Report No: AUS7544 . Republic of Djibouti Poverty and Social Impact Analysis: Strengthening Safety Nets in Djibouti . February 2015 . . Acknowledgements The PSIA team from the World Bank comprises Stefanie Brodmann, (Senior Economist and task team leader), Harold Coulombe (Consultant), Robert Bacon (Consultant), Ines Rodriguez Caillava (Consultant) and Angela Elzir (Junior Professional Associate). Paolo Verme (Senior Economist), Abdoulaye Sy (Country Economist) and Ilhelm Salamon (Senior Economist) are part of the extended team. The International Monetary Fund (IMF) is represented by Abdurahman Aden. On the Djibouti side, the national technical committee was composed of Amina Warsama, Mouna Ahmed Ragueh, and Zeinab Ahmed Houssein (Secretary of State responsible for National Solidarity), Almis Mohamed Abdillahi and gentlemen (Ministry of Budget), Idriss Abdillahi Orah (Ministry of Economy and Finance, responsible for Industry) Houmed Gaba- Omar (Ministry of Energy), Aref Omar Wahib (Ministry of Transport) and Yacin Abdi Farid (Department of Statistics and Demographic Studies, DISED). The team is grateful to HE Zahra Youssouf Kayad (Secretary of State responsible for National Solidarity), Simon Mibrathu (Secretary General, Ministry of Budget), Idriss Ali Sultan (Director, DISED), Sekou Konate Tidiani (Statistician, DISED) Homa Fotouhi (Resident Representative, World Bank) and Yasser El-Gammal (Manager Social Protection, World Bank) for their guidance in the preparation of this study. . Standard Disclaimer: . This volume is a product of the staff of the International Bank for Reconstruction and Development/ The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. . . Contents Acknowledgements ............................................................................................................................ 2 Tables ...................................................................................................................................................... v Figures ..................................................................................................................................................... v EXECUTIVE SUMMARY ............................................................................................................................ 1 What is the Current Nature of Tax Exemptions in Djibouti? .............................................................. 2 What is the Impact of the Current Tax Exemptions on Household Welfare? .................................... 3 What is the Context of such Reform: Winners and Losers? ............................................................... 4 What is the Current Role of Social Safety Nets in Djibouti? ............................................................... 5 What is the Impact of the Reforming Tax Exemptions and Safety Nets on Poverty? ........................ 6 COUNTRY AND REFORM CONTEXT ......................................................................................................... 9 DATA SOURCES AND TARGETING ........................................................................................................... 9 EFFECTIVENESS OF THE CURRENT SOCIAL SAFETY NET ........................................................................ 12 Who are the poor? ............................................................................................................................ 12 Female-headed households .............................................................................................................. 14 Human capital and poverty............................................................................................................... 16 Capacity to respond to shocks, including savings and strategies for coping .................................... 17 Effectiveness of the social safety net................................................................................................ 18 EFFECTIVENESS OF TAX EXEMPTIONS AS AN UNTARGETED SAFETY NET ............................................ 24 The import and consumption of petroleum products in Djibouti .................................................... 24 Petroleum products: prices, costs, and taxes ................................................................................... 24 Gasoil prices and transport costs ...................................................................................................... 28 Who consumes energy products? .................................................................................................... 29 Who benefits from tax exemptions? ................................................................................................ 30 Impact of subsidy reforms ................................................................................................................ 32 SIMULATING REFORM OPTIONS ........................................................................................................... 36 Budget and simulations .................................................................................................................... 36 References ............................................................................................................................................ 41 Annex 1: Members of the PSIA Technical Team and Participants in Meetings .................................... 42 Annex 2. PMT Approach and its Efficiency ........................................................................................... 44 Annex 3. Population Demographics...................................................................................................... 51 Annex 4: Survey of Variable Costs OF Passenger Road Transportation ............................................... 52 Annexe 5. Social Assistance Project « Bourse Familiale pour les ménages pauvres et vulnérables des régions de l'intérieur » .......................................................................................................................... 54 Annex 6. Capacity to Respond to Shocks .............................................................................................. 55 iv Tables Table 1. Population Demographics ....................................................................................................... 12 Table 2. Percentage of Vulnerable Populations by Quintile ................................................................. 13 Table 3. Shocks Faced by Households and Capacity to Respond (% of Households).......................... 17 Table 4. Coverage of Transfer Programs .............................................................................................. 19 Table 5. Distribution of Benefits (Targeting Accuracy) ....................................................................... 20 Table 6. Under-coverage and Leakage - Total Poor ............................................................................. 20 Table 7. Relative Incidence - All Households ...................................................................................... 21 Table 8. Generosity - Direct and Indirect Beneficiaries ....................................................................... 21 Table 9. Impact of Programs on Poverty Measures – Simulating the Absence of the Program ........... 22 Table 10. Cost-Benefit Ratios - Upper Poverty Line ............................................................................ 22 Table 11. Decomposition of the Impact of the different Programs....................................................... 23 Table 12. Imports and Domestic Consumption of Petroleum Products by Djibouti in 2012 (Million Liters) .................................................................................................................................................... 24 Table 13. Prices of Transportation Fuels in 2012 ($/liter) .................................................................... 25 Table 14. Retail Prices and Discretionary Taxes for Petroleum Products in 2013 (DF/liter) ............... 27 Table 15. Retail Petroleum Product Prices with and without the Discretionary Tax Element (December 2013) DF/liter ..................................................................................................................... 27 Table 16. Results of Simulation – Range of Retail Prices (DF/liter) .................................................... 28 Table 17. Shares of Operating Cost of a Bus Fleet in Developing Countries ....................................... 29 Table 18. Percentage of Households which Own a Car or Motorbike.................................................. 30 Table 19. Expenditures per Household (in DF) .................................................................................... 31 Table 20. Expenditure on Subsidized Products over Total Expenditures (in %) .................................. 32 Table 21. The Total Impact on the Population’s Well-Being (in millions DF) .................................... 32 Table 22. The Impact on the Per Capita Well-Being (in DF) ............................................................... 33 Table 23. The Impact on Well-Being (in %) ........................................................................................ 33 Table 24. The Impact of the Reform on the Government Revenue (in millions of DF) ....................... 34 Table 25. The Reform, the Destitution Headcount, and the Gini Index ............................................... 35 Table 26. Budgetary Sources (Preliminary Proposal) ........................................................................... 36 Table 27. Definition of the Different Transfer Schemes....................................................................... 37 Table 28. Recommended Calorie Intake ............................................................................................... 38 Table 29. Effect on Destitution Headcount of the Different Transfer Schemes ................................... 39 Table 30. Effect on Destitution Gap of the Different Transfer Schemes .............................................. 40 Table 31. Coverage Rate ....................................................................................................................... 40 Figures Figure 1. Percentage of Urban Population by Quintiles ....................................................................... 13 Figure 2. Level of Education by Quintiles (Percentage) ....................................................................... 14 Figure 3. Type of Job by Quintile (Percentage) .................................................................................... 14 Figure 4. Level of Education by Gender (Percentage).......................................................................... 15 Figure 5. Type of Job by Gender of Household Head (Percentage) ..................................................... 15 Figure 6. Proportion of Household Heads Working and Looking For Work (Percentage) .................. 15 Figure 7. School Attendance by Age and Quintiles .............................................................................. 16 Figure 8. School Attendance by Age Group and Gender...................................................................... 16 Figure 9. Cost-Benefit Ratios ............................................................................................................... 23 Figure 10. Utilization of Different Energy Sources by Location (in %) ............................................... 30 Figure 11. Utilization of Different Energy Sources of Energy, by Quintile (in %) .............................. 30 Figure 12. The Total Impact on the Population Well-Being (in DF) .................................................... 33 Figure 13. The Impact on Well-Being (in %) ....................................................................................... 34 Figure 14. The Impact of the Reform on the Government Revenue (in DF) ........................................ 35 v EXECUTIVE SUMMARY 1. This Poverty and Social Impact Analysis (PSIA) is part of a broader dialogue on energy tax reform and strengthening social safety nets in Djibouti. As part of a possible reform of energy taxes in Djibouti, the government of Djibouti has sought the support of the World Bank to better understand how such a policy reform can be pro-poor. Through this technical assistance, a cross-sectoral Bank team, in close cooperation with the International Monetary Fund (IMF), has supported the government of Djibouti to help understand the following questions: • What is the Current Nature of Tax Exemptions in Djibouti? • What is the Impact of the Current Tax Exemptions on Household Welfare? • What is the Context of such Reform: Winners and Losers? • What is the Current Role of Social Safety Nets in Djibouti? • What is the Impact of the Reforming Tax Exemptions and Safety Nets on Poverty? 2. The study was designed and implemented by a multisectoral committee composed of various stakeholder institutions, including the Ministry of Economy and Finance, the Ministry of Budget, the Secretary of State responsible for National Solidarity (SESN), the Department of Statistics and Demographic Studies (DISED), the Ministry of Energy, and the Ministry of Transport, with whom the teams of the Bank and the IMF collaborated throughout the process of preparation of the study. Technical meetings were held on January 30, February 2, May 25, May 28, and May 29, 2014, in Djibouti to discuss the various scenarios of reform, obtain additional information, and present preliminary quantitative results. Consultation meetings were held on July 2 and November 15, 2014, to present the findings and discuss possible reform options. 3. This executive summary condenses the main findings of the study. The study is available as a separate report with more analyses and background information. The study is based on data from a representative household survey which includes detailed information on household expenditures and receipt of certain cash and in-kind benefits (EDAM 3-2012). The tables in this executive summary show 2014 prices, with inflation rates of 2.5 and 2.9 for 2013 and 2014, respectively. 4. The third Enquête Djiboutienne auprès des ménages (EDAM 3) was conducted in 2012 and has a nationally representative sample of the sedentary population composed of 5,880 households with 31,686 individuals. The EDAM 3 questionnaire covers many aspects: demography, education, employment, mortality, governance, housing, access to basic social services, durable goods ownership, and finally, expenditures and revenues. Of particular importance for this study is information on household expenditure on tax-exempt food (flour, rice, oil, sugar, and milk); certain fuel items (kerosene, butane, and fuel expenditure on transport); and electricity, as well as information on cash and in-kind benefits. The EDAM 3 dataset has been used to compute total expenditure aggregates of households based on which the DISED has produced their recent poverty profile, yielding 40.8 percent of poverty and 23 percent of extreme poverty.1 5. As is common for household surveys, the EDAM 3 data is representative only of the sedentary population. The EDAM 3 sample leaves out the nomad and homeless populations (population flottante) and individuals living in collective households (hotels, prison, military camps, and orphanages). According to the most recent census conducted in 2009, Djibouti’s total population was 818,159 1 The EDAM 3 sample slightly underestimates the size of households, and the level of average per capita total household expenditure is therefore slightly overvalued in this survey. Since this study focuses primarily on expenditure quintiles, the effect of this general overvaluation is marginal. Furthermore, and in contrast to the recently updated national poverty profile that combines data from EDAM 3 and the Budget and Consumption Survey (EBC), this study uses data from EDAM 3 and its expenditure aggregate only. The aggregate used in this study, however, is highly correlated with that used for the poverty profile produced by the DISED and we do not see any conflict between the analysis in this study and the figures recently approved by the government. 1 individuals, of which 161,132 were nomads and 149,022 either lived in collective households or were homeless. Having household surveys solely covering the sedentary population is standard practice since surveying nomad and homeless populations creates important conceptual and logistic issues. 6. Five quintiles based on per capita expenditure have been constructed based on the per capita expenditure welfare index. The first quintile includes the poorest 20 percent of the sedentary Djibouti population; the second quintile includes the next 20 percent, and so on up to the top quintile with the richest 20 percent of the population. For the purpose of this study, the destitution2 line is defined as the upper limit of the first quintile. Therefore the destitution head-count rate is de facto set to 20 percent. WHAT IS THE CURRENT NATURE OF TAX EXEMPTIONS IN DJIBOUTI? 7. Djibouti is vulnerable to major risks to growth and macroeconomic stability, including fuel and food price shocks and natural disasters such as droughts and floods. Poverty has been exacerbated by drought conditions since 2007—the worst in 60 years. The drought is estimated to have affected at least half the rural population, with annual economic losses of 3.9 percent of gross domestic product (GDP) over the period 2008–2011 and a substantial flow of refugees from neighboring countries that also suffer from drought. 8. Universal tax exemptions were introduced in response to the food crisis and to shield the population from price shocks on essential food products. Djibouti depends massively on imports to meet its food needs and a large fraction of the population faces food insecurity. Practically all food items are imported and increases in international food prices directly affect Djibouti’s poor people, who spend up to three-quarters of their income on food. Due to severe and prolonged droughts, at least 20 percent of the capital’s population and three-quarters of rural households are vulnerable to severe or moderate food insecurity, according to the Emergency Food Security Assessment carried out by the World Food Programme in 2013. In response to the stark food price increases, the government has exempt five essential food items from domestic consumption tax since 2008. 9. Similarly, discretionary price adjustments on certain energy products (super, kerosene, and diesel) have been in operation since 2009. The government’s Department of Customs and Excise, after consultation with oil companies, performs a monthly adjustment of prices at the pump to minimize the negative impact of fluctuating international prices of super, kerosene, and diesel. According to estimates of the IMF, Djibouti foregoes an estimated 2 percent of GDP (2011) on certain energy products.3 10. The government is currently considering abandoning the use of the discretionary tax element on certain fuel products (super and diesel) for private consumers; the privileges for other exempt groups such as the military and embassies would remain. At the time of analysis (based on prices of December 2013), such a reform would have resulted in a small fall in super prices and an increase of around 13 percent for diesel. Crude oil prices have recently fallen substantially and this is relevant to the calculations shown. In December 2013, Brent crude sold for about US$110 a barrel and it remained around that level until July 2014. Since then it has steadily declined until falling to around US$50 a barrel in January 2015. There is considerable uncertainty about the course of prices in 2015. 11. Before the drop in oil prices, the government had not taken any firm decision, in part due to fears of increasing inflation with increasing fuel prices (based on the previously higher oil prices). In addition, there are concerns over the impact on the poor as well as the middle class and on certain sectors (transport, fisheries, and bakeries) in particular. The impact of fuel subsidy reforms on the transport sector is of particular concern to the government. Ticket prices for public transport are set by the state and have more or less been stable since 2006. The bus and taxi fleet is outdated and current discussions center on decreasing the cost of transport by updating the fleet. The government is considering pre-financing new vehicles, which the bus and taxi operators would pay back over time, 2 In this report we try to avoid the terms “poverty line” and “poverty headcount” in order to differentiate our analysis from the poverty profile produced by the DISED 3 De Broek, M., A Kangur, and R. Kpodar. 2012. Djibouti: Fuel Price Subsidy Reform. IMF. 2 thereby reducing the consumption of fuel. 12. If the government wanted to abandon the discretionary tax, this would be the time for action. With falling oil prices, an elimination of the discretionary tax elements would not necessarily lead to higher prices for consumers. In fact, given the low prices seen in early 2015, removal of discretionary tax on diesel would be small in comparison to the fall in underlying costs—so that the effect of its removal will be negligible and the effect on bus prices will be easily absorbed. If bus operators do not lower their prices at all then their margins will increase. 13. However, with the elimination of discretionary tax on fuel products, the government would relinquish a tool to smooth fuel prices in times of price increases and decreases. With falling oil prices, government tax revenues will decrease accordingly. The removal of discretionary tax at this point would lower the tax revenue further. It is likely that the government has adjusted the magnitude of the discretionary tax since January 2014, which would warrant further analysis. Furthermore, going forward, an analysis of the optimal tax structure would be warranted. 14. Overall, the following analysis based on December 2013 prices confirms that a negative tax on fuel products effectively subsidizes the better-off. Any reform of the current energy tax system should be pro-poor and social safety nets would be the channel to reinvest savings in pro-poor policies. WHAT IS THE IMPACT OF THE CURRENT TAX EXEMPTIONS ON HOUSEHOLD WELFARE? Distribution of Subsidies on Energy Products 15. The following analysis includes all the tax-exempt fuel products available in the household survey. The survey does not differentiate between diesel and super (lumped together as carburant in the EDAM 3 questionnaire), but data from the Enquête de Budget et Consommation (EBC) survey (an urban-only survey done in 2013) show that around two-thirds of spending by households on carburant is on diesel. Furthermore, the survey shows that almost all direct spending on carburant is by the richest quintile. The simulations assume that the price of fuel purchases will increase from DF 215 to DF 242 per liter. 16. Car ownership and utilization of public transport is a strong indication of welfare. Car ownership is not widespread in Djibouti—only 6 percent of households own a car and 1 percent owns a motorcycle. One-fourth of the richest quintile owns a car while car ownership is basically negligible in the other quintiles. Most cars are owned by urbanites. Not surprisingly, carburant is essentially consumed by urban households and the richest quintile. Utilization of public transport (buses, taxis, and school buses) is also highest among the richer quintiles. Only 12 percent of the poor (first quintile) use public transport compared with 60 percent in the richest two quintiles. More than half of the population in urban areas makes use of public transport but less than 10 percent in rural areas. Utilization of school transportation is also highly skewed toward richer and urban households. 17. Djibouti households spend about DF 7.96 million on subsidized fuel products (that is, fuel at the pump, public, and school transport), about 6.75 percent of their total annual expenditure. On an average, households spend DF 25,400 on fuel at the pump, DF 27,400 on public transport, and DF 30,600 on school transport. Tax exemptions on fuel products do not benefit the poorest as they consume little fuel and hardly use public transportation. Possession of cars and motorbikes is essentially limited to the fifth quintile, which consumes DF 96,847 per household on fuel at the pump, about 4 percent of the total annual household expenditure. Spending on public and school transport is also considerably lower in the poorest quintile (DF 2,142 and DF 2,381 per household, respectively) than in the richest quintile (DF 49,837 per household). Already the second quintile spends considerably more on public transport than the very poor. For the poor, expenses on fuel and public and school transport amount to less than 2 percent each of the overall household expenses (DF 4,522), whereas the richest quintile spends about 8 percent of total household expenditure (DF 197,643) on these fuel products. 3 Distribution of Subsidies on Food Products 18. Poor households spend relatively more on tax-exempt food products than richer households. Household expenses on tax-exempt food products amount on average to DF 153,629 per household, which is equivalent to 12.4 percent of total household spending. Of these basic food items, sugar is the most consumed item in terms of expenditure (DF 37,622). Although rice consumption is higher, only a tiny fraction of rice is actually tax exempt and therefore has been excluded from our analysis. Tax- exempt products are relatively more important for the poor, as the expenditure share of these products is much higher for the very poor than for the very rich. In the poorest households, 19 percent of the total expenses correspond to tax-exempt food products, while these products account for less than 7 percent of the richest households’ total expenses. WHAT IS THE CONTEXT OF SUCH REFORM: WINNERS AND LOSERS? Impact of Abandoning the Discretionary Tax on Retail Prices of Fuel Products 19. The proposal to abandon the use of discretionary tax on certain fuel products is currently under consideration by the government. Other tax rates could be varied by legislation, as at present, but would normally be stable for lengthy periods. Allowable costs along the supply chain could also be varied if justified by the circumstances of the entities involved. To simulate the effect of removing the discretionary tax element on prices, it is assumed that all other tax rates and costs remain at the levels of December 2013. The removal of discretionary tax would have resulted in a small fall in gasoline and kerosene prices but an increase of around 13 percent for diesel. The comparison between the before and after prices in December 2013 is possible because the government’s action with respect to the determination of the retail price (and the associated discretionary tax) is a known fact. Simulating the effect of removing the discretionary tax under different circumstances is possible, but it is not possible to give a ‘before’ calculation since it is not known what the government would have decided to do with retail prices had it kept the discretionary tax. Impact of Fuel Subsidy Reform on Household Welfare, Government Budget, Poverty, and Inequality 20. Abandoning discretionary tax on super and diesel retail prices would imply a loss of DF 510.8 million (or 0.2 percent of GDP4) for the population. 21. For fuel bought directly at a pump, the impact of the reform on poor households is negligible, but it increases with welfare and represents the highest loss among rich households (DF 2,734 per capita), equivalent to 0.5 percent of household spending. The poorest two quintiles spend considerably less on public transport than the richer quintiles; this is partly due to the fact that the poor live in areas with no public transport available, such as the rural areas. However, the same conclusion holds when restricting the analysis only to urban areas. The impact of the reform on the poorest 40 percent is less than DF 80 per capita on public transport, compared to more than DF 400 among the richest 20 percent. In terms of household spending, this would amount to a loss of 0.3 percent of welfare for the poorest 20 percent and 0.8 percent for the richest quintile. The middle class would experience the largest reduction in well-being—about 0.12 percent. 22. The impact of the reform on government budget would result in a gain, the highest coming from fuel bought directly at the pump. The impact of the reform on government budget would result in a total gain of DF 408.6 million (or 0.16 percent of GDP). Sixty percent of that gain would come from fuel sold at the pump (96 percent of the gain from fuel will originate from the richest households) and the remaining 40 percent from public transport. It should be noted that since we assume a price elasticity of 0.2, the amount gained by the government is less than the loss incurred by the different households. 23. As the poor spend most of their income on food-related products, the elimination of tax exemptions on fuel products would reduce inequality but with no apparent impact on poverty. The 4 GDP for 2013 is estimated at US$1.456 billion. 4 elimination of tax exemptions on fuel would not affect the poorest because the consumption of this product is negligible among the poor. On the other hand, the consumption of this product is one of the highest among the subsidized products in rich households, and an elimination of tax exemption would result in a reduction in inequality by 0.12 percentage points. 24. Results of the PSIA show that an elimination of tax exemption on fuel at the pump offers potential for higher government revenues without impacting poverty. An increase of prices on public transport would increase poverty, but at a lower rate than increases on school transport. Impact of Introducing Consumer Tax on Basic Food Items 25. The government is not considering levying consumer tax on basic food items and the simulations below are merely for illustrative purposes. As mentioned above, among the basic food items that are tax exempt, only a certain quality/type is exempt (for example, broken rice). For rice, only 6 percent of the imported rice is exempt, but about 88 percent of flour, about 60 percent of sugar and edible oil, and about 50 percent of powdered milk products are exempt. The implicit subsidy represents 7 percent of the unsubsidized price. 26. Introducing consumer taxes would imply a loss of DF 558.7 million (or 0.22 percent of GDP) for the population. The per capita values indicate that the loss would be considerably higher for the richest in absolute terms. Overall, the impact of the reform on the poorest 20 percent would imply a decrease in well-being by DF 500 or 1.06 percent of household spending. For the richest quintile, the loss would be equivalent to DF 1,836 or 0.33 percent of household spending. This comparison shows that the poor spend more in relative terms on tax-exempt food products. Therefore, introducing consumer taxes would affect poverty. WHAT IS THE CURRENT ROLE OF SOCIAL SAFETY NETS IN DJIBOUTI? 27. A public safety net typically consists of publicly sponsored programs that provide income or in-kind support and access to basic social services to the poorest and vulnerable members of society. In addition, there are other types of transfers such as contributory social insurance like pensions, labor market programs, subsidies on food and fuel products, and private transfers (remittances) among households. To assess the effectiveness of these transfers to protect the poorest and vulnerable from shocks, a number of indicators are presented, for example, coverage, targeting accuracy, and generosity. 28. Untargeted tax exemptions (implicit subsidies) reach a wider part of the population than targeted programs. Tax exemptions on basic food items reach the majority of the poor (77.3 percent in the first quintile) and almost the majority of individuals in the other quintiles. Tax exemptions on certain fuel products, on the other hand, benefit only 17 percent of the poorest quintile but more than 82 percent of the richest. About a quarter of the population in the poorest quintile benefits from food rations, making it a program with relatively effective targeting. Compensation for health expenditure disproportionately benefits the richer quintiles. Very few households (less than 10 percent) benefit from pensions. Finally, 21 percent of Djibouti households receive private transfers (international or national) and these transfers mainly benefit the poorest households. 29. The intended beneficiaries of social safety net programs should be the poor. Therefore, the performance of such programs can be assessed by estimating program leakage. One way to measure such leakage is by determining the share of total transfers received by non-poor beneficiaries. In a well- targeted progressive program, the poor (bottom quintile) receive the highest share of transfers; this share declines as welfare increases. In Djibouti, food rations and cash transfers generally fit this description as the poor receive most of the transfers. Over half of the transfers for food rations are received by the poor (bottom quintile). In contrast, tax exemptions on food and fuel items predominantly benefit the urban population and non-poor, making these programs regressive. In fact, the majority of food and fuel subsidy resources (85 percent and 97 percent, respectively) are received by those living in urban areas and by those from the richest two quintiles (57 percent and 89 percent, respectively). However, only 15 percent of food subsidy benefits and less than 3 percent of fuel subsidy benefits go to 5 beneficiaries living in rural areas and only 10 percent and less than 1 percent, respectively, are received by those in the poorest quintile. Pensions and compensation for health care expenditure transfers are received mainly by non-poor beneficiaries and the population living in urban areas. 30. The generosity of social safety net programs in Djibouti is generally very low. The lower the generosity, the less important the value of the transfer is for the welfare of the beneficiaries. On the other hand, the higher the generosity, the higher its importance as a source of welfare for the beneficiaries. The generosity and size of transfers are, therefore, important design features of social safety net programs as they will have an impact on the poverty and other intended objectives of the programs. In fact, low generosity will limit the impact on poverty. Only two programs (pensions and private transfers from family and friends—which strictly speaking are not social assistance programs), out of the seven types of programs available, seem to have an impact on the consumption levels of the population in general. On the contrary, by focusing on the poorest quintile, food rations also have a significant effect even if private transfers are by far the most efficient vehicle. In fact, the impact of cash transfers from the government or NGOs and tax exemptions on food on the welfare of the poorest quintile is extremely modest and that of tax exemptions on fuel items is negligible. 31. In line with low targeting accuracy and low generosity, social safety net programs in Djibouti are generally small and inadequate in reducing the poverty gap. The ratio is above 1 for pension, which means that the average transfer is higher than the poverty gap and a strong indication that beneficiaries depend on this transfer as a source of income, which is usually expected for social insurance programs. On the other hand, the generosity of subsidies (both fuel and food) is negligible in closing the poverty gap; hence, a ratio of less than 0.1. WHAT IS THE IMPACT OF THE REFORMING TAX EXEMPTIONS AND SAFETY NETS ON POVERTY? 32. Discretionary energy taxes have benefitted the better-off in times of higher fuel prices (the analysis in this study is based on December 2013 prices). An elimination of tax exemptions on fuel products would reduce inequality but would not have any apparent impact on poverty. To reduce poverty, savings from a possible tax reform and other funding resources could be rechanneled toward the poor and vulnerable. To reduce poverty, however, effective targeting of the poor is key. The government with the support of the Bank is currently developing a social registry to increase equity in the distribution of resources, and promoting greater social inclusion for the most vulnerable groups. Over the course of the technical assistance provided to the government of Djibouti, a number of policy recommendations have emerged and some have already been taken into consideration in the design of a stronger social protection system. These recommendations are derived from the sections below and include: • Savings on energy tax reforms and other funding resources, including those spread over a number of very small safety net programs, should be channeled to a cash-transfer program targeting the poorest; • A Proxy Means Test (PMT) should be used to determine the households’ poverty score, and all safety net programs should target the poorest (as defined by the PMT) rather than targeting rural households based on geography;5 • Similarly, current and future safety net programs should first target poor households based on the relative poverty score, and then use other (categorical) factors to determine program eligibility. Impact of Tax Reform on Household Welfare and Government Revenue 33. As the poor spend most of their income on food-related products, the elimination of tax exemptions on such products would have the highest impact on destitution and inequality, while the elimination of tax exemptions on fuel products would reduce inequality but with no apparent impact on 5This functionality will be part of the forthcoming social registry which will be used to identify, classify, and target households that would be considered poor or vulnerable, to improve the delivery of assistance to them. 6 destitution. However these effects would be minimal, almost negligible. Globally, a reform of taxes on fuel and food products alone would not have a very significant impact on destitution and no impact on inequality. In particular, the destitution rate would increase by 0.17 percentage points from 20.00 to 20.17 percent. The elimination of tax exemptions on flour would increase destitution by 0.05 percentage points (from 20.00 to 20.05 percent), and inequality by 0.05 percentage points (from 45.13 to 45.18 percent). The effect of the elimination of the discretionary tax adjustment on fuel would not affect the poorest and would in fact result in a reduction of inequality by 0.12 percentage points. This is explained by the fact that the consumption of this product is negligible among the poor, while it is one of the highest consumed products among the subsidized products in rich households. 34. Among the poorer quintiles, the loss in welfare as a result of the reform would be the highest on food-related items; while it would be the highest on fuel products among the richer quintiles. In terms of food-related products, the reform would result in a significant loss of welfare among the poorest quintile (1.12 percent of total spending) but this loss decreases as welfare increases. On the other hand, the reform would result in a minimal loss among the top quintile for fuel products, and this loss decreases as welfare decreases and becomes negligible for the first quintile. 35. The impact of the reform on government budget would result in a gain, the highest coming from fuel. The impact of the reform on government budget would result in a total gain of DF 856 million (or 0.33 percent of GDP): 28 percent of that gain would come from fuel (96 percent of the gain from fuel will originate from the richest households), 18 percent from flour, and 15 percent from sugar. The highest gain in the government budget will come from the richest households (54 percent). This decreases as welfare decreases to reach the lowest share among poor households (5 percent). This is consistent with the previous finding that the highest loss of welfare in the population would come from fuel, and particularly among the rich. The most important revenue gain to the government would come from increasing the price of fuel, while the least would come from increasing the price of cooking oil. Likely Impact of Compensation Policies through Social Safety Nets Programs 36. Any reform necessitates an efficient system for targeting the poor and vulnerable. The government, with support from development partners, is currently in the process of reforming the social protection system. A key element to strengthening social safety nets in Djibouti is the creation of a social registry of poor and vulnerable households, which will be used to target the poor and serve as a single platform used by all social assistance programs, resulting in significant cost savings and substantial improvements in targeting the poorest households. 37. The preceding analysis has shown that tax exemptions on food and fuel items are regressive, and that the poor benefit the least. The government is considering strengthening the social protection system through a targeted cash-transfer system. Based on discussions with the SESN, reform options based on a number of transfer schemes and budget envelopes are presented. The study shows the results of 54 different simulations on the welfare effects of various reform options on poverty headcount and poverty gap. The reform options consist of transfer schemes implemented depending on the location and/or quintile targeted; whether it is done at the individual or household level; and according to the total budget to be transferred: either 1, 2, or 3 billion Djibouti francs. • Destitution headcount. The largest decline in poverty headcount is achieved when targeting the first quintile. Most of the schemes that target the first quintile achieve a significant reduction in poverty headcount, especially with a budget of DF 2 or 3 billion. In fact, with a budget of DF 3 billion it would be possible to more than halve the poverty headcount using any of the schemes that target the first quintile. A program targeting only the rural areas would be less efficient. • Destitution gap. The optimal scheme in terms of reducing the poverty gap would be the ‘first quintile with a 4-step transfer’ in which the amount transferred depends on the standard of living of households in the first quintile. In this way, the poorest 5 percent would receive the most, then the following 5 percent would receive a bit less, and so on. A unique simple transfer targeting the first quintile might be simpler to implement and would also provide satisfactory results. 7 • Coverage. In a scheme focusing only on the population living in rural areas, only 62.4 percent of individuals in the first quintile and quite a few non-poor households would receive a transfer (11.8 percent of individuals of the second quintile and 4.3 percent in the third quintile). Any such leakage necessarily makes rural-based criteria less efficient at reducing poverty than one focusing solely on the first quintile. To reduce the poverty gap, targeting the first quintile is more efficient than any of the schemes focusing on rural households. There should be an exclusive transfer scheme focusing on the first quintile. 8 COUNTRY AND REFORM CONTEXT 1. Djibouti faces multiple development challenges, and recent economic growth has not led to improvements in the welfare of the people. The increased growth of the past decade has relied in part on one-time events such as an economic resurgence after the political turmoil of the 1990s, the establishment of foreign military bases, and significant inflows of foreign investment that financed the construction of the new port and hotel infrastructure. Yet this growth pattern has not alleviated high levels of poverty or unemployment. Although poverty figures are difficult to pinpoint due to data limitations, available evidence indicates that poverty is widespread and worsening in the face of the current drought—the worst in 60 years. In terms of human development, Djibouti is ranked 170th out of 187 countries (United Nations Development Programme 2014). 2. Djibouti depends heavily on imports to meet its food needs, and a large fraction of the population faces food insecurity. Practically all food items are imported and increases in international food prices directly affect Djibouti’s poor people, who spend up to three-quarters of their income on food. Due to severe and prolonged droughts, at least 20 percent of the capital’s population and three-quarters of rural households are vulnerable to severe and moderate food insecurity (World Food Program 2013). On average, 17.8 percent of Djiboutians are affected by acute malnutrition, which varies between 14 and 26 percent across regions, and about 5.7 percent by severe acute malnutrition. On average, 29.7 percent suffer from chronic malnutrition, exceeding 40 percent in certain regions. About one in four children are underweight (SMART 2013). 3. In this context, social safety nets become crucially important to alleviate the devastating effects of poverty, and the government of Djibouti has made considerable progress in building a social safety net. Social safety nets (noncontributory transfer programs targeted to the poor or vulnerable) have an immediate impact on extreme poverty and inequality. They enable households to manage risks and make better investments in their future, often helping to prevent malnutrition and underinvestment in education. Djibouti is only at the beginning of institutionalizing an efficient social safety net. To date, the scale and funding of existing social safety net programs remains inadequate to protect most poor and vulnerable groups. Spending is limited, programs are fragmented and largely uncoordinated, and there is no (non-emergency) large-scale social safety net at the national level. Increasing coverage of social safety net programs is thus paramount so that the country can eventually benefit from reforming the policies to reduce the large expenditures going to untargeted subsidies at present. 4. In response to the food crisis and to shield the population from price shocks on essential food and certain energy products, universal tax exemptions were introduced; yet, since these are untargeted, these exemptions benefit mostly the middle and higher income class. According to IMF estimates, Djibouti foregoes 0.5 percent of GDP (2009) on certain food items (rice, edible oil, sugar, flour, and powdered milk) and an estimated 2 percent of GDP (2011) on certain energy products (IMF 2012). Such untargeted tax breaks on fuel benefit mainly those who are better off. According to IMF estimates, there is significant leakage of diesel and gasoline price reductions, and households in the highest income quintile receive 12 times more of these benefits than the two poorest quintiles combined. Gasoline, diesel, and kerosene are subject to a mix of ad valorem (value added tax [VAT]) and specific taxes (royalties, excise). A discretionary component is added, allowing the government to target monthly prices. Between 2010 and 2013, these goods were all taxed but net taxes have fallen. DATA SOURCES AND TARGETING 5. The study is based on data from a representative household survey (2012) which includes detailed information on household expenditures as well as on receipt of certain cash and in-kind benefits. The third Enquête Djiboutienne auprès des ménages (EDAM 3) was conducted in 2012 by the DISED, the national statistical agency. EDAM 3 has a nationally representative sample of the sedentary population composed of 5,880 households with 31,686 individuals. The EDAM 3 9 questionnaire covers many aspects: demography, education, employment, mortality, governance, housing, access to basic social services, durable goods ownership, and finally, expenditures and revenues. Of particular importance for this study is information on household expenditure on tax- exempt food (flour, rice, oil, sugar, and milk); certain fuel items (kerosene, butane, and fuel expenditure on transport); and electricity as well as information on cash and in-kind benefits. The EDAM 3 dataset has been used to compute total expenditure aggregates of households based on which the DISED has produced their recent poverty profile, yielding 40.8 percent of poverty and 23 percent of extreme poverty. 6. As is common for household surveys, the EDAM 3 data is representative of the sedentary population only. The EDAM 3 sample covers the sedentary population, leaving out the nomad and homeless populations (population flotante) as well as individuals living in collective households (hotels, prison, military camps, orphanages). According to the most recent census conducted in 2009, Djibouti’s total population was 818,159 individuals, of which 161,132 were nomads and 149,022 either lived in collective households or were homeless. Having household surveys covering solely sedentary population is standard practice since surveying nomad and homeless populations creates important conceptual and logistic issues. 7. Access to a representative sample of the nomad population extracted from the 2009 Census allowed supplementing certain analysis. The census-based nomad sample was used to construct the proxy means test (PMT) index. However, as the census dataset does not have any information on specific spending on tax-exempt items or on access to benefits, most of the analysis had to rely only on the sedentary population. Finally, for the homeless population the only information that could be provided by the DISED was a simple count according to gender and age. 8. A series of corrective factors were applied to both datasets to perform analysis on the actual 2014 population. First, the EDAM 3 sample weight has been corrected for a slight under- coverage. Second, the population growth rate (2.8 percent per year) was applied to both the 2009 Census and 2012/13 EDAM 3 data up to 2014. Therefore, the analysis is based on the following population size (excluding collective households): • Sedentary: 536,851 • Nomad: 184,990 • Homeless: 139,514 9. To assess the effectiveness of the current safety nets (section IV) and the effectiveness of the tax exemptions (section V), five quintiles based on per capita expenditure have been constructed.6 Sections IV and V are based on the EDAM 3 data since this is the only database with information on the current safety nets and expenditure on tax-exempt food and petroleum items. Based on the per capita expenditure welfare index already computed by the DISED and the World Bank and the ranking of per capita expenditure, five quintiles have been constructed. The first quintile includes the poorest 20 percent of the sedentary Djiboutian population, the second quintile includes the next 20 percent, and so on up to the top quintile with the richest 20 percent of the population. Translated into Djibouti francs, the ranges of per capita expenditure for each quintile are the following: • Quintile 1 : DF 277,231 6 Following a complete revision of the expenditure-based household welfare index undertaken recently by the World Bank Djibouti Country Team (in full collaboration with the DISED), our report was also revised to take into account those changes. Although the revisions were important, the actual ranking of the different households from the poorest to the richest was not modified enough to change our conclusions. In fact, the revised figures found in the latest version of this report are barely the same as the previous version. 10 10. Simulations of the welfare effects of various reform options (section VI) use the EDAM 3 data as well as data from the 2009 Census. For these simulations, quintiles are based on the PMT formula that has been constructed for the social registry, which will be implemented by the SESN. 11. For the purpose of this study, the poverty line is defined as the upper limit of the first quintile. Using the PMT formula and the sedentary sample only, the poverty line is DF 77,926 per capita. 12. Different targeting approaches are possible (self-targeting, geographical targeting) but after discussion with the Djiboutian authorities and based on experience in the region, it was agreed that the PMT approach would be used to target the poorest households. The PMT approach calls for the construction of a synthetic measure of household welfare based on a series of easily observable indicators (at least easier to observe than expenditure or income) but strongly correlated with expenditure or income. Those easily observable indicators could be dwelling characteristics (water source, access to electricity, wall, roof and floor material, type of sanitation); demographic composition of the household; socioeconomic characteristics (employment, education, age, and gender) of the household head as well as ownership of durable goods or animals. The relationship between the actual per capita expenditures (in log) and those correlates is supposed to be linear and is estimated by the ordinary least square (OLS) technique. From the estimated coefficients, it is straightforward to compute expected per capita expenditure on which our analysis would be performed. Technically, we need to estimate equation (1) on the 5,880 households from the EDAM 3. = + + (1) where represents the actual per capita expenditure for household i (in log); represents the j independent variables for household i; is the coefficient to be estimated for the j variables ; is the constant; and is the error term. Estimated by OLS techniques, equation (1) would generate the constant and coefficients and therefore , the estimated PMT measure of welfare. That can be summarized by the following equation: = = exp + (2) In the analysis, a given household would be considered poor if its measure of welfare is lower than a predetermined poverty line and non-poor otherwise. In this study, we consider the upper limit of the first quintile as the poverty line (DF 77,962 per capita). Further details on the PMT approach as well as a series of results on its efficiency can be found in annex 2. 11 EFFECTIVENESS OF THE CURRENT SOCIAL SAFETY NET WHO ARE THE POOR? 13. This section shows characteristics of the sedentary population by quintile to understand how well programs are targeted to the poor and inform possible reform options. Table 1 shows that the data has been divided into five consumption quintiles and that the first quintile is fixed as ‘poor’ in the analysis. The rural population dominates the poorest quintile (67.9 percent of the rural population are classified as ‘poor’ according to this definition) while only 2.2 percent of the rural households belong to the richest quintile. Over half of the total expenditures of all households (50.4 percent) accrues to the richest quintile. In comparison, the poorest quintile consumes less than one-twentieth (4.3 percent) of the total even if it constitutes 20 percent of the population. Similarly, rural areas represent 16.1 percent of the sedentary population but only 5.6 percent of the total expenditure and 54.8 percent of the poor individuals. Table 1. Population Demographics Quintiles of Per Capita Expenditure Total Q1 Q2 Q3 Q4 Q5 Share of total 100.0 20.0 20.0 20.0 20.0 20.0 population Share of poor 100.0 100 0.0 0.0 0.0 0.0 population Share of urban 100.0 10.8 20.2 22.4 23.2 23.4 population Share of rural 100.0 67.9 18.7 7.7 3.6 2.2 population Share of total 100.0 4.3 9.8 14.4 21.2 50.4 expenditures Area of Residence Region Urban Rural Djibouti Ali Sabieh Dikhil Tadjourah Obock Arta Share of total 83.9 16.1 73.1 5.6 7.1 7.4 2.8 4.1 population Share of poor 45.2 54.8 31.8 13.7 19.4 18.7 8.2 8.2 population Share of urban 100.0 0.0 87.2 4.4 3.0 2.7 1.1 1.6 population Share of rural 0.0 100.0 0.0 11.9 28.3 31.7 11.2 16.9 population Share of total 94.4 5.6 86.5 2.9 3.2 3.9 1.2 2.3 expenditures Source: World Bank calculation based on the EDAM 3. 14. Welfare is not uniformly spread across groups. Some strata of the Djiboutian population are more likely to be in the poorest quintile than others (see Table in annex 3). Important correlates of welfare are the following: • Location. This is the most important correlate of poverty. There is a clear regional pattern: richer populations are more likely to live in urban areas and poorer in rural areas (see figure 1). About 90 percent of individuals in the richest three quintiles live in urban areas compared with 40 percent in the poorest quintile. About 73 percent of the population lives in urban Djibouti city and the rest distributed among the five regions Ali Sabieh, Dikhil, Tadjourah, Obock, and Arta (see Table 1). Among the poor (first quintile), only 32 percent live in urban Djibouti city. • Levels of literacy and education. Overall, poverty is associated with lower levels of literacy of household heads (12.7 percent in the poorest quintile compared with 67.8 percent in the richest quintile) and also lower levels of education (in the poorest quintile, 89.0 percent have 12 no education at all, 7.4 percent have completed primary education, and less one percent have completed high school or tertiary education, see figure 2). • In the case of household heads, not working or working in menial jobs. This is associated with greater poverty. Poorer quintiles have lesser attachment to the labor market (32.9 percent in the poorest quintile hold a job compared with 71.1 percent in the richest quintile) and lower status jobs (only 7.5 percent in the poorest quintile work in the public sector compared with almost 40 percent in the richest quintile; see figure 3). For the total population aged between 15 and 65 years, employment rates are even lower (15.9 in the first quintile against 41.8 for the richest quintile). Employment rates are particularly low in rural areas (17.6 percent against 30.0 in urban areas). Conversely, the proportion of people looking for a job (at least ready to take up one if available) is negatively correlated with the employment rate • Household size. Poorer households tend to be larger and with more young members. The average household size among the poorest three quintiles is 6.1 persons; it is 5.6 among the fourth quintile and 4.5 among the richest quintile. The average age in the poorest quintile is 22.1 compared with 27.5 in the richest. • Vulnerable populations, such as households with disability, old-age and orphans, or female- headed households. These are not associated with a higher risk of living in poor households. Table 2 shows the proportion of those aged 60 years and older, 70 years and older, disabled and orphans, and households with female heads in each quintile. There is no apparent relationship between vulnerability and poverty. Table 2. Percentage of Vulnerable Populations by Quintile Female- headed Age 60+ Age 70+ Handicapped Orphans Households Quintile Poorest 4.6 1.8 0.0 0.5 24,9 Second 3.7 1.4 0.1 0.9 20,1 Third 3.8 1.4 0.1 0.9 20,0 Fourth 4.2 1.4 0.1 1.0 23,2 Richest 4.6 1.4 0.1 0.9 21,8 Source: World Bank calculation based on the EDAM 3. Note: Full table and source/notes are available in annex 5. Figure 1. Percentage of Urban Population by Quintiles (Percentage) 120 95.1 97.4 100 89.8 78.1 80 60 39.7 40 20 0 Poorest Second Third Fourth Richest Source: World Bank data. 13 Figure 2. Level of Education of Household Heads by Quintiles (Percentage) 70.0 60.0 50.0 College 40.0 High School 30.0 Middle Scool 20.0 Primary 10.0 0.0 Poorest Second Third Fourth Richest Source: World Bank data. Figure 3. Type of Job of Household Heads by Quintile (Percentage) 80.0 70.0 60.0 50.0 Menial job 40.0 Self-employed 30.0 Wage, private 20.0 Wage, public 10.0 0.0 Poorest Second Third Fourth Richest Source: World Bank data. FEMALE-HEADED HOUSEHOLDS 15. Female-headed households are not subject to higher poverty risk despite their disadvantageous socioeconomic characteristics. The distribution of female-headed households is rather uniform across the welfare distribution. Nevertheless, female-headed households are at a disadvantage in terms of socioeconomic characteristics. Figure 4 shows that almost 90 percent of female heads of households have no educational attainment, in contrast to 55 percent of their male counterparts. About 30 percent of female heads of households work mainly as day laborers or are self- employed (see figure 5). Overall, this points to a more extensive support system for these female- headed households. Interestingly, among the poorest quintile, the heads of about 40 percent of both female- and male-headed households are looking for work. 14 Figure 4. Level of Education by Gender of Household Head (Percentage) 50.0 40.0 Percentages 30.0 College 20.0 High school 10.0 Middle school 0.0 Primary Male Female Gender Source: World Bank data. Figure 5. Type of Job by Gender of Household Head (Percentage) 80.0 70.0 60.0 Percentage 50.0 Menial job 40.0 30.0 Self-employed 20.0 Wage, private 10.0 Wage, public 0.0 Male Female Gender Source: World Bank data. Figure 6. Proportion of Household Heads Working and Looking For Work (Percentage) Female headed households Male-headed households 60.0 80.0 50.0 70.0 15.2 29.2 60.0 21.3 40.0 28.8 50.0 28.5 36.2 34.7 37.7 30.0 34.3 40.0 34.5 58.5 20.0 30.0 40.3 44.8 26.0 20.0 35.6 10.0 16.0 25.0 11.2 12.9 10.0 7.2 0.0 0.0 Poorest Second Third Fourth Richest Poorest Second Third Fourth Richest Looking for work Working Looking for work Working Source: World Bank data. 15 HUMAN CAPITAL AND POVERTY 16. Education supports resilience to inter-generational transmission of poverty but poorer children are less likely to be enrolled in school. School attendance for all age groups (6 to 13, 14 to 18, and 19 to 25) goes up with welfare quintile (see Figure 7). Among the richer quintiles, about 80 percent of 6 to 18-year-olds are in school compared with less than 60 percent among the poorest 6 to 13-year-olds and about 45 percent among the poorest 14 to 18-year-olds. School attendance is also correlated with location: more than 75 percent of 14 to 18-year-olds in urban areas are enrolled compared with only 28 percent in rural areas (see Table in annex 3). The gender gap in school attendance increases with age (see Figure 8). There is a very small gap for the 6–13 age group but the gap increases for older children. However, the gender gap is much smaller than in many African countries, particularly Muslim ones. Analphabetism is widespread—about half the population is illiterate—but it is also strongly correlated with welfare. Only 23 in the lowest quintile can read and write compared with 71 percent in the richest quintile (see Table in annex 3). Figure 7. School Attendance by Age Group and Quintiles 90.0 83.1 85.2 81.6 80.0 74.3 70.0 60.0 56.0 50.0 6 to 13 years old 40.0 14 to 18 years old 30.0 19 to 25 years old 20.0 10.0 0.0 Poorest Second Third Fourth Richest Source: World Bank data. Figure 8. School Attendance by Age Group and Gender 100.0 76.5 72.6 74.8 80.0 65.3 Percentage 60.0 40.0 33.9 24.5 20.0 0.0 6 to 13 years old 14 to 18 years old 19 to 25 years old Age group Male Female Source: World Bank data. 16 CAPACITY TO RESPOND TO SHOCKS, INCLUDING SAVINGS AND STRATEGIES FOR COPING 17. Poor and rural households are especially vulnerable to shocks, in particular, food price increases. Over the past 12 months, households in the lowest quintile were affected by an average of 1.55 shocks and households in the highest quintile by an average of 0.67 shocks (see Table 3). Rural households are faced with more than twice the number of shocks than urban households (1.79 compared with 0.70), with households in Obock reporting more than three average shocks. Among the households that faced a shock, 40 percent were affected by high food prices while 8 percent cited the loss of cattle due to drought and 7 percent cited fuel and transport price rises (see annex 6). Again, these numbers differ across welfare quintiles. For example, 56 percent of households in the first quintile faced the shock of high food prices compared with 31 percent in the richest quintile. Loss of cattle mainly affects poor households in rural areas, with 29 percent citing loss of cattle due to drought and another 14 percent citing loss of cattle unrelated to drought, compared with less than 2 percent among the richest quintile. High cost of fuel/transportation is cited by eight percent of the population across all quintiles. Table 3. Shocks Faced by Households and Capacity to Respond (% of Households) Average Shock Led to Shock Could Not Do Able to # of Decrease in Diminished Anything to Recover from Shocks Revenue/Loss of Capacity to Compensate the Shock Goods Get Enough Effect of the Food Different Shocks Poorest 1.55 80.8 83.6 59.1 26.5 Second 0.88 81.2 84.4 52.3 27.8 Third 0.71 71.1 80.1 43.5 34.2 Fourth 0.71 70.6 77.6 39.1 38.1 Richest 0.67 63.7 64.9 36.6 51.6 Djibouti 0.66 73.8 80.0 39.2 36.3 Ali Sabieh 0.65 86.4 85.8 66.3 57.9 Dikhil 1.60 54.9 59.7 57.5 25.3 Tadjourah 1.73 85.8 86.1 68.6 29.4 Obock 3.30 77.3 80.3 40.3 21.9 Arta 0.70 77.2 79.3 54.7 34.9 Urban 0.70 73.1 78.3 42.2 38.1 Rural 1.79 77.5 80.5 60.3 25.5 Male 0.88 72.9 77.8 48.5 35.0 Female 0.98 79.4 82.6 44.0 32.6 Total 0.91 74.4 78.9 47.5 34.4 Source: World Bank calculation based on the EDAM 3. Note: With the exception of column 2, averages are across households that faced at least one shock over the past 12 months. 18. Shocks affect the ability of households, both poor and non-poor, to have access to enough food. Following a shock, about 81 percent of households in the poorest quintile and 64 percent of households in the richest quintile experienced a decrease in revenue or a loss of goods, including animals (see Table 3). Among the households affected by shocks, 83 percent of the poorest households and 65 percent of the richest households report not having enough food. There is no major difference between rural and urban households (80 and 78 percent, respectively), while the percentage of female-headed households (83 percent) without sufficient food following a shock is slightly higher than male-headed households (78 percent). 17 19. Low resilience is associated with poverty and rural areas. Households in lower quintiles and rural areas have less capacity to recover from shocks. Among the poorest households, 59 percent report not being able to do anything to compensate the effect of the shock, whereas only 36 percent of the richest households were in this situation. Similarly, vulnerability is higher in rural than urban areas (60 and 42 percent, respectively). Conversely, the most resilient households tend to be the richest and urban households. More than half of the households in the fifth quintile could recover from losses caused by a shock, while only 26 percent of the poorest households were able to recover. Similarly, urban households (38 percent) report more resilience than rural ones (25 percent). 20. There does not seem to be a clear pattern on the resilience of female- and male-headed households. On average, female-headed households are slightly more vulnerable to shocks (0.98 average shocks) than male headed-households (0.88 average shocks), in particular due the death of a family member or the loss of cattle. As a consequence of a shock, a larger proportion of female- headed households (83 percent) than male-headed households (78 percent) report not having enough food. Male-headed households are less able to compensate for shocks compared with female-headed households: 48 percent of male-headed households compared with 44 percent of female-headed households report not being able to do anything to compensate the effect of a shock. For example, female-headed households report being more likely to compensate for the loss of a family member than male-headed households (difference of thirteen percentage points). Conversely, male-headed households seem slightly better able to recover from a shock than female-headed households, in particular after the loss of a family member. Again, this points to the existence of a ‘family-based safety net’ for female-headed households. EFFECTIVENESS OF THE SOCIAL SAFETY NET 21. The public safety net typically consists of publicly sponsored programs that provide income or in-kind support and access to basic social services to the poorest and vulnerable members of society. In addition, there are other types of transfers such as contributory social insurance like pensions, labor market programs, subsidies on food and fuel products, as well as private transfers (remittances) among households. To assess the effectiveness of these transfers to protect the poorest and vulnerable from shocks, a number of indicators are presented, for example, coverage, targeting accuracy, and generosity. 22. Untargeted tax exemptions (implicit subsidies) reach a wider part of the population than targeted programs. Table 4 shows the percentage of the population (by area, region, and welfare quintile) receiving seven types of transfers: pensions (private or public), compensation for health care expenditure, food rations, cash transfers from the government or NGOs, publicly provided food and fuel subsidies, and private transfers received from family and friends. Tax exemptions on basic food items reach the majority of poor (77.3 percent in the first quintile) and almost the majority of individuals in the other quintiles. Tax exemptions on certain fuel products, on the other hand, benefit only17 percent of the poorest quintile but more than 82 percent of the richest. About a quarter of the population in the poorest quintile benefits from food rations, making it a program with a relatively effective targeting, partly because the program mainly targets rural areas. Compensation for health expenditure disproportionately benefits the richer quintiles. Very few households (less than 10 percent) benefit from pensions. Finally, 21 percent of Djiboutian households receive private transfers (international or national) and these transfers mainly benefit the poorest households. 18 Table 4. Coverage of Transfer Programs Transfers All Compensati Food from Food Fuel transfers Pensio on for Remittanc Ratio Governme Subsidi Subsidi ns Health Care es ns nt or es es Expenditure NGOs 97,4 Total 8.5 4.3 8.1 2.1 94.7 58.2 21.0 Area of residence Urban 8.9 4.8 1.7 1.0 97.6 67.3 18.9 98.4 Rural 6.4 1.6 41.2 7.5 79.8 10.9 32.5 92.5 Region Djibo 98.3 7.9 4.8 1.2 0.9 97.4 74.1 18.5 uti Ali 98.5 Sabie 22.4 3.7 22.5 3.8 96.1 7.0 31.8 h Dikhil 5.1 1.9 28.3 4.3 93.9 8.9 25.1 97.1 Tadjo 91.7 11.8 2.0 24.5 5.6 75.6 11.7 37.6 urah Obock 6.1 3.3 36.1 12.3 89.2 3.6 24.4 98.4 Arta 1.3 3.8 28.7 3.1 83.9 50.4 13.5 90.9 Quintiles of per capita consumption Q1 5.3 1.3 27.0 5.8 77.3 17.1 29.7 94.2 Q2 8.6 3.5 8.6 1.3 98.1 48.2 23.6 96.9 Q3 10.5 4.2 2.5 1.5 99.7 66.9 20.2 98.1 Q4 8.5 5.5 1.3 0.9 99.3 76.2 18.3 98.8 Q5 9.6 6.9 1.1 0.8 99.2 82.5 13.5 99.1 Source: World Bank calculation based on the EDAM 3. 23. Looking at the transfers in terms of benefits incidence shows that untargeted tax exemptions have high benefit leakage and are regressive compared to some targeted transfers which appear generally more progressive. Table 5 shows the distribution of benefits by area of residence, and by welfare quintile) of the same seven types of transfers previously mentioned. The intended beneficiaries of social safety net programs should be the poor, so the performance of such programs can be assessed by estimating program leakage. One way to measure such leakage is by determining the share of total transfers received by non-poor beneficiaries. In a well-targeted progressive program, the poor (bottom quintile) receive the highest share of transfers; this share declines as welfare increases.7 In Djibouti, food rations and cash transfers generally fit this description as the rural population and the poor receive most of the transfers. Over half of the transfers for food rations are received by the poor (bottom quintile); almost 80 percent are in the bottom two quintiles and live in rural areas (84.2 percent of benefits received), which is consistent with the fact that such transfers are targeted to rural areas. In contrast, tax exemptions on food and fuel items predominantly benefit the urban population and non-poor, making these programs regressive. In fact, the majority of food and fuel subsidy resources (85 percent and 97 percent, respectively) are received by those living in urban areas and by those from the richest two quintiles (57 percent and 89 percent, respectively). However, only 15 percent of food subsidy benefits and less than 3 percent of fuel subsidy benefits go to beneficiaries living in rural areas and only 10 percent and less than 1 percent, respectively, are received by those in the poorest quintile. Pensions and compensation for health care expenditure transfers are received mainly by the population living in urban areas and non-poor beneficiaries. 7 World Bank. 2012. Inclusion, Dignity, and Resilience: The Way Forward for Social Safety Net Reform in the Middle East and North Africa Region. Human Development Network. Washington, DC: World Bank. 19 Table 5. Distribution of Benefits (Targeting Accuracy) Area of Residence Quintiles of Per Capita Consumption Urban Rural Q1 Q2 Q3 Q4 Q5 Pensions 85.6 14.4 4.5 10.9 17.5 19.3 47.9 Compensation for health 88.0 12.0 3.5 11.1 17.2 23.4 44.9 care expenditure Food rations 15.8 84.2 56.2 21.2 10.7 4.3 7.5 Transfers from 44.7 55.3 45.1 16.9 13.6 11.7 12.7 government or NGOs Food subsidies 84.9 15.1 9.5 15.4 17.8 22.0 35.3 Fuel subsidies 97.2 2.8 0.6 3.4 7.5 13.5 75.0 Remittances 75.6 24.4 20.8 15.4 20.7 15.0 28.1 Source: World Bank calculation based on the EDAM 3. Note: Benefits' incidence is the transfer amount received by the group as a percent of total transfers received by the population. Specifically, benefits' incidence is (Sum of all transfers received by all individuals in the group)/(Sum of all transfers received by all individuals in the population). Aggregated transfer amounts are estimated using household size- weighted expansion factors. 24. The majority of the poor are not covered by the social safety nets programs in Djibouti, leading to high exclusion (under-coverage or proportion of the poor not benefitting from the program) and inclusion errors (leakage or proportion of beneficiaries who are non-poor) and low targeting accuracy (see Table 6). ‘Targeting differential’ (last column in Table 6) measures the difference between the coverage rate of the poor and beneficiary leakage (participation rate for non- poor). Therefore, a negative number would mean that there are more non-poor than poor beneficiaries covered by the program. Coverage of the poor for pension (5.3 percent), compensation for health care expenditure (1.3 percent), cash transfers from the government or NGOs (5.8 percent), and fuel subsidies (17.1 percent) are very low, and therefore under covered. Leakage (both in terms of beneficiaries and benefits) and targeting differential are very high. On the other hand, coverage of the poor for food-related transfers (food ration and subsidies) is relatively higher. Table 6. Under-coverage and Leakage - Total Poor Leakage (% Coverage Under- Leakage Targeting of of the coverage (benefits) differential beneficiaries) poor (1) (2) (4) (5) = (1) - (3) (3) Direct and indirect beneficiaries Pensions 5.3 94.7 87.5 95.5 -82.1 Compensation for health care 1.3 98.7 93.8 96.5 -92.4 expenditure Food rations 27.0 73.1 33.4 43.8 -6.5 Transfers from government or 5.8 94.2 44.0 54.9 -38.3 NGOs Food subsidies 77.3 22.6 83.7 90.4 -6.3 Fuel subsidies 17.1 82.8 94.1 99.4 -76.9 Remittances 29.7 70.2 71.7 79.2 -42.0 Source: World Bank calculation based on the EDAM 3. Note: Under-coverage refers to the percentage of poor individuals that do not receive transfer. Leakage refers to the percentage of individuals that receive transfers and are not poor. Sample of all households. Under-coverage and leakage are calculated across this sample, setting as expansion factor the household expansion factor multiplied by the household size. The targeting differential is the difference between the coverage rate and the participation rate for non-poor. 25. The generosity of social safety nets programs in Djibouti is generally very low. Tables 7 and 8 illustrate the generosity of the transfers (size or value of the transfer) received by a given group relative to the welfare of all households in that group (relative incidence) or relative to the total welfare of beneficiaries in that group (generosity). The lower the generosity, the less important is the value of the transfer on the welfare of the beneficiaries. On the other hand, the higher the generosity, 20 the higher its importance as a source of welfare to the beneficiaries. The generosity and size of transfers are, therefore, important design features of social safety programs as they will have an impact on the poverty and other intended objectives of the programs. In fact, low generosity will limit the poverty impact. Only two programs (pensions and private transfers from family and friends – which strictly speaking are not social assistance programs), out of the seven types of programs available, seem to have an impact on the consumption level of the population in general. On the contrary, by focusing on the poorest quintile food rations also have a significant effect even if private transfers are by far the most efficient vehicle. In fact, the impact on the welfare of the poorest quintile from cash transfers from the government or NGOs and tax exemptions on food is extremely modest and that of tax exemptions on fuel items is negligible. Table 7. Relative Incidence - All Households Area of Residence Quintiles of Per Capita Consumption Total Urban Rural Q1 Q2 Q3 Q4 Q5 Pensions 2.4 2.1 5.8 3.4 3.6 3.9 2.9 3.1 Compensation for 0.1 0.1 0.3 0.2 0.2 0.2 0.2 0.2 health care expenditure Food rations 0.2 0.0 3.5 4.4 0.7 0.3 0.1 0.1 Transfers from 0.1 0.0 0.6 1.0 0.2 0.1 0.1 0.0 government or NGOs Food subsidies 0.4 0.3 0.9 1.1 0.8 0.6 0.5 0.3 Fuel subsidies 0.3 0.3 0.2 0.1 0.2 0.2 0.3 0.6 Remittances 2.7 2.2 11.3 18.1 5.9 5.3 2.6 2.1 Source: World Bank calculation based on the EDAM 3. Note: Relative incidence is the transfer amount received by a group as a share of the total welfare aggregate of the group. Relative incidence is calculated by setting as expansion factor the household expansion factor multiplied by the household size. Incidence expressed in local currency units (LCU). Table 8. Generosity - Direct and Indirect Beneficiaries Area of Residence Quintiles of Per Capita Consumption Total Urban Rural Q1 Q2 Q3 Q4 Q5 Pensions 26.5 23.9 78.2 53.7 40.5 37.6 35.0 31.9 Compensation for health 2.7 2.5 9.3 12.4 5.9 5.5 3.8 2.9 care expenditure Food rations 8.7 4.3 10.7 20.0 8.6 10.1 5.5 5.7 Transfers from 6.4 4.7 8.8 19.7 12.5 5.7 5.6 3.6 government or NGOs Food subsidies 0.4 0.3 1.0 1.2 0.8 0.6 0.5 0.3 Fuel subsidies 0.4 0.4 0.8 0.3 0.3 0.3 0.4 0.8 Remittances 17.1 14.7 33.9 61.6 25.3 26.5 14.2 19.2 Source: World Bank calculation based on the EDAM 3. Note: Generosity is the mean value of the share transfer amount received by all beneficiaries in a group as a share of the total welfare aggregate of the beneficiaries in that group. Generosity is calculated by setting as expansion factor the household expansion factor multiplied by the household size. Generosity expressed in LCU. 26. Social safety nets in Djibouti appear to have limited impact on poverty and inequality, which could be explained by the combination of low targeting accuracy and generosity. Table 9 illustrates the impact of different transfers on poverty, using the headcount index or poverty rate (FGT0) and poverty gap index (FGT1) measures as well as the impact on inequality using the Gini coefficient. The poverty rate is the share of population that is counted as poor; the poverty gap index (expressed as a percentage of the poverty line) sums up the extent to which, on average, individuals fall below the poverty line; and the Gini coefficient measures income distribution and ranges from 0 (perfect equality) to 1 (perfect inequality). For example, without the food ration program, poverty rate would increase by 0.4 percentage points, the poverty gap would increase by 0.58 percentage points and the Gini coefficient would increase from 46.5 to 46.7. The two programs with the highest impact 21 on poverty are pensions and private transfers as the poverty rate would increase by 1.5 and 2.1 percentage points, respectively, in the absence of this program. Table 9. Impact of Programs on Poverty Measures – Simulating the Absence of the Program Difference Difference Change FGT0 FGT1 Gini Percentage Percentage (%) (%) Points Points Indicator 20.00 8.58 46.5 Indicator without listed transfer Pensions 21.50% 1.50 9.71% 1.13 47.2 0.7 Compensation for health care 20.08% 8.61% 46.5 0.0 expenditure 0.08 0.02 Food rations 20.44% 0.44 9.16% 0.58 46.7 0.3 Transfers from government or 20.09% 8.71% 46.6 0.1 NGOs 0.09 0.13 Food subsidies 20.14% 0.14 8.71% 0.13 46.6 0.1 Fuel subsidies 20.00% 0.00 8.59% 0.01 46.4 -0.1 Remittances 22.10% 2.10 10.41% 1.83 47.5 1.0 Source: World Bank calculation based on the EDAM 3. 27. In Djibouti, the program with the highest cost-benefit ratio is food rations followed by cash transfers from the government or NGOs (targeted transfers). In contrast, the one with the lowest cost-benefit ratio is tax exemptions on fuel products closely followed by compensation for health care expenditure (untargeted transfers). Using the poverty line, defined by the upper limit of the first quintile, Table 10 and Figure 9 illustrate the cost-benefit ratio of the seven types of transfers in Djibouti. This ratio measures the poverty gap reduction in local currency (Djibouti franc - DF) for every unit spent in the given program. For example, for the food rations transfers, the program with the highest ratio, every DF 1 spent in the program leads to DF 0.64 reduction in the poverty gap. However, for the tax exemption on fuel products, the transfer with the lowest ratio, every DF 1 spent in the program leads to only DF 0.01 reduction in the poverty gap. Table 10. Cost-Benefit Ratios - Upper Poverty Line Simulated Total Cost- Poverty Gap Actual Difference Amount Benefit Without Poverty Gap (dPG) Spent in the (dPG0/X) Transfer Program (X) Pensions 4 067 3 593 474 3 605 0.13 Compensation for health 3 604 3 593 10 213 0.05 care expenditure Food rations 3 835 3 593 241 377 0.64 Transfers from government 3 648 3 593 54 104 0.52 or NGOs Food subsidies 3 646 3 593 53 544 0.10 Fuel subsidies 3 596 3 593 3 484 0.01 Remittances 4 359 3 593 765 4 162 0.18 Source: World Bank calculation based on the EDAM 3. Note: All values are in DF. Cost-benefit is the poverty gap reduction in DJ for each unit (1 DF) spent in the social program. Amounts in millions of DF. 22 Figure 9. Cost-Benefit Ratios Poor Aide Alimentaire Aide Gouv et ONG Transfert privé Pension Exonération - aliment Prise en charge - Santé Exonération - Carburant 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 DF reduction in poverty gap for each 1 DF spent in the program Source: World Bank data. 28. In line with the low targeting accuracy and low generosity, social safety net programs in Djibouti are generally small and inadequate in reducing the poverty gap. Table 11 shows the generosity ratio, the ratio between the average transfers and the poverty gap, and the program size for each of the seven transfer programs in Djibouti. The ratio is above 1 for pension, which means that the average transfer is higher than the poverty gap and a strong indication that beneficiaries depend on this transfer as a source of income, which is usually expected for social insurance programs. On the other hand, the generosity of subsidies (both fuel and food) is negligible in closing the poverty gap; hence, a ratio of less than 0.1. Table 11. Decomposition of the Impact of the different Programs Average Poverty Gap Targeting Transfers (in Ratio (in DF) DF) Pensions 0.60 78,987.7 35,238.5 2.2 Compensation for health care 0.06 9,294.2 33,424.6 0.3 expenditure Food rations 0.76 8,684.8 34,942.9 0.2 Transfers from government or NGOs 0.63 9,342.8 33,814.5 0.3 Food subsidies 0.10 1,068.9 33,716.6 0.0 Fuel subsidies 0.01 1,550.6 33,490.3 0.0 Remittances 0.68 36,839.3 36,733.6 1.0 Source: World Bank calculation based on the EDAM 3. 23 EFFECTIVENESS OF TAX EXEMPTIONS AS AN UNTARGETED SAFETY NET THE IMPORT AND CONSUMPTION OF PETROLEUM PRODUCTS IN DJIBOUTI 29. Djibouti relies entirely on imports for the supply of petroleum products and domestic consumption is dominated by diesel. All imports come through Djibouti port. Of these imports, a substantial portion is re-exported, and in addition, there are large volumes in transit to Ethiopia. Furthermore, a large fraction of the net imports are destined for foreign armies with bases in Djibouti, for international airlines, and for maritime transport. Although there are no official figures for imports and consumption of petroleum products, a recent study for the government of Djibouti surveyed all parties involved in the import and sale of petroleum products and produced a reconciliation for 2012. The resulting data is shown in Table 12, and this indicates that domestic consumption is dominated by diesel. 30. The same study also presented a forecast of domestic consumption until 2017 in which diesel was expected to reach 91 million liters, kerosene 17 million liters, gasoline 9 million liters, and fuel oil 8 million liters. These figures indicate that the change in taxation of diesel will be particularly important in terms of government revenue. The domestic consumption of petroleum products is divided between households and businesses that pay all taxes and duties and a number of parties that receive some tax exemptions. These include certain businesses, embassies, and the Republican Guard. Table 12. Imports and Domestic Consumption of Petroleum Products by Djibouti in 2012 (Million Liters) Total imports 530 Re-export 283 Consumption by foreign military, airlines, and shipping 163 Domestic consumption 84 Diesel (gasoil) 61 Gasoline(super) 6 Kerosene (pétrole lampant) 11 Fuel oil 6 Source: Cap Gemini Consulting. January 2014. PETROLEUM PRODUCTS: PRICES, COSTS, AND TAXES 31. Fuel prices, especially of gasoline, are higher in Djibouti than in neighboring countries. Transportation fuel prices in Djibouti can be compared to those from other non-oil-producing countries in the region in 2012. With the exception of Eritrea, prices in Djibouti, especially gasoline prices, were higher than neighboring countries (see Table 13). This is largely due to the very small size of the domestic market, resulting in loss of economies of scale. Diesel prices are nearer to those of neighboring countries and this is in part due to discretionary tax offsets that have been used for diesel. 32. The retail prices of petroleum products in Djibouti are regulated by the Ministry of Finance according to a formula, which includes predetermined and discretionary elements (costs and taxes). The complete price and cost structure is reviewed monthly. Costs include an import component and various domestic items. The allowable amounts for domestic costs are changed occasionally, while tax rates are fixed except for a discretionary component (ajustement en faveur de l’Etat) that is used to smooth out fluctuations in retail prices that would otherwise be induced by fluctuations in the import cost. This component can be either positive (extra tax) or negative (tax offset). 33. The exact determination of the smoothed retail price is made not according to a formula but depends on judgments made by the government. In principle, such an approach to smoothing 24 out import cost fluctuations could result in no extra, long-run net benefit or cost to the government. However, there have been lengthy periods in which the discretionary component has been negative (because of the low final retail price set by the government) resulting in tax revenue persistently below that which would have resulted from the application of the non-discretionary tax structure. The government is now considering abandoning the discretionary tax component so that retail prices would be predictably linked to the allowable costs and the import cost of the products. This would mean that the full tax revenue implied by the formula would be collected from retail sales. 34. At present, kerosene is provided through two routes, yielding the same retail price due to a discretionary tax element. Besides the established marketing of kerosene, the government has recently made an arrangement with the La Société de Distribution et de Vente de Kérosène (SDVK) to distribute kerosene nationwide (although at present it serves only Djibouti-ville and suburbs). The government allows the SDVK to include a fee in the price charged to build its network and also exempts the price from domestic consumption tax (TIC) and VAT. The price does include a discretionary tax element so that the retail price of this kerosene supply is the same as the general retail price of kerosene. Table 13. Prices of Transportation Fuels in 2012 ($/liter) Country Super Gasoline Diesel (Gasoil) Djibouti 1.8 1.2 Ethiopia 1.1 0.9 Kenya 1.4 1.3 Eritrea 2.5 1.7 Lebanon 1.1 0.9 Tanzania 1.3 1.3 Source: GTZ: https://www.energypedia.info/index.php/International_Fuel_Prices. The Determination of the Retail Price 35. The various components of the pricing formula are set by the government in agreement with the oil companies.8 • The price for delivery at Djibouti port. FOB prices in international markets are collected monthly, as averages of daily FOB prices for the preceding month quoted in Platt’s Oilgram Price Report. An exporter’s margin, cost of shipping and insurance, and port fees are added. The commercial margin is updated every six months based on invoice information about actual FOB prices paid by oil companies in the preceding six months. • Duties and taxes include a domestic consumption tax (TIC) and VAT at rates set by legislation—currently 26 and 7 percent, respectively. Excise duties are set by legislation, and royalties are determined annually in the budget. • The transport, operational, and storage costs and the commercial margins of distributors and service station operators are set by the government. These can be changed after discussion with the relevant parties. • A discretionary tax component is used to smooth out retail prices in the face of volatile FOB costs and to reduce inflationary pressure on low-income households in times of rising international prices. This component varies each month and is calculated so as to produce a desired retail price linked to the other costs by the formula. 8 De Broek, M., A Kangur, and R. Kpodar. Djibouti: Fuel Price Subsidy Reform. IMF. May 2012. 25 36. The price build-up for retail gasoline is illustrated with the case of gasoline (super) for December 2013 (December 11, 2013–January 10, 2014) in Djibouti francs per liter. \Category Rate in DF PF (FOB price) [FOB de reference] 137.57 MF (maritime freight) [Fret maritime] 3.24 EM (exporters margin) [Marge commercial de l’exportateur] 4.36 PC (CIF price) = PF+MF+EM [Prix CIF Djibouti] 145.18 SE (extra storage cost) [Surcoût de stockage] 2.60 FP (port fees) [Redevances portuaires] 0.68 PP (price at port) = PC+SE+FP [Prix quai Djibouti] 148.46 TC (domestic consumption tax) = 0.26 * PP [Taxe intérieur à la consummation] 38.60 TE (excise duty) [Surtaxe] 49.50 TR (royalty) [Péréquation de prix] 32.13 TD (discretionary tax adjustment) [Ajustement en faveur de l’Etat] 5.839 DD (various distribution costs) [Passage en depot et distribution] 11.65 PV (price subject to VAT) = PP+TC+TE+TR+TD+DD [Prix de vente hors TVA aux 286.18 stations] TV (VAT) = 0.07* PV [TVA facturée( collectée)] 20.033 CT (terminal transport cost) [Transport terminal] 1.76 PS (price received at service station) = PV+TV+CT [Vente TTC aux stations services] 307.98 RM (retail margin) [Revente (stations services)] 7.02 PR (retail price) = PS+RM [Prix aux detail] 315.00 Source: World Bank data. 37. The regulated prices for petroleum products sold to special groups also allow for various exemptions on taxes, and these exemptions are expected to continue so that tax revenue from these groups is lower than that from the purely retail market. Such groups include the French military (50 percent exemption on domestic consumption tax and 50 percent exemption on excise duties); exempt businesses (zero domestic consumption tax and zero excise tax); embassies and domestic security forces (zero domestic consumption tax, zero VAT, zero excise tax, and zero royalty). The effect of these exemptions is to forgo tax revenue that would otherwise have accrued to the budget had the products been sold at the same price as to the general public. 38. The government recently entered into an arrangement with the SDVK to distribute kerosene in order to make it more widely available. A service fee was included in the price markup on top of other port and distribution charges. To encourage the company to set up the network required to expand the market for kerosene, the government exempted this kerosene from the domestic consumer tax (TIC), excise duty, and VAT. However, royalty was charged and the discretionary tax was set so as to bring the retail price to the same level as the charge for the traditional method of selling kerosene. 39. The retail prices and discretionary tax elements for 2013 are shown in Table 14. The data for gasoline prices illustrate how the retail price was smoothed by varying the discretionary tax through the year. In January 2013, the adjustment was negative, indicating that the government was holding down retail prices by forgoing a certain amount of tax revenue. By the end of the year, the retail price had risen slightly, but the discretionary tax had become positive, indicating that the government was now collecting some extra tax revenue. For diesel, the government was forgoing a certain amount of tax revenue throughout the year to hold retail prices down. It is important to note that the government during this period was still a net recipient of tax revenue from diesel. For kerosene, the government collected some tax revenue through royalty as well as a small amount from the (positive) discretionary tax. The net effect of this sales arrangement for kerosene was that the total tax revenue per liter was much lower than for gasoline and for diesel. 26 Table 14. Retail Prices and Discretionary Taxes for Petroleum Products in 2013 (DF/liter) Month Gasoline Kerosene (SDVK) Diesel Discretionary Tax Discretionary Tax Discretionary Tax Other Taxes Other Taxes Other Taxes Retail Price Retail Price Retail Price FOB Price FOB Price FOB Price January 139.27 310.00 -11.47 141.70 137.66 195.00 8.68 7.00 135.08 210.00 -28.47 75.40 February 139.27 310.00 -11.47 141.70 142.02 195.00 4.29 7.00 139.79 210.00 -34.44 76.65 March 139.27 310.00 -11.47 141.70 148.65 200.00 2.61 7.00 146.46 215.00 -38.23 78.69 April 142.58 310.00 -5.90 140.59 136.31 190.00 5.04 7.00 135.44 210.00 -28.92 75.51 May 142.58 310.00 -5.31 141.26 128.29 190.00 13.12 7.00 127.98 210.00 -19.46 73.57 June 137.57 310.00 1.17 139.94 127.88 190.00 13.53 7.00 128.47 210.00 -20.08 73.70 July 137.57 310.00 1.17 139.94 129.52 190.00 11.88 7.00 131.44 212.00 -21.98 74.60 August 137.57 315.00 5.85 140.26 134.89 195.00 11.47 7.00 136.16 215.00 -25.16 76.02 September 137.57 315.00 5.84 140.26 138.39 195.00 7.95 7.00 136.73 215.00 -25.88 76.16 October November 137.57 315.00 5.84 140.26 136.50 195.00 9.85 7.00 136.42 215.00 -25.49 76.08 December 137.57 315.00 5.84 140.26 136.25 195.00 10.01 7.00 136.09 215.00 -25.07 76.00 Source: Ministry of Finance. 40. The calculation of the discretionary element is similar for the exempt organizations but it has to be interpreted differently. The price calculation for the Republican Guard is given as an illustration. In December 2013, the price at the port (PP) for gasoline was DF 148.46; domestic tax, excise tax, royalty, and VAT were all zero; distribution costs (DD) were DF 11.65; terminal transport cost was DF 1.76; and retail margin was DF 7.02. The sum of these costs was DF 168.89 and this was the charge to the Republican Guard. Relative to the retail price of DF 315, this involved a tax exemption of DF 146.10 that can be seen as a transfer from one segment of the government to another. Impact of Abandoning the Discretionary Tax on Retail Prices 41. The proposal currently under consideration by the government is to abandon the use of the discretionary tax element. Other tax rates could be varied by legislation, as at present, but would normally be stable for lengthy periods. Allowable costs along the supply chain could also be varied if justified by the circumstances of the entities involved. To simulate the effect of removing the discretionary tax element on prices, it is assumed that all other tax rates and costs remain at the levels of December 2013. The calculation shows the effect on the price of kerosene that is supplied by the SDVK. 42. For gasoline and diesel, the removal of the discretionary tax component has two effects on the retail price that would have to be charged. First, when the discretionary tax is positive, its removal would contribute to lowering of the retail price by this amount. Second, because VAT at 7 percent is charged on the discretionary tax element, there would be a further lowering of the retail price. Similarly, when the discretionary tax is negative, its removal would raise retail prices by 1.07 times the amount of the discretionary tax. For kerosene sold by the SDVK, there is no VAT element so that the resulting price change would be equal to the amount of the discretionary tax. Table 15 illustrates the effects on retail prices if the tax had been removed in December 2013. Table 15. Retail Petroleum Product Prices with and without the Discretionary Tax Element (December 2013) DF/liter Gasoline Kerosene (SDVK) Diesel Before After % Change Before After % Change Before After % Change 315.00 308.73 -2.0 195.00 184.90 -5.2 215.00 241.82 +12.5 Source: Authors’ calculations based on World Bank data. 27 43. The removal of the discretionary tax element would have resulted in a small fall in gasoline and kerosene prices but an increase of around 13 percent for diesel. The comparison between the before and after prices in December 2013 is possible because the government’s action with respect to the determination of the retail price (and the associated discretionary tax) is a known fact. Simulating the effect of removing the discretionary tax under different circumstances is possible, but it is not possible to give a ‘before’ calculation since it is not known what the government would have decided to do with retail prices had it kept the discretionary tax. 44. As an illustration of the range of retail prices that might be experienced if the discretionary tax were abandoned, simulations of the impacts of a 20 percent increase and a 20 percent decrease in the FOB prices (relative to the levels of December 2013) are constructed and the results are shown in Table 16. It is assumed that all other costs, taxes, and duties remain unchanged, and in all cases there is no discretionary tax element. The results of the calculations show gasoline price varying by between ±12 percent, kerosene prices by ±15 percent, and diesel prices by ±15 percent. The larger fixed elements for gasoline (excise duty and royalty) mean that the percentage swing in the FOB price (which is similar for all three products) is damped down more than is the case for diesel and for kerosene 45. Crude oil prices have recently fallen substantially and this is relevant to the calculations shown. In December 2013, Brent crude sold for about US$110 a barrel and it remained around that level until July 2014. Since then it has steadily declined until falling to around US$70 a barrel in December 2014 (i.e., a drop of 36 percent). There is considerable uncertainty about the course of prices in 2015, with some analysts expecting a small recovery to around $80 a barrel. 46. The simulations of the effect of abandoning the discretionary tax focused on December 2013 actual FOB prices, and included a sensitivity analysis of a 20% drop of those prices (Table 17). This drop would have corresponded to a Brent crude price of about $90 a barrel. The actual fall has been almost twice that allowed in the sensitivity analysis. Simple extrapolation indicates that, in the absence of discretionary taxes, a 40% fall in FOB prices would result in gasoline prices of 235 DF/liter, kerosene prices of 130 DF/liter, and diesel prices of 169 DF/liter. Table 16. Results of Simulation – Range of Retail Prices (DF/liter) Gasoline Kerosene (SDVK) Diesel December 2013 FOB 308.70 184.90 241.80 December 2013 FOB plus 20% 345.80 212.20 278.50 December 2013 FOB minus 20% 271.60 157.70 205.10 Source: Authors’ calculations based on World Bank data. GASOIL PRICES AND TRANSPORT COSTS 47. The analysis of the impact of removing the discretionary tax element on households proceeds through the use of an expenditure survey, coupled with the calculated changes in petroleum product prices. The shares of total household expenditure allocated to each of the three petroleum products is directly available from the household expenditure survey and can be combined with predicted price changes to estimate the expenditure change required to purchase the same amounts of each product. 48. Besides these direct effects on household budgets, there are indirect effects caused by the impact of rising petroleum product prices on other goods and hence on the household budget. Without a detailed input-output table it is not possible to quantify all such links, but the most important link for petroleum products in Djibouti is transport costs. In particular, costs of travel by bus or taxi can be an important component of household expenditure, so it is important to consider the link between product prices and transport costs. 49. As gasoil (diesel) is used as fuel for commercial transportation vehicles, the key question is the nature of the link between the gasoil price and the price of transportation services. Bus 28 fares are regulated and have changed very little in the last decade. However, it is likely that a jump in fuel costs brought about by the abolition of the discretionary tax, which was holding down costs by about 12 percent in December 2013, could provide an opportunity for the bus sector to ask for higher ticket prices to cover their increased costs. Many factors might enter such a negotiation, including previous loss of profitability caused by the government holding prices steady for a long period. A full justification of an allowable fare increase would require detailed analysis of the economics of the bus and taxi sectors. In the absence of such a detailed study, a first approximation can be obtained by combining the fuel share in total costs with the percentage increase in fuel costs. 50. There is some evidence on the share of fuel costs in the total costs of operating a bus fleet from other countries that can serve as a marker for any assumption that is made for the case of Djibouti. ESMAP (2011) refers to a study in India where the share of fuel cost in Andhra Pradesh amounted to 37 percent of total costs. A study by the World Bank analyzed factors affecting bus performance in middle- to low-income countries and provided values indicative of the range of cost breakdown as shown in Table 17. The following remarks from the study are relevant in the present case: “In the case of informal small-scale operation using rehabilitated or locally fabricated buses, financed by overseas remittances, depreciation and interest costs are much less (only about 10 percent of total costs), while driver and other staff costs can be relatively more (20-30 percent) due to the higher number of people employed per unit of capacity (often including the owner).” Table 17. Shares of Operating Cost of a Bus Fleet in Developing Countries Cost Item Proportion of Operating Cost (%) Variable costs Fuel 20–30 Lubricating oil 1–5 Tires 5–10 Spares 5–10 Fixed costs Driver and other platform staff 10–15 Other labor About 5 Depreciation and interest 20–30 Overheads and other costs 5–15 Source: www.worldbank.org/transport/roads/rdt_docsannex1.pdf. 51. Table 17 indicates that fuel costs can range between about 50 and 75 percent of variable costs depending on circumstances. A survey carried out by the SESN and DISED in Djibouti-ville in 2014 indicated that, averaged over all forms of passenger road transport, fuel accounted for 80 percent of variable costs (see annex 4 for details) and that there was little variation in this ratio among the different forms of passenger transport. The closeness of these figures suggests that it is reasonable to assume that fuel costs in Djibouti are about 30 percent of total operating costs (the high end of the range given in Table 17 corresponding to the 75 percent share of variable costs). 52. Combining the information on the possible increase in diesel prices (12.5 percent) relative to their level in December 2013 with a fuel cost share of 30 percent suggests that an increase of 4 percent in passenger fares would be justified to allow bus companies to overcome the effects of higher fuel prices. If the government decided to permit a larger price rise, perhaps to allow for ‘catch-up’ with previous cost increases, there is more likelihood of public opposition to the change. WHO CONSUMES ENERGY PRODUCTS? 53. Car ownership and utilization of public transport is a strong indication of welfare. Car ownership is not widespread in Djibouti: only 6 percent of households own a car and 1 percent owns a motorcycle. One-fourth of the richest quintile owns a car while car ownership is basically negligible in the other quintiles (see Table 18). Most cars are owned in urban areas. Not surprisingly, carburant is only consumed only by urban households (see figure 10) of the richest quintile (see Figure 11). 29 Utilization of public transport (bus, taxis, and school buses) is also highest among the richer quintiles. Only 10 percent of the poor use public transport compared with almost 60 percent in the richest 2 quintiles. About half the population in urban areas makes use of public transport but less than 10 percent in rural areas. Among school children, only 5 percent of children in the poorest quintile take the school bus compared with 30 percent in the richest 3 quintiles. Table 18. Percentage of Households which Own a Car or Motorbike Quintile Area Total Poorest Second Third Fourth Richest Urban Rural Car 0.0 0.4 1.7 3.0 25.3 7.1 0.7 6.1 Motorbike 0.1 0.5 0.6 1.3 3.3 1.3 0.2 1.2 Source: World Bank calculation based on the EDAM 3. Figure 10. Utilization of Different Energy Sources by Location (in %) 100 80 80 59 60 51 50 37 Urban 40 Rural 20 10 13 14 7 7 2 1 1 1 0 Bus/taxi School Carburant Kerosene Gas Electricity Coal transport Source: World Bank data. Figure 11. Utilization of Different Energy Sources of Energy, by Quintile (in %) 100 88 89 80 58 Poorest 60 Second 42 45 40 Third 23 21 25 20 12 7 10 Fourth 3 6 0 0 Richest Bus/taxi School Carburant Kerosene Gas Electricity Coal transport Source: World Bank data. WHO BENEFITS FROM TAX EXEMPTIONS? 54. In response to the food crisis and to shield the population from price shocks on essential food and certain energy products, universal tax exemptions were introduced. Djibouti depends heavily on imports to meet its food needs, and a large fraction of the population faces food insecurity. Practically all food items are imported and increases in international food prices directly affects Djibouti’s poor people, who spend up to three-quarters of their income on food. Due to severe and prolonged droughts, at least 20 percent of the capital’s population and three-quarters of rural households are left vulnerable to severe and moderate food insecurity (World Food Program 2013). According to IMF estimates, Djibouti foregoes 0.5 percent of GDP (2009) on certain food items (rice, 30 edible oil, sugar, flour, and powdered milk) and an estimated 2 percent of GDP (2011) on certain energy products (IMF, 2012). 55. The following analysis includes all the tax-exempt fuel and food products available in the household survey. Among the basic food items that are tax exempt, only a certain quality/type is exempt (for example, broken rice). For rice, only 6 percent of the imported rice is exempt, but about 88 percent of flour, about 60 percent of sugar and edible oil, and about 50 percent of powdered milk products are exempt. For flour, the implicit subsidy represents 7 percent of the unsubsidized price. The survey does not differentiate between diesel and super (lumped together as carburant), but data from the Enquete de Budget et Consomation (EBC) survey (an urban only survey done in 2013) show that around two-thirds of spending by households on carburant is on diesel. Furthermore, the survey shows that almost all direct spending on carburant is by the richest quintile. For carburant (super and diesel), the implicit subsidy represents 11 percent of the unsubsidized price. For public transport, the analysis assumes a fuel cost share of 30 percent, which corresponds to passing on 4 percent to passengers. 56. Poor households spend relatively more on tax-exempt products, with the exception of fuel and public transport. Household expenses on tax-exempt products amount on average to DF 267,531 per household, which is equivalent to 22.8 percent of total household spending. Table 19 shows household expenses on tax-exempt products and Table 20 shows the proportion of household spending toward those tax-exempt products. Kerosene is the most consumed item in terms of expenditures (DF 42,777), followed by rice (DF 36,671) and sugar (35 670 DF). Tax-exempt products are relatively more important for the poor, as the expenditure share of these products is much higher for the very poor than for the very rich. In the poorest households, 30.7 percent of the total expenses correspond to tax-exempt products, while these tax-exempt products account for just 19.6 percent of the richest households’ total expenses. 57. Tax exemptions on fuel products do not benefit the poor as they consume little fuel and hardly use public transportation. As shown in Table 18, possession of cars and motorbikes is essentially limited to the fifth quintile, which consumes DF 91,847 per household on fuel, while the very poor do not consume fuel at all. Spending on public transport is also considerably lower in the poorest quintile (DF 2,030 per household) than in the richest quintile (over DF 47,000 per household). Already the second quintile spends considerably more on public transport than the very poor (see Table 19). For the poor, expenses on public transport amount to less than 1 percent of overall household expenses. The middle three quintiles spend about 2 percent of overall expenses on buses/taxis and about 3 percent on school transport (see Table 20). Table 19. Expenditures per Household (in DF) Powder- Cooking Public School Rice Flour Sugar Fuel Kerosene Total ed Milk Oil Transport Transport Poorest 19,688.78 4,029.88 16,549.11 7,833.32 21,041.96 0.00 5,146.00 2,030.48 2,257.50 78,577.02 Second 32,621.33 16,296.78 25,599.48 16,661.67 34,811.94 473.25 31,646.14 12,511.09 12,507.85 183,129.53 Third 36,437.36 25,199.99 24,215.76 19,422.76 38,164.19 584.79 48,174.52 24,036.83 30,107.64 246,343.84 Fourth 40,280.14 33,436.31 26,876.72 22,548.07 39,593.50 3,880.76 58,707.99 36,030.61 44,456.45 305,810.56 Richest 49,970.80 52,381.60 30,163.75 31,489.47 42,742.87 91,847.41 63,248.31 47,251.26 48,289.03 457,384.50 Total 36,671.28 27,996.93 24,982.63 20,337.09 35,670.25 24,102.90 42,776.72 25,960.58 29,033.01 267,531.41 Source: World Bank calculation based on the EDAM 3. 31 Table 20. Expenditure on Subsidized Products over Total Expenditures (in %) Powdered Cooking Public School Rice Flour Sugar Fuel Kerosene Total Milk Oil Transport Transport Poorest 7.7 1.6 6.5 3.1 8.2 0.0 2.0 0.8 0.9 30.7 Second 4.9 2.5 3.9 2.5 5.3 0.1 4.8 1.9 1.9 27.7 Third 3.9 2.7 2.6 2.1 4.1 0.1 5.2 2.6 3.2 26.3 Fourth 3.2 2.7 2.1 1.8 3.2 0.3 4.7 2.9 3.5 24.4 Richest 2.1 2.2 1.3 1.4 1.8 3.9 2.7 2.0 2.1 19.6 Total 3.1 2.4 2.1 1.7 3.0 2.1 3.7 2.2 2.5 22.8 Source: World Bank calculation based on the EDAM 3. IMPACT OF SUBSIDY REFORMS 58. This section presents a simulation that eliminates the discretionary tax elements on each of the subsidized products discussed in the previous section and then shows the potential impact of this removal on household welfare, government budget, poverty, and inequality. Household Welfare 59. The reform would imply a loss of DF 1,014 million (or 0.66 percent of GDP) for the population; however, the per capita values indicate that the loss would be higher for the richest. This is explained by fact that the richest have a higher consumption level than the poor as mentioned in the previous section. Table 21 shows the impact of the reform on the welfare of the population for each quintile, as illustrated in Figure 12 and Table 22, and Table 22 shows the impact of the reform on the per capita welfare of each quintile. For fuel, the impact of the reform on poor households is negligible, but it increases with welfare and represents the highest loss among rich households. In fact, among the poorest households, the reform would imply an important loss in welfare from food- related products—flour, followed by sugar—while for the richest the loss will be the highest for fuel, followed, by far, by flour (also seen in per capita basis in Table 22). This is consistent with the finding in the previous sections that the richest households spend a significant share of their income on fuel. Table 21. The Total Impact on the Population’s Well-Being (in millions DF) Rice Powdered Milk Flour Cooking Oil Sugar Poorest 0.0 -3.0 -23.5 -6.9 -17.5 Second 0.0 -10.7 -32.2 -13.1 -25.6 Third 0.0 -17.1 -31.7 -15.8 -29.2 Fourth 0.0 -25.0 -38.6 -20.2 -33.3 Richest 0.0 -49.7 -55.1 -35.8 -45.7 Total 0.0 -105 -181.1 -91.8 -151.3 Public School Fuel Kerosene Total Transport Transport Poorest 0 0 -1.4 -1.6 -53.8 Second -1.0 0 -7.7 -7.7 -97.9 Third -1.3 0 -15.4 -19.3 -129.9 Fourth -9.2 0 -25.4 -31.3 -183.0 Richest -277.5 0 -42.3 -43.2 -549.3 Total -288.9 0 -92.2 -103.1 -1 014 Source: World Bank calculation based on the EDAM 3. 32 Figure 12. The Total Impact on the Population Well-Being (in DF) Poorest Second Third Fourth Richest Total 0.0 -200.0 -400.0 Food Energy -600.0 Total -800.0 -1,000.0 -1,200.0 Source: World Bank data. Table 22. The Impact on the Per Capita Well-Being (in DF) Powdered Cooking Public School Rice Flour Sugar Fuel Kerosene Total Milk Oil Transport Transport Poorest 0.00 -28 -219 -64 -163 0 0 -13 -15 -501 Second 0.00 -99 -299 -121 -238 -9 0 -72 -72 -911 Third 0.00 -160 -296 -148 -273 -12 0 -144 -180 -1,211 Fourth 0.00 -232 -359 -188 -309 -86 0 -236 -291 -1,700 Richest 0.00 -464 -515 -335 -427 -2,592 0 -395 -404 -5,131 Total 0.00 -196 -337 -171 -282 -538 0 -172 -192 -1,889 Source: World Bank calculation based on the EDAM 3. 60. Among the poorer quintiles, the loss in welfare as a result of the reform would be the highest on food related items; while it would be the highest on fuel products among the richer quintiles. Table 23 and Figure 13 show the impact of the reform on the welfare of the population as a proportion of total spending by quintile and for each subsidized product. In terms of food-related products, the reform would result in a significant loss of welfare among the poorest quintile (1.12 percent of total spending) (see Figure 13) but this loss decreases as welfare increases. On the other hand, the reform would result in a minimal loss among the top quintile for fuel products, and this loss decreases as welfare decreases and becomes negligible for the first quintile. Table 23. The Impact on Well-Being (in %) Powdere Cooki Kero- Public School Rice Flour Sugar Fuel Total d Milk ng Oil sene Transport Transport Poorest 0.00 -0.06 -0.49 -0.14 -0.37 0.00 0.00 -0.03 -0.03 -1.12 Second 0.00 -0.10 -0.29 -0.12 -0.23 -0.01 0.00 -0.07 -0.07 -0.90 Third 0.00 -0.11 -0.20 -0.10 -0.18 -0.01 0.00 -0.10 -0.12 -0.81 Fourth 0.00 -0.11 -0.16 -0.08 -0.14 -0.04 0.00 -0.11 -0.13 -0.77 Richest 0.00 -0.09 -0.10 -0.06 -0.08 -0.49 0.00 -0.08 -0.08 -0.98 Total 0.00 -0.09 -0.16 -0.08 -0.14 -0.26 0.00 -0.08 -0.09 -0.91 Source: World Bank calculation based on the EDAM 3. 33 Figure 13. The Impact on Well-Being (in %) Poorest Second Third Fourth Richest Total 0.00 -0.20 -0.40 Food Energy -0.60 Total -0.80 -1.00 -1.20 Source: World Bank data. Government Budget 61. The impact of the reform on government budget would result in a gain, the highest coming from fuel. Table 24 shows the impact of the reform on government budget from the different subsidized products. The impact of the reform on government budget would result in a total gain of DF 811 million (or 0.53 percent of GDP): 28 percent of that gain would come from fuel (96 percent of the gain from fuel will originate from the richest households), 18 percent from flour, and 15 percent from sugar. The highest gain in government budget will come from the richest households (54 percent). This decreases as welfare decreases to reach the lowest share among poor households (5 percent). This is consistent with the previous finding that the highest loss of welfare in the population would come from fuel, and particularly among the rich. Figure 14 shows the impact on government revenues as the price of each subsidized product increases. The most important revenue gain to the government would come from increasing the price of fuel, while the least would come from increasing the price of cooking oil. It should be noted that since we assume a price elasticity of 0.2, the amount gained by the government is less than the loss incurred by the different households. Table 24. The Impact of the Reform on the Government Revenue (in millions of DF) Powdered Rice Flour Cooking Oil Sugar Milk Poorest 0.0 2.4 18.8 5.5 14.0 Second 0.0 8.5 25.8 10.4 20.5 Third 0.0 13.7 25.4 12.7 23.4 Fourth 0.0 20.0 30.9 16.2 26.7 Richest 0.0 39.8 44.1 28.7 36.6 Total 0.0 84.4 144.9 73.5 121.1 Public School Fuel Kerosene Total Transport Transport Poorest 0.0 0.0 1.1 1.3 43.0 Second 0.8 0.0 6.2 6.2 78.4 Third 1.0 0.0 12.3 15.5 103.9 Fourth 7.4 0.0 20.3 25.1 146.4 Richest 222.0 0.0 33.8 34.6 439.5 Total 231.2 0.0 73.8 82.5 811.2 Source: World Bank calculation based on the EDAM 3. 34 Figure 14. The Impact of the Reform on the Government Revenue (in DF) 900 811 800 700 600 500 400 300 231 200 145 121 84 73 74 82 100 0 0 0 Rice Powder Flour Cooking Sugar Fuel Kerosene Public School Total Milk Oil Transport Transport Source: World Bank data. Poverty and Inequality 62. As the poor spend most of their income on food-related products, the elimination of tax exemptions on such products would have the highest impact on destitution and inequality, while the elimination of tax exemptions on fuel products would reduce inequality but with no apparent impact on poverty. However these effects would be minimal, almost negligible. Table 25 shows the impact of the reform on destitution and inequality. The reform would not have a very significant impact on destitution and no impact on inequality. In particular, the poverty rate would increase by 0.17 percentage points from 20.00 to 20.17 percent. Similarly, the Gini index would decrease by 0.02 percentage points from 45.13 to 45.11 percent. The elimination of tax exemptions on flour would increase destitution by 0.05 percentage points (from 20 to 20.05 percent), and inequality by 0.05 percentage points (from 45.13 to 45.18 percent). The effect of the elimination of the discretionary tax adjustment on fuel would not affect the poorest and would in fact result in a reduction of inequality by 0.12 percentage points. This is explained by the fact that the consumption of this product is negligible among the poor, while it is one of the highest consumed products for rich households among the subsidized products. Table 25. The Reform, the Destitution Headcount, and the Gini Index Poverty Change in Gini Variation Level Poverty Index in Gini Pre Reform 20.00 . 45.13 . Rice 20.00 0.00 45.13 0.00 Powdered Milk 20.00 0.00 45.13 0.00 Flour 20.05 0.05 45.18 0.05 Cooking Oil 20.01 0.00 45.15 0.01 Sugar 20.01 0.01 45.17 0.04 Fuel 20.00 0.00 45.01 -0.12 Kerosene 20.00 0.00 45.13 0.00 Public Transport 20.00 0.00 45.13 0.00 School Transport 20.00 0.00 45.14 0.00 Post Reform 20.17 0.17 45.13 -0.02 Source: World Bank calculation based on the EDAM 3. 35 SIMULATING REFORM OPTIONS BUDGET AND SIMULATIONS 63. A rough estimate of the available budget to simulate reform options is based on current budget estimates and half the savings from removing tax exemptions. Table 26 shows an estimation of a budget that could be transferred by the SESN. At this stage, knowing the exact amount to be transferred is not that important and simulations are mainly illustrative. 64. Based on discussions with the SESN, 6 families of transfer scheme either at individual, equivalent adult, or household level yield a total of 18 schemes. For each scheme, we assume that the total budget to be transferred is DF 1, 2, or 3 billion for a grand total of 54 different simulations. Table 26. Budgetary Sources (Preliminary Proposal) Sources Amount (in millions of DF) FNS (50% of total) 1,000 WFP 300 Subsidies (50% of total) 400 Local NGOs 400 SESN 120 Total 2,220 Source: World Bank calculation based on the EDAM 3. 65. Table 27 defines 6 families of transfer schemes: • Rural + urban outside Djibouti-ville: all individuals away from Djibouti-ville are targeted irrespective of whether they are from urban or rural areas. • Rural only. • Rural + urban in first quintile: all rural + individuals that are in first quintile (nationally defined). • First quintile with unique transfer: any households (urban or rural) found in the first quintile are targeted. Targeted households received the same transfer, whether they are the poorest or the ‘richest’ within the first quintile. • First quintile with 2 steps: same as the one above but the amount transferred depends on whether you are the first or second decile. • First quintile with 4 steps: same as above but with 4 different amounts, one for the poorest 5 percent, 5–10 percent, 10–15 percent, and 15–20 percent. That scheme is closer to an optimal one where the amount transferred would be different for each household, depending on the PMT level, the poorest receiving the most, then the second poorest receiving a bit less, and so on. However, such a scheme would be rather difficult to implement. • The total budget allotted to the different groups was based on the relative size of their poverty gap. The following was used: o 0–5%: 41.0% of total budget o 5–10%: 33.9% o 10–15%: 19.0% o 15–20%: 6.2% 66. Those 6 schemes could be implemented at individual or household level. In the latter case, the amount transferred is the same for any household meeting the selection criteria irrespective of the household size. On the other hand, the ‘individual schemes’ depend on household size. For example, a nine-member household would receive three times the amount received by a three-member household. An intermediate measure is based on equivalence adults (Eq.Ad.). Each member is then weighted by the calorie requirement as defined in Table 28. 36 Table 27. Definition of the Different Transfer Schemes Transfer Selection Criteria Beneficiary Amount Transferred Per No. Unit 1 Rural + urban outside Individual 6,935 Djibouti-ville 2 Individual (in Eq.Ad.) 9,268 3 Household 35,826 4 Rural only Individual 11,560 5 Individual (in Eq.Ad.) 15,717 6 Household 54,940 7 Rural + urban in first quintile Individual 7,675 8 Individual (in Eq.Ad.) 10,259 9 Household 42,550 10 First quintile with unique Individual 9,306 transfer 11 Individual (in Eq.Ad.) 12,418 12 Household 58,748 13 First quintile with 2 steps Individual percentile 10: 13,960 percentile 20: 4,673 14 Individual (in Eq.Ad.) percentile 10: 18,811 percentile 20: 6,176 15 Household percentile 10: 90,133 percentile 20: 28,863 16 First quintile with 4 steps Individual percentile 5: 15,245 percentile 10: 12,670 percentile 15: 7,066 percentile 20: 2,288 17 Individual (in Eq.Ad.) percentile 5: 20,629 percentile 10: 17,001 percentile 15: 9,313 percentile 20: 3,031 18 Household percentile 5: 110,709 percentile 10: 73,607 percentile 15: 42,777 percentile 20: 14,417 Source: World Bank calculation based on the EDAM 3. 37 Table 28. Recommended Calorie Intake Category Age (years) Average energy Equivalence scale allowance per day (kcal) Infants 0–0.5 650 0.22 0.5–1.0 850 0.29 Children 1–3 1,300 0.45 4–6 1,800 0.62 7–10 2,000 0.69 Males 11–14 2,500 0.86 15–18 3,000 1.03 19–25 2,900 1.00 25–50 2,900 1.00 51+ 2,300 0.79 Females 11–14 2,200 0.76 15–18 2,200 0.76 19–25 2,200 0.76 25–50 2,200 0.76 51+ 1,900 0.66 Source: Recommended Dietary Allowances, 10th edition. Washington, D.C. National Academy Press. 1989. Destitution Headcount 67. The largest decline in destitution headcount is achieved when targeting the first quintile. Table 29 shows the destitution headcount (P0)—poverty defined as first quintile—after transferring three different budget amounts (DF 1, 2, and 3 billion) according to the 18 schemes defined. Overall, with a total budget of DF 1 billion, the effect on destitution headcount is limited if we concentrate mainly on rural households without taking into account urban households from the first quintile. The largest decline in destitution headcount with a DF 1 billion budget is scheme 10–12, which targets the first quintile and transfers a uniform amount. 68. With a larger budget of DF 3 billion, it would be possible to almost halve the destitution headcount using any of the schemes that target the first quintile. Any scheme that targets the first quintile (schemes number 7 to 18) achieves a significant reduction in destitution headcount with a 3 billion budget. Using such a destitution headcount as a measure of efficiency, however, it is not clear whether an individual or a household-based scheme is more efficient at reducing destitution. The main problem using a destitution headcount to assess the different schemes is that no weight is given when an extremely poor household receives an important transfer while remaining below the destitution line. Actually we can imagine an extreme case where all the poorest households would be much better off but still poor if none of the amount transferred makes them go over the poverty line. Because of that, we should focus on destitution gap as a measure of poverty. Destitution Gap 69. To reduce the destitution gap, targeting the first quintile is more efficient than any of the schemes focusing on rural households. The destitution gap index (P01) estimates the depth of destitution by considering how far, on average, the poor are from that poverty line. It is defined as the average destitution gap in the population as a proportion of the destitution line. In a graph presenting the cumulative welfare function, this is the area below the poverty line and on the left hand side of the function. Before any transfer, the destitution gap index (PGI or P1) associated with the destitution line used in Table 30 is measured as 6.9 percent (last line in the table). On average, the poor individual has expenditures (as measured by the PMT) 6.9 percent below the destitution line (DF 77,926 per capita). 38 70. Table 30 clearly shows that there should be an exclusive transfer scheme focusing on the first quintile. Importantly, such a transfer would be much more efficient at reducing poverty than any of the schemes focusing on rural households (schemes 1–9). The preferred scheme to reduce the destitution gap would be targeting the first quintile with a 4-step transfer amount depending on poverty. Schemes number 16 or 17 would be by far the best: focusing on the first quintile but with the amount transferred being dependent on poverty (as defined by the PMT). In this case, the poorest 5 percent would receive more than the poorest 10 percent, and so on (see Table 27). In a scheme focusing on ‘rural only’, Table 31 shows that only 62.4 percent of individuals in the first quintile, and quite a few non-poor households, would receive a transfer (11.8 percent of individuals of Q2, 4.3 percent in Q3, and so on). Any such leakage (households receiving transfers although they were not targeted) necessarily makes rural-based criteria less efficient at reducing poverty than one focusing solely on Q1. Alternatively, the ‘rural + urban in first quintile’ would have the advantage of covering everybody in Q1 but the leakage would remain similar (i.e. the non-poor in rural areas), making such a scheme only slightly better (no exclusion error but the same inclusion error) that the ‘rural only’ one (see Table 31). In summary, by far the best scheme would be number 16 or 17. Table 29. Effect on Destitution Headcount of the Different Transfer Schemes Transfer Selection Criteria Beneficiary DF 1 billion DF 2 billion DF 3 billion No. 1 Rural + urban outside Individual 18.4 17.0 15.2 Djibouti-ville 2 Individual (in Eq.Ad.) 18.4 17.0 15.2 3 Household 18.5 16.9 15.4 4 Rural only Individual 18.7 16.9 13.8 5 Individual (in Eq.Ad.) 18.7 16.7 14.0 6 Household 18.3 16.6 14.5 7 Rural + urban in first Individual 16.5 13.3 10.8 quintile 8 Individual (in Eq.Ad.) 16.3 13.3 10.6 9 Household 16.6 13.7 11.1 10 First quintile with unique Individual 15.8 12.0 8.5 transfer 11 Individual (in Eq.Ad.) 16.0 12.0 8.9 12 Household 15.5 11.8 8.4 13 First quintile with 2 steps Individual 17.8 14.3 7.7 14 Individual (in Eq.Ad.) 17.7 14.6 8.3 15 Household 17.8 12.9 8.1 16 First quintile with 4 steps Individual 18.9 16.2 5.8 17 Individual (in Eq.Ad.) 18.9 16.0 7.2 18 Household 18.7 14.0 7.8 Without transfer 20.0 20.0 20.0 Source: World Bank calculation based on the EDAM 3. 39 Table 30. Effect on Destitution Gap of the Different Transfer Schemes Transfer Selection Criteria Beneficiary DF 1 billion DF 2 billion DF 3 billion No. 1 Rural + urban outside Individual 5.5 4.3 3.3 Djibouti-ville 2 Individual (in Eq.Ad.) 5.6 4.4 3.4 3 Household 5.7 4.6 3.8 4 Rural only Individual 5.0 3.4 2.2 5 Individual (in Eq.Ad.) 5.0 3.5 2.3 6 Household 5.3 4.0 3.1 7 Rural + urban in first Individual 5.0 3.4 2.2 quintile 8 Individual (in Eq.Ad.) 5.0 3.5 2.3 9 Household 5.1 3.8 2.7 10 First quintile with unique Individual 4.6 2.9 1.6 transfer 11 Individual (in Eq.Ad.) 4.7 3.0 1.7 12 Household 4.6 2.9 1.8 13 First quintile with 2 steps Individual 4.4 2.1 0.6 14 Individual (in Eq.Ad.) 4.5 2.2 0.8 15 Household 4.3 2.3 1.1 16 First quintile with 4 steps Individual 4.4 1.9 0.3 17 Individual (in Eq.Ad.) 4.4 2.0 0.4 18 Household 4.2 2.0 0.8 Without transfer 6.9 6.9 6.9 Source: World Bank calculation based on the EDAM 3. Table 31. Coverage Rate Rural + Urban Rural Only Rural + Urban in First Quintile outside Djibouti- First Quintile ville Area Urban 12.8 0.0 9.0 9.0 Rural 100.0 100.0 100.0 77.5 Region Djibouti 0.0 0.0 6.8 6.8 Ali Sabieh 100.0 34.3 53.4 41.1 Dikhil 100.0 64.3 72.2 56.7 Tadjourah 100.0 69.4 75.2 64.6 Obock 100.0 65.5 73.1 70.0 Arta 100.0 66.9 75.0 50.8 Quintile Poorest 75.3 62.4 100.0 100.0 Second 27.8 11.8 11.8 0.0 Third 17.6 4.3 4.3 0.0 Fourth 9.9 1.6 1.6 0.0 Richest 3.6 0.4 0.4 0.0 Total 26.9 16.1 23.7 20.0 Source: World Bank calculation based on the EDAM 3. 40 References • Cap Gemini Consulting. January 2014. • De Broek, M., A Kangur, and R. Kpodar. Djibouti: Fuel Price Subsidy Reform. IMF. May 2012. • Direction de la Statistique et des Etudes Démographiques. 2012. Enquête Djiboutienne Auprès des Ménages (EDAM 3-IS). • GIZ, International Fuel Price Database, https://www.energypedia.info/index.php/International_Fuel_Prices • Subcommittee on the Tenth Edition of the Recommended Dietary Allowances, Food and Nutrition Board, Commission on Life Sciences, National Research Council Recommended Dietary Allowances, 10th edition. Washington, D.C. National Academy Press. 1989. • World Bank. 2011. Best operational and maintenance practices for city bus fleets to maximize fuel economy : energy efficient cities initiative. Energy Sector Management Assistance Program ; briefing note 010/11. Guidance note. Washington, DC: World Bank. http://documents.worldbank.org/curated/en/2011/01/16259489/best-operational- maintenance-practices-city-bus-fleets-maximize-fuel-economy-energy-efficient-cities- initiative • World Bank. 2012. Inclusion, Dignity, and Resilience: The Way Forward for Social Safety Net Reform in the Middle East and North Africa Region. Human Development Network. Washington, DC: World Bank. • www.worldbank.org/transport/roads/rdt_docsannex1.pdf (link not working) 41 Annex 1: Members of the PSIA Technical Team and Participants in Meetings Meeting 1: January 30, 2014 Technical team: • Idriss Abdillahi Orah (Ministry of Finance) • Almis Mohamed Abdillahi (Ministry of Budget) • Zeinab Ahmed Houssein (SESN) • Iltreh Osman Iltreh (SESN) • Idriss Ali Soultan (DISED) – substituting for Yacin Abdi Farid (DISED) • Mohamed Seif (IMF) – substituting for Abdourahman Aden (IMF) • Stefanie Brodmann (WB) • Harold Coulombe (WB) Observation: • Amina Warsama (SESN) • Members of the social registry team of the SESN Meeting 2: February 2, 2014 Technical team: • Idriss Abdillahi Orah (Ministry of Finance) • Mohamed Djibril Mahamoud (Ministry of Budget) – substituting for Almis Mohamed Abdillahi • Mouna Ahmed (SESN) – substituting for Zeinab Ahmed Houssein and Iltreh Osman Iltreh • Yacin Abdi Farid (DISED) • Abdourahman Aden (IMF) • Stefanie Brodmann (WB) • Harold Coulombe (WB) Meeting 3: May 25, 2014 Technical team: • Idriss Abdillahi Orah (Ministry of Finance) • Almis Mohamed Abdillahi (Ministry of Budget) • Houmed-Gaba Omar (Ministry of Energy) • Mouna Ahmed (SESN) – substituting for Zeinab Ahmed Houssein and Iltreh Osman Iltreh • Yacin Abdi Farid (DISED) • Stefanie Brodmann (WB) • Harold Coulombe (WB) Observation: • Amina Warsama (SESN) • Members of the social registry team of the SESN • Ines Rodriguez Caillava (WB) Meeting 4: May 28, 2014 Technical team: • Yacin Abdi Farid (DISED) • Harold Coulombe (WB) Observation: • Amina Warsama (SESN) • Statisticians (SESN) • Ines Rodriguez Caillava (WB) 42 Meeting 5: May 29, 2014 Technical team: • Mohamed Said Seif (IMF) • Omar Wahib Aref (Ministry of Transport) • Marta Dormal (Ministry of Finance) • Houmed-Gaba Omar (Ministry of Energy) • Simon Mibrat (Ministry of Budget) • Idriss Abdillahi Orah (Ministry of Finance) • Yacin Abdi Farid (DISED) • Fatouma Awaleh Osman (Ministry of Transport) • Stefanie Brodmann (WB) • Harold Coulombe (WB) Observation : • Amina Warsama (SESN) • Ines Rodriguez Caillava (WB) Meeting 6 : July 2, 2014 In Djibouti: • Simon Mibrathu (Ministry of Budget) • Idriss Abdillahi Orah (Ministry of Finance) • Yacin Abdi Farid (DISED) • Amina Warsama, Mouna Ahmed Ragueh, Zeinab Ahmed Houssein (SESN) • Abdourahman Aden (IMF) In Washington: • Stefanie Brodmann (WB) • Harold Coulombe (WB) • Paolo Verme (WB) • Robert Bacon (WB) • Ines Rodriguez Caillava (WB) Meeting 7 : November 15, 2014 • M. Almis Mohamed Abdillahi (Ministry of Budget) • M. Houmed-Gaba Omar (Ministry of Energy) • M. Omar Wahib Aref (Ministry of Transport) • Mme Amina Warsama (SESN) • Mme Mouna Ahmed (SESN) • M. Zeinab Ahmed Houssein (SESN) • Mme Stefanie Koettl-Brodmann (WB) 43 Annex 2. PMT Approach and its Efficiency Proxy Means Test9 - Djibouti Introduction Le Gouvernement de la République de Djibouti, ainsi que plusieurs organismes non- gouvernementaux, cherchent les meilleures politiques afin d’améliorer le bien-être de la population Djiboutienne en général, et des individus les plus pauvres et vulnérables en particulier. Les différents programmes actuellement en place utilisent différents critères de ciblages ce qui entraîne des coûts administratifs trop élevés et parfois des incohérences quant aux différentes populations ciblées. Une approche plus cohérente entre les différents programmes devrait permettre d’obtenir des programmes de lutte contre la pauvreté plus efficace et à un meilleur rapport coût-efficacité. Différentes approches de ciblage sont possible (auto-ciblage, ciblage géographique, ciblage catégorielle etc.) mais après concertation entre le gouvernement et ses différents partenaires il a été arrêté que l’approche dite « Proxy Means Test » (PMT) serait celle utilisée à Djibouti. L’approche PMT consiste à construire une mesure de bien-être des ménages à partir de la collecte d’indicateurs multiples qui sont plus faciles à observer que le revenu (consommation), mais qui sont fortement corrélés avec le revenu (consommation). Ces indicateurs multiples servent à établir une note (score) qui détermine si la famille devrait recevoir une aide ou pas. La formule PMT inclura en général des caractéristiques des ménages comme sa taille et sa composition, la qualité de son logement, la propriété des biens de consommation durables ou bien le niveau d’instruction des membres du ménage. La présente note technique fait suite à une mission de la Banque mondiale qui a eu lieu à Djibouti- ville du 10 au 19 avril 2013 et mise à jour lors d’une dernière mission en mai 2014. La mission a reçu la collaboration étroite de la Direction de la statistique et des études démographiques (DISED) et du Secrétariat d’état à la solidarité nationale (SESN). L’objectif principal consistait à estimer cette fonction PMT qui devrait être utilisée dans le cadre du projet de Registre unifié des pauvres. Données La troisième Enquête Djiboutienne auprès des ménages (EDAM 3) a été conduite en 2012 par la DISED. Cette enquête est à la base du dernier profil de pauvreté qui est en cours de finalisation par la DISED. L’EDAM 3 a un échantillon représentatif de la population nationale sédentaire de 31 686 individus répartis au sein de 5 880 ménages. Le questionnaire de l’EDAM 3 couvre une multitude de facettes socio-économiques des ménages : démographe, éducation, emploi, mortalité, gouvernance, logement, accès aux services sociaux de base, possessions du ménage, dépenses et revenus. A partir des données de cette enquête, un agrégat des dépenses totales des ménages par habitant a été calculé et a été à la base du profil de pauvreté. Les analystes de la DISED et de la Banque mondiale ont trouvé des taux de pauvreté de 16,9% (pauvreté extrême) et 33,7% (pauvreté relative)10. Par contre, l’EDAM3 ne couvre ni la population nomade ou la population dite « flottante », i.e. les sans-abri. Selon le recensement général de la population et de l’habitat (RGPH) de 2009, sur une population totale de 818 159 personnes, la population nomade est de 161 132 et la population flottante est de 149 022. Afin de prendre en compte ces populations non-sédentaires dans notre analyse, nous avons complété l’EDAM3 par des données du dernier RGPH. Pour la population nomade, nous avons 9 L’expression Proxy Means Test est quelques fois traduite par contrôles indirects du niveau des ressources ou bien par test de revenu par approximation. Par contre, nous continuerons à utiliser Proxy Means Test (PMT) étant donné que les gens sont plus familiers avec ce vocable. 10 Voir Poverty and Gender Diagnostic Paper, 2014. 44 accès à un échantillon des données désagrégées du RGPH tandis que dans le cas de la population flottante (parfois appelée « particulière » ou « sans-abri ») nous avons uniquement un décompte de cette population. A partir de ces différentes sources de données nous avons construit un système de pondération qui tient en compte le poids relatif de ces trois différentes populations et de la croissance de la population depuis les collectes de données. Ainsi nos analyses se baseront sur une population totale de 827 857 selon la répartition suivante : • Sédentaire : 523 359 • Nomade : 173 585 • Sans-abri : 130 913 Méthodologie La méthodologie dite PMT consiste simplement à trouver une série de caractéristiques des ménages facilement observables et qui sont corrélées avec le niveau de dépenses des ménages. Cette relation est supposée linéaire et estimée par la méthode des moindres carrés ordinaires (MCO). A partir des coefficients obtenus, il est possible de calculer la valeur espérée des dépenses de ces mêmes ménages. En termes plus techniques, nous estimons l’équation (1) à partir des données sur les 5880 ménages de l’EDAM 311. représente la valeur réelle des dépenses des ménages i per capita (en log), représente les j variables indépendantes pour les ménages i, sont les coefficients à estimer pour les j variables , est la constante et est le terme d’erreur de l’équation à estimer. = + + (1) L’estimation de l’équation (1) par la méthode des MCO génèrera la constante et les coefficients . À partir de ceux-ci, le PMT score est simplement calculé comme la valeur espérée selon l’équation suivante : = = exp + (2) Notons que les coefficients estimés sont les poids qui seront utilisée lors du calcul des scores PMT et ainsi détermineront si un ménage sera bénéficiaire ou non. Un ménage sera considéré comme pauvre si est inférieure au seuil de pauvreté et non-pauvre dans le cas contraire. Similairement, un ménage sera considéré comme bénéficiaire si est inférieure au même seuil; et non-bénéficiaire si cette valeur est égale ou supérieure au seuil. Le seuil de pauvreté utilisé pourrait être un des deux seuils se retrouvant dans le dernier profil de pauvreté, mais pourrait aussi être un niveau arbitraire décidé par l’organisme de mise en place du programme de transfert ciblé. Dans la présente note, nous utiliserons une série de seuils définie par les différents déciles. Même si la valeur de se veut la meilleure estimation des dépenses il est indéniable que ces valeurs seront différentes. Ce type d’estimation n’ai jamais parfait. Le tableau 1 présente les différents cas possibles. Certains ménages seront considérés comme pauvres (non-pauvres) et bénéficiaires (non-bénéficiaires) et seront ainsi des exemples de ciblage réussi (S1 et S2 dans le tableau 1). Par contre, il y aura assurément des cas où des ménages pauvres seront considérés comme non-bénéficiaires et inversement, des ménages non-pauvres qui recevront des transferts. Dans le premier cas, nous parlerons d’erreur d’exclusion (erreur de type II) tandis que le deuxième cas sera considéré comme des erreurs d’inclusion (erreur de type I). A partir de cet ensemble de possibilités, nous définirons les concepts suivants : 11Étant donné que la relation (1) est estimée sur une mesure de dépenses des ménages, cette le calcul de l’équation PMT sera basé uniquement les données de l’EDAM3. Par contre, les coefficients servant au calcul de l’équation (2) seront appliqués sur toutes les données, autant de l’EDAM3 que du RGPH 2009. 45 • Taux de couverture (coverage) : M1/N • Taux de fuite (leakage) : E2/M1*100 • Taux de sous-couverture (undercoverage) : E1/N1*100 Ainsi, dans notre recherche de la meilleure fonction PMT, nous chercherons à minimiser à la fois le taux de fuite et le taux de sous-couverture. Autrement dit, nous chercherons à maximiser les cas de ciblage réussi. Tableau 1. Erreurs de Type I et II Pauvre Non pauvre Total Bénéficiaire Ciblage réussi (S1) Erreur d’inclusion (E2) M1 Non Bénéficiaire Erreur d’exclusion (E1) Ciblage réussi (S2) M2 Total N1 N2 N Résultats A partir du questionnaire de l’EDAM-3, nous avons défini une série de caractéristiques des ménages et estimé différents modèles PMT. Les variables utilisées sont définies au tableau A2 en annexe. A partir de ces variables, nous avons estimé une fonction PMT pour l’ensemble du pays.12 Les résultats de cette régression se trouvent au tableau A1 de l’annexe. Avec un R2 de 0,67, nous sommes heureux de constater que le pouvoir explicatif de ce modèle est très élevé et devrait ainsi permettre un meilleur ciblage. A partir de ce modèle estimé au niveau national, nous avons calculé les taux de couverture, de sous-couverture et de fuite pour une série de seuils de pauvreté définis comme les bornes supérieures des différents décile de la distribution de dépenses. Le Tableau 2 montre que si les autorités décident que les bénéficiaires seront la population sous la médiane des revenus selon la mesure de bien-être , la formule PMT couvrira 51,7% de la population avec une sous-couverture de 15,4% et une fuite de 18,1%. Cela permettra un ciblage réussi de 66,4%. Ces taux de réussite se comparent très avantageusement avec les différentes expériences de PMT ailleurs dans les autres pays en développement. Il est à noter que lorsque le seuil de pauvreté (i.e. le décile) utilisé augmente, plus faibles seront les erreurs de ciblage. Tableau 2. Taux de couverture, fuite et sous-couverture, modèle national unique Sous- Efficacité du Décile Couverture couverture Fuite ciblage 1 9,1 42,6 36,3 21,1 2 22,5 28,8 36,7 34,4 3 33,0 24,7 31,5 43,8 4 43,7 18,1 25,0 56,9 5 51,7 15,4 18,1 66,4 6 62,8 12,4 16,3 71,3 7 71,5 9,7 11,6 78,7 8 81,3 6,0 7,4 86,6 9 91,4 3,1 4,5 92,4 12 Au lieu d’utiliser un modèle de régression unique pour l’ensemble du pays, nous avons aussi estimé trois modèles distincts selon le milieu de résidence : Djibouti-ville, autres zones urbaines et zones rurales. Il est possible de montrer que les résultats sont très semblables. Étant donné que l’utilisation de trois systèmes de pondérations différents rendrait la logistique beaucoup plus lourde sans une amélioration significative du ciblage, nous suggérons fortement l’utilisation d’un modèle unique. 46 La ventilation de nos résultats selon le milieu et la région de résidence (Tableau 36) montre que la couverture d’un programme ciblé sera beaucoup plus grande en milieu rural qu’en milieu urbain. Si on suppose que 20 ou 40 pourcent (i.e. le premier ou le deuxième quintile) de la population nationale serait ciblé, la couverture serait presque complète (respectivement 80 et 94 pourcent) en milieu rural tandis qu’elle serait de 11 et 34 pourcent en milieu urbain. En particulier les cas de fuite (i.e. les ménages bénéficiaires qui n’auraient pas dû l’être) serait minimisé Tableau 3. Taux de couverture, fuite et sous-couverture, selon le milieu et la région de résidence Quintile 1 Quintile 2 Quintile 3 Quintile 4 Couv. Sous Fuite Couv. Sous Fuite Couv. Sous Fuite Couv. Sous Fuite Milieu Urbain 11,4 56,5 58,9 34,1 27,0 33,4 56,1 16,3 20,3 77,8 7,4 8,8 Rural 80,3 5,9 20,4 94,0 1,4 9,2 97,9 1,0 4,7 99,5 0,2 1,8 Région Djibouti 8,9 67,0 67,8 30,4 30,3 35,0 52,5 18,3 20,7 75,3 8,4 9,0 Ali Sabieh 46,4 22,1 17,8 72,6 8,6 10,2 86,0 5,6 6,9 95,7 2,0 3,2 Dikhil 60,1 10,7 18,5 81,9 5,4 10,2 92,6 3,0 7,0 98,4 0,4 2,2 Tadjourah 66,6 6,6 28,7 82,8 2,3 19,5 92,5 0,5 12,4 97,8 0,4 5,3 Obock 71,2 5,7 21,1 80,9 4,0 8,7 91,9 2,7 5,6 98,0 0,9 3,2 Arta 55,6 8,5 34,1 81,8 4,0 23,0 92,1 2,5 14,7 98,5 0,8 7,2 Note: the three columns below “Quintile 1” suppose that the upper limit of the poorest quintile is the destitution line. Similarly, the other groups of columns suppose that the respective upper limits of the different quintiles are the destitution lines. Questionnaire du recensement à venir Étant donné que cette fonction PMT a été construite à partir des données de l’enquête EDAM3, il est nécessaire que la formulation des questions dans le recensement nécessaire à la construction du Registre unifié des pauvres soit la même que dans l’enquête EDAM3. Les questions nécessaires sont les suivantes: • Région de résidence (question IM01 dans le questionnaire EDAM3) • Age et sexe des membres du ménage (I02 & I04) • Éducation du chef de ménage (E05) • Éducation de la conjointe du chef (E05) • État matrimonial du chef (E07) • Matériaux des murs (CL02) • Matériaux du toit (CL03) • Matériaux du plancher (CL04) • Source d’énergie pour l’éclairage (CL07) • Principale source d’eau (CL09) • Type de toilettes (CL14) • Possession d’un téléphone portable (P03) • Possession d’un poste radio (P05) • Possession d’un téléviseur (P06) • Possession d’un réfrigérateur (P09) • Possession de camelins (P31A) • Possession de bovins (P31B) 47 Conclusion L’efficacité du ciblage dépend beaucoup du pouvoir explicatif de la fonction PMT. Avec le R2 très élevé obtenu pour la régression multivariée à la base de la fonction PMT (Annexe A1), il est attendu que nous avons un haut taux de précision qui nous donne espoir qu’un ciblage basé sur cette spécification sera efficace. Linear regression Number of obs = 5880 F( 34, 275) = 139.28 Prob > F = 0.0000 R-squared = 0.6747 Root MSE = .60739 (Std. Err. adjusted for 276 clusters in grappe) | Robust lnw | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- r_dji | .248529 .0354774 7.01 0.000 .1786871 .3183708 toil_trou | .3888291 .0485081 8.02 0.000 .2933347 .4843236 toil_latsim | .4532654 .0463145 9.79 0.000 .3620894 .5444415 toil_latame | .4584193 .0524071 8.75 0.000 .3552493 .5615893 toil_wc | .620787 .0532736 11.65 0.000 .5159111 .7256629 taille | -.2774993 .0138618 -20.02 0.000 -.3047881 -.2502105 taille2 | .0088266 .0008325 10.60 0.000 .0071877 .0104656 radio | .1185628 .0191032 6.21 0.000 .0809557 .15617 garcon610 | .0304748 .0133286 2.29 0.023 .0042357 .0567139 frigo | .187904 .0227822 8.25 0.000 .1430543 .2327536 garcon1117 | .0366247 .0136504 2.68 0.008 .0097522 .0634973 fille1117 | .0408917 .0132266 3.09 0.002 .0148534 .06693 homme1860 | .0354592 .0104893 3.38 0.001 .0148097 .0561088 femme1860 | .0448654 .0120357 3.73 0.000 .0211715 .0685592 electricite | .1355706 .0322414 4.20 0.000 .0720994 .1990418 portable | .2716365 .0274699 9.89 0.000 .2175584 .3257145 chef_prima~e | .1295366 .0245002 5.29 0.000 .0813047 .1777685 chef_college | .1424231 .0267136 5.33 0.000 .0898341 .1950122 chef_lycee | .191069 .0305344 6.26 0.000 .1309581 .2511799 chef_postbac | .3037089 .0402782 7.54 0.000 .2244162 .3830016 chef_veuf_~v | -.1126706 .0231306 -4.87 0.000 -.1582062 -.0671349 chef_age3549 | .0642248 .026373 2.44 0.016 .0123062 .1161434 chef_age5099 | .0877339 .0298676 2.94 0.004 .0289357 .1465322 conjointe_~e | .0524683 .0210194 2.50 0.013 .011089 .0938476 eau_courint | .3611416 .04666 7.74 0.000 .2692854 .4529977 mur_adobe | .0655207 .0213075 3.08 0.002 .0235742 .1074672 eau_courext | .1991182 .0419228 4.75 0.000 .1165878 .2816486 eau_fontaine | .1639344 .0487089 3.37 0.001 .0680446 .2598241 tv | .0744375 .0299037 2.49 0.013 .0155683 .1333067 toit_tole | .2301234 .0502317 4.58 0.000 .1312359 .329011 toit_bois | .2167163 .0458935 4.72 0.000 .1263691 .3070635 toit_beton | .3165959 .0568208 5.57 0.000 .2047369 .428455 sol_ciment | .1528517 .0268133 5.70 0.000 .1000664 .205637 sol_carrel | .3999771 .0355944 11.24 0.000 .329905 .4700493 _cons | 11.40121 .0529656 215.26 0.000 11.29694 11.50548 48 Variables utilisées durant la formulation de la fonction PMT, avec moyenne par ménage, par type de population Nom des Codificatio Sédentair Sans- Nomade Description des variables variables n e abris Ménage réside à Djibouti-ville r_dji 0/1 0.708 0.000 1 Ménage réside dans la région d’Ali Sabieh r_ali 0/1 0.059 0.219 0 Ménage réside dans la région de Dikhil r_dik 0/1 0.074 0.107 0 Ménage réside dans la région de Tadjourah r_tad 0/1 0.084 0.420 0 Ménage réside dans la région d’Obock r_obo 0/1 0.034 0.196 0 Ménage réside dans la région d’Arta r_art 0/1 0.042 0.058 0 Taille du ménage taille en nombre 5.624 5.566 1 Taille au carré taille2 en nombre 38.950 38.240 1 Nombre d’enfants âgés entre 0 et 5 enfant05 en nombre 0.905 0.747 0 Nombre de garçons âgés entre 6 et 10 garcon610 en nombre 0.429 0.389 0 Nombre de filles âgées entre 6 et 10 fille610 en nombre 0.369 0.357 0 Nombre de garçons âgés entre 11 et 17 garcon1117 en nombre 0.407 0.405 0 Nombre de filles âgées entre 11 et 17 fille1117 en nombre 0.395 0.348 0 Nombre d'hommes âgés entre 18 et 60 homme1860 en nombre 1.391 1.499 0 Nombre de femmes âgées entre 18 et 60 femme1860 en nombre 1.492 1.553 0 Nombre d'hommes âgés de 61 ou plus homme61plus en nombre 0.107 0.155 0 Nombre de femmes âgées 61 ou plus femme61plus en nombre 0.130 0.113 0 Chef du ménage est une femme chef_femme 0/1 0.220 0.168 0 Chef du ménage n'est pas allé à l'école chef_noneduc 0/1 0.618 0.922 1 Chef du ménage est allé école primaire chef_primaire 0/1 0.123 0.004 0 Chef du ménage est allé au collège chef_college 0/1 0.111 0.055 0 Chef du ménage est allé au Lycée chef_lycee 0/1 0.082 0.020 0 Chef du ménage est allé à l’université chef_postbac 0/1 0.066 0.000 0 (BAC+3/4) Chef du ménage est veuf ou divorcé chef_veuf_div 0/1 0.169 0.137 0 Chef du ménage est âgé entre 15 et 34 chef_age1534 0/1 0.187 0.193 0 Chef du ménage est âgé entre 35 et 49 chef_age3549 0/1 0.438 0.337 1 Chef du ménage est âgé de 50 ou plus chef_age5099 0/1 0.375 0.470 0 conjointe_scol Conjointe du chef est allée à l’école 0/1 0.197 0.035 0 arisee Pas de conjointe conjointe_non 0/1 0.242 0.208 0 Type de matériaux des murs: adobe mur_adobe 0/1 0.237 0.000 0 Type de matériaux des murs: brique mur_brique 0/1 0.165 0.000 0 Type de matériaux des murs: tôle mur_tole 0/1 0.346 0.000 0 Type de matériaux des murs: pierre mur_pierre 0/1 0.051 0.000 0 Type de matériaux des murs: autres mur_autre 0/1 0.201 1.000 1 Type de matériaux de la toiture: tôle toit_tole 0/1 0.612 0.000 0 Type de matériaux de la toiture: bois toit_bois 0/1 0.168 0.000 0 Type de matériaux de la toiture: béton toit_beton 0/1 0.096 0.000 0 49 Type de matériaux de la toiture: paille toit_paille 0/1 0.124 1.000 1 Type de matériaux du sol: ciment sol_ciment 0/1 0.425 0.000 0 Type de matériaux du sol: carrelage sol_carrel 0/1 0.140 0.000 0 Type de matériaux du sol: terre sol_terre 0/1 0.435 1.000 1 Ménage utilise l’électricité electricite 0/1 0.529 0.000 0 Source d’eau potable: courante, intérieur eau_courint 0/1 0.263 0.000 0 Source d’eau potable: courante, extérieur eau_courext 0/1 0.409 0.000 0 Source d’eau potable: fontaine publique eau_fontaine 0/1 0.136 0.000 0 Source d’eau potable: forage eau_forage 0/1 0.037 0.000 0 Source d’eau potable: autres eau_autre 0/1 0.155 1.000 1 Type de toilette: WC avec chasse d’eau toil_wc 0/1 0.107 0.000 0 Type de toilette: latrine améliorée toil_latame 0/1 0.113 0.000 0 Type de toilette: latrine simple toil_latsim 0/1 0.380 0.000 0 Type de toilette: trou avec cloture toil_trou 0/1 0.247 0.000 0 Type de toilette: nature toil_nature 0/1 0.153 1.000 1 Ménage possède un portable portable 0/1 0.599 0.070 0 Ménage possède un radio radio 0/1 0.351 0.677 0 Ménage possède un télévisueur tv 0/1 0.422 0.009 0 Ménage possède un réfrigérateur frigo 0/1 0.329 0.000 0 Ménage possède un réchaud rechaud 0/1 0.293 0.000 0 Ménage possède des camelins camelin 0/1 0.046 0.000 0 Ménage possède des bovins bovin 0/1 0.034 0.000 0 Ménage possède des ovins ovin 0/1 0.147 0.000 0 50 Annex 3. Population Demographics Table 1. Socioeconomic Characteristics of the Sedentary Population (in %) Looking for jobs Handicapped School 14-18 School 19-25 Independent School 6-13 Menial job Literacy School Work Wage Quintile Poorest 1.9 56.0 42.7 13.9 23.2 23.1 15.9 36.0 33.9 29.7 36.5 Second 1.4 74.3 66.9 25.2 43.6 44.4 23.2 31.4 47.9 23.6 28.5 Third 1.3 81.6 71.4 29.3 50.3 51.7 26.2 32.6 57.1 22.4 20.5 Fourth 1.4 83.1 82.8 34.5 61.1 62.0 29.7 28.3 70.0 19.3 10.7 Richest 1.1 85.2 84.4 38.5 71.3 71.6 41.8 22.4 81.1 13.5 5.3 Region Djibouti 1.4 78.6 75.1 31.9 57.4 58.5 30.9 28.3 67.1 17.9 15.0 Ali Sabieh 1.3 73.9 65.2 24.0 41.3 40.2 23.9 29.3 47.8 32.2 19.9 Dikhil 0.9 63.7 62.8 25.0 35.2 33.1 16.9 29.4 54.9 30.2 14.9 Tadjourah 2.4 62.5 42.1 11.4 29.8 29.9 17.3 39.0 58.0 17.8 24.2 Obock 1.1 66.0 54.4 11.6 21.1 22.9 22.7 37.2 49.8 31.6 18.5 Arta 1.3 58.6 49.2 16.4 35.1 34.4 23.9 37.6 52.3 26.4 21.2 Area Urban 1.4 78.4 75.7 32.0 57.1 57.8 30.0 28.5 66.0 18.8 15.2 Rural 1.5 56.2 28.2 5.9 17.8 18.1 17.6 37.1 50.0 27.0 22.9 Gender of household head Male 1.3 75.3 72.5 30.6 52.8 53.3 29.4 27.5 66.5 17.0 16.5 Female 1.7 70.8 59.4 23.6 46.9 48.2 23.6 38.1 55.4 31.8 12.8 Total 1.4 74.7 70.1 29.0 51.6 52.2 28.2 29.7 64.6 19.5 15.8 51 Annex 4: Survey of Variable Costs OF Passenger Road Transportation A survey of passenger road transportation vehicles was carried out in Djibouti-ville during April 2014 with the objective of providing a breakdown of variable costs of running these vehicles. The share of fuel costs in variable costs was required to estimate the passing on of an increase in diesel prices that would be brought about by the removal of the discretionary tax element in these prices. The survey was designed and administered by the DISED. Fifty vehicles were surveyed, of which 26 respondents were the owners (propriétaire) and 24 were employees (employé). The vehicles were of four types: 18 buses, 12 mini-buses, 10 taxis (berline), and 10 land cruisers. The average age of these vehicles was 5.3 years. Urban routes were serviced by 58 percent of these vehicles and various interurban routes by the other vehicles. Only one vehicle used gasoline (super) as a fuel, while all the rest used diesel (gasoil). The average daily fuel consumption was 54 liters. A number of questions in the survey related to variable costs based over different time intervals. Fuel costs were given daily, maintenance costs (cleaning, oil change, tire repair, lights, battery, and small repairs) over a 3-month period, and major replacements and repairs (purchase of tires, batteries) were given on an annual basis. Other annual costs included license fee (vignette), insurance, and other taxes. All these have been converted into daily equivalents and are shown in Table 1. Fuel costs are the major component of variable costs accounting for 80 percent of all daily costs. Table 1. Daily Cost of Passenger Transportation Vehicles (in DF) Cost Element DF per Day Fuel 11,582 Maintenance and small repairs 1,378 Replacement parts and large repairs 776 License fee 87 Insurance 430 Taxes and other costs 127 Total 14,380 Share of fuel in total variable costs 80% Source: World Bank calculation based on the EDAM 3. The shares of fuel costs by type of transportation are shown in Table 2, indicating that fuel costs were most important for land cruisers and least important for taxis. However, the household expenditure survey does not indicate the shares of expenditure to various types of passenger transportation, so the fact that the proportions are similar for all classes of vehicle indicates that using the overall fuel share of 80 percent would be a reasonable estimate for all classes of vehicle. Table 2. Shares of Fuel in Variable Costs by Vehicle Type Vehicle Type Share of Fuel Costs (%) Bus 78 Mini-bus 81 Land cruiser 85 Taxi 73 Source: World Bank calculation based on the EDAM 3. Finally, the survey includes information on average daily receipts, distinguishing between periods when schools were open and periods when schools were closed. Table 3Error! Reference source not 52 found. indicates that there is a substantial difference in the average daily receipts between days when schools are open and days when schools are closed. The average daily receipts of DF 18,254 can be compared to the average daily variable cost of DF 14,380. The difference between the two has to cover labor costs, depreciation and interest costs on the vehicles, and the entrepreneur’s margin. Table 3. Average Daily Receipts of Passenger Transportation Vehicles Period DF per Day All days 18,524 Days with schools open 21,044 Days with schools closed 16,904 Source: World Bank calculation based on the EDAM 3. 53 Annexe 5. Social Assistance Project « Bourse Familiale pour les ménages pauvres et vulnérables des régions de l'intérieur » Table 1. Distribution of Vulnerable Populations by Quintile (Percentage) Age60 Age70 Handicapped Orphan Quintile Poorest 4.6 1.8 0.0 0.5 Second 3.7 1.4 0.1 0.9 Third 3.8 1.4 0.1 1.0 Fourth 4.2 1.4 0.1 0.9 Richest 4.6 1.4 0.1 0.8 Region Djibouti 3.6 1.2 0.1 0.8 Ali Sabieh 5.6 2.4 0.1 1.3 Dikhil 4.7 1.5 0.0 0.5 Tadjourah 6.6 2.4 0.1 0.8 Obock 5.5 1.9 0.1 0.9 Arta 5.4 2.4 0.2 0.7 Area Urban 3.8 1.3 0.1 0.9 Rural 5.8 2.2 0.1 0.5 Gender of household head Male 3.7 1.3 0.1 0.6 Female 6.6 2.6 0.1 2.2 Total 4.2 1.5 0.1 0.8 # of cases 6791 2418 133 521 Notes: All variables are in % and are defined based on the 2009 Census. Age60: proportion of the population aged 60 years and above. Age70: proportion of the population aged 70 years and above. Handicapped: proportion of the population declaring a ‘handicap majeur’. Orphans: proportion of the children aged 17 years and below with both parents dead. 54 Annex 6. Capacity to Respond to Shocks Table 1. Percentage of Households that Faced a Shock in the Last 12 Months Diseases/Medical expenses Loss of livestock related to Average number of shocks Bad harvest non-related to Bad quality/irregularity of Job loss/wage decrease Rent payment/increase Loss of livestock non- Bad harvest related to Electricity/gas outage Fuel/transport prices High price of food related to drought Death of a family member/funerals Debt repayment Insecurity/theft drinking water Another shock No shock drought drought drought Fire Quintile Poorest 1.55 5.3 4.4 5.2 56.6 8.1 1.4 6.3 8.3 1.8 1.9 14.3 29.7 5.9 4.5 1.1 2.1 0.5 Second 0.96 4.8 4.8 5.2 44.3 6.4 0.9 3.4 4.1 1.9 1.3 3.1 7.3 1.0 0.6 0.0 0.5 0.5 Third 0.73 4.3 3.8 4.7 36.4 6.3 1.5 3.7 3.0 2.7 0.4 1.2 3.7 0.4 0.6 0.3 0.6 0.3 Fourth 0.71 3.6 3.3 4.6 37.5 6.9 1.9 3.6 2.3 2.6 0.7 1.3 1.7 0.3 0.3 0.2 0.1 0.0 Richest 0.67 2.3 5.2 4.8 31.0 8.2 2.7 3.4 1.8 2.9 0.9 1.9 0.9 0.6 0.7 0.2 0.0 0.1 Region Djibouti 0.66 3.9 4.1 4.6 33.1 6.3 1.7 3.6 2.7 2.5 0.8 0.8 0.3 0.4 0.3 0.2 0.2 0.2 Ali Sabieh 0.65 1.0 2.8 3.3 39.7 1.2 0.3 3.2 1.9 2.2 0.5 1.2 7.1 0.5 0.1 0.0 0.0 0.0 Dikhil 1.6 8.6 3.1 4.1 63.4 21.9 3.3 3.6 8.0 2.4 0.8 10.2 20.1 5.2 3.7 0.3 1.1 0.0 Tadjourah 1.73 1.2 3.3 9.3 67.4 5.2 0.6 2.9 8.9 1.5 1.7 13.4 45.8 4.6 4.7 1.2 1.2 1.4 Obock 3.3 10.6 19.1 7.9 87.7 15.8 5.5 21.0 13.5 3.7 7.9 48.4 53.2 14.1 12.3 1.8 8.0 0.4 Arta 0.7 1.2 2.7 1.8 37.0 3.6 1.4 3.3 0.4 1.7 0.4 2.4 13.3 1.0 0.2 0.0 0.0 0.0 Area Urban 0.7 3.9 4.6 4.8 35.0 6.5 2.0 3.9 2.6 2.7 0.9 0.9 0.6 0.4 0.4 0.2 0.3 0.2 Rural 1.79 3.9 3.3 5.2 64.5 10.4 0.6 4.6 8.9 1.1 1.8 18.5 41.1 6.8 5.4 0.9 1.8 0.6 55 head Total Male Third Quintile Fourth Female Second Richest Poorest Gender of household 2.59 2.99 3.17 4.65 7.49 Total (all households) 0.91 0.98 0.88 Average number of shocks Total (households that faced 70.6 73.7 64.11 81.62 81.08 a shock) 3.9 3.8 4.0 Job loss/wage decrease 1.6 2.8 3.4 4.2 3.8 Job loss/wage decrease 4.3 5.0 4.2 Diseases/Medical expenses 3.4 2.3 3.3 4.1 3.5 Diseases/Medical expenses Death of a family 4.9 8.3 4.0 Death of a family member/funerals 3.2 2.5 3.7 3.7 3.7 member/funerals High price of food 40.6 41.4 40.4 21.2 27.0 26.3 37.0 46.7 High price of food 7.3 5.7 7.7 Fuel/transport prices 4.5 4.1 4.1 4.1 5.3 Fuel/transport prices 1.8 1.8 1.7 Rent payment/increase 2.1 1.7 1.0 0.8 1.2 Rent payment/increase 4.1 4.0 4.1 56 Debt repayment 2.9 2.8 3.4 2.4 5.5 Debt repayment Bad quality/irregularity of 3.8 4.2 3.7 Bad quality/irregularity of drinking water 1.3 1.9 2.3 3.1 6.6 drinking water Table 2. Percentage of Households that Faced a Loss after a Shock in the Last 12 Months 2.4 2.3 2.4 Electricity/gas outage 1.7 1.6 1.5 1.5 1.6 Electricity/gas outage 1.0 1.4 1.0 Insecurity/theft 0.6 0.6 0.1 0.9 1.6 Insecurity/theft Loss of livestock non- 4.3 4.7 4.1 Loss of livestock non-related related to drought 1.1 0.8 0.9 2.6 12.5 to drought Loss of livestock related to 8.3 7.7 10.3 Loss of livestock related to drought 0.5 1.4 2.7 6.1 25.6 drought Bad harvest non-related to 1.6 2.3 1.4 Bad harvest non-related to drought 0.3 0.1 0.2 0.9 4.6 drought Bad harvest related to 1.3 1.8 1.2 Bad harvest related to drought 0.3 0.2 0.4 0.5 3.5 drought Fire 0.4 0.7 0.2 0.2 0.0 0.3 0.0 1.0 Fire Another shock 0.6 0.9 0.5 0.0 0.1 0.5 0.4 1.4 Another shock No shock 0.3 0.8 0.1 0.1 0.0 0.0 0.2 0.8 No shock Loss of livestock non-related Total (households that faced Diseases/Medical expenses Loss of livestock related to Bad harvest non-related to Bad quality/irregularity of Job loss/wage decrease Rent payment/increase Total (all households) Bad harvest related to Electricity/gas outage Fuel/transport prices High price of food Death of a family member/funerals Debt repayment Insecurity/theft drinking water Another shock to drought No shock a shock) drought drought drought Fire Region 3.3 3.3 3.3 24.9 4.3 1.4 3.1 2.2 1.9 0.5 0.5 0.2 0.3 0.2 0.2 0.2 0.2 Djibouti 2.94 73.77 3.3 3.3 3.3 24.9 4.3 1.4 3.1 2.2 1.9 0.5 0.5 0.2 0.3 0.2 0.2 0.2 0.2 Ali Sabieh 3.16 86.38 0.9 1.1 1.4 38.3 0.4 0.3 2.9 0.1 0.3 0.2 1.0 6.2 0.5 0.1 0.0 0.0 0.0 Dikhil 5.35 54.94 4.6 2.3 2.1 36.1 9.0 2.2 2.2 4.9 1.2 0.6 7.7 12.8 2.8 1.4 0.0 0.9 0.0 Tadjourah 9.08 85.79 0.7 2.2 6.3 58.6 2.9 0.2 2.3 8.7 0.9 1.7 12.7 44.7 4.2 4.6 1.2 1.2 1.2 Obock 14.91 77.29 7.5 15.1 5.9 68.5 10.9 4.7 17.0 8.6 1.8 4.9 39.3 42.4 11.6 10.0 1.3 3.6 0.4 Arta 3.09 77.17 1.0 1.9 0.8 29.4 3.1 0.4 2.4 0.3 0.7 0.4 2.2 9.0 0.7 0.2 0.0 0.0 0.0 Area Urban 3.07 73.15 3.3 3.5 3.3 26.3 4.3 1.6 3.4 2.1 1.8 0.6 0.6 0.3 0.2 0.2 0.2 0.2 0.1 Rural 8.18 77.54 2.0 2.4 3.7 50.7 4.9 0.4 3.6 6.5 0.8 1.4 15.7 35.1 5.2 4.2 0.7 1.3 0.6 Gender of household head Male 3.85 72.9 3.1 3.2 2.4 30.2 4.6 1.3 3.3 2.8 1.5 0.7 3.4 6.4 1.0 0.9 0.2 0.4 0.1 Female 4.75 79.41 2.9 3.8 6.5 33.6 4.0 1.8 3.7 3.5 1.8 1.1 3.7 9.2 1.8 1.3 0.7 0.7 0.6 Total 4.05 74.43 3.1 3.3 3.3 31.0 4.4 1.4 3.4 2.9 1.6 0.8 3.5 7.0 1.2 1.0 0.3 0.4 0.2 57 Table 3. Percentage of Households that Did Not Have Enough Food after a Shock in the Last 12 Months Loss of livestock non-related Total (households that faced Diseases/Medical expenses Loss of livestock related to Bad harvest non-related to Bad quality/irregularity of Job loss/wage decrease Rent payment/increase Total (all households) Bad harvest related to Electricity/gas outage Fuel/transport prices High price of food Death of a family member/funerals Debt repayment Insecurity/theft drinking water Another shock to drought No shock a shock) drought drought drought Fire Quintile Poorest 7.64 83.43 4.0 3.4 3.7 50.5 6.4 1.3 5.9 6.2 1.5 1.6 12.1 25.0 4.1 3.2 1.0 1.5 1.0 Second 4.81 85.61 4.4 3.9 3.6 39.4 4.2 0.8 2.7 3.1 1.4 0.8 2.3 5.5 0.9 0.6 0.0 0.4 0.2 Third 3.58 83.82 3.7 2.9 3.6 32.1 4.6 1.1 3.3 2.0 1.5 0.3 0.7 2.7 0.2 0.4 0.3 0.6 0.0 Fourth 3.03 73.92 3.1 1.9 2.3 31.4 5.4 1.8 3.1 1.5 1.4 0.3 0.5 1.3 0.3 0.2 0.0 0.0 0.0 Richest 2.67 68.26 1.5 3.1 2.6 23.2 4.4 1.7 2.4 0.7 1.7 0.4 1.3 0.4 0.3 0.3 0.2 0.0 0.1 Region Djibouti 3.12 80 3.3 2.9 3.1 28.5 4.9 1.4 3.0 1.9 1.8 0.4 0.6 0.2 0.3 0.2 0.2 0.2 0.2 Ali Sabieh 3.13 85.85 0.8 0.8 1.4 37.8 0.3 0.2 3.2 0.6 0.0 0.2 1.0 6.2 0.5 0.1 0.0 0.0 0.0 Dikhil 5.39 59.66 5.6 2.0 2.1 46.1 8.5 2.0 2.6 3.8 0.8 0.4 5.0 8.8 2.0 1.2 0.0 0.7 0.0 Tadjourah 9.2 86.13 1.1 1.9 5.9 58.6 4.3 0.5 2.3 8.5 0.9 1.7 13.1 44.9 4.4 4.6 1.2 1.2 1.2 Obock 15.41 80.26 7.4 15.5 4.8 77.0 12.6 5.1 18.8 7.4 1.6 5.2 39.3 41.7 10.2 8.7 1.2 5.0 0.4 Arta 3.13 79.35 1.0 2.3 0.6 31.5 2.2 0.6 2.5 0.1 0.8 0.2 2.1 8.5 0.7 0.2 0.0 0.0 0.0 Area Urban 3.23 78.3 3.3 3.1 3.0 29.7 4.7 1.6 3.3 1.8 1.7 0.5 0.6 0.3 0.3 0.2 0.2 0.3 0.2 Rural 8.38 80.45 2.6 2.4 3.5 55.5 6.1 0.4 4.0 6.2 0.6 1.4 14.8 33.6 4.8 3.8 0.7 1.4 0.6 58 Total Male Third Quintile Fourth Female Second Poorest Gender of household head 4.21 4.89 4.02 Percentage of households Total (all households) that could not do anything to 51 Total (households that faced 38.62 49.46 59.32 compensate the effect of 77.8 78.93 82.61 shocks a shock) 3.2 2.8 3.3 Job loss/wage decrease 44.6 37.9 45.0 52.5 Job loss/wage decrease 3.0 3.7 2.8 Diseases/Medical expenses 47.3 40.3 32.6 46.8 Diseases/Medical expenses Death of a family Death of a family 3.1 6.0 2.3 45.4 39.8 46.6 56.9 member/funerals member/funerals 34.6 36.2 34.2 High price of food 37.1 43.9 49.0 41.6 High price of food 5.0 4.8 5.0 Fuel/transport prices 48.2 62.1 42.1 49.3 Fuel/transport prices 1.4 1.5 1.3 Rent payment/increase 33.4 61.0 56.9 46.6 Rent payment/increase 59 3.4 3.5 3.4 Debt repayment 21.1 31.6 29.3 31.8 Debt repayment Bad quality/irregularity of 2.6 3.5 2.4 Bad quality/irregularity of drinking water 52.1 68.1 79.9 88.7 drinking water 1.5 1.7 1.4 Electricity/gas outage 66.8 59.9 58.5 53.2 Electricity/gas outage 0.7 1.1 0.6 60.0 Insecurity/theft 38.5 67.1 46.2 Insecurity/theft Loss of livestock non-related 3.3 3.9 3.2 Loss of livestock non-related to drought 76.1 76.5 87.5 64.3 to drought Loss of livestock related to 6.7 9.2 6.0 Loss of livestock related to drought 86.6 69.5 79.3 70.6 drought Bad harvest non-related to 1.1 1.8 0.9 Bad harvest non-related to drought 12.3 65.6 49.2 57.7 drought Table 4. Percentage of Households that Could Not Do Anything to Compensate the Effect of a Shock in the Last 12 Months Bad harvest related to 0.9 1.3 0.8 Bad harvest related to drought 63.5 59.8 34.0 57.5 drought 0.3 0.6 0.2 Fire 0.0 0.0 76.4 46.6 Fire 0.5 0.7 0.4 Another shock 40.3 41.1 39.7 61.5 Another shock 0.3 0.8 0.1 No shock 0.0 45.5 No shock 100.0 Loss of livestock non-related that could not do anything to Diseases/Medical expenses Loss of livestock related to Bad harvest non-related to Bad quality/irregularity of Percentage of households compensate the effect of Job loss/wage decrease Rent payment/increase Bad harvest related to Electricity/gas outage Fuel/transport prices High price of food Death of a family member/funerals Debt repayment Insecurity/theft drinking water Another shock to drought No shock drought drought drought shocks Fire Richest 36.34 58.2 46.3 37.1 33.7 36.3 40.1 25.6 46.5 68.0 51.9 42.0 40.3 57.5 26.1 0.0 0.0 0.0 Region Djibouti 39.2 41.8 38.5 36.9 40.1 33.5 33.4 21.0 57.1 57.9 37.7 39.4 27.6 27.8 10.9 0.0 0.0 0.0 Ali Sabieh 66.33 73.5 65.5 75.7 61.7 63.1 47.8 69.8 100.0 97.8 66.5 43.2 68.5 100.0 100.0 Dikhil 57.48 58.0 55.7 63.6 48.0 74.8 79.6 25.8 91.8 70.9 49.9 48.8 56.8 45.7 32.5 64.4 21.6 Tadjourah 68.65 76.7 66.6 68.9 44.2 77.2 100.0 60.6 96.8 67.4 68.8 89.4 88.0 87.1 94.3 100.0 81.7 100.0 Obock 40.32 54.9 39.4 30.0 12.5 37.3 45.5 22.7 81.3 80.4 72.6 68.5 50.5 46.5 47.1 50.6 85.1 100.0 Arta 54.68 85.4 63.3 46.9 44.4 54.0 63.1 77.2 100.0 57.7 100.0 65.6 86.7 100.0 0.0 Area Urban 42.15 45.2 41.8 40.9 42.0 37.8 43.2 27.0 60.1 62.2 47.5 44.8 42.9 30.5 18.4 0.0 31.0 3.9 Rural 60.32 57.2 50.4 61.1 38.8 68.9 61.0 32.7 92.5 66.8 64.8 70.4 73.4 61.3 61.4 81.6 71.7 100.0 Gender of household head Male 48.54 46.1 46.4 49.7 42.8 47.7 46.8 28.2 79.3 64.9 61.5 68.1 71.0 56.0 53.7 51.0 62.8 63.7 Female 44.01 52.8 32.9 37.0 34.7 39.6 35.6 28.3 59.8 54.0 32.7 59.5 73.3 53.4 48.3 28.0 35.2 35.6 Total 47.47 47.5 43.0 45.0 41.0 46.3 44.3 28.2 74.5 62.6 53.2 66.0 71.6 55.2 52.1 40.4 54.0 46.2 60 Table 5. Percentage of Households that Recovered from Losses Caused by a Shock Loss of livestock non-related Diseases/Medical expenses Loss of livestock related to Bad harvest non-related to Bad quality/irregularity of Job loss/wage decrease Rent payment/increase Bad harvest related to Electricity/gas outage Fuel/transport prices High price of food Death of a family member/funerals Debt repayment Insecurity/theft drinking water Another shock to drought No shock drought drought drought Total Fire Quintile Poorest 26.78 17.3 32.9 30.0 33.1 19.8 33.2 26.3 18.8 43.0 29.7 14.2 14.9 15.9 18.5 52.2 21.9 45.4 Second 28.31 13.9 27.0 38.2 25.7 32.9 34.9 13.7 25.5 11.9 15.2 2.4 18.4 31.8 42.0 0.0 41.2 22.1 Third 31.27 35.4 36.0 38.6 31.8 37.2 61.4 49.3 16.6 17.7 61.1 17.7 17.8 23.0 27.2 70.1 41.0 Fourth 36.89 36.4 38.0 43.2 33.9 34.8 58.9 53.5 19.5 33.7 3.4 34.7 23.8 17.7 36.5 100.0 40.3 0.0 Richest 49.48 47.0 47.5 60.3 49.6 32.8 55.8 48.9 52.5 46.2 57.0 24.5 47.3 46.8 40.4 100.0 100.0 100.0 Region Djibouti 36.27 30.3 40.1 49.4 33.5 37.4 63.4 46.0 36.6 35.4 46.6 30.7 72.7 62.1 64.7 91.8 80.2 72.3 Ali Sabieh 57.89 46.3 49.9 38.4 65.0 36.1 100.0 29.3 5.4 2.2 74.8 40.3 42.5 46.8 0.0 Dikhil 25.29 14.4 23.4 29.5 24.7 16.8 20.6 14.5 14.1 17.8 18.3 14.1 10.3 3.3 4.0 0.0 0.0 Tadjourah 29.37 32.0 38.2 31.1 35.7 28.4 39.8 50.0 6.6 58.3 0.0 15.3 16.6 12.7 12.4 44.7 29.5 19.2 Obock 21.94 30.2 25.6 16.5 31.8 18.0 22.7 19.4 16.0 19.3 17.7 7.9 12.1 17.3 27.1 38.1 7.3 0.0 Arta 34.92 30.5 45.5 44.3 41.8 47.0 36.9 19.4 31.8 31.0 0.0 33.4 10.5 0.0 100.0 61 Loss of livestock non-related Diseases/Medical expenses Loss of livestock related to Bad harvest non-related to Bad quality/irregularity of Job loss/wage decrease Rent payment/increase Bad harvest related to Electricity/gas outage Fuel/transport prices High price of food Death of a family member/funerals Debt repayment Insecurity/theft drinking water Another shock to drought No shock drought drought drought Total Fire Area Urban 38.15 31.0 40.8 49.5 36.1 37.6 53.8 43.9 34.8 33.7 42.5 34.8 49.0 60.4 62.4 92.0 56.5 69.5 Rural 25.5 15.1 17.5 17.2 31.9 13.5 23.9 15.9 10.3 20.4 8.6 11.1 15.0 11.1 14.2 36.1 7.8 18.0 Gender of household head Male 35.01 29.0 37.1 45.7 34.7 30.1 52.8 38.0 24.4 33.4 27.2 13.5 17.9 14.9 20.9 61.6 18.9 49.1 Female 32.6 24.4 38.6 38.4 35.2 35.4 49.2 37.3 22.2 29.0 41.7 20.7 14.6 34.3 33.3 67.5 50.3 45.5 Total 34.44 28.0 37.5 43.0 34.8 31.0 52.0 37.9 23.9 32.5 31.4 15.3 17.0 20.9 24.6 64.4 28.9 46.8 62