41892 MADAGASCAR ­ An Evaluation of the Welfare Impact of Higher Energy Prices in Madagascar Africa Region Working Paper Series No. 106 June 28, 2007 Abstract I n this paper we estimate the effect of a rise in petroleum prices on living standards in Madagascar combining information on expenditure patterns from the Enquete Aupres des Menages 2005 with an input-output model describing how petroleum price increases propagate across economic sectors. We identify both a direct welfare effect (heating and lighting one's house become more expensive) and an indirect effect (the price of food and anything else which has to be transported from factory to shop rises). We find that, a 17 percent rise in oil prices produces, on average, a 1.75 percent increase in household expenditures (1.5 percent for high-income households, 2.1 for the households in the bottom expenditure quintile). Circa 60 percent of the increase in expenditures is due to the indirect effect, mostly via higher food prices. Although energy price increases hurt the poor more in percentage terms, subsidizing would involve a substantial leakage in favor of higher income households. This raises the issue of identifying more cost-effective policies to protect the poor households against energy price increases. JEL classification system: D57, H2, Q4, R2. Key Words: energy prices; oil; price subsidies; input-output analysis; Madagascar. Authors' Affiliation and Sponsorship Noro Aina Andriamihaja, Economist, AFTP1 (World Bank) nandriamihaja@worldbank.org Giovanni Vecchi*, (Università di Roma "Tor Vergata") giovanni.vecchi@uniroma2.it The Africa Region Working Paper Series expedites dissemination of applied research and policy studies with potential for improving economic performance and social conditions in Sub-Saharan Africa. The Series publishes papers at preliminary stages to stimulate timely discussion within the Region and among client countries, donors, and the policy research community. The editorial board for the Series consists of representatives from professional families appointed by the Region's Sector Directors. For additional information, please contact Paula White, managing editor of the series, (81131), Email: pwhite2@worldbank.org or visit the Web site: http://www.worldbank.org/afr/wps/index.htm. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s), they do not necessarily represent the views of the World Bank Group, its Executive Directors, or the countries they represent and should not be attributed to them. An Evaluation of the Welfare Impact of Higher Energy Prices in Madagascar Noro Andriamihaja Giovanni Vecchi* June 28, 2007 (*) This paper is the product of a joint Government and World Bank effort to investigate the poverty and social impacts of key policy reforms and economic events. The authors would like to thank the PSIA Group at the IMF and Benu Bidani and Stefano Paternostro of the World Bank for advice and comments. They are also grateful to Brian A'Hearn and Nicola Amendola, for many helpful discussions. Thanks also to Laza Razafiarison for help at various stages of the project. Contact address: giovanni.vecchi@uniroma2.it Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 1 Contents 1. Introduction ...........................................................................................................................................2 2. General Background.............................................................................................................................3 3. The Welfare Impact of Higher Energy Prices........................................................................................6 3.1. The Direct Welfare Effect.................................................................................. 7 3.2. Indirect Welfare Effect....................................................................................... 9 3.3. Total Welfare Effect......................................................................................... 13 4. Summary And Final Remarks.............................................................................................................14 5. References..........................................................................................................................................15 6. Appendices .........................................................................................................................................16 List of Tables Table 1 - Direct Welfare Effect of Price Changes .....................................................................................8 Table 2 - Indirect Price and Real Income Effects by Sector....................................................................13 Table 3 - Total Welfare Effect of Energy Price Changes ........................................................................14 List of Figures Figure 1 - International versus Domestic Crude Oil Prices - Madagascar, 2003-2006 ............................3 Figure 2 - Domestic Prices of Selected Petroleum Products ....................................................................5 Figure 3 - Impact of the increase of international oil price on imports (as % of GDP)..............................6 Figure 4 - Difference between urban and rural energy consumption patterns ..........................................8 Figure 5 - Components of Direct Effect by Quintile of Per Capita Expenditure.........................................9 Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 2 1. INTRODUCTION Typical analyses of the economic impact of oil price movements treat them as macroeconomic supply shocks affecting inflation and output. Not less important is their distributional impact on different segments of the population, operating through relative prices and real incomes. A rise in petroleum prices is not simply bad news for the economy as a whole, but is particularly bad news for poorer households. Over the past few years, Madagascar has experienced a substantial increase in oil prices. The relatively low price of international crude oil of USD 29.8 a barrel in December 2003, has increased by about 150 percent, to reach USD 62 a barrel by December 2006. To make things worse, the local currency, the Ariary has depreciated considerably during the period 2003-2005. A US dollar bought 1,277 Ariary in January 2003, which became 2,488.5 Ariary by mid 2004. Beginning in the second half of 2004 the Ariary showed some tendency to appreciate to reach 2,013 at the end of December 2006. The exchange rate has since stabilized around 2,050 to the US dollar. Fiscal regimes did not help. Oil products are subject to specific taxes (TPP, Taxes sur les Produits Pétroliers), while petroleum products are subject to VAT. The oil tax rates were adjusted upward in 2002 and 2006. Diesel was affected the most by a 230% increase in TTP in 2002 and 179% in the budget law 2006. Given the dynamics of (i) international crude oil prices, (ii) the exchange rate, and (iii) the fiscal regimes, it comes as no surprise that domestic prices of petroleum products increased significantly between 2003 and the first half of 2006. Prices of gasoline, diesel and kerosene increased on average by 145 percent between December 2003 and December 2006. Higher energy prices have adverse consequences for the poor. Real income losses may be substantial, as higher oil prices not only imply higher prices for petroleum products directly consumed by households, but also higher prices of other goods which use petroleum as an intermediate good in the production process. In fact, previous studies show that the latter, indirect effect is just as important or more so than the direct effect.1 With the goal of shielding the purchasing power of poor households, governments may consider subsidizing petroleum prices. However, the introduction of price subsidies raises a number of issues that need careful consideration. First, the introduction of price subsidies is not neutral from the distributive standpoint. The key questions here are: (i) Are the poor the real beneficiaries of the price subsidies?; (ii) What is the exact extent of their benefit,?, and (iii) Is the overall effect progressive or regressive? Second, there is a concern about the consequences of price subsidies in terms of allocational efficiency. Are subsidies the most effective/efficient way of protecting the real income of the poor? In the presence of binding budget constraints, subsidies are likely to divert resources from other social expenses, which may be more effective in reaching the poor. Moreover, by altering the structure of relative prices, subsidies may affect the incentives for households to use their energy efficiently. Third, there are fiscal considerations. Even if price subsidies are not financed by reductions in other social expenditures, they may eventually cause fiscal distress (increase in budget deficit and debt). Thus, it can be argued, that they may lead to adjustment policies, e.g. increases in taxes with offsetting effects. 1 See Coady and Newhouse (2005), Kpodar (2006). Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 3 To begin addressing the above issues in the context of Madagascar, this paper analyzes how higher petroleum prices impact households, focusing most of the attention on the poor. The simulation exercise is based on an input-output model á la Coady and Newhouse (2005). This approach, besides being relatively easy to implement, efficiently combines micro- and macro-data which are commonly available for most countries. The paper is organized as follows. Section 2 provides background information on the present pricing regime in Madagascar, the petroleum sector market structure, and the pattern of price increases. Sector 3 outlines the method used to asses the welfare effect of higher oil prices, and presents the estimates. Section 4 contains final remarks. 2. GENERAL BACKGROUND Madagascar is a net oil importer. Madagascar stopped importing crude oil in 2005, after the refinery in Tamatave was closed. It imported, however, 540,106 metric tons of final products in 2006. Besides transport, the lighting and electricity generation sectors are the second largest users of petroleum. More than half of the electricity produced in Madagascar is derived from fuel. More than 85 percent of the rural population use kerosene for lighting. The market is dominated by four companies: Total, Shell, Galana and Jovenna. Prices were set by the Malagasy Petroleum Office (OMH), the regulatory agency till June 2004, based on a formula with monthly adjustments. In early 2004, when oil prices started moving upwards, the Government froze prices for a couple of months. Prices were expected to be fully liberalized by October 2004, but this date was moved forward to July 2005, after which prices at the pump have been set by the operators, with no oversight whatsoever by the government. Oil prices have increased dramatically since the beginning of 2004. Figure 1 shows that between January 2003 and December 2006 international prices rose significantly (+1.7 percent per month), if not steadily. From 31.3 USD a barrel in Jan, the price of crude oil peaked in July 2006 (72.45 USD), and closed at 60 USD in December 2006. Figure 1 - International versus Domestic Crude Oil Prices ­ Madagascar 2003-2006 400 International price - usd/bbl (left axis) Domestic price - ariary/bbl (left axis) CPI (right axis) 140 x 300 x denI 0) deIn 0) eci 10 = eci 10 = Prl 03 Oi 20 Prre 03 20 e udrC anJ( 120 um 200 nsoC anJ( 100 100 January 2003 January 2004 January 2005 January 2006 January 2007 Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 4 Figure 2 shows the dynamic of prices for a selection of petroleum products. Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 5 Figure 2 - Domestic Prices of Selected Petroleum Products 3000 premium gasoline regular gasoline diesel 2500 lamping kerosene 2000 e li/yrr)et Pric 1500 (Aria 1000 500 Jan Jan Jan Jan Jan Jan Jan Dec 2000 2001 2002 2003 2004 2005 2006 2006 Source: OMH ­ Malagasy Petroleum Office. In order to calculate the effect of the exchange rate on the increase of domenstic prices of energy products we define as follows. Let poil (i) be the international price of a given energy product at the $ beginning of year i, and (i) be the exchange rate (ariary/dollar) at the beginning of year i. The change in international price in ariary during the year i is given by: poil = (i +1) poil (i +1)-(i) poil (i) Ary $ $ The (simulated) change of international price in Ariary, assuming a constant exchange rate during the year i is: poil (sim) = (i) poil (i +1)-(i) poil (i) Ary $ $ The effect of the exchange rate on the change in domestic prices is therefore given by: Exchange Rate Effect = poil -poil (sim) Ary Ary poil Ary Table 1 shows the proportion of the domestic price change due to the variation of the exchange rate for selected oil products Table 1 ­ Exchange rate effect on domestic price (percent) 2004 2005 2006 Aircraft gasoline 92 100 23 Lamping kerosene 61 24 30 Premium gasoline 66 26 26 Regular gasoline 71 24 30 Diesel 66 23 29 Fuel Oil 99 12 93 Total 60 22 32 Source: OMH and authors' calculation. Note: domestic prices are assumed to be a linear transformation of international prices. Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 6 The increase in international oil prices caused a net deterioration of the terms of trade by 5.1% in 2004 and 6.2% in 2005. In addition, the net impact of these international price movements is estimated at 1.57% of GDP in 2005 (Figure 3): this is a measure of the aggregate burden to the economy caused by the increase of the international oil product. Figure 3 - Impact of the increase of international oil price on imports (as % of GDP) 2 1.75 PDGfo 1.5 1.25 %sa 1 ct pa .75 Im .5 .25 0 2002 2003 2004 2005 2006 year Source: OMH and authors' calculation. 3. THE WELFARE IMPACT OF HIGHER ENERGY PRICES The effects of increasing petroleum prices on household welfare are twofold. First, the direct effect: households are affected through increases in the price of petroleum products they consume directly, e.g. kerosene for lighting and/or cooking, gasoline for private transport, etc.. Second, the indirect effect: households are affected through increases in the prices of other goods and services, as higher energy costs are passed through onto the consumer. One way of assessing the scale of the two effects is by decomposing the total welfare effect (TWE) into its two components, namely the direct welfare effect (DWE) and the Indirect Welfare Effect (IWE), and estimate them separately: (1) TWE = DWE + IWE Let woil (j=1,..n) denote the budget share for the j-th petroleum product, and poil its price variation j j (expressed in percent). The DWE of price changes in petroleum products can be calculated by multiplying the price variation by the corresponding household budget share and aggregating across goods: J (2) DWE = p oil j woil j j=1 where J is the number of petroleum products consumed by the households. Equation (2) expresses DWE as a percentage of total household expenditure. The procedure is more involved for the IWE. Below, we provide a broad outline of the strategy for estimating IWE, while section 3.2 provides a full account. Let poil denote an aggregate (scalar) measure of the price of petroleum products: Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 7 J J (3) poil = poil with j j j =1 j=1 j=1 where the last term is quantity-share based. The change in this aggregate price is: J (4) poil = poil j j j=1 Now let us define q as a (row) vector of changes in consumer prices: (5) q = f poil( ) Equation (5) makes clear that consumer prices for non-petroleum goods and services depend on prices of petroleum goods, as the latter enter the production of the former as intermediate goods. Once estimates for equation (5) are available, calculating IWE is a simple matter: S (6) IWE = f (p )w oil j j j=1 where f poil ( ) j denotes the change in price for the j-th good (or service) consumed by the household. The difficulty, as we shall see in section 3.2, is in calculating q, since the function f ( ) j in equation (6) is a mapping between producer and consumer prices that must account for the production structure of various sectors of the economy, in particular, for the different intensity in the use of petroleum products as inputs by various sectors. In sections 3.1 and 3.2 we discuss equations (2) and (6) in detail, and obtain estimates for both the direct and indirect effects. 3.1. The Direct Welfare Effect Measuring the direct effect of a change in petroleum product prices on households' welfare is a relatively straightforward matter. Schematically, it requires three steps. (i) Identifying the petroleum products directly consumed by households. According to the EPM 2005, information is available for electricity (électricité), gasoline (essence), diesel (gas-oil), kerosene (pétrole). (ii) Identifying the price increases for each petroleum product. Data on prices are available from the OMH. (iii) Estimating the direct impact of price increases on households by multiplying the budget share of each petroleum product by its percentage price increase. The top panel of Table 1 shows household budget shares for electricity, gasoline, diesel and kerosene. Overall, petroleum products absorb, on average, 2.6 percent of the household budget. However, Table 1 shows that the consumption of energy products differs significantly across households according to their expenditure levels. For example, kerosene ­ the single most important product with a 1.9 percent budget share ­ accounts for as much as 3.2 percent of expenditures among the poorest households, but only 1 percent among households in the upper quintile. Consumption of gasoline, by contrast, is negligible for most households, but for the 0.3 percent of expenditures among the richest 20 percent. Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 8 Similarly, the budget share for electricity increases dramatically with the household's living standard, from 0.06-0.10 percent in the bottom two quintiles of total expenditure to 1.17 percent for the richest 20 percent of households. Figure 4 investigates the differences in consumption patterns among urban and rural households. Three features stand out. First, the budget share allocated to electricity is always higher for urban households than rural, with a clear pattern across income quintiles. Second, rural households allocate a higher percentage of total consumption to kerosene. Third, the overall pattern in budget shares for energy products differs according to living standards: poor rural households allocate a higher percentage of their total expenditure to energy than poor urban households, while the reverse is true for the richest 40 percent. Figure 4 - Difference between urban and rural energy consumption patterns 2 1 0 poorest 2nd 3rd 4th -1 richest electricity gasoline diesel kerosene total Note: the graph charts the average difference between urban and rural budget shares for energy products, by national per capita expenditure quintiles. A positive bar implies a higher consumption in urban than in rural areas. Table 1 - Direct Welfare Effect of Price Changes Per capita expenditure Q1 Q5 quintiles (poorest) Q2 Q3 Q4 (richest) All Household Budget Shares (%) Electricity 0.06 0.10 0.21 0.45 1.17 0.48 Gasoline 0.02 0.00 0.00 0.02 0.27 0.08 Diesel 0.19 0.10 0.08 0.07 0.08 0.10 Kerosene 3.18 2.31 1.96 1.64 1.04 1.89 All 3.46 2.51 2.25 2.19 2.56 2.55 Direct Welfare Effect (% of total household 1.19 0.86 0.73 0.64 0.50 0.74 expenditure) Mean consumption of petroleum products 1.00 1.08 1.16 1.34 3.10 1.69 (ratio to bottom quintile) Notes: Budget shares are based on the EPM 2005. Quintiles are based on the national distribution of per capita annual expenditures. The estimation of the direct welfare effect is based on price increases observed during 2005 (11 percent for gasoline, 13 percent for diesel, and 36 percent for kerosene). The change in the price of electricity is assumed to be one-third of the average change in petroleum prices. Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 9 Table 1 shows estimates of the direct welfare effect, as defined in equation (2). The average direct impact of the increase in the price of energy products on total household expenditure is 0.74 percent. The table shows that the loss of purchasing power is higher for the poor (1.2 percent) than for the rich (0.5 percent). Part of this difference is attributable to differences in the total budget share of all fuels, while the rest is due to the budget share structure. It is worth mentioning that the estimates in Table 1 are effectively a Laspeyres index that hold budget shares fixed, overstating the loss in welfare. This issue will be taken up again in the final section. Figure 5 - Components of Direct Effect by Quintile of Per Capita Expenditure 1.3 Electricity Gasoline Diesel 1.0 Kerosene reu pendit 0.8 exlatot (%) of 0.5 Share 0.3 0.0 poorest 2nd 3rd 4th richest Figure 5 clearly shows the regressive nature of the welfare effect of energy price increases. The poor are affected most severely (1.2 percent of their total expenditure), mostly by the increase in price of kerosene, with other fuels and electricity playing a marginal role only. The negative impact on welfare decreases monotonically with per capita expenditure. Rich households are affected the least (0.5 percent of total expenditure), as a consequence of the increase in prices of both kerosene and electricity. 3.2. Indirect Welfare Effect Measuring the indirect effect of a change in petroleum product prices on households' welfare is not a straightforward matter. In this section we use the price-shifting model introduced in Coady and Newhouse (2005). The building blocks of this model are (i) an input-output matrix, and (ii) household expenditure patterns, available from the recent EPM 2005. The Price-Shifting Model We start by assuming that the technology of the economy is fully described by the input-output matrix (IO matrix, henceforth). The IO matrix describes the use of sectoral inputs in the production of sectoral outputs, and, in particular, it provides information on the use of petroleum products as inputs in each sector of the economy. In the case of Madagascar, the IO table contains 30 economic sectors, and was last estimated for the year 1995.2 A stylized representation of the IO matrix is as follows: 2 See the Appendix. Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 10 a11 a21 L a1 S (7) A = a21 a22 M aS O 1 aSS In our application, the matrix A is a 30 ×30 square matrix. The generic element of A, aij (i,j=1, 2, ..., S) represents the cost of the i-th input per 1 unit of value of the j-th output. Units of output are defined so that they have a user price of unity. Appendix A shows that the aij coefficient represents the change in the cost of producing one unit of j due to a unit change in the price of input i. 3 The next step is modeling each sector of the economy as a producer of an "aggregate" or composite commodity. There are as many composite commodities as the number of economic sectors in the IO matrix. For instance, sector "CN01 Agriculture" produces the "agricultural commodity", sector "CN02 Animal Production" produces the "animal production commodity" and so on. Each sector is assumed to face a certain market structure, which determines the mechanism through which changes in input prices are passed onto output prices. Precisely, we assume that three market structures are enough to describe the sectors of the economy (notation is as follows: p denotes the 1 × S vector of producer prices net of sales taxes and/or tariffs, while q denotes the 1 × S vector of consumer prices): A) Cost-push sectors. These are sectors where higher input costs are pushed fully on to output prices. This is likely to occur for most government services, construction, public utilities, trade and transportation, and retail and wholesale trade. In general, one would expect this pricing scheme to apply to non-traded commodities. The formula: (8) qcp = pcp + tcp According to equation (8), the final price paid by consumers (qcp) is equal to the price set by producers (pcp) plus sales or excise taxes (tcp) imposed by the government B) Traded sectors. These are sectors that compete with internationally traded goods. Foreign goods compete with domestic goods, therefore higher input costs cannot be passed onto output prices. (9) qts = pworld + tts In equation (9) consumer prices (qts) are determined by world prices (pworld), and by trade taxes (tts, inclusive of tariffs and sales taxes). C) Controlled sectors. These are sectors where output prices are controlled by the government: (10) qcs = p* In equation (10) consumer prices (qcs) are determined by pricing controls (p*). For the sake of simplicity, domestic taxes are set to zero. Having defined the price-setting equations, we now turn to modeling the mechanisms through which factor price changes are passed onto output prices. Changes in prices for traded sectors are given by: 3A sketch of the proof is contained in the Appendix. Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 11 (11) qts = pworld + tts where both pworld and tts on the right-hand side are assumed to be exogenous. For controlled sectors the formula for the changes in prices is most simple, and is obtained from equation (10): (12) qcs = p* where the right-hand side variable is exogenously determined by the government pricing controls. Finally, the change in consumer prices in the cost-push sector is given by: (13) qcp = pcp + tcp A problem arises in calculating the term pcp. Producer prices depend on all factor prices of intermediate goods. Let w denote the vector containing the prices of production factors not included in the IO table (wages, for instance) and let q be the vector of intermediate goods included in the IO table; equation (13) can then be re-written as follows: (14) qcp = pcp(w,q)+ tcp Equation (14) shows that the difficulty in calculating qcp arises because output prices q are in fact input prices for certain industries: this is what the term pcp(q) represents. How do changes in q pass on to final prices? For simplicity we will ignore changes in w and will focus on q, so that pcp(w,q) is reduced to pcp(q). The solution suggested by Coady and Newhouse (2005) is based on the assumption that each of the composite commodities described above is made up of a certain proportion of cost-push, traded and controlled commodities. Let , and denote these proportions, respectively. To illustrate, let us assume that the producers of the "agricultural commodity" buy s percent of inputs from producers in the cost-push sectors, s percent of inputs from the traded sector and s percent from the controlled sector. The suffix s refers to the sector, and ranges from 1 to 30. These proportions should, obviously, sum to unity and never be negative: (15) 0 (s,s,s) 1 and s+s+ s=1 (s = 1, ... 30) The change in the price of the j-th commodity (the "agricultural commodity") can be expressed as a linear combination of the three market structures identified above: pcpj (q)= iaijqcpj + iaijqtsj + iaijqcsj S S S (16) i=1 i=1 i=1 (j=1,...,S) Equation (16) can be compacted by using matrix notation: (17) pcp = qcp A + qts A + qcs A where pcp is a 1 × S row vector, , and are diagonal S × S matrices, qcp is 1 × S, qts is 1 × S, and qcs is 1 × S. Equation (17) gives the changes in the producer prices for the controlled sectors. Now substitute (11), (12) and (13) in equation (17): (18) pcp = pcp A + tcp A + pworld A + tts A + p* A Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 12 The reduced form of equation (18) is given by: (19) pcp = tcpAV + pworldAV + ttsAV + p*AV where to simplify the notation we let V = (I ­ A)-1. Equation (19) gives the vector of producer price changes in the cost-push sectors. The matrix V captures both the direct and indirect effects of input price changes on output price changes. The equation we are interested in is consumer price changes: (20) q = qcp +qts+qcs In our application to Madagascar we further simplify the model by assuming that (i) the only exogenous price changes are changes in the controlled sector (in other words, we assume no changes in producer prices abroad pworld = 0, no changes in either taxes or tariffs tts = 0 and tcp = 0); (ii) all petroleum products are within the controlled sector (this poses restrictions on the matrices , and ), and (iii) all other products are cost-push sectors. As a consequence, equation (11) is reduced to qts= 0, and equation (13) becomes qcp = pcp. Under hypotheses (i)-(iii), equation (20) can be re-written as: (21) q = pcp +p* Finally, we substitute (19) in (21): (22) q = p*AV + p* = p*[AV + ] Equation (22) is key for the evaluation of the impact of a change in energy prices on consumer prices for the range of sectors available in the I/O table. The Estimates Table 2 shows the impact of higher petroleum prices on the prices in other sectors. Multiplying these induced price increases (column 2) by the corresponding household budget shares (column 1) and aggregating across goods and services gives the percentage increase in costs due to the indirect price increases. According to the estimates inTable 2, the generalized increase in prices in goods and services caused by a 17 percent increase in petroleum prices implies a drop in household real income of 1 percent (row total of column 3). This is the estimate of what was referred to as the "indirect welfare effect" in previous paragraphs. also shows that the single most important contribution comes from products in the food industry, which accounts for almost two-thirds of the indirect effect. The aggregate of all food-related sectors, accounts for almost 80 percent of the overall indirect effect. The second highest source is the textile sector, which accounts for some 7 percent of the indirect effect. Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 13 Table 2 - Indirect Price and Real Income Effects by Sector Budget Shares Price Impact on Percent of Total Sector (%) Effect Expenditure Impact (%) (%) (%) (1) (2) (1)×(2) /100 Agriculture 10.17 0.29 0.03 2.94 Farming 0.20 1.13 0.00 0.23 Agro-industry 1.53 2.75 0.04 4.19 Food industry 39.53 1.61 0.64 63.38 Beverage 0.70 1.29 0.01 0.90 Tabac 2.15 0.95 0.02 2.03 Fats 2.00 3.48 0.07 6.93 Chemical 2.30 2.21 0.05 5.06 Textile 2.97 2.44 0.07 7.22 Metal 0.00 1.69 0.00 0.00 Electrical products 0.00 1.08 0.00 0.00 Paper 0.01 2.81 0.00 0.03 Leather 0.66 1.49 0.01 0.98 Transport (people) 0.94 3.14 0.03 2.94 Transport (other) 0.02 1.26 0.00 0.03 Telecommunication 0.06 0.87 0.00 0.05 Trade 0.30 1.04 0.00 0.31 Services (coll.) 3.15 0.87 0.03 2.73 Services (indiv.) 0.18 0.36 0.00 0.06 Total 66.87 1.00 100.00 Note: Budget shares are derived from EPM 2005 based on commodity groupings that match the more aggregated input-output table sectoral breakdown available for the year 1995. Quintiles are based on the national distribution of per capita annual expenditures. The estimation of the indirect welfare effect is based on price increases observed during 2005 (11 percent for gasoline, 13 percent for diesel, and 36 percent for kerosene). The change in the price of electricity is assumed to be one-third of the average change in petroleum prices. 3.3. Total Welfare Effect After calculating separately the direct effect by aggregating real income changes across petroleum products, and the indirect effect, by aggregating real income changes across all other commodities, the total effect can be calculated as the sum of the two separate effects. presents the key estimates for understanding the distributional impact of higher petroleum prices in Madagascar. A number of findings are worthy of a comment. First, the top panel of the table shows that the the total welfare effect (row 3) is significant in magnitude: an average increase of 17 percent in energy prices causes a loss equal to 1.75 percent of total household expenditure. The loss is greater for housholds in the bottom quintile (2.14 percent). It decreases for households in the upper quintile (1.52 percent). The incidence of the increase in oil prices is therefore unambiguosly regressive. Second, circa 60 percent of the total welfare loss is due to the indirect welfare effect. This implies that the main channel through which households are affected by higher energy prices is the indirect effect on non-oil prices (especially food prices, as noted above). The combination of (i) a relatively high sensitivity of food prices to oil prices (Table 2, column 2), and (ii) a large budget share devoted to food, column 1) is largely responsible for the prevalence of the indirect welfare effect shown inTable 3. Third, the share of the indirect effect is lowest for the poorest households, accounting for 44 percent of the total welfare effect, compared to 67 percent for the households in the top quintiles. This result is driven by the pattern observed for the direct effect. Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 14 Table 3 - Total Welfare Effect of Energy Price Changes (% of total household expenditure) Per capita expenditure Q1 Q5 quintiles (poorest) Q2 Q3 Q4 (richest) All Direct Welfare Effect 1.19 0.86 0.73 0.64 0.50 0.74 Indirect Welfare Effect 0.95 1.01 1.01 1.03 1.02 1.01 Total Welfare Effect 2.14 1.87 1.74 1.67 1.52 1.75 IWE as % of total 44 54 58 62 67 58 Share of the burden Direct Welfare Effect 13.4 16.0 17.8 20.4 32.4 100 Indirect Welfare Effect 6.9 11.7 15.5 21.2 44.7 100 Total Welfare Effect 9.2 13.2 16.3 20.8 40.4 99.9 Note: Budget shares are derived from EPM 2005 based on commodity groupings that match the more aggregated input-output table sectoral breakdown available for the year 1995. Quintiles are based on the national distribution of per capita annual expenditures. The estimation of the total welfare effect is based on price increases observed during 2005 (11 percent for gasoline, 13 percent for diesel, and 36 percent for kerosene). The change in the price of electricity is assumed to be one-third of the average change in petroleum prices. The bottom panel of translates the percentage effects shown in the top panel into shares of aggregate real expenditure loss borne by each quintile. Although energy products may absorb a higher proportion of the total consumption budget of low-income households, high-income households typically consume larger quantities of these products. Hence, it is important to assess the distribution of the total loss, in absolute terms (that is in local currency units), by quintile. This is precisely what is shown in the bottom panel of. The top two quintiles, representing only 40 percent of the population, account for circa 60 percent of the total loss, compared to circa 22 percent for the bottom two quintiles. The pattern observed for the total welfare effect mirrors the pattern for both indirect and direct effects. The main finding here is that although price increases hurt the poor more in percentage terms (third rows of Table 3), subsidizing energy prices would involve a substantial leakage in favor of higher income households (last row of Table 3). This empirical finding raises the issue of identifying more cost- effective policies to protect the poor households against energy price increases. 4. SUMMARY AND FINAL REMARKS High on most governments' agendas is the issue of how to deal with the adverse consequences of higher energy prices on living standards of the poorest segments of the population. A hotly debated question is whether governments should introduce price controls and/or intervene in other ways to mitigate the social costs of oil price increases. In the absence of compelling, unambiguous theoretical arguments, assessing the advisability of price controls must be based on the empirical evidence. In this paper we assess the distributional impact of the increase in petroleum prices during 2005, by estimating the impact on household real expenditures. The main findings can be summarized as follows. First, a 17 percent rise in the price of energy products leads to a 1.75 percent average decrease in real expenditure. This percentage is higher for low-income households (2.1 percent) than for for high- income households (1.5 percent). This implies that the benefits of introducing energy price subsidies would be progressive, i.e. in percentage terms, subsidies would benefit poor households more than rich households. However, subsidizing would involve substantial leakage in favor of high-income households, who account for more than 60 percent of the total burden of price increases. This raises the issue of identifying more cost-effective policies to protect poorest households. The relatively large size Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 15 of the "leakage to the rich" implied by our estimates, suggests that improvements in the ability to target social transfers and/or expenditure should be a priority. The analysis carried out in the paper suggests that subsidies on kerosene represent one such option worth exploring. Consistent with our results is an overall re-thinking of the system of safety nets. It is well to bear in mind the limitations of the method used here. A first simplification in our analysis is the assumption that all "composite goods" (with the exception of energy products) are produced in cost- push sectors. As noted by Kopdar (2006), this is likely to over-emphasize the importance of the indirect price effect, though the bias is hard, if at all possible, to quantify. Second, our analysis does not take into account the fact that consumers are likely to change their consumption behaviour in response to the initial price shock. Nor does it account for the fact that producers are also likely to modify their use of factors of production. To deal with substitution effects would require a different approach, based on the estimation of a computable general equilibrium model, a task which goes beyond the scope of this paper ­ see Banks, Blundell and Lewbel (1996). That said, it is probably a fair claim, that given the relatively low substitutability of petroleum products for both consumers and producers in the context of Madagascar, the results obtained in the paper may be deemed reasonable, and possibly realistic, at least, for the short term. 5. REFERENCES Banks, J., R. Blundell, and A. Lewbel (1996), "Tax Reform and Welfare Measurement: Do We Need Demand System Estimates?" Economic Journal, vol. 106, No. 438, pp. 1227­41. Coady, D. and D. Newhouse (2005), "Evaluating the Distribution of the Real Income Effects of Increases in Petroleum Product Prices in Ghana", International Monetary Fund, mimeo. Gupta, S, M. Verhoeven, R. Gillingham, C. Schiller, A. Mansoor, J. P. Cordoba (2000), Equity and Efficiency in the Reform of Price Subsidies. A Guide for Policymakers. International Monetary Fund. Kpodar, K. (2006), "Distributional Effects of Oil Price Changes on Household Expenditures: Evidence from Mali.", IMF Working Paper, WP/06/91. International Monetary Fund. Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 16 6. APPENDICES Appendix A - On the interpretation of the coefficients of the Input-Output matrix This Appendix shows that the aij coefficient of the IO table represents the change in the cost of producing one unit of j due to a unit change in the price of input i. According to equation (7) in the text, aij denotes the cost of input i to produce one unit of output j. Now re-scale the unit of measurement of each output so that total value of each output is equal to one: S (23) v j = a ij + AVj =1 j i=1 where vj is the total value of the j-th output, and AVj is the added value for output j. Let bij denote the quantity of input i required to produce one unit of output j, as defined above. By definition we have: (24) aij = bij pi The elasticity of the monetary cost of producing j with respect to the price of the input i is: (25) vj × pi pi QED pi = bij × = bij pi = aij v j v j Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 17 Appendix B ­ The IO table for Madagascar 1995 This Appendix shows the IO table for 1995, based on a compilation of data from the Household Survey, the Ministry of Economy and Finance, the Ministry of Trade, the Ministry of Agriculture, and the Central Bank of Madagascar. External trade data are from the National Institute for Statistics and other survey to enterprises, NGOs, the banking sector and insurance companies. agric. farming hunting fishing agro mining energy food bever. tabac fats chem. textile wood nonmetal metal agriculture 0.0137 0.0879 0.0000 0.0141 0.3896 0.0000 0.0000 0.3250 0.1476 0.2160 0.0700 0.0049 0.0514 0.0000 0.0000 0.0000 farming 0.0194 0.0019 0.0000 0.0000 0.0000 0.0000 0.0000 0.2464 0.0000 0.0000 0.0000 0.0000 0.0001 0.0000 0.0000 0.0000 hunting 0.0076 0.0328 0.0000 0.0386 0.0018 0.0002 0.0000 0.0004 0.0003 0.0092 0.0025 0.0035 0.0001 0.1023 0.0042 0.0110 fishing 0.0000 0.0516 0.0000 0.0000 0.0000 0.0000 0.0000 0.0261 0.0000 0.0000 0.0197 0.0000 0.0000 0.0000 0.0000 0.0000 agro 0.0014 0.0000 0.0000 0.0000 0.0042 0.0016 0.0000 0.0271 0.0166 0.0000 0.0131 0.0013 0.0000 0.0000 0.0000 0.0000 mining 0.0001 0.0000 0.0000 0.0000 0.0000 0.0000 0.3247 0.0194 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 energy 0.0035 0.0214 0.0497 0.0635 0.1253 0.0756 0.0180 0.0479 0.0433 0.0189 0.0918 0.0251 0.0549 0.0282 0.0704 0.0241 food 0.0000 0.2091 0.0000 0.0003 0.0000 0.0000 0.0000 0.1132 0.0000 0.0000 0.0000 0.0002 0.0012 0.0000 0.0000 0.0000 beverage 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0008 0.0000 0.0000 0.0002 0.0000 0.0000 0.0000 0.0000 tabac 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0535 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 fats 0.0000 0.0257 0.0000 0.0000 0.0000 0.0000 0.0000 0.0041 0.0000 0.0000 0.3818 0.0253 0.0000 0.0000 0.0000 0.0000 chemical 0.0628 0.0393 0.0093 0.0045 0.1134 0.0594 0.0335 0.0094 0.0419 0.0108 0.1810 0.5608 0.0072 0.0295 0.0076 0.0620 textile 0.0006 0.0024 0.0023 0.0091 0.0007 0.0003 0.0006 0.0001 0.0006 0.0008 0.0035 0.0016 0.5602 0.0008 0.0000 0.0000 wood 0.0047 0.0000 0.0000 0.0000 0.0000 0.0297 0.0027 0.0003 0.0012 0.0016 0.0000 0.0000 0.0006 0.1657 0.0000 0.0107 nonmetal 0.0058 0.0000 0.0000 0.0000 0.0000 0.0000 0.0010 0.0000 0.0000 0.0000 0.0022 0.0000 0.0004 0.0014 0.0311 0.0015 metal 0.0110 0.0112 0.0346 0.0241 0.0579 0.0221 0.0791 0.0041 0.0612 0.0445 0.0060 0.0247 0.0144 0.0583 0.0579 0.5472 electrical pr. 0.0000 0.0005 0.0000 0.0003 0.0034 0.0000 0.0164 0.0001 0.0012 0.0000 0.0000 0.0000 0.0000 0.0018 0.0267 0.0118 paper 0.0003 0.0025 0.0026 0.0040 0.0077 0.0020 0.0086 0.0088 0.0069 0.0896 0.0084 0.0153 0.0061 0.0017 0.0397 0.0033 leather 0.0001 0.0007 0.0008 0.0021 0.0197 0.0095 0.0173 0.0006 0.0497 0.0101 0.0089 0.1025 0.0090 0.0041 0.0269 0.0107 construction 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 transstuff 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 transpeople 0.0008 0.0148 0.0056 0.0064 0.0022 0.0068 0.0085 0.0107 0.0164 0.0051 0.0050 0.0404 0.0102 0.0003 0.0411 0.0199 transother 0.0021 0.0026 0.0073 0.0129 0.0142 0.0404 0.0070 0.0005 0.0315 0.0253 0.0070 0.0308 0.0050 0.0123 0.0204 0.0488 telecom 0.0001 0.0014 0.0011 0.0041 0.0011 0.0008 0.0008 0.0007 0.0052 0.0052 0.0033 0.0073 0.0047 0.0009 0.0112 0.0023 trade 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 banks 0.0006 0.0042 0.0078 0.0095 0.0028 0.0016 0.0011 0.0007 0.0099 0.0103 0.0095 0.0043 0.0072 0.0016 0.0069 0.0013 servicefirm 0.0027 0.0274 0.0265 0.0375 0.0878 0.1506 0.0064 0.0050 0.0666 0.0722 0.0096 0.0101 0.0104 0.0937 0.0553 0.0195 servicecoll 0.0032 0.0121 0.0430 0.0597 0.0472 0.1295 0.0227 0.0092 0.0525 0.0480 0.0176 0.0347 0.0152 0.0705 0.0651 0.0244 serviceind 0.0095 0.0340 0.0234 0.0331 0.0480 0.0355 0.0074 0.0003 0.0287 0.0171 0.0449 0.0192 0.0049 0.0276 0.0248 0.0111 serviceap 0.0040 0.0047 0.0115 0.0044 0.0000 0.0075 0.0000 0.0000 0.0000 0.0080 0.0000 0.0000 0.0000 0.0000 0.0000 0.0002 (continued on next page) Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 18 The IO table for Madagascar 1995 (continued) electric. pr. paper leather constr. transstuff transpeople transother telecom trade banks servicefirm servicecoll serviceind serviceap agriculture 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0906 0.0000 0.0006 farming 0.0000 0.0000 0.0009 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0196 0.0000 0.0003 hunting 0.0000 0.0068 0.0002 0.0040 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0112 0.0004 0.0000 fishing 0.0000 0.0000 0.0403 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0586 0.0000 0.0002 agro 0.0000 0.0001 0.0140 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0187 0.0000 0.0006 mining 0.0000 0.0000 0.0234 0.0214 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0066 0.0000 0.0000 energy 0.0056 0.0468 0.0421 0.0203 0.1780 0.1483 0.0398 0.0297 0.0533 0.0089 0.0123 0.0120 0.0198 0.0324 food 0.0000 0.0000 0.0331 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.1787 0.0000 0.0018 beverage 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0305 0.0005 0.0000 tabac 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 fats 0.0000 0.0000 0.0001 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0122 0.0000 0.0000 chemical 0.0003 0.2271 0.0935 0.0735 0.0204 0.0241 0.0062 0.0014 0.0018 0.0000 0.0003 0.0043 0.0009 0.0090 textile 0.0000 0.0003 0.0001 0.0002 0.0000 0.0001 0.0000 0.0000 0.0000 0.0000 0.0000 0.0041 0.0005 0.0002 wood 0.0021 0.0000 0.0173 0.0877 0.0050 0.0039 0.0000 0.0000 0.0021 0.0000 0.0000 0.0014 0.0006 0.0025 nonmetal 0.0258 0.0006 0.0036 0.1776 0.0000 0.0000 0.0000 0.0034 0.0000 0.0000 0.0000 0.0000 0.0005 0.0003 metal 0.0128 0.0111 0.0181 0.3708 0.1068 0.2425 0.0470 0.0073 0.0009 0.0022 0.0287 0.0105 0.0028 0.0182 electrical pr. 0.7410 0.0242 0.0291 0.0222 0.0030 0.0029 0.0059 0.0000 0.0004 0.0000 0.0005 0.0005 0.0004 0.0004 paper 0.0135 0.4596 0.0167 0.0065 0.0004 0.0004 0.0764 0.0642 0.0065 0.0285 0.0056 0.0011 0.0002 0.0392 leather 0.0146 0.0283 0.0951 0.0259 0.0001 0.0002 0.0069 0.0397 0.0006 0.0062 0.0005 0.0006 0.0003 0.0176 construction 0.0000 0.0000 0.0000 0.0009 0.0000 0.0000 0.0013 0.0047 0.0000 0.0000 0.0000 0.0002 0.0003 0.0032 transstuff 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 transpeople 0.0070 0.0056 0.0137 0.0126 0.0074 0.0123 0.0114 0.0042 0.0102 0.0149 0.0209 0.0044 0.0010 0.0257 transother 0.0099 0.0380 0.0004 0.0131 0.0234 0.0172 0.0029 0.0000 0.0172 0.0000 0.0040 0.0062 0.0000 0.0009 telecom 0.0081 0.0044 0.0067 0.0050 0.0067 0.0069 0.1124 0.0000 0.0123 0.0140 0.0096 0.0014 0.0007 0.0082 trade 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 banks 0.0021 0.0060 0.0041 0.0045 0.0587 0.0985 0.0406 0.0004 0.0094 0.0112 0.0075 0.0011 0.0015 0.0552 servicefirm 0.0098 0.0460 0.0239 0.0155 0.0249 0.0980 0.1059 0.0236 0.0612 0.0255 0.0113 0.0629 0.0034 0.0559 servicecoll 0.0079 0.0492 0.0492 0.0100 0.0244 0.0321 0.0735 0.0855 0.0247 0.0064 0.0212 0.0096 0.0077 0.1094 serviceind 0.0058 0.0028 0.0298 0.0198 0.0315 0.0402 0.0423 0.0134 0.0251 0.0275 0.0249 0.0010 0.0008 0.0352 serviceap 0.0000 0.0000 0.0000 0.0022 0.0024 0.0017 0.0011 0.0000 0.0000 0.0000 0.0000 0.0022 0.0000 0.0000 Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 19 Africa Region Working Paper Series Series # Title Date Author ARWPS 1 Progress in Public Expenditure Management January 1999 C. Kostopoulos in Africa: Evidence from World Bank Surveys ARWPS 2 Toward Inclusive and Sustainable March 1999 Markus Kostner Development in the Democratic Republic of the Congo ARWPS 3 Business Taxation in a Low-Revenue June 1999 Ritva Reinikka Economy: A Study on Uganda in Comparison Duanjie Chen with Neighboring Countries ARWPS 4 Pensions and Social Security in Sub- October 1999 Luca Barbone Saharan Africa: Issues and Options Luis-A. Sanchez B. ARWPS 5 Forest Taxes, Government Revenues and January 2000 Luca Barbone the Sustainable Exploitation of Tropical Juan Zalduendo Forests ARWPS 6 The Cost of Doing Business: Firms' June 2000 Jacob Svensson Experience with Corruption in Uganda ARWPS 7 On the Recent Trade Performance of Sub- August 2000 Francis Ng and Saharan African Countries: Cause for Hope Alexander J. Yeats or More of the Same ARWPS 8 Foreign Direct Investment in Africa: Old November 2000 Miria Pigato Tales and New Evidence ARWPS 9 The Macro Implications of HIV/AIDS in South November 2000 Channing Arndt Africa: A Preliminary Assessment Jeffrey D. Lewis ARWPS 10 Revisiting Growth and Convergence: Is December 2000 C. G. Tsangarides Africa Catching Up? ARWPS 11 Spending on Safety Nets for the Poor: How January 2001 William J. Smith Much, for How Many? The Case of Malawi ARWPS 12 Tourism in Africa February 2001 Iain T. Christie D. E. Crompton ARWPS 13 Conflict Diamonds February 2001 Louis Goreux ARWPS 14 Reform and Opportunity: The Changing Role March 2001 Jeffrey D. Lewis and Patterns of Trade in South Africa and SADC ARWPS 15 The Foreign Direct Investment Environment March 2001 Miria Pigato in Africa ARWPS 16 Choice of Exchange Rate Regimes for April 2001 Fahrettin Yagci Developing Countries Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 20 Africa Region Working Paper Series Series # Title Date Author ARWPS 18 Rural Infrastructure in Africa: Policy June 2001 Robert Fishbein Directions ARWPS 19 Changes in Poverty in Madagascar: 1993- July 2001 S. Paternostro 1999 J. Razafindravonona David Stifel ARWPS 20 Information and Communication Technology, August 2001 Miria Pigato Poverty, and Development in sub-Saharan Africa and South Asia ARWPS 21 Handling Hierarchy in Decentralized Settings: September 2001 Navin Girishankar A. Governance Underpinnings of School Alemayehu Performance in Tikur Inchini, West Shewa Yusuf Ahmad Zone, Oromia Region ARWPS 22 Child Malnutrition in Ethiopia: Can Maternal October 2001 Luc Christiaensen Knowledge Augment The Role of Income? Harold Alderman ARWPS 23 Child Soldiers: Preventing, Demobilizing and November 2001 Beth Verhey Reintegrating ARWPS 24 The Budget and Medium-Term Expenditure December 2001 David L. Bevan Framework in Uganda ARWPS 25 Design and Implementation of Financial January 2002 Guenter Heidenhof Management Systems: An African H. Grandvoinnet Perspective Daryoush Kianpour B. Rezaian ARWPS 26 What Can Africa Expect From Its Traditional February 2002 Francis Ng Exports? Alexander Yeats ARWPS 27 Free Trade Agreements and the SADC February 2002 Jeffrey D. Lewis Economies Sherman Robinson Karen Thierfelder ARWPS 28 Medium Term Expenditure Frameworks: February 2002 P. Le Houerou From Concept to Practice. Preliminary Robert Taliercio Lessons from Africa ARWPS 29 The Changing Distribution of Public February 2002 Samer Al-Samarrai Education Expenditure in Malawi Hassan Zaman ARWPS 30 Post-Conflict Recovery in Africa: An Agenda April 2002 Serge Michailof for the Africa Region Markus Kostner Xavier Devictor ARWPS 31 Efficiency of Public Expenditure Distribution May 2002 Xiao Ye and Beyond: A report on Ghana's 2000 S. Canagaraja Public Expenditure Tracking Survey in the Sectors of Primary Health and Education Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 21 Africa Region Working Paper Series Series # Title Date Author ARWPS 33 Addressing Gender Issues in Demobilization August 2002 N. de Watteville and Reintegration Programs ARWPS 34 Putting Welfare on the Map in Madagascar August 2002 Johan A. Mistiaen Berk Soler T. Razafimanantena J. Razafindravonona ARWPS 35 A Review of the Rural Firewood Market August 2002 Gerald Foley Strategy in West Africa P. Kerkhof, D. Madougou ARWPS 36 Patterns of Governance in Africa September 2002 Brian D. Levy ARWPS 37 Obstacles and Opportunities for Senegal's September 2002 Stephen Golub International Competitiveness: Case Studies Ahmadou Aly Mbaye of the Peanut Oil, Fishing and Textile Industries ARWPS 38 A Macroeconomic Framework for Poverty October 2002 S. Devarajan Reduction Strategy Papers : With an Delfin S. Go Application to Zambia ARWPS 39 The Impact of Cash Budgets on Poverty November 2002 Hinh T. Dinh Reduction in Zambia: A Case Study of the Abebe Adugna Conflict between Well Intentioned Bernard Myers Macroeconomic Policy and Service Delivery to the Poor ARWPS 40 Decentralization in Africa: A Stocktaking November 2002 Stephen N. Ndegwa Survey ARWPS 41 An Industry Level Analysis of Manufacturing December 2002 Professor A. Mbaye Productivity in Senegal ARWPS 42 Tanzania's Cotton Sector: Constraints and December 2002 John Baffes Challenges in a Global Environment ARWPS 43 Analyzing Financial and Private Sector January 2003 Abayomi Alawode Linkages in Africa ARWPS 44 Modernizing Africa's Agro-Food System: February 2003 Steven Jaffee Analytical Framework and Implications for Ron Kopicki Operations Patrick Labaste Iain Christie ARWPS 45 Public Expenditure Performance in Rwanda March 2003 Hippolyte Fofack C. Obidegwu Robert Ngong ARWPS 46 Senegal Tourism Sector Study March 2003 Elizabeth Crompton Iain T. Christie ARWPS 47 Reforming the Cotton Sector in SSA March 2003 Louis Goreux John Macrae Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 22 Africa Region Working Paper Series Series # Title Date Author ARWPS 48 HIV/AIDS, Human Capital, and Economic April 2003 Channing Arndt Growth Prospects for Mozambique ARWPS 49 Rural and Micro Finance Regulation in June 2003 William F. Steel Ghana: Implications for Development and David O. Andah Performance of the Industry ARWPS 50 Microfinance Regulation in Benin: June 2003 K. Ouattara Implications of the PARMEC LAW for Development and Performance of the Industry ARWPS 51 Microfinance Regulation in Tanzania: June 2003 Bikki Randhawa Implications for Development and Joselito Gallardo Performance of the Industry ARWPS 52 Regional Integration in Central Africa: Key June 2003 Ali Zafar Issues Keiko Kubota ARWPS 53 Evaluating Banking Supervision in Africa June 2003 Abayomi Alawode ARWPS 54 Microfinance Institutions' Response in June 2003 Marilyn S. Manalo Conflict Environments: Eritrea- Savings and Micro Credit Program; West Bank and Gaza ­ Palestine for Credit and Development; Haiti ­ Micro Credit National, S.A. AWPS 55 Malawi's Tobacco Sector: Standing on One June 2003 Steven Jaffee Strong leg is Better than on None AWPS 56 Tanzania's Coffee Sector: Constraints and June 2003 John Baffes Challenges in a Global Environment AWPS 57 The New Southern AfricanCustoms Union June 2003 Robert Kirk Agreement Matthew Stern AWPS 58a How Far Did Africa's First Generation Trade June 2003 Lawrence Hinkle Reforms Go? An Intermediate Methodology A. Herrou-Aragon for Comparative Analysis of Trade Policies Keiko Kubota AWPS 58b How Far Did Africa's First Generation Trade June 2003 Lawrence Hinkle Reforms Go? An Intermediate Methodology A. Herrou-Aragon for Comparative Analysis of Trade Policies Keiko Kubota AWPS 59 Rwanda: The Search for Post-Conflict Socio- October 2003 C. Obidegwu Economic Change, 1995-2001 AWPS 60 Linking Farmers to Markets: Exporting Malian October 2003 Morgane Danielou Mangoes to Europe Patrick Labaste J-M. Voisard AWPS 61 Evolution of Poverty and Welfare in Ghana in October 2003 S. Canagarajah the 1990s: Achievements and Challenges Claus C. Pörtner Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 23 Africa Region Working Paper Series Series # Title Date Author AWPS 62 Reforming The Cotton Sector in Sub- November 2003 Louis Goreux Saharan Africa: SECOND EDITION AWPS 63 (E) Republic of Madagascar: Tourism Sector November 2003 Iain T. Christie Study D. E. Crompton AWPS 63 (F) République de Madagascar: Etude du November 2003 Iain T. Christie Secteur Tourisme D. E. Crompton AWPS 64 Migrant Labor Remittances in Africa: Novembre 2003 Cerstin Sander Reducing Obstacles to Development Samuel M. Maimbo Contributions AWPS 65 Government Revenues and Expenditures in January 2004 Francisco G. Guinea-Bissau: Casualty and Cointegration Carneiro Joao R. Faria Boubacar S. Barry AWPS 66 How will we know Development Results June 2004 Jody Zall Kusek when we see them? Building a Results- Ray C. Rist Based Monitoring and Evaluation System to Elizabeth M. White Give us the Answer AWPS 67 An Analysis of the Trade Regime in Senegal June 2004 Alberto Herrou- (2001) and UEMOA's Common External Arago Trade Policies Keiko Kubota AWPS 68 Bottom-Up Administrative Reform: Designing June 2004 Talib Esmail Indicators for a Local Governance Scorecard Nick Manning in Nigeria Jana Orac Galia Schechter AWPS 69 Tanzania's Tea Sector: Constraints and June 2004 John Baffes Challenges AWPS 70 Tanzania's Cashew Sector: Constraints and June 2004 Donald Mitchell Challenges in a Global Environment AWPS 71 An Analysis of Chile's Trade Regime in 1998 July 2004 Francesca Castellani and 2001: A Good Practice Trade Policy A. Herrou-Arago Benchmark Lawrence E. Hinkle AWPS 72 Regional Trade Integration inEast Africa: August 2004 Lucio Castro Trade and Revenue Impacts of the Planned Christiane Kraus East African Community Customs Union Manuel de la Rocha AWPS 73 Post-Conflict Peace Building in Africa: The August 2004 Chukwuma Challenges of Socio-Economic Recovery and Obidegwu Development AWPS 74 An Analysis of the Trade Regime in Bolivia August 2004 Francesca Castellani in2001: A Trade Policy Benchmark for low Alberto Herrou- Income Countries Aragon Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 24 Africa Region Working Paper Series Series # Title Date Author Lawrence E. Hinkle AWPS 75 Remittances to Comoros- Volumes, Trends, October 2004 Vincent da Cruz Impact and Implications Wolfgang Fendler Adam Schwartzman AWPS 76 Salient Features of Trade Performance in October 2004 Fahrettin Yagci Eastern and Southern Africa Enrique Aldaz- Carroll AWPS 77 Implementing Performance-Based Aid in November 2004 Alan Gelb Africa Brian Ngo Xiao Ye AWPS 78 Poverty Reduction Strategy Papers: Do they December 2004 Rene Bonnel matter for children and Young people made Miriam Temin vulnerable by HIV/AIDS? Faith Tempest AWPS 79 Experience in Scaling up Support to Local December 2004 Jean Delion Response in Multi-Country Aids Programs Pia Peeters (map) in Africa Ann Klofkorn Bloome AWPS 80 What makes FDI work? A Panel Analysis of February 2005 Kevin N. Lumbila the Growth Effect of FDI in Africa AWPS 81 Earnings Differences between Men and February 2005 Kene Ezemenari Women in Rwanda Rui Wu AWPS 82 The Medium-Term Expenditure Framework: April 2005 Chukwuma The Challenge of Budget Integration in SSA Obidegwu countries AWPS 83 Rules of Origin and SADC: The Case for June 2005 Paul Brenton change in the Mid Term Review of the Trade Frank Flatters Protocol Paul Kalenga AWPS 84 Sexual Minorities, Violence and AIDS in July 2005 Chukwuemeka Africa Anyamele Ronald Lwabaayi Tuu-Van Nguyen, and Hans Binswanger AWPS 85 Poverty Reducing Potential of Smallholder July 2005 Paul B. Siegel Agriculture in Zambia: Opportunities and Jeffrey Alwang Constraints AWPS 86 Infrastructure, Productivity and Urban July 2005 Zeljko Bogetic Dynamics Issa Sanogo in Côte d'Ivoire An empirical analysis and policy implications AWPS 87 Poverty in Mozambique: Unraveling Changes August 2005 Louise Fox and Determinants Elena Bardasi, Katleen V. Broeck Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 25 Africa Region Working Paper Series Series # Title Date Author AWPS 88 Operational Challenges: Community Home August 2005 N. Mohammad Based Care (CHBC) forPLWHA in Multi- Juliet Gikonyo Country HIV/AIDS Programs (MAP) forSub- Saharan Africa AWPS 90 Kenya: Exports Prospects and Problems September 2005 Francis Ng Alexander Yeats AWPS 91 Uganda: How Good a Trade Policy September 2005 Lawrence E. Hinkle Benchmark for Sub-Saharan-Africa Albero H. Aragon Ranga Krishnamani Elke Kreuzwieser AWPS 92 Community Driven Development in South October 2005 David Everatt Lulu Africa, 1990-2004 Gwagwa AWPS 93 The Rise of Ghana''s Pineapple Industry November 2005 Morgane Danielou from Successful take off to Sustainable Christophe Ravry Expansion AWPS 94 South Africa: Sources and Constraints of December 2005 Johannes Fedderke Long-Term Growth, 1970-2000 AWPS 95 South Africa''s Export Performance: December 2005 Lawrence Edwards Determinants of Export supply Phil Alves AWPS 96 Industry Concentration in South African December 2005 Gábor Szalontai Manufacturing: Trends and Consequences, Johannes Fedderke 1972-96 AWPS 97 The Urban Transition in Sub-Saharan Africa: December 2005 Christine Kessides Implications for Economic Growth and Poverty Reduction AWPS 98 Measuring Intergovernmental Fiscal May 2006 Navin Girishankar Performance in South Africa David DeGroot Issues in Municipal Grant Monitoring T.V. Pillay AWPS 99 Nutrition and Its determinants in Southern July 2006 Jesper Kuhl Ethiopia - Findings from the Child Growth Luc Christiaensen Promotion Baseline Survey AWPS 100 The Impact of Morbidity and Mortality on September 2006 Zara Sarzin Municipal Human Resources and Service Delivery AWPS 101 Rice Markets in Madagascar in Disarray: September 2006 Bart Minten Policy Options for Increased Efficiency and Paul Dorosh Price Stabilization Marie-Hélène Dabat, Olivier Jenn-Treyer, John Magnay and Ziva Razafintsalama Madagascar: An Evaluation of the Welfare Impact of Higher Energy Prices 26 Africa Region Working Paper Series Series # Title Date Author AWPS 102 Riz et Pauvrete a Madagascar Septembre 2006 Bart Minten AWPS 103 ECOWAS- Fiscal Revenue Implications of April 2007 Simplice G. Zouhon- the Prospective Economic Partnership Bi Agreement with the EU Lynge Nielsen AWPS 104(a) Development of the Cities of Mali June 2007 Catherine Challenges and Priorities Farvacque-V. Alicia Casalis Mahine Diop Christian Eghoff AWPS 104(b) Developpement des villes Maliennes June 2007 Catherine Enjeux et Priorites Farvacque-V. Alicia Casalis Mahine Diop Christian Eghoff AWPS 105 Assessing Labor Market Conditions In June 2007 David Stifel Madagascar, 2001-2005 Faly H. Rakotomanana Elena Celada AWPS 106 An Evaluation of the Welfare Impact of June 2007 Noro Andriamihaja Higher Energy Prices in Madagascar Giovanni Vecchi WB21847 C:\Documents and Settings\WB21847\My Documents\Working Paper Series\No.106- Assessing Labor Market Cond in Madag\AWPS 106-REVISED.na gv madagascar9.17.07pw.doc 09/17/2007 5:20:00 PM