Poverty and Shared Prosperity in Moldova: Progress and Prospects 1 Report No. 105722-MD Poverty Reduction and Shared Prosperity in Moldova: Progress and Prospects Poverty and Equity Global Practice World Bank Group May 2016 Poverty and Shared Prosperity in Moldova: Progress and Prospects Table of Contents Acknowledgments.......................................................................................................................................... viii Executive Summary...........................................................................................................................................ix Overview..............................................................................................................................................................x I. Introduction.......................................................................................................................................................1 II. Progress in poverty reduction and shared prosperity.......................................................................................3 Poverty declined in Moldova ....................................................................................................................3 ... and the country reduced inequality and boosted shared prosperity................................................5 Progress was underpinned by high upward economic mobility...........................................................7 Yet, Moldova is one of the poorest countries in Europe.........................................................................8 III. What drove poverty reduction and shared prosperity?.................................................................................10 Economic growth was volatile, but positive and pro-poor overall.....................................................10 Labor markets aided the progress mainly by nonagricultural wage increases..................................11 Public transfers, mainly pensions, drove some improvement in living standards............................16 Migration and remittances have shaped growth, poverty reduction, and shared prosperity..........17 IV. Abiding challenges and the issue of sustainability.........................................................................................21 Spatial and cross-group inequalities persist...........................................................................................21 Important and increasing risks to sustainable progress persist...........................................................30 V. Conclusion......................................................................................................................................................36 VI. References.....................................................................................................................................................38 Annex A. Characteristics of the poor and bottom 40..................................................................................40 Annex B. Nonmonetary poverty.....................................................................................................................45 iii Poverty and Shared Prosperity in Moldova: Progress and Prospects Figures Figure 1. Poverty rate and GDP per capita, latest available...................................................................x Figure 2. Welfare growth of the bottom 40 and average population, latest available........................x Figure 3. GDP growth, 2001–15...............................................................................................................1 Figure 4. Real GDP index (2007 = 100)...................................................................................................1 Figure 5. Employment rate by demographics and location, 2000–14..................................................2 Figure 6. Activity rate by demographics and location, 2000–14..........................................................2 Figure 7. Poverty headcount ratio, 2007–14............................................................................................3 Figure 8. Poverty rates, by urban and rural areas, %..............................................................................5 Figure 9. Poverty rates, by region, %........................................................................................................5 Figure 10. Poverty gap, by urban and rural areas...................................................................................5 Figure 11. Poverty gap squared, by urban and rural areas..........................................................................5 Figure 12. Real consumption per capita growth, by group...................................................................6 Figure 13. Dynamics of consumption per capita inequality.................................................................6 Figure 14. Real consumption growth, urban areas......................................................................................6 Figure 15. Real consumption growth, rural areas...................................................................................6 Figure 16. Real consumption growth, the bottom 40 and total population, by country...................7 Figure 17. Households in poverty that escaped poverty or the nonpoor who fell into poverty, %......7 Figure 18. Intragenerational mobility, by share of the population, Moldova, 2007–14.....................8 Figure 19. GDP per capita and $5.00-a-day poverty, Europe and Central Asia, latest available data......9 Figure 20. Welfare group decomposition, by country, latest available data........................................9 Figure 21. GDP growth decomposition...................................................................................................10 Figure 22. GDP and household income and consumption growth.....................................................11 Figure 23. Growth incidence curve, total population, 2007–14...........................................................11 Figure 24. Datt-Ravallion decomposition of changes in the decline in the poverty rate..................11 Figure 25. Decomposition of income growth, the bottom 40, by income source, 2007–14, %........12 Figure 26. Decomposition of changes in poverty, by income source, 2007–14, % points.................12 Figure 27. Reason for unemployment or inactivity, 15–65 age-group, 2014......................................12 Figure 28. Underemployment among the employed ages 15+, 2014...................................................12 iv Poverty and Shared Prosperity in Moldova: Progress and Prospects Figure 29. Employment status of the poor..............................................................................................13 Figure 30. Employment status of the bottom 40.....................................................................................13 Figure 31. Population distribution, manufacturing and investment, by location, 2014....................13 Figure 32. Share of employment abroad in total employment, %........................................................13 Figure 33. Employment in agriculture.....................................................................................................14 Figure 34. Employed populations, by sector...........................................................................................14 Figure 35. Employment sector of the poor..............................................................................................14 Figure 36. Employment sector of the bottom 40....................................................................................14 Figure 37. Average monthly salary earnings for employees, by sector................................................15 Figure 38. Monthly labor income, bottom 40 and top 60......................................................................15 Figure 39. Share of low-intensity agriculture among adults (aged 15+)..............................................15 Figure 40. Real growth, average monthly pension, % year-on-year.....................................................16 Figure 41. Real growth, average monthly pension, by welfare group, % year-on-year......................16 Figure 42. Income structure of the poor, %.............................................................................................16 Figure 43. Income structure of the bottom 40, %...................................................................................16 Figure 44. Income structure, poor, by area..............................................................................................17 Figure 45. Income structure, bottom 40, by area....................................................................................17 Figure 46. Real year-on-year growth of social assistance, by group.....................................................17 Figure 47. Share of social assistance in overall income, by group........................................................17 Figure 48. Adult population (15+) who are active in Moldova or abroad, 2000–14..........................18 Figure 49. Remittances as a share of monthly disposable household income, 2006–14, %..............19 Figure 50. Remittances as a share of GDP, 2014.....................................................................................20 Figure 51. Household welfare ranking before (left) and after (right) remittances.............................20 Figure 52. Urban population share in selected countries, %.................................................................21 Figure 53. Type of settlement, poor and nonpoor.......................................................................................22 Figure 54. Type of settlement, bottom 40 and top 60..................................................................................22 Figure 55. Age composition of the poor..................................................................................................24 Figure 56. Age composition of the bottom 40.........................................................................................24 v Poverty and Shared Prosperity in Moldova: Progress and Prospects Figure 57. Poverty rate, by household composition....................................................................................24 Figure 58. Dependency ratio, the poor and nonpoor..................................................................................24 Figure 59. Composition, remittance households....................................................................................25 Figure 60. Poverty rate, by age..................................................................................................................25 Figure 61. Living conditions and access to utilities, poor.....................................................................25 Figure 62. Living conditions and access to utilities, bottom 40............................................................25 Figure 63. Level of education of among poor and nonpoor adult population...................................26 Figure 64. Level of education of among bottom 40 and top 60 adult population..............................26 Figure 65. Population, by health status....................................................................................................27 Figure 66. Nonmonetary indicators.........................................................................................................29 Figure 67. Nonmonetary poor, by region................................................................................................29 Figure 68. Share of people deprived in each dimension........................................................................30 Figure 69. Contribution of each dimension to multidimensional poverty.........................................30 Figure 70. Overlap, nonmonetary and monetary poor.........................................................................30 Figure 71. Nonmonetary poverty, by subjective well-being and consumption quintiles..................30 Figure 72. Demographic composition, 2000–30....................................................................................31 Figure 73. Age-gender pyramid, 2015–60...............................................................................................31 Figure 74. Adult (15+) dependency ratio: inactive population relative to active population, 2013.........32 Figure 75. Pension system dependency ratio, number of pensioners per contributor......................32 Figure 76. Pension coverage, population above standard retirement age...........................................32 Figure 77. Replacement rates, selected countries...................................................................................33 Figure 78. Replacement rates, 2013–2069...............................................................................................33 Figure 79. Remittances to Moldova, 1995–2015.....................................................................................33 Figure 80. Value added in agriculture and GDP growth, 2001–14......................................................34 Figure 81. Composition of the consumption of the poor and nonpoor..............................................35 Figure 82. Composition of the consumption of the bottom 40 and top 60........................................35 Figure 83. Consumer price indicators, year-on-year, 2007–15.............................................................35 vi Poverty and Shared Prosperity in Moldova: Progress and Prospects Tables Table 1. Moldovan labor emigrants, by labor sector in home and destination countries, 2012.......40 Table 2. Multidimensional poverty index: dimensions and indicators................................................40 Boxes Box 1. National Poverty Measurement Methodology............................................................................4 Box 2. Transport and Household Welfare in Moldova..........................................................................22 Box 3. Out-of-Pocket Health Spending...................................................................................................27 vii Poverty and Shared Prosperity in Moldova: Progress and Prospects Acknowledgments This report has been prepared by María E. Dávalos (Senior Economist, TTL), Tu Chi Nguyen (consultant), and Mikhail Matytsin (consultant) of the World Bank Poverty Global Practice. The team is grateful for comments received from World Bank colleagues, including Reena Badiani, Hanan Jacoby, Ruth Hill, and Ruslan Piontkivsky, and for comments received in early consultations in Chișinău in June 2015. We thank Qimiao Fan (Country Director), Carolina Sánchez-Páramo (Practice Manager), and Alexander Kremer (Country Manager) for their guidance and support. Most importantly, we thank the management and staff of the National Bureau of Statistics of Moldova for their hard work in producing survey data for Moldova and for their collaboration with the World Bank team. viii Poverty and Shared Prosperity in Moldova: Progress and Prospects Executive Summary Moldova has experienced rapid economic growth in the past decade, which has been accompanied by reductions in poverty and good performance in shared prosperity. Nonetheless, Moldova remains one of the poorest countries in Europe and faces challenges in sustaining the progress. Analysis of the Household Budget Surveys from 2007 to 2014 shows that economic growth was volatile, revealed once more in developments during 2015, but, overall, positive and pro-poor. Economic growth was driven generally by private consumption, fueled by remittances, and household consumption expanded accordingly. Public and private transfers, namely, pensions and remittances, had an important role in reducing poverty. Moldovan labor markets contributed to the progress, mostly through productivity increases rather than job creation, given that employment fell driven by increasingly high inactivity rates. On average, the declines in employment were partly offset by the higher wages in the nonfarm sectors. The challenges for progress include spatial and cross-group inequalities, particularly because of unequal access to assets, services and economic opportunities. Moreover, strengthening the persistently weak labor markets to boost employment, especially in the nonfarm sectors, is critical for sustaining progress toward the twin goals of reducing poverty and expanding shared prosperity and for addressing the problems associated with an aging population in a fiscally responsible manner. Accordingly, ensuring the viability of the pension system and improving social assistance are necessary areas of reform, particularly in a context of fiscal pressures, the aging population, and the great vulnerability of the poor to shocks. The Moldova Poverty Assessment 2016 includes three prongs of analysis: this report, which explores trends and the drivers of poverty and shared prosperity, and the accompanying analyses, “A Jobs Diagnostic for Moldova” and “Structural Transformation of Moldovan Small-Holder Agriculture and Its Poverty and Shared Prosperity Impacts.” The jobs diagnostic explores the main labor demand and supply challenges in Moldova in more detail, while the analysis of structural transformation focuses on the agricultural sector and whether it can become a driver of progress. ix Poverty and Shared Prosperity in Moldova: Progress and Prospects Overview Overview Moldova has experienced rapid economic growth in the past decade, Moldova accompaniedhasby experienced rapid in significant progress economic growth poverty reduction inshared and the past decade, accompanied by prosperity. significant The economy progress in poverty has been growing reduction at 5 percent and annually shared since 2000. Atprosperity. the same time, The economy has been the national growing poverty rate at 5 percent dropped annually sincefrom 68 to 2000. At27thepercent between same time, the2000 and 2004 national poverty rate dropped from and continued the downward trend to 11.4 percent in 2014. Similarly, inequality, 68 to 27 percent between 2000 and 2004 and continued the downward trend to 11.4 percent in measured as the Gini coefficient, declined from 0.3 to 0.23 between 2007 and 2014, 2014. Similarly, and the consumptioninequality, growth measured as the of the bottom 40Gini coefficient, outpaced declined that of the top 60 from 0.3 to 0.23 between 2007 in 2009- and 2014, 2014. and Given the consumption pro-poor growth, thegrowth countryof the bottom experienced 40 outpaced a dynamic processthatofof the top 60 in 2009-2014. high Given upward pro-poor economic mobilitythe growth, andcountry experienced little churning (that is, a dynamic process contemporary of high upward economic movements in and out mobility of poverty). and Its achievements little churning (that is,in poverty reductionmovements contemporary and shared prosperity in and out of poverty). Its have been impressive given its economic level and compared with other countries achievements in poverty reduction and shared prosperity have been impressive given its economic in Europe and Central Asia (Figure 1 and Figure 2). Yet, with one of the highest level andrates poverty compared with other in the region countries – 41 percent in population of its Europe and Central lived below Asia (Figure 1 and Figure 2). Yet, the regional line one with of the of 5 USD highest a day (2005poverty in the region rates Moldova PPP) in 2014, needs– to41 percent continue ofprogress. this its population lived below the regional line of 5 USD a day (2005 PPP) in 2014, Moldova needs to continue this progress. Figure 1. Poverty rate and GDP per capita, Figure 2. Welfare growth of the bottom 40 and latest 1. Figure available Poverty rate and GDP per capita, latest average population, Figure 2. latest available Welfare growth of the bottom 40 and available average population, latest available 90 10% 90 [CELLRA 8% 80 [CELLRA 80 NGE] [CELLRA 6% PPP) 70 NGE] 4% NGE] [CELLRA  PPP) 70 2% 60 NGE] 2005 0% 60  2005 50 -2% USD 40 50 [CELLRA -4% Bottom 40%  USD 30 [CELLRA NGE] -6% Total Population  (5(5 40 -8% Macedonia, FYR (2009-… NGE] [CELLRA [CELLRA rate 20 [CELLRA -10%  rate Russian Federation… Bosnia&Herzegovina… Slovak Republic (2007-… 30 NGE] NGE] NGE] Slovenia (2007-2012) Kazakhstan (2009-2013) Georgia (2009-2014) Albania (2008-2012) Armenia (2009-2014) Romania (200.7-2012) Turkey (2008-2013) Ukraine (2009-2014) Moldova (2009-2014) Belarus (2009-2014) Montenegro (2009-2014) Serbia (2008-2013) Bulgaria (2007-2012) Tajikistan (2012-2014) Poverty 10 Kyrgyz Republic… Poverty 20 [CELLRA 0 [CELLRA [CELLRA [CELLRA [CELLRA NGE] 0 5,000 10,000 15,000 NGE] [CELLRA 20,000 25,000 NGE] 10 NGE] [CELLRA NGE] [CELLRA [CELLRA [CELLRA NGE] NGE] NGE] GDP per  capita (2011[CELLRA  PPP) NGE] 0 NGE] NGE] 0 5,000 10,000 15,000 20,000 25,000 GDP per capita (2011 PPP) Source: ECAPOV database harmonization as of April 2016, Europe and Central Asia Team for Statistical Development, World Bank, DC. database harmonization as of April 2016, Europe and Central Asia Team for Statistical Development, World Bank, Washington, DC. Source: ECAPOV Washington, The past progress has been driven mainly by pensions and remittances. Economic growth in Moldova has been mostly driven by private consumption, which was in turn fueled by remittances. 5.7 percentage points to GDP growth in 1999–2014. Private consumption contributed as much as x Remittances account for 26 percent of GDP in 2014, making Moldova one of the countries most dependent on remittances. More than 25 percent of Moldovan households received remittances, Poverty and Shared Prosperity in Moldova: Progress and Prospects The past progress has been driven mainly by pensions and remittances. Economic growth in Moldova has been mostly driven by private consumption, which was in turn fueled by remittances. Private consumption contributed as much as 5.7 percentage points to GDP growth in 1999–2014. Remittances account for 26 percent of GDP in 2014, making Moldova one of the countries most dependent on remittances. More than 25 percent of Moldovan households received remittances, which make up around 18 percent of their income. Remittances helped lift many households, particularly rural ones, out of poverty, and contributed to 21.6 percent of the income growth of the bottom 40 in 2010–14. Similarly, as the government raised pensions to support vulnerable categories, especially during the global crisis, pensions has become the main driver of poverty reduction and shared prosperity, accounting for more than 30 percent of the consumption growth of the bottom 40 in 2007–14. Labor markets contributed to the progress, but mostly through nonagricultural wage increases rather than employment creation. Employment has been declining (from 55 percent in 2000 to below 40 percent in 2014), especially in rural areas. This trend is associated with rising inactivity driven by increasing migration and early retirement among the aging population. With the exceptions of agriculture after 2012 and a few trade-related sectors such as sales, tourism, and transport, employment in most sectors have been on the decline. In contrast, the share of people working in low-intensity agriculture (less than 20 hours a week) has been rising steadily, from 13 percent to 24 percent. Wage growth has been positive, but mainly in the nonagricultural sectors and less in agriculture where the majority of the poor and the bottom 40 work. As a result of this differential wage growth, combined with the shift toward subsistence farming (especially among the poor and bottom 40), the gap in labor income (including wages and self-employment earnings) between the bottom 40 and the top 60 has shown limited signs of narrowing. There are concerns about the sustainability of past achievements as the poor and bottom 40 continue to lack the necessary capital to advance. The slow growth of the agriculture sector and the limited access to markets, non-farm jobs, and modern services mean that people in rural areas are persistently poorer. Coverage of heating, piped water, sewage is limited among the rural population (and the poor and bottom 40). The poor and bottom 40 also have much less educational attainment and own smaller plots of land, which limit their opportunities to gain (better) jobs and improve productivity. And even though they have similar health profiles as the nonpoor, they are less likely to have medical insurance, hence suffer from low quality healthcare or high out-of-pocket spending, which drives them further into poverty in the long term. The same constraints apply to the ethnic minorities, who, because of language barrier or disparities in access to services, end up in worse welfare situation. These dimensions of wellbeing, including educational, health, employment, and housing status, determine a person’s capacity to enjoy decent social and economic living standards. This report measures the deprivations of individuals along these dimensions and finds that 24 percent of the population are considered multidimensionally poor in 2014, more than twice the number of those who are monetary poor, and there has been limited improvement since 2007. xi Poverty and Shared Prosperity in Moldova: Progress and Prospects Continued progress in poverty reduction and shared prosperity face tremendous risks due to long-term fiscal pressures and high volatility in the economy. Driven by declining fertility and accelerating emigration among the young population, the population of Moldova is shrinking and aging rapidly. Combined with low labor force participation, the contribution base of the pension system will contract, hence undermining the system sustainability and reducing pension coverage of the retirees. Even though pensions have not been generous and often insufficient to lift many of the elderly out of poverty – the poverty rate among the elderly is higher than the population average, their potential contraction may jeopardize the economic security of the elderly. At the same time, the economic slowdown in the European Union and Russia is hampering the remittance inflows and growth rates of remittances is expected to be slower than in the past, making it less likely for households to rely on this income source to sustain their consumption. Finally, the agriculture sector is subject to high volatility due to climate and external demand shocks. Since the poor rely on agriculture as a major part of their income and consumption, any fluctuations in the sector will affect their wellbeing directly. Meanwhile, social assistance has limited potential to serve as a safety net for the poor and vulnerable. The main targeted programs are Ajutor Social (social aid) and a heating allowance program, both of which are relatively well targeted, but the coverage is not wide and the benefits low, which constrains the ability of the programs to respond to widespread downturns in household income. These challenges point to the need to promote a more vibrant domestic labor market to lead future progress in poverty reduction and shared prosperity. This involves creating more (and better) jobs and enhancing access to education, health, and services to allow individuals to access those jobs. Some of these policies need to pay additional attention to the abovementioned structural issues, in particular: (i) aging – efforts to promote active and healthy aging can help people work longer and reduce the looming economic dependency ratio growth; (ii) regional and group disparities – policies to ensure equitable opportunities for rural populations and ethnic minorities can allow people to better contribute to the economy; and (iii) economic and climatic shocks – policies to increase adaptation and mitigation of climatic shocks, including through social assistance programs to protect the vulnerable in times of needs. Measures to help households manage and adapt to risks need to be complemented with labor market policies aimed at diversifying household income sources – particularly for the poor who rely disproportionately more on agricultural income. The report is structured as follows. The first section lays out the overall macroeconomic environment of Moldova in recent years and briefly summarizes the report findings. Section II describes trends in poverty reduction and shared prosperity in Moldova. Section III provides analysis of the main factors behind the progress. Section IV continues by assessing the sustainability of the progress and pointing out the remaining challenges and risks. The last section discusses policy implications and concludes. xii Poverty and Shared Prosperity in Moldova: Progress and Prospects I. Introduction 1. Introduction Moldova has experienced rapid economic growth in the past decade, despite volatility. The country has grown on average by 5 percent annually since 2000 (Figure 3), much more rapidly than other countries in the region (Figure 4). Even with the sharp contraction during the global financial Moldova has experienced rapid economic growth in the past decade, despite volatility. The country has crisis of grown on2008–09, 5 percent continued average byMoldova annually sinceto grow quickly3), 2000 (Figure until much 2014, more averaging 5.4 other rapidly than percent growth countries in in 2010–14. the 4). Even growth Its economic region (Figure trajectory with the sharp has been contraction increasingly during volatile, crisis the global financial however. As a Moldova small and of 2008–09, open economy, continued to growMoldova quickly has 2014, not borne until only several averaging external 5.4 percent economic growth shocks in 2010–14. Itsin the past, growth but also economic trajectory has been climatic shocks thatincreasingly volatile,affected have particularly As however.its a small andsector, open economy, agricultural as well as Moldova the wider has borne notIn economy. only several external economic shocks in the past, but also climatic shocks that have particularly affected its 2015, a confluence agricultural of events sector, as well pushed as the wider In 2015,into the economy economy. a downturn a confluence (−0.5 of events percent pushed growth in the economy gross into a domestic product downturn [GDP] (−0.5 percent in 2015) growth and in gross projections domestic of little product [GDP] growth in 2016. in 2015) The main and projections offactors behind little growth in 2016. this Thepoor recent factors behind mainperformance this are recent poor weaker performance external are weaker flows, large-scale external bank flows, fraud, andlarge-scale a drought, bank all of fraud, and a drought, all of which took place in an environment which took place in an environment of political instability. 1 of political instability. 1 Figure 3. GDP growth, 2001–15 Figure 4. Real GDP index (2007 = 100) 140 10 8 6 120 4 2 100 0 -2 2000 2002 2004 2006 2008 2010 2012 2014 80 -4 2007 2008 2009 2010 2011 2012 2013 2014 2015 -6 Albania Bosnia & Herzegovina Georgia Croatia -8 Moldova Macedonia, FYR Romania Serbia Moldova ECA Slovak Republic Slovenia Sources: World Development Indicators database, World Bank; Sources: World Bank 2016b, based on WDI and IMF WEO. World Bank 2016a. Labor market Labor marketoutcomes outcomes are are weak in Moldova, weak in Moldova, inactivity and and rates rates inactivity are high. Employment are high. has been Employment has declining, especially in rural areas, driven by increasing migration and early retirement been declining, especially in rural areas, driven by increasing migration and early retirement among among the aging population. Even though the adult population (aged 15+) has stabilized in recent years, the share of people the aging working orpopulation. looking for a Even though job abroad has the adult population increased (agedin from 4.3 percent 15+) 2000 has stabilized to 10 percent in in 2014, recent years,to leading the share a of people declining working orIn active population. looking addition, for thea retirement job abroad agehas increased in Moldova from 4.3low is relatively percent (56 forin 2000 and women to 10 61 for men) and, post-retirement employment is low — 30.5 percent in 2014. percent in 2014, leading to a declining active population. In addition, the retirement age in Moldova 2 As a result, the employment rate decreased drastically is relatively low (56 for 55 from percentand women in 2000 61 tofor below men) 40and, percent in 2014, with the post-retirement biggest decrease employment in rural is low—30.5 areas (22 percentage2 points) ( Figure 5). Unemployment is low, fluctuating around 3 percent during the percent in 2014. As a result, the employment rate decreased drastically from period, but inactivity has been on the rise, from 40 percent in 2000 to 59 percent in 2014 (Figure 6). 55 percent in 2000 to below 40 percent in 2014, with the biggest decrease in rural areas (22 percentage points) (Figure 5). 1 World Bank (2016a). Unemployment 2 is low, fluctuating around 3 percent during the period, but inactivity has been on the Labor Force Survey. rise, from 40 percent in 2000 to 59 percent in 2014 (Figure 6). 1 1 World Bank (2016a). 2 Labor Force Survey. Poverty and Shared Prosperity in Moldova: Progress and Prospects Figure 5. Employment rate by demographics and Figure 6. Activity rate by demographics and location, 2000–14 location, 2000–14 60 65 60 55 55 50 50 45 45 40 40 35 35 30 30 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 urban rural men women urban rural men women Source: World Bank calculations based on the Moldova National Bureau of Statistics (NBS) data. How has Moldova fared in poverty reduction and shared prosperity in recent years? National poverty How has estimates Moldova reveal fared in a downward poverty trend reduction in poverty and shared in the 2000s prosperity and earlier. in recent This is good news. Ityears? National is important, however, to understand poverty estimates revealtheatrends by focusing downward trendon povertywelfare in various in the indicators 2000s and across groups earlier. This, digging is gooddeeper news. It into the forces behind the progress so far, and exploring whether these driving factors are sustainable so that is important, however, to understand the trends by focusing on various welfare indicators across the prospects will be equally positive among the less well off. groups, This paper digging deeper aims to assess theinto recentthe forces trends behind and drivers of the progress poverty decline andso shared far, and exploring whether prosperity—the twin goalsthese of driving the Worldfactors Bank—in sustainable areMoldova so that and the the prospects potential will be challenges ahead. equally It relies positive on data of the among the Household less well Budget off. Survey (HBS) from 2007 to the latest year available, 2014, produced by the Moldova National Bureau of Statistics (NBS). This paper aims to assess the recent trends and drivers of poverty decline and shared from this twin prosperity—the Findings of the goalsshow assessment World Bank—in Moldova and the potential challenges the following: • Moldova ahead. exhibited It relies data performance on good in reducing of the Household poverty Budget and inequality Survey (HBS) and boosting from 2007 shared to theprosperity. latest year This progress was underpinned by high upward economic mobility. available, 2014, produced by the Moldova National Bureau of Statistics (NBS). • Economic growth was volatile, but positive and pro-poor overall. Public and private transfers, namely, pensions Findings remittances, and this from assessment had show an important role in reducing poverty. Moldovan labor markets the following: contributed to the progress, mostly through productivity increases rather than job creation, given that  employment Moldova exhibited good declined over performance the period, driven inbyreducing high andpoverty and inequality rising inactivity rates. and boosting shared • C hallenges remain, prosperity. and the was This progress prospects are not too underpinned by favorable. high upward Spatial and cromobility. economic ss-group inequalities persist, particularly in the unequal access to assets and services among, for example, rural areas and ethnic  minorities. Economic growth was volatile, but positive and pro-poor overall. Public and private transfers, namely, pensions • Important and growingandrisks remittances, to sustainable an important had progress persist.role in reducing Remittances poverty. may and pensions Moldovan labor not remain crucial forces behind progress in the future, given the changing external markets contributed to the progress, mostly through productivity increases rather than job environment and the fiscally unsustainable creation, given that system. pension employmentIn addition, the less declined well-off over are increasingly the period, driven vulnerable by high and rising shocks. to climate inactivity This report is accompanied by two additional parts that are also critical to understanding two of the rates. that Moldova faces: weak labor markets (explored in “A Jobs Diagnostic for Moldova”) and a low challenges  Challenges productivity remain, and agricultural the sector inprospects which many are not of the favorable. toopoor Spatial and are concentrated cross-group (explored inequalities in “Structural Transformation of Moldovan persist, particularly Small-Holder in the unequal access Agriculture andand to assets servicesand Its Poverty Shared among, Prosperity for example,Impacts” ). rural areas Together, the three parts represent the Moldova Poverty Assessment 2016. They will provide significant inputs to the and ethnic minorities. more comprehensive approach of the Moldova Systematic Country Diagnostic, which precisely explores the  Important main constraints,and growing across risks to to the economy, sustainable achieving progress progress in persist. poverty Remittances reduction and and pensions shar may not ed prosperity. 3 Emerging remainmessages crucialindicate that creating forces behind jobs is progress incritical to raising the future, people’s given living standards the changing in aenvironment sustainable way external and in Moldova and to alleviating pressures related to aging and the fiscal system. the fiscally unsustainable pension system. In addition, the less well-off are increasingly vulnerable This document is organized as follows. The next section (section II) describes trends in poverty reduction to climate and shared shocks. prosperity in Moldova. Section III provides analysis of the main factors behind the advances. To This the assess prospects, report this discussion is accompanied byis followed, in sectionparts two additional an analysis IV, by that are alsoof the remaining critical challenges and to understanding risks. The last section discusses policy implications and concludes. two of the challenges that Moldova faces: weak labor markets (explored in “A Jobs Diagnostic Moldova”) for See 3 and a low productivity agricultural sector in which many of the poor are concentrated World Bank (2016b). 2 14 Systematic Country Diagnostic, which precisely explores the main constraints, across the economy, to achieving progress in poverty reduction and shared prosperity.3 Emerging messages indicate that creating jobs is critical to raising people’s and standards living Poverty in Moldova:way in a sustainable Shared Prosperity in Moldova Progress and to and Prospects alleviating pressures related to aging and the fiscal system. Progress in poverty This document is organized as follows. The next section (section II) describes trends in poverty reduction and shared prosperity in Moldova. Section III provides analysis of the main factors behind 2. reduction and shared the advances. To assess the prospects, this discussion is followed, in section IV, by an analysis of the remaining challenges and risks. The last section discusses policy implications and concludes. II. prosperity Progress in poverty reduction and shared prosperity Poverty declined in Moldova . . . The poverty Poverty rate significantly declined declined in Moldova between 2007 and 2014. Over the period, in Moldova... the national poverty rate fell from 26.0 percent to 11.4 percent (Figure 7), although the downward The trendpoverty during the declined rate significantly stagnated in Moldova global financial between crisis 2007 and of 2008–09. 4 2014. This Over the period, the national is a continuation, albeit at a poverty offrom rate fell slower rate, 26.0 percent the progress madeto 11.4 percent in the early(Figure 2000s,7),when thethe although downward national trendrate poverty stagnated droppedduring from the global financial crisis of 2008–09.4 This is a continuation, albeit at a slower rate, of the progress made in 68.0 the to 2000s, early when the (2000–04), 27.0 percent after national poverty peaking rate dropped from 68.0 the following 1998 to 27.0 crisis percent involving (2000–04), the after Russian peaking Federation. following theSimilarly, 1998 crisis absolute poverty involving at the the Russian World Bank Federation regional . Similarly, poverty absolute line of poverty $5.00 at the purchasing World Bank power parity (PPP) U.S. dollars per day almost halved, to 40.7 percent, and absolute poverty40.7 regional poverty line of $5.00 purchasing power parity (PPP) U.S. dollars per day almost halved, to ($2.50 percent, and absolute poverty ($2.50 PPP a day), at 2.9 percent, was almost eradicated. (See box 1 for technical PPP a day), at 2.9 percent, was almost eradicated. (See box 1 for technical information on the information on the calculation of the poverty rate.) calculation of the poverty rate.) Figure 7. Poverty headcount ratio, 2007–14 100% 13% 12% 12% 11% 12% 12% 12% 12% 80% 38% 36% 38% 40% 43% 42% 60% 48% 47% 40% 38% 39% 38% 40% 38% 40% 20% 36% 38% 12% 13% 12% 9% 7% 6% 0% 4% 3% 2007 2008 2009 2010 2011 2012 2013 2014 <$2.50 $2.50-$5.00 $5.00-$10 >$10 National pov line Source: World Bank calculations based on the HBS. Note: The national poverty rate and the World Bank poverty rate rely on two separate consumption aggregates. The figure relies on national and regional poverty lines. 3 See World Bank (2016b). 4 The poverty line was set at MDL 104.67 per equivalent adult per month in 2014. 15 4 The poverty line was set at MDL 104.67 per equivalent adult per month in 2014. 3 Poverty and Shared Prosperity in Moldova: Progress and Prospects Box 1. National Poverty Measurement Methodology Poverty in Moldova is measured using a basic needs approach and relies on consumption expenditure as an indicator of living standards. Consumption is the preferred welfare indicator because it is more accurately measured and less exposed to misreporting. To provide better comparability across the country, the consumption aggregate is modified in several ways, as follows: • Expenditures for durables and rent are excluded from the aggregate because there is not sufficient data to estimate correctly the stream of services from durables and imputed rent for the owners of housing. • The differences in energy tariffs are considered, and these price distortions are corrected to account for the actual benefit that the household receives. • Nominal expenditure has been adjusted for inflation as well as for regional price differences through a Paasche price index constructed using data collected in the survey and information from the official consumer price index. • To capture the economies of scale within the same households, equivalence scales have been adopted. The former scale of the Organisation for Economic Co-operation and Development is used: 1 for the first household member, 0.7 for any other adult, and 0.5 for children ages below 15. The poverty line is set using cost of basic need methods. The food poverty line is set to meet the minimum energy requirement of 2,282 calories per day per average person, which corresponds to 3,004 calories per day per adult equivalent. The structure of the food bundle is taken directly from the survey and corresponds to the actual set of consumed food by the groups in the HBS from the second to the fourth deciles. The nonfood component of poverty lines is calculated as the share of nonfood expenditures of households the total expenditures of which are equal to the food poverty line. A standard set of poverty measures proposed by Foster, Greer, and Thorbecke (1984) is used to determine the poverty rate. The poverty methodology has deteriorated for various reasons. First, the sampling frame is outdated because the 2004 census continues to be used. The results of the 2014 census have not yet been released. This is an important consideration, given the large changes the country has experienced in the past decade, including migration. The poverty line also needs to be updated to reflect changes in consumption patterns, particularly because poverty has declined. Efforts to carry out this updating are under way at the NBS. Finally, there is high nonresponse at the national level, at 34 percent by December 2013, and response rates declined by 2.6 percent between 2006 and 2013. This is driven by refusal rates, which peaked in October 2010 at 22.3 percent from a baseline of 10.4 percent in January 2006 and stood at 18.1 percent in December 2013. Starting from an already low response rate in 2006, the response rate continues to be extremely low in urban areas (41 percent), particularly in Chișinău (24 percent), after small but systematic declines. In rural areas and other regions, the changes are much smaller, and the response levels are higher. Source: World Bank, based on an NBS note on poverty measurement in September 2007 and an NBS note on nonresponse analysis. Poverty has declined throughout the country, but regional disparities persist. In 2007–09, a period of sharp GDP contraction, the progress in poverty in urban and rural areas diverged. Poverty increased in rural areas and declined in urban area (Figure 8). In 2010–14, both poverty rates fell by half, and rural poverty remained at three times the level of urban poverty (16.3 percent and 5.0 percent, respectively). A similar gap existed between the urban Chișinău, the capital, which accounts for 23 percent of the population, and the rest of the country. Although the gap has shrunk in the last decade, poverty in other regions is still more than five times the rate in Chișinău (Figure 9). Poverty rates in the north, center, and south followed similar trends, although they diverged from the same starting point in 2007. 4 rates in the north, center, and south followed similar trends, although they diverged from the same rates in the north, center, and south followed similar trends, although they diverged from the same starting point in 2007. starting point in 2007. Poverty and Shared Prosperity in Moldova: Progress and Prospects Figure 8. Poverty rates, by urban and rural areas, % Figure 9. Poverty rates, by region, % Figure 8. Poverty rates, by urban and rural areas, % Figure 9. Poverty rates, by region, % 40% 40% 40% 40% 30% 31.3% 30.3% 30% 30% 31.3% 25.8% 30.3% 30% 25.8% 21.9% 20% 20% 18.4% 21.9% 20% 16.3% 20% 18.4% 16.3% 10.4% 11.4% 10% 10% 10% 10% 10.4% 11.4% 5.0% 0% 0% 5.0% 0% 2007 2008 2009 2010 2011 2012 2013 2014 0% 2007 2008 2009 2010 2011 2012 2013 2014 2007 2008 2009 2010 2011 2012 2013 2014 2007 2008 2009 2010 2011 2012 2013 2014 Central North Total Rural Urban Central South North Chisinau Total Rural Urban South Chisinau Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. The depth and severity of poverty in Moldova have also declined in the past decade. Similar The The to the depth depth andand headcount severity severity of poverty ratio, the ofpoverty poverty in Moldova in Moldova gap have also have also declined and the squared declined in the poverty in the past decade. gap past Similar indicators to decade. Similar the headcount improved in both ratio, to thethe poverty gapratio, headcount and the squared the poverty poverty gap gap andindicators the improved squared poverty 5 in both gapurban and rural indicators areas in 2007– improved in both urban 14 and (Figure 10 rural in 52007–14 areas11). and Figure This means (Figure 10 that the and Figure well-being 11). of the people 5 This means with that thebelow consumption well-being of the the poverty urban people line, withrural and consumption particularly areas in 2007–14 the poorest ofbelow the the poor, (also Figure poverty10 and line, improved Figure 11). particularly significantly This the even means poorest though theof the the that people well-being poor, also may have of the improved remained people with even poor. significantly consumption though the below poverty the may people have line, particularly remained poor. the poorest of the poor, also improved significantly even though the people may have remained poor. Figure 10. Poverty gap, by urban and rural areas Figure 11. Poverty gap squared, by urban and rural areas Figure 10. Poverty gap, by urban and rural areas Figure 11. Poverty gap squared, by urban and rural areas 10% 4% 10% 8% 4% 3% 8% 6% 3% 6% 2% 4% 2% 4% 2% 1% 2% 1% 0% 0% 0% 2007 2008 2009 2010 2011 2012 2013 2014 2007 2008 2009 2010 2011 2012 2013 2014 0% 2007 2008 2009 2010 2011 2012 2013 2014 2007 2008 2009 2010 2011 2012 2013 2014 Total Rural Urban Total Rural Urban Total Rural Urban Total Rural Urban Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. . . . and the country reduced inequality and boosted shared prosperity . . . and the country reduced inequality and boosted shared prosperity ... and the Relatively country larger reduced increases inequality in consumption and growth boosted among shared the less prosperity well off supported the Relatively decline larger increases in inequality. in consumption The World Bank goal ofgrowth among boosting shared the less prosperity well aims off supported to ensure the that growth Relatively larger increases in consumption growth among the less well off supported the decline in decline the reaches in inequality. less off wellBank TheandWorld monitored isboosting of boosting Bank goal through shared anaims prosperity indicator that aims to ensure measures the that incomegrowth inequality. The World goal of shared prosperity to ensure that growth reaches the lessor reaches consumption well theis less off and wellamong growth monitored off and through is indicator people an monitored in the bottom thatthrough an 40 percent measures the indicator of the income orthat measuresdistribution consumption consumption the income growth among in or a consumption people country in the (the growth bottom bottom40 among percent 40). In people of in theconsumption the consumption Moldova, bottom 40 percent distributiongrowth ofamong the consumption in a country (the thebottom 40). bottom distribution In 40Moldova, outpacedin a consumption growth country (the growth among bottomamong the 40). In bottom 40 outpaced consumption growth among the top 60 percent of the consumption theMoldova, top 60 percentconsumption growth among of the distribution (the topthe 60)bottom in 2007–1440 outpaced (Figure distribution (the top 60) in 2007–14 (Figure 12). In 2010, growth rate among the bottom 40 rose slightly, consumption 12). In while the2010, growth growth rate among among the rate top 60 the among top 60 percent theDuring declined. bottom the40 of the restrose distribution slightly, of the (the while the rate period, consumption top 60) among among in 2007–14 the top the bottom (Figure 4060 In60 12).top declined. and 2010, During grew growth in rate the rest parallel. aamong Asof the period, result ofthe thesebottom dynamics roseamong 40across consumption slightly, the groupsthewhile bottom capturedthe rate in40 theandamong HBS, top 60thegrew consumptiontop in60 declined. declined inequality During (Figure the rest theGini of The 13). period, consumption coefficient, among for example, the from declined bottom and 400.23 0.3 to intop 60 grew in 2007–14. 5 The poverty gap is measured as the average distance between the welfare of the poor and the poverty threshold. 5 The poverty gap is measured as the average distance between the welfare of the poor and the poverty threshold. 5 17 The poverty gap is measured as the average distance between the welfare of the poor and the poverty threshold. 17 5 parallel. As a result of these dynamics across the groups captured in the HBS, consumption parallel. inequalityAs a result declined of these (Figure dynamics 13). The across the Gini coefficient, groups for captured example, in from declined the 0.3 HBS, consumption to 0.23 in 2007– inequality 14. declined (Figure 13). The Gini coefficient, for example, declined from 0.3 to 0.23 in 2007– Poverty 14. and Shared Prosperity in Moldova: Progress and Prospects Figure 12. Real consumption per capita growth, Figure 13. Dynamics of consumption per by group Figure 12. Real consumption per capita growth, Figure capita inequality 13. Dynamics of consumption per by group 10% 0.32 inequality capita 4.0 8% 10% 0.32 0.30 4.0 6% 8% 3.5 0.30 0.28 4% 6% 3.5 0.28 0.26 3.0 2% 4% 0.26 0.24 3.0 0% 2% 2008 2009 2010 2011 2012 2013 2014 2.5 -2% 0% 0.24 0.22 Gini (LHS) P90/P10 (RHS) 2.5 -4% -2% 2008 2009 2010 2011 2012 2013 2014 0.22 0.20 Gini (LHS) P90/P10 (RHS) 2.0 -6% -4% Total Bottom 40 Top 60 2007 2008 2009 2010 2011 2012 2013 2014 0.20 2.0 Total Bottom 40 Top 60 2007 2008 -6% World Bank calculations Source: based on the HBS. Source: World Bank 2009 2010 2011 calculations based2012 2013 on the 2014 HBS. Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. Consumption growth in urban and rural areas among the bottom 40 was higher overall than Consumption Consumption among growth the more growth in urban well in and urban off, rural but and with areas rural among areas significant the amongbottom the 40 fluctuations. Inhigher was bottom overall 40 was particular, than higher the among40 overall bottom thein than more well off, but with significant fluctuations. In particular, the bottom 40 in urban areas faced sharp among urban the areas fluctuations more faced in wellfluctuations sharp consumption, off, whilebut with people fluctuations. significant in consumption, progress among while in progress rural areas wasIn particular, among more people stable. the inGDP The bottom rural areas decline 40 inin was urban more 2012 areas faced stable. affected The the sharp GDP urban fluctuations decline bottom in in 2012 40 more consumption, than affected people thewhile in ruralurbanprogress areas, bottom although among 40 former the people more thaninpeople rural benefited moreareas infromwas rural more the areas, in The stable. increase although pensions the GDP anddecline former social in 2012 assistance benefited more affected in 2012 from the urban (Figure 14 andbottom increase in pensions Figure 1540 more ). Overall, and than people consumption social in rural growth assistance in was 2012 greater among although areas,(Figure the 14 the bottom and former Figure 1540 than benefited among ). Overall, more the top from thegrowth consumption60, helping to increasereduce was inequality pensions in greater and the amongin both urban social bottom and rural assistance 40 than in areas. 2012 among (Figure the top 1460, andhelping 15). Figure to Overall, reduce consumption inequality in bothgrowth was rural urban and greater among the bottom 40 than areas. among the top 60, helping to reduce inequality in both urban and rural areas. Figure 14. Real consumption growth, urban areas Figure 15. Real consumption growth, rural areas 12% 14. Real consumption growth, urban areas Figure 10% 15. Real consumption growth, rural areas Figure 12% 10% 10% 8% 10% 8% 8% 6% 8% 6% 6% 4% 6% 4% 2% 4% 4% 2% 0% 2% 0% 2008 2009 2010 2011 2012 2013 2014 2% -2% 0% 2008 2009 2010 2011 2012 2013 2014 2008 2009 2010 2011 2012 2013 2014 -2% -4% 0% -2% 2008 2009 2010 2011 2012 2013 2014 -4% -6% -2% -4% Total Bottom 40 Top 60 Total Bottom 40 Top 60 -6% -8% -4% Total Bottom 40 Top 60 Source: World BankTotal Bottom calculations based the HBS. Top 60 on 40 -8% World Bank calculations based on the HBS. Source: Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. Overallprogress Overall progress shared in in prosperity shared was remarkable prosperity in Moldova was remarkable relative to in Moldova other countries relative to otherin Europe countries and Central Overall in Europe Asia. progress and Moldova in shared Central was Asia. among prosperity Moldova thewas leaders was in the remarkable among shared the leadersprosperity in Moldova indicator relative in the shared in other to the region prosperity (realin countries indicator annualized growth was close to 5 percent in 2008–13), although the total growth in consumption was much in Europe the region slower (real (belowand Central annualized 2 percent) Asia. (Figure Moldova growth was 16).6 As was among close a result, to 5 the the gap percent positive leaders in the inbetween 2008–13), shared the although growth prosperity of the indicator total the bottom 40 andin growth region the total the (real annualized population growth was the highest inwas close together Moldova, to 5 percent in 2008–13), with the although the total growth in Kyrgyz Republic. 18 18 6 The bottom 40 in the region is defined based on a harmonized consumption aggregate, which is different from the consumption aggregate used in Moldova. As a result, the growth rate of the bottom 40 here may be different from the rate in the rest of the analysis. 6 consumption was much slower (below 2 percent) (Figure 16).6 As a result, the positive gap between consumption the growth of was much slower the bottom (below 40 and 2 percent) the total (Figure population was16 ).6 highest the As a result, the positive in Moldova, gap between together with the the growth of the Kyrgyz Republic. bottom 40 and the Poverty total and Shared population was Prosperity the in highest Moldova: in Progress Moldova, and together Prospects with the Kyrgyz Republic. Figure 16. Real consumption growth, the bottom 40 and total population, by country 10% 16. Real consumption growth, the bottom 40 and total population, by country Figure 8% 10% 10% Bottom 40% Total Population 6% 8% 8% Bottom 40% Total Population 4% 6% Bottom 40% Total Population 6% 2% 4% 4%0% 2% -2% 2%0% -4% -2% 0% -6% -4% -2% -8% -6% -4% -10% -8% 4) … (2007-… -6% Bosnia&Herzegovina (2007-2012) (2007-2012) (2007-2012) (2007-2012) (2007-2012) (2007-2012) (2007-2012) -10% (2007-2012) (2009-2014) (2007-2012) (2009-2013) (2009-2014) (2007-2012) (2007-2012) (2007-2012) (2007-2012) (200.7-2012) (2007-2012) (2007-2012) (2009-2014) (2007-2012) (2007-2012) (2008-2012) (2008-2013) (2012-2014) (2008-2013) (2009-2013) (2007-2012) (2007-2012) (2007-2012) (2007-2012) (2009-2014) (2009-2014) (2007-2012) (2009-2014) (2009-2014) Croatia (2009-2012) (2007-2012) Montenegro (2009-2014) (2007-2012) (2007-2012) Lithuania (2007-2012) (2008-2013) Albania (2008-2012) (2007-2012) (2009-2014) Bulgaria (2007-2012) (2007-2012) Romania (200.7-2012) Turkey (2008-2013) (2012-2014) Ukraine (2009-2014) (2009-2014) Moldova (2009-2014) (2009-2013) (2007-2012) Russian Federation (2007-2012) Kazakhstan (2009-2013) 4) … (2007-… Kyrgyz Republic (2009-201 (2009-201 Bosnia&Herzegovina (2007-2012) (2007-2012) (2007-2012) (2007-2012) (2007-2012) (2007-2012) (2007-2012) (2009-2014) (2009-2014) (2007-2012) (2007-2012) (2007-2012) (2007-2012) (2007-2012) (2009-2013) (2007-2012) (2007-2012) (2007-2012) (2007-2012) (2008-2012) (2007-2012) (200.7-2012) (2009-2014) (2008-2013) (2008-2013) (2012-2014) (2007-2012) (2007-2012) (2009-2013) (2007-2012) (2007-2012) (2009-2014) (2007-2012) (2009-2014) (2009-2014) (2009-2014) Bosnia&Herzegovina (2007-2011) Federation Federation Tajikistan Poland Belarus Serbia Hungary Georgia Latvia Estonia Republic Armenia Macedonia, FYR Slovenia Kazakhstan Montenegro Albania Armenia Estonia Denmark Poland Lithuania Ireland Netherlands Georgia Greece Spain France Finland Moldova Italy Slovenia Latvia Kingdom Austria Bulgaria Tajikistan Republic FYR Portugal Hungary Belgium Sweden Turkey Ukraine Serbia Republic Republic Belarus Romania Montenegro Kazakhstan Albania Armenia Estonia Lithuania Denmark Poland Ireland Netherlands Greece Spain France Georgia Italy Finland Moldova Slovenia Latvia Kingdom Austria Bulgaria Tajikistan Portugal Republic Hungary Belgium FYR Sweden Ukraine Serbia Turkey Republic Republic Belarus Romania Slovak Macedonia, Russian Macedonia, Russian Slovak Czech Kyrgyz United Slovak Czech Kyrgyz United Source: ECAPOV database harmonization as of April 2016, Europe and Central Asia Team for Statistical Development, World Bank, Washington, Source: DC. database harmonization as of April 2016, Europe and Central Asia Team for Statistical Development, World Bank, ECAPOV Washington, DC. Progress was underpinned by high upward economic mobility Progress Progress was was underpinned underpinned byupward by high high upward economic economic mobility mobility Moldova presented one of the highest levels of upward economic mobility across the region Moldova over Moldova presented the presented past decade.one one of ofPrior thethe highest to levels global thelevels highest of upward economic of upward crisis, economiceconomic upward mobility mobilitymobilityacross in region across the theover Moldova region was the over the past past decade.25 impressive: Priordecade. percent Prior to the global to of theeconomicthe populationglobal crisis, upward movedeconomic out crisis, mobility upward in Moldova of extreme, mobility in was impressive: $2.50-a-day Moldova poverty 25 percent (Figure was of 17). impressive: the population 25 percent moved outof of the population extreme, moved $2.50-a-day out poverty of extreme, (Figure 17). $2.50-a-day The country experienced The country experienced a dynamic process of high upward economic mobility and little churningpoverty (Figure a 17). dynamic process The of high country upward economic experienced mobility a dynamic churning and littlehigh 7 (that is, contemporary and little in movements and out (that is, contemporary of poverty). 7 movements in process and out of upward of poverty). economic mobility churning (that is, contemporary movements in and out of poverty).7 Figure 17. Households in poverty that escaped poverty or the nonpoor who fell into poverty, % Figure 40% 17. Households in poverty that escaped poverty or the nonpoor who fell into poverty, % early time period (circa 2001 to 2008) late time period (circa 2008 to 2010) 40% 30% early time period (circa 2001 to 2008) late time period (circa 2008 to 2010) 30% 20% 20% 10% 10% 0% 0% KAZ UKR ALB MNE POL MKD HUN KAZ POL UKR MNE MDA ROU ARM BLR TUR GEO RUS LTU SRB SVK MDA TUR GEO ROU ARM BLR SRB RUS SVK -10% KAZ UKR ALB MNE POL MKD HUN KAZ POL UKR MNE MDA ROU ARM BLR TUR GEO RUS LTU SRB SVK MDA TUR GEO ROU ARM BLR SRB RUS SVK -10% -20% -20% upwards mobility, share 2.5 USDline downwards mobility, share 2.5 USDline upwards mobility, share 2.5 USDline downwards mobility, share 2.5 USDline Source: Cancho et al. 2015, using World Bank estimates based on ECAPOV database harmonization as of February 2014, Europe and Central 6 The bottom Statistical forin Asia Team40 Development, the region World is defined Bank, based Washington, on a harmonizedDC. consumption aggregate, which is different from the Note: Based on a linear probability model with the dependent variable equal to 1 if the household experienced upward mobility (cross 6 The bottomaggregate consumption 40 in the used regionin Moldova. is defined As a basedresult, on a the growth harmonized rate any threshold) and zero otherwise. Significance evaluated at the 10 percent level. See the bottom of source consumption the 40 here aggregate, paper for more may be which information different is different on the from the rate in the rest consumption methodology. The of the Bank aggregate World analysis. used in Moldova. regional $2.50-a-day As a result, poverty the growth rate of the bottom 40 here may be different from line is used. the 7 Dávalos rate in the and rest Meyer (2015). analysis. of the the After the crisis, rate of upward mobility continued to rise: a large share of poor 7 Dávalos and Meyer (2015). households were able to improve their living standards and escape poverty, while few 19 nonpoor households fell into poverty. The share of 19people who remained poor at the national poverty line was low (3 percent), while the share of those who moved out of poverty was higher than the share of those who fell into poverty. The same dynamics can be observed for the regional 7 Dávalos and Meyer (2015). poverty lines of $2.50 and $5.00 a day, although churning is more common at the higher poverty lines (Figure 18). However, many households in Moldova remain 7 vulnerable to shocks and thus to falling into poverty. In particular, 9 percent of the 2007 nonpoor (defined according to the national poverty line) were poor by 2014, when the consumption of 21 percent of the population was below Note: Based on a linear probability model with the dependent variable equal to 1 if the household experienced upward mobility (cross any threshold) and zero otherwise. Significance evaluated at the 10 percent level. See the source paper for more information on the methodology. The World Bank regional $2.50-a-day poverty line is used. After the crisis, the rate of upward mobility continued to rise: a large share of poor Poverty and Shared Prosperity in Moldova: Progress and Prospects households were able to improve their living standards and escape poverty, while few nonpoor After households the crisis, the rate of fell into poverty. upward The shareto mobility continued of people rise: who a large remained share poor at thewere of poor households national able to improve poverty wasliving their line low (3standards percent),and escape while the poverty, share ofwhile those few who nonpoor moved households fell into out of poverty waspoverty. higher The share of people who remained poor at the national poverty line was low (3 percent), while the share of than the share of those who fell into poverty. The same dynamics can be observed for the regional those who moved out of poverty was higher than the share of those who fell into poverty. The same dynamics poverty can lines of be observed for$2.50 and $5.00 the regional a day, poverty although lines of $2.50 churning and $5.00 is more a day, common although at theishigher churning poverty more common the (Figure lines at 18). However, higher poverty many lines (Figure households 18). in Moldova However, many householdsremain vulnerable in Moldova remainto vulnerable shocks and tothus to shocks and thus falling falling into to poverty. into In poverty. In 9 particular, particular, percent of9 percent the 2007of nonpoor nonpoor according the 2007 (defined (defined according to the to the national poverty poverty national line) wereline) were poor poor by by 2014, 2014, whenwhen the consumption the consumption of 21of 21 percent percent ofof thethe populationwas was below population below $5.00 a day (the regional poverty line). $5.00 a day (the regional poverty line). Figure 18. Intragenerational mobility, by share of the population, Moldova, 2007–14 9 28 8 68 21 88 83 32 9 21 2 19 3 10 National line $2.50 $5.00 $10 Stayers Upward mobility Downward mobility Never poor Source: World Bank calculations based on the HBS. Note: The data are calculated according to the regional poverty lines (2005 PPP) and refer to the lower-bound mobility estimates following the Dang et al. (2011) synthetic panel methodology. See Cancho et al. (2015) for more information on the methodology. Yet, Moldova is one of the poorest countries in Europe Yet, is amongis Moldova Moldova one the of the countries poorest with countries the highest inin Europe poverty rates Europe and Central Asia and the poorest in Europe. Its moderate $5.00-a-day poverty rate is lower than other countries in Moldova the region among is at the similar countries GDP, with but is the highest among poverty the highest rates in the in Europe region, and Central 40 percent in 2013 Asia and 19 (Figure the ). poorest in Europe. Its moderate $5.00-a-day poverty rate is lower than other countries in the region at Because similar it isbut GDP, is unequal less among the than other highest countries in the in percent region, 40 the region, Moldova in 2013 (Figurehad 19). a relatively Because it islow less extreme unequal $2.50-a-day than poverty in other countries rate, 6.0 percent, the region, Moldova 2012 in had and a small a relatively middle $2.50-a-day low extreme class (above $10.00-a-day poverty rate, 6.0 percent, in 201211.7 consumption), and a small middle percent. class (above Nonetheless, $10.00-a-day a large share of consumption), 11.7 percent. the population—41.9 Nonetheless, percent in 2012— a large share of the was concentrated population—41.9 among the vulnerablepercent in 2012—was ($5.00–$10.00-a-day concentrated(Figure consumption) among 20).the vulnerable ($5.00–$10.00-a-day consumption) (Figure 20). 20 8 Poverty and Shared Prosperity in Moldova: Progress and Prospects Figure 19. GDP per capita and $5.00-a-day poverty, Europe and Central Asia, latest available data 30,000 GDP per capita (left axis) Poverty rate (right axis) 90 Poverty rate (5 USD 2005 PPP) GDP per capita (2011 PPP) 80 25,000 70 20,000 60 50 15,000 40 10,000 30 20 5,000 10 - 0 Kosovo Lithuania Moldova Malta Kyrgyz Republic Romania Hungary Ukraine Latvia Armenia Montenegro Poland Estonia Georgia Albania Belarus Croatia Kazakhstan Greece Portugal Slovenia Czech Republic Turkey Slovak Republic Serbia Russian Federation Tajikistan Bulgaria Macedonia, FYR Sources: ECAPOV database harmonization as of April 2016, Europe and Central Asia Team for Statistical Development, World Bank, Washington, DC; World Development Indicators database, World Bank. Figure 20. Welfare group decomposition, by country, latest available data Belarus Slovenia Ukraine Czech Republic Hungary Kazakhstan Slovak Republic Russia Poland Montenegro Serbia Lithuania Estonia Croatia Moldova Latvia Turkey Bulgaria Albania Romania Macedonia FYR Armenia Kosovo Kyrgyz Republic Georgia 0 10 20 30 40 50 60 70 80 90 100 Extreme poor (below US$2.50 PPP) Poor (between US$2.50-5 PPP) Vulnerable (between US$5-10 PPP) Middle Class (above US$10 PPP) Source: ECAPOV database harmonization as of April 2016, Europe and Central Asia Team for Statistical Development, World Bank, Washington, DC. 21 9 Poverty and Shared Prosperity in Moldova: Progress and Prospects What drove poverty 3. reduction and shared prosperity? What drove poverty reduction and shared prosperity? Given the progress in reducing poverty and boosting shared prosperity, this section seeks to identify the drivers Given of progress. the progress It focuses in reducing on and poverty resolving whether boosting economic shared growth prosperity, was pro-poor this section and which seeks to identify the drivers sourcesof progress. resolving whether It focuses on income, of income—labor economic growth public transfers, was pro-poor and which or remittances—drove the sources of positive income—labor income, public transfers, or remittances—drove the positive performance. performance. Economic growth was volatile, but positive and pro-poor overall Economic growth was volatile, but positive and pro-poor overall Economic growth in Moldova has been mostly consumption driven. Fueled by remittances, private Economic growth in Moldova has been mostly consumption driven. Fueled by remittances, consumption contributed as much as 7 percentage points to GDP growth in 1999–2008, that is, prior to the crisis, and is contributed private consumption economic as much a larger contributor as 7 than percentage exports points (Figure 21).8 to GDP growth in 1999–2008, that is, prior to the economic crisis, and is a larger contributor than exports Figure 1).1 Figure 21. GDP growth decomposition 20 10 Percent 0 1996-1998 1999-2008 2009-2014 Average -10 -20 Private Consumption Public Consumption Fixed Investments Change in stocks Exports Imports Sources: World Bank 2016b; World Bank calculations based on national accounts. Given the large contribution of private consumption to GDP, GDP growth was closely mirrored by both household income and consumption growth. Household income and consumption trends followed the GivenGDP overall the trend contribution largeclosely (Figure 22).of private However, consumption this to GDP, also means households GDP from suffered growth was closely the volatility in the mirroredThe economy. bytwoboth household downturns income as a result and of the consumption global growth. economic crisis and theHousehold income drought, in 2009 and, and 2012 respectively, drove down household welfare growth . In 2009, households were able to smooth consumption trends followed the overall GDP trend closely (Figure 2). However, this also means consumption even as income households fell sharply. suffered from the volatility in the economy. The two downturns as a result of the global economic crisis and the drought, in 2009 and 2012, respectively, drove down household welfare growth. In 2009, households were able to smooth consumption even as income fell sharply. 8 See World Bank (2016b). 10 Poverty and Shared Prosperity in Moldova: Progress and Prospects Figure 22. GDP and household income and consumption growth 10% 5% 0% 2008 2009 2010 2011 2012 2013 2014 -5% Figure 22. GDP and household income GDP per capita and consumption growth growth Consumption growth Income growth -10% 10% Source: World Bank calculations based on data of the World Development Indicators database and the HBS. Consumption 5% growth was pro-poor, and changes in the distribution thus contributed more than the growth in average consumption to the decline in poverty rates. From 2007 to 2014, consumption growth was 0% 23. overall, positive Figure Growth but average consumption incidence curve, the bottom amongFigure 20 percent of the 24. Datt-Ravallion distribution (the decomposition bottom 20) of changes grew by more than 2008 2007–142009 total population, 10 percent, surpassing the 2010growth rate 2011 in theamong higher decline in 2012 consumption the poverty rate2013groups ( 201423). This Figure positive, pro-poor growth led to progress in poverty reduction, which was therefore driven by changes in both -5% mean growth and redistribution. According to the Datt-Ravallion (1992) decomposition, changes 2007-2008 -2.0 in 2.6distribution led, overall, to the povertyGDP perin decline 2007–14, capita growth except for brief periods, in two 2008-2009 Income growth 2008–09 and Consumption 2012–13 growth -1.41.3 (Figure 24). In -10% 2009–10, after the global financial crisis, the 4 percentage point decline in poverty -4.0 2009-2010 can-0.4 be almost fully Source: World Bank calculations based on data of the World Development Indicators 2010-2011 database and the explained by changes in distribution. In 2013–14, poverty would have increased, given the decline in HBS.-2.3 -2.1 consumption growth, had it not been for distribution effects.2011-2012 The distribution effects were relatively -1.1 0.2 stronger in 2012-2013 -0.9 -3.0 both urban and rural areas. 2013-2014 -2.4 1.1 2007-2014 (Total) -11.1 -3.3 Figure 23. Growth incidence curve, Figure 2007-2014 4. Datt-Ravallion 2(Rural) -11.6 decomposition -3.4 of changes total population, 2007–14 (Urban) in the poverty in the decline 2007-2014 -10.8 rate -2.6 10 2007-2008 Growth Distribution -2.0 2.6 Source: World Bank calculations based on the HBS. 2008-2009 Source: World Bank calculations based on the HBS. -1.41.3 8 2009-2010 -4.0 -0.4 (%)% rate Labor markets aided the progress mainly by nonagricultural wage increases 6 2010-2011 -2.3 -2.1 growthrates 4 2011-2012 -1.1 0.2 ed growth Domestic 2 labor income growth contributed to income growth among the bottom 40 and to 2012-2013 -0.9 -3.0 Annualized 2013-2014 poverty reduction. Overall, because of a lack of job creation and rising inactivity-2.4 1.1 rates, the Annualize 0 2007-2014 (Total) -11.1 -3.3 contribution -2 of labor income was driven by the 1 5 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 97 nonagricultural 2007-2014 (Rural) sector mostly through -11.6 -3.4 wage increases (Figure 5 and Figure 6). Before the economic crisis, the faster growth 2007-2014 (Urban) -10.8 of nonagricultural -2.6 -4 earnings relative to agricultural earnings placed many agricultural workers into the bottom 40. After Consumption per capita percentiles Consumption per capita percentiles Growth Distribution the crisis, agricultural employment and earnings picked up, but this contributed little. Agricultural Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. income was subject to fluctuations over the period, which undermined its contribution to welfare improvements (see below). Labor markets aided the progress mainly by nonagricultural wage increases Labor markets aided the progress Domestic labor income growth contributed to income growth among the bottom 40 and to mainly by nonagricultural wage increases poverty reduction. Overall, because of a lack of job creation and rising inactivity rates, the contribution Domestic labor of income contributed labor growth income was driventoby the nonagricultural income growth among sector mostly the bottom through 40 and wage to poverty increases (Figure reduction. Overall,5 and Figure because 6). Before of a lack the economic of job creation crisis, and rising the faster inactivity growth rates, of nonagricultural the contribution of labor income was driven by the nonagricultural sector mostly through wage increases (Figure earnings relative to agricultural earnings placed many agricultural workers into the bottom 40. After25 and Figure 26). Before the economic the crisis, agriculturalcrisis, the faster growth employment of nonagricultural and earnings earnings picked up, but relative to agricultural this contributed earnings little. Agricultural placed many agricultural workers into the bottom 40. 2 After the crisis, agricultural employment and earnings income picked up,was but subject to fluctuations this contributed over the period, little. Agricultural income which undermined was subject its contribution to fluctuations to welfare over the period, which improvements undermined (see below).to welfare improvements (see below). its contribution 11 Poverty and Shared Prosperity in Moldova: Progress and Prospects Figure25. Figure Decompositionof 25.Decomposition ofincome incomegrowth, growth, Figure26. Figure Decompositionof 26.Decomposition ofchanges inpoverty, changesin poverty,by by the income source, 2007–14, % points bottom the bottom 40, by income source, 2007–14,% 8% 40, by income source, 2007–14, % income source, 2007–14, % points 8% 2007-2010 2010-2014 4 2007-2010 2010-2014 4 6% 3 2007-2010 2010-2014 6% 3 2007-2010 2010-2014 2 2 4% 1 4% 1 2% 0 2% 0 Other Pensions employment earnings assistance employment Remittances earnings -1 Other Pensions employment earnings assistance employment Remittances earnings Non-agri -1 Non-agri 0% -2 0% -2 employment assistance Other Pensions Remittances employment earnings earnings employment assistance Other Pensions Remittances employment earnings earnings -3 Non-agri Non-agri Agri -2% -3 Non-agri Non-agri Agri Social -2% Social -4 Agri -4 Agri Non-agri -4% Agri -5 Non-agri -4% Agri Social -5 Social Agri -6 Agri -6% -6 -6% Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. The inactivity inactivity rate The inactivity rate rose raterose rose by 19 percentage points in 2000–14 in , , driving the decline in The byby 19 19 percentage percentage points points 2000–14 in 2000–14 driving , driving thethe decline decline in in employment employment employment rates. rates. The decline in labor force participation is explained by international migration rates. The decline inThe labordecline force in labor force participation participation is explained 9 is by explained international migrationmigration by international and the associated and the and the associated reduction associated in informalreduction reduction in informal in informal In employment. 9 employment. employment. 9In 2014, for example, 20 percent of inactive In 2014, for 2014, for example, percent 20 20example, of percent inactive ofmeninactive were abroad men were men were for searching abroad jobssearching abroad or working searching for jobs jobs or for(Figure working or 27). Early working (Figure 27). retirement (Figure 27). Early was another Early retirement retirement was was another factor another factor contributing factor to the high contributing inactivity contributing to rate. the to the high In 2014, inactivity rate. among inactive high inactivity In 2014, rate. In 2014, among menamong and womeninactive inactiveinmen men and theand women 15–65women in the age-group, 15–65 in the 15–65 age- 39 percent age- and 47 group, percent 39 percent were and group, 39 percent and 47 percent were pensioners, respectively. The sharp increase in the inactivity ages 57 pensioners,47 percent were respectively. pensioners, The sharp respectively. increase The in thesharp increase inactivity ratein the inactivity among people rate among rate and above amongwas people people ages higher 57 57 and agesthan and above was inabove previous was higher highermore years; than in than in previous thanprevious years; more years; 60 percent of people more than thanages 60 percent 60 60 and percent of above were of people ages 60 and above were inactive. Among women, family responsibilities is an important Of equal inactive. people 10 ages Among 60 and women, above were family responsibilities inactive. 10 10 Among women, is an important family reason responsibilities behind is an inactivity. important concern reason is the prevalence of underemployment, especially among those who are self-employed and behindinactivity. reasonbehind inactivity.Of equalconcern Ofequal concernis isthe theprevalence prevalenceof underemployment,especially ofunderemployment, especiallyamong among those those who work in rural areas or agriculture. This points to the possibly lower quality of jobs in these who are those who self-employed and are self-employed those who and those work in who work in rural rural areas areas or agriculture. This or agriculture. points to This points to the the areas possibly(Figure lower 28). quality of jobs in these areas (Figure 28). possibly lower quality of jobs in these areas (Figure 28). Figure27. Figure Reasonfor 27.Reason forunemployment unemploymentor orinactivity, inactivity, Figure28. Figure 28.Underemployment amongthe Underemploymentamong theemployed employed 15–65 age-group, 2014 15–65 age-group, 2014 ages 15+, 2014 ages 15+, 2014 6% 6% 70 70 6% 20% 60 20% 20% 60 47% 50 50 47% 47% 40 39% 40 39% 30 30 39% 20 22% 20 22% 1% 10 10 22% 1% 25%1% 00 18% 18% 25% 25% Female 18% Male Female Male Education Female Family p Male y responsibilities Disease Edcuation Pensioner Family responsibilities Edcuation Family responsibilities %working % lessthan workingless than40 40hours/week hours/week Voluntary inactive Disease Abroad for work Pensioner Disease Off season agriculture Other Pensioner % wishing to work morehours % wishing to work more hours Voluntaryinactive Voluntary inactive Abroad for work Abroad for work Source: World Bank 2016, using estimates based on LFS data. Source: Ronnås 2015, based on LFS data. Source: World Bank 2016, using estimates based on LFS data. Source: Ronnås 2015, based on LFS data. 9See World Bank (2014). In Moldovan labor force statistics, people who are working or searching for work abroad are 9 See World Bank (2014). In Moldovan labor force statistics, people who are working or searching for work abroad are Among considered considered those inactive.who work, the less well-off experienced an increasingly lower-quality labor market inactive. engagement. Being less well educated, the bottom 40 and the poor had fewer opportunities and less favorable 10 World Bank (2016c). 10 World Bank (2016c). outcomes in the labor market. The structure of employment was quite different for people at the top and the 24 24 the employment and unemployment rates were close in bottom of the distribution, although, in aggregate, the two groups. The bottom 40 and poor households were employed less often and self-employed more often, especially in the agricultural sector, which usually provides lower-quality jobs (Figure 29 and Figure 9 See World Bank (2014). In Moldovan labor force statistics, people who are working or searching for work abroad are considered inactive. 10 World Bank (2016c). 12 employment and unemployment rates were close in the two groups. The bottom 40 and poor households were employed less often and self-employed more often, especially in the agricultural sector, which usually provides lower-quality jobs (Figure 29 and Figure 30). Overall, the increase in self-employment among the adult population Poverty and Shared was alarming Prosperity because in Moldova: this signaled greater Progress and Prospects informality. This may have partly derived from the rigid labor market regulations in Moldova, which impose 30). Overall, strong the increase restrictions in self-employment on dismissal among the and high overtime adult population premiums. Progress was alarming is being made because this signaled by the greater informality. government, however, in aligning the labor code with European Union (EU) regulations. Figure 29. Employment status of the poor Figure 30. Employment status of the bottom 40 60% 60% 18% 21% 17% 19% 40% 40% 22% 25% 22% 25% 20% 38% 36% 20% 39% 40% 26% 24% 29% 28% 0% 0% Non-poor Poor Non-poor Poor T60 B40 T60 B40 2007 2014 2007 2014 Employee Self-employed Employee Self-employed Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. There is a lack of employment opportunities in nonfarm sectors, especially in rural areas. There is a lack of employment opportunities in nonfarm sectors, especially in rural areas. Although Although the majority of the population lives in rural areas, most manufacturing and investment the majority of the population lives in rural areas, most manufacturing and investment activities take activities place take in the place two the two in cities, large large Bălți and cities, Bălți (Figure Chișinău șinău and Chi31). (Figure The 31). The concentration concentration of of economic development economic in development the capital may servein tothe theserve capital may perpetuate to perpetuate competitive the competitive gap between firms in gap between the capital firms and in elsewhere firms the capital and representand afirms elsewhere constraint the represent on and development a constraint on the development of nonagricultural economic of activities nonagricultural outside the big cities. economicThe activities lack of economic outside the diversification big cities. Thein rural lack ofareas, combined economic with poorin diversification access rural to jobs in urban areas, areas, combined led to has with poortwo important access to jobstrends. in urban One is has areas, the led high and important to two rising rate of migration trends. One is the from highrural areas (Figure and rising32). rateOfof migration concern particular from ruralis areas the migration (Figure 32). of Of rural youth. concern particular Aroundis 23.6 the percent migration ofof rural youth ages 15–24 are working abroad, compared with only 15.7 percent in the domestic rural youth. Around 23.6 percent of rural youth ages 15–24 are working abroad, compared with only economy, which points to a lack 15.7 of attractive percent in theemployment opportunities domestic economy, which among points to a people young lack of in rural areas.employment The other important attractive 11 trend is a return to subsistence farming (see below11 ). opportunities among young people in rural areas. The other important trend is a return to subsistence farming (see below). Figure 31. Population distribution, manufacturing Figure 32. Share of employment abroad in total and investment, by location, 2014 employment, % 100% 30 80% 25 20 60% 11 Ronnås (2015). 15 40% 10 20% 25 5 All areas Rural Urban 0% 0 Population Manufacturing Investments 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 production Chisinau + Balti Rest of Moldova Source: Ronnås 2015, based on NBS data. Source: Ronnås 2015, based on LFS data. Note: Investments refer to investments in long-term tangible assets (active material). Therehas There hasbeenbeen aa large large shift shift in in employment employment out out of agriculture, of agriculture, whichwhich is not is not reflected reflected in in increases in increases other sectors;in other rather, sectors; rather, subsistence subsistence farming farming is expanding. is expanding. Since 2006, the shareSinceof 2006, workers share the in of agriculture has declined workers in from 34 to 30has agriculture percent, declined partly from because 34 to 30 of emigration percent, partly among working-age because of emigrationpeople in rural areas among (Figure 33). 12 Employment in other sectors has12 been declining, with the exception of trade and transport, working-age people in rural areas (Figure 33). Employment in other sectors has been declining, with the exception of trade and transport, which expanded only slightly (Figure 34). In contrast, the 11 Ronnås (2015). share 12 of people Employment working in agriculture hasin low-intensity experienced agriculture a rebound (less since 2012, partly asthan 20 a result hours of new a week) investment has in the been sincerising sector 2007, but it is unclear if this rebound will be sustainable. steadily, from 13 percent to 24 percent. This agricultural work is often conducted by the owners of the plots and, so, is not officially considered employment. 13Without formal employment, there is a risk these part-time agricultural workers will not be eligible for pensions and the associated benefits. There has been a large shift in employment out of agriculture, which is not reflected in increases in other sectors; rather, subsistence farming is expanding. Since 2006, the share of workers in agriculture has declined from 34 to 30 percent, partly because of emigration among Poverty and Shared Prosperity in Moldova: Progress and Prospects working-age people in rural areas (Figure 33).12 Employment in other sectors has been declining, with the which exception expanded trade and of slightly only transport, (Figure which 34). In expanded contrast, onlyof the share slightly people(Figure working 34). In contrast, the in low-intensity agriculture share of people working in low-intensity agriculture (less than 20 hours a week) has been rising (less than 20 hours a week) has been rising steadily, from 13 percent to 24 percent. This agricultural work is steadily, often percent from 13 by conducted to 24 percent. the owners This of the plots agricultural and, so, is notwork is often officially conducted considered by the owners employment. of formal Without the plots and, there employment, so, is is nota risk theseconsidered officially part-time agricultural Withoutwill employment.workers not be formal eligible forthere employment, pensions is a and the associated benefits. risk these part-time agricultural workers will not be eligible for pensions and the associated benefits. Figure 33. Employment in agriculture Figure 34. Employed populations, by sector 35% 1200 Employed people (000s) 1000 30% 800 25% 600 400 20% 200 15% 0 % of employment 2006 2007 2008 2009 2010 2011 2012 2013 2014 10% Agriculture, forestry; Fishery 5% % people working less than 20 Industry hours Construction 0% Trade; Hotels and restaurants Transportation, Communications 2006 2007 2008 2009 2010 2011 2012 2013 2014 Public administration; Education; Health and social work Other Source: World Bank calculations based on NBS and LFS data. Source: NBS data. The poor and the bottom 40 are more likely to be employed in agriculture than in services; The of thepoor and 40 bottom the bottom are employed 40inare the more likelysector agricultural to be(Figure employed35 and agriculture in Figure 36). than Given theinhigh services; this this concentration concentration has increased in recent years. Almost 80 percent of the poor has increased in recent years. Almost 80 percent of the poor and 70 percent of the bottom and 70 percent informality in agriculture relative to other sectors, this dependence on agricultural employment 40 are employed in the agricultural sector (Figure 35 and Figure 36). Given the high informality in agriculture means 12 that the Employment informality in agriculture rate among has experienced the poor a rebound and since the 2012, bottom partly 40 is as a result ofhigh. new The volatility investment and in the sector relative to other sectors, this dependence on agricultural employment means that the informality rate among seasonality since the 2007, but poor and of agriculture bottomin40 it is the unclear if this Moldova, is high. including rebound frequent will be sustainable. The volatility andclimatic shocks, of in seasonality addition to agriculture intrade bans including Moldova, by Russia frequent for certain climatic agricultural shocks, products, in addition to trade means bans that these workers suffer large fluctuations in 13 26 by Russia for certain agricultural products, means that income and are highly vulnerable. In 2013, these workers suffer large fluctuations in income and are farmers and agricultural highly workers, vulnerable. 13 In together, accounted 2013, farmers and agricultural for 40 percent of the poor. Overall, although farm income continues to make workers, together, accounted for 40 percent of the poor. Overall, although farm income continues up a significant part of to make theaincome up significant income of smallholder part of thehouseholds, of smallholder its importance households, has declined itsin importance recent years, has declined from in recent years, 30 percent from 30 percent in 2007 to 18 percent in 2013, because of the drought in in 2007 to 18 percent in 2013, because of the drought in 2007 and the global crisis in 2008–09, 2007 and the global crisis in 2008– 09, among among other other factors. factors. 14 14 Figure 35. Employment sector of the poor Figure 36. Employment sector of the bottom 40 22% 15% 22% 43% 36% 8% 26% 15% 45% 41% 10% 15% 13% 18% 18% 14% 77% 69% 64% 51% 59% 39% 37% 45% Non-poor Poor Non-poor Poor T60 B40 T60 B40 2007 2014 2007 2014 Agriculture Mining & manufacturing Services Agriculture Mining & manufacturing Services Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. Although wage Although wagegrowth growth slowed slowed during duringthe crisis, the it has crisis, been it has positive been since. positive Average since. wages Average recovered from wages zero growth in 2011 to 4 percent in 2012 and almost 6 percent in 2014. The growth was mainly in the recovered from zero growth in 2011 to 4 percent in 2012 and almost 6 percent in 2014. The growth nonagricultural sectors, which explains the larger contribution of nonagricultural labor income to income was mainly growth and in the nonagricultural poverty reduction (see Figurewhich sectors, 25 andexplains Figurethe26).larger contribution However, becauseof ofnonagricultural the concentration of the labor income to income growth and poverty reduction (see Figure poor and the bottom 40 in low-productivity agriculture and the fact that agricultural wages 25 and Figure 26). However, are persistently because of the concentration of the poor and the bottom 40 in low-productivity agriculture and the 13 The number of the employed in agriculture ranges from 200,000 during the winter to over 400,000 during the second and third quarters. that fact are There large seasonalwages agricultural also are variations persistently in the lower number of hours than worked perwages in services 2015). and industry, the gap in week (Ronnås labor 14 income Möllers providewages (including et al. (2016) andon more detail earnings conditionsfrom amongself-employment) between small farms and their impact the bottom 40 and the on poverty. top 60 has shown limited signs of narrowing in recent 14 years (Figure 37 and Figure 38). Figure 37. Average monthly salary earnings for Figure 38. Monthly labor income, bottom 40 employees, by sector and top 60 Although wage growth slowed during the crisis, it has been positive since. Average wages recovered from zero growth in 2011 to 4 percent in 2012 and almost 6 percent in 2014. The growth was mainly in the nonagricultural sectors, which explains the larger contribution of nonagricultural labor income to income growth and poverty reduction Poverty (see Figure and Shared 25 and Prosperity Figure 26). in Moldova: However, Progress and Prospects because of the concentration of the poor and the bottom 40 in low-productivity agriculture and the lower than fact that wages in services agricultural and wages are industry, the persistently in labor gapthan lower income wages (including in services wages and and industry, earnings gap in from self- the employment) between the labor income (including bottom wages 40 and the and earnings fromtop 60 has shown limited self-employment) betweensigns of narrowing the bottom 40 andin recent years the (Figure 37 and Figure 38). top 60 has shown limited signs of narrowing in recent years (Figure 37 and Figure 38). Figure 37. Average monthly salary earnings for Figure 38. Monthly labor income, bottom 40 employees, by sector and top 60 600 300 USD 2005 PPP USD 2005 PPP 250 400 200 150 200 100 50 Agriculture Industry Services B40 T60 0 0 2007 2008 2009 2010 2011 2012 2013 2014 2007 2008 2009 2010 2011 2012 2013 2014 Source: World Bank calculations based on NBS data. Source: World Bank calculations based on the HBS. Note: Average gross monthly earning represents the relation between the gross amounts for employees by the economic and social units 13 The number (remuneration Note:the ofand fund) Average employed the numbergross ofin monthly agriculture employees. earning ranges represents the relation from 200,000 during the winter to over 400,000 during the second between and third quarters. There gross theare amounts also for employees large seasonal by the variations ineconomic and of hours worked per week (Ronnås 2015). the number social units (remuneration fund) and the number of employees. 14 Möllers et al. (2016) provide more detail on conditions among small farms and their impact on poverty. More broadly, More given given the nature the semisubsistence broadly, of the farm Figure 39. Share of low-intensity agriculture semisubsistence sector, its potential to be a driver of nature of the farm sector, its potential progress 27 is limited. to be a among adults (aged 15+) Semisubsistence farming is a core component driver of progress is limited. Semisubsistence of rural 30% livelihood strategies, and this is likely to persist farming is a core component of rural livelihood in the medium 25% and longer term. Smallholder farming is prevalent in strategies, and this is likely to persist in the medium 20% agricultural work. The 2011 General Agricultural Census revealed that and more longer than term. Smallholder half the farmingless farms cultivate is prevalent than 0.5 15% inabout agricultural work. The 10% hectares, and 95 percent use an2011 thanAgricultural General area less 3 hectares. Census 5% Small family farmsrevealed produce that more 71 around than half the percent of farms total 0% output.15 less agricultural cultivate 0.5 hectares, than production Livestock is and about 95 also primarily 2007 2008 2009 2010 2011 2012 2013 2014 managed by percent use an 16 smallholders. area Most less smallholder than 3 hectares. farmsSmall are Poor B40 Non-poor T60 family Subsistence subsistence farms. farms produce farm households around accounted 71 percent for of total 74 percent of households all farm output. agricultural 15 in 2013, Livestock a rise from production 73 Source: World Bank calculations based on the HBS. is also percent in 2007. primarilySmallholder managed farms byare smallholders. more likely to switch 16 Most to subsistence farming (39 percent) smallholder farms are subsistence than to commercial farms. Subsistence farm (13 farming percent). households Adults are accounted increasingly for 74 percent of engaging in low-intensity farming (less than 20 hours a week), especially among all farm households in 2013, a rise from 73 percent in 2007. Smallholder farms are more likely to the poor and the bottom 40 (Figure 39).17 switch to subsistence farming (39 percent) than to commercial farming (13 percent). Adults are Dependence increasingly on subsistence engaging in low-intensity farming undermines farmingthe(less thanof ability hours a week), 20households enhance among to especially the poor their welfare. 17 and the bottom 40 (Figure 39). Subsistence farming is often labor intensive; the level of productivity is typically low and, in Moldova, steadily declining. Dependence This outcome is linked to on subsistence missing farming investmentsthe undermines and a lack of ability of capital and credit households availability, to enhance their which have resulted in low-yield technologies and poor use of fertilizers and pesticides. welfare. Subsistence farming is often labor intensive; the level of productivity is typically low 18 Subsistence farming and, in are often steadily households Moldova, small and woman-headed declining. This outcome andischaracterized linked to missingby lower educational investments attainment of capitaland and a lack and older household heads with health problems, which means that the members of these households have few credit availability, which have resulted in low-yield technologies and poor use of fertilizers and alternatives in the labor 18 market. Lacking a dynamic land rental system, smallholders have limited opportunities pesticides. to commercializeSubsistence andfarming increase the size of are households often their small farms. Lowand woman-headed nonfarm also characterized income and means they by lower educational attainment and older household do not have sufficient resources to mechanize or invest in inputs to raise productivity. heads with health problems, 19 which means that the members of these households have few alternatives in the labor market. Lacking a dynamic land rental system, smallholders have limited opportunities to commercialize and increase the size of their farms. Low nonfarm income also means they do not have sufficient resources to mechanize or 15 Volk et al. (2015). 16 World Bankinvest (2015a). in inputs to raise productivity.19 17 This analysis draws on Möllers et al. (2016). 18 Munoz et al. (2015). 19 Public transfers, mainly pensions, drove some improvement in living standards Möllers et al. (2016). Income growth and poverty reduction were partly driven by a pension increase. They contributed to reducing poverty by 4.8 percentage points in 2007–14. Average pensions rose 50 15 20 This partly led to an increase in the share of percent cumulatively in real terms in that period. Poverty and Shared Prosperity in Moldova: Progress and Prospects Public transfers, mainly pensions, drove some improvement in living standards Income growth and poverty reduction were partly driven by a pension increase. They contributed to pensions in total income among the poor and the bottom 40. More importantly, pensions were the reducing poverty by 4.8 percentage points in 2007–14. Average pensions rose 50 percent cumulatively in real mainin terms driver people lifting 20 that period. out of This partly ledpoverty and outin to an increase of the the bottom share 40. In particular, of pensions the pension in total income among the poor increases in 2009 and the bottom 40. More importantly, pensions were the main driver lifting people out of poverty the and 2010 generated a reduction in the share of pensioners who were among and out of the poor and bottom the 40. In bottom 40, particular, theleading pensionto increases a drop in in the and pension average 2009 among a 2010 generated these groups reduction in(Figure 40 and the share of pensioners Figure who 41); were in other among thewords, poor andthe composition the bottom 40, of those in the leading to a various drop in groups changed. the average The government pension among these groups raised 40 (Figure pensions and Figure and 41social assistance ); in other that words, supported the composition vulnerable of those categories of the groups population in the various changed. The government raised pensions and social assistance that supported vulnerable significantly, though mostly in urban areas. However, because real pensions fell slightly in 2011, categories of the population significantly, though back pensioners moved mostly into urban in the areas. bottom 40However, and among because real pensions the poor, causing fell slightly a rise in thein average 2011, pensioners moved back into the bottom pension among these groups. 40 and among the poor, causing a rise in the average pension among these groups. Figure 40. Real growth, average monthly pension, Figure 41. Real growth, average monthly % year-on-year pension, by welfare group, % year-on-year 20% 25% 20% 15% 15% 10% 10% 5% 0% 5% -5% 2008 2009 2010 2011 2012 2013 2014 -10% 0% 2008 2009 2010 2011 2012 2013 2014 Poor Non-poor -5% Bottom 40 Top 60 Source: World Bank calculations based on NBS data. Source: World Bank calculations based on the HBS. The Theexpanding expanding role pensions ofof role pensions is is reflected the in in reflected structure the ofof structure household household incomes. incomes.The share The of pensions share the inof budgets pensions inof the the poor rose budgets from of the poor23.2 percent rose to 27.7 from 23.2 percent percent during to 27.7 the period. percent A similar during the period. pattern A was observable among the bottom 40 (Figure 42 and Figure 43). Pensions are a key income source: around 28 similar pattern was observable among the bottom 40 (Figure 42 and Figure 43). Pensions are a key percent and 20 percent of total household income among the poor and the bottom 40 were accounted for by income in pensions 2014,around source: 28 percent respectively, comparedand 20 percent with of total 17 percent household among incomeand the nonpoor among the60. the top poor and The dependence on pension the bottom income 40 were was much larger accounted for byamong urban pensions in households, while the 2014, respectively, dependence compared onpercent remittances was with 17 greater among among rural households the nonpoor and the top (Figure 60. The 44 and Figure on dependence 45). pension income was much larger among urban households, while the dependence on remittances was greater among rural households (Figure Figure 44 and 42. 45).structure of the poor, % Income Figure Figure 43. Income structure of the bottom 40, % 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% Non-poor Poor Non-poor Poor T60 B40 T60 B40 2007 2014 2007 2014 Labor income Agriculture income Labor income Agriculture income Pensions Social assistance Pensions Social assistance Remittances Other Remittances Other Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. 20 Figure 44. NBS data. Income structure, poor, by area 29Figure 45. Income structure, bottom 40, by area 100% 16 100% 80% 80% 60% 60% 2007 2014 2007 2014 Labor income Agriculture income Labor income Agriculture income Pensions Social assistance Pensions Social assistance Remittances Other Remittances Other Poverty and Shared Prosperity in Moldova: Progress and Prospects Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. Figure 44. Income structure, poor, by area Figure 45. Income structure, bottom 40, by area 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% Rural poor Urban poor Rural poor Urban poor Rural Urban Rural Urban 2007 2014 2007 2014 Labor income Agriculture income Labor income Agriculture income Pensions Social assistance Pensions Social assistance Remittances Other Remittances Other Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. Socialassistance Social assistance targeted, isis targeted, but the but benefits the are are benefits small and and small insufficient to protect insufficient the poor the and to protect poor vulnerable. and vulnerable. The share of social assistance in the budgets of poor households rose from 3.6 percent in The share of social assistance in the budgets of poor households rose from 3.6 percent to 4.4 percent toSocial 2007–14. assistance 4.4 percent provided in 2007–14. a cushion Social assistance households toprovided in 2009 a cushion and 2012, when to households in 2009 total and household consumption stagnated following the contraction of GDP during the 2012, when total household consumption stagnated following the contraction of GDP during thecrisis in 2009 and during the drought of 2012. The real growth in social assistance was high among both the poor and the bottom 40 during these crisis in 2009 and during the drought of 2012. The real growth in social assistance was high among two periods. Meanwhile, social assistance did not grow among the top 60 and the nonpoor in 2012, indicating both that the poorwas targeting the bottom andeffective 46). Asthese 40 during (Figure twothe a result, share Meanwhile, periods. social assistance of social assistance did not among the almost doubled grow from poor, among 4 the top 60 percent to and the nonpoor 8 percent in 2012, in 2007–13, indicating although thatagain it fell targeting was effective in 2014 (Figure 46). declined even as consumption As a result, (Figure main of the share 47). The social programs targeted assistance almost doubled are Ajutor Social among (social the poor, aid) andfrom 4 percent a heating to 8 program, allowance percent in 2007–13, although it fell again in 2014 even as consumption declined both of which specifically address people most in need and are relatively well targeted. However, (Figure 47). Thecoverage is main not targeted wide, which programs limits the are Ajutor ability Social of the (social to programs aid) and a heating respond to sharpallowance downturns program, both of such as in in consumption whichThe 2014. the benefits level of address specifically peopleare alsoin most not and areto sufficient need providewell relatively security However, full targeted. coverage categories for the vulnerable is of the population. There is substantial room for additional support. not wide, which limits the ability of the programs to respond to sharp downturns in consumption such as in 2014. The level of the benefits are also not sufficient to provide full security for the Figure 46. Real year-on-year growth of social Figure 47. Share of social assistance in overall vulnerable categories of the population. There is substantial room for additional support. assistance, by group income, by group 50% 10% 8% 30% 6% 10% 30 4% -10% 2008 2009 2010 2011 2012 2013 2014 2% -30% 0% 2007 2008 2009 2010 2011 2012 2013 2014 -50% Poor Non-poor Poor Non-poor Bottom 40 Top 60 Bottom 40 Top 60 Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. Migration and remittances have shaped growth, poverty Migration and remittances have shaped growth, poverty reduction, and shared prosperity and shared prosperity reduction, 21 21 Migrationisis Migration substantial,and substantial, largeshare andaalarge shareof migrants are ofmigrants are labor labor migrants. migrants. Although Although precise precise data are data are lacking onlacking labor on labor migrants migrants from Moldova, from Moldova, variousvarious estimates estimates point point to a high to a high shareshare of the of the working-age population working-age looking for jobs population and working looking abroad. for jobs and abroad. to According working the LFS,to According a peak of labor the LFS, emigrants, of labor 394,500, a peak was registered emigrants, in 2005. 394,500, was 22 The results registered in of surveys 2005. 22 Theconducted by the Center results of surveys conducted of Sociological by the Center Research of and Sociological 21 draws onand Research This subsection (2016).suggest that one-quarter of the economically active population Marketing Prokhorova Labor 22was migrants working abroad in mid-2006.23 15 are defined as people ages and above who to are According looking a more or working abroad. for workInternational recent Labour Organization survey on labor force migration in Moldova, 460,000 people, 17 or 17 percent of the working-age population, were working abroad in 2012.24 The share of the economically active population involved in labor emigration grew from 8 percent to 27 percent in 2000–05. In recent years, the Poverty and Shared Prosperity in Moldova: Progress and Prospects Marketing suggest that one-quarter of the economically active population was working abroad in mid- 2006.23 According to a more recent International Labour Organization survey on labor force migration in Moldova, 460,000 people, or 17 percent of the working-age population, were working abroad in 2012.24 The share of the economically active population involved in labor emigration grew from 8 percent to 27 percent in 2000–05. In recent years, the number of people working abroad stabilized, but the economically active population continues to decrease, hence raising the share of migrants among the active population (Figure 48).25 Figure 48. Adult population (15+) who are active in Moldova or abroad, 2000–14 2,000 3,500 Thouseand people Thousand people 1,500 3,400 1,000 3,300 500 3,200 0 3,100 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Population 15+, working or looking for work abroad (left) Population 15+, active (left) Population 15+ (right) Sources: World Bank calculations based on World Population Prospects: The 2015 Revision (database), Population Division, Department of Economic and Social Affairs, United Nations, New York, http://esa.un.org/unpd/wpp/; NBS data. Migrantsare Migrants areusually menwho usuallymen who work work inin low-skilled low-skilled jobs. The profile of The profile of labor labor emigrants emigrants changed changed over the over the 2000s, 2000s, the reflecting reflecting the shifts in shifts the in the composition composition of the population. of the population. Recentshows Recent research research thatshows the average age of a typical that emigrant the average age of a typicalsubstantially, increased from 30.5–31.0 emigrant increased years substantially, from 30.5–31.0to in 2000–02 35.0–36.0 years yearsto in 2000–02 in 2010–12. According to LFS data, men dominated among emigrants, accounting for about 35.0–36.0 years in 2010–12. According to LFS data, men dominated among emigrants, accounting 67.6 percent of the total registered in 2012. The main group of migrants was represented by people employed for about 67.6 percent of the total registered in 2012. The main group of migrants was represented in low-skilled jobs (61.9 percent), by people especially employed in construction in low-skilled (56.5 jobs (61.9 percent especially percent), of total migrants). About(56.5 in construction one-third percentwere of employed as unqualified labor (32.5 percent), and only 17.9 percent were employed in services total migrants). About one-third were employed as unqualified labor (32.5 percent), and only 17.9 and trade (Table 1).26 In recent years, however, there has been an increase in the26 share of emigrants with higher educational attainment, percent were mirroring the employed improvement in services in educational In recent (Table 1). within and tradeattainment theyears, however, general there has been an population. increase in the share of emigrants with higher educational attainment, mirroring the improvement in educational Table attainment 1. Moldovan labor within the general emigrants, population. by labor sector in home and destination countries, 2012 Table 1. Moldovan labor emigrants, by labor sector in home and destination countries, 2012 Labor sector Labor sector In home In home country, country, % % In destination In destination country, % country, % Agriculture and Agriculture andforestry forestry 43.3 43.3 2.8 2.8 Mining 9.5 2.6 Mining Construction 13.1 9.5 56.5 2.6 Construction Trade and commerce 11.6 13.1 9.7 56.5 Services Trade and(hospitality commerce sector) 2.2 11.6 3.6 9.7 Transport and communication Services (hospitality sector) 5.0 2.2 3.6 3.6 Household services and assistance 0.2 18.7 Transport and communication Other activities 15.1 5.0 2.5 3.6 Household Total services and assistance 100.0 0.2 100.0 18.7 Sources: Prokhorova 2016; calculations based on Vremiș et al. 2012. Other activities 15.1 2.5 The unique geographic location of Moldova 100.0 Total has provided the population with 100.0 many Sources: Prokhorovafor opportunities 2016; calculations based migration. on Vremiș Situated et al. 2012. between the EU and Russia, many Moldovans choose these 23two emigration keyMahmoud, Lücke, (2007). Currently, in the Commonwealth of Independent States (which directions. and Pinger attracts 24 63 percent of all Moldovan emigrants), the most popular migration destination is Russia (56 ILO (2012). 25 These shares are based on population numbers estimated from the 2004 Census. The 2014 Census, which has yet to be released may percent indicate the labor ofpopulation lower migrant overall, higherstock). In the EU, share of migrants which to working ageaccounts population. for approximately 30 percent of all 26 Ibid. Moldovan emigrants, the dream migration country is Italy (19 percent), followed by Poland and 18 26 Ibid. 32 Poverty and Shared Prosperity in Moldova: Progress and Prospects The unique geographic location of Moldova has provided the population with many opportunities for migration. Situated between the EU and Russia, many Moldovans choose these two key emigration directions. Currently, in the Commonwealth of Independent States (which attracts 63 percent of all Moldovan emigrants), the most popular migration destination is Russia (56 percent of the labor migrant stock). In the EU, whichThe Romania. accounts for approximately migration outflows toward 30the percent Moldovan emigrants, of alldestinations—Russia two main andthe dream migration Italy—differ not country is Italy only (19 percent), in absolute followed numbers, but also inPoland by and Romania. gender composition andThe migration the rural or urbanoutflows toward background the two main of the destinations—Russia and Italy—differ not only in absolute numbers, but also migrants. For example, emigrants oriented toward Russia are more typically men from rural areas,in gender composition and the rural or urban background of the migrants. For example, emigrants oriented toward Russia are more typically whereas women represented 68 percent of the emigrants to Italy. While, in Italy, the typical migrant men from rural areas, whereas women represented 68 percent of the emigrants to Italy. While, in Italy, the jobs include typical migrant babysitting jobs include andbabysitting catering, foreign workers and catering, in Russia foreign workersarein mostly employed mostly in Russia are employed in 27 construction. 27 Relatively older migrants prefer the EU, while younger migrants prefer the construction. Relatively older migrants prefer the EU, while younger migrants prefer the Commonwealth 28 Independent of Commonwealth of Independent States. 28 States. Migrants Migrants to Russia areto Russia more are more likely likely to stay to stay there, thegiven there, given legalthe possibility of legal possibility of obtaining Russian citizenship, while migrants to Europe are more obtaining Russian citizenship, while migrants to Europe are more likely to return to Moldova in their likely to return to Moldova These retirement. in their retirement. differences These have differences implications forhave implications remittance for remittance behavior. behavior. in the EU Migrants working countries earn more Migrants working in and the EU send larger amounts countries earn moreof money and sendhome, largerwhile thoseof amounts money in working Russia home, send relatively while small amounts of money. 29 those working in Russia send relatively small amounts of money. 29 The migration The migration patterns resulted in patterns resulted in a surgein asurge remittances.Because inremittances. Becauseof thelimited ofthe productive capacity of limitedproductive capacity the of thewhich economy, economy, which was was reflected inreflected in low productivity low productivity and little employment creation, creation, and little employment remittance inflows remittance inflows were the primary force driving the boom in private consumption were the primary force driving the boom in private consumption and the surge in imports during and the surge in the 2000s. imports during Remittances the are 2000s. Remittances a critical are a critical source of foreign source currency of foreignthey in Moldova: currency in Moldova: are second they are and before after exports foreign direct second after investment, exports loans, and before and external foreign assistance on direct investment, the list loans, andof sources external of foreign assistance onexchange. the list In 2010– 14, remittances accounted for around 20 percent of the income growth of of sources of foreign exchange. In 2010–14, remittances accounted for around 20 percent of the bottom 40. 2006–08, they In the surpassed social of income growth protection the bottompayments 40. In households to 2006–08, through they pensions, surpassed socialchild allowances, protection compensation, payments to and other social support (Figure 49 ). households through pensions, child allowances, compensation, and other social support (Figure 49). Figure 49. Remittances as a share of monthly disposable household income, 2006–14, % 30 25 20 15 10 5 0 2006 2007 2008 2009 2010 2011 2012 2013 2014 urban rural all areas social protection payments Source: World Bank calculations based on the HBS. Theeconomy The economy was was highly dependent highly dependent remittances. on on As aAs share remittances. of GDP, a share remittance of GDP, inflows remittance tripled from 11.5 inflows percent to a peak tripled from 11.5 of 32.8 percent percent in 2000–06, to a peak in of a period of 32.8 percent substantial 2000–06, poverty a period of reduction, substantial but also declines in poverty employment. By 2014, remittances represented reduction, but also declines in employment. By 2014, remittances represented 26 percent of GDP, the Kyrgyz 30 30 26 percent of GDP, and, after Tajikistan and Republic, Moldova was and, after Tajikistan the Kyrgyz and the most remittance-dependent Republic, Moldova was country the mostinremittance-dependent the region (Figure 50country ). Overall, about a fourth of the population benefited from remittances in 2014. Among the nonpoor, 26.7 percent received in the region (Figure 50). Overall, about a fourth of the population benefited from remittances in remittances, and remittances accounted for 54.6 percent of their incomes. Although a smaller share among the poor received remittances, 14.9 percent in 2014, those who did were highly dependent on them, deriving more than half of their incomes from these flows. The share of the inflows in disposable household income was two 27 Hristev et al. (2009). 28 ILO (2012). times 29 greater among rural households receiving remittances relative to corresponding urban households. Prohnițchi and Lupușor (2013). 27 HBS data this period does not include data on household incomes, thereby curtailing the use of poverty for(2009). 30 Hristev et al. ILO (2012). and income growth exercises to determine the role of remittances more clearly. decompositions 28 29 Prohnițchi and Lupușor (2013). 30 HBS data for this period does not include data on household33 incomes, thereby curtailing the use of poverty decompositions and income growth exercises to determine the role of remittances more clearly. 19 percent of their incomes. Although a smaller share among the poor received remittances, 14.9 percent in 2014, those who did were highly dependent on them, deriving more than half of their incomes from these flows. The share of the inflows in disposable household income was two times Poverty and Shared Prosperity in Moldova: Progress and Prospects greater among rural households receiving remittances relative to corresponding urban households. Figure 50. Remittances as a share of GDP, 2014 50% 40% 30% 20% 10% 0% Source: World Bank calculations based on data of the Development Economics Prospects Group. Alongsidepensions, Alongside pensions, migration migration and andremittances remittances contributed contributed to lifting people to lifting out of people poverty. out Remittance- of poverty. receiving households were more likely to be in rural areas (56 percent). Remittances Remittance-receiving households were more likely to be in rural areas (56 percent). Remittances helped many households helped poverty escape and boost their many households welfare. escape poverty Indeed, in 2014, and boost 17.9 their percent welfare. of the nonpoor Indeed, would in 2014, 17.9 have been of poor percent had they not received remittances. Among remittance-receiving households, 60 percent would have been in the nonpoor would have been poor had they not received remittances. Among remittance-receiving the bottom quintile without remittances. Remittances helped move 75 percent of these households to higher households, income percent51). 60 (Figure groups would They alsobeen have helpedin reduce the bottom quintile without inequality—the remittances. Remittances Gini coefficient—from 0.29 to 0.22.31 helped move 75 percent of these households to higher income groups (Figure 51). They also helped reduce inequality—the However, Gini coefficient—from not all households 0.22.31 The share of emigrants in the total population 0.29 to income. benefit from remittance varies across the country. For example, the highest emigration rates were registered in southern Moldova, a However, not all households benefit from remittance income. The share of emigrants in the poorer region. Thus, in Gagauzia, up to 34 percent of the adult population works and resides abroad. The total population share is much lower in across varies the country. the northern part ofFortheexample, country.the highest temigration However, rates were he survey conducted in registered 12 rural raions in in southern Moldova, a poorer region. Thus, in Gagauzia, up to 34 percent of the 2008 by the Center of Sociological Research and Marketing found that around 25 percent of emigrant adult population cannotabroad. works and resides households count on share is much Theremittances fromlower in the 32 northern part of the country. However, abroad. Therefore, although the northern and central regions do not the survey send out conducted in 12most the raions in they ruralemigrants, 2008account by the for the majority Center of remittance of Sociological households, Research and 31 percent and 44 percent, respectively. 33 Marketing found that around 25 percent of emigrant households cannot count on remittances from abroad.32 Therefore, although the northern and central regions do not send out the most emigrants, Figure 51. Household they account welfare for the majority of ranking remittancebefore (left) and households, 31 after (right) percent remittances and 44 percent, respectively.33 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% 1 2 3 4 5 1 2 3 4 5 31 The welfare estimates with Consumption and without quintile remittances before do not take into account household remittances coping Consumption behaviors. quintile after remittances 32 Hristev et al. (2009). Non-remittance HH Remittance HH Non-remittance HH 33 Data of the 2007 European Bank for Reconstruction and Development’s survey on remittances in Moldova. Remittance HH Source: World Bank calculations based on the HBS. 34 IV. Abiding challenges and the issue of sustainability The remaining challenges that could affect the prospects for poverty reduction and shared prosperity include (1) spatial and cross-group inequalities and (2) increasing risks to sustainable progress. 31 The welfare estimates with and without remittances do not take into account household coping behaviors. 32 Hristev et al. (2009). 33 Data of the 2007 European Bank for Reconstruction and Development’s survey on remittances in Moldova. Spatial and cross-group inequalities persist 20 The poor and bottom 40 are concentrated in rural areas Moldova is a rural country, and poverty is a rural phenomenon. Around two-thirds of the population are estimated to live in Poverty and Shared Prosperity in Moldova: Progress and Prospects Abiding challenges 4. and the issue of sustainability The remaining challenges that could affect the prospects for poverty reduction and shared prosperity include (1) spatial and cross-group inequalities and (2) increasing risks to sustainable progress. Spatial and cross-group inequalities persist The poor and bottom 40 are concentrated in rural areas Moldova is a rural country, and poverty is a rural phenomenon. Around two-thirds of the population are estimated to live in rural areas.34 Most people live in the north (31.5 percent) and the center (30.5 percent). Although the share of the rural population has declined, partly driven by labor migration, the urbanization profile reveals a greater similarity with Central Asia than with Eastern Europe (Figure 52). Rural poverty stands at 19 percent compared with urban poverty, at 5 percent. Of the bottom 40, 75 percent live in rural areas, while 84 percent of the poor live in rural areas; only 7 percent and 2 percent, respectively, live in large cities (Figure 53 and Figure 54). This is consistent with a higher poverty rate in rural areas relative to cities. Part of the higher poverty rates in rural areas may be associated with the nature of income sources. Rural people are dependent on agricultural income and remittances, which are more volatile compared with other income sources, making these people more vulnerable to poverty (see below and Figure 42 and Figure 43). Figure 52. Urban population share in selected countries, % 100 80 60 40 20 0 1990 2014 2050 Sources: Prokhorova 2016; calculations based on NBS data. Residence in rural Figure 53. Type areas also of settlement, explains poor the poor’s limited and nonpoor access Figure 54. Typeto ofmarkets, jobs, settlement, and40 bottom modern and topservices 60 . The lack of road infrastructure and transport services plays a crucial role in preventing people from connecting to employment 13% and economic5%opportunities, markets, health13% care, and education;9%the extent to 27% 26% 14% 31% 34% 18% which this is a constraint on rural areas in Moldova requires more research (see below and Box 2). Thus, 17% 18% 19% 20% although a greater share of the 19% the rural population population is living in rural areas, 20% receives only a quarter 81% 73% 34 Preliminary results 70% of the 2014 census show urban areas accounting for 34.2 percent of the70% population (995,227 people), compared with 54% 65.8 percent (1,918,054 people) in rural 54% 50% estimates of the urban and areas. In June 2015, the NBS produced lower 46%rural populations: 1,507,3000 (42.4 percent) and 2,047,900 (57.6 percent), respectively. See “Demographic Situation in the Republic of Moldova in 2014,” NBS, Chișinău, http:// www.statistica.md/newsview.php?l=en&idc=168&id=4787. Non-poor Poor Non-poor Poor T60 B40 T60 B40 21 2007 2014 2007 2014 Rural Small towns Big cities Rural Small towns Big cities 40 20 0 Poverty and Shared Prosperity in Moldova: Progress and Prospects of the amount of water supplied to the urban 1990 population. 2014 2050 Similarly, urban residents receive almost four times more Sources: gas than Prokhorova the ruralbased 2016; calculations population. 35 on NBS data. Figure 53. Type of settlement, poor and nonpoor Figure 54. Type of settlement, bottom 40 and top 60 13% 5% 13% 9% 27% 26% 14% 31% 34% 17% 18% 18% 19% 20% 19% 20% 70% 81% 73% 70% 54% 54% 50% 46% Non-poor Poor Non-poor Poor T60 B40 T60 B40 2007 2014 2007 2014 Rural Small towns Big cities Rural Small towns Big cities Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. Residence in rural areas also explains the poor’s limited access to markets, jobs, and modern services. The lack of road infrastructure and transport services plays a crucial role in people from and Box 2. Transport preventing Household connecting Welfare to employment andin Moldova economic opportunities, markets, health care, and education; the extent to which this is a constraint on rural areas in Moldova requires more The literature suggests that transport infrastructure and policies—direct transport infrastructure research (see below investments, priceand Box 2). Thus, instruments, although a greater and regulations—can sharebeneficial induce of the population outcomesis living households in rural among and areas, rural growth, thefoster firms, population and reduce only receives a quarter poverty throughof the five amount of water supplied key mechanisms: to the (a) lowering urban costs transport (including population. time costs), Similarly, which promotes urban residents trade four receive almost andtimes structural change more gas in rural than the economies, localpopulation. 35 creates agglomeration effects, and leads to higher productivity; (b) improving access to input, output, and labor Box 2. Transport markets and and public Household health Welfare care and in Moldova education services, especially in remote and rural areas, which favors The literature social suggests inclusion andthat transport provides infrastructure better matches toand policies—direct the transport skills and needs infrastructure of individuals; investments, (c) cutting the prices of price instruments, and regulations—can induce beneficial outcomes among households consumption goods and services; (d) creating jobs in road construction and maintenance and and firms, foster in public growth, and reduce and private poverty services; transport through five andkey mechanisms: (a) (e) expanding lowering transport connectivity across costs (including regions time costs), and sectors to promote which promotes trade and structural change in local economies, creates agglomeration effects, and leads to mobility and enhance productive capacity.a higher productivity; (b) improving access to input, output, and labor markets and public health care and education services, especially in remote and rural areas, which favors social inclusion and provides better These beneficial effects largely depend on supportive conditions in other sectors. Thus, linking unemployed rural workers to nonfarm jobs may require training to trim the skill mismatch, and manufacturing, trade, 35 World Bank (2016b). and service development needs an enabling environment for doing business (Berg et al. 2015). The link between better transport infrastructure and poverty 36 reduction through greater farm production can be mediated by solving issues of land ownership, access to credit, trade barriers, and labor mobility (Starkey and Hine 2014). Transport policies may also affect households and locations differently, which often leads to ambiguous aggregate effects. For instance, households in a village closer to an urban center are likely to pay lower prices for consumer goods (manufacturing and services) because of lower transport costs (Emran and Hou 2013). But they may also face higher prices for housing and agricultural products because of their urban proximity. The relocation of activities from one place to another induced by changes in the transport network may have potential gains in one place and losses in the other (Berg et al. 2015). Rural residents with greater access to capital and resources, particularly more well educated individuals, are more able to adapt to changing market conditions and use new economic opportunities. For this reason, transport upgrades are less likely to benefit the chronically poor, at least in the short run (Benjamin et al. 2002; Duncan 2007; Starkey and Hine 2014). Given the large urban-rural welfare gap in Moldova, exploring how connectivity is associated with household consumption, income, and employment outcomes in rural areas is key. This may help expand the knowledge base within a multicriteria approach to prioritize road investments through a welfare and equity lens. 35 World Bank (2016b). 22 Poverty and Shared Prosperity in Moldova: Progress and Prospects This initial research focuses on a specific aspect of transport infrastructure and policies: the distance of households to roads. Using new geographic information system–based data on the distance of communities to three types of roads, local, republican, and magistral (highways), merged with data on households and individuals from the 2008–13 HBSs can help identify the links between distance to roads, household welfare, and labor market outcomes. Proper instruments to deal with the potential endogeneity of transport investments were not found; so, this paper relies on repeated cross-sections with raion- and year-fixed effects and assumes that the road network and the quality of the network rarely change within a short period. The results on the effect of distances to local, republic, and magistral roads on outcome variables among rural households and individuals are mixed and nonrobust. These outcome variables include real and spatially deflated rural household per capita consumption, labor income, agricultural income among individuals, rural household nonfarm income, the probability of poverty based on the absolute poverty line and subjective measures, an individual’s probability of employment, and the probability of employment in agriculture, conditional on being employed. The results suggest other aspects of transport infrastructure and policies should be examined, such as prices and transport services, which may matter more than distance to roads. Research is thus needed to identify how road quality, transport costs, and the market for transport services affect, especially, rural welfare. Given the heterogeneous effects of transport connectivity on labor markets and household welfare, the impact may be muted, and unpacking the household and individual benefits of enhanced road networks would be crucial. Source: Kupets, Olga (2016), Background paper, with Abla Safir, World Bank, Washington, DC. a. Berg et al. (2015) supply a recent review of the literature on transport policies and development. Starkey and Hine (2014) provide a review of existing research on the relationship between transport and poverty reduction (covering about 360 studies). The same literature warns about possible negative externalities generated by transport activities such as air pollution, road accidents, congestion, impact on health, displacement of the poor, and degradation of ecosystems. Ethnic and language minorities have less access to services and labor market opportunities36 Ethnic and language minorities face significant barriers to using public services. Moldova is an ethnically diverse country. Around one-quarter of the population is of other ethnic groups, such as Gagauzian, Romanian, Russian, and Ukrainian.37 There is also substantial diversity in the languages spoken as the first language; over a quarter of the population is not exposed to Moldovan at home. In regions where non- Moldovan groups are more prominent, such as in Gagauzia, children receive substandard instruction in the national language at school. In 2011, 10 percent of high school graduates in Gagauzia failed to achieve even a minimum passing score in their Romanian language exams, and were in jeopardy of not graduating. Part of the reason is that there are not enough qualified Romanian-speaking teachers in Gagauzia or enough teachers willing to teach in Gagauzia to help improve the quality of learning of the national language. As the government rolls out the open governance initiative through e-services (all in the national language), the barriers minority groups will face in using these services will become greater. The barriers translate into lower labor market opportunities and welfare outcomes. These gaps in education represent disparities in human capital, which undermine the capacity to obtain quality employment. Given that data on minority groups are scarce and tend to be underrepresented in national surveys, there is little information about the magnitude of the disparities. For example, the Roma, who account for a small share of the population (around 0.4 percent, though they are likely underreported), fare poorly relative to the general population on almost all welfare indicators. One in five Roma are unable to read or write, and 36 Adapted from World Bank (2016b). 37 Based on data of the 2004 census. 23 Poverty and Shared Prosperity in Moldova: Progress and Prospects about a third complete primary school only. One-third of Roma households live in an insecure dwelling, and more than 80 percent do not benefit from basic housing conditions such as the supply of potable water and sanitation facilities.38 Households with children and elderly are less well off than the rest of the population While the gender profile is about the same among the bottom 40 and the poor relative to the rest of the population, the age profile is different. There are no major differences in gender characteristics across the bottom 40, the poor, the nonpoor, and the top 60. Poverty has been declining across all age-groups in recent years, but the elderly (ages 65 and above) and children are slightly more likely to be living in poverty, and, together, they make up 40 percent of the population. Children account for a higher portion of the poor and the bottom 40 than of the more well off groups. The share of the population below working age is 19 percent among the poor compared with 16 percent among the nonpoor (Figure 55). Children are also more likely to be in the bottom 40 (Figure 56). Among the poor, the share of pensioners—the population above the upper limit of working age—is also relatively greater. Indeed, regression results show that people ages 55 and over are substantially more likely to be poor (see annex A). Nonetheless, some pensioners managed to rise to the top 60, indicated by an expansion in the share of pensioners in this group (beyond the effect of aging). Figure 55. Age composition of the poor Figure 56. Age composition of the bottom 40 18% 18% 21% 21% 16% 16% 19% 19% 16% 16% 24% 13% 13% 22% 24% 22% 64% 55% 59% 59% 53% 53% 65% 60% 60% 56% 64% 55% 65% 57% 57% 56% 18% 24% 24% 25% 25% 28% 28% 19% 19% 28% 28% 22% 22% 18% 19% 19% Non-poor Non-poor Poor Poor Non-poor Non-poor Poor Poor T60 T60 B40 B40 T60 T60 B40 B40 2007 2007 2014 2014 2007 2007 2014 2014 Retirement age age Working Retirement age Retirement age Working age Working age Children Children Retirement Working age age Children Children Source: World Source: Bank calculations World Bank calculations based based on on the the HBS. HBS. Source: World Bank Source: World based on calculations based Bank calculations the HBS. on the HBS. Poverty is Poverty is thus concentrated among thus concentrated among larger householdsand larger households andhouseholds households with withchildren children andand elderly elderly members. members. Households Households with and with children children elderlyand elderly and members members and elderly-only elderly-only households households show higher poverty higher poverty showrelative rates to the rates relative overall to the overall population (Figurepopulation 57). They seem 57). (Figure alsoThey to beseem alsosensitive to be more more to economic sensitive tosuch downturns as the2008–09 economic downturns global such as crisis, the 2012 the2008–09 drought, global and crisis, thethe 2014 2012 slowdown. drought, and theLarger 2014 households with more children slowdown. tend to be Larger households poorer, with moreand theirtend children to be in progress poverty poorer, andreduction is slower. their progress Similarly, poor in poverty households exhibited a higher child dependency ratio, 47 percent in 2014, compared reduction is slower. Similarly, poor households exhibited a higher child dependency ratio, 47 percent with 36 percent among the nonpoor. The old-age dependency ratio among poor households is also slightly in 2014, compared with 36 percent among the nonpoor. The old-age dependency ratio among poor higher (14 percent versus 11 is also slightly(higher percent, respectively) households Figure 58(14). percent versus 11 percent, respectively) (Figure 58). Figure 57. Poverty rate, by household composition Figure 58. Dependency ratio, the poor and nonpoor 50% 50% 60 60 40% 40% 50 50 30% 30% 40 40 20% 20% 30 30 10% 20 20 10% 10 10 0% 0% 2007 0 0 2008 2009 2007 2008 2010 2011 2009 2010 2012 2013 2011 2012 2013 2014 2014 Non-poor Non-poor Poor Poor Non-poor Non-poor Poor Poor HH without HH children or without children or elderly elderly HH with HH with children children Young dependency ratio Young dependency ratio Old-age Old-age dependency dependency ratio ratio HH with HH with elderly elderly 2007 2007 2014 2014 HH with children HH with and elderly children and elderly Elderly-only Elderly-only HH HH Source: World Bank Source: World calculations based Bank calculations based on on the the HBS. HBS. Source: World Source: World Bank Bank calculations based on calculations based the HBS. on the HBS. 38 Cace et al. (2007). Migration and aging affect the demographics and, potentially, poverty. The share of people 24 who live in elderly-only households rose from 9 percent to 11 percent in 2007–14, likely driven by migration. The elderly, however, are less likely than children to benefit from remittances. Families with children are more likely to receive remittances, or it may be that having children and facing a Poverty and Shared Prosperity in Moldova: Progress and Prospects Migration and aging affect the demographics and, potentially, poverty. The share of people who live in elderly-only households rose from 9 percent to 11 percent in 2007–14, likely driven by migration. The elderly, however, are less likely than children to benefit from remittances. Families with children are more likely to receive remittances, or it may be that having children and facing a higher risk of poverty are key push factors behind migration (Figure 59). The elderly benefit less from remittances. Though 25 percent of the population received remittances in 2014, only 10 percent of the people in elderly households did so. This means that the latter are highly dependent on pensions, which will become less reliable in the future, raising the risk of poverty among the elderly (see below). The poverty rate has been falling sharply among elderly day (2005 PPP). households, which 2060, Byare stillthe share among the of poorer elderly groups will have more 60). (Figure than In doubled, from 2013, half 12 the percent elderly to 27 were living on less 39 percent than of PPP). day (2005 $5.00 the population, a day By 2060, (2005 which the By PPP). will share 2060,have of important elderly the will share of implications have more elderly than will for have poverty doubled, overall. from more than 12 percent doubled, to 12 from 27 percent to 27 percent percent Figure 59. of the the population, of Composition, population, which which remittancewill havehave will important important households implications implications Figure 60. for poverty Poverty for rate, overall.overall. poverty by age 39 39 Figure 100% 59. Composition, remittance households 40% 60. Poverty rate, by age Figure 100% 80% 40% 35% 80% 60% 35% 30% 60% 40% 30% 25% 40% 20% 25% 20% 20% 0% 20% 15% 0% Non-remittance HH Remittance HH 15% 10% HH without Non-remittance HH children or elderly HH Remittance 0-14 15-24 25-34 HH with children 10% 5% 35-44 45-54 55-64 HH without HH children or elderly with elderly 0-14 65+ 15-24 25-34 HH with children 5% 0% 35-44 45-54 55-64 HH with children and elderly HH with elderly Elderly-only HH 65+ 2009 2010 2011 2012 2013 2014 2007 2008 HH with children and elderly 0% calculationsHH Elderly-only Source: World Bank based on the HBS. 2008 2007 Bank Source: World 2009 2010 calculations based2011 on the 2012 HBS. 2013 2014 Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. The poor and bottom 40: less stock of physical and human capital and less access to services The Thepoor The poor and poor and bottom havebottom 40: 40: less less accessless stock stockservices. to basic of physical of physical and Theyand less capital human human have capital access and to less and access less housing accessto services utilities, to services cold and hot The The water, poor poor sewerage, have have less less central access access gas, to basic and to basic sanitation services. services. facilities They They have have less (Figure lessto access 61 andto access housing Figure housing62). utilities, This utilities, cold andmay cold hot partly and sewerage, water, derive hot from water, the fact sewerage, that centralhouseholds gas, and at the bottom sanitation of facilitiesthe distribution (Figure 61 and frequently Figure 62).live in This rural may areas partly central gas, and sanitation facilities (Figure 61 and Figure 62). This may partly derive from the fact that households with derive at the less developed from bottom the fact of the utility services. that households distribution at the frequently bottom live of areas in rural the distribution frequently with less developed live services. utility in rural areas with Figureless 61.developed utility services. Living conditions and access to Figure 62. Living conditions and access to utilities, Figure 61. poor Living conditions and access to utilities, Figure 62. bottom Living40 conditions and access to utilities, poor utilities, bottom 40 Poor Telepho B40 Telepho ne ne ne ne Non-poor Poor Telepho T60 B40 Telepho Poor Non-poor Heating B40 T60 Heating Non-poor Poor 2007 T60 2007 Heating B40 Heating 2014 2007 2014 2007 Poor Non-poor B40 T60 Toilet inside Toilet inside 2014 2014 Non-poor Poor T60 B40 Toilet inside Toilet inside Poor Non-poor B40 T60 Sewage Sewage Non-poor Poor T60 B40 Sewage Sewage Poor Non-poor B40 T60 water water Hot Hot Hot Hot Non-poor Poor T60 B40 water water Pumped Poor Non-poor Pumped B40 T60 water water Non-poor Pumped Poor T60 Pumped B40 water water Non-poor 0% 20% 40% 60% 80% 100% T60 0% 20% 40% 60% 80% 100% Source: World Bank 0% 40% on 20% based calculations 60% 80% 100% the HBS 0% Source: World 20% 40% Bank calculations 60% based on 80% the HBS 100% Source: World Bank calculations based on the HBS Source: World Bank calculations based on the HBS 39 World Bank (2016c). 39 World Bank (2016c). 25 39 World Bank (2016c). 40 40 The poor Poverty andand Sharedbottom 40 have Prosperity fewer assets. in Moldova: ProgressFamilies at the top of the distribution live in bigger and Prospects houses. The average area of apartments and houses overall and per capita is greater among the top The poor 60 and and the bottomrelative nonpoor 40 havetofewer assets. Families the bottom 40 and the poor. at the Nonpoor top of households the distribution have live in an average bigger houses. The average area ofmore of 23 percent apartments and houses home area. In rural overall and per areas where capita nearly ishouseholds all greater amongown the topowners land, 60 and ofthe nonpoor larger relative to the bottom 40 and the poor. Nonpoor households have an average plots are less poor. The nonpoor own an average of 51 percent more farmland than the poor. of 23 percent more home area. In rural areas where nearly all households own land, owners of larger plots are less poor. The nonpoor own The an poorof average and bottom 51 percent 40 have more farmlandmuchthan less the educational poor. attainment. The average educational attainment is quite high in Moldova, as in most of former central planning economies; only 1 The poor percent and of bottom people ages4012have andmuch aboveless educational report they do not attainment. have primaryThe average educational education. attainment the is Nonetheless, quite high in Moldova, as in most of former central planning economies; only 1 percent of people ages 12 level of education varies widely across the population. The bottom 40 and the poor tend to have and above report they do not have primary education. Nonetheless, the level of education varies widely primary across theor secondary The population. education bottom at40most, and thewhile thetend poor top to and primary 60have the nonpoor more commonly or secondary educationhaveat most, tertiary education (Figure 63 and Figure 64 ). Similarly, the heads in vulnerable households while the top 60 and the nonpoor more commonly have tertiary education (Figure 63 and Figure 64). Similarly, are usually less heads the well educated (annex in vulnerable A). For example, households are usually less in people households well educated with heads (annex A). who have at least For example, peoplea in households who have with headsdegree college or university areat 27least a college percent lessor university likely to be degree poor. are This percent less 27inequality in likely to be poor. educational This inequality in educational attainment may be a result of a lack of affordability attainment may be a result of a lack of affordability in education. The poor spent 15 percent of in education. thepoor The spent 15 percent of the expenditure of the nonpoor on education in 2014, although there are more children expenditure of the nonpoor on education in 2014, although there are more children in poorer in poorer households (see above). This inequality in education significantly affects people at the bottom of households the (seeby distribution above). Thisthem rendering inequality in education less competitive onsignificantly affects people at the bottom of the the labor market. distribution by rendering them less competitive on the labor market. Figure 63. Level of education of among poor Figure 64. Level of education of among bottom 40 and nonpoor adult population and top 60 adult population 28% 27% 27% 23% 48% 44% 41% 51% 19% 21% 19% 20% 21% 20% 20% 20% 51% 53% 53% 58% 32% 35% 38% 28% Non-poor Poor Non-poor Poor T60 B40 T60 B40 2007 2014 2007 2014 Tertiary Secondary Primary Tertiary Secondary Primary Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. Peopleat People atthe the bottom bottom the of of distribution the distributionhave less less have access to health access care. Health to health outcomes care. Health are poor outcomes across are the poorcountry. the number across The country.of people The with of number various peopledegrees of disability with various or chronic degrees disease of disability oris equallydisease chronic distributed between the bottom 40 and the top 60. Self-rated health status is not necessarily is equally distributed between the bottom 40 and the top 60. Self-rated health status is worse among the not 40 bottom and the poor relative to the top 60 and the nonpoor, although the share of the population that reports they necessarily are worse experiencing among a bad the bad or very bottom life is 40 two and the times poor among greater relativethe the top to bottom 4060 and the relative nonpoor, to the top 60, and although the share of the gap is even wider among the poor relative to the nonpoor (Figure 65). Nonetheless, the poor is the population that reports they are experiencing a bad or very bad life two less spend times on greater health care.among the bottom The poor spent 27 40 relativeof percent tothe theexpenditure top 60, and of the gap the is even wider nonpoor among on health the poor in 2014. Higher spending relative toonthehealth nonpooris not always (Figure 65positive, the that but, given ). Nonetheless, poor the poor spend and less on the nonpoor health care.have The a similar poor health spent 27 percent of the expenditure of the nonpoor on health in 2014. Higher spending on health is not can profile, it does offer an indication of the quality of health services to which the poor have access and afford. One-quarter of the socially vulnerable working-age population has, for instance, no adequate access always positive, but, given that the poor and the nonpoor have a similar health profile, it does offer to health care services.40 This is partly explained by the lower coverage of medical insurance among these an indication vulnerable of the quality population of health categories; only services 70 percent to of which the poor the poor havehave access medical and cancompared afford. One- insurance, with 80 quarter of the socially vulnerable working-age percent of the nonpoor. This has important implications for the burden of out-of-pocket (OOP) to population has, for instance, no adequate access health spending (Box 3). 41 40 Molodikova (2008). 26 health care services.40 This is partly explained by the lower coverage of medical insurance among these vulnerable population categories; only 70 percent of the poor have medical insurance, compared with 80 percent of the nonpoor. This has important implications for the burden of out- of-pocket (OOP) health spending (Box 3Poverty ). and Shared Prosperity in Moldova: Progress and Prospects Figure 65. Population, by health status 80% 2014 2007 60% 40% 20% 0% Non-poor Poor Non-poor Poor Non-poor Poor Non-poor Poor Bad health Disabled Chronic disease Medical insurance Source: World Bank calculations based on the HBS. Box 3. Out-of-Pocket Health Spending The poorer segments of Moldovan society—the bottom quintile—may appear to be more well protected fromBox 3. Out-of-Pocket health-related shocks because Health they are Spending less likely to incur OOPs health fees and face catastrophic health expenditures. This conclusion would be mistaken, for two main reasons. The poorer segments of Moldovan society—the bottom quintile—may appear to be more well First, inpatient-outpatient protected from health-related service use is shocks two times because lower they are lessamong likely to the bottom incur OOPs 20 households health fees and than face among the top health catastrophic 20; indeed, it is lowestThis expenditures. among the former. conclusion While would be the inequalities mistaken, in outpatient for two service use in main reasons. 2008 were somewhat reduced in the following five years, outpatient-inpatient service utilization usage rates were still inpatient-outpatient First, strongly positively correlated service withuse is thetwo levels times household oflower among consumption, the bottom income, and wealth. 20 households than amongin Similarly, the 20; indeed, top according 2012, the to it Multiple is lowest Indicator among Cluster Surveys, the former. While the the bottom 20 in inequalities among women outpatient were service fouruse in 2008 times more were likelysomewhat reduced in to forgo antenatal the care following during five years, pregnancy and, outpatient-inpatient if they sought care, were service utilization 5.5 times more usage likely to rates receive were still strongly inadequate positively antenatal correlated supervision, that with the than is, less levels of household four consumption, antenatal visits. Obviously, income, such and wealth. behavior negativelySimilarly, affects in 2012, according maternal health outcomes,to the and,Multiple Indicator Cluster as a consequence, Surveys, the bottom the bottom 20 among women 20 wereamong 2.7 timeswomen more were four likely times to stay in more likely to the hospital forgo longer antenatal because postpartum care during pregnancy of delivery and, if they complications. In sought addition, care, were relative to the 5.5 times top more 20, they likely a showed to receive inadequate probability of visiting antenatal supervision, a health provider that that after birth is, less wasthan0.6 four antenatal visits. Obviously, such behavior negatively affects maternal times lower. These findings highlight that the poorest segments of the population use health services the least health outcomes, and, as a and consequence, frequently forgo the bottom treatment 20 when among inwomen were because need partly 2.7 times ofmore a lack likely to stay ininsurance of adequate the hospitaland longer other postpartum financial barriers, because which thus of delivery complications. have a negative In addition, effect on health outcomes. relative to the top 20, they showed a probability of visiting a health provider after birth that was 0.6 times lower. These findings highlight Second, that the because poorest poorer segmentshouseholds live close to of the population usethe poverty health servicesline theand least experience poorer living and frequently forgo standards, treatment even when in need low relatively partly because of a OOP health-related lack of adequatecan payments push them insurance into poverty and other financial . In figure barriers, B3.1 Pen Parade), (the thus which the smooth have a negative effectline on along health the top represents pre-OOP expenditure consumption levels, outcomes. from the poorest to the richest individuals who reported positive OOP expenditures for medical services. Consumption Second, because levels are measured poorer on the vertical households axis as live close tomultiples the poverty of theline poverty and line. The Pen poorer Parade shows experience living that approximately 8 standards, even relatively low health-related OOP payments can push them into poverty. Inline percent of those who reported medical expenses were living below the poverty figurein 2013. B3.1 The (the vertical lines show Pen Parade), the amounts the smooth of OOP line along expenditures the top represents for medicalexpenditure pre-OOP services reported by the consumption individuals levels, from interviewed. the poorest The lines to theshow that the richest net consumption individuals who reportedof medical services positive OOPwere below the pre- expenditures for OOP consumption medical services. line for all households. Consumption levels areSome people who measured on thestarted off above vertical axis asthe poverty line multiples in the of the first poverty and second line. The poorest quintiles Pen Parade fell below shows the poverty line8as that approximately a result of percent of medical those who reported Some expenditures. medicalpeople who expenses werewere living already below below the the poverty poverty lineline had in to2013. pay forThe vertical medical lines show services and, as the a amounts result, fellof OOP expenditures deeper into poverty. OOP medical services for payments are most reported prevalent by among the individuals the richest population The interviewed. lines show segments. Higher that the net OOP consumption spending by the of medical services were below the pre-OOP consumption line for all households. Some people who started off above the poverty line in the first and second poorest quintiles fell below the poverty line 40 Molodikova (2008). as a result of medical expenditures. Some people who were already below the poverty line had to pay for medical services and, as a result, fell deeper 42 into poverty. OOP payments are most prevalent among the richest population segments. Higher OOP spending by the rich signals that people are using higher incomes to purchase better access and higher-quality services. 27 Poverty and Shared Prosperity in Moldova: Progress and Prospects rich signals that people are using higher incomes to purchase better access and higher-quality services. Figure B3.1. Effects of OOP health payments on household consumption, 2013 Figure B3.1. Effects of OOP health payments on household consumption, 2013 8 pre-OOP consumption post-OOP consumption 6 smultipleofPL 4 Consumptiona 2 0 0 .2 .4 .6 .8 1 Cumulative proportion of population, ranked from poorest to richest Source: Data of the 2007–13 HBS. Source: Data of the 2007–13 HBS. Moreover, the health care financing system is failing to provide protection from catastrophic health payments the Moreover, health and, care financing consequently, fromsystem is failing to provide impoverishment protection mainly among from rural catastrophic residents health payments and residents of the northern and southern regions. Indeed, in 2007–13, especially regional inequalities, although inequalities also consequently, and, exist between from impoverishment urban mainly among and rural locations, widened,rural andresidents and residents geographical of the equity gaps northern became and more southern regions. pronounced. Indeed, Along within health 2007–13, especially financing regional issues, inequalities, these although inequalities may inequalities also of arise because exist between a lack of information and awareness or because of the inadequacy of a supply network and other supply-side factors. urban and rural locations, widened, and geographical equity gaps became more pronounced. Along with health financing Source: issues, Adapted from Worldthese Bank inequalities 2015b. may arise because of a lack of information and awareness or because of the inadequacy of a supply network and other supply-side factors. Source: Adapted Alternative of Bank from World measures 2015b. poverty focusing on nonmonetary indicators show that the incidence of poverty is much greater and more persistent than measured by monetary indicators. Although Moldova relies on monetary consumption–based measures, using nonmonetary welfare measures can provide additional Alternative information onmeasures of poverty trends in living focusing standards, on in particularly a country in indicators nonmonetary show have which remittances that fueled the of poverty incidence growth. consumption This is much initial greater attempt and more persistent at a multidimensional povertythan measured measure by monetary for Moldova involves the construction of an index based on several dimensions of well-being to explore whether and the extent to which indicators. individuals households Moldova and Although relies in face deprivations on thesemonetary consumption–based areas.41 The measures, dimensions considered in the indexusing are weighted equally and relate to health, education, labor, and housing following the international literature nonmonetary welfare measures can provide additional information on trends in living standards, (Table 2). Each dimension is assessed through indicators chosen to reflect the material or human capital particularly that deprivations in a undermine country in which capacityhave remittances a person’s fueled to enjoy consumption decent growth. social and Thisliving initial standards. economic attempt Individuals are considered multidimensionally poor if they live in households deprived in more than one- multidimensional at a of third Other for poverty measure the weighted indicators. Moldova indicators, dimensions, involves the construction and thresholdsof an index could based onin be explored several dimensions alternative ways to define well-being of the to explore multidimensional whether and the extent to which individuals and measure. households face deprivations in these areas.41 The dimensions considered in the index are weighted equally and relate to health, education, labor, and housing following the international literature (Table 2). Each dimension is assessed through indicators chosen to reflect the material or human 41 Using the Alkire and Foster (2011) methodology. 28 capital deprivations that undermine a person’s capacity to enjoy decent social and economic living standards. Individuals are considered multidimensionally poor if they live in households deprived in Poverty and Shared Prosperity in Moldova: Progress and Prospects Table 2. Multidimensional poverty index: dimensions and indicators Weight Dimension Indicator Description 1/4 Health Bad health At least one household member reports bad to very bad health status No insurance At least one household member is without health insurance 1/4 Education Low education, working age There is no working age—15–56 among women and 15–62 Weight Dimension Indicator among men—householdDescription member with secondary education or 1/4 Health Bad health more member reports bad to very bad health status At least one household No insurance Behind in compulsory At school- At least member least one household one child is ages without health 12–15 hasinsurance not completed primary school 1/4 Education Low education, ing, schoolworking age There is no working or, agesage—15–56 15–18, has among women and not completed 15–62 lower among school secondary age men—household member with secondary education or more 1/4 Employment Labor force participation, Active ratio is less than 0.5 among working-age members Behind in compulsory At least one child ages 12–15 has not completed primary school or, working age schooling, school age ages 15–18, has (15–56 among women not completed and 15–62 lower secondary among men) school 1/4 Employment LaborUnemployment and Active force participation, underem- than 75 At least ratio is less 0.5 percent of active working among working-age membersage members (15–56 (15–56 ployment, working age among womenamong working age and 15–62womenamong and 15–62 among men) are unemployed or men) Unemployment and At least 75 percent of active working underemployed less than(15–56 age members (working among 40 hours a week but want to underemployment, women and 15–62 work among more) men) are unemployed or underemployed working age (working less than 40 hours a week but want to work more) 1/4 Housing House material The house is not made of brick or stone 1/4 Housing House material The house is not made of brick or stone Living Living space space Living space is Living space is insufficient insufficient deprivation ToiletToilet deprivation There is no indoor toilet There is no facility indoor toilet facility Heating There is no heating, or heating is provided by coal, wood stove, or Heating There is no heating, or heating is provided by coal, wood stove, other solid materials Sewerage Household lacks other or any solid access tomaterials a sewerage system WaterSewerage Household lacks a water connection Household lacks any access to a sewerage system Water Household lacks a water connection The multidimensional poverty index reveals that 24 percent of the population was poor in 2014 The . This represented small multidimensional improvement poverty fromthat index reveals the 31 24 percent percentin 2007. of In this period, the population was those poor who in 2014. This remained multidimensionally poor were deprived in more dimensions of the represented small improvement from the 31 percent in 2007. In this period, those who remained index (higher intensity), multidimensionally crisis deprived poor were especially during the global in more and the 2012 drought. dimensions 42 of the index If multidimensional (higher poverty intensity), is defined by a especially stricterthe during crisis global that criterion, is,and the 2012that households drought. 42 If multidimensional have deprivations in more thanpoverty is defined 50 percent byweighted of the a stricter criterion, index indicators, the result might be considered a measure of severe poverty. The shareindicators, that is, households that have deprivations in more than 50 percent of the weighted index of the result might be considered a measure of severe poverty. The share households living in severe poverty remained stable at around 8 percent (Figure 66). of households living in severe poverty remained stable at around 8 percent (Figure 66). Similar to monetary poverty, multidimensional poverty tends to be concentrated in rural Similar areas, andto monetary the rate is poverty, multidimensional much higher in areas outside poverty Chiș inău. to tends be concentrated Nonetheless, althoughin rural areas, and the monetary rate is much higher in areas outside Chi ș in ă u. Nonetheless, although monetary poverty poverty is most extensive in the south, the region has made significant progress in multidimensional extensive is most in the south, poverty. the region Almost has made 100,000 significant people progress out have moved in multidimensional of multidimensional poverty. Almost 100,000 poverty. The people have moved out of multidimensional poverty. The multidimensional poverty multidimensional poverty rate is now lower in the south than in the north and central regions rate is now lower in the south than 67).north and central regions (Figure 67). in the (Figure Figure 66. Nonmonetary indicators Figure 67. Nonmonetary poor, by region 40% 50% 500,000 50% 30% 48% 400,000 40% 46% 20% 300,000 30% 44% 10% 42% 200,000 20% 0% 40% 100,000 10% 20072008200920102011201220132014 - 0% Non-monetary poverty North Central South Chisinau Severe non-monetary poor Number of poor (2007) Number of poor (2014) Intensity (right axis) Poverty rate (2007) Poverty rate (2014) Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. 42 Intensity is the weighted average of the deprivations in the index. 42 Intensity is the weighted average of the deprivations in the index. 29 44 Deprivations Poverty in health and Shared and housing Prosperity conditions in Moldova: Progress are and Prospects high. More than 60 percent of particularly Deprivations the populationin health faced and housing deprivations conditions in these are particularly indicators high. More during the period than 60 considered percent (Figure of 68). Deprivations However, in health there faced the population housing deprivations have been and small advancesconditions in these are in indicators particularly during the these dimensions high. as households More period considered than have more 60 (Figure percent 68). access to of the population However, health faced there insurance, deprivations have small beenwater, pumped these in advances and indicators in these sewerage, andduring the period dimensions as households their contributionconsidered (Figure have more 68). to multidimensional However, access to poverty there have been health has small insurance, decreased advances pumped (Figure in these dimensions as households have more access to health insurance, 69). water, and sewerage, and their contribution to multidimensional poverty pumped has decreased (Figure 69). their contribution to multidimensional poverty has decreased (Figure 69). water, and sewerage, and Figure 68. Share of people deprived in each Figure 69. Contribution of each dimension to dimension Figure 68. Share of people deprived in each multidimensional poverty Figure 69. Contribution of each dimension to dimension 80% multidimensional poverty 2007 2014 14% 17% 80% 7% 10% 60% 14% 17% 2007 2014 7% 36% 60% 40% 10% 35% 36% 35% 40% 20% 43% 38% 20% 0% 43% 38% 0% 2007 2014 2007 Health Housing Education 2014 Employment Housing Health Education Employment Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. The majority of the multidimensionally poor are not poor according to the consumption- The majority based povertyof the multidimensionally measure. The poor identified poor thenot byare poor according nonmonetary to the andconsumption- indicators the monetary The majority of the multidimensionally poor are not poor according to the consumption-based poverty based poverty indicators were measure. identical in The only poor 12.7 identified percent of by the the nonmonetary possible cases in indicators 2007, measure. The poor identified by the nonmonetary indicators and the monetary indicators and, and given the monetary that were identical in indicators 12.7declined poverty only were identical percent in only significantly, of the possible 12.7 this percent overlap cases of the in 2007, possible represented and, only given cases in 2007, 5.5 monetary percent that ofand, given cases poverty in that monetary 2014. declined At least significantly, this poverty half the declined reduction significantly, in monetary this overlap poverty aroserepresented among only people 5.5 who percent were not of cases in 2014. multidimensionally overlap represented only 5.5 percent of cases in 2014. At least half the reduction in monetary poverty arose At least poor, half the whereas among reduction the share people whoin ofmonetary were those poverty who werearose not multidimensionally among poor,people multidimensionally whereaswho poor, butnot were the share not multidimensionally poor of those by were who poor, the monetary multidimensionally poor, but whereas was measure not poor by the constant share of (Figure the monetary those who measure werepersistence 70). This was constant multidimensionally (Figure of multidimensional 70). poor, but not This poor poverty persistence by the reflects of multidimensional monetary long-lasting measurereflects poverty deprivations long-lasting was in constant living (Figuredeprivations standards 70). This that may in living persistence stem fromstandards of a lack ofthat multidimensionalmay access tostem poverty from markets a lack reflects and of access for to markets long-lasting services, and services, deprivations example, for example, in living education, education, standards health, that may and other health, stemrather services, and other a lack from than services, of access a lack rather of income. than There and to markets a lack of income. services, for is, nonetheless, a There is, nonetheless, example, strong a strong education, correlation correlation health, between and between other multidimensional services, multidimensional rather povertythan and a lackpoverty of income. rankings ofand rankings subjective of subjective There is,well-being nonetheless, anda well-being and consumption, strong correlation consumption, whichwhich between suggests suggests that richer multidimensional households poverty that richer households and and and people who of rankings people who subjective enjoy enjoy higher living well-being higher living and standards standards do appreciate their advantages in life (Figure 71). consumption, do which appreciate their suggests that advantages richer in life households (Figure 71). and people who enjoy higher living standards do appreciate their advantages in life (Figure 71). Figure 70. Overlap, nonmonetary and monetary Figure 71. Nonmonetary poverty, by subjective Figure 70. Overlap, nonmonetary and monetary well-being poor and consumption Figure 71. Nonmonetary quintiles poverty, by subjective poor well-being 50% and consumption quintiles 56% 40% 50% 70% 56% 30% 40% 70% 13% 20% 30% 19% 6% 13% 18% 10% 20% 13% 19% 6% 6% 18% 0% 10% 13% 2007 2014 6% Bad 2 3 4 Good 0% Poor 2007(both) 2014 Bad 2 Quintile 3 4 Good Non-monetary poor, monetary non-poor Poor (both) Quintile Monetary poor, non-monetary non-poor Non-monetary poor, monetary non-poor Subjective wellbeing Consumption per capita Non-poor (both) Monetary poor, non-monetary non-poor Source: WorldNon-poor (both) based on the HBS. Bank calculations Subjective Source: World Bank wellbeing Consumption calculations based per capita on the HBS. Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. 45 to sustainable progress persist Important and increasing risks 45 Fiscal pressure because of aging and weak labor markets will limit the role of public transfers The population of Moldova is shrinking and aging rapidly. Driven by declining fertility and accelerating emigration among the young population, the population shrunk by 16 percent in 2000–15. This is equivalent to a reduction of around 670,000 people. The share of the working-age population (ages 15–64) rose from 66 percent to 71 percent during the period, creating a demographic dividend for the economy. This occurred 30 The population of Moldova is shrinking and aging rapidly. Driven by declining fertility and accelerating emigration among the young population, the population shrunk by 16 percent in 2000– 15. This is equivalent to a reduction of around 670,000 people. The share of the working-age Poverty and Shared Prosperity in Moldova: Progress and Prospects population (ages 15–64) rose from 66 percent to 71 percent during the period, creating a demographic dividend for the economy. This occurred mainly because of a rather rapid decline in mainly because of a rather rapid decline in the share of the population below working age, from 24 percent the in share 2000 to the of17 population percent belowThe in 2015. working shareage, from of the 24 percent above population in 2000 to 17 percent working in 2015. age (ages The over) rose 65 and share of the moderately, from 10 percent in 2000 to 12 percent in 2015 (Figure 72). As a result, both the childin population above working age (ages 65 and over) rose moderately, from 10 percent dependency 2000 to 12 percent in 2015 (Figure 72). As a result, both the child dependency ratio and ratio and the total dependency ratio declined during the period, while the old-age dependency ratio rose the total dependency slightly, ratio declined providing favorableduring the period, demographic while the for conditions old-age dependency economic ratio growth. rose slightly, However, the situation is providingeventually expected favorable to reverse as large demographic cohorts approach conditions for economicretirement growth. age. By 2030, the However, the share of the situation iselderly will have increased rapidly to 17 percent, more than the share of children. expected eventually to reverse as large cohorts approach retirement age. By 2030, the share of the elderly will have increased rapidly to 17 percent, more than the share of children. Figure 72. Demographic composition, 2000–30 5,000 Children Working age Elderly 4,500  Children Working age Elderly 4,000  4,000 [CELLRANGE] 24% Thousand people 3,500  Thousand people 17% 3,000  3,000 [CELLRANGE] 15% 2,500  2,000  66% [CELLRANGE] Thousan [CELLRANGE] 71 % 1,500  1,000  2,000 [CELLRANGE] 67% 500  10% [CELLRANGE] 12% ‐ 1,000 17% [CELLRANGE] 2000 [CELLRANGE] 2015 2030 - [CELLRANGE] 2000 2015 2030 Source: World Bank calculations based on World Population Prospects: The 2015 Revision (database), Population Division, Department of Economic and Social Affairs, United Nations, New York, http://esa.un.org/unpd/wpp/. Note: The population forecast is based on the medium fertility, normal mortality, and normal migration scenario, which assumes fertility rates follow a trend from high to low, then fluctuate around the replacement rate (2.1 children per woman), while life expectancy at birth and migration follow historical trends in each country. The estimates do not include Transnistria. Ifthese If these trends trends continue, continue, Moldova Moldova willstark will face stark demographic facedemographic challengeschallenges affecting in growth affecting prospects sharedin prospects and growth and prosperity. shared The prosperity. population The population is expected to shrink isbyexpected another to 25 shrink another by or percent, 25 1.2 million people, by Because 2060. or percent, people, by will the reduction 1.2 million 2060.be concentrated Because among the reduction the will beyoung population, concentrated among Moldovan the youngsociety will also be much older (Figure 73). The share of the elderly is anticipated to increase population, Moldovan society will also be much older (Figure 73). The share of the elderly from 12 percent is to 30 percent in 2015–60, anticipated to increaseand fromwomen 60 ages to 12 percent and 30 70 will percent account and in 2015–60, for women segment a majorages 60 and of70the population. will account for a major segment of the population. Accelerating population aging is likely to putand shared Accelerating population aging is likely to put pressure on the progress in economic growth prosperity. pressure on the progress in economic growth and shared prosperity. Figure 73. Age-gender pyramid, 2015–60  Male Female 46 People, 1,000s Source: World Bank calculations based on World Population Prospects: The 2012 Revision (database), Population Division, Department of Economic and Social Affairs, United Nations, New York, http://esa.un.org/unpd/wpp/. Note: The population forecast is based on the medium fertility, normal mortality, and normal migration scenario, which assumes fertility rates follow a trend from high to low, then fluctuate around the replacement rate (2.1 children per woman), while life expectancy at birth and migration follow historical trends in each country. The estimates do not include Transnistria. Low labor force participation and a shrinking working-age population have led to a large 31 dependency ratio. Moldova has one of the lowest employment rates in the region. It is third after Bosnia and Herzegovina and Kosovo. This is driven by a persistent fall in labor force participation among both men and women (see above). As a result, the country has an exceptionally high People, 1,000s Source: World Bank calculations based on World Population Prospects: The 2012 Revision (database), Population Division, Department of Economic and Social Affairs, United Nations, New York, http://esa.un.org/unpd/wpp/. Note: The population forecast is based on the medium fertility, normal mortality, and normal migration scenario, which assumes fertility rates follow a trend from high to low, then fluctuate around the replacement rate (2.1 children per woman), while life Poverty and Shared Prosperity in Moldova: Progress and Prospects expectancy at birth and migration follow historical trends in each country. The estimates do not include Transnistria. Low labor Low force participation labor force participation and a shrinking and a shrinking working-age working-agepopulation population have ledto haveled toaalarge largedependency ratio. Moldova dependency has one ratio. of the Moldova lowest has one employment of the lowest rates in the region. employment It is rates in thethird after region. ItBosnia is thirdand Herzegovina after and Kosovo. This is driven by a persistent fall in labor force participation among both men and women (see Bosnia and Herzegovina and Kosovo. This is driven by a persistent fall in labor force participation above). As a result, the country has an exceptionally high dependency ratio. On average, there are 1.6 inactive both among for adults each men andadult active women (see74). above). As a result, the country has an exceptionally high (Figure dependency ratio. On average, there are 1.6 inactive adults for each active adult (Figure 74). Figure 74. Adult (15+) dependency ratio: inactive population relative to active population, 2013 1.6 1.6 1.4 1.4 1.2 1.2 1.0 1.0 0.8 0.6 0.8 0.4 0.6 0.2 0.4 0.0 Russian… United… Bosnia and… 0.2 France Hungary Italy Denmark Portugal Austria Slovenia Kyrgyzstan Estonia Romania Serbia Kazakhstan Sweden Tajikistan Cyprus Uzbekistan Bulgaria Malta Croatia Turkmenistan Latvia Moldova Germany Spain Greece Turkey Czech Republic Belarus Belgium Montenegro Netherlands Georgia Armenia Lithuania Ireland Finland Slovakia Poland Ukraine Luxembourg Azerbaijan Albania Macedonia 0.0 Source: Bussolo, Koettl, and Sinnott 2015. demographic aging Given demographic agingandandthe labor the market labor marketsituation, sustaining situation, sustaining pension the the system pension involves a high system fiscal burden. involves a high Pension fiscal spending has risen burden. Pension substantially. spending Thesubstantially. has risen cost of pensions The jumped from 5 percent to 8 cost of pensions jumped from 5 percent to 8 percent of GDP in 2002–14, which fueled the fiscal deficit of upThe percent of GDP in 2002–14, which fueled the fiscal deficit of up to 1 percent of GDP. to 1spending is percent ofgreater relatively GDP. Thethanspending in manyis relatively countries European greater thanthat in are manymore advanced European in economic countries that aredevelopment more and characterized by fewer problems in population aging, such as Croatia, Estonia, and the Slovak Republic.43 advanced in economic development and characterized by fewer problems in population aging, such shrinkingof Because share of the the rising population, working-age pensionand theelderly shrinking system dependencyworking-age ratio—thepopulation, the pension system number of pensioners as Croatia, Estonia, and the Slovak Republic.43 Because of the rising share of the elderly and dependency ratio—the per contributor—is number expected toof pensioners increase fromper77 contributor—is percent to a peak expected of 108 increase topercent infrom 77 percent to a 2014–55 (Figure 75). Labor migrants do not tend to contribute to the pension system, exacerbating the system, peak of 108 percent in 2014–55 (Figure 75). Labor migrants do not tend to contribute to the pension exacerbating decreasing 43 the decreasing pension contribution rates. Given that inactivity and informality are expanding, (2016c). contribution rates. Given that inactivity and informality are expanding, fewer World Bank pension fewer people will contribute people will contribute to the to the pension pension system system and and thusundermining thus undermining systemsustainability sustainability system and reducing and pension coverage of the retiree population (Figure 47 76). reducing pension coverage of the retiree population (Figure 76). Figure 75. Pension system dependency ratio, Figure 76. Pension coverage, population above number of pensioners per contributor standard retirement age 110% 100% 90% 100% 80% 90% 70% 80% 60% 50% 70% 40% 60% 30% 2014 2018 2022 2026 2030 2034 2038 2042 2046 2050 2054 2058 2062 2066 2070 2014 2018 2022 2026 2030 2034 2038 2042 2046 2050 2054 2058 2062 2066 2070 Sources: World Bank 2016c; calculations based on PROST (Pension Reform Options Simulations Toolkit), World Bank; data of the National Social Insurance House. The The declining declining pension pension coverage coverage andand replacement replacement ratesrates may may increase increase the the risk ofrisk of old-age old-age poverty. Pensions play Pensions role an important poverty. an reducing play in importantpoverty. particularly They are poverty. role in reducing They important in reducing are particularly old-age important in poverty. reducing old-age poverty. Without pensions, the poverty rate in Moldova would have increased by a a factor of Without pensions, the poverty rate in Moldova would have increased by a factor of 11 overall and 14 among factor of 11the elderly. overall and Pension 44 replacement a factor of rates—the 14 among the ratio elderly.44 of old-age Pension pensions replacement to the average rates—the ratio of wage—are already low and are on a declining trend because of shrinking contributions old-age pensions to the average wage—are already low and are on a declining trend because (Figure 77 andof Figure 78). shrinking 43 contributions World Bank (2016c). (Figure 77 and Figure 78). Addressing this challenge requires broader reforms 44 World Bank (2016c). and cannot rely on the government budget to bridge the gap between contributions and benefits. 32 Figure 77. Replacement rates, selected countries Figure 78. Replacement rates, 2013–2069 70 30% 60 The declining pension coverage and replacement rates may increase the risk of old-age poverty. Pensions play an important role in reducing poverty. They are particularly important in reducing old-age poverty. Without pensions, the poverty rate in Moldova would have increased by a 44 factor of 11 overall and a factor of 14 among the elderly. Poverty Pension and Shared replacement Prosperity rates—the in Moldova: ratio and Progress of Prospects old-age pensions to the average wage—are already low and are on a declining trend because of Addressing this challenge shrinking contributions requires (Figure broader 77 and Figure reforms and cannot 78). Addressing rely on the this challenge government requires budget broader reformsto bridge the andbetween gap contributions cannot rely and benefits. on the government budget to bridge the gap between contributions and benefits. Figure 77. Replacement rates, selected countries Figure 78. Replacement rates, 2013–2069 70 30% 60 50 25% 40 20% 30 20 15% 10 10% 0 5% 0% 2013 2017 2021 2025 2029 2033 2037 2041 2045 2049 2053 2057 2061 2065 2069 2011 2030 Source: World Bank 2016c. Sources: World Bank 2016c; calculations based on PROST (Pension Reform Options Simulations Toolkit), World Bank; social reports. Remittances: unsustainable drivers of poverty reduction and shared prosperity Remittances are an unsustainable driver of Moldova’s economic growth. Remittance income 44 World Bank (2016c). 45 Remittances: unsustainable drivers of poverty reduction and shared prosperity can be highly volatile, and this especially affects48 households that are highly dependent on them. A Remittances are an unsustainable driver of Moldova’s large share of the remittances received in Moldova are sent economic from the EU growth. 45 Remittance and Russia. income can be Remittances highly volatile, and this especially affects households that are highly dependent dipped significantly during the financial crisis of 2008–09 and have been decreasing since 2013 on them. A large share of the remittances received in Moldova are sent from the EU and Russia. Remittances dipped significantly during (Figure 79). Because of the slowdown in the economic activity of Russia beginning in 2014, the financial crisis of 2008–09 and have been decreasing since 2013 (Figure 79). Because of the slowdown in remittances to Moldova dropped by 30 percent in 2015. The biggest reduction was in the transfer of the economic activity of Russia beginning in 2014, remittances to Moldova dropped by 30 percent in 2015. rubles. The The reduction biggest slowin impact of was growth on remittances the transfer of rubles. measured The impact in U.S. dollars of slow has been growth driven by the on remittances measured in valuation U.S. dollarseffects has been the appreciation of driven of the effects by the valuation against dollar of the currencies the appreciation of dollar remittance-source of the against the currencies countries of (contributing remittance-source 13.8 percentage countries points 13.8 (contributing to the total decline percentage in transfers points in 2015), to the total especially decline the in 2015), in transfers ruble, as the especially wellruble, as byasthe well as by thein reduction actual transfers reduction in actual (16.2 percentage transfers points). Many (16.2 percentage households points). Many households depend heavily depend heavily on remittanceincome onremittance income for consumption; for consumption; so,so, lower inflows lower inflows hamper can can suchsuch hamper consumption. Less foreign currency income among households and domestic exporters consumption. Less foreign currency income among households and domestic exporters because ofbecause of declining remittances also remittances pressures, fuels inflationary declining further affecting also fuels inflationary the welfare pressures, further of households. affecting Remittance the welfare flows are not likely of households. to continue to expand at a rapid pace as in the past given the already high share of the Remittance flows are not likely to continue to expand at a rapid pace as in the past given the already population of Moldova abroad the and of high share fact the that second-generation population of Moldova abroad migrants and themay bethat fact less second-generation attached to their homemigrantscountry may and thus reduce their remittances. be less attached to their home country and thus reduce their remittances. Figure 79. Remittances to Moldova, 1995–2015 35 2500 30 2000 25 20 1500 15 1000 10 500 5 0 0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Remittances (RHS) Remittances as a share of GDP (LHS) Sources: World Bank Development Economics Prospects Group; 2015 data: Moldova Central Bank, World Economic Outlook Database, International Monetary Fund, Washington, DC, http://www.imf.org/external/pubs/ft/weo/2011/02/weodata/index.aspx. Volatility because of climate shocks is increasing, particularly among the poor and bottom 40 45 TheWorld Bank (2016b). agricultural sector, critical to the livelihoods of the rural poor, is highly volatile. 33 are increasingly driving fluctuations in Episodes of drought, the most recent in the summer of 2015, agriculture valued added, household consumption, and overall GDP (Figure 80). A majority of farm households are smallholders, who tend to be poorer and have less buffer against shocks. More than Poverty and Shared Prosperity in Moldova: Progress and Prospects Volatility because of climate shocks is increasing, particularly among the poor and bottom 40 The agricultural sector, critical to the livelihoods of the rural poor, is highly volatile. Episodes of drought, the most recent in the summer of 2015, are increasingly driving fluctuations in agriculture valued added, household consumption, and overall GDP (Figure 80). A majority of farm households are smallholders, who tend to be poorer and have less buffer against shocks. More than one-third of all farm households reported they faced difficulties in paying for the food needed to ensure decent nutrition among household members over the previous year. The numbers are higher among subsistence farm households. Subsistence farm households consume more than 99 percent of their farm production, while other smallholder farm of their farm households production, consume around while other smallholder 80 percent. This means farm thathouseholds consume there is little around 80 left for income percent. 46 generation. 46 This means that there is little left for income generation. Figure 80. Value added in agriculture and GDP growth, 2001–14 45 35 25 15 Percent 5 -5 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 -15 -25 GVA in Agriculture, y/y GDP growth, y/y Source: World Bank 2016b, using World Bank estimates based on national accounts. The The structure structure ofof incomes incomes mean mean thethe poor poor andand thethe bottom bottom 4040are are moremore vulnerable vulnerable toto various various shocks. The shocks. poor and The poor bottom 40 and bottom depend more40 than depend themore nonpoor the nonpoor than and the top 60 and onthe top 60 onincome, agricultural agricultural pensions, and social assistance. income, pensions, Theandhigh share social of agricultural assistance. The high incomes share in of overall household agricultural incomes income—19.2 percent among the in overall household poor percentamong and 22.9 percent income—19.2 amongthe thebottom poor and40—make these incomes 22.9 percent among more and vulnerable volatile 40—make the bottom theseto external factors, incomes such as weather more volatile andconditions vulnerableand food to pricefactors, external fluctuations. such asAmong weather farm households, conditions the average price nonpoor and food fluctuations. Among farm households, the average nonpoor household has 2.8 sources of income, and many household has 2.8 sources of income, while the average poor household has 2.3 sources of income, poor while the averagerely households poor almost exclusively household has 2.3 on one income sources source. of income, andThemanyrelative households of poorimportance the rely various income almost sources in total income is revealed through the application of a Hirschman-Herfindahl index.47 Across all exclusively on one income source. The relative importance of the various income sources in total years, poor households are characterized by a higher Hirschman-Herfindahl47 index value, indicating a stronger income is revealed through the application of a Hirschman-Herfindahl index. Across all years, poor concentration of income streams in fewer sources. This also reflects the lack of nonfarm opportunities in rural 48 households areas. in the riseby are characterized The slowdown in a higher Hirschman-Herfindahl nonagricultural wages index value, and the volatility indicating income, of agricultural a stronger coupled with 48 concentration of income streams in fewer sources. This also reflects limited employment growth, meant that the share of labor in income was declining among the poor and the the lack of nonfarm 40. Forin opportunities bottom rural areas. example, Thethe among slowdown poor, the inshare the rise in nonagricultural of nonagricultural wages labor incomeand the volatility in total incomeof increased agricultural income, coupled with limited employment growth, meant from 38 percent to 34 percent in 2007–14. Without fiscal support, the incomes of the poor and the that the share of labor in bottom 40 cannot income was as grow they did declining in 2007–14; among the poor indeed, and the they may even bottom 40. Forcontract. example, In contrast, among the the nonpoor poor, the shareand the top 60 have a healthier income of nonagricultural labor composition. income in total They rely increased income more on direct from 38labor income percent to 34and remittances. percent in 2007–14. Without fiscal support, the incomes of the poor and the bottom 40 cannot grow as they did in Linked 2007–14; partly to climate indeed, they may shocks, fluctuations even contract. in food In contrast, theprices nonpooraffect and the thepoor top disproportionately. 60 have a healthier In 2014, the poor spent 60 percent of their consumption expenditure income composition. They rely more on direct labor income and remittances. on food, compared with 45 percent among the nonpoor. Conversely, the nonpoor spent more of their consumption expenditure on services and durables Linked (Figure 81).partly The same climate to patterns are shocks, observed among fluctuations the bottomin food 40 and pricesthe top 60 affect (Figurethe 82). poor In recent years, the prices of food have disproportionately. Inrisen 2014, much more the poor than spent 60the pricesof percent of their services ). These changes (Figure 83expenditure consumption on food,in food prices are affecting compared the 45 with poor and the percent bottom among the40 much more, nonpoor. potentially Conversely, the squeezing nonpoor spent out spending more of on other needs, their including health care and education, that are important for long-term consumption expenditure on services and durables (Figure 81). The same patterns are observed well-being. 46 Möllers et al. (2016). 47 See Herfindahl (1950); Hirschman (1945, 1964). 48 etal. Möllerset 46 Möllers (2016). al. (2016). 47 See Herfindahl (1950); Hirschman (1945, 1964). 34 48 Möllers et al. (2016). 50 among the bottom 40 and the top 60 (Figure 82). In recent years, the prices of food have risen much more than the prices of services (Figure 83). These changes in food prices are affecting the poor and the bottom 40 much more, potentially squeezing out spending on other needs, including health care Poverty and Shared Prosperity in Moldova: Progress and Prospects and education, that are important for long-term well-being. Figure 81. Composition of the consumption of Figure 82. Composition of the consumption of the poor and nonpoor the bottom 40 and top 60 100% 100% 90% 90% 80% 80% 70% 70% 60% 60% 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% Non-poor Poor Non-poor Poor T60 B40 T60 B40 2007 2014 2007 2014 Food Clothing Housing Services Other Food Clothing Housing Services Other Source: World Bank calculations based on the HBS. Source: World Bank calculations based on the HBS. Figure 83. Consumer price indicators, year-on-year, 2007–15 20 15 10 Percent 5 0 2007 2008 2009 2010 2011 2012 2013 2014 2015 -5 Foodstuff products Non-foodstuff products Services -10 Source: World Bank calculations based on NBS data. 51 35 Poverty and Shared Prosperity in Moldova: Progress and Prospects Conclusion Moldova has made good progress in reducing poverty and boosting shared prosperity. However, the drivers of this progress in the past are likely to exercise a much more limited role. • Aside from increases in nonagricultural wages, pensions have been the biggest contributor to income growth among the bottom 40, as well as to poverty reduction. However, pensions are not the most efficient means to target the neediest. Moreover, given population aging, the shrinking workforce, and currently low employment rates and worker productivity, increases in pension spending will be fiscally unsustainable.49 • Remittances are volatile and are not likely to continue to grow at the high pace of the past given the external context and the already high share of the population living abroad. Strengthening domestic labor markets is thus becoming more critical in the effort to maintain the progress toward reaching the twin goals of reducing poverty and boosting shared prosperity and to address the problems associated with an aging population in a fiscally sustainable manner. The accompanying analysis, “A Jobs Diagnostic for Moldova,” explores in more detail the main challenges facing the labor market and the areas for eventual action. Addressing issues of governance, which have taken center stage in Moldova in recent years, is critical to creating an environment in which the necessary reforms and private sector-led job creation are possible. Because of the large share of the population, particularly the poor and the bottom 40, in rural areas and in the agricultural sector, ensuring that rural areas enjoy the proper conditions for the development of the nonfarm and farm economies is key. The accompanying analysis, “Structural Transformation of Moldovan Small-Holder Agriculture and Its Poverty and Shared Prosperity Impacts,” examines the main challenges in the agricultural sector, focusing on the nature and potential or lack of potential in smallholder agriculture. Policies to promote healthier labor markets need to take account of and address the structural challenges to the extent possible, as follows: • Aging: The country needs to prepare for the rapidly aging population. By 2060, the population is projected to have dropped by 29 percent, while the share of the elderly (ages 65+) will have tripled to 30 percent. Given the substantial migration, low fertility, and weak labor markets, the demographic dividend may vanish before the country has reaped the benefits. This raises serious questions about the ability of society and the economy to support a growing elderly population.50 Efforts to reduce old-age mortality, accompanied by policies to improve education and health care and to promote active aging, can allow people to work longer and contribute more to the economy. • Regional disparities: The gaps in welfare and access to services across urban and rural areas and among ethnic minorities call for more active engagement by the government to enhance the provision and quality of services in remote areas. Failing to address these barriers would risk widening inequality and undermining economic mobility and harmonization across the country. Improving the equitable access 49 World Bank (2016c) discusses the pension system and relevant options in depth. 50 World Bank (2016c) explores the challenges involved in aging in Moldova in depth. 36 Poverty and Shared Prosperity in Moldova: Progress and Prospects to and the quality of education, particularly among the less well off, is also key to increasing productivity and the opportunities to find jobs. Equity in access to services, including health care, without the currently high OOP expenditures is also critical to permitting individuals to build up their human capital. • Significant vulnerability: The vulnerability to external and climate shocks seems to be increasing, and this will affect more heavily those households dependent on the agricultural sector. The poor are more highly exposed to such shocks. Mitigation and measures to help households adapt to climate shocks are needed. Social assistance can also be improved. The targeting of social assistance has been enhanced through the Ajutor Social and heating allowance programs, although they still represent a small share of total spending and cover only 4 percent and 6 percent of the total population, respectively. There is likewise scope for improvements within the overall expenditure envelope, including program consolidation to provide room to expand the coverage of social assistance programs. Previous analyses of poverty and equity in Moldova point to challenges similar to those discussed in this report. More than a decade ago, a similar document identified challenges related to the lack of sufficient investment in human capital accumulation by the poor, large spatial inequalities in living standards, dependence on pensions, increasing inactivity, and dependence on subsistence agriculture, substantial migration, and the poor coverage of social assistance.51 Addressing these challenges once and for all is critical for sustainable growth, poverty reduction, and shared prosperity in coming years. 51 World Bank (2004). 37 Poverty and Shared Prosperity in Moldova: Progress and Prospects References Alkire, Sabina, and James Foster. 2011. “Counting and Multidimensional Poverty Measurement.” Journal of Public Economics 95 (7–8): 476–87. Benjamin, Dwayne, Loren Brandt, Paul Glewwe, and Guo Li. 2002. “Markets, Human Capital, and Income Inequality in an Economy in Transition: The Case of Rural China.” In Inequality around the World, edited by Richard B. Freeman. International Economic Association Series. Houndsmills, Basingstoke, United Kingdom: Palgrave-Macmillan. Berg, Claudia N., Uwe Deichmann, Yishen Liu, and Harris Selod. 2015. “Transport Policies and Development.” Policy Research Working Paper 7366, World Bank, Washington, DC. Bussolo, Maurizio, Johannes Koettl, and Emily Sinnott. 2015. Golden Aging: Prospects for Healthy, Active, and Prosperous Aging in Europe and Central Asia. Europe and Central Asia Studies Series. Washington, DC: World Bank. Cace, Sorin, Vasile Cantarji, Nicolae Sali, and Marin Alla. 2007. “The Situation of Roma in the Republic of Moldova.” United Nations Development Programme, Chișinău, Moldova. Cancho, César A., María E. Dávalos, Giorgia Demarchi, Moritz Meyer, and Carolina Sánchez-Páramo. 2015. “Economic Mobility in Europe and Central Asia: Exploring Patterns and Uncovering Puzzles.” Policy Research Working Paper 7173, World Bank, Washington, DC. Dang, Hai-Anh, Peter Lanjouw, Jill Luoto, and David McKenzie. 2011. “Using Repeated Cross-Sections to Explore Movements in and out of Poverty.” Policy Research Working Paper 5550, World Bank, Washington, DC. Datt, Gaurav, and Martin Ravallion. 1992. “Growth and Redistribution Components of Changes in Poverty Measures: A Decomposition with Applications to Brazil and India in the 1980s.” Journal of Development Economics 38 (2): 275–95. Dávalos, María E., and Moritz Meyer. 2015. “Moldova: A Story of Upward Economic Mobility.” Policy Research Working Paper 7167, World Bank, Washington, DC. Duncan, Tyrrell. 2007. “Findings from Studies of Poverty Impacts of Road Projects.” Asian Development Bank, Manila. http://www.adb.org/sites/default/files/evaluation-document/35848/files/sape-prc-impacts- road-projects.pdf. Emran, M. Shahe, and Zhaoyang Hou. 2013. “Access to Markets and Rural Poverty: Evidence from Household Consumption in China.” Review of Economics and Statistics 95 (2), 682–97. Foster, James, Joel Greer, and Erik Thorbecke. 1984. “A Class of Decomposable Poverty Measures.” Econometrica 52 (3): 761–66. Herfindahl, Orris C. 1950. “Concentration in the U.S. Steel Industry.” PhD dissertation, Columbia University, New York. Hirschman, Albert O. 1945. National Power and the Structure of Foreign Trade. Berkeley, CA: University of California Press ———. 1964. “The Paternity of an Index.” American Economic Review 54 (5): 761–62. 38 Poverty and Shared Prosperity in Moldova: Progress and Prospects Hristev, Eugene, Georgeta Mincu, Maya Sandu, and Mateusz Walewski. 2009. “The Effects of Migration and Remittances in Rural Moldova.” CASE Network Studies and Analyses 389, CASE-Center for Social and Economic Research, Warsaw. ILO (International Labour Organization). 2012. “Moldova and Ukraine: Effective Governance of Labour Migration and Its Skill Dimensions.” ILO, Geneva. Kupets, Olga. 2016. “Road connectivity, welfare and labor market outcomes in rural Moldova.” Background paper for the Moldova Poverty Assessment, World Bank, Washington, DC. Lücke, Matthias, Toman Omar Mahmoud, and Pia Pinger. 2007. “Patterns and Trends of Migration and Remittances in Moldova.” June, International Organization for Migration, Chișinău, Moldova. Möllers, Judith, Thomas Herzfeld, Simone Piras, and Axel Wolz. 2016. “Structural Transformation of Moldovan Small-Holder Agriculture and Its Poverty and Shared Prosperity Impacts.” Background paper for the Moldova Poverty Assessment, World Bank, Washington, DC. Molodikova, Irina. 2008. “Trends in the Field of Social Policies and Welfare Reforms in Ukraine and Moldova.” Background Report 3, Centro Studi di Politica Internazionale, Rome. Moroz, Victor, Alexandru Stratan, Anatolie Ignat, Eugenia Lucasenco. 2015. “Country Report: Republic of Moldova.” Agricistrade research report (March), National Institute for Economic Research, Chișinău, Moldova. Prohnițchi, Valeriu, and Adrian Lupușor. 2013. “Options for Harnessing Emigrants’ Remittances and Savings for the Development of the Republic of Moldova.” February–April, International Organization for Migration and United Nations Development Programme, Chișinău, Moldova. Prokhorova, Anna. 2016. “Migration and Remittances in Moldova.” Background paper for the Moldova Poverty Assessment, World Bank, Washington, DC. Ronnås, Per. 2015. “Productive Employment in Moldova: Trends and Challenges.” Background paper for the Moldova Poverty Assessment, World Bank, Washington, DC. Starkey, Paul, and John Hine. 2014. “Poverty and Sustainable Transport: How Transport Affects Poor People, with Policy Implications for Poverty Reduction, a Literature Review.” Overseas Development Institute, UN‐ Habitat, Partnership on Sustainable Low Carbon Transport, and U.K. Department for International Development, London. Volk, Tina, Emil Erjavec, Ilona Rac, and Miro Rednak. 2015. “Agricultural Policy in the European Union’s Eastern Neighbours.” Agricistrade research report, University of Ljubljana, Slovenia. Vremiș, Maria, Viorica Craievschi-Toartă, Eugeniu Burdelnii, Anne Herm, and Michel Poulain. 2012. Extended Migration Profile of the Republic of Moldova. Chișinău, Moldova: International Organization for Migration. World Bank. 2004. “Recession, Recovery, and Poverty in Moldova.” Report 28024-MD, World Bank, Washington, DC. ———. 2014. “Informal Employment in Moldova: Characteristics and Policy Measures.” Policy Note, World Bank, Washington, DC. ———. 2015a. “Republic of Moldova: Food Security Assessment; Analysis of the Current Situation and Next Steps. Report ACS13175, World Bank, Washington, DC. ———. 2015b. “How to Reduce Out-of-Pocket Payments (Oops) in the Health Sector in Moldova?” Policy Note, World Bank, Washington, DC. ———. 2016a. “Moldova Macro Poverty Outlook.” April, World Bank, Washington, DC. ———. 2016b. Moldova Systematic Country Diagnostic. World Bank Report. Washington, DC: World Bank. ———. 2016c. Economic Security of the Elderly in Moldova. World Bank Report. Washington, DC: World Bank. 39 Poverty and Shared Prosperity in Moldova: Progress and Prospects Annex A. Characteristics of the poor and bottom 40 Table A.1. Headcount ratio and Gini, by location 2007 2014 Squared Squared Poverty Poverty Poverty Poverty Poverty Poverty headcount gap Gap Gini headcount gap Gap Gini Total 25.8 5.9 2.1 29.8 11.4 1.5 0.3 23.4 Urban 18.4 3.6 1.1 29.4 5.0 0.5 0.1 22.7 Rural 31.3 7.6 2.8 28.4 16.3 2.3 0.5 21.6 North 30.4 6.9 2.3 27.1 11.6 1.6 0.4 21.6 Central 30.2 7.5 2.8 30.7 14.9 1.8 0.4 22.4 South 29.5 6.6 2.3 27.7 16.7 2.4 0.5 20.8 Chisinau 11.4 2.2 0.7 28.2 2.6 0.4 0.1 22.9 Source: World Bank calculations based on the HBS Table A.2. Headcount ratio, by household and individual characteristics 2007 2014 Poverty Distribution Distribution Poverty Distribution Distribution headcount of the Poor of Population headcount of the Poor of Population Location Urban 18.4 30.0 42.2 5.0 18.8 43.2 Rural 31.3 70.0 57.8 16.3 81.2 56.8 North 30.4 34.9 29.7 11.6 29.8 29.2 Central 30.2 33.2 28.4 14.9 36.7 28.1 South 29.5 21.9 19.2 16.7 28.0 19.1 Chisinau 11.4 10.0 22.8 2.6 5.5 23.6 Gender of the household head Male 25.9 66.1 66.0 12.1 69.6 65.9 Female 25.7 33.9 34.0 10.2 30.4 34.1 Household head’s age 15-19 4.9 0.1 0.4 1.9 0.3 1.7 20-24 15.0 1.3 2.2 4.6 2.1 5.3 25-29 17.0 3.1 4.7 9.2 5.6 6.9 30-34 21.1 6.4 7.8 9.4 6.9 8.4 35-39 24.7 9.6 10.0 12.6 10.8 9.8 40-44 23.9 10.1 10.9 9.6 8.8 10.4 40 Poverty and Shared Prosperity in Moldova: Progress and Prospects 2007 2014 Poverty Distribution Distribution Poverty Distribution Distribution headcount of the Poor of Population headcount of the Poor of Population 45-49 24.1 13.3 14.3 13.4 15.4 13.0 50-54 25.0 13.3 13.8 9.8 11.7 13.6 55-59 23.9 10.8 11.7 13.4 13.0 11.1 60-64 28.7 5.9 5.3 14.7 25.2 19.6 65+ 35.5 26.1 19.0 Household’s head education Primary 41.4 40.4 25.2 21.1 42.0 22.7 Secondary 26.8 17.5 16.9 12.0 18.5 17.7 Tertiary 18.8 42.1 57.9 7.5 39.5 59.7 Age group 0-13 28.2 18.6 17.1 13.0 17.4 15.3 14-14 28.8 2.2 1.9 18.0 1.9 1.2 15-19 24.4 8.6 9.1 10.7 5.9 6.3 20-24 21.1 6.8 8.4 10.1 5.4 6.1 25-29 21.4 5.0 6.0 6.9 4.1 6.7 30-34 23.0 5.2 5.8 12.2 5.8 5.4 35-39 25.0 5.4 5.6 10.4 4.7 5.2 40-44 24.2 5.4 5.8 12.8 6.3 5.6 45-49 24.1 7.2 7.7 9.8 5.4 6.3 50-54 21.7 6.6 7.8 11.3 8.6 8.7 55-59 23.0 6.4 7.1 9.5 7.6 9.1 60-64 27.4 4.0 3.7 10.8 7.7 8.2 65+ 34.7 18.7 13.9 13.8 19.1 15.9 Education (adult 15+) Primary 33.3 59.1 45.9 16.7 60.8 41.6 Secondary 24.4 15.7 16.7 11.4 16.3 16.3 Tertiary 17.4 25.2 37.5 6.2 22.9 42.0 Employment status (adult 15+) Employed 22.0 47.5 54.5 9.6 49.5 56.8 Unemployed 34.7 4.4 3.2 16.3 4.1 2.8 Inactive 28.7 48.1 42.3 12.7 46.4 40.4 Number of children 0-14 years old no children 23.7 46.4 50.7 10.6 52.8 57.0 1 22.6 22.6 25.8 8.0 16.1 23.0 2 29.6 20.2 17.6 15.0 20.5 15.6 3 or more 47.0 10.7 5.9 27.9 10.7 4.4 children Household size 1 25.8 8.1 8.1 9.0 9.2 11.6 2 23.8 20.1 21.8 10.7 27.1 29.0 41 Poverty and Shared Prosperity in Moldova: Progress and Prospects 2007 2014 Poverty Distribution Distribution Poverty Distribution Distribution headcount of the Poor of Population headcount of the Poor of Population 3 17.4 15.6 23.2 7.6 13.7 20.7 4 23.7 23.3 25.4 11.1 21.4 21.9 5 35.7 17.6 12.7 19.3 19.0 11.3 6 40.8 9.1 5.8 19.5 7.1 4.1 7 or more 53.0 6.1 3.0 20.8 2.5 1.4 Source: World Bank calculations based on the HBS Table A.3. Welfare distribution, by country quintiles, location and household head characteristics 2007 2014 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Location Urban 12.3 16.5 19.3 23.6 28.3 9.8 14.9 19.5 25.1 30.8 Rural 25.6 22.6 20.5 17.4 14.0 27.8 23.9 20.3 16.1 11.9 North 22.4 23.0 21.8 19.4 13.5 20.1 21.4 22.9 20.5 15.1 Central 24.9 20.7 19.0 17.4 18.0 25.9 22.7 19.9 16.5 15.1 South 23.3 25.6 20.3 18.0 12.7 25.9 26.8 19.4 17.2 10.7 Chisinau 7.9 10.6 18.6 25.7 37.2 8.0 9.5 16.9 25.9 39.7 Gender of the household head Male 20.7 19.9 20.1 20.8 18.5 21.0 19.9 20.1 19.8 19.3 Female 18.6 20.3 19.8 18.3 23.0 18.1 20.2 19.7 20.5 21.5 Household head’s age 15-19 4.9 2.4 14.5 14.8 63.4 19.9 5.2 0.0 24.2 50.7 20-24 11.7 12.2 11.9 17.6 46.6 9.2 8.2 16.2 20.4 46.0 25-29 16.6 16.2 17.7 17.5 32.0 17.6 10.5 20.4 19.3 32.2 30-34 21.0 16.9 25.8 15.6 20.7 20.0 16.3 20.3 20.9 22.6 35-39 26.1 20.0 18.9 19.1 15.9 24.8 21.8 22.5 17.6 13.3 40-44 19.7 19.9 21.6 18.1 20.6 27.0 20.6 16.4 17.1 18.9 45-49 18.1 20.6 19.0 22.2 20.0 20.5 18.5 18.2 24.5 18.4 50-54 20.5 18.5 21.6 17.7 21.7 21.8 18.5 15.3 20.3 24.0 55-59 16.9 20.0 17.7 23.9 21.5 16.3 23.0 19.8 21.1 19.9 60-64 19.4 19.2 20.0 20.6 20.8 18.2 20.0 23.5 16.3 21.9 65+ 21.6 24.5 19.9 21.7 12.3 18.0 23.7 23.3 21.1 13.9 Household’s head education Primary 30.6 24.9 19.8 16.3 8.4 31.9 23.2 20.1 15.5 9.3 Secondary 21.0 22.5 19.6 16.5 20.4 20.5 20.7 25.2 17.4 16.2 Tertiary 15.0 17.2 20.2 22.6 25.0 15.3 18.6 18.4 22.5 25.2 Source: World Bank calculations based on the HBS. Note: Individuals are ranked by consumption per capita. 42 Poverty and Shared Prosperity in Moldova: Progress and Prospects The characteristics of individuals and household heads correlate with poverty status. Table A.4. Household and individual characteristics correlations with poverty status Correlates with poverty status Population average 2007 2014 2007 2014 Female 0.00 −0.01 54.4% 54.3% (0.01) (0.00) Age group (reference group = 0–14) 15–24 0.01 0.01 17.5% 12.5% (0.01) (0.01) 25–34 −0.01 0.00 11.8% 12.2% (0.01) (0.01) 35–44 0.01 0.02* 11.4% 10.8% (0.01) (0.01) 45–54 0.03* 0.02** 15.5% 14.9% (0.01) (0.01) 55–64 0.06*** 0.03* 10.9% 17.2% (0.02) (0.01) 65+ 0.11*** 0.04** 13.9% 15.9% (0.02) (0.02) Urban −0.10*** −0.11*** 42.2% 43.2% (0.03) (0.02) Household size 0.04*** 0.01* 3.44 3.11 (0.01) (0.01) Number of children 0.02 0.03** 0.82 0.69 (0.01) (0.01) Number of elderly 0.05*** 0.03* 0.31 0.32 (0.02) (0.02) Constant 0.12*** 0.09*** (0.03) (0.02) Observations 16,589 11,741 Adjusted R 2 0.051 0.042 Source: World Bank calculations based on the HBS. 43 Poverty and Shared Prosperity in Moldova: Progress and Prospects Table A.5. Household head characteristics correlations with poverty status Correlates with poverty status Population average 2007 2014 2007 2014 Female −0.01 −0.03 ** 34% 34.1% (0.02) (0.01) Age group (reference group = 15–24) 25–34 0.08** 0.06*** 12.4% 12.1% (0.03) (0.02) 35–44 0.14*** 0.09*** 20.9% 18.2% (0.03) (0.02) 45–54 0.13*** 0.10*** 28.1% 23.5% (0.03) (0.02) 55–64 0.10*** 0.08*** 17.0% 24.6% (0.03) (0.02) 65+ 0.10*** 0.05** 19.0% 19.6% (0.04) (0.02) Education level (reference group = without primary) Primary −0.09 0.03 6.0% 3.6% (0.06) (0.11) Secondary incomplete −0.11 −0.14 17.5% 18.6% (0.07) (0.11) Secondary complete −0.23*** −0.20* 16.9% 17.7% (0.06) (0.12) Secondary professional −0.25*** −0.20* 26.2% 29.1% (0.07) (0.11) College −0.31*** −0.27** 16.4% 14.7% (0.07) (0.11) University −0.37*** −0.28** 15.3% 16.0% (0.06) (0.11) Sector of employment (reference group = inactive or unemployed) Agriculture 0.01 0.03 32.6% 44.5% (0.03) (0.02) Industry −0.06** −0.05** 16.4% 13.4% (0.03) (0.02) Services −0.10*** −0.03 (0.02) (0.02) 27.3% 28.8% Constant 0.42 *** 0.25** (0.07) (0.12) Observations 16,589 11,741 Adjusted R 2 0.073 0.064 Source: World Bank calculations based on the HBS. 44 Poverty and Shared Prosperity in Moldova: Progress and Prospects Annex B. Nonmonetary poverty Table B.1. Correlations between the multidimensional poverty indicators and monetary poverty status Probit monetary poor=1 2007 2008 2009 2010 2011 2012 2013 2014 Bad health 0.127*** 0.208*** 0.085 0.207*** 0.135** 0.155*** 0.116 0.174** (0.049) (0.048) (0.053) (0.058) (0.056) (0.060) (0.072) (0.074) No insurance 0.141*** 0.090** 0.180*** 0.261*** 0.184*** 0.080 0.159** 0.199*** (0.048) (0.046) (0.054) (0.054) (0.056) (0.065) (0.065) (0.076) Low education, working age 0.259*** 0.261*** 0.378*** 0.314*** 0.322*** 0.187** 0.354*** 0.205** (0.066) (0.069) (0.069) (0.073) (0.078) (0.077) (0.092) (0.088) Behind in compulsory school- 0.409*** 0.133 0.315** 0.256 -0.031 0.383** 0.422* 0.176 ing, school age (0.157) (0.167) (0.158) (0.161) (0.161) (0.190) (0.217) (0.255) Labor force participation, 0.051 0.081 0.009 -0.085 0.060 0.088 0.075 -0.053 working age (0.065) (0.062) (0.072) (0.077) (0.081) (0.081) (0.097) (0.093) Unemployment and underem- 0.116 0.213*** 0.247*** 0.207** 0.327*** 0.324*** 0.360*** 0.180* ployment, working age (0.074) (0.075) (0.074) (0.081) (0.080) (0.086) (0.094) (0.104) House material 0.027 0.207*** 0.252*** 0.155*** 0.201*** 0.027 0.206*** 0.117 (0.054) (0.053) (0.059) (0.059) (0.060) (0.064) (0.069) (0.075) Toilet 0.023 0.146 0.147 -0.034 0.083 0.221* 0.509*** 0.198 (0.103) (0.092) (0.111) (0.102) (0.122) (0.122) (0.173) (0.138) Water 0.106 0.050 -0.064 0.071 -0.096 0.169** 0.010 0.147 (0.079) (0.067) (0.068) (0.072) (0.067) (0.074) (0.081) (0.096) Living space 0.386*** 0.335*** 0.263*** 0.383*** 0.306*** 0.355*** 0.475*** 0.362*** (0.048) (0.047) (0.054) (0.055) (0.055) (0.061) (0.069) (0.077) Heat 0.227*** 0.254*** 0.331*** 0.539*** 0.614*** 0.372*** 0.093 0.410*** (0.076) (0.074) (0.088) (0.083) (0.106) (0.113) (0.162) (0.149) Sewage 0.429*** 0.372*** 0.610*** 0.511*** 0.520*** 0.420*** 0.607*** 0.405*** (0.105) (0.089) (0.094) (0.093) (0.086) (0.093) (0.099) (0.112) Constant -1.496*** -1.591*** -1.775*** -1.997*** -2.183*** -2.037*** -2.503*** -2.324*** (0.064) (0.060) (0.065) (0.080) (0.081) (0.089) (0.104) (0.110) Number of observations 16,589 16,420 15,066 14,379 14,659 13,974 12,354 11,741 Adjusted R2 0.092 0.108 0.161 0.168 0.168 0.138 0.198 0.141 Note: *** .01, ** .05, * .1 45 Poverty and Shared Prosperity in Moldova: Progress and Prospects 46