Report No: AUS0000425 . Afghanistan Trends in Poverty and Inequality 2007-17 . July 2018 . POV . . Document of the World Bank © 2017 The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved This work is a product of the staff of The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Attribution—Please cite the work as follows: “World Bank. 2016. Trends in Poverty and Inequality 2007-17. © World Bank.” All queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. AFGHANISTAN: TRENDS IN POVERTY AND INEQU ALITY 2007-17 1 SUMMARY A severe slow-down in Afghanistan’s economic growth characterized the period between 2012 and 2016. This sharp deceleration can be attributed to the combined effects of the drawdown of international military forces and a sharp fall in associated international spending, reduction of aid, and increasing conflict and political instability. These trends are reflected in the increasing vulnerability of the Afghan population, as widespread deteriorations in welfare are evidenced in the sharp increase in poverty rates to 55 percent in 2016-17. Many inequalities persist in Afghanistan, between regions, cities and rural areas, and rich and poor Afghans. Poverty headcount rates increased in every region between 2011-12 and 2016-17 and the deterioration in welfare was experienced across the distribution, among the poorest households, as well as among the most well-off. These distributional changes imply that while the intensity of poverty has increased between 2011-12 and 2016-17, inequality has declined, as the welfare loss among the top of the distribution has been relatively larger than that at the bottom of the distribution. Demographic characteristics remain strongly correlated with poverty headcount rates. Poverty rates increase steadily with household size and households of larger size are both more prevalent and face a higher poverty rate. Education (or the lack thereof) is another important correlate of poverty in Afghanistan. Low levels of educational attainment are pervasive and households with illiterate heads account for 74 percent of the population, facing poverty rates of 63 percent on average, compared with headcount rates of 40 percent among households with literate heads. While unemployment of the head of household is correlated with higher poverty, employment is no guarantee against poverty. Roughly half the population belonging to households with employed heads lives in poverty. Few have access to productive or remunerative employment. Afghans living in households where the household head is employed in agriculture are likely to face higher poverty rates (63 percent) and account for a third of the poor population. More broadly, almost 60 percent of the population belongs to households where the head of household holds vulnerable employment, or in other words, is self-employed or works on own-account, is a day laborer or is an unpaid worker. 1 This report was prepared by a team from the Poverty and Equity Global Practice of the World Bank and included Nandini Krishnan (Senior Economist), Christina Wieser (Economist), and Zihao (Tobias) Wang (Consultant). CONTEXT, 2007-2017: DECLINE IN ECONOMIC GROWTH AND A DETERIORATING SECURITY SITUATION The overall macro-economic and security context in the country since 2007 can be broken into two distinct phases, pre- and post- the 2014 security transition. While the pre-transition phase was characterized by higher economic growth and a relatively stable security situation, since 2014, growth has stagnated, and the security situation continues to deteriorate. With the withdrawal of international troops and the steady decline in aid (both security and civilian aid) since 2012, the economy witnessed a huge shock to demand, one that it is still struggling to recover from. The economy remains heavily aid-dependent: aid covers almost 60 percent of Afghanistan’s on-budget civilian expenditures and 80 percent of on-budget development expenditures. In this context, the 2016-17 poverty estimates are the first direct estimates of welfare since the security transition. Figure 1 plots per capita GDP and annual GDP growth between 2007 and 2016.2 Between 2007 and 2012, GDP per capita increased from $381 to $691, with economic growth averaging 11.2 percent per year. Since then, the Afghan economy has grown at an average of 2.1 percent between 2013 and 2016, and GDP per capita in 2016 remains $100 below its 2012 levels. This economic slowdown has been accompanied by a deterioration in security since 2014 and economic activity (as measured by new firm registrations, Figure 2) has been adversely affected. Figure 1: Economic growth and per capita GDP has declined since Figure 2: The deterioration in the security situation has adversely 2012 affected economic activity GDP per capita ($) Battle-related deaths Real GDP growth, annual (%) New firm registrations 800 25 20000 18000 700 20 16000 600 GDP per capita ($) 14000 % growth GDP 500 15 12000 400 10000 300 10 8000 6000 200 5 4000 100 2000 0 0 0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Source: World Bank, ADU, November 2017, based on Source: World Bank, ADU, November 2017; data from the National Statistics and Information UNAMA Authority (NSIA) 2 Afghanistan’s economic growth is projected to increase slightly to 2.6 percent in 2017, and assuming no further deterioration in security, to 3.2 percent in 2018. World Bank, 2017. Afghanistan Development Update, November 2017. The deteriorating security situation has led to large-scale population displacements (Figure 3), and has coincided with the return of more than a million Afghans. Since 2007, the number of injuries and deaths has increased five-fold, and in 2016, more than 650,000 Afghans were internally displaced due to conflict. At the same time, 2016 witnessed the return of more than a million documented and undocumented Afghan refugees, primarily from Pakistan and Iran (Figure 4). Internal displacement and large-scale return within a difficult economic and security context pose risks to welfare, not only for the displaced, but also for the population at large, putting pressure on service delivery systems and increasing competition for already scarce economic opportunities. Figure 3: More than 650,000 Afghans were internally displaced by Figure 4: More than a million Afghans returned in 2016 alone conflict in 2016 800000 140000 Documented returns Undocumented returns 700000 120000 1,200,000 600000 100000 1,000,000 Injuries + Deaths 500000 80000 800,000 IDPs 400000 691,581 60000 300000 600,000 40000 200000 400,000 663,295 100000 20000 564,443 200,000 372,577 0 0 2007 2009 2011 2013 2015 2017 - 58,460 58,817 Injuries+Deaths Conflict-induced IDPs 2015 2016 2017 Source: UNOCHA, UNAMA Source: UNHCR, IOM Sector specific trends in growth suggest further causes for concern. While agriculture’s contribution to GDP has declined steadily from around 30 percent in 2007 to 22 percent in 2016, it remains an important sector as a source of livelihoods for the rural poor, in influencing the affordability of basic food items for the population, and its significant inputs into the manufacturing sector (Figure 5). Characterized by significant annual fluctuations, the agricultural sector grew, on average, 8 percent per year between 2007 and 2012. Since then, its annual growth rate has fallen sharply to 0.1 percent on average. Potentially related, the Afghanistan Living Conditions Survey (ALCS) 2016-17 survey period coincided with an increase in food price inflation, which climbed to 10.7 percent year-on-year in May 2017 (World Bank, ADU, November 2017). Figure 5: Annual growth rates by sector (%) Agriculture Industry Services 80.0 60.0 40.0 20.0 0.0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 -20.0 Source: World Bank staff estimates for ADU, November 2017, based on data from the NSIA TRENDS IN POVERTY, 2007-2017 In 2007-08, the national average threshold for the cost of covering basic needs, the poverty line was 1,261 Afs per person per month. This threshold, consistently defined and valued at 2016-17 survey prices, increases to 2,056 Afs per person per month. Relative to 2007, this represents a 64% increase in the cost of basic needs, comprised of a 68% increase in the cost of the basic food bundle (benchmarked at 2,100 kilocalories per person per day), and a 58% increase in the cost of non-food necessities. Using current exchange rates, the national poverty line is roughly equivalent to US$30 per capita per month, or US$1 per person per day.3 Afghanistan has experienced a sharp increase in poverty since 2011-12. Figure 6 plots the national, urban and rural poverty headcount rates based on the new comparable series (using the three surveys where direct estimation of poverty is possible) against GDP per capita.4 Poverty headcount rates measure the share of the population whose monthly per capita expenditure falls below the poverty line. At the national level, these headcount rates increased from 34 percent in 2007-08 to 38 percent in 2011-12, followed by a sharp rise to 55 percent in 2016-17 (Table 1). Rural poverty remains consistently higher than urban poverty, although the deterioration in welfare has become more widespread. While the increase in poverty in the first period, 2007- 3 No Purchasing Power Parity estimates are used for poverty estimates for Afghanistan; using DEC alternative conversion factor LCU to US$. 4 These trends take into account methodological improvements defined consistently over time, and therefore differ from previously released estimates. Table: Comparable poverty trend series (Old and revised) 2007-08 2011-12 2013-14* 2016-17 Old series, excluding Helmand and Khost 36 36 39 Revised series, all provinces 34 38 55 Source: NSIA, World Bank staff estimates, NRVA 2007, 2011 and ALCS 2016; *Imputation estimates 2011, was driven by an increase in rural poverty, in the second period, both urban and rural poverty rates have increased substantially. These trends are consistent with the large economic contraction the country has experienced since 2012. The period 2007 to 2011 was characterized by a large increase in GDP per capita (which grew 63 percent relative to its 2007 value), whereas during the latter period, 2012 to 2016, GDP per capita actually fell (Figure 7). Even during this period, there was a small increase in poverty despite robust growth, because growth was not pro-poor. With the continued decline in GDP per capita, poverty rates have gone up significantly. Table 1: Poverty headcount, poverty gap, and squared poverty gap, 2007-2017 Indicator 2007-08 2011-12 2016-17 Poverty rate 34% 38% 55% Urban poverty rate 26% 25% 42% Rural poverty rate 36% 42% 59% Poverty gap 7% 10% 15% Squared poverty gap 2% 4% 6% Source: NSIA, World Bank staff estimates, NRVA 2007, 2011 and ALCS 2016 Figure 6: Trends in poverty and GDP per capita, 2007-2017 Figure 7: GDP Per capita (% change) GDP per capita ($) National poverty rate (%) 70.0 63.3 Urban poverty rate (%) Rural poverty rate (%) 60.0 800 70.0 50.0 700 60.0 600 40.0 50.0 500 30.0 40.0 400 30.0 20.0 300 200 20.0 10.0 100 10.0 0.0 0 0.0 2007-2011 2011-2016 -10.0 -5.0 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Source: NSIA, World Bank staff estimates, NRVA 2007, 2011 and Source: NSIA, World Bank staff ALCS 2016 estimates As economic growth has lagged behind population growth, not to mention the additional 2.3 million Afghan returnees since 2015, the country has experienced a large increase in the number of poor people (Figure 8). In 2016-17, almost 13 million rural Afghans lived below the poverty line. At the same time, urban poverty has become more widespread, with the number of urban poor more than doubling since 2007, and consequently, 18 percent of Afghanistan’s poor now live in urban areas (Figure 9). The increase in urban poverty since 2011 was concentrated in Kandahar, Kabul, Herat, Balkh and Kunduz. In 2016-17, these provinces together accounted for 80 percent of the urban poor, with Kabul alone accounting for about half of all the urban poor. In part, this trend may be driven by IDPs and returnees turning to urban centers in search of security, jobs and services. If this trend continues, the pressure on urban centers will likely increase. Figure 8: Estimated number of poor people Figure 9: Share of the poor living in urban and rural areas Urban Rural 84.82 85.38 81.82 2007-08 2011-12 2016-17 18.18 15.18 14.62 Rural 7129355 8815166 12979997 Urban 1275920 1509359 2883535 2007-08 2011-12 2016-17 Source: NSIA, World Bank staff estimates, NRVA Source: NSIA, World Bank staff estimates, NRVA 2007, 2011 and ALCS 2016 2007, 2011 and ALCS 2016 This large deterioration in welfare at a poverty line roughly equivalent to US$1 a day (using exchange rates) arguably constitutes a humanitarian crisis. The trends in poverty closely match an increase in food insecurity, from 30 percent in 2011/12 to 45 percent in 2016/17, driven by an increase in the proportion of severely and very severely food insecure population. A large proportion of the Afghan population was vulnerable to poverty even in past years. Figure 10 plots the estimated poverty rates if per capita expenditures fell by 10 percent, 20 percent, and 30 percent respectively in each of the survey years. The current poverty estimates in 2016/17 are roughly what would have obtained in 2007 and 2011 had the population experienced an income shock of around 20 percent of their per capita expenditures. Considerable vulnerability remains in the population today: a 10 percent decline in per capita expenditures would push poverty rates up to 63 percent in 2016/17. Even the richest quintile, earning the equivalent of US$50 a month would come close to the poverty line with a 30 percent income shock. Finally, it is unclear the extent to which poverty estimates and trends are affected by the exclusion of districts deemed to insecure for fieldwork. In 2016, 45 such districts were excluded from the reserve sample, accounting for almost 10% of the national population.5 5 These include Alasay and Tagab (Kapisa), Chaki Wardak, Jaghatu, Sayydabad and Nerkh (Wardak), Khar War and Azra (Logar), Bala Murghab and Ghormach (Badghis), Kot, Hesarak, Sher Zad, Pachir Wagam, Deh Bala, Dur Baba, and Lalpoor (Nangarhar), Ajristan and Nawur (Ghazni), Baghran, Garm Ser, Musa Qala, Nawzad and Sangin Qala (Helmand), Shahid-e-Hassas and Khas Urozgan (Urozgan), Qush Tepa and Darzab (Jawzjan), Shorabak, Reg Shiga, Miyanishin, Nesh and Maruf (Kandahar), Daichopan, Shemel Zayi and Kakar/Khak-e-Afghan (Zabul), Chapa Dara, Nari and Ghazi Abad (Kunarha), Tala Wa Barfak, Dahana-e-Ghuri, and Burka (Baghlan), Alishing and Bad Pakh (Laghman), Jani Khel, Zurmat and Wuza Jadran (Paktya), Bala Buluk, Gulistan and Pur Chaman (Farah), Sozma Qala Figure 10: High vulnerability to income shocks baseline 10% 20% 30% 80.6% 72.1% 67.5% 66.5% 63.3% 56.3% 55.1% 54.5% 46.9% 44.0% 38.3% 33.7% 2007 2011 2016 Source: NSIA, World Bank staff estimates, NRVA 2007, 2011 and ALCS 2016 Distributional changes in welfare This deterioration in welfare was experienced across the distribution, among the poorest households, as well as among the most well-off. Table 2 displays the cut-off points of each of the expenditure quintiles in 2016 prices which also shows that expenditure ranges in each of the quintiles narrow substantially between 2011-12 and 2016-17. We observe that almost all Afghans in the 3rd expenditure quintile show monthly expenditures below the poverty line of 2,056 Afs per month while in 2007-08 and 2011-12, roughly the bottom 2 expenditure quintiles are below the poverty line of the respective year. Table 2: Expenditure intervals for each quintile by survey year Expenditure intervals (in 2016 Afs) 2016-17 2011-12 2007-08 Bottom 20% 331 - 1,228 330 - 1,394 269 - 1,531 2 1,229 - 1,606 1,394 - 1,866 1,531 - 1,992 3 1,606 - 2,078 1,866 - 2,425 1,992 - 2,549 4 2,079 - 2,900 2,425 - 3,379 2,549 - 3,486 Top 20% 2,900 - 34,103 3,379 - 50,164 3,486 - 49,362 Source: NSIA, World Bank staff estimates, ALCS 2016 These distributional changes imply that while the intensity of poverty has increased between 2011-12 and 2016-17 (Figure 11), inequality has declined (Figure 12), as the welfare loss among the top of the distribution has been relatively larger than that at the bottom of the distribution (albeit from very different baseline levels). The poverty gap index measures the extent of poverty as the average distance between the per capita expenditure levels of the population and the poverty line, assuming the non-poor have a zero shortfall, and is and Kohistanat (Sar-e-Pul), Almar, Qaisar and Kohistan (Faryab). expressed as a proportion of the poverty line. As Figure 11, shows, the intensity of poverty has doubled at the national, urban and rural level since 2007. On average, the gap between per capita expenditures and the poverty line was 0.15 times the poverty line (calculated across the population).6 Another interpretation of the poverty gap index is that it provides a measure of the aggregate size of the monetary transfer required to bring the poor out of poverty, assuming perfect targeting were possible. Assuming a national population of 29 million in 2016-17 and using the poverty line of 2,056 Afs per capita per month, a poverty gap index of 0.15 or 15 percent of the poverty line, implies an average transfer of 310 Afs per person per month would be needed to eliminate poverty (and the total budget needed would be 131 million US$ per month, targeted to the poor). The squared poverty gap (Table 1, also known as the poverty severity index), on the other hand, puts a higher weight on the poorest households, those that are further away from the poverty line by averaging the squares of the poverty gaps relative to the poverty line. The squared poverty gap increased by 2 percentage points across survey years, indicating that more people are further away from the poverty in 2016-17 compared to 2007-08. Figure 12 plots the trends in expenditure inequality as measured by the Gini coefficient. The Gini index measures the extent to which the distribution of consumption among individuals or households differs from a perfectly equal one. A value of 0 represents absolute equality with everybody consuming the same amount, a value of 1 absolute inequality, where all consumption is concentrated in one person. The welfare decline experienced across the distribution (shown in Error! Reference source not found.) is reflected here as a decline in the Gini index in 2016-17 relative to 2011-12, driven by the decline in rural inequality, stemming from a compression of living standards. Figure 11: Trends in the intensity of poverty (Poverty gap index) Figure 12: Trends in expenditure inequality (Gini coefficient) National Urban Rural National Urban Rural 0.33 0.16 0.15 0.31 0.29 0.30 0.29 0.11 0.10 0.28 0.29 0.10 0.08 0.07 0.25 0.25 0.05 0.05 2007-08 2011-12 2016-17 2007-08 2011-12 2016-17 Source: NSIA, World Bank staff estimates, NRVA Source: NSIA, World Bank staff estimates, NRVA 2007, 2011 and ALCS 2016 2007, 2011 and ALCS 2016 6 Among the poor, this gap is larger, on average 27 percent of the poverty line. Seasonal variation in welfare Figure 13: Seasonal vs. quarterly poverty rates in 2016-17 Poverty has always had a seasonal character in Afghanistan, with winter months being Quarter Season characterized by a deterioration in welfare7. 67.5% 65.9% 62.7% This pattern holds in 2016-17 as well, with a 55.7% sharp increase in poverty in the autumn and 48.7% 48.2% 46.3% 45.2% winter (Figure 13). In part, this trend could be driven by increases in prices, particularly food prices, over the survey period, peaking in quarter 4, which roughly corresponds to winter (Figure 14 and Figure 15). They may also be explained by a decline in income-generating opportunities from agriculture; and by a Q1, Q2, Q3, Q4, decline in the local availability of food items SPRING SUMMER AUTUMN WINTER in the market during the winter months. Source: NSIA, World Bank staff estimates, ALCS 2016 Figure 14: Quarterly food and non-food price inflation, survey + Figure 15: Quarterly average inflation, as measured by CPI, 2016-17 Nonfood CPI, 2016-17 (Base Q1=100) Survey food price index Overall CPI Food CPI Non-food CPI Non-food price index 110.0 100.06 108.0 100.04 106.0 100.02 104.0 100.00 102.0 99.98 100.0 99.96 98.0 99.94 96.0 1 2 3 4 Q1 Q2 Q3 Q4 Source: NSIA, World Bank staff estimates, ALCS Source: NSIA, World Bank staff estimates 2016 7 While the methodology for poverty measurement divides the survey period into quarters, these quarters closely track seasons in the country, with quarter 1 roughly coinciding with Spring, and quarter 4 with Winter. However, seasonal poverty rates should be used with caution as poverty estimates are based on quarters and not seasons. WHO ARE AFGHANISTAN’S POOR? Household and individual demographic and socio-economic characteristics are important correlates of poverty. This section provides some descriptive statistics on the key correlates of poverty in Afghanistan, while describing the prevalence of these characteristics among the poor and the population as a whole. Demographic Characteristics Consistent with past NRVA surveys, demographic characteristics are strongly correlated with poverty headcount rates. First, poverty rates increase steadily with household size (Figure 16). While a third of households with 1 to 5 members live below the poverty line, roughly 60 percent of households with 8 members or more are poor. Furthermore, smaller households with 1 to 5 members make up only 13 percent of the total population, whereas households with 8 or more members make up more than 60 percent of the population (Figure 17). Households of larger size therefore, are both more prevalent and face a higher poverty rate. Poverty also rises with increasing dependency. Figure 18 plots the share of the population living below the poverty line by child dependency and total dependency ratios. Given the demographic distribution of the country, with roughly 40 percent of the population below the age of 14, the bulk of dependency is accounted for by children, and as a result, the prevalence of poverty is very similar when comparing child dependency or total dependency. As with household size, households with very high dependency, for instance, 3 or more dependents to each working age household member, face rates of poverty as high as 70 percent. Figure 16: Poverty rates by household size, 2016-17 Figure 17: Population share and poor population by household size, 2016-17 Poverty rate Poor population by HH size Population share by HH size 62.1% 61.5% 58.2% 5,000,000 30.0% 50.1% 4,000,000 25.0% 34.6% 20.0% 3,000,000 15.0% 2,000,000 10.0% 1,000,000 5.0% 0 0.0% 10-12 members 8-9 members 1-5 members 6-7 members 13 or more 8-9 MEMBERS 1-5 MEMBERS 6-7 MEMBERS MEMBERS 13 OR MORE members MEMBERS 10-12 Source: NSIA, World Bank staff estimates, ALCS 2016 Source: NSIA, World Bank staff estimates, ALCS 2016 Figure 18: Poverty headcount rates, by dependency ratios, 2016-17 HEADCOUNT RATE BY DEPENDENCY RATIOS 71.4% 71.3% 61.9% 61.1% 59.9% 59.2% 44.1% 43.6% BELOW 1 1-1.9 2-2.9 3 OR ABOVE poverty headcount child dependency ratio poverty headcount dependency ratio Source: NSIA, World Bank staff estimates, ALCS 2016 Note: The child dependency ratio is defined as the number of children aged 0-14 over the population in the most productive ages (15-64). The total dependency ratio is defined as the number of children aged 0-14 and elderly aged 65 and above over the population in the most productive ages (15-64) Education and Labor Market Outcomes Education (or the lack thereof) is another important correlate of poverty in Afghanistan. With only 36 percent of household heads being literate, low levels of educational attainment are pervasive. Households with illiterate heads account for 74 percent of the population, facing poverty rates of 63 percent on average, compared with headcount rates of 40 percent among households with literate heads. Breaking it down further, it becomes evident that the lack of education is both highly correlated with poverty as well as highly prevalent. Approximately 73 percent of the population belongs to households where the head of household has no education (Figure 19). These households account for 82 percent of the poor, facing a poverty rate of 61 percent on average. While poverty does fall with increasing education of the head of household, households where heads have more than secondary education account for only 5 percent of the population. Finally, having an educated household head does not eliminate the risk of poverty: on average, such households have a chance of one in three of being poor. The lack of a strong link between higher education and lower poverty likely reflects the pervasive lack of productive employment opportunities. Overall, 38 percent of the population belongs to households whose heads are either unemployed, under-employed or inactive (Table 3). About 42 percent of the poor population belongs to these types of households. In other words, the employment status of the head of the household does not sharply differentiate poor households from non-poor households. While poverty rates are highest among households with heads who are unemployed (59 percent), they remain high irrespective of the employment status of the head. Figure 19: Poverty rates and share in population, by education level of head of household, 2016-17 Poverty rate Share of poor population Share of poor population 81.9 72.8 61.4 42.9 34.0 29.8 13.1 9.0 8.1 7.1 5.2 2.8 NO EDUCATION COMPLETED COMPLETED UPPER COMPLETED PRIMARY SCHOOL SECONDARY SCHOOL UNIVERSITY OR TEACHER OR TECHNICAL COLLEGE Source: NSIA, World Bank staff estimates, ALCS 2016 Table 3: Poverty rates, share in population, by employment status of head of household, 2016-17 Employment status of head of Share of poor Share of total household Poverty rate population population Employed 51.1 57.6 61.4 Underemployed 63.0 17.6 15.2 Unemployed 58.8 13.9 12.9 Inactive 57.2 10.8 10.3 Source: NSIA, World Bank staff estimates, ALCS 2016 To understand why the poverty rate among households with employed heads is more than 50 percent, we must understand the average quality of employment. Figure 20 plots poverty rates, share of the poor population and share of the total population by the type of job held by employed household heads. Approximately 59 percent of the population belongs to households where the head of household holds vulnerable employment, or in other words, is self-employed or works on own-account, is a day laborer or is an unpaid worker. Only 17 percent of the population belongs to households where heads hold salaried employment or work as employers. It is only when the head of household has a salaried job (15 percent of the population) that poverty rates fall below 50 percent. In contrast, 56 percent of the population belongs to households with heads who are self- employed or day laborers, whose poverty rates are as high as 53 and 66 percent respectively. Figure 20: Poverty rates, share in population, by type of job, employed head of household, 2016-178 Poverty rate Share of poor population Share of total population 65.7 53.3 44.9 44.0 41.6 16.1 15.0 13.4 11.5 DAY LABOURER SALARIED WORKER SELF-EMPLOYED Source: NSIA, World Bank staff estimates, ALCS 2016 While employment of the household head in agriculture continues to be associated with higher poverty rates (63 percent), and accounts for a third of the poor population (and 29 percent of the total population) with employed heads, other sectors are also characterized by high poverty rates. Industry accounts for 11 percent of the poor (and total) population with employed heads of household, with a poverty rate of 58 percent; and the services sector9, with 29 percent of the poor (and 35 percent of the total) population with employed heads, has poverty rates of 45 percent. Figure 21 also highlights the high vulnerability of poverty for households engaged in the agricultural and livestock sector as well as those in wage or day labor (“other work”). Households who derive their main source of income from agriculture, livestock or non-formal employment (such as wage or day labor) have high poverty rates of 65, 66 and 68 percent respectively; almost 60 percent of poor Afghans live in such households. Afghans living in households with a main income source from health, education, or general government services; as well as transport have the lowest poverty rates, yet, only 13 percent of Afghans live in such households. Furthermore, 2 percent of Afghans live in households with a vulnerable main income source (i.e. borrowing, zakat, dowry, begging). Poverty rates for those households are, unsurprisingly, high at 71 percent. On the other hand, remittances provide some monetary relief; Afghans living in households with remittances (about 5 percent of Afghans) as a main income source are relatively better-off with poverty rates of 43 percent. 8 There are too few households with household heads who are unpaid family workers to precisely estimate poverty. They are however, in the same range of poverty rates as households whose household head is self- employed. 9 Service sector includes the following industries: Wholesale and retail trade; transport, storage, communication; financing, insurance, real estate; and community, social and personal services. 64.7 AGRICULUTRE 22.6 19.1 65.7 LIVESTOCK 12.0 10.0 46.8 PRODUCTION AND 4.8 MANUFACTURING 5.6 Poverty rate HEALTH, EDUCATION, 39.9 GENERAL 8.6 GOVERNMENT 11.8 SERVICES 39.0 TRANSPORT 4.7 6.6 Source: NSIA, World Bank staff estimates, ALCS 2016 Figure 21: Poverty rates, share in population, by income source, 2016-17 45.7 SECURITY-RELATED 5.1 Share of poor population SERVICES 6.1 43.1 TRADE 10.7 13.6 68.4 OTHER LABOR 24.4 19.4 Share of total population 70.7 VULNERABLE 2.7 INCOME/TRANSFER 2.0 43.1 REMITTANCES FROM 3.9 MIGRANTS 4.9 ANNEX: THE ALCS AND WELFARE MEASUREMENT IN AFGHANISTAN One of the main objectives of the Afghanistan Living Conditions Surveys (ALCS, formerly the National Risk and Vulnerability Assessment or NRVA) is to provide information on welfare and living standards, on their evolution over time, and their distribution over households. Of particular importance is the measurement and tracking of welfare amongst the poorest segments of the population, and ALCS survey data provide the principal means for estimating the extent and severity of poverty in Afghanistan. Measuring Poverty in Afghanistan: The Cost of Basic Needs Approach The measure of welfare adopted to assess population living standards is based on household expenditures. An individual is considered as poor if their level of consumption expenditures is not sufficient to satisfy basic needs, or in other words, if their consumption expenditure falls below the minimum threshold identified by the poverty line. In line with international standards, the official absolute poverty line for Afghanistan is estimated following the Cost of Basic Needs (CBN) approach and it was set using the NRVA 2007-08. The CBN absolute poverty line represents the level of per capita consumption at which the members of a household can be expected to meet their “basic needs” in terms of both food and non-food consumption.10 To assess the evolution of wellbeing over time, the 2007-08 poverty line was updated to 2011-12 and 2016-17 prices for each of the survey years to reflect changes in the cost of living. Figure 1 below briefly describes the data sources and the estimation methodology, and more details are described in Annex X. It is important to note that the detailed consumption expenditure module, which allows for direct estimation of poverty, was not included in the ALCS 2013-14. Rather, survey-to-survey imputation techniques were used to predict poverty rates for this survey year. The 2016-17 estimates introduced improvements in the methodology, which have been consistently taken backwards to 2011-12 and 2007-08 (survey to survey imputation estimates for 2013-14 have not yet been revised). These comprise of three important changes: (i) In the interest of increased transparency, and in line with international good practice, non-food thresholds are inflated from their 2007- 08 benchmark levels using non-food inflation rates as measured by the official CPI; (ii) Improvements and changes in the survey questionnaire have required small changes in the definition of the welfare measure, which have been consistently revised for all survey years; and (iii) NSIA has made the decision in the interest of transparency to include all provinces in national estimates; while indicating clearly provinces for which estimates are deemed to be of inadequate quality due to the security situation or concerns about data quality. These revisions imply that current estimates may differ from previously released numbers. 10 More specifically, the food component of the poverty line captures the cost of consuming 2,100 Kcal per day following the typical food consumption patterns of the relatively poor; the non-food component of the poverty line is estimated as the median non-food expenditure of individuals with food consumption around the food poverty line. For more details, see methodology note. Figure 1: Data sources for poverty measurement and methodology 2007-08 NRVA: 2013-14 ALCS: 2011-12 NRVA: 2016-17 ALCS: BENCHMARK IMPUTATION OF POVERTY ESTIMATES POVERTY ESTIMATES POVERTY ESTIMATES POVERTY ESTIMATES •Define welfare •Food poverty threshold •Survey did not include •Food poverty threshold aggregate priced 2007 food bundle the detailed prices 2007 food bundle •Define reference at 2011 prices expenditure modules at 2016-17 prices population (2nd-5th •Non-food poverty • Survey to survey •Change: Non-food decile) threshold was re- imputation methods thresholds calculated •Define reference estimated based on were used to estimate based on inflating 2007 quantity bundle 2011 food thresholds poverty non-food threshold •Food threshold: Cost of •Helmand and Khost using non-food inflation purchasing 2100 dropped as measured by CPI kilocalories per capita •Comparable poverty •Consistency: Welfare per day following the trends (32 provinces) aggregate and poverty same consumption line made consistent to pattern as the reference reflect improvements in bundle questionnaire •Non-food threshold: • Coverage: All provinces Average non-food to be included in expenditures of national estimates, with households whose per data quality capita food expenditure assessments is close to the food threshold •Define poverty line Source: For more details, please refer to the methodology note Poverty lines Poverty lines in Afghanistan are estimated at the regional-urban/rural strata level, and the national poverty line is the population weighted average of these regional-strata lines. The classification of provinces into regions for this purpose is shown in Table 1 below. These 8 regions, when split into urban-rural strata, yield 14 region-strata classifications (some regions do not have urban strata), and therefore 14 poverty lines. Table 1: Regions and provinces West Central South East Northeast North West Southwest Central Kabul Ghazni Kunarha Badakshan Balkh Badghis Helmand Bamyan Kapisa Khost Laghman Baghlan Faryab Farah Kandahar Daykundi Logar Paktika Nangarhar Kunduz Jawzjan Herat Nimroz Ghor Panjsher Paktya Nooristan Takhar Samangan Urozgan Parwan Sar-e-Pul Zabul Wardak Figure 2 shows the revised poverty lines for Afghanistan after implementing the methodological changes described above. In 2007-08, the national average threshold for the cost of covering basic needs, the poverty line was 1,261 Afs per person per month. This threshold, consistently defined and valued at 2016-17 survey prices, increases to 2,056 Afs per person per month. Relative to 2007, this represents a 64% increase in the cost of basic needs, comprised of a 68% increase in the cost of the basic food bundle (benchmarked at 2,100 kilocalories per person per day), and a 58% increase in the cost of non-food necessities. Figure 2: Poverty lines (weighted national average), Figure 3: Implied average inflation (relative to 2007 to 2017 2007=100) based on poverty lines Food Non-food Total Food Non food Total Afs per person per month 168.43 2,056 163.69 1,758 157.67 146.37 1,261 1,188 139.16 1,034 868 130.01 706 724 555 100.00 2007-08 2011-12 2016-17 2007-08 2011-12 2016-17 Source: NSIA, World Bank staff estimates, NRVA Source: NSIA, World Bank staff estimates, NRVA 2007, 2011 and ALCS 2016 2007, 2011 and ALCS 2016