WESTERN BALKANS REGULAR ECONOMIC REPORT No.17 | Spring 2020 The Economic and Social Impact of COVID-19 POVERTY AND HOUSEHOLD WELFARE “You and me” by Tanja Burzanovic (Montenegro) The RER No. 17 is a collection of notes on the Economic and Social Impact of COVID-19 that will be pub- lished in three parts. The first part was launched on April 29 and focused on the macroeconomic impact of COVID-19. This second part shows how the macroeconomic impact affects the people in the region. It discusses the social impact of COVID-19 in the Western Balkans in six separate RER notes on poverty and welfare, labor, health, education, air pollution, and social protection. The third part, to be launched in early June, will focus on specific economic policy response areas—fiscal, external, and financial sector—and the crisis impact on the private sector as reported by firms. THE ECONOMIC AND SOCIAL IMPACT OF COVID-19 How COVID-19 Could Affect Poverty and Household Welfare in the Western Balkans1 • COVID-19 and containment measures will halt poverty reduction and welfare improvements in the Western Balkans. • The crisis is likely to affect household welfare mainly through fewer jobs and loss of labor income in the most affected sectors, as well as lower international remittances, in addition to impacts on health and access to services. • In the absence of government responses, more than 400,000 people in the region could fall into poverty, and the middle class would shrink substantially. • Many of those who could become impoverished because of the COVID-19 crisis are not receiving social protection unless programs are expanded, and informal workers without access to unemployment insurance and other support to the formal economy are particularly at risk. • Beyond public health responses, governments in the region should minimize the negative economic and poverty impact by providing basic utility and social services and alleviating income shocks, particularly among the most vulnerable workers. Recent poverty reduction and welfare interrupt and partly reverse recent progress. improvements are being interrupted as the This note illustrates possible distributional COVID-19 crisis forces Western Balkan and poverty impacts of COVID-19 in the economies into recession. Before COVID-19 Western Balkans.4 It discusses transmission hit early in 2020, most of the six countries channels: how COVID-19 may affect the poor in the region had made notable progress in and specific groups in the region. It shows, in reducing poverty and boosting household a quantitative simulation exercise, losses that income.2 Yet in 2019, about 19.7 percent of could be caused by reduced workhours, labor the region’s population was still living on less earnings, and remittances. Since the situation than $5.50 a day (2011 PPP) per person— is rapidly evolving and highly uncertain, the the standard upper-middle-income country simulations rely on likely scenarios informed poverty threshold—and sizable numbers were by the information currently available. They close to becoming poor.3 The crisis will likely demonstrate that without carefully designed policy responses, COVID-19 could have 1 This note was prepared by the World Bank poverty and equity team in the Western Balkans, comprising Leonardo Lucchetti, severe negative consequences on poverty Trang Nguyen, Ana Maria Oviedo, and Jayne Yoo. Enrique and household welfare in the region. While Blanco Armas, Alexandru Cojocaru, Facundo Cuevas, Alan Fuchs Tarlovsky, Edith Kikoni, Maria Ana Lugo, Marc Tobias everyone is being affected by COVID-19, poor Schiffbauer, Dhiraj Sharma, Gallina Vincelette, and Salman Zaidi provided helpful comments and suggestions. and vulnerable groups may be particularly hurt 2 For simplicity, in this note income is generally the welfare measure, although consumption is used for Kosovo and Bosnia and Herzegovina. same basis. The regional rate excludes Bosnia and Herzegovina due 3 Throughout this note, poverty estimates are based on $5.50 (2011 to lack of comparable data. PPP) per person per day for all countries except for Bosnia and 4 Recommendations on what countries could do to minimize the Herzegovina, where poverty is based on KM 205 per month poverty and distributional impacts can be found in the companion threshold. The regional poverty rate is calculated as a population- RER notes, including the note titled “Options for social protection weighted average of 2019 estimates of country poverty using the responses to the COVID-19 crisis in the Western Balkans.” Poverty and Household Welfare  |  1 WESTERN BALKANS REGULAR ECONOMIC REPORT NO.17 HOW COVID-19 COULD AFFECT POVERTY AND HOUSEHOLD WELFARE IN THE WESTERN BALKANS because they tend to have less access to services For people in the Western Balkans, labor and fewer savings and coping mechanisms. income losses will be the main way the economic impacts of the pandemic are transmitted. They will probably first affect Channels for transmitting aggregate those employed in vulnerable sectors but COVID-19-induced shocks to will eventually spread to other parts of the individuals economy across formal and informal sectors. In addition to the direct health impacts, In fact, workers in hospitality, entertainment, there are many ways the COVID-19 crisis transport, non-food retail services, and some may affect personal and household welfare, manufacturing (e.g., apparel, automobiles) including:5 were affected immediately. Before COVID-19, about 48-65 percent of workers in Western 1. fewer jobs and thus labor income as Balkan countries were employed in sectors that demand declines and supplies are are moderately or highly affected by the crisis disrupted; (see Annex Table A.1) and they will continue to feel the pain, either by job losses or lower 2. lower nonlabor income, particularly income. A separate note on employment remittances from those who have “Western Balkans Labor Markets and the emigrated to the EU; COVID-19 Shock” for this RER edition discusses in details sectors most acutely affected 3. possible declines in real income if supply and how much they are affected in each country. chain disruptions lead to price increases, rationing of basic consumption goods like The self-employed, informal workers, and food, or both; those with casual jobs or less secure contracts are in particular danger. They are more likely 4. pressure on service delivery and disruptions than formal wage earners on permanent in education, health, and social services contracts to lose their jobs, work fewer hours, (discussed in other RER notes), at the or be furloughed. Workers in the gig economy risk of failure to address the needs of and those unable to work remotely are also such vulnerable groups as children from highly vulnerable. Many of these workers, disadvantaged backgrounds, the elderly, often informal, have no social insurance and the disabled). and will not be eligible for unemployment benefits when they lose their jobs. Government E  specially for poor households that lack response measures that help formal sector savings, insurance, and coping mechanisms, firms to retain jobs will not directly reach an extended loss of income can damage their informal workers. Many of these informal or ability to invest in learning, nutrition, and self-employed workers are not beneficiaries of health of children and mothers, with serious the last-resort social assistance reserved for the consequences in the longer run. poorest, leaving them particularly at risk. In sectors highly affected by COVID-19, about 5 “Poverty and distributional impacts of COVID-19: Potential 14 to 33 percent of workers in Western Balkan channels of impact and mitigating policies”, World Bank Poverty and Equity Global Practice, April 3, 2020. economies are self-employed and therefore 2  |  Poverty and Household Welfare THE ECONOMIC AND SOCIAL IMPACT OF COVID-19 particularly vulnerable (Figure 1). This share The poor and vulnerable may suffer large is particularly high in Albania and Kosovo. economic and social costs. A large share of Worker skill levels in Kosovo, Montenegro, the nonpoor population is at risk of falling North Macedonia, and Serbia are slightly lower into poverty. The COVID-19 outbreak will in affected than in other sectors. The average hit urban dwellers hard, especially those whose wage in affected sectors is also typically lower jobs are precarious (informal) and are less likely (excluding agriculture, which is considered not to benefit from measures to protect formal affected). jobs. Limited savings, low insurance coverage, and relatively less access to public services Nonlabor income is also likely to fall because may exacerbate the long-term negative effects. remittances will drop. In 2019 remittances Moreover, women and minority groups, who are accounted for 2‒10 percent of GDP through the traditionally more affected by domestic violence Western Balkans, with Bosnia and Herzegovina and social discrimination, may be hit harder by and Kosovo particularly dependent on them.6 COVID-19 containment measures. Childcare Emigrant workers, especially those whose and school closings could increase the burden employment is temporary or informal, may on mothers who need to stay home, negatively lose jobs in European countries hit hard by affecting women’s work. Social isolation, the pandemic and lockdowns. Some of those confinement measures, travel restrictions, and emigrants will return home, at risk of being higher stress are already increasing domestic unemployed without access to social protection. violence, as indicated by multiple media and Remittance inflows to families in the Western CSO reports.7 Because the elderly are relatively Balkans will fall. more likely to have health complications, they are disproportionately affected by the outbreak. This issue is particularly relevant in the Western Balkans, where the population has Figure 1.  Up to one-third of workers in affected sectors are self-employed and particularly vulnerable to income losses, mainly in Albania and Kosovo Distribution of employment by type, moderately and highly impacted sectors, percent 100 90 80 70 60 50 40 30 20 10 0 Moderately Highly Moderately Highly Moderately Highly Moderately Highly Moderately Highly Moderately Highly impacted impacted impacted impacted impacted impacted impacted impacted impacted impacted impacted impacted sectors sectors sectors sectors sectors sectors sectors sectors sectors sectors sectors sectors Albania Bosnia and Herzegovina Kosovo Montenegro North Macedonia Serbia J Employee J Self-employed J Unpaid family worker Source: SILC and HBS analysis. 6 World Bank (2019) Western Balkans Regular Economic 7 See, e.g., https://eca.unwomen.org/en/news/stories/2020/4/ Report: Fall 2019. http://documents.worldbank.org/curated/ statement-ed-phumzile-violence-against-women-during- en/643781570478210132/pdf/Rising-Uncertainties.pdf pandemic. Poverty and Household Welfare  |  3 WESTERN BALKANS REGULAR ECONOMIC REPORT NO.17 HOW COVID-19 COULD AFFECT POVERTY AND HOUSEHOLD WELFARE IN THE WESTERN BALKANS aged in recent years. Finally, the risk to health one or two quarters hardest hit by the outbreak and incomes is higher for the Roma and other and containment measures. See Annex 2 for ethnic minorities who have less access to basic full technical details of the methodology. services and economic opportunities. It is estimated that, in the Western Balkans, without considering government response The possible magnitude of COVID-19 measures, the COVID-19 crisis would push distributional impacts in the Western at least 400,000 people into poverty, and up Balkans to 950,000 if the crisis is prolonged. Figure This note uses simulation analysis to assess 2 shows how the number of people in three the likely poverty impact of the pandemic categories could change due to COVID-19: and containment measures. The simulations All six countries would have thousands more look at first-order approximation of the impact, citizens falling into poverty—and reversing the before any government responses, through progress made in reducing poverty in the past multiple channels: income from each sector, five to eight years. In the first simulated scenario, type of work, and household private transfers. COVID-19 is estimated to increase the poverty Poverty is based on income for Albania, rate (by upper-middle-income standards)10 by Montenegro, North Macedonia, and Serbia; 1‒4 percentage points (pp) in all six countries consumption is the welfare measure for Bosnia (Figure 3). Poverty increases more if the crisis and Herzegovina and Kosovo.8,9 Economic lasts through August, as shown in scenario 2, sectors are grouped into three categories: for example, by 3 pp in Montenegro to 10 pp (1) relatively unimpacted, (2) moderately in Kosovo. The share of the vulnerable (living impacted, and (3) highly impacted (see Table on income of $5.50 to $11/day in 2011 PPP) A.1). Self-employed workers are assumed to be is estimated to go up in North Macedonia, relatively more impacted than salaried workers. Montenegro, and Serbia, and in the other The simulations assume two scenarios: in countries, movement from vulnerability scenario 1, the outbreak and the associated into poverty surpasses the movement from containment measures last until end-June, and the middle class into vulnerability. In all the in scenario 2, they last through August. The countries, the crisis also reduces the share of the estimates are average impacts for the whole population who are economically secure (the year, and the impacts will be most severe in the middle and upper classes) by 3‒5 pp in scenario 1, and about 6‒10 pp in scenario 2. 8 Income data in household surveys often do not capture top incomes. Before the simulation, all incomes are now-casted to 2019 values based on growth in real GDP per capita, assuming Income losses are likely across the entire country-specific passthrough rates. income distribution. Average household 9 In some surveys, income data can be less reliable than consumption data and the two do not always align. For instance, in BiH, income is simulated to fall by 3 to 12 percent household income has systematically been about a third lower than household consumption in part because imputations are applied to across the six countries. Figure 4 illustrates the consumption but not income measures. Thus, results for BIH are declines across the income distribution due to underestimated, should be considered a lower bound of the impact, and are not fully comparable with the impact on poverty elsewhere COVID-19. Income losses that individuals in the region. Nevertheless, the simulation of income shocks in BiH is still informative if the results are considered with this caveat and families in the poorest quintile could suffer (World Bank. 2015. Rebalancing Bosnia and Herzegovina: A Systematic Country Diagnostic. Washington, D.C.: World Bank 10 Because comparable data are lacking, poverty in BiH is based on a Group). KM 205 monthly threshold. 4  |  Poverty and Household Welfare THE ECONOMIC AND SOCIAL IMPACT OF COVID-19 Figure 2. The COVID-19 outbreak will substantially increase the number of poor people Simulated change in numbers in poverty (P), vulnerable (V), and middle class (M-C) due to COVID-19 (number of people in thousands) 350 150 -50 -250 -450 -650 P V M-C P V M-C P V M-C P V M-C P V M-C P V M-C Albania Bosnia and Herzegovina Kosovo North Macedonia Montenegro Serbia J Scenario 1 minus baseline Q Scenario 2 minus baseline Source: World Bank simulations based on 2017 income data from the SILC for Albania, North Macedonia, and Serbia; 2015 income data from the SILC in Montenegro; 2015 HBS for BiH, and 2017 HBS for Kosovo. Note: P = Poor, V = Vulnerable, and M-C = Middle-class. Income measures in the SILC and consumption measures in the HBS are not strictly comparable. Baseline = poverty as estimated in 2019 based on household survey data in earlier years and growth in real GDP per capita. Scenario 1 = crisis lasts one quarter; Scenario 2 = crisis lasts two quarters. Since poverty in BiH is based on a KM 205 per capita monthly threshold and the per capita welfare measure is reduced to 2007 prices, vulnerability and middle-class cannot be calculated and results are not comparable to other countries. Figure 3. The COVID-19 outbreak could push a large share of vulnerable people into poverty. Simulated impact of the COVID-19 crisis by economic class (percent of population) 100 90 25.4 22.6 19.9 26.7 22.1 19.9 80 48.3 42.2 47.8 46.5 42.5 52.5 53.6 51.1 50.8 70 36.5 60 37.8 39.1 52.6 50 56.3 55.4 40 34.5 30.8 33.5 34.6 32.2 29.4 31.0 30.7 28.3 30 20 39.6 43.6 35.5 27.5 23.3 22.9 10 12.9 14.6 17.9 21.6 16.9 19.5 18.1 19.5 21.3 18.2 20.0 11.8 0 Scenario Scenario Baseline Scenario Scenario Scenario Scenario Scenario Scenario Baseline Baseline Baseline Baseline Scenario Scenario Baseline Scenario Scenario 1 2 1 2 1 2 1 2 1 2 1 2 Albania Bosnia and Herzegovina Kosovo North Macedonia Montenegro Serbia J Poor J Vulnerable J Middle-class Source: World Bank simulations based on 2017 income data from the SILC for Albania, North Macedonia, and Serbia; 2015 income data from the SILC in Montenegro; 2015 HBS in BiH, and 2017 HBS in Kosovo. Note: See Figure 2 for additional notes. Figure 4. Household income will fall across the income distribution. Simulated income loss due the COVID-19 (%) by welfare distribution quintiles. 0 -2 -4 -6 -8 -10 -12 -14 -16 -18 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Albania Bosnia and Herzegovina Kosovo North Macedonia Montenegro Serbia ▬ Scenario 1 ▬ Scenario 2 Source: World Bank simulations based on 2017 income data from the SILC for Albania, North Macedonia, and Serbia; 2015 income data from the SILC in Montenegro; 2015 HBS in Bosnia and Herzegovina, and 2017 HBS in Kosovo. Note: see Figure 2. Poverty and Household Welfare  |  5 WESTERN BALKANS REGULAR ECONOMIC REPORT NO.17 HOW COVID-19 COULD AFFECT POVERTY AND HOUSEHOLD WELFARE IN THE WESTERN BALKANS Figure 5. Many of the new poor are not receiving any social protection benefits. Simulated percentage of the population not receiving any social protection support 100 90 80 70 60 50 40 30 20 10 0 Albania Kosovo Montenegro J Originally poor not covered by SP Q People who entered poverty and are not covered by SP (Scenario 1) Q People wo entered poverty and are not covered by SP (Scenario 2) Source: World Bank simulations based on 2017 income data from the SILC for Albania, North Macedonia, and Serbia; 2015 income data from the SILC in Montenegro; 2015 HBS in BiH, and 2017 HBS in Kosovo. Note: see Figure 2. range from 3 percent in BiH to 5 percent Many of the new poor may be left without in Kosovo in scenario 1, or double that in social protection support unless programs scenario 2. In most Western Balkan countries, are expanded. Informal workers losing their compared to the rest of the population, the jobs do not receive unemployment insurance poor rely more on nonlabor income and less on benefits. While social assistance programs in income from work in the sectors most affected the Western Balkans are generally well-targeted, by COVID-19. Still, since poor households coverage of the poor has been limited. Few of had little to start with, a loss of income can the new poor are receiving any social protection jeopardize their ability to meet basic needs benefit—for example, more than 75 percent and invest in critical human development. The of them in Albania, North Macedonia, and second to fifth quintiles are estimated to face Montenegro (Figure 5). The group of new larger declines in income, because more of their poor that are not eligible for unemployment total income comes from the most affected insurance or social assistance would be sectors. particularly vulnerable without policies to support them. Those working in the most affected sectors experienced particularly high increases in poverty rates because of COVID-19, even Results by country though they were often not among the poorest before the crisis. For example, in Albania:  COVID-19 and containment Montenegro, only 6 percent of those working measures are expected to reverse recent poverty in the most affected sectors (category 3 in Table reduction in Albania. Albania closed activity A.1) were poor before, but the pandemic is in hospitality, theaters, and nonessential trade estimated to increase the poverty rate among in March and limited the movement of people them to 9 percent in scenario 1 and 13 percent both inside and abroad. Many industries, in scenario 2. among them textiles, mining, call centers, and construction, have been forced to reduce production to comply with social distancing 6  |  Poverty and Household Welfare THE ECONOMIC AND SOCIAL IMPACT OF COVID-19 requirements and because demand has dropped Kosovo: Th e impacts of COVID-19 on poverty as export orders collapse. Albanian emigrants are likely to be substantial, as economic activity are also likely to suffer from this global in Kosovo has been brought to a standstill and crisis, which reduces their incomes and thus remittances plummeted. A significant number remittances. The simulations produced for this of people are likely to lose their jobs, especially note estimate that poverty in Albania could go those on fixed-term contracts, seasonal workers, up by 4‒8 pp, to about 40 percent in the first and those working in the informal economy. scenario and 44 percent in the second scenario. Kosovo has a large construction sector, which is Thus, in the more optimistic scenario, poverty particularly sensitive because it is an important would go back to the 2012 rate and, in the source of income for the urban poor and more pessimistic scenario, it would equal the vulnerable. Indeed, the simulation conducted 2005 rate. Since most of the people working in for this note suggest that 57,000‒148,000 agriculture are already poor, and the simulation people could become poor, while the middle assumes no income decrease in agriculture, the class could shrink by as many as 100,000 increase in poverty is mainly the result of urban people. This is equivalent to poverty rate going people in services losing a significant share of up by 4‒10 pp. Although social assistance is their earnings. an important source of income for the poor (about 15 percent in 2017) and will to some Bosnia and Herzegovina:  The slowdown due extent protect the poorest from crisis effects, to the global COVID-19 outbreak, and the more than half the people falling into poverty associated potential loss of jobs and earnings, will do not benefit from any current social program likely have a short-term effect on poverty. The (though about 40 percent receive pensions), it latest available BiH poverty data are for 2015, will be necessary to expand the coverage and when poverty was an estimated 16 percent, very possibly the generosity of the cash transfer close to the 14 percent poverty rate estimated program to ease the effects of the collapse in for 2007. Simulations show that consumption incomes. will likely fall by 3.5 to 6.7 percent across the entire distribution, so poverty is predicted Montenegro: Th  e COVID-19 pandemic to rise from about 11.8 percent in 2019 to could raise the poverty rate in Montenegro by 12.9 to 14.6 percent in 2020. Depending on an estimated 1.5‒3 pp, depending on whether whether the crisis lasts one or two quarters, the crisis lasts one or two quarters, but most about 35,000 to 85,000 individuals are likely of the new poor may be unprotected. The to fall into poverty. Among workers in the most result would be to reverse the past 4‒6 years affected sectors before the pandemic, poverty of poverty reduction, bringing poverty back up was about 9 percent. However, that rate will to the 2015, or even 2013, level. From about likely increase to 11 percent in scenario 1 and 9,000 to 20,000 citizens could fall into poverty 15 percent in scenario 2. It is worrying that due to the economic shock. Among workers more than 60 percent of those likely to fall in the most affected sectors (see Table A.1), into poverty due to the COVID-19 crisis are only 6 percent were poor before the pandemic, currently not covered by any social protection but the rate is estimated to go up to 9 percent program. in scenario 1 and 13 percent in scenario 2. Average household income could fall by 5 Poverty and Household Welfare  |  7 WESTERN BALKANS REGULAR ECONOMIC REPORT NO.17 HOW COVID-19 COULD AFFECT POVERTY AND HOUSEHOLD WELFARE IN THE WESTERN BALKANS to 10 percent—more in the higher quintiles shock. It is estimated that poverty in Serbia will because more of their total income comes from go up by 2‒4 pp, reaching about the poverty the sectors most affected. The share of the levels in 2016‒17. Among workers in the most population considered vulnerable (not poor but affected sectors (see Table A.1), only 8 percent not economically secure either) may go up from were poor before the pandemic, but that could about 28 percent to an estimated 29‒31 percent. rise to an estimated 12 percent in scenario 1 Worryingly, more than 80 percent of the and 19 percent in scenario 2. In both scenarios, Montenegrins who could become poor because average household income could fall by 5 to of the COVID-19 crisis do not currently have 10 percent—more in the middle quintiles who any social protection support. With its fiscal rely on income from the sectors most affected. buffers limited, Montenegro’s short-term policy The share of the population considered responses would likely be targeted but should vulnerable (not poor but also not economically consider relieving adverse impacts on poor and secure) is estimated to go up from about 31 vulnerable workers and families. percent to 33‒35 percent. Many of the new poor may be left without social protection North Macedonia: Th  e combined effect of support unless the government revises benefit lower labor incomes and lower remittances due programs. About 66 percent of those Serbians to COVID-19 will likely increase poverty to who could become impoverished because of the pre-2017 or even pre-2015 levels, depending COVID-19 crisis currently receive no social on the duration of the crisis. Simulations protection benefits. Short-term measures to predict the poverty rate to rise from about mitigate impacts on the poor and vulnerable 17 percent in 2019 to about 20‒23 percent would be important to complement aggregate in 2020, and from 55,000 to over 130,000 macroeconomic policy responses. Macedonians (depending on whether the crisis lasts one or two quarters) would fall into poverty. Before the pandemic, poverty was only Broad policy directions 4 percent among workers in the most affected sectors, but the rate is estimated to hit about 8 The simulations show that in addition to the in scenario 1 and 15 percent in scenario 2. The health impacts, poverty could significantly impact of the virus outbreak will likely differ increase and the middle class shrink across along the income distribution; incomes will the board. Most of the new poor and vulnerable fall by about 5 to 8 percent among those in the do not currently get social protection benefits. lowest quintile, and 7 to 13 percent or more Clearly, the first priority is to contain the spread for the rest of the population. Moreover, more of the virus and protect the health system, as than 75 percent of those who are likely to fall all countries are doing now. But as extreme into poverty are currently not covered by any measures can have severe negative social social protection program. and economic impacts, particularly for the urban vulnerable and poor, minimizing those Serbia: S  erbia’s recent poverty reduction could losses is the second policy priority. Indeed, be partly reversed in 2020: depending on the governments in all six countries are moving crisis duration, 125,000 to 327,000 citizens to do so. Most of the measures announced could fall into poverty due to the economic support formal sector workers or beneficiaries 8  |  Poverty and Household Welfare THE ECONOMIC AND SOCIAL IMPACT OF COVID-19 of existing social assistance programs, leaving can resume relatively quickly after mobility vulnerable informal workers unprotected. The restrictions are lifted. More importantly, broad policies outlined here touch upon two income support responses need to be rolled areas: providing basic social services, including out for workers that may not be covered by preserving human capital investment, and firm support programs because they are self- alleviating income shocks, especially for more employed or work in the informal sector. vulnerable workers. Specific responses are Examples of such measures include expansion discussed in detail in other notes of this edition of safety net programs, public work programs, of the RER. and adjusting unemployment insurance. First, governments must ensure basic services, food security, shelter, and basic social services for the most vulnerable. Despite severe lockdown measures and reduced economic activity, providing basic services like utilities, connectivity, waste management, and infrastructure maintenance is still essential in part ensure that people respect the lockdown. Moreover, many people in the Western Balkans are not in the labor force and will not be covered by most of the economic rescue measures. Among them are the most vulnerable—the elderly, the disabled, and ethnic minorities— whose situation could worsen when remittances fall. It will be important that responses, including in-kind supports, ensure that a minimum level of support and services is available to them. Moreover, policies to support continued learning for all is critical to preserve investments in human capital and avoid exacerbating learning gaps for children from disadvantaged backgrounds. Second, policy responses need to protect livelihoods and facilitate recovery, for informal as well as formal sector workers. Minimizing firm exits and job destruction is a first step and can have important positive spillover effects for the informal sector via both supply and demand channels. These measures are also needed to ensure that a minimum level of economic activity is maintained or Poverty and Household Welfare  |  9 WESTERN BALKANS REGULAR ECONOMIC REPORT NO.17 HOW COVID-19 COULD AFFECT POVERTY AND HOUSEHOLD WELFARE IN THE WESTERN BALKANS Annex 1. Employment in sectors affected and those unaffected by COVID-19 Table A.1. Distribution of employment by sector and estimated degree of impact on sector growth Employment distribution by sector (% of the employed) ALB BIH KOS MKD MNE SRB Agriculture, hunting, forestry, and fishing 33.4 17.1 7.8 11.1 6.2 8.2 Mining and quarrying 1.0 2.4 1.9 0.9 0.6 1.9 Manufacturing 12.1 11.9 10.6 22.2 5.7 22.1 Utilities 1.9 2.8 3.8 3.4 5.4 2.7 Construction 7.9 10.9 17.0 8.0 5.3 4.3 Wholesale and retail trade 11.2 13.0 15.5 13.2 19.4 12.4 Transportation 3.8 6.4 4.1 4.8 9.1 6.1 Accommodation and food service activities 6.0 5.0 5.1 5.2 6.3 3.7 Information and communication 0.9 - 2.3 1.9 2.2 3.3 Financial and insurance activities 0.7 1.9 2.3 1.5 2.5 2.0 Real estate activities 0.9 0.3 0.2 0.2 0.2 Professional, scientific, and technical activities 1.5 - 1.5 2.1 2.7 2.7 Administrative and support service activities 2.6 - 2.1 2.5 2.0 3.7 Public administration 4.9 6.6 4.5 6.3 11.7 5.6 Education 5.6 5.1 8.7 5.9 8.9 7.5 Human health and social work 3.7 5.1 4.5 5.3 6.5 6.7 Art, entertainment, and recreation 0.9 - 1.5 1.6 2.9 2.1 Other service activities 1.4 10.0 5.7 2.7 2.1 3.2 Activities of households as employers 0.4 0.4 0 0.2 0.2 0.7 Activities of extraterritorial organizations 0.1 0.5 0.8 0.9 0.2 0.6 Categories of sectors by impact on sector growth Highly impacted 24.0 19.3 31.7 25.0 33.7 25.5 Moderately impacted 23.8 39.2 33.6 37.7 22.8 37.6 Relatively unimpacted 52.2 41.5 34.7 37.2 43.6 36.6 Source: Calculations based on SILC and HBS data. 10  |  Poverty and Household Welfare THE ECONOMIC AND SOCIAL IMPACT OF COVID-19 Annex 2. Detailed description of the methodology This note employs simulation analysis to members. In the other two countries, the most assess the possible poverty and distributional recent surveys available are the 2015 Household impact of the COVID-19 outbreak in the Budget Survey (HBS) for BiH12 and the 2017 six Western Balkan countries. Simulations HBS for Kosovo. These ask about household consider first approximation of the impact incomes, spending on final consumption (goods through multiple channels: income from each and services), socio-economic and geographical sector of employment, type of work, and characteristics of households, characteristics of household private transfers. Some impacts household members, participation in the labor are not incorporated in the simulations, market, housing characteristics, and social among them spill-over and second-order inclusion and subjective opinion on health and effects and behavioral responses. Moreover, health care services. Poverty is based on per the methodology does not measure the direct capita household income in Albania, North impact of loss of earnings because of illness; the Macedonia, Serbia, and Montenegro and on impact on consumption because of changes in per capita household consumption in BiH and prices or higher out-of-pocket health care costs; Kosovo. and the nonmonetary dimension of welfare loses. The analysis relies on likely labor market impacts, by sector and employment type, The analysis uses the latest available before any government responses. Economic household surveys in the six Western Balkan sectors of activity are grouped into (1) relatively countries. In Albania, North Macedonia,11 unimpacted, (2) moderately impacted, and (3) and Serbia, the most recent data available is highly impacted (see Table A.1). Self-employed from the 2018 Survey on Income and Living workers are less likely to be enrolled in social Conditions (SILC), which refers to 2017 security schemes and may not be covered by income. The analysis for Montenegro uses government emergency response measures. 2015 income data from the 2016 SILC, which Varying shocks are assumed within these sectors asks about household income, housing and and by employment type (see Table A.2). We living conditions, health, social exclusion, and assume in scenario 1 that the self-employed socioeconomic characteristics of household in group 3 will lose income for one quarter Table A.2. Labor Income Losses by Sector and Employment Type (%) Scenario 1 Scenario 2 Wage employed Self-employed Wage employed Self-employed Highly impacted 12.5 25 25 50 Moderately impacted 6.25 12.5 12.5 25 Unimpacted 0 0 0 0 12 Agency for Statistics of Bosnia and Herzegovina, Republika Srpska 11 State Statistical Office of the Republic of North Macedonia. Survey Institute for Statistics, and Institute for Statistics of Federation on Income and Living Conditions: Microdata file. Skopje, North Bosnia and Herzegovina. Household Budget Survey: Microdata Macedonia: State Statistical Office. file. Poverty and Household Welfare  |  11 WESTERN BALKANS REGULAR ECONOMIC REPORT NO.17 HOW COVID-19 COULD AFFECT POVERTY AND HOUSEHOLD WELFARE IN THE WESTERN BALKANS (25 percent of annual income), group 2 will lose half of their income this quarter (12.5 percent), and incomes in group 1 will not be affected. The income shock for a wage employee is assumed to be smaller: 12.5 percent in group 3, 6.25 percent in group 2, and no impact in group 1. In scenario 2 we assume that the pandemic lasts for two quarters, and income losses are double those in scenario 1. We also assume that private transfers will be reduced by 40 percent during a quarter of the year. The impacts are simulated by subtracting the loss in worker earnings and household remittances from per capita household income and comparing these scenarios with the baseline. 12  |  Poverty and Household Welfare Western Balkans Regular Economic Report No.17 | Spring 2020 Poverty and Household Welfare View this report online: www.worldbank.org/eca/wbrer You and me by Tanja Burzanovic (Montenegro) Dr. Tatjana Burzanovic has a wide experience in the fields of graphic design, graphics in architecture, interior design. She has worked as an art editor, interior designer and graphic designer at various levels. Many of her art exhibitions have taken place at different places. She has received many awards for her arts and literary works. She has published a book with a title The Interrelation between Art Worlds, with the support from the Embassy of India for Austria and Montenegro in Vienna. Her artistic philosophy includes displaying of interrelationship between art worlds (spatial and temporal arts). The artist thus meditates between nature and the sprits and yet stems from the absolute idea and serves the goal of realization of absolute sprit. ‘Grasping the meaning through the form’ is a task of the art set by a contemporary thinker to demonstrate that building forms and creating sense are two simultaneous, intertwined, and absolutely inseparable processes in Arts. Without that recognition it is not possible to take any further step in investigating the nature of art and literature. She believes that art is a way to search the truth. Art is inseparable from searching the truth. People forge ideas, people mold dreams, and people create art. To connect local artists to a broader audience, the cover of this report and following editions will feature art from the Western Balkan countries.