Ethiopia Economic Update 7 SPECIAL TOPIC: Poverty & Household Welfare in Ethiopia, 2011-16 i Acknowledgements The macro section of this Ethiopia Economic Update was prepared by Nora Dihel, Zerihun Getachew Kelbore, and Samuel Mulugeta. The special topic of this Ethiopia Economic Update was prepared by a team consisting of Tom Bundervoet, Berhe Beyene, and Arden Finn. The report was prepared under the general guidance of Mathew Verghis, Pierella Paci, Nataliya Mylenko, and Carolyn Turk. Comments were received from Jean-Pierre Chauffour. Special thanks go out to Manex Bule Yonis and Asmelash Haile Tsegay for invaluable advice and assistance on data processing and GIS mapping. The team is thankful to Ethiopia’s Central Statistics Agency for making available the data upon which this report is based. The findings, interpretations, and conclusions expressed herein are those of the authors, and do not necessarily reflect the views of the World Bank’s Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this report, nor does it endorse the boundaries, colors, denominations, and other information shown on any map in this report. ii Table of Contents Acknowledgements.................................................................................................. i Abbreviations and Acronyms ....................................................................................... v Executive Summary .................................................................................................. vi Recent Economic Developments and Outlook ............................................................. vii Special Topic: Poverty and Household Welfare in Ethiopia, 2011–16 .......................... xi PART ONE 1 1. Macroeconomic Developments and Outlook ......................................................... 1 1.1 1.1 Global and Regional Trends ............................................................................... 2 1.2 1.2 Recent Economic Developments and Outlook ...................................................... 3 1.3 1.3. Ethiopia’s Economic Prospects in FY2019 and Beyond ........................................ 32 PART TWO 2 2. Special Topic: Poverty and Household Welfare in Ethiopia, 2011–16 ....................... 38 2.1 2.1 Introduction ....................................................................................................... 39 2.2 2.2 Trends in Household Welfare, 2011–16 ............................................................... 39 2.3 2.3 Patterns of Growth in Consumption Expenditure ................................................... 46 2.4 2.4 Brief Snapshot of Current Monetary Living Standards ........................................... 53 2.5 2.5 Perspectives on Poverty Reduction Going Forward ................................................ 55 Annex A Trends in Mean Consumption Expenditures .................................................. 57 Annex B The “Five Ethiopias” .................................................................................... 60 REFERENCES ......................................................................................................... 61 List of Figures Figure 1.1 Economic Activity, Contribution to GDP Growth ............................................ 3 Figure 1.2 Crop Growth during the Main (Meher) Season ............................................. 4 Figure 1.3 Productivity Growth of Major Crops ............................................................. 4 Figure 1.4 Sectoral Shares in Gross Domestic Product ................................................... 9 Figure 1.5 Service Sector Performance ......................................................................... 11 Figure 1.6 Inflation ..................................................................................................... 12 Figure 1.7 Broad Money and Reserve Money ................................................................ 13 Figure 1.8 Contribution to Broad Money Growth .......................................................... 14 Figure 1.9 Composition of Domestic Credit .................................................................. 14 Figure 1.10 National Bank of Ethiopia Advances to the Government ................................ 15 Figure 1.11 Government Spending Moderated ................................................................ 20 iii Figure 1.12 Revenue Growth Remains Weak .................................................................. 21 Figure 1.13 Public Debt, FY2018 .................................................................................... 25 Figure 1.14 Exports of Goods and Services ..................................................................... 26 Figure 1.15 Number of Firms ........................................................................................ 27 Figure 1.16 Average Production and Exports .................................................................. 27 Figure 1.17 Imports of Goods ....................................................................................... 27 Figure 1.18 Balance of Payments ................................................................................... 28 Figure 1.19 Nominal Exchange Rate of the Birr against the U.S. Dollar ............................ 29 Figure 1.20 Real Effective Exchange Rate Index ............................................................... 29 Figure 1.21 Share of Manufacturing FDI in Total FDI, 2010/08–2016/17 ........................ 30 Figure 1.22 Gross Official Foreign Exchange Reserves ................................................... 32 Figure 2.1 Consumption Increased between 2011 and 2016, Particularly in Urban Areas ... 41 Figure 2.2 Consumption Increased in All Agro-Ecological Zones, Particularly in Urban Areas .. 43 Figure 2.3 Calorie Consumption Showed a Modest Increase between 2011 and 2016 ... 44 Figure 2.4 Consumption Growth Was Zero for the Poorest and Strongly Positive for the Richest ................................................................................................. 47 Figure 2.5 Growth Was Strong and Positive in Urban Areas but Lower and Variable in Rural Areas ................................................................................................ 48 Figure 2.6 Growth Was Negative for the Bottom 15 Percent in Rural Areas between 2005 and 2011 .................................................................................................... 48 Figure 2.7 Patterns of Growth Differed Widely across Agro-Ecological Zones ................. 52 Figure 2.8 Real Consumption Increased Everywhere Except the First Decile .................... 54 Figure 2.9 Monetary Living Standards Remain Low for the Majority of the Population ...... 54 List of Tables Table 1.1 Annual Growth Forecast .............................................................................. Table 2.1 Household Consumption Increased in All Regions Except Afar and Amhara ... 34 42 iv v Abbreviations and Acronyms CAB current account balance CPI Consumer Price Index DSA Debt Sustainability Analysis ECX Ethiopian Commodity Exchange ESS Ethiopia Socioeconomic Survey FDI foreign direct investment FDRE Federal Democratic Republic of Ethiopia FEWS NET Famine Early Warning System Network FY fiscal year G&S goods and services GDP gross domestic product GERD Grand Ethiopian Renaissance Dam HCES Household Consumption Expenditure Survey HH households IMF International Monetary Fund kcal kilocalories MOFEC Ministry of Finance and Economic Cooperation MOFED Ministry of Finance and Economic Development NBE National Bank of Ethiopia NPC National Planning Commission PPP purchasing power parity REER real effective exchange rate RHS right-hand side SNNPR State of Southern Nations, Nationalities, and Peoples' Region SOEs State-owned enterprises UN DESA United Nations Department of Economic and Social Affairs VCI Vegetation Condition Index WEO World Economic Outlook WTO World Trade Organization vi Executive Summary vii RECENT ECONOMIC DEVELOPMENTS AND OUTLOOK Recent Economic Developments Ethiopia’s real gross domestic product (GDP) growth, while still strong, decelerated to 7.7 percent in FY2018. A slowdown in industrial growth, mainly driven by lower growth in construction due to foreign exchange shortages and higher prices of imported construction materials, coupled with weaker performance of the manufacturing and the agriculture sectors, explains to a large extent the growth deceleration. Services sector exhibited strong growth in FY2018 while manufacturing underperformed. In addition to the prevailing structural- and trade logistics– related challenges, severe foreign exchange shortages and political unrest affected the manufacturing sector over the past three years, with the magnitude of the effect peaking in FY2018. As a result, manufacturing growth has dropped drastically, from 24.7 percent in FY2017 to 5.5 percent in FY2018.¹ Large and small-scale industries underperformed dramatically. By contrast, growth of services increased from 7.5 percent in FY2017 to 8.8 percent in FY2018, with real GDP growth driven mainly by services (3.4 percent). Agricultural growth decelerated from 6.7 percent in FY2017 to 3.5 percent in FY2018, mainly due to slower growth of crop production. On the demand side, government and private consumption declined by about 1 percentage point each. 1 The drastic fall in growth may also be explained by measurement issues. See box 1.1, in part one, for details on inconsistencies in GDP estimates in the System of National Accounts. viii Inflation declined in the first half of FY2019 reversed in recent months, with the REER showing after surging to double digits in FY2018. a quarter-on-quarter appreciation of 4.8 percent Since peaking at 16.8 percent in June 2018, due in the first quarter of FY2019, as the inflation to the expansion of public sector credit in 2017, differential with trading partners remains high. broad money growth, pass-through of the October 2017 devaluation, and political disruptions The government continued to adhere to (which adversely affected distribution networks), a cautious fiscal policy stance in FY2018. inflation has been on a declining trend. It reached According to the latest official data, the federal a low of 10.4 percent in December 2018. The government’s fiscal deficit amounted to 3 percent of price deceleration has been more rapid for food GDP in FY2018, lower than the 3.4 percent recorded products, which decreased from 17.8 percent in in FY2017. Although expenditure was implemented June 2018 to 11.4 percent in December 2018, as budgeted, revenue decreased as a percentage of than nonfood items, whose prices decreased only GDP and in real terms, mainly due to the slow pace marginally (from 11.1 to 9.1 percent during the of tax reforms. The federal government fiscal deficit period). There was a slight uptick in inflation to that accumulated during FY2018 was financed 10.8 percent in January 2019. through external—mainly concessional—financing and domestic financing with large repayments of The National Bank of Ethiopia (NBE) has cash balances and residuals. resumed its policy of gradual depreciation of the birr against the U.S. dollar in FY2019, Total exports increased in FY2018, thanks after keeping the exchange rate largely to the good performance of service exports. constant in FY2018 following the October Goods exports declined due to the weak performance 2017 devaluation. Following the 15 percent of coffee. Coffee represented 30 percent of goods one-off devaluation in October 2017, the exports and decreased by 5 percent in FY2018, nominal exchange rate was kept largely constant mainly because of a drop of more than 10 percent for the remaining period of FY2018. The official in coffee prices. Export increases were recorded in exchange rate against the U.S. dollar depreciated the leather, textile and garments, chemical, and by less than 1 percent between October 2017 electricity sectors, reflecting in part the coming on and June 2018. The NBE allowed the birr to stream of the industrial parks. However, at US$2.9 depreciate by about 2.9 percent against the U.S. billion, manufacturing exports remain well below dollar during the first six months of FY2019. The the targeted US$7 billion. The underperformance premium between the official and parallel market of goods exports was mainly due to structural exchange rates widened in FY2018, reaching a and competitiveness issues, such as rigid labor peak of about 30 percent toward the end of the and product markets, including an overvalued fiscal year. However, there was a sharp narrowing exchange rate. However, Ethiopia’s total exports of the premium in June/July (reaching about 5 expanded by 13.1 percent in FY2018, after years percent), following the government’s call for the of underperformance. This sharp recovery in exports public to exchange foreign currency holdings was driven by the growth in service exports, which in commercial banks. The premium has again accounted for half of total exports and expanded by started to widen in recent months, returning to 26.6 percent, mainly due to the strong performance its previous level of about 30 percent. The real of Ethiopian Airlines. Despite these positive effective exchange rate (REER) depreciated by 7.1 developments, total exports of goods and services percent in FY2018, reversing the appreciating do not exceed 10 percent of GDP , significantly below trends of the past six years, mostly aided by the the 24 percent expected from a country the size of devaluation. However, the depreciating trend has Ethiopia at this level of development. ix Following an increase of 13 percent in percent compared with 54.9 percent in FY2017. FY2018, total exports continued to expand The increase was mainly driven by external debt, by 18 percent in the first half of FY2019, which increased from 27.5 to 29.2 percent of GDP , driven by services. Led by the performance of while domestic debt marginally increased from Ethiopian Airlines, service exports increased by 41 27.4 to 27.8 percent of GDP . Despite the strong percent during the first half of FY2019 compared outturn in service exports in FY2018, external with the same period in FY2018. By contrast, debt vulnerabilities continue to arise from lower- merchandise exports decreased by 10 percent, than-expected export performance, keeping the reflecting the continued poor performance of two export-related indicators in the 2018 Debt Ethiopia’s major export items, including coffee, Sustainability Analysis (DSA) above their respective oilseeds, and pulses. Manufacturing export thresholds in the baseline. As a result, the 2018 performance was relatively strong, with exports DSA maintained that Ethiopia remains at high of textiles and electronics posting growth rates of risk of external debt distress, as was the case in more than 35 percent. the 2017 DSA. Ethiopia’s overall public debt showed an Poverty continued to decline, particularly increase in FY2018 but moderated in the in urban areas, supported by strong first quarter of FY2019. Public debt declined GDP growth. According to the most recent from 57 percent of GDP in June 2018 to 54 percent Household Living Standards Survey, 27 percent in the first quarter of FY2019. External debt, which of Ethiopians lived below the international constituted about 55 percent of total public debt, poverty line (US$1.9 per capita per day, 2011 decreased from 29.2 percent of GDP to 27.8 purchasing power parity (PPP)) in 2015/16, percent, while domestic public debt decreased a decline from 34 percent in 2010/11. The from 27.8 to 26.1 percent of GDP between reduction in poverty was strong in urban areas, June and September 2018. In FY2018, public but substantially slower in rural areas, parts of debt as a percentage of GDP increased to 57 which were affected by the El Niño drought. Ethiopia’s real GDP growth remains strong despite decelerating to 7.7 percent in FY2018. x Outlook national development plan, so they will require public-private partnerships. The economic prospects for FY2019 and the medium term are expected to remain The federal government approved the FY2019 stable. The current reform agenda focusing on budget with a deficit targeted at 3.3 percent boosting private sector participation aims to support of GDP . Although the deficit targeted for FY2019 is growth. Annual real GDP growth is projected to be higher than the fiscal deficit for FY2018, the FY2019 around 7.9 percent in FY2019 and 8.2 percent in budget assumes a significant improvement in revenue the medium term. The reform agenda proposed by collection while maintaining expenditures, especially the new prime minister is expected to address some capital expenditures, at the FY2018 level. Of the macroeconomic imbalances, while moderate fiscal total budget approved for FY2019 (Br 346.9 billion), deficits and prudent monetary policy are expected 26.4 percent is allocated to recurrent expenditure, to reduce the rate of inflation and keep it in the 32.8 percent to capital expenditure, and 39 percent single digits. Merchandise exports could recover to regional transfers. Sixty-six percent of the budget in the medium term, as large investment projects, is allocated to pro-poor expenditures focusing on such as the railway to the Port of Djibouti and large education, health, industrial development, and the power dams (with potential for electricity exports) Urban Productive Safety Net. become operational. The government is expected to continue On the supply side, growth in all sectors is its cautious fiscal policy stance. Government projected to be robust. Agriculture is projected revenue is projected to increase gradually to continue growing at about 4 percent, as national following tax reforms and improvements in tax main season crop production for 2018/19 is compliance, while government expenditure, expected to be average, following forecasts of mainly capital expenditure, is expected to stabilize. average rainfall during the main season (meher) As a result, the fiscal deficit would remain at 3 and belg season, except for localized areas in the percent in FY2019 and further improve to reach northeastern and eastern parts of the country, which 2.9 percent in FY2020. Although the general may face below-average seasonal rainfall (FEWS government fiscal policy stance is expected to NET 2018). Industrial growth is expected to decline remain relatively cautious (the FY2019 budget slightly in FY2019, to about 11 percent. The service does not allow for any new investment projects), sector is projected to continue expanding, due to domestic revenue may be insufficient to finance the spillover effects from growth in other sectors. the infrastructure investments foreseen in the national development plan. This would require In line with the government’s reform, the innovative forms of finance for development. private sector is expected to play an active role. Private investment is projected to increase by Poverty is expected to continue its steady 10.6 percent, while government investment (public decline on the back of sustained economic investment) is expected to decelerate. Household growth and better climatic conditions for and government consumption are projected to agriculture. However, the pace of poverty decelerate temporarily. Although the general reduction is expected to be relatively slow, given government fiscal policy stance is expected to the weak transmission between overall GDP remain relatively cautious (the FY2019 budget growth and poverty reduction in Ethiopia. Based does not allow for any new investment projects), on past elasticities, poverty as measured by the domestic revenue may be insufficient to finance international poverty line is projected to decrease the infrastructure investments foreseen in the to 22 percent by 2020. xi Risks and Challenges medium-term growth. The rising risk to external debt sustainability may impact Ethiopia’s access The medium-term sustainability of the to external finance. growth model poses an important risk. The key economic challenge relates to limited Political disruption associated with social competitiveness, which may constrain the unrest could derail the new reform agenda development of manufacturing, creation of jobs, and negatively impact growth through lower foreign and increase of exports. The pursuit of sound, direct investment, tourism, and exports. The outlook private sector–led and export-oriented economic for poverty reduction, while broadly positive, also policies is therefore critical for achieving Ethiopia’s faces significant downside risks, given continued economic and social ambitions. localized turmoil and large-scale displacement. As the special topic of this seventh Economic Update Rising external imbalances constitute highlights, achieving stronger and broader-based major challenges to the economy. The lack of improvements in household welfare will require external openness, rising international oil prices, strengthening the link between overall economic and an overvalued exchange rate (even after the growth and increased opportunities for social and 2017 devaluation) will continue to have adverse economic advancement for Ethiopia’s young and effects on competitiveness, affecting short-to- rapidly growing population. SPECIAL TOPIC: POVERTY AND HOUSEHOLD WELFARE IN ETHIOPIA, 2011-16 Improvements in Household The increase in consumption was broadly shared across regional states, except for Welfare, Challenges in Reaching Amhara and Afar, where median household the Rural Poor consumption levels decreased. There was a pronounced urban-rural difference within regional Monetary welfare levels of Ethiopian states, with urban consumption levels growing households continued to improve between everywhere but rural consumption increasing only 2011 and 2016 despite adverse climatic in Tigray, Gambella, and Harari (in a statistically circumstances linked to the El Niño drought. significant fashion). Household welfare levels Median household consumption expenditures increased in the highland and lowland regions of per adult equivalent increased at the national, Ethiopia, with a sharper increase in the highlands. urban, and rural levels, although the increase was a lot stronger in urban than rural areas. The Household consumption growth between consumption expenditure of urban households 2011 and 2016 was stronger at the upper increased at an annual average of 6 percent, while end of the distribution. At the national level, that of rural households increased at less than 1 the poorest 10 percent of the population did not percent per year. Overall, Ethiopian households’ experience any consumption growth between consumption levels increased by 2 percent per 2011 and 2016, while the top 10 percent grew at year between 2011 and 2016. an average rate of 5 percent per year. The national xii picture largely reflects the dynamics in rural areas, was strong growth across the distribution, although where the real consumption of the bottom 20 percent it was particularly strong for the better-off. As a result did not change between 2011 and 2016 (figure ES.1, of the growth patterns, inequality in consumption panel b), a finding similar to the rural consumption levels somewhat increased, but it remains low in dynamic during 2005–11. In urban areas, there regional comparison. Figure ES.1. Strong Growth in Urban Areas, but Low and Variable Growth in Rural Areas, 2011–16 (average annual growth rates of consumption by percentile and urban/rural location) A Urban B Rural 8 8 7 7 Annual mean growth rate (%) Annual mean growth rate (%) 6 6 5 5 4 4 3 3 2 2 1 1 0 0 -1 -1 -2 -2 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 Percentiles Percentiles 95% confidence bounds 95% confidence bounds Sources Household Consumption Expenditure Surveys 2011, 2016; World Bank staff calculations Despite 20 years of growth and development and US$2.2 a day, respectively (figure ES.2). that have substantially reduced poverty Between 2005 and 2016, while consumption levels levels, the monetary living standards of were increasing overall, the bottom 15 percent in Ethiopian households remain low. This is rural areas did not experience real consumption especially true in rural areas and mainly for the growth. The severity of rural poverty was higher bottom 40 percent. Expressed in U.S. dollars and in 2016 than in 2005, implying the existence of adjusted for purchasing power, Ethiopians in the two a segment of the rural population that has yet to bottom quintiles of consumption lived on US$1.3 benefit from Ethiopia’s development (FDRE 2017). xiii Figure ES.2. Monetary Living Standards Remain Low for the Majority of the Population (daily expenditures per capita, US$ PPP) 8 7.2 Expenditures per capita per day 7 6 5 3.6 4 2.8 3 2.2 2 1.3 1 0 Q1 Q2 Q3 Q4 Q5 Quintiles of consumption per capita Sources Household Consumption Expenditure Survey 2016; World Bank staff calculations. Education and Jobs—Key to it is difficult to imagine that the living standards of today’s poor children or youth will improve Boosting Household Welfare much as they enter adulthood and start their own Going Forward families. Devising policies or interventions to keep children from poor rural families in school longer Further improving the welfare levels of will be crucial in any attempt to share the benefits Ethiopian households will require, among of growth more widely. others, more effective investments in human capital, especially education. In a In a context of increasing land scarcity, scenario of development and transformation, the labor market will become increasingly children of poor households would accumulate important as a transmission mechanism more education and be able to move out and between aggregate growth and poverty diversify into more productive activities, breaking reduction. The Government of Ethiopia the intergenerational transmission of poverty. acknowledges this and expanding employment Better-educated youth from rural areas tend to opportunities is a priority area in the second move to urban areas and improve their material Growth and Transformation Plan. The challenge conditions. Despite the government’s large is daunting. Over the next 10 years, the working- investments in education, few youths in the rural age population in Ethiopia will grow at two million areas of Ethiopia complete primary school. per year, dramatically increasing the population’s According to the 2015/16 Welfare Monitoring demand for jobs. Even if Ethiopia’s manufacturing Survey, 22 percent of youths ages 15-24 years drive turns out to be highly successful, the in rural Ethiopia had completed primary school, number of jobs created will not come close to dropping to 14 percent in the bottom quintile (the the number required. Successfully addressing the poorest in rural areas). Overall, the median years employment challenge will require interventions, of education obtained by rural youth (ages 15- improvements, and innovations in many sectors 24) in Ethiopia amounts to five (lower for those in and dimensions and a critical review of existing poor households). With these education statistics, employment policies and strategies. 1 Part 1: Macroeconomic Developments and Outlook 2 1.1 Global and Regional Trends Global growth is projected at 3.7 percent of the commodity price slump. Although the for 2018, continuing the steady expansion region’s economic recovery continues, it does so underway since mid-2016. However, the at a slower pace than expected (0.4 percentage expansion has become less balanced and may point lower than the April 2018 forecast). On have peaked in some major economies. Downside the supply side, the moderate recovery reflected risks to global growth have risen in recent months higher oil prices and better agricultural conditions in a context or rising political uncertainties. The following droughts. On the demand side, growth reversal of capital flows to emerging market was supported by consumer spending amid eased economies with weaker fundamentals and higher inflation and public investment—especially among political risk has become more pronounced, while non-resource-rich countries. However, the external trade barriers are on the rise. Although financial environment is becoming less favorable for Sub- market conditions remain accommodative Saharan Africa, as global trade and industrial in advanced economies, they could tighten production are losing momentum. The prices rapidly if, for example, trade tensions and policy of metals and agricultural commodities fell due uncertainty were to intensify. Monetary policy is to concerns about trade tariffs and weakening another potential trigger, and tighter financial demand prospects, while oil prices were on an conditions in advanced economies could cause upward trend and are likely to remain elevated disruptive portfolio adjustments, sharp exchange in 2019. Financial market pressures intensified rate movements, and further reductions in capital in some key emerging markets and developing inflows to emerging markets (IMF 2018b). economies. Concern about dollar-denominated debt has risen among emerging markets amid a With risks shifting to the downside, there stronger U.S. dollar. is greater urgency for policies to enhance prospects for strong and inclusive growth. Public debt levels remained high and According to the 2018 World Economic Outlook, continued to rise in several Sub-Saharan avoiding protectionist reactions to structural African countries. Changes in the composition change and finding cooperative solutions that of debt—characterized by growing liabilities promote continued growth of trade in goods and owed to non–Paris Club governments and private services remain essential to preserve the global creditors—have increased the vulnerability of expansion. With shrinking excess capacity and public debt sustainability to weaker currencies mounting downside risks, many countries need to and higher global interest rates. Although the rebuild fiscal buffers and strengthen their resilience external positions of oil exporters improved, they to an environment in which financial conditions weakened in metals exporters and non-resource- could tighten suddenly and sharply (IMF 2018b). rich countries. Economic growth in Sub-Saharan Africa is Growth in Sub-Saharan Africa is expected estimated to have picked up to 2.7 percent to rise to 3.3 percent in 2019. Despite in 2018 from 2.3 percent in 2017, barely relatively solid growth, average growth per above population growth (World Bank Group capita will remain weak, pointing to continued 2018). The acceleration relative to 2017 reflects slow progress in poverty reduction. Structural stronger global growth, higher commodity prices, constraints hinder a stronger rebound in the and improved capital market access, following region’s largest economies, and growth is efforts to improve fiscal balances in the aftermath expected to rise moderately in 2020 to reach 3.6 3 percent. Risks to the growth outlook are tilted to the commodity prices, an abrupt tightening of global downside, reflecting vulnerabilities related to tighter financial conditions, and escalating trade tensions global financial conditions, the reversal of capital between major economies. The main domestic risks inflows, and higher financing costs. Key external are fiscal slippage, domestic conflicts, and weather risks include an unexpectedly sharp decline in shocks (World Bank Group 2018). 1.2 Recent Economic Developments and Outlook Real Sector in Ethiopia Real gross domestic product (GDP) growth, growth deceleration. For example, main season while still strong, decelerated to 7.7 percent crop production, which covers about 90 percent in FY2018 (Figure 1.1. ). A slowdown in industrial of total crop production, recorded growth of 5.4 growth, mainly driven by lower growth in construction percent, which was lower than the growth rate due to foreign exchange shortages and higher reached in FY2017 (a recovery year) (Figure 1.2.). prices of imported construction materials, coupled Similarly, although crop productivity growth was with weaker performance of the manufacturing and generally strong, it remained below the growth in the agriculture sectors explains to a large extent the FY2017 (Figure 1.3.). Figure 1.1. Economic Activity, Contribution to GDP Growth (percent) 10.4 10.6 10.3 11 10.2 10.1 0.9 0.9 0.6 1.3 1.2 9 0.7 8.1 3.5 0.6 2.0 0.7 7.7 1.6 0.9 1.1 2.8 1.1 7 0.5 0.7 0.3 3.4 2.2 3.2 1.8 5 4.4 2.6 0.2 2.5 0.1 1.3 0.7 1.5 3 0.7 2.8 0.7 0.8 1.7 3.1 1.0 2.5 0.4 1 2.2 2.3 2.6 0.9 1.3 0.0 -0.9 0.0 -0.1 -0.1 -1 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 2017/18 Agriculture Manufacturing Construction Other industry Trade & hotel Transport & communications Other services GDP Source National Planning Commission. 4 Figure 1.2. Crop Growth during the Main (Meher) Season (percent) 20 16 12 8 4 0 -4 -8 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 2017/18 Cereal Pulses Oilseeds Grains Source Central Statistical Agency. Figure 1.3. Productivity Growth of Major Crops (percent) 8 6 4 2 0 -2 2014/15 2015/16 2016/17 2017/18 Teff Wheat Maize Sorghum Source Central Statistical Agency 5 Real GDP growth was mainly driven by Manufacturing underperformed . In services. Of the 7.7 percent real GDP growth, addition to the prevailing structural- and trade the service sector contributed 3.3 percentage logistics–related challenges, foreign exchange points, followed by industry at 3.1 percentage shortages and political unrest severely affected points, and agriculture at 1.3 percentage points. the manufacturing sector over the past three Despite good crop production (5.4 percent years, with the magnitude of the effect peaking annual growth), agricultural value added in FY2018. As a result, manufacturing growth dropped by half. Compared with the previous dropped drastically, from 24.7 percent in FY2017 year, the growth contribution of the agriculture to 5.5 percent in FY2018. Large and small-scale sector dropped by 50 percent, and that of industries underperformed dramatically in FY2018 industry dropped by 34 percent. compared with their historical five-year average growth rates. Between 2012/13 and 2016/17, Industrial sector growth slowed. Industrial large and small-scale industries registered average sector growth slowed due to underperformance growth of about 22 and 9.9 percent, respectively. in the manufacturing and construction subsectors. However, the growth rates of large and small- Between 2012/13 and 2016/17, the industrial scale industries in 2017/18 (6 and 4.6 percent, sector grew on average by 22 percent. However, respectively) were far below the historical trends, the growth rate dropped to 12.2 percent in dropping by 73 and 54 percent, respectively, 2017/18, showing a cut by 40 percent from the leading to a drop in manufacturing growth. The growth rate in 2016/17, and a drop by 45 percent drastic fall in growth may also be explained by from the historical five-year average growth of the measurement issues; box 1.1 discusses how to sector. The construction subsector, the main driver deal with the questions raised by inconsistencies in of industrial sector growth, grew by 28 percent GDP estimates in the System of National Accounts. between 2012/13 and 2016/17. However, the By contrast, service sector growth increased from growth rate in 2016/17 declined to 15.7 percent, 7.5 percent in FY2017 to 8.8 percent in FY2018. showing a drop of 24 percent compared with the Agricultural growth decelerated from 6.7 percent previous year, and a drop of 44 percent from the in FY2017 to 3.5 percent in FY2018, mainly due five-year historical growth rate. to slower growth of crop production. Box 1.1 How to Deal with Inconsistencies in GDP Estimates in the System of National Accounts? In principle, a change of base year in the changes in the size of GDP , growth rates, national accounts involves changing the price and sectoral contributions. and quantity base for the individual price and quantity relatives and updating the weights In FY2018, the National Planning Commission used in aggregating the individual quantity (NPC) revised the base year from 2010/11 relatives into sub indexes. At the same time, to 2015/16. Due to rebasing, nominal GDP the base year change serves to reconcile has shown a 2.6 percent increase (figure the different estimates of gross domestic B1.1.1), while GDP at constant market prices product (GDP) to enable methodological increased by 94 percent, from Br 810 billion and conceptual reviews and improvements. to Br 1.7 trillion over the same period. The recent rebasing in Ethiopia has led to 6 Figure B1.1.1 Comparison of Nominal GDP in 2010/11 and 2015/16 Base Years 1.58 1.57 1.56 GDP (birr, trillions) 1.55 2.6% 1.54 1.53 1.52 1.51 1.50 Base year 2010/11 Base year 2015/16 The revision of GDP at constant prices by An in-depth sectoral analysis reveals that activity has brought significant changes in value added in small and cottage industries the value added of all activities, most notably increased by 51 percent; value added in in the service sector: private households hotels and restaurants increased by 25 with employed persons increased by more percent; and the value addition in financial than nine-fold; in the industrial sector, intermediation decreased by 17 percent. The construction and small and cottage industries 2016/17 growth rate for small and cottage increased by more than twofold; agriculture industries of 2.5 percent was revised to 36.9 doubled; and most activities have seen percent in 2018; by contrast, for education, similar increases (figure B1.1.2). As a result, the 2016/17 growth rate of 11 percent was GDP in constant prices doubled. revised down to -3.2 percent in 2018. Most importantly, after the rebasing in The growth estimates for 2017/18 bring 2016/17, real GDP growth for 2016/17 was out further inconsistencies. Agriculture reported to reach 10.9 percent. However, value addition in constant prices was the growth rate for 2016/17 was reduced to estimated to increase only by 3.5, despite 10.1 percent in the recent NPC report. In the crop production growth of 5.4 percent; new GDP estimate for 2016/17, value added furthermore, livestock value addition was for agriculture and industry was revised estimated to drop by 0.6 percent, from 4.2 upward by 2 percent, while value added percent growth in the previous year, without for services was reduced by 0.2 percent. a proper explanation. Incorrect GDP estimates lead to policy mismanagement. 7 Figure B1.1.2 Change in 2015/16 GDP between the 2010/11 and 2015/16 Base Years, by Industry GDP at constant market prices Taxes on products GVA at constant basic prices Private households with employed persons Other community, social & personal services Health & social work Education Public administration & defense Real estate, renting & business activities Financial intermediation Transport & communications Hotels & restaurants Wholesale & retail trade Construction Electricity & water Small-scale & cottage industries Large & medium scale manufacturing Manufacturing Mining & quarrying Fishing Forestry Animal farming & hunting Crop Agriculture, hunting & forestry -200% 0% 200% 400% 600% 800% 1,000% It is evident that, due to the change in the base highest drops are observed in the shares year from 2010/11 to 2015/16, the shares of hotels and restaurants and real estate, of industry and agriculture increased while renting, and business activities. The declines the share of services declined. Yet, services in wholesale and trade and hotels and continue to account for the highest share in restaurants might be due to the high cost GDP in both base years. In the agriculture of inputs that may have reduced the value sector, the share of crops declined, while in added despite the expansion of activities. the service sector, the shares of wholesale The highest increases are observed in small- and retail trade and hotels and restaurants scale and cottage industries and construction declined in the 2015/16 base year. The (figure B1.1.3). 8 Figure B1.1.3 Percentage Shares of 2015/16 GDP , between 2010/11 and 2015/16 Base Years, by Industrial Classification Agriculture, hunting & forestry Crop Animal farming & hunting Forestry Fishing Industry Mining & quarrying Manufacturing Large & medium-scale manufacturing Small-scale & cottage industries Electricity & water Construction Services Whole sale & retail trade Hotels & restaurants Transport & communications Financial intermediation Real estate, renting & business activities Public administration & defense Education Health & social work Other community, social & personal services Private households with employed persons 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0 50.0 Base year 2010/11 Base year 2015/16 In general, incorrect GDP estimates can lead to foreign exchange reserves. Technical assistance mismanagement of monetary policy, including to the NPC and other institutions could help inadequate measures to deal with inflation, and address remaining issues with the rebasing and can provide confusing signals for managing forthcoming GDP forecasts. 9 Ethiopia has experienced a remarkable workforce from agriculture to manufacturing and shift from agriculture to services and services that is consistent with a long tradition industry over the past 15 years. The share in development economics. Traditionally, poor of agriculture in GDP declined from 55 percent countries undergo a process of structural in 2003/04 to 35 percent in 2017/18, while the change, with labor reallocating from traditional, share of industry more than doubled, from 13 to low-productivity sectors of the economy toward 27 percent, and the share of services increased modern, high-productivity sectors to achieve from 33 to 39 percent (figure 1.4). However, high levels of aggregate productivity (Lewis Ethiopia did not witness the movement of the 1954; Chenery et al 1986). Figure 1.4. Sectoral Shares in Gross Domestic Product (percent) 100 90 80 70 60 50 40 30 20 10 0 2001/02 2002/03 2003/04 2004/05 2005/06 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2000/01 2006/07 2007/08 2008/09 2009/10 2016/17 2017/18 2019/20 1990/00 Agriculture Industry Services Source National Planning Commission. Note The values for 2019/20 are targets of the second Growth and Transformation Plan. The reallocation of labor from agriculture sector that would result in rapid migration of labor to manufacturing has been constrained by to the urban informal service sector. Between 2005 the slow pace of manufacturing expansion. and 2013, the share of employment in agriculture Agriculture continues to dominate employment, declined by 7.5 percentage points, whereas the with a minimal decline in the employment share share of agriculture value added dropped by 11.5 in agriculture between 2005 and 2013. This points percentage points. Over the same period, the to inadequate expansion of the manufacturing employment share of the service sector increased 10 by 8.7 percentage points, absorbing the labor recent unemployment data. After a long period released from agriculture and industry, and of decline, there was an increase in urban increased its share of value added in GDP by 5.4 unemployment in Ethiopia between 2016 and percentage points. The longer-run sustainability 2018, mainly in female and youth unemployment of the Ethiopian model is further undermined by (box 1.2). Box 1.2 Labor Market Developments and Poverty in Ethiopia: Increasing Unemployment, Especially among Women and Youth Getting a job is often an important pathway three consecutive years of decline, urban out of poverty in Ethiopia. The official unemployment increased in 2018, with the urban unemployment rate increased from female unemployment rate (26.4 percent) 17.5 percent in 2012 to 19.1 percent in more than double the male rate (12.2 2018 (figure B1.2.1). Interestingly, after percent). Figure B1.2.1 Unemployment Rate in Urban Areas 30 26.4 24.2 24.1 23.8 23.8 25 19.1 Unemployment Rate 20 17.5 17.4 16.8 16.9 15 11.4 12.2 11.3 10.4 9.4 10 5 0 March 2012 April 2014 March 2015 April 2016 June 2018 Both sexes Male Female Source Central Statistical Agency 2018. Young people are also particularly affected 2018 compared with 19 percent of young men. by unemployment. At 25.3 percent, the unemployment rate for youth is higher than the Among the regions, Dire Dawa has the highest national average of 19.1 percent, with female unemployment rate at 25.3 percent, followed youths bearing the brunt of unemployment: 30.9 by Tigray at 21.5 percent, Addis Ababa at 20.2 percent of young women were unemployed in percent, and Amhara at 19.7 percent. 11 The service sector contributed 44 percent wholesale and trade to the overall GDP growth of the real GDP growth. With a growth and the service sector GDP growth were 9 and 30 contribution of 3.4 percent, the service percent, respectively. The growth rate of transport sector was the leading contributor to the and communication dropped to single digits, at 7.7 percent growth of GDP in FY2018. In the 6.4 percent, for the first time since 2010/11, and service sector, wholesale and trade contributed decreased by more than twice compared with the 22 percent to the overall GDP growth and 50 level in FY2017 (15.1 percent). Accordingly, the percent to the growth in the service sector GDP. The contributions of the transport and communication performance of the service sector and the wholesale subsector to overall GDP dropped by 55 percent, and trade subsector showed a huge improvement from 0.7 to 0.3 percentage point. Similarly, the compared with FY2017, when the service sector’s contribution of the same subsector to service sector growth contribution was 3 percent (30 percent of GDP dropped by 58 percent, from 2 percentage the overall growth), whereas the contributions of points to 0.8 percentage point (figure 1.5). Figure 1.5. Service Sector Performance (percent) 10 9 8 7 6 5 4 3 2 1 0 Real estate, renting & business activities Transport & communications Financial intermediation Other community, social & personal services Hotels & restaurants Services Whole sale & retail trade Public administration & defense Education Private households with employed persons Health & social work -1 2016/17 2017/18 Source National Planning Commission. 12 Inflation and Monetary Policy in Ethiopia Inflation showed a declining trend in the in December 2018. The price deceleration has first half of FY2019, after surging to double been more rapid for food products, whose digits in FY2018. Since reaching 16.8 percent price inflation decreased from 17.8 percent in in June 2018, due to the expansion of public June 2018 to 11.4 percent in December 2018, sector credit in 2017, broad money growth, pass- compared with nonfood items, whose price through of the October 2017 devaluation, and inflation decreased only marginally (from 11.1 to political disruptions (which adversely affected 9.1 percent during the period). There was a slight distribution networks), inflation has been on a uptick in inflation, to 10.8 percent in January declining trend. It reached a low of 10.4 percent 2019 (figure 1.6). Figure 1.6. Inflation (year-on-year, percent) 25 20 15 10 5 0 Jul-13 Oct-13 Jan-14 Apr-14 Jul-14 Oct-14 Jan-15 Apr-15 Jul-15 Oct-15 Jan-16 Apr-16 Jul-16 Oct-16 Jan-17 Apr-17 Jul-17 Oct-17 Jan-18 Apr-18 Jul-18 Oct-18 Jan-19 General Food Non-food Source Central Statistical Agency. Although growth in monetary aggregates was strong in FY2018 despite the NBE’s restrictive was strong in FY2018, there was a significant policy measures implemented in the aftermath of reduction during the first half of FY2019. the October 2017 devaluation. The NBE lowered Growth in reserve money (base money), which its reserve money target for FY2018, from 22 to 16 is the operating target for the National Bank of percent; limited credit expansion to sectors other Ethiopia’s (NBE’s) monetary policy, dropped to than exports and manufacturing to 16 percent; 12.8 percent in December 2018 compared with and raised the minimum interest rate from 5 19.1 percent in June 2018. Accordingly, broad to 7 percent, in an effort to contain the pass- money expansion was significantly curtailed (from through effect of the October 2017 devaluation. 29.2 percent in June 2018 to 22.2 percent in Although the NBE managed to reduce reserve December 2018). Growth in monetary aggregates money growth from about 24 percent in the 13 months before the devaluation to 19.1 percent broad money supply was also strong, reaching in June 2018, it missed the target of 16 percent 29.2 percent in June 2018 compared with 28.8 announced following the devaluation. Growth in percent in June 2017 (figure 1.7). Figure 1.7. Broad Money and Reserve Money (year-on-year, percent) 45 40 35 30 25 20 15 10 5 0 Jul-14 Sep-14 Nov-14 Jan-15 Mar-15 Jul-15 Sep-15 Nov-15 Jan-16 Mar-16 Jul-16 Sep-16 Nov-16 Jan-17 Mar-17 Jul-17 Sep-17 Nov-17 Jan-18 Mar-18 May-18 Jul-18 Sep-18 Nov-18 May-15 May-16 May-17 Broad money supply (M2) Reserve money (M0) Source National Bank of Ethiopia. Expansion of domestic credit was the main previous year’s growth (23.3 percent). Of the driver of growth in broad money supply nongovernmental sectors, credit growth to state- (M2). Domestic credit expanded by 24.3 percent owned enterprises (SOEs) remained broadly stable, in June 2018—slowing from the 28.7 percent as in the previous fiscal year, at around 21 percent, expansion in June 2017—while net foreign assets reflecting continued slowdown in investments showed only a modest increase of 3.5 percent. by SOEs. Growth in credit to the private sector Credit to the government increased by 19.4 increased from 27.7 to 30.1 percent (figure 1.8). percent—significantly less than its growth rate of Looking at the composition of domestic credit, about 80 percent the previous year—reflecting the share of SOEs continued to decline, as was the government’s cautious fiscal policy stance in the case in FY2017, while the share of the private FY2018. Credit to the nongovernmental sectors sector increased slightly. The share of the central grew by 25.1 percent, slightly higher than the government decreased slightly (figure 1.9). The share of the private sector in domestic credit has increased, while the share of SOEs and the central government decreased in FY2018. 14 Figure 1.8. Contribution to Broad Money Growth (year-on-year, percent) 40 35 30 25 20 15 10 5 0 -5 -10 -15 Jul-14 Oct-14 Jan-15 Apr-15 Jul-15 Oct-15 Jan-16 Apr-16 Jul-16 Oct-16 Jan-17 Apr-17 Jul-17 Oct-17 Jan-18 Apr-18 Jul-18 Oct-18 Net foreign assets Domestic credit Other items, net Broad money supply (M2) Source National Bank of Ethiopia. Figure 1.9. Composition of Domestic Credit (percent) 100 90 37.3 36.5 35.3 35.8 36.9 80 39.8 43.0 70 60 50 40 53.8 55.7 55.0 50.7 50.1 45.6 50.8 30 20 10 11.4 9.4 9.0 7.8 9.7 13.5 13.0 0 FY2012 FY2013 FY2014 FY2015 FY2016 FY2017 FY2018 Central government SoEs Private sector Sources National Bank of Ethiopia. Note SOEs = state-owned enterprises. 15 Treasury bills and NBE advances are the few fiscal years, amounting to about 1.3 percent main instruments for the government’s of GDP . This significant level of borrowing from domestic financing of its deficit. In FY2018, NBE has obviously been an important factor in the government borrowing through T-bills covered inflationary pressures experienced over the past about 60 percent of total government borrowing; few years. It has been compounded by nonmarket- the remaining 40 percent was filled through NBE based borrowing through T-bills, which relies advances (figure 1.10). In the previous fiscal year, heavily on a limited number of captive investors NBE advances covered about 63 percent. NBE (predominantly public bodies) that are buying the advances have remained significant over the past bills at significantly below market rates (box 1.3). Figure 1.10. National Bank of Ethiopia Advances to the Government (percent of gross domestic product) 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 FY2010 FY2011 FY2012 FY2013 FY2014 FY2015 FY2016 FY2017 FY2018 Source National Bank of Ethiopia. Box 1.3 Reforming Government Domestic Financing Mechanisms: Toward the Development of a Domestic Government Debt Market Domestic borrowing has become an 0.5 percent of gross domestic product (GDP) important source of financing the deficit about five years ago, has been increasing during the past five years. Deficit financing over the past years, reaching more than 2 through domestic borrowing, which was below percent of GDP last fiscal year (figure B1.3.1). 16 Figure B1.3.1 Fiscal Deficit and Financing (percent of GDP) 5 4 3 2 1 0 -1 -2 -3 -4 FY2010 FY2011 FY2012 FY2013 FY2014 FY2015 FY2016 FY2017 FY2018 FY2019 External borrowing Domestic borrowing Privatization proceeds Others & residuals Fiscal deficit Source Ministry of Finance and Economic Cooperation. The number of instruments the government draw from the NBE in any given fiscal year. uses for domestic borrowing has remained The National Bank of Ethiopia Establishment very limited. There are two main domestic Proclamation, which was amended in 2008, financing instruments for the government: states that the amount of advances and advances from the National Bank of Ethiopia credit that NBE extends to the government (NBE) and Treasury bills. Direct advances from for each fiscal year will be determined the NBE involve the provision of an overdraft through consultations between the NBE facility to the government and are essentially and the government, and that it should be monetary financing of the deficit. Over the consistent with the maintenance of price and past decade, on average nearly 60 percent exchange rate stability. The interest rate the of the government’s domestic borrowing was government pays on such advances from the through such advances from the NBE. NBE has remained unchanged at 3 percent for more than a decade. Moreover, these There are no statutory limits set on the amount advances have been the major driver of of direct advances that the government can reserve money growth (figure B1.3.2). 17 Figure B1.3.2 Contribution to Reserve Money Growth (year-on-year, %) 70 50 30 10 -10 -30 Jul-14 Oct-14 Jan-15 Apr-15 Jul-15 Oct-15 Jan-16 Apr-16 Jul-16 Oct-16 Jan-17 Apr-17 Jul-17 Oct-17 Jan-18 Apr-18 Jul-18 Oct-18 Other item, net Advances to Development Bank of Ethiopia Advances to the government Net foreign assets Reserve money Source National Bank of Ethiopia. Although the government started accounted for over 95 percent of the total selling T-bills in 1995, the T-bill market T-bill outstanding balances. has remained undeveloped. The number of participants in T-bill auctions Investment options for the pension funds has remained limited, and the yield are restricted through their respective on T-bills has been very low. In recent establishment proclamations such that years, T-bill sales have been limited to a T-bills are the only instrument in which handful of captive investors, including the they can invest their funds. The pension pension funds for public and private sector funds are essentially required to buy T-bills employees and very few state-owned using the funds they mobilize. The return on enterprises (including the Development the T-bills is below the minimum interest rate Bank of Ethiopia). Over the past five years, paid for deposits and well below the inflation T-bill holdings of the pension funds and rate, meaning negative returns in real terms the Development Bank of Ethiopia together (figure B1.3.3). 18 Figure B1.3.3 Interest Rates and Inflation (percent) 25 20 15 10 5 0 FY2014 FY2015 FY2016 FY2017 FY2018 Saving deposits T-bills GERD bond Inflation Sources National Bank of Ethiopia. Note Grand Ethiopian Renaissance Dam (GERD) bonds are issued to raise money for the construction of the dam. As part of the developmental state debt accumulated through such financing model, the government ’s strategy has been increasing rapidly over the past for financing its deficit from domestic few years. The stock of domestic debt of the sources over the past years focused on government, which was about 11 percent using sources of funds with relatively of GDP five years ago, rose to 14 percent low cost. By making use of administrative in FY2018. Of the total domestic debt, measures, the government has been able to 51 percent constitutes accumulated direct obtain financing domestically at substantially advances from the NBE, and 37 percent is below market interest rates. The stock of T-bill holdings (figure B1.3.4). Figure B1.3.4 Stock of Government Domestic Debt 350 16 300 14 12 250 Percent of GDP 10 Birr, billions 200 8 150 6 100 4 50 2 0 0 FY2012 FY2013 FY2014 FY2015 FY2016 FY2017 FY2018 NBE advance T-bills Bonds Total (RHS) Sources Ministry of Finance and Economic Cooperation. Note GDP = gross domestic product; NBE = National Bank of Ethiopia. 19 Domestic financing of the government financing raised through market-based is one of the areas that the government means, with the aim of eventually phasing intends to look into in its current out the distortionary financing mechanisms, reform agenda. The government’s current including direct advances from the NBE. financing model contributes to financial repression, which greatly hinders the To have a functioning market for development of a vibrant financial market. government debt securities, there are Moving away from the existing distortionary also other measures that should be financing mechanisms toward a more taken. One such measure is revising the market-based means of financing will make directive that forces private commercial significant contributions to the development banks to divert 27 percent of their fresh loan of the financial sector. However, as the disbursements for the purchase NBE bills at government’s financing needs have grown a low interest rate. Although immediately significantly over the years, an immediate eliminating this requirement might create shift to market-based financing mechanisms liquidity management challenges for the poses the risk of facing very high borrowing NBE, gradually phasing out the requirement is costs for the government. Thus, the shift needs essential, so that private commercial banks will to be gradual. In this regard, starting to issue have sufficient liquidity to be active participants T-bills based on market rates (for example, in the government’s securities market. In through auctions) only for a small portion addition, it might also be worthwhile to look of the financing needs initially could be a into the establishment proclamations of the good starting point. Then, the government social security funds in a way that could can gradually increase the portion of its broaden their investment options. Fiscal Policy in Ethiopia The federal government continued to recent social unrest, all sources of fiscal revenue adhere to a cautious fiscal policy stance in (direct taxes, indirect taxes, domestic taxes, and FY2018. According to official data, the federal foreign taxes) decreased in real terms and as a government’s fiscal deficit stood at 3.0 percent of share of GDP . Total revenue was 13 percent lower GDP in FY2018, which was lower than its level in than budgeted in FY2018. Hence, the tax-to-GDP FY2017 (3.4 percent of GDP) (figure 1.11). The ratio declined from nearly 12 percent in FY2017 deficit declined following a drop in expenditure to 10.7 percent in FY2018. This puts Ethiopia in and revenue collection. the bottom third of Sub-Saharan African countries in terms of tax effort.² Revenues and grants Revenue growth slowed in FY2018. Due to dropped from 14.7 percent of GDP in FY2017 weak tax administration, generous tax incentives to 13.1 percent of GDP in FY2018, mainly due and exemptions, and the disruptive impact of to underperformance in tax revenue collection ² The average tax-to-GDP ratio for Sub-Saharan Africa was 16.2 percent in 2015 (IMF 2018a). 20 (figure 1.12). Total revenue fell to its lowest level to unrest in the first half of FY2018. Currently, in more than a decade, reaching 13.0 percent of various studies, including by the World Bank, focus GDP . Tax revenue fell from 11.7 percent of GDP on improving tax administration efficiency and in FY2017 to 10.7 percent of GDP in FY2018, increasing tax collection. Tax revenue could be mainly due to the decline in indirect taxes, the increased from the current level of 11 percent of largest source of tax revenue in Ethiopia. Domestic GDP through reforming the existing tax system indirect taxes fell from 3.4 percent of GDP in and improving the efficiency of tax administration. FY2017 to 3.0 percent of GDP in FY2018, and As the average tax gap (tax potential) for most foreign trade taxes dropped to 3.2 percent of GDP Sub-Saharan African countries (including Ethiopia) in FY2018. The weak performance in domestic falls between 3 and 5 percent of GDP , addressing indirect taxes could be attributed to inefficiencies inefficiencies appears to be a priority for domestic in tax administration and low compliance due resource mobilization (IMF 2018a). Figure 1.11. Government Spending Moderated 20.0 17.2 18.8 16.2 17.8 15.1 16.2 15.4 17.3 14.5 17.0 15.7 18.2 15.6 17.9 14.7 18.0 13.1 16.1 0.0 18.0 -0.5 16.0 -0.9 14.0 -1.2 -1.2 -1.0 12.0 -1.6 -1.1 -1.5 -1.8 10.0 -2.0 -1.6 -1.6 -2.2 -2.0 8.0 -1.9 -2.5 6.0 -2.5 -2.4 -2.3 -2.9 4.0 -2.5 -3.0 2.0 -3.0 - -3.3 -3.5 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 2017/18 Revenue & grants Total expenditure Fiscal balance incl. grants (RHA) Primary deficit (RHA) Source Ministry of Finance and Economic Cooperation. 21 Figure 1.12. Revenue Growth Remains Weak (percent of GDP) 13.1 13.4 13.9 13.4 14.7 14.8 15 14.0 1.68 12.2 2.63 1.9 1.20 12 1.9 2.2 2.54 1.56 4.16 4.00 4.3 4.2 3.61 9 4.5 4.4 3.20 4.13 6 3.57 3.41 3.6 3.7 3.05 3.0 3.0 3 4.3 4.74 4.58 4.44 4.43 3.7 3.8 4.1 0 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 2017/18 Direct tax Domestic indirect tax Foreign trade tax Non tax revenue Domestic Revenue Source Ministry of Finance and Economic Cooperation. With revenue mobilization underperforming concessional borrowing financed one-third of the in FY2018, expenditure cuts helped fiscal deficit, while domestic borrowing, including maintain the deficit within reasonable direct advances from the NBE equal to 1.1 percent margins, but general government spending of GDP , financed the remaining two-thirds. increased, driven by increases in recurrent expenditures. Total expenditure fell from 18.0 The overall public sector deficit (including percent of GDP in FY2017 to 16.1 percent of GDP SOEs) was further consolidated in FY2018. in FY2018 (figure 1.12). However, despite a decline The overall public sector primary deficit is in capital expenditures by 1.8 percentage points in estimated at 6.7 percent of GDP in FY2018 (which percent of GDP , recurrent expenditures increased is nearly the same as FY2017, 6.6 percent of by 19 percent (year-on-year) in FY2018 (keeping GDP). This shows that the financing needs of the the recurrent expenditure stable as a percentage nonfinancial public sector continued to decline, of GDP), mainly due to growth in interest payments following the government’s effort to refrain from and charges, which increased by about 40 percent. nonconcessional borrowing by SOEs. Yet, a The decline in capital expenditure is attributable complete picture of the public sector is difficult to a slowdown in development expenditure, to establish, as the government does not publish which fell by 9 percent in FY2018 compared with consolidated fiscal accounts that include all SOEs. FY2017. The low revenue performance impacted government expenditure. Year-on-year capital On January 18, 2018, the government expenditure growth declined by 6 percent. Thus, approved a supplementary budget of Birr the FY2018 fiscal deficit remained within the target 14 billion (US$514.7 million), equivalent to anticipated in the annual budget. External, largely 0.6 percent of GDP. The supplementary budget 22 was introduced to finance (i) rehabilitation of universities, (iii) transition of the state broadcaster people affected by conflicts and natural disasters, to digitalization, and (iv) the Youth Revolving Fund (ii) student enrollment in the 11 newly opened (box 1.4). Box 1.4 Why the Supplementary Budget? Program budgeting and more broadly Analytically, the program can be a means to performance budgeting has a long history group inputs around objectives and evaluate across the globe. Many governments have effectiveness and efficiency in combination implemented similar reforms, with varied with output and performance data. experiences. The program concept is flexible in scope, but it can serve multiple In view of the importance of program objectives. The program can be a means budgeting, the Ministry of Finance and to enable more efficient use of resources Economic Cooperation prepared the Program within the program, by aggregating activity Budget Manual (2010). The manual states that budgets and providing more flexibility to “although planning and budget process should program managers to apply these resources be thorough and attempt to anticipate the needs during the year. And it can be a vehicle for of the next year, not all future circumstances reorganizing administrative units into more can be foreseen with accuracy.” As a result, it coherent structures focused on objectives. allows two types of budget adjustments: 1 Budget transfer: moving budget funds between public bodies, budget institutions, projects, or items of expenditure, without changing the total approved budget. 2 Budget supplement: the total approved budget can be increased with the approval of the House of People’s Representatives on the recommendation of the Council of Ministers. The manual states the conditions for the approval of the supplementary budget as follows. During a budget year, while an approved budget is in the process of implementation, it is possible that: a. An unforeseen or urgent need for increased expenditure arises (for example, a natural disaster) b. A new project, not included in the original approved budget, is approved for commencement during the budget year c. Additional resources become available (for example, from external assistance or loans) that can fund increased total expenditure, including any new projects. 23 Further, Financial Administration Proclamation government has issued supplementary No. 648/2009 Article 2(12) states that budgets for the past eight years in a row. In “Supplementary Budget” means a budget the first year after the transition from line- approved in situations where the revenue item budgeting to program budgeting, the budget appropriated for activities of the government approved an additional budget government to be carried out in a fiscal of Br 7.5 billion (10 percent of the total federal year is not sufficient, or where a budget is budget for that year). For instance, since required for an activity of the government to 2014/15, a supplementary budget equivalent which budget is not appropriated or where to on average 5.5 percent of the total approved the expenditure budget appropriated for an budget has been endorsed each year. The activity is not sufficient. supplementary budget for 2017/18, Br 14 billion, amounts to 4.3 percent of the annual With this provision in the Program Budget budget approved in June 2017, which is lower Manual (MOFED 2010) and Financial compared with the preceding two fiscal years, Administration Proclamation (2009), alongside 6.2 percent in 2014/15 and 7.6 percent in the program budgeting it has adopted, the 2015/16 (figure B1.4.1). Figure B1.4.1 Supplementary Budget and Its Share of the Total Annual Budget 20 12 18 16 10 14 8 12 10 6 8 6 4 4 2 2 0 0 2007/08 2010/11 2011/12 2014/15 2015/16 2016/17 2017/18 Suppl. Budget, (birr, billions) % of Federal Budget (RHS) Source Negarit Gazzetta Budget Proclamations, various years. Total public spending increased in FY2018. public sector in FY2018. According to the NBE In the absence of consistent and comprehensive monetary survey, the two state-owned banks, fiscal reports encompassing all extra-budgetary Commercial Bank of Ethiopia and Development accounts and SOE operations,³ the overall fiscal Bank of Ethiopia, disbursed 61 percent of all new policy impulse of the public sector cannot be loans from the banking system in FY2018. The precisely estimated. However, available data government continued to require that private indicate a major stimulus effort by the broader banks purchase NBE bills equivalent to 27 percent 24 of their loan portfolios to finance the public sector. percent of GDP in June 2018 to 54 percent. As a result, SOEs received 43.7 percent of all new External debt4 decreased from 29.2 percent of loans disbursed by the banking system in FY2018, GDP to 27.8 percent, while domestic public debt up from 39.7 percent in FY2017. Support to SOEs decreased from 27.8 to 26.1 percent of GDP represented 83 percent of the total change in loan between June and September 2018. In FY2018, disbursement, up from just 12 percent in FY2017. public debt as a percentage of GDP increased to SOEs in the manufacturing, export, transport, 57 percent compared with 54.9 percent in FY2017 communications, mines, power, and water sectors (figure 1.13). The increase was mainly driven by were the main beneficiaries of new loans. external debt, which increased from 27.5 to 29.2 percent of GDP , while domestic debt marginally Although the federal government has increased, from 27.4 to 27.8 percent of GDP . maintained financial discipline, the broader Borrowing from multilateral institutions (largely public sector shows serious financial on concessional terms) accounted for over half vulnerabilities. The government’s mix of fiscal, of the increase in external debt, while about one- monetary, exchange rate, and structural policies third of the increase came from borrowing from has generated persistent inflation, overvaluation of commercial banks. The composition of public debt the birr, large external current account imbalances, remained largely unchanged, with a 55:45 ratio foreign exchange shortages, and an elevated risk between external and domestic debt. of debt distress. Although the rapid expansion of public infrastructure investment has boosted Despite the increase in the level of external firm productivity, this approach is reaching its debt, debt service expenditures moderated limit in terms of external debt sustainability and in FY2018, although they remained high. crowding out the private sector in the credit and External debt service payments as a ratio of foreign exchange markets. Notwithstanding exports declined to 18.7 percent, from 21 recent improvements in public debt management, percent in FY2017, as a result of strong export including tighter control over SOE borrowing, performance (which was entirely attributable to persistent external imbalances have contributed service exports). Notwithstanding the strong service to an elevated risk of external debt distress, which export performance in FY2018, Ethiopia’s external is not expected to self-correct in the near term. debt vulnerabilities mainly arise from lower-than- All three major credit rating agencies assigned expected outturns from the export sector. In the Ethiopia a single B-rating, as the country’s 2018 Debt Sustainability Analysis (DSA), the two obligations are considered speculative and subject indicators related to exports—the present value to high credit risk. of public and publicly guaranteed external debt- to-exports and external debt service–to-exports— Ethiopia`s overall public debt showed an remained above their respective thresholds in the increase in FY2018 but moderated in the baseline. As a result, the 2018 DSA maintained first quarter of FY2019. Public debt moderated that Ethiopia remains at “high risk” of external in the first quarter of FY2019, declining from 57 debt distress, as was the case in the 2017 DSA. 3 Adopting international best practices for government finance statistics remains an urgent priority. 4 External debt includes Saudi and United Arab Emirates deposits at the NBE but excludes Ethiopian Airlines debt. 25 Figure 1.13. Public Debt, FY2018 70 25 60 20 Debt services-to-eports ratio 50 Percent of GDP 40 15 30 10 20 5 10 0 0 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 2017/18 Public debt External debt Ext. Debt service to export ratio (RHA) Source Ministry of Finance and Economic Cooperation Debt Bulletin. The government has taken several steps nonconcessional borrowing decreased from to address external debt sustainability US$5.8 billion (9 percent of GDP) in FY2015 challenges, including exercising stronger and US$2.2 billion (3 percent of GDP) in controls on nonconcessional borrowing by FY2014 to US$0.6 billion (0.7 percent of GDP) SOEs. Ethiopia has put in place a legal debt in FY2018. In addition, the implementation management framework that is comprehensive of the FY2017 Public Debt Management and and ranks among the best performing countries Guarantee Issuance Directive establishes strict in the World Bank ’s Debt Management oversight by MOFEC of new nonconcessional Performance Assessment. The FY2017 Public borrowing by SOEs. MOFEC is also tasked Debt Management and Guarantee Issuance with monitoring compliance with the FY2018 Directive was effectively implemented in Budget Law and controlling import-intensive FY2018. The directive establishes stricter investment projects financed on nonconcessional controls by the Ministry of Finance and Economic terms. The government is taking proactive Cooperation (MOFEC) on the contracting of steps to renegotiate the terms on the stock of new nonconcessional borrowing by SOEs. It nonconcessional debt. considerably strengthened MOFEC’s monitoring of the FY2018 Budget Law and controls over Ethiopia’s External Sector import-intensive investment projects financed on nonconcessional borrowing terms, including Following an increase of 13 percent in by SOEs. FY2018, total exports continued to expand by 18 percent in the first half of FY2019, Tightening of nonconcessional borrowing. driven by services. Led by the performance of The government has tightened control over Ethiopian Airlines, service exports increased by 41 nonconcessional borrowing. The level of percent during the first half of FY2019 compared 26 with the same period in FY2018. In contrast, Total exports increased in FY2018, thanks merchandise exports decreased by 10 percent, to the good performance of service exports reflecting the continued poor performance of (figure 1.14). Despite the underperformance Ethiopia’s major export items, including coffee, of goods exports, mainly due to structural and oilseeds, and pulses. Manufacturing export competitiveness issues, such as rigid labor and performance was relatively strong, with exports product markets including an overvalued exchange of textiles and electronics posting growth rates rate, Ethiopia’s total exports expanded by 13.1 of more than 35 percent. Despite these positive percent in FY2018, after years of underperformance. developments, total exports do not exceed 10 This sharp recovery in exports was driven by growth percent of GDP , significantly below the 24 percent in service exports, which accounted for half of total expected from a country the size of Ethiopia at exports and expanded by 26.6 percent, mainly due this level of development. to the strong performance of Ethiopian Airlines. Figure 1.14. Exports of Goods and Services (percent of GDP) 18 16 14 8.6 12 10 6.7 7.3 6.5 5.9 8 4.5 4.6 6 4.0 3.6 3.4 4 8.1 6.5 6.0 5.7 6.0 6.8 4.8 2 4.1 4.2 5.0 0 2008/09 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 2017/18 Service exports Good exports Source National Bank of Ethiopia. Goods exports declined due to the weak Ethiopia exported electronics products for the first performance of coffee. Coffee represented time in FY2018. A closer look at firm-level data 30 percent of goods exports and decreased by reveals that in 2016 exporting firms represented 5 percent in FY2018, mainly because of a drop less than 4.5 percent of total manufacturing firms of more than 10 percent in coffee prices. Export (figure 1.15). Exports accounted for about 2.7 increases were recorded in the leather, textile and percent of total manufacturing production (figure garments, chemical, and electricity sectors, reflecting 1.16). In 2017, industrial parks accounted for 2.4 in part the coming on stream of the industrial parks. percent of total exports. 27 Figure 1.15. Number of Firms Figure 1.16. Average Production and Exports 3,000 60 2,500 50 2,000 40 Millions 1,500 30 1,000 20 500 10 0 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Non-exporting Exporting Average production value Average export value Source Ethiopian Revenue and Customs Authority. The compression of goods imports played an debt service obligations, despite the overall relatively important role in reducing the current account low external debt–to-GDP ratio of 32 percent. Over deficit (figures 1.17 and 1.18). Low levels of the medium term, expansion of exports is critical for exports resulted in liquidity constraints related to Ethiopia’s external debt sustainability. Figure 1.17. Imports of Goods (percent of gross domestic product) 35 30 5.1 5.7 0.4 6.1 4.4 4.8 0.4 4.8 4.7 25 4.8 0.7 0.4 0.4 4.4 0.8 4.3 0.3 5.0 4.3 4.2 4.0 4.2 4.7 20 4.4 3.2 0.4 5.2 4.9 4.6 1.9 0.3 4.5 3.4 15 4.5 2.3 3.1 7.0 6.7 7.3 2.7 7.2 8.2 7.2 10 6.1 5.6 5 9.6 10.7 8.6 6.8 7.5 8.7 9.4 7.5 6.2 0 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 2017/18 Capital goods Consumer goods Fuel Intermediate goods Others Service imports Total imports Source National Bank of Ethiopia. 28 Figure 1.18. Balance of Payments (percent of GDP) 25 2.3 1.3 20 3.4 4.0 3.2 3.9 8.2 4.4 15 2.4 4.2 4.8 3.3 2.8 6.4 5.8 2.5 2.6 4.2 2.6 3.4 5.2 4.4 10 4.1 3.2 2.3 1.4 2.6 1.8 1.3 5 9.0 8.6 7.6 8.3 7.5 7.5 7.3 6.9 7.2 0 -1.0 -0.7 -0.2 -0.1 -1.2 -0.7 0.0 -5 -4.0 -4.3 -10 -6.5 -5.3 -6.5 -7.9 -8.2 -11.5 -10.4 -15 -14.9 -16.2 -14.4 -20 -16.4 -17.6 -19.2 -17.9 -21.0 -19.8 -25 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 2017/18 Private transfers Official transfers FDI Loans Change in reserves CAB/GDP Trade balance G&S Sources National Bank of Ethiopia. Note CAB = current account balance; FDI = foreign direct investment; G&S = goods and services; GDP = gross domestic product. Challenges to exports range from supply allowed the birr to depreciate by about 2.9 percent issues and quality of products, to marketing against the U.S. dollar during the first six months channels, branding, and so forth. The of FY2019. The premium between the official and business environment seems to favor incumbent parallel market exchange rates widened in FY2018, firms and deter new entrants to export businesses. reaching a peak of about 30 percent toward the The export sector lacks dynamism in firm entry and end of the fiscal year. However, there was a sharp exit. Other factors, such as low entrepreneurship narrowing of the premium in June/July (reaching and low regulatory quality for promoting the private about 5 percent), following the government’s sector, may also explain Ethiopia’s poor export call for the public to exchange foreign currency performance. holdings in commercial banks. The premium has again started to widen in recent months, returning After keeping the exchange rate largely to its previous level of about 30 percent. constant in FY2018, following the October 2017 devaluation, the NBE resumed its The real effective exchange rate (REER) policy of gradual depreciation of the birr depreciated by 7.1 percent in FY2018, against the U.S. dollar in FY2019 (figure reversing the appreciating trends of the 1.19). Following the 15 percent one-off devaluation past six years, although the appreciating in October 2017, the nominal exchange rate was trend has resumed in recent months (figure kept largely constant for the remaining period of 1.20). The depreciation of the REER was largely FY2018. The official exchange rate against the owing to the 15 percent one-off devaluation of the U.S. dollar depreciated by less than 1 percent nominal exchange rate of the birr against the U.S. between October 2017 and June 2018. The NBE dollar in October 2017. However, the trend has 29 reversed in recent months, with the REER showing According to the International Monetary Fund’s a quarter-on-quarter appreciation of 4.8 percent estimates, the REER was overvalued by 12-18 in the first quarter of FY2019, as the inflation percent in September 2018 (IMF’s Article IV differential with trading partners remains high. Consultation Report, 2018). Figure 1.19. Nominal Exchange Rate of the Birr against the U.S. Dollar 39 36 33 30 27 24 21 18 15 Jul-13 Oct-13 Jan-14 Apr-14 Jul-14 Oct-14 Jan-15 Apr-15 Jul-15 Oct-15 Jan-16 Apr-16 Jul-16 Oct-16 Jan-17 Apr-17 Jul-17 Oct-17 Jan-18 Apr-18 Jul-18 Oct-18 Official Parallel market Source National Bank of Ethiopia. Figure 1.20. Real Effective Exchange Rate Index (2000 = 100) 180 160 140 120 100 80 60 40 20 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Source National Bank of Ethiopia. Currently, the focus seems to be on large sector in manufacturing. Although FDI inflows foreign direct investment (FDI) and state- to the manufacturing sector are growing, partly owned firms, without commensurate due to the establishment of industrial parks, the attention to the role of the domestic private total FDI stock remains small, but it has grown 30 overtime. FDI inflows to the manufacturing bottlenecks in transport and telecommunications sector constituted more than 80 percent of total infrastructure, human capital shortages, erratic FDI inflows over the past 10 years (figure 1.21). policy changes, and political favoritism. However, it is yet to be seen whether multinational companies will have a positive spillover effect on The Ethiopian authorities are aware of the productivity of local firms, as the multinational these features of their economy, and efforts companies tend to be confined to industrial zones at diversifying and increasing exports are where only a few domestic firms are located. ongoing. For example, to support exports and Interviews with Ethiopian exporters brought up encourage the private sector, on October 10, two recurrent themes: constraints on accessing 2017, the NBE devalued the birr by 15 percent foreign exchange, which impede the imports of and relaxed foreign exchange controls.5 The new intermediate inputs required for export-oriented leadership seems to be committed to pursuing production (especially for non-endowment reforms that improve the country’s export firms situated outside industrial parks) and high performance. For such efforts to be effective, turnover of the labor force (especially for firms including for poverty reduction (Box 1.5), it is established in industrial parks). The interviews also important to understand the main impediments raised several well-known challenges linked to to export growth and diversification. Figure 1.21. Share of Manufacturing FDI in Total FDI, 2007/08–2016/17 (U.S. dollars, millions) 4,500 4,000 3,500 3,000 2,500 2,000 1,500 1,000 500 0 2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 Total FDI Manufacturing FDI Sources Ethiopian Investment Commission. Note FDI = foreign direct investment. 5 Foreign exchange controls were relaxed with the issuance in October of two NBE directives on (i) external loans and suppliers’ credit, and (ii) retention and utilization of export earnings and inward remittances. Under the first directive, any domestic investor can now access an external loan, if the investor generates foreign currency. The debt-equity ratio to access a loan from foreign sources has been revised to 60-40, replacing the hitherto 50-50 ratio. Under the second directive, exporters of goods and services are authorized to retain 30 percent of the proceeds of their exports indefinitely in their forex account (replacing the previous 10 percent threshold). 31 Box 1. Trade and Poverty Reduction in Ethiopia The expansion of international trade has mitigate the risks faced by the poor. For been essential for development and poverty instance, agricultural commodity exchanges reduction. Trade as a proportion of global can improve farmers’ understanding of gross domestic product has approximately market conditions and enable them to doubled since 1975. But trade is not an receive better prices for their crops. All coffee end in itself. People measure the value and sesame products from Ethiopia that are of trade by the extent to which it delivers not directly exported by farmers are traded better livelihoods, through higher incomes, at the Ethiopian Commodity Exchange (ECX) greater choice, and a more sustainable by law. The ECX reaches 2.4 million small future, among other benefits. The number of farmers through farmer cooperative unions, people living in extreme poverty around the which are members of the ECX. By effectively world has fallen by around one billion since using mobile technology, agricultural 1990. Without the growing participation of commodity exchanges can become a developing countries in international trade, platform where farmers can receive up- and sustained efforts to lower barriers to to-date market information and help them the integration of markets, it is difficult to optimize decisions related to the sale of their see how this reduction could have been produce. Kenya’s Agricultural Commodities achieved. Exchange, which deals with nontraditional export commodities, such as maize and By lowering trade costs for deeper integration beans, conveys daily information on the of markets and improving the enabling prices of some 20 commodities collected environment, trade can play a key role in from market vendors around the country poverty reduction in Ethiopia. For example, to farmers and market intermediaries, such a recent study suggests that the gains from as traders, through SMS text messages trade in remote regions of Ethiopia tend to be and daily radio bulletins. Similarly, the captured by intermediaries, thus highlighting ECX provides market data in various ways, the importance of reducing transaction including to rural ticker boards. Expanding costs and intensifying the poverty impact of and improving the use of this system can integration policies to ensure that the rural lead to significantly increased income poor benefit from trade. through better access to information and services in Ethiopia: 75 percent of farmers By effectively using technology, trade can reported higher income as a result of Kenya’s help link producers to upstream buyers and Agricultural Commodities Exchange. Source World Bank and WTO (2017). 32 Gross foreign exchange reserves remained while net international reserves fell below US$1 low, at US$2.85 billion at end the of FY2018 billion. Reserves are below adequate levels and, taken (figure 1.22). Official gross international reserves together with the exchange rate and debt sustainability decreased to US$2.9 billion at end-June 2018 (below assessments (DSAs), the Ethiopia’s external position is two months of projected goods and services imports), weaker than warranted by fundamentals. Figure 1.22. Gross Official Foreign Exchange Reserves (US$, millions) 4,500 3.5 4,000 3,500 3.0 3,000 2.5 2,500 2,000 2.0 1,500 1,000 1.5 500 0 1.0 Mar-10 Aug-10 Jan-11 Jun-11 Nov-11 Apr-12 Sep-12 Feb-13 Jul-13 Dec-13 May-14 Oct-14 Mar-15 Aug-15 Jan-16 Jun-16 Nov-16 Apr-17 Sep-17 Feb-18 Jul-18 Dec-18 Foreign Reserve (mill $) Reserve in months of import (right axis) Source National Bank of Ethiopia. 1.3. Ethiopia’s Economic Prospects in FY2019 and Beyond The economic prospects for FY2019 and imbalances, while moderate fiscal deficits and the medium term are expected to remain prudent monetary policy are expected to reduce stable. The current reform agenda envisaging a the rate of inflation and keep it in the single greater role for the private sector aims to support digits. Merchandise exports could recover in the growth. Annual real GDP growth is projected to medium term, as large investment projects, such be around 7.9 percent in FY2019 and 8.2 percent as the railway to the Port of Djibouti and large in the medium term (table 1.1). The reform power dams (with potential for electricity exports), agenda proposed by the new prime minister become operational. is expected to address some macroeconomic 33 On the supply side, growth in all sectors is The federal government approved the projected to be robust. Agriculture is projected FY2019 budget with a deficit targeted to continue growing at about 4 percent, as national at 3.3 percent of GDP . Although the deficit main season crop production for 2018/19 is targeted for FY2019 is higher than the fiscal expected to be average, following forecasts of deficit for FY2018, the FY2019 budget assumes average rainfall during the main (meher) and a significant improvement in revenue collection belg seasons. Localized areas in the northeastern while maintaining expenditures, especially and eastern parts of the country may face below- capital expenditures, at the FY2018 level. Of average seasonal rainfalls (FEWS NET 2018). the total budget approved for FY2019 (Br 346.9 Industrial growth is expected to decline slightly in billion), 26.4 percent is allocated to recurrent FY2019, to about 11 percent. The service sector expenditure, 32.8 percent to capital expenditure, is projected to continue expanding, due to the and 39 percent to regional transfers. Sixty-six spillover effects from growth in other sectors. percent of the budget is allocated to pro-poor expenditures focusing on education, health, In line with the government’s reform, the industrial development, and the Urban Productive private sector is expected to play an active Safety Net. role. Private investment is projected to increase by 10.6 percent, while government investment (public The government is expected to continue its investment) is expected to decelerate. Household and cautious fiscal policy stance. Government government consumption are projected to decelerate revenue is projected to increase gradually, temporarily. Although the general government fiscal following tax reforms and improvements in policy stance is expected to remain relatively cautious tax compliance, while government expenditure, (the FY2019 budget does not allow for any new mainly capital expenditure, is expected to investment projects), domestic revenue may be stabilize. As a result, the fiscal deficit would insufficient to finance the infrastructure investments remain at the 3 percent observed in FY2019 foreseen in the national development plan, which and further improve to reach 2.9 percent in will require public-private partnerships. FY2020. 34 Table 1.1: Ethiopia: Macro poverty outlook indicators (annual percent changes unless indicated otherwise) 2015 2016 2017 2018 2019f 2020f 2021f Real GDP growth, at constant market prices 10.4 7.6 9.5 8.0 7.9 8.2 8.3 Private consumption 9.9 1.8 5.0 4.9 4.0 4.8 5.0 Government consumption 3.8 13.3 20.4 26.4 9.7 7.0 6.9 Gross fixed capital investment 26.2 10.6 12.4 0.8 10.2 10.5 10.5 Exports, goods and services -11.2 -8.1 7.4 8.3 7.5 8.0 8.3 Imports, goods and services 21.9 0.0 -5.1 1.5 1.8 2.0 2.0 Real GDP growth, at constant factor prices 10.9 9.2 10.2 7.7 7.9 8.2 8.3 Agriculture 6.4 2.6 6.7 3.5 4.0 4.5 5.0 Industry 19.8 22.8 18.7 12.2 11.0 11.0 11.5 Services 11.1 8.6 8.3 8.7 9.3 9.4 8.7 Inflation (Consumer Price Index) 10.1 7.3 7.1 13.1 9.0 8.0 8.0 Current account balance (% of GDP) -11.9 -10.2 -8.2 -6.5 -6.2 -5.9 -5.3 Fiscal balance (% of GDP) -2.4 -2.3 -3.2 -3.0 -3.0 -2.9 -3.0 Debt (% of GDP) 54.0 55.0 57.7 58.7 57.9 62.9 62.0 Primary balance (% of GDP) -2.0 -1.9 -2.7 -2.5 -2.4 -2.3 -2.3 International poverty rate 27.3 26.5 25.2 24.2 23.2 22.2 21.3 a, b ($1.9 in 2011 PPP) Lower middle-income poverty rate 62.2 61.3 60.0 58.8 57.7 56.7 55.6 a,b ($3.2 in 2011 PPP) Upper middle-income poverty rate 85.0 83.8 82.0 80.4 78.9 77.4 76.0 a,b ($5.5 in 2011 PPP) Sources World Bank staff estimates. Note GDP = gross domestic product; PPP = purchasing power parity. Poverty is expected to continue its steady main meher harvest, benefiting rural producers decline on the back of sustained strong and easing urban food inflation. Based on past growth. Good rains in 2018 in most parts of elasticities, poverty measured by the international the country boosted the small belg harvest and poverty line is expected to decrease to 22 percent are expected to result in a better-than-average by 2020. 35 Risks and Challenges devaluation) will continue to have adverse effects on competitiveness, affecting short-to-medium-term The medium-term sustainability of the growth. The rising risk to external debt sustainability growth model poses an important risk. may impact Ethiopia’s access to external finance. The key economic challenge relates to limited competitiveness, which may constrain the Political disruption associated with social development of manufacturing, creation of jobs, unrest could derail the new reform agenda and increase of exports. The pursuit of sound, and negatively impact growth through lower FDI, private sector–led and export-oriented economic tourism, and exports. The outlook for poverty policies is therefore critical for achieving Ethiopia’s reduction, while positive, also faces significant economic and social ambitions. downside risks, given continued localized turmoil and large-scale displacement. Ensuring that Rising external imbalances constitute major growth translates into increased opportunities for challenges to the economy. The lack of external social and economic advancement for Ethiopia’s openness, rising international oil prices, and an young and rapidly growing population will be a overvalued exchange rate (even after the 2017 defining challenge in the coming years. Box 2.4 How Does Poverty (Reduction) in Ethiopia Compare with Other Countries? Over the past 10 to 15 years, poverty US$1.9 poverty line amounted to about 27 in Ethiopia has decreased substantially percent in 2015/16, substantially lower than on the back of strong growth. How does would be estimated based on its per capita the current state of poverty in Ethiopia GDP . Countries with similar per capita GDP compare with that of other countries, in a levels tend to have higher poverty rates (figure static and dynamic sense? That is, conditional B2.4.1). However, the extent to which growth on Ethiopia’s level of development, is the in GDP per capita in Ethiopia translates into country’s poverty rate high or low? And poverty reduction has been low: the “poverty- has there been “enough” poverty reduction, elasticity of growth,” a measure of the extent given the exceptionally strong gross to which GDP growth decreases poverty, domestic product (GDP) growth? To answer amounted to -0.22 between 2005 and 2015. these questions, we use the World Bank’s That means that a 1 percent increase in per international poverty line of US$1.9 per capita GDP was accompanied with a 0.22 capita per day (in purchasing power parity percent decrease in the poverty rate. Among terms) to compare Ethiopia with a selection a sample of comparator countries that also of other countries. had at least two surveys during the same time period (2005–15), only Zambia had a lower Although Ethiopia has a relatively low responsiveness of poverty to growth (figure poverty rate for its income level, it also has B2.4.2). The semi-elasticity, which measures weak transmission between aggregate the percentage point change in poverty for a 1 growth and poverty reduction. The poverty percent change in per capita GDP , is similarly rate for Ethiopia based on the international low: between 2005 and 2015, a 1 percent 36 increase in per capita GDP was accompanied poverty.a Uganda’s semi-elasticity, for by a 0.08 percentage point reduction in instance, was four times higher. Figure B2.4.1 Ethiopia Has a Low Poverty Rate for Its Income Level (linear association between GDP per capita and international poverty rates) 80 Congo. Dem. Rep Madagascar Malawi Burundi Guinea Bissau Moambique 60 Lesotho Zambia Rwanda Poverty (1.9 USD) Sierra Leone Togo Benin Niger Burina Faso Uganda Sengal 40 Liberia Chad Kenya Congo, Rep. Guinea Sao Tome ETHIOPIA Solomon IslandsCote d'Ivoire Haiti Lao PDR 20 Zimbabwe Comoros Cameroon India Micronesia South Africa Nepal Yemen Rep. Bangladesh Namibia Gambia, The Vanuatu Ghana Tajikstan Myanmar Guatemala Pakistan Indonesia Tuvalu Nicaragua Tunisia 0 West 6 7 8 9 10 Log of GDP per capita (2011 PPP) Sources World Development Indicators 2018; World Bank staff calculations. Note The countries included are low- or middle-income countries with at least one poverty survey since 2010. Figure B2.4.2 But Ethiopia Has Weak Transmission between Growth and Poverty Reduction (growth-elasticity of poverty, 2005–15) -2.00 -1.50 -1.00 -0.50 0.00 Vietnam Ghana Uganda Bangladesh Kenya Rwanda -0.22 Ethiopia -0.08 Elasticity Semi-elasticity Zambia Sources World Development Indicators 2018; World Bank staff calculations. Note The countries included are low- or middle-income countries with at least two poverty surveys between 2005 and 2015. 37 The reasons behind Ethiopia’s low were true, continued economic growth conversion rate between growth and could lead to much more poverty reduction poverty reduction are not entirely clear. in the future. Research shows that countries with low levels of initial development (high initial a. The growth-elasticity of poverty is poverty rates) tend to have lower growth- notoriously sensitive to the baseline poverty elasticities, as do countries with high level of development. If initial levels of levels of inequality (Bourguignon, 2003; consumption are low, growth rates of Ravallion, 2012). While Ethiopia definitely consumption will be relatively high for the had low levels of initial development (by far same absolute change, which will lead the lowest of the countries included in figure to an underestimation of the growth- 2.4.1), it also had among the lowest levels elasticity of poverty. As such, growth- of inequality. It is possible that the baseline elasticities tend to be higher in richer level of development in Ethiopia was so low countries. The semi-elasticity partly avoids that growth has increased incomes of the this pattern and does not automatically poor but not yet to the level of pulling them increase when a country grows richer above the poverty line. If that hypothesis (Klasen and Misselhorn 2008). 38 Part 2: Special Topic: Poverty & Household Welfare in Ethiopia, 2011–16 39 2.1 Introduction Poverty in Ethiopia deceased substantially expenditures as a proxy for household income.7 over the past decade, although by less than We draw on the two most recent Household what would have been expected based the Consumption Expenditure Surveys (HCES), country’s on strong economic growth. Despite the 2010/11 round and the 2015/16 round. adverse climatic circumstances linked to the 2015/16 Although the special topic focuses mainly on El Niño drought, poverty reduction has continued in descriptive trends, the World Bank is currently recent years. According to the government’s Interim preparing a deep-dive analysis of poverty and Report on 2015/16 Poverty Analysis Study, the share its drivers in Ethiopia for the forthcoming 2019 of the population living below the national poverty Poverty Assessment. line decreased from 30 percent in 2011 to 24 percent in 2016 (FDRE 2017).6 The special topic proceeds as follows. The next section sketches the trends in household This special topic complements the consumption expenditures between 2011 and government’s Interim Report on 2015/16 2016, disaggregated by urban/rural location, Poverty Analysis Study, by taking a closer region, and agro-ecological zone. Section look at some of the household-level welfare 2.3 discusses the distribution of growth and its trends between 2011 and 2016. Rather than implications for the change in inequality. Section focusing on binary poverty status (whether a 2.4 expands on the situation in terms of monetary household is above or below the poverty line), living standards for the most recent year for which the analysis focuses on household consumption data are available (2016). Section 2.5 concludes. 2.2 Trends in Household Welfare, 2011–16 This section discusses changes in household has a consumption level above the median. The consumption expenditures between 2011 choice of the median over the mean is because, and 2016. The analysis mainly focuses on median unlike the mean, the median is not sensitive to consumption levels rather than means (although extremely high values, which makes the median means are presented in annex A). The median preferable when the focus is on the living standards represents the level of consumption below or above of lower-income groups. The median is also more which lies the consumption of half the population. In robust to measurement errors in consumption. other words, half the population has a consumption Box 2.1 explains the construction of the household level lower than the median, while the other half consumption aggregate. 6 The national poverty line was determined at Br 7,184 per adult equivalent per year (in December 2015 prices). 7 In countries where regular wage employment is rare and most people make their living from the land or as self-employed workers, consumption is considered a better welfare indicator than income. This is largely because of the difficulty in measuring incomes when income flows are erratic and unpredictable, or when large shares of households do not earn cash income (for instance, pure subsistence farmers) (World Bank 2005). 40 Box 2.1 Consumption Aggregation and Adjustments The analysis presented in this 8 to July 7), by randomly allocating sample Economic Update is based on detailed households to different months. consumption data included in the Household Consumption Expenditure To adjust for price variations across Surveys (HCES) (2010/11 and 2015/16). time and space, spatial and temporal All consumption of food and nonfood items price deflators are used. First, nominal is included, regardless of whether these consumption is adjusted for price differences items are purchased on the market, come across reporting levels, by using the spatial from own production, or were received as deflators provided by FDRE (2012, 2017). gifts. For own-consumption and gifts, the Second, spatially-deflated consumption levels quantities consumed are valued at prevailing are expressed in December prices (December prices in the enumeration area. Although 2010 and December 2015), by using the consumption is expressed on an annual food and nonfood Consumer Price Indexes basis, the reference period used during data provided by the Central Statistics Agency. The collection varies based on the nature of the food and nonfood Consumer Price Indexes consumption item. For example, information are also used to bring the December 2010 on food and food-related items was asked consumption expenditure (2010/11 HCES) to twice a week using the “last three days” December 2015 prices. Finally, to adjust for and “last four days” as reference periods variations in household size and composition, (households are visited twice during the consumption expenditure is divided by the HCES). For house rent, durable goods, officially-used adult-equivalent scales, which clothing, health and education expenditures, are based on calorie requirements and vary and so forth, the “last three months” and by age and sex. This exercise should result “last 12 months” were used as references. in a consumption aggregate that can be Imputed rent is included in the consumption compared through space and time. The aggregate.a To capture the effect of seasonal consumption aggregates used in this special variations, the data were collected over a topic section are the official ones used by the 12-month span (Hamle 1 to Sene 30/July National Planning Commission. Sources DRE (2012, 2017); Central Statistical Agency 2018. a. Imputed rent was already included in the HCES data set obtained from the Central Statistical Agency and as such was not calculated by the World Bank. 41 Strong Consumption Growth in the median consumption per adult amounted to Urban Areas, Relatively Weak US$1,301 per year in 2015/16.9 The increase in in Rural Areas consumption was substantially stronger in urban areas, where the median annual consumption per Overall, solid household consumption adult increased by 32 percent, from Br 10,750 to growth was observed between 2011 and Br 14,230.10 In rural areas, median consumption 2016, particularly in urban areas. At the increased by about 7 percent (from Br 9,331 to Br national level, the median annual consumption 9,944). Real consumption levels in 2015/16 were per adult increased from Br 9,520 to Br 10,657, a approximately 43 percent higher in urban than 12 percent increase.8 Expressed in U.S. dollars and in rural areas (figure 2.1). The results are similar corrected for differences in the purchasing power if mean consumption is used instead of median of a dollar between Ethiopia and the United States, consumption (annex A). Figure 2.1. Consumption Increased between 2011 and 2016, Particularly in Urban Areas (median annual consumption per adult, December 2015 prices) 16,000 14,230 Median annual consumption per adult 12,000 10,657 10,750 (December 2015 birr) 9,520 9,944 9,331 8,000 4,000 0 National Urban Rural 2011 2016 Source Household Consumption Expenditure Surveys 2011, 2016; World Bank staff calculations. Consumption increased in all regions Afar and from Br 9,395 to Br 9,219 (a decrease except Afar and Amhara. Annual median of about 2 percent) in Amhara.11 Consumption consumption per adult equivalent decreased increased in all the other regions, although the from Br 9,031 to Br 8,503 (a fall of 6 percent) in magnitude of the increase varied: the highest 8 All consumption values in this and the next section are expressed in December 2015 prices. 9 The purchasing power parity conversion factor for private consumption was Br 8.189 for US$1 in 2015 (World Bank 2018). 10 This is in line with the National Planning Commission’s interim poverty report, which shows that urban poverty decreased by 11 percentage points between 2011 and 2016, compared with 5 percentage points in rural areas (FDRE 2017). 11 The forthcoming Poverty Assessment will examine in detail the drivers of consumption movements between 2011 and 2016. It is possible that the decline in median consumption in Afar and Amhara was related to the El Niño drought. Mean, as opposed to median, consumption increased in all regions, including Afar and Amhara, although not in a statistically significant fashion (see annex A). 42 increase in median consumption was observed Amhara with a 56 percent rise. The lowest urban in Harari (49 percent), followed by Dire Dawa consumption increase was observed in Somali (27 percent) and Gambella (25 percent). 12 (10 percent). Rural consumption levels increased Urban consumption increased in all regions, but significantly in Tigray, Gambella, and Harari. In there was large variation in the magnitude: the the other regions, the change in rural consumption highest increase happened in Dire Dawa, where between 2011 and 2016 was not statistically consumption increased by 66 percent, followed by different from zero (table 2.1 and annex A). Table 2.1: Household Consumption Increased in All Regions Except Afar and Amhara (regional median annual consumption per adult equivalent, December 2015 birr) Total Urban Rural Region 2011 2016 % 2011 2016 % 2011 2016 % change change change Tigray 9,308 10,749 15.5 13,786 15,665 13.6 8,604 9,705 12.8 Afar 9,031 8,503 -5.9 11,317 15,339 35.5 8,277 7,892 -4.7 Amhara 9,395 9,219 -1.9 10,289 16,095 56.4 9,301 8,758 -5.8 Oromia 9,748 10,993 12.8 10,758 14,090 31.0 9,615 10,894 13.3 Somali 9,197 10,195 10.9 11,052 12,143 9.9 8,868 10,128 14.2 Benishangul-Gumuz 9,671 10,641 10.0 11,640 14,659 25.9 9,506 9,971 4.9 SNNPR 9,278 9,972 7.5 10,308 14,089 36.7 9,169 9,692 5.7 Gambella 9,134 11,382 24.6 10,304 13,862 34.5 8,837 10,210 15.5 Harari 11,255 16,739 48.7 12,448 18,392 47.8 10,638 15,607 46.7 Addis Ababa 10,377 12,718 22.6 10,377 12,718 22.6 - - - Dire Dawaa 9,610 12,203 27.0 9,540 15,876 66.4 9,733 11,280 15.9 Sources Household Consumption Expenditure Surveys 2011, 2016; World Bank staff calculations. Note SNNPR = State of Southern Nations, Nationalities, and Peoples’ Region. Consumption increased across the major B, plots the distribution of the agro-ecological zones.) agro-ecological zones in Ethiopia. The agro- Urban consumption increased significantly in all five ecological zones are based on elevation and rainfall agro-ecological zones. However, in rural areas, volume and variability and are labeled as (i) moisture- consumption increased in the highlands (drought- reliable highlands, (ii) drought-prone highlands, prone and moisture-reliable) but did not change in (iii) moisture-reliable lowlands, (iv) drought-prone a statistically significant fashion in the lowland or lowlands, and (v) pastoral areas. (map B.1, in annex pastoral areas (figure 2.2). 12 Harari and Dire Dawa are predominantly urban, and Addis Ababa is completely urban. The findings are consistent with the higher consumption growth observed in urban areas in general. 13 Although overall (urban and rural combined) and urban mean consumption increased in all five agro-ecological zones, rural consumption decreased in the drought-prone lowlands (and increased in the other ecological zones, although the magnitude was small). See annex A. 43 Figure 2.2. Consumption Increased in All Agro-Ecological Zones, Particularly in Urban Areas (median annual consumption per adult equivalent, December 2015 birr) 20,000 Median annual consumption per adult 18,422 16,000 16,204 15,617 (December 2015 birr) 12,000 13,877 13,535 12,731 10,967 11,101 10,572 10,385 10,512 8,000 9,907 9,875 8,971 9,716 9,438 9,301 9,028 8,971 8,683 4,000 0 Drought Drought Moisture Pastoralist Drought Drought Moisture Moisture Pastoralist prone prone reliable prone prone reliable reliable highland lowland highland highland lowland highland lowland Urban Rural 2011 2016 Source Household Consumption Expenditure Surveys 2011, 2016; World Bank staff calculations. Trends in calorie intake roughly conform intake decreasing most notably in the rural areas to the trends in consumption expenditures. of Amhara and Benishangul-Gumuz. Looking at At the national level, the median daily calorie agricultural zones, the drought-prone lowlands consumption per adult increased from 2,668 experienced a significant decrease in calorie to 2,809 kilocalories (kcal) (an increase of 5.3 intake. Although this may have been a result percent). It increased from 2,518 to 2,666 kcal of the 2015/16 El Niño drought, quantitative in urban areas and from 2,725 to 2,831 kcal research has so far failed to find a clear negative in rural areas (figure 2.3).14 There is substantial impact of the drought on household-level variation across regions, with median calorie outcomes (box 2.2). 14 It may seem counterintuitive that calorie intake is lower in urban than in rural areas. However, the nature of work in rural areas (working the fields) requires a higher intake of calories compared with the nature of most work in urban areas. In addition, urban households tend to consume more expensive sources of calories, such as teff and animal products. 44 Figure 2.3. Calorie Consumption Showed a Modest Increase between 2011 and 2016 (median daily calorie consumption per adult equivalent) 3,000 2,809 2,931 2,668 2,666 2,725 Median and daily calorie consumption per adult (kcal) 2,518 2,500 2,000 1,500 1,000 500 0 National Urban Rural 2011 2016 Source Household Consumption Expenditure Surveys 2011, 2016; World Bank staff calculations. Box 2.2 The (Non) Impact of the El Niño Drought? In 2015/16, Ethiopia was hit by the El Niño agricultural production in 2015/16 decreased drought, which was labeled the worst drought only marginally (and remained higher than in five decades. The failure of two consecutive overall production levels two years earlier in rainy seasons in 2015 led to a sharp increase 2013/14).b The food security indicators in the in humanitarian requirements, with more than Welfare Monitoring Survey tell a largely similar 10 million Ethiopians in need of humanitarian story: the share of the Ethiopian population food aid, on top of the chronically food- who experienced a food shortage in the 12 insecure Productive Safety Net Programme months prior to the survey decreased from 22 caseload of eight million.a Government and percent in 2011 to 10 percent in 2016 (table development partners mounted a large-scale B2.2.1). The food gap—the number of months humanitarian response, which was credited a household experiences food shortage— with averting any loss of life due to starvation. remained the same but, given that the gap only refers to those households who actually Despite the severity of the drought, it is experienced food shortages, it also decreased difficult to observe its aggregate effects at the aggregate level. At the regional level, in official statistics. Chronic malnutrition only Benishangul-Gumuz experienced a self- decreased between 2011 and 2016; acute reported increase in food insecurity from a malnutrition remained unchanged; and overall low base. 45 Table B2.2.1: Food Security Improved between 2011 and 2016 incidence of food shortages and average duration of food shortages—food gap) 2011 2016 Region Food shortage (%) Food gap (months) Food shortage (%) Food gap (months) Tigray 13.2 3 11.9 2.5 Afar 7.7 5.2 9 3.8 Amhara 23.2 3.1 10.4 3 Oromia 16 3.1 10.5 3.6 Somali 30.3 4.4 6 3 Benishangul-Gumuz 5.6 2.1 8.5 2.8 SNNPR 35 3.4 12.6 3.2 Gambella 31.6 2.6 3.8 1.2 Harari 8 3.2 0 - Addis Ababa 7.8 4 1.1 3.6 Dire Dawa 13.5 1.6 7.7 2 National 21.6 3.3 10.2 3.3 Sources Welfare Monitoring Surveys 2011, 2016; World Bank staff calculations. Note The food gap is only calculated for those households who reported a food shortage. SNNPR = State of Southern Nations, Nationalities, and Peoples’ Region. Figure B2.2.1 Bad Rains… Figure B2.2.2 Good Vegetation… (rainfall deviations from long-term averages) (VCI) 4 100 Distribution Z-scores Rain 80 2 Vegetation (VCI) 60 0 40 -2 20 -4 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Sources Household Consumption Expenditure Surveys Sources Sohnesen 2018, based on data from NASA’s 2011, 2016; World Bank staff calculations. National Oceanic and Atmospheric Administration. Note SNNPR = State of Southern Nations, Note VCI measures the state of vegetation in a given Nationalities, and Peoples’ Region. year compared with similar periods in the previous years. VCI = Vegetation Condition Index. 46 Figure B2.2.3. Food Aid Targeted Locations with Unusually Poor Vegetation (likelihood of receiving food aid as a function of the VCI) 5 4 HH received free food 3 2 1 0 0 20 40 60 80 100 Vegetation meher harvest (VCI) Sources Sohnesen 2018. Note Local polynomial smoothing between receiving food aid and the VCI. HH = households; VCI = Vegetation Condition Index. a. Based on the 2016 Humanitarian Requirements Document. b. Based on the Ethiopian Demographic and Health Surveys (2011 and 2016) and Agricultural Sample Surveys (2011–16). c. When using a self-reported indicator of drought exposure, there is a large negative impact of (self-reported) drought exposure on consumption. This is likely due to endogeneity: people who have had a bad year are more likely to report having been exposed to shocks. 2.3 Patterns of Growth in Consumption Expenditure Consumption Growth Concentrated for each percentile of the distribution, ranging in the Upper Parts of the from the poorest 1 percent to the richest 1 percent. Distribution Growth for the bottom 20 percent was between 0 and 1 percent per year, in contrast to the top The patterns of consumption growth of the distribution where growth rates reached a between 2011 and 2016 were very different maximum of just under 6 percent per year. The at the bottom and top of the distribution. overall average increased by 2.4 percent per Figure 2.4 shows the average annual percentage year, while the median (50th percentile) grew at change in consumption between 2011 and 2016 2 percent per year, on average. 47 Figure 2.4. Consumption Growth Was Zero for the Poorest and Strongly Positive for the Richest (average annual growth rate of consumption by percentile between 2011 and 2016) 6 Annual mean growth rate (%) 5 4 3 2 1 0 -1 -2 0 10 20 30 40 50 60 70 80 90 100 Percentiles 95% confidence bounds Source Household Consumption Expenditure Surveys 2011, 2016; World Bank staff calculations. Growth rates were strong and positive variable. Figure 2.5, panel b, shows that growth across the whole urban distribution. was zero or negative for the poorest households in Annual growth across the urban consumption rural areas. The growth rate picked up to around distribution was always above 3 percent and 1 percent at the median, but then dropped back to became increasingly strong toward the upper just above zero until around the 80th percentile. As end (figure 2.5, panel a). The growth rate of the was the case for urban areas, the highest growth mean in urban areas was almost two-and-a-half rates were experienced at the top of the rural times the national average, at 5.9 percent per distribution, although they were relatively lower year. There was a similar average growth rate at around 3 percent per year. The rural average for the urban median. grew at 1 percent per year, a rate that was almost six times lower than in urban areas. Rural growth In contrast, rural consumption growth was statistically different from zero only in Tigray, rates were generally quite low and more Gambella, and Harari. 48 Figure 2.5. Growth Was Strong and Positive in Urban Areas but Lower and Variable in Rural Areas (average annual growth rate of consumption by percentile between 2011 and 2016) Urban Rural 8 8 7 7 Annual mean growth rate (%) Annual mean growth rate (%) 6 6 5 5 4 4 3 3 2 2 1 1 0 0 -1 -1 -2 -2 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 Percentiles Percentiles 95% confidence bounds 95% confidence bounds Source Household Consumption Expenditure Surveys 2011, 2016; World Bank staff calculations. The pattern of growth in rural areas is become more impoverished. However, it does a continuation of the pattern observed mean that whoever was in the bottom 15 percent between 2005 and 2011. Consumption of of rural welfare in 2016 had lower monetary the bottom 15 percent in rural areas contracted living standards than whoever was in the bottom between 2005 and 2011 (figure 2.6) (World Bank 15 percent in 2005. The indicators of poverty 2015), much the same as it did between 2011 severity that will be presented later are consistent and 2016 (figure 2.5, panel b). Given the cross- with this pattern. These numbers suggest the sectional nature of the HCES data, this does not existence of a segment of the rural population necessarily imply that households who were that has not benefited from Ethiopia’s economic poor to begin with (in 2005) have increasingly growth (box 2.3). Figure 2.6. Growth Was Negative for the Bottom 15 Percent in Rural Areas between 2005 and 2011 (average annual growth rates of rural consumption by percentile, 2005–11) 6 5 Annual mean growth 4 3 rate (%) 2 1 0 -1 -2 0 10 20 30 40 50 60 70 80 90 100 Percentiles 95% confidence bounds Source Household Consumption Expenditure Surveys 2005, 2011; World Bank staff calculations. 49 Box 2.3 Chronic and Transitory Poverty in Ethiopia: Evidence from the Ethiopia Socioeconomic Survey The Household Consumption Expenditure that is always poor is one that is observed to Surveys interview a different sample of be below the poverty line in all three rounds a households in different years. Although of the ESS. A household that is usually poor this allows the surveys to get information is one in which the average of consumption on the change in poverty over time, it does expenditure over the three rounds of the not allow studying the fate of individual ESS is below the poverty line, although the households over time. The Ethiopia household is not poor in all three rounds. Socioeconomic Survey (ESS), implemented Transitory poverty is associated with those in 2012, 2014, and 2016, follows the same households that are occasionally poor— sample of households in three successive average consumption expenditure over the survey rounds. The longitudinal nature of the three rounds of the ESS is above the poverty ESS means that households can be separated line, but the household is poor in at least into those that are chronically poor and one round. those that experienced transitory poverty. The importance of this for policy purposes is Overall, 16 percent of the population in highlighted by the fact that an appropriate rural areas and small towns in Ethiopia policy response to chronic poverty would focus was chronically poor, according to the on increasing the attainment of and returns to definition outlined above. This was the assets (human and physical) of the poor. significantly lower than the 31 percent of Transient poverty would be better tackled households that experienced transitory through initiatives that focus on insurance and poverty over the period. Economic mobility income stabilization. Accordingly, conclusions (moving above and below the poverty line about longer-run welfare depend on how over time) was very high over the period much mobility is present over time in society and goes some way to explaining why the (Lipton and Ravallion 1995). transitory component is so high for these households. Overall, taking account of Chronic poverty status is assigned to chronic and transitory poverty, about half the households who are always poor or population in rural areas and small towns usually poor over the three rounds of the experienced at least one spell of poverty ESS (2012, 2014, and 2016). A household between 2012 and 2014. 50 Figure B2.3.1. Chronic and Transitory Poverty over ESS 2012 to ESS 2016: Households in Rural Areas and Small Towns 100% 90% 80% 37.2 40.9 52.8 53.3 70% 58.2 70.1 60% 50% 40% 43.0 31.5 30% 31.4 29.8 33.7 20% 23.6 10% 15.7 19.9 27.6 17.0 6.2 8.1 0% Rural & small Amhara Oromia SNNP Tigray Others towns Chronic poor Transient poor Never poor Sources Staff calculations from ESS 2012, 2014, and 2016. Note ESS = Ethiopia Socioeconomic Survey; SNNPR = State of Southern Nations, Nationalities, and Peoples’ Region. There were large regional differences in located in moisture-reliable highlands were the extent of chronic poverty.b Chronic disproportionately more likely to have been in poverty rates were highest in the State chronic poverty between 2012 and 2016. of Southern Nations, Nationalities, and Peoples’ Region (SNNPR) and Amhara, Apart from being more likely to be located at 28 and 20 percent of the rural and small- in SNNPR and Amhara, chronically town population, respectively. Six percent of this poor households in Ethiopia differ from population in Oromia was in chronic poverty transitory poor or never poor households between 2012 and 2016, while the share in in several characteristics (table B2.3.1). Tigray was 8 percent. Rates of transitory poverty Chronically poor households are larger, were a bit more evenly spread throughout the contain more young children, and have regions, ranging from 24 percent in Oromia higher dependency ratios than transitory poor to 43 percent in Amhara. A little over half of households and those that were never poor.c all households were never poor between 2012 They have less arable land, fewer assets (as and 2016 (shown by the grey bars in figure measured by their score on a composite asset B2.3.1). This aggregate number hides some index), and are more likely to be headed large regional differences. Only 37 percent of by a male with little to no education.d They households in Amhara were not poor in any one are more likely to have experienced food of the three rounds of the ESS, compared with shortages, and their children are less likely 58 percent in Tigray and 70 percent in Oromia. to go to school. Surprisingly, chronically poor Shifting the focus to the five agro-ecological households are not significantly more remote zones of Ethiopia shows that households than those households that did not experience 51 poverty between 2012 and 2016. The average and never poor were 16 and 14 kilometers, chronically poor household was located about respectively. Proximity to the nearest population 14 kilometers from the nearest road, while the center of 20,000 people or more also did not corresponding distances for the transitory poor vary much between the groups. Table B2.3.1: Profiles by Poverty Status, Nonurban Households, 2012 to 2016 Characteristic (1) (2) (3) Diff. Diff. Chronic Transitory Never poor (1) versus (1) versus poor poor (2) (3) Household Household size 59 5.3 4.9 *** *** Dependency ratio (%) 101 89 74 * *** Land per adult (hectares) 0.32 0.43 0.57 *** *** Asset index -0.4 -0.2 0.2 ** *** Distance to nearest road (kilometers) 13.7 16.7 14.0 *** HH ever in PSNP (W1 to W3) (%) 25.2 20.8 19.0 * Not enough food last 12 months (%) 47.4 35.6 27.8 *** *** Children ages 7-12 enrolled (%) 63.4 69.2 74.6 ** *** Household head Age (years) 43.8 46.0 44.4 ** ** Female (%) 17.5 18.2 22.0 ** *** No education (%) 70.3 71.5 57.3 ** *** Primary (%) 29.2 25.7 35.2 * * Secondary (%) 0.4 2.4 4.2 *** *** Post-secondary (%) 0.0 0.5 2.9 ** *** Sources Staff calculations from ESS 2012, 2014, and 2016. Note ESS = Ethiopia Socioeconomic Survey; HH = household; PSNP = Productive Safety Net Programme; W = Wave. *** significant difference at 1% level, ** 5% level, * 1% level. a. The poverty line in this analysis is set at the 25th percentile of the 2016 ESS distribution of consumption expenditure, which roughly corresponds to the official 2016 poverty rate. b. In this context, the region in which a household is located is the region that was recorded in 2012. c. The definition of the dependency ratio comes from the Central Statistical Agency of Ethiopia (2017) and is “…the population that is not of working age (<15 and >64) divided by total number of working age persons (15-64 years). The value is then multiplied to express it in percent. Households with no working persons were excluded in the dependency ratio computation.” d. The conventional wisdom is that female-headed households are more likely to be poor. This is mistaken. 52 Although growth was higher in the from Br 26,000 to Br 33,000 (per adult equivalent upper half of the distribution, most of per year). the population experienced an increase in consumption between 2011 and 2016. In urban areas, there was consumption Only in the first decile (the poorest 10 percent growth across the distribution. Consumption of the population) was there no growth in real expenditure grew by Br 800 for the poorest 10 consumption levels: consumption expenditure in percent, and by almost Br 14,000 for the richest the poorest decile was constant at around Br 3,800 10 percent of urban households.15 Figure 2.8, per adult equivalent per year (in December 2015 panel b confirms that although every urban prices). Small increases took place for deciles 2 to decile experienced a real increase in consumption 8, while average consumption in decile 9 jumped expenditure, growth was strongest at the top. from Br 15,700 to Br 18,000. The biggest real Growth in rural households was a lot flatter—the gains took place for the richest 10 percent, whose top 10 percent of rural households saw an increase average real consumption expenditure increased of Br 2,400 on average between 2011 and 2016. Figure 2.7. Real Consumption Increased Everywhere Except the First Decile (real annual consumption expenditure, by decile, between 2011 and 2016, December 2015 birr) 40,000 Overall 2011 Overall 2016 35,000 Average annual consumption 30,000 25,000 20,000 15,000 10,000 5,000 0 Poorest 2 3 4 5 6 7 8 9 Richest Consumption decile Urban Rural 60,000 60,000 50,000 50,000 40,000 40,000 30,000 30,000 20,000 20,000 10,000 10,000 0 0 Poorest 2 3 4 5 6 7 8 9 Richest Poorest 2 3 4 5 6 7 8 9 Richest Urban 2011 Urban 2016 Rural 2011 Rural 2016 Source Household Consumption Expenditure Surveys 2011, 2016; World Bank staff calculations. 15 This is in contrast to 2005–11, during which the bottom 10 percent of the urban population contracted in consumption expenditures (World Bank 2015). 53 Given that consumption growth between with, experienced faster consumption growth 2011 and 2016 was higher in the upper between 2011 and 2016. half of the distribution, inequality increased somewhat. The Gini coefficient of The slight increase in inequality should not inequality increased from 0.29 in 2011 to 0.33 be a cause for concern, at least for now. This in 2016 (FDRE 2017). This inequality level is still pattern is a familiar one in fast-growing and urbanizing low and lower than the average for Sub-Saharan countries and can be expected to continue until rural African countries. An urban-rural decomposition and urban welfare levels start to converge through shows that there was a substantial increase in increased migration to urban areas. It would only be the contribution of inequality between urban and a cause for concern in a scenario of restricted labor rural areas to overall inequality; the contribution of mobility, where workers from rural areas cannot move inequality within urban and rural areas decreased. freely to more economically dynamic urban areas. In In other words, inequality increased because urban that case, urban-rural income gaps could be expected households, who were already better-off to begin to increase further, pushing up spatial inequality. 2.4 Brief Snapshot of Current Monetary Living Standards Despite strong poverty reduction over standards are low for the bottom 80 percent of the past 20 years, standards of living in the population but sharply increase in the fifth Ethiopia remain low. In the 2015/16 survey, quintile (figure 2.9). The fifth quintile is structurally median household consumption per capita different from the other quintiles, in the sense that it amounted to Br 704 per month.16 Converting this is predominantly urban, better educated, and much amount to U.S. dollars and accounting for the more likely to be wage-employed in the public differences in purchasing power, this means that sector or self-employed in the formal private sector. half the population in Ethiopia lives on consumption In comparative perspective, poverty in Ethiopia is expenditure of purchasing power parity (PPP) US$86 low relative to the country’s income level, but the per month or less.17 For rural areas, this amounts extent to which growth has translated into poverty to US$82 (PPP adjusted). Overall, monetary living reduction has also been low. 16 Thisfigure is expressed on a per capita basis. As a result, it is different from the figure presented in section 2.2, which uses a per adult equivalent basis. 17 The PPP conversion factor for private consumption was Br 8.189 for US$1 in 2015 (World Bank 2018). 54 Figure 2.8. Monetary Living Standards Remain Low for the Majority of the Population (daily expenditures per capita, US$ PPP) 8 7.2 Expenditures per capita per day 7 6 5 3.6 4 2.8 3 2.2 2 1.3 1 0 Q1 Q2 Q3 Q4 Q5 Quintile of consumption per capita Sources Household Consumption Expenditure Survey 2016; World Bank staff calculations. Note PPP = purchasing power parity. Although Ethiopia has managed to continue report, poverty severity in rural areas did not change its progress on poverty reduction in adverse between 2011 (poverty severity of 3.2) and 2016 circumstances (the El Niño drought), the (poverty severity of 3.1) and remains higher than in severity of poverty has remained stubbornly 2005 (2.7) (FDRE 2017).18 There is also significant sticky. Poverty severity is an indicator of the regional variation. Poverty severity increased in the distance that separates the poor from the poverty rural areas of Oromia, Harari, and Dire Dawa.19 line, and it gives more weight to the very poor It decreased in Afar, Somali, Benishangul-Gumuz, (those households with the lowest consumption SNNP and Gambella while it remained the same expenditures). According to the official poverty in Tigray and Amhara (figure 2.10). Figure 2.9. Rural Poverty Severity increased in Oromia, Harari and Dire Dawa (Rural poverty severity across regions: 2011 and 2016) 0.050 0.040 Severity of poverty 0.030 0.020 0.010 0.000 Tigray Afar Amhara Oromia Somali B.Gumuz SNNP Gambella Harari Dire Dawa 2011 2016 Source Household Consumption Expenditure Surveys 2011, 2016; World Bank staff calculations 18 FDRE, 2017. 19 The increase in poverty severity does not mean that poverty increased (poverty decreased). Rather, it means that the (fewer) people who were poor in 2016 were on average poorer than the people who were poor in 2011. 55 2.5 Perspectives on Poverty Reduction Going Forward The analysis presented in this Economic In an ideal scenario, the children of Update is largely descriptive and, as such, extremely poor households would cannot make too strong a statement on accumulate more education and be able to the way forward for poverty reduction in move out and diversify into more productive Ethiopia. The upcoming Poverty Assessment activities, breaking the intergenerational will provide a deep analysis of poverty trends transmission of poverty. Better-educated young and make more links to public action. There are people from rural areas tend to move to urban nevertheless some important questions to extract areas and substantially improve their material from the recent Welfare Monitoring Survey and conditions. Despite the improvements in access to HCES data and existing academic and policy education, few youths in rural areas in Ethiopia research. These questions relate to the role of complete primary school. According to the 2015 agricultural development in reaching the poorest Welfare Monitoring Survey, 22 percent of youth of the poor, the state of education in rural Ethiopia, ages 15-24 years in rural Ethiopia had completed and population dynamics and job creation. primary school, dropping to 14 percent in the bottom quintile (the poorest in rural areas). Overall, Although improvements in agriculture have the median years of education obtained by rural driven poverty reduction over the past 15 youth (ages 15-24) in Ethiopia amounts to five years (World Bank 2015), it begs the question (lower for those in poor households). With these as to what extent the smallholder-focused education statistics, it is difficult to imagine that the poverty reduction strategy has been able to living standards of today’s poor children or youth reach the poorest of the poor in rural areas. will improve much as they enter adulthood and start Successive HCES surveys show that the bottom 10- their own families. Devising policies or interventions 15 percent in rural areas experienced no growth to keep children from poor rural families in school in consumption between 2005 and 2015, resulting longer will be crucial in any attempt to share the in rural poverty severity that was higher in 2015 benefits of growth more widely. than in 2005. The chronic poor in rural Ethiopia have less land and bigger families, meaning that In a context of increasing land scarcity, the little land they have will become even more the labor market will become increasingly subdivided in the future generation. All in all, this important as a transmission mechanism suggests that there will not be an agriculture-based between aggregate growth and poverty pathway out of poverty for Ethiopia’s chronic poor.20 reduction. The Government of Ethiopia Although social programs like the Productive Safety acknowledges this and expanding employment Net Programme provide much-needed relief to opportunities is a priority area in the second extremely poor rural families, the question remains Growth and Transformation Plan. The challenge whether this is, on its own, sufficient to lift ultra- is daunting: over the next 10 years, the working- poor households to a sustainably better standard age population in Ethiopia will grow at two million of living while allowing them to invest in the human per year, dramatically increasing the population’s capital of their children. demand for jobs. Even if Ethiopia’s manufacturing 20 Atleast not in own-account agriculture on small and degraded fields. Agricultural wage labor on other farms or commercial farms could potentially be a pathway. 56 drive turns out to be highly successful, the number per woman (in 2016) (Ethiopia Demographic and of jobs created will not come close to the number Health Surveys 2000, 2011, and 2016). In addition, required. Successfully addressing the employment the fertility decline has stalled in recent years: the total challenge will require interventions, improvements, fertility rate was largely similar in 2016 (4.6) to what and innovations in many different sectors and it was in 2011 (4.8). The persistently high fertility dimensions and a critical review of existing rate means that future youth cohorts will become employment policies and strategies. larger and the challenges for education and job creation will become bigger. To illustrate, there are A final reflection, linked to the employment currently 21 million youths (ages 15-24) in Ethiopia. challenge, relates to issues of fertility and By 2030, there will be close to 28 million (UN DESA family planning. Progress on poverty reduction, job 2015). Achieving faster declines in fertility through creation, and schooling is complicated by persistently improved access to family planning, girls’ education, high fertility rates. Although the total fertility rate and better female labor force participation will be in Ethiopia declined from 5.5 in 2000 to 4.6 in required to manage the twin challenges of education 2016, rural fertility remains high, at 5.2 children and employment in the decades to come. 57 Annex A: Trends in Mean Consumption Expenditures ANNEXES 58 Figure A.1. Mean Annual Consumption per Adult, 2011 and 2016 (December 2015 birr) 20,000 18,649 11,014 Mean annual consumption per adult 16,000 13,901 10,433 12,500 (December 2015 birr) 12,000 11,008 8,000 4,000 0 National Urban Rural 2011 2016 Source Household Consumption Expenditure Surveys 2011, 2016; World Bank staff calculations. Table A.1: Regional Mean Annual Consumption per Adult, 2011 and 2016 (December 2015 birr) Total Urban Rural Region 2011 2016 % 2011 2016 % 2011 2016 % change change change Tigray 11,630 14,108 21.3 17,691 20,536 16.1 10,074 12,038 19.5 Afar 10,641 12,902 21.2 13,945 18,645 33.7 9,298 10,463 12.5 Amhara 10,944 12,340 12.8 14,325 21,879 52.7 10,464 10,557 0.9 Oromia 10,947 12,022 9.8 13,891 18,080 30.2 10,504 11,022 4.9 Somali 10,565 11,714 10.9 12,942 14,470 11.8 10,004 11,063 10.6 Benishangul-Gumuz 11,435 13,373 17.0 15,124 18,524 22.5 10,832 12,112 11.8 SNNPR 10,725 12,204 13.8 13,391 18,049 34.8 10,414 11,157 7.1 Gambella 10,334 13,855 34.1 12,477 17,945 43.8 9,325 11,745 26.0 Harari 13,264 21,059 58.8 15,344 24,028 56.6 11,397 17,479 53.4 Addis Ababa 12,831 16,237 26.5 12,831 16,237 26.5 - - - Dire Dawa 11,268 17,428 54.7 11,617 20,718 78.3 10,532 11,393 8.2 Sources Household Consumption Expenditure Surveys 2011, 2016; World Bank staff calculations. Note The increase in mean consumption is statistically significant in all regions except Somali. For rural areas, the increase in mean consumption is statistically different from zero only in Tigray, Gambella, and Harari. SNNPR = State of Southern Nations, Nationalities, and Peoples’ Region. 59 Figure A.2. Mean Annual Consumption per Adult in the Five Agro-Ecological Zones, Urban and Rural (December 2015 birr) 25,000 24,876 Mean annual consumption per adult 20,000 20,913 20,341 (December 2015 birr) 17,651 17,421 15,000 15,838 14,495 13,973 13,664 12,458 12,081 11,860 10,880 11,410 10,839 10,688 10,319 10,081 10,112 10,000 9,310 5,000 Drought Drought Moisture Pastoralist Drought Drought Moisture Moisture Pastoralist prone prone reliable prone prone reliable reliable highland lowland highland highland lowland highland lowland 0 Urban Rural 2011 2016 Sources Household Consumption Expenditure Surveys 2011, 2016; World Bank staff calculations.t Note The increase in mean consumption is statistically significant in all agro-ecological zones. For rural areas, the increase in mean consumption is statistically different from zero only in the highlands (drought-prone and moisture-reliable). 60 Annex B: The "Five Ethiopias" Map B.1. Agro-Ecological Zones of Ethiopia Drought prone, highland Drought prone, lowland Humid moisture reliable, lowland Moisture reliable, highland Pastoralist N/A 0 100 200 300 400 500km Sources : Note The moisture-reliable highlands-cereal and moisture-reliable highlands-enset are lumped into one group for the analysis. The nonclassified parts of Somali and Afar are considered pastoral. 61 References Bachewe, F., F. Yimer, B. Minten, and P. Dorosh. 2016. “Agricultural Prices during Drought in Ethiopia: An Assessment Using National Producer Data (January 2014 to January 2016).” ESSP Working Paper 88, International Food Policy Research Institute, Washington, DC. Bourguignon, F. 2003. “The Growth-Elasticity of Poverty Reduction: Explaining Heterogeneity across Countries and Time-Periods.” World Bank, Washington, DC. Chenery, H., S. Robinson, and M. Syrquin. 1986. Industrialization and Growth: A Comparative Study. Washington, DC: World Bank. FDRE (Federal Democratic Republic of Ethiopia). 2012. Ethiopia’s Progress Towards Eradicating Poverty: An Interim Report on Poverty Analysis Study 2010/11. Addis Ababa, Ethiopia: National Planning Commission. ———. 2017. Ethiopia’s Progress Towards Eradicating Poverty: An Interim Report on 2015/16 Poverty Analysis Study. Addis Ababa, Ethiopia: National Planning Commission. FEWS NET (Famine Early Warning System Network). 2018. “Ethiopia Food Security Outlook: October 2018 to May 2019.” FEWS NET, Washington, DC. Hirvonen, K., T. Sohnesen, and T. Bundervoet. 2018. “Impact of Ethiopia’s 2015 Drought on Child Undernutrition.” ESSP Working Paper 114, International Food Policy Research Institute, Washington, DC. IMF (International Monetary Fund). 2018a. Sub-Sharan Africa Regional Outlook, April. Washington, DC: IMF. ———. 2018b. World Economic Outlook. Washington, DC: IMF. Klasen, S., and M. Misselhorn. 2008. “Determinants of the Growth Semi-Elasticity of Poverty Reduction.” Discussion Papers 176, Ibero America Institute for Economic Research, University of Göttingen, Germany. Lewis, W. A. 1954. “Economic Development with Unlimited Supplies of Labour”. The Manchester School 22(2): 139-191 Lipton, M., and M. Ravallion. 1995. “Poverty and Policy.” In Handbook of Development Economics, ed. 1, vol. 3, edited by J. Behrman and T. N. Srinivasan, 2551–2657. Amsterdam: North Holland. Ravallion, M. 2012. “Why Don’t We See Poverty Convergence?” American Economic Review 102 (1): 504–23. 62 Sohnesen, T. 2018. “Measurement and Economic Impact of Drought: The Case of Ethiopia’s El-Nino Drought.” UN DESA (United Nations Department of Economic and Socail Affairs). 2015. World Population Prospects, 2015 Revision. New York: UN DESA. World Bank. 2005. Introduction to Poverty Analysis. Washington, DC: World Bank Institute. ———. 2015. Ethiopia Poverty Assessment. Washington, DC: World Bank. World Bank 2018 World Bank Group. 2018. Africa’s Pulse, No. 17, April. Washington, DC: World Bank. World Bank and WTO (World Trade Organization). 2017. The Role of Trade in Ending Poverty. Washington, DC: World Bank.