of old-age security going forward. 5. Conclusion and Policy Implications SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 5. Conclusion and Policy Implications 47 Sri Lanka has an impressive track record of reducing poverty and sharing prosperity more broadly. Between 2012/13 and 2016 alone, the $3.20 poverty rate fell from 16.2 percent to 11 percent, continuing pro- gress from the previous decade. Trends in a range of non-monetary indicators of well-being, such increased access to basic services like electricity, wider asset ownership, and decreased indebtedness, are sugges- tive of broader progress, particularly among poor households. However, while growth was inclusive it was less pro-poor: the average per capita consumption growth rate accelerated across the distribution over the same period, but the growth rate of the bottom 40 percent remained below that of the total population. COVID-19 has led to significant welfare losses and continues to present a formidable challenge to Sri Lanka’s economy and people. Sri Lanka had just started to recover from the Easter Sunday attacks of 2019 when the COVID-19 pandemic hit. The shock came primarily through the labor market, as many vulnerable workers without job security and proper safety nets experienced jobs and earnings losses. Poverty is expected to have increased to 11.7 percent in 2020, though extensive mitigation efforts likely helped absorb some of the impact of the shock. The largest impact was experienced in the services sec- tor, but there was wide variation across subsectors. While the new poor—those who fell into poverty as a result of the pandemic—are more likely to be urban, the COVID-19 crisis does not fundamentally shift the nature of poverty in Sri Lanka. The potential impact of the COVID-19 crisis on inequality could be significant, particularly as the pan- demic likely widened gaps in access to education. In the short term, the labor market impact has been more severe for poor households and is likely to raise inequality. Moreover, there are still many areas where the impact of the crisis is not well understood. In addition to a widening of gaps in access to edu- cation—the result of the closure of schools for much of 2020—the pandemic could reduce social mobil- ity to the extent that better education serves as a ladder out of poverty. In fact, analysis has repeatedly pointed to education as a key determinant of higher productivity and access to better jobs. Before the COVID-19 outbreak, Sri Lanka experienced a dynamic decade that further transformed the structure of the economy and sources of livelihoods, and the process provides clues about where fur- ther economic transformation could come from. Poverty reduction in recent years was mainly driv- en by improvements in nonfarm earnings, which is consistent with trends in jobs and earnings growth. Economic transformation entails the movement of workers from low- to high-productivity sectors and could help reinforce the drivers behind progress. However, the increase in productivity in Sri Lanka came primarily from increases in within-sector productivity and much less from reallocation effects, imply- ing that workers moved out of agriculture into other sectors of low productivity. Education was high- lighted as a key correlated of being engaged in remunerative nonfarm activities and is generally closely associated with access to better jobs. Given this track record, the constraints to economic transforma- tion and poverty reduction are expected to remain broadly valid in a post-COVID-19 world, though some challenges could intensify. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 5. Conclusion and Policy Implications 48 The main message of this report is that maximizing the potential of economic transformation to cre- ate more and better jobs will contribute to sustainable poverty reduction and shared prosperity. The priorities for economic transformation are structured around the following four themes: (i) increasing agricultural productivity through diversification; (ii) addressing the constraints to accessing remuner- ative nonfarm jobs; (iii) raising the quality of jobs, especially in the informal sector; and (iv) facilitating spatial transformation and strengthening inclusion. The main conclusions and policy recommendations are summarized below. Increasing agricultural productivity through diversification • Diversification could benefit poorer farmers, many of whom are engaged in low-productivity paddy farming. Increasing the productivity of paddy and shifting toward a higher-value, export-oriented crop mix would help increase productivity and earnings. • Agricultural interventions would further benefit from a mix of programs besides the provision of fertilizer subsidies; these programs could incentivize farmers to adopt climate-smart technolo- gies, invest in better agro-logistics, or access value chains. • Promoting equitable access to farming resources for female farmers can help close the income gap. This includes facilitating access to land as well as other agricultural inputs. Addressing the constraints to accessing nonfarm jobs • Education emerged as an overall important determinant of participation in nonfarm activities. That is, greater educational attainment drives households’ livelihoods choices toward more remu- nerative opportunities (including public sector employment), whereas less education results in greater reliance on farm and elementary nonfarm activities. This is relevant because diversifica- tion into low-return activities will likely not help increase incomes. Addressing the skills barrier to high-paying nonfarm employment could contribute to better welfare among the poor. • Tourism has abundant potential to support income growth in rural areas, as it has job-creation potential for the less skilled and requires relatively little investment. Tourism has a long and diver- sified supply chain, as it includes many different inputs and output activities, such as small-scale agriculture, handicrafts, and transport and other services, all of which can contribute to poverty reduction. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 5. Conclusion and Policy Implications 49 Raising the quality of jobs, especially in the informal sector • Widespread informal employment is associated with inferior working conditions, limited job secu- rity, and heightened risk of poverty due to low earnings. Yet given the complex operating environ- ment for firms created by stringent labor regulations, high cost of compliance, and overlapping regulations, the benefits of formalization may be low if the constraints to accessing finance are not lifted. Reforms could focus on increasing productivity and creating jobs, by addressing the causes and consequences of informality, rather than targeting informality itself. This would also support a resilient recovery in a post-pandemic world. • Human capital can be further improved by closing the learnings gaps, improving access to higher education, and by investing in skills that can cater to the demands of the private sector, which can in turn help improve labor market outcomes. Education is also highlighted as a major factor that explains an overwhelming share of the wage gap between formal and informal workers. Facilitating spatial transformation and strengthening inclusion • Spatial transformation is intrinsically linked to economic transformation. The cornerstone of spatial transformation and inclusion lies in strengthening public service delivery, particularly in education, health care and water supply, where the biggest gaps remain. Lagging regions perform more poorly in these areas. • A strong social protection system can contribute to economic transformation. Improving the tar- geting performance of the programs and making the system more adaptive can help build resil- ience to shocks among the poor and vulnerable. Sri Lanka faces a challenging road ahead with the country and the world still in the midst of the COVID-19 pandemic. Containing the health crisis remains a prerequisite for returning to normalcy, and as uncertainties abound the path to recovery could be a long one. Sri Lanka’s economy is expected to gradually recover, with a projected GDP growth rate of 3.4 percent in 2021. The share of people living on less than $3.20 per day is expected to decline accordingly to 10.9 percent. While the newly poor in Sri Lanka are more urban and slightly more educated than the old poor, COVID-19 did not fundamentally change the nature of poverty in the country. Restoring lost livelihoods and creating new ones will be critical for a resilient and sustainable recovery, which can be facilitated by the priorities to accelerate economic transformation. While the new poor are more urban and slightly more educated, the nature of poverty did not fundamentally change since COVID-19. An immediate priority for resilient recovery will likely be to bring back jobs and livelihoods. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 5. Conclusion and Policy Implications 50 Policy measures could aim to strike a balance between those that support the recovery and those that aim to include the most vulnerable in the recovery process. The latter could be supported by efforts to reverse the impact of the pandemic and mitigate its consequences for long-run inequality. The crisis is also occurring at a time when several other mega trends are emerging in Sri Lanka—these include the demographic transition to an aging society and anticipated changes to the nature of work brought about by digital technology. These underlying trends reinforce the imperative for establishing conditions to create more and better jobs and thus contribute to sustainable poverty reduction and shared prosperity. SRI LANKA POVERTY ASSESSMENT: Accelerating Economic Transformation 51 References Avilla, Montserrat and Yeon Soo Kim. 2019. “Tourism’s World Bank. 2009. World Development Report: Reshaping Potential to Create Jobs and Reduce Poverty: Case Studies Economic Geography. Washington, DC: World Bank. and Lessons Learned.” Unpublished working paper. World Bank. 2012. “Sri Lanka–Enterprise Survey 2011.” Central Bank. 2019. Economic and Social Statistics of Sri https://datacatalog.worldbank.org/dataset/sri-lank Lanka. Colombo: Central Bank of Sri Lanka. https://www. a-enterprise-survey-2011. cbsl.gov.lk/sites/default/files/cbslweb_documents/statis- World Bank. 2013. “What Is the Cost of a Bowl of Rice? The tics/otherpub/ess_2019_e.pdf. Impact of Sri Lanka’s Current Trade and Price Policies Ceriani, Lidia, Gabriela Inchauste, and Sergio Olivieri. 2015. on the Incentive Framework for Agriculture.” Agriculture “Understanding Poverty Reduction in Sri Lanka: Evidence and Environmental Services Discussion Paper 2, World from 2002 to 2012/13.” Policy Research Working Paper Bank, Washington, DC. 7446, World Bank, Washington, DC. World Bank. 2019a. “IDA Special Theme: Jobs and Economic DCS (Department of Census and Statistics). 2017. “Labour Transformation.” http://documents1.worldbank.org/ Force Survey Annual Report—2017.” http://www.sta- curated/en/381791564075012778/pdf/Special-Them tistics.gov.lk/Resource/en/LabourForce/Annual_ e-Jobs-and-Economic-Transformation.pdf. Reports/2017-2ndVersion.pdf. World Bank. 2019b. Sri Lanka Human Capital Development: DCS (Department of Census and Statistics). 2019. “Sri Lanka Realizing the Promise and Potential of Human Capital. Labour Force Survey Annual Report—2018.” http:// Washington, DC: World Bank. www.statistics.gov.lk/LabourForce/StaticalInformation/ World Bank. 2020a. Global Economic Prospects, January 2020: AnnualReports/2018. Slow Growth, Policy Challenges. Washington, DC: World De Mel, Suresh, David McKenzie, and Christopher Woodruff. Bank. DOI: 10.1596/978-1-4648-1468-6. 2011. “Getting Credit to High Return Microentrepreneurs: World Bank. 2020b. “Informality, Job Quality, and Welfare The Results of an Information Intervention.” World Bank in Sri Lanka.” Background paper to Sri Lanka Poverty Economic Review 25, no. 3: 456–85. Assessment, World Bank, Washington, DC. Diao, Xinshen, Margaret McMillan, and Dani Rodrik. 2017. World Bank. 2020c. “Integrated State-Owned Enterprises “The Recent Growth Boom in Developing Countries: Framework: Sri Lanka Assessment.” World Bank. A Structural Change Perspective.” NBER Working Washington, DC. Paper 23132, National Bureau for Economic Research, World Bank. 2021a. “Agricultural Productivity, Cambridge, MA. Diversification, and Gender.” Background paper to Sri Hirimuthugodage, D., Madurawala, S., Senaratne, A., Lanka Poverty Assessment, World Bank, Washington, DC. Wijesinha, A. and Edirisinghe, C. 2014. Female entrepre- World Bank. 2021b. “Sri Lanka Development Update neurship and the role of business development services in pro- Spring 2021.” World Bank, Washington, DC. https:// moting small and medium women entrepreneurship in Sri thedocs.worldbank.org/en/doc/15b8de0edd4f39c- Lanka. Institute of Policy Studies of Sri Lanka and Oxfam c7a82b7aff8430576-0310062021/original/ International, Sri Lanka. SriLanka-DevUpd-Apr9.pdf. ILO (International Labour Organization). 2018. “Sri Lankan World Bank. 2021c. “Sri Lanka Poverty Update.” Background Female Migrant Workers and the Family Background paper to Sri Lanka Poverty Assessment, World Bank, Report.” International Labour Organization. https:// Washington, DC. www.ilo.org/colombo/whatwedo/publications/ World Bank. 2021d. “The Impact of COVID-19 on Livelihoods WCMS_632484/lang--en/index.htm. and Poverty in Sri Lanka.” Background paper to Sri Lanka National Transport Commission. 2017. National Transport Poverty Assessment, World Bank, Washington, DC. Statistics 2017. Colombo: National Transport Commission. World Bank. 2021e. “The Rural Nonfarm Sector and Liveli- Newhouse, David Locke, Pablo Suarez Becerra, and Dung hood Strategies in Sri Lanka.” Background paper to Sri Doan. 2016. “Sri Lanka Poverty and Welfare: Recent Lanka Poverty Assessment, World Bank, Washington, DC. Progress and Remaining Challenges.” World Bank, World Bank. Forthcoming. “Sri Lanka Country Economic Washington, DC. Memorandum.” World Bank, Washington, DC. Ranaraja, Shyamali. 2020. “Assessment of Gender Equality Zainudeen, Mohamed Hafiz. 2016. “Land Administration and the Legal Framework of Sri Lanka.” Background in Sri Lanka. Issues and Challenges.” Paper prepared for report for Sri Lanka Systematic Country Diagnostic presentation at World Bank Conference on Land and Update. World Bank, Washington, DC. Poverty, Washington, DC, March 14–18, 2016.