Gainers and Losers from Trade Reform in Morocco Martin Ravallion and Michael Lokshin1 Development Research Group, World Bank Abstract: We use Morocco's national survey of living standards to measure the short-term welfare impacts of prior estimates of the price changes attributed to various agricultural trade reform scenarios for de-protecting cereals -- the country's main foodstaple. We find small impacts on mean consumption and inequality in the aggregate. There are both winners and losers and (contrary to past claims) the rural poor are worse off on average after de-protection. We decompose the aggregate impact on inequality into a "vertical" component (between people at different pre-reform welfare levels) and a "horizontal" component (between people at the same pre-reform welfare). There is a large horizontal component, which dominates the vertical impact of full de-protection. The diverse impacts reflect a degree of observable heterogeneity in consumption behavior and income sources, with implications for social protection policies. World Bank Policy Research Working Paper 3368, August 2004 The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Policy Research Working Papers are available online at http://econ.worldbank.org. 1 For helpful discussions on this topic we are grateful to Touhami Abdelkhalek, Jennie Litvack, Rachid Doukkali and Chris Ward. The assistance of Nithin Umapathi and Dimitri Kaltsas is gratefully acknowledged. 2 1. Introduction As a water-scarce country, Morocco does not have much natural advantage in the production of water-intensive crops such as most cereals, including wheat, which is used to produce the country's main food staples. The desire for aggregate self-sufficiency in the production of food staples has led in the past to governmental efforts to foster domestic cereal production, even though cereals can be imported more cheaply. Since the 1980s, cereal producers have been protected by tariffs on imports as high as 100%. There have been concerns that the consequent reallocation of resources has hurt consumers and constrained the growth of production and trade. Reform to the current incentive system for cereals has emerged as an important issue on the policy agenda for Morocco (World Bank, 2003). The major obstacles to reform stem from concerns about the impacts on household welfare, particularly for the poor. There has been very little careful research into who will gain and who will lose from such reforms. Nonetheless, there has been much debate about the equity implications. It is generally agreed that urban consumers are likely to gain from lower cereal prices. More contentious are the welfare distributional impacts in rural areas. Defenders of the existing protection system have argued that there will be large welfare losses to the rural economy from trade reform. Critics have argued against this view, claiming that the bulk of the rural poor tend to be net consumers, and so lose out from the higher prices due to trade protection. They argue that the rural poor are likely to gain from the reform, while it will be the well off in rural areas who tend to be net producers who will lose; see for example, Abdelkhalek (2002) and World Bank (2003). This paper studies the household welfare impacts of the relative price changes induced by specific trade policy reform scenarios for cereals in Morocco. Past analyses of the welfare 3 impacts have been highly aggregated, focusing on just one or a few categories of households. Here we estimate impacts across 5,000 sampled households in the Morocco Living Standards Survey for 1998/99. This allows us to provide a detailed picture of the welfare impacts, so as to better inform discussions of the social protection policy response to trade liberalization. Past approaches to studying the welfare impacts of specific trade reforms have tended to be either partial equilibrium analyses, in which the welfare impacts of the direct price changes due to tariff changes are measured at household level, or general equilibrium analyses, in which second-round responses are captured in a theoretically consistent way but with considerable aggregation across household types. In general terms, the economics involved in both approaches is well known. And both approaches have found numerous applications. We combine these two approaches. In particular, the price changes induced by the trade- policy change are simulated from a general equilibrium analysis done for a Joint Government of Morocco and World Bank Working Group. We take the methods and results of that analysis as given and carry them to the Moroccan Living Standards survey. Our approach respects the richness of detail available from a modern integrated household survey, allowing us to go well beyond the highly aggregative types of analysis one often finds. We not only measure expected impacts across the distribution of initial levels of living, but we also look at how they vary by other characteristics, such as location. We are thus able to provide a reasonably detailed "map" of the predicted welfare impacts by location and socio-economic characteristics. In studying the distributional impacts of trade reform we make a distinction between the "vertical impact" and "horizontal impact." The former concerns the way the mean impacts vary with level of pre-reform income -- how the reform affects people at different pre-reform incomes. The horizontal impact relates to the disparities in impact between people at the same 4 pre-reform income. As argued in Ravallion (2004), many past discussions of the distributional impacts of trade and other economy-wide reforms have tended to focus more on the vertical impacts, analogously to standard practices in studying the "benefit incidence" of tax and spending policies. However, as we will demonstrate here, this focus may well miss an important component of a policy's distributional impact, arising from the horizontal dispersion in impacts at given pre-reform incomes. We show how the impact of a policy on a standard inequality measure can be straightforwardly decomposed into its vertical and horizontal components. The former tells us how much of the change in total inequality can be accounted for by the way in which mean impacts conditional on pre-reform income vary with the latter. If there is no difference in the proportionate impact by level of income, then the vertical component is zero. The horizontal component tells us the contribution of the deviations in impacts from their conditional means. Only when the impact of the reform is predicted perfectly by pre- reform income will the horizontal component be zero. We study the relative importance of these two components of our predicted distributional impact of trade reform in Morocco. The following section discusses our approach in general terms. Section 3 presents our results in detail, while section 4 reviews the main findings. 2. Measuring and explaining the welfare impacts of reform using micro data We use pre-existing estimates of the household-level welfare impacts of the price changes generated by a computable general equilibrium (CGE) model. The CGE analysis generates a set of price changes; these embody both the direct price effects of the trade-policy change and indirect effects on the prices of both traded and non-traded goods once all markets respond to the reform. Standard methods of first-order welfare analysis are used to measure the gains and losses at household level. 5 Our focus here is very much on the short-term welfare impacts. In keeping with the limitations of the preceding general equilibrium analysis, our approach does not capture the dynamic effects of trade reform through labor market adjustment and technological innovation. Nor does it capture potential gains to the environment.2 The specifics of our approach to estimating welfare impacts at the household level can be outlined as follows.3 Each household has preferences over consumption and work effort represented by the utility function ui(qi ,Li) where qi is a vector of the quantities of d d commodities demanded by household i and Li is a vector of labor supplies by activity, including supply to the household's own production activities.4 The household is assumed to be free to choose its preferred combinations of qi and Li subject to its budget constraint. d The household owns a production activity that generates a profit i(pi ) = max[pi qi - ci(qi )] where pi is the vector of supply prices, and ci(qi ) is the s s s s s s household-specific cost function.5 The indirect utility function of household i is given by: vi[pi , pi , wi ] = max[ui(qi , Li ) pi qi = wiLi + i( pi )] s d d d d s (1) (qi ,Li ) d where pi is the price vector for consumption, wi is the vector of wage rates. d 2 Though it is not a subject of the present analysis, arguments are also made about adverse environmental impacts arising from the expansion of protected cereal production into marginal areas, It is claimed that scarce water resources have also been diverted into soft wheat production. For further discussion see World Bank (2003). 3 There are many antecedents of our approach in the literatures on both tax reform and trade reform, though there are surprisingly few applications to point to in the ex ante assessment of actual reform proposals. For another example see Chen and Ravallion (2004). Hertel and Reimer (2004) provide a useful overview of the strengths and weaknesses of alternative approaches to assessing the welfare impacts of trade-policies, including references to empirical examples for developing countries. 4 We make the standard assumptions that goods have positive marginal utilities while labor supplies have negative marginal utilities. 5 On can readily include input prices in this cost function; see Chen and Ravallion (2004) for a more general formulation. In the present context this makes no difference to the subsequent analysis so we subsume factor prices in the cost function to simplify notation. 6 We take the predicted price impacts from the CGE model as given for the analysis of household-level impacts. In measuring the impacts we are constrained of course by the data, which do not include prices and wages. However, this limitation does not matter to calculating a first-order approximation to the welfare impact in a neighborhood of the household's optimum. Taking the differential of (1) and using the envelope property (whereby the welfare impacts in a neighborhood of an optimum can be evaluated by treating the quantity choices as given), the gain to household i (denoted gi ) is given by the money metric of the change in utility: gi dui m s s dpij s d d dpij d n dwk ) (2) vi = [pijqij s - pijqij ]+ (wk Lsik j=1 pij pij d k=1 wk where vi is the marginal utility of income for household i (the multiplier on the budget constraint in equation 1) and Lsik is the household's "external" labor supply to activity k. (Notice that gains in earnings from labor used in own production are exactly matched by the higher cost of this input to own-production.) The proportionate changes in prices are weighted by their corresponding expenditure shares; the weight for the proportionate change in the j'th selling price is pijqij , the revenue (selling value) from household production activities in sector j; s s similarly - pij qij is the (negative) weight for demand price changes and wkLsik is the weight for d d changes in the wage rate for activity k. The term pijqij - pij qij gives (to a first-order s s d d approximation) the welfare impact of an equi-proportionate increase in the price of commodity j. Equation (2) is the key formula we will use for calculating the welfare impacts at household level, given the predicted price changes. In the specific model we will use (as discussed later), real wage rates are fixed. So the last term on the right hand side of (2) drops out. (We discuss likely implications of relaxing this assumption in section 3.5.) 7 Notice that by applying the calculus in deriving (2) we are implicitly assuming small changes in prices. Relaxing this requires more information on the structure of the demand and supply system; see for example Ravallion and van de Walle (1991). This would entail considerable further effort, and the reliability of the results will be questionable given the aforementioned problem of incomplete price and wage data. For the same reason, we will have little choice but to largely ignore geographic differences in the prices faced, or in the extent to which border price changes are passed on locally. Having estimated the impacts at household level, we can study how they vary with pre- reform welfare, and what impact the reform has on poverty and inequality. Let yi denote the pre-reform welfare per person in household i while yi = yi + gi is its post-reform value, where * gi is the gain to household i. (Ideally, yi will be an exact money-metric of utility, though in practice it can be expected that it is an approximation given omitted prices or characteristics.) The distribution of post-reform welfare levels is y1 , y2,...yn . By comparing standard summary * * * measures of poverty or inequality for this distribution with those for the pre-reform distribution, y1 , y2,...yn , we can assess overall impacts. Of obvious interest is to see how the gains vary with pre-reform welfare. Is it the poor who tend to gain, or is it middle-income groups or the rich? However, it is important to recognize that the assignment of impacts to the pre-reform distribution is very unlikely to be a degenerate distribution, with no distribution of its own. There will almost certainly be a dispersion in impact at given pre-reform welfare. This will arise from (observable and unobservable) heterogeneity in characteristics and prices. It could also arise from errors in the 8 welfare measure. Averaging across the distribution of impacts at given pre-reform welfare, one can calculate the conditional mean impact given by: gi = Ei(gi y = yi ) c (3) where the expectation is formed over the conditional distributions of impacts. By including a subscript i in the expectations operator in (3), we allow the possibility that the horizontal dispersion in impacts is not identically distributed. In our empirical implementation, equation (3) will be estimated using a non-parametric regression. Taking these observations a step further, we can think of the overall impact on inequality as having both vertical and horizontal components.6 This is straightforward for the mean log deviation (MLD) -- an inequality measure known to have a number of desirable features.7 The mean log deviation defined on the distribution of post-reform welfares y1 , y2,...yn is given by: * * * I* = 1 n ln( * (4) n y* / yi ) i=1 n where y* = yi / n is mean post-reform welfare. Similarly, * i=1 I = 1 n ln( (5) n y / yi ) i=1 is the pre-reform MLD. (In both (4) and (5) it is assumed that yi > 0 and yi > 0 for all i. Thus * I*- I is the change in inequality attributable to the reform. The proposed decomposition of the overall change in inequality can then be written as: 6 Antecedents to this type of decomposition can be found in the literature on horizontal equity in taxation. In the context of assessing a tax system, Auerbach and Hassett (2002) show how changes in an index of social welfare can be decomposed into terms reflecting changes in the level and distribution of income, the burden and progressivity of the tax system and a measure of the change in horizontal equity. 7 For further discussion of the MLD see Bourguignon (1979) and Cowell (2000). MLD is a member of the General Entropy class of inequality measures. 9 I* - I = 1 n 1+ g / y = 1 n 1+ g / y + 1 n 1+ gi / yi c (6) n ln1+gi c i=1 / yi n ln1+gi i=1 / yi n ln1+gi i=1 / yi vertical component + horizontal component The vertical component is the contribution to the change in total inequality (I*- I ) of the way in which mean impacts vary with pre-reform welfare levels. If there is no difference in the proportionate impact by level of welfare ( gi / yi = g / y for all i) then the vertical component is c zero. The horizontal component is the contribution of the deviations in impacts from their conditional means. If the impact of the reform is predicted perfectly by pre-reform welfare ( gi = gi for all i) then the horizontal component is zero. c We also want to try and explain the differences in impacts in terms of observable characteristics of potential relevance to social protection policies. The way we have formulated the problem of measuring welfare impacts above allows utility and profit functions to vary between households at given prices. To try to explain the heterogeneity in measured welfare impacts we can suppose instead that these functions vary with observed household characteristics. The indirect utility function becomes: vi( pi , pi , wi ) = v( pi , pi , wi, x1 , x2 ) = max[u(qi , Li, x1 ) pi qi - wiLi = i ] s d s d d d d i i i (7) where i = ( pi , x2 ) = max[pi qi - c(qi , x2 )]. Note that we allow the characteristics that s s s s i i influence preferences over consumption ( x1 ) to differ from those that influence the profits from i own-production activities ( x2 ). i The gain from the price changes induced by trade reform, as given by equation (7), depends on the consumption, labor supply and production choices of the household, which depend in turn on prices and characteristics, x1 and x2 . For example, households with a higher i i proportion of children will naturally spend more on food, so if the relative price of food changes 10 then the welfare impacts will be correlated with this aspect of household demographics. Similarly, there may be differences in tastes associated with stage of the life cycle and education. There are also likely to be systematic covariates of the composition of welfare. Generically, we can now write the gain as: gi = g( pi , pi ,wi, x1 , x2 ) s d i i (8) However, we do not observe the household-specific wages and prices. So we must make further assumptions. In explaining the variation across households in the predicted gains from trade reform we assume that: (i) the wage rates are a function of prices and characteristics as wi = w( pi , pi , x1 , x2 ) and (ii) differences in prices faced can be adequately captured by a d s i i complete set of regional dummy variables. Under these assumptions, and linearizing (8) with an additive innovation error term, we can write down the following regression model for the gains: gi = 1x1 + 2x2 + kDki + i i i (9) k where Dki =1 if household i lives in county k and Dki = 0 otherwise and i is the error term. 3. Measured welfare impacts of trade reform in Morocco 3.1 The predicted price changes and the survey data The price changes (implied by trade reform) we use here were generated by a CGE model that was commissioned by a joint working group of the Ministry of Agriculture, Government of Morocco, and the World Bank, as documented in Doukkali (2003). The model was constructed with the aim of realistically representing the functioning of the Moroccan economy around 1997- 98. The model was explicitly designed to assess the aggregate impacts of de-protecting cereals in Morocco. In addition to allowing for interactions between agriculture and the rest of the 11 economy (represented by six sectors), the model is quite detailed in its representation of the agricultural sector. It allows for 16 different crops or groups of crops, three different livestock activities, 13 major agro-industrial activities, six agro-ecological regions, and within each region the model distinguishes between rainfed agriculture and four types of irrigated agriculture. The model has two types of labor, both with fixed real wage rates. Four policy simulations are undertaken. The simulations then differ in the extent of the tariff reductions for cereals, namely 10% (Policy 1), 30% (Policy 2), 50% (Policy 3) and 100% (Policy 4). In all cases, the government's existing open-market operations, which attempt to keep down consumer prices by selling subsidized cereals, are also removed.8 The loss of revenue from a 50% tariff cut approximately equals the saving on subsidies. Table 1 gives the predicted prices changes for various trade liberalization scenarios, based on Doukkali (2003).9 As one would expect, the largest price impact is for cereals, though there are some non-negligible spillovers into other markets, reflecting substitutions in consumption and production and welfare effects on demand. Some of these spillover effects are compensatory. For example, some producer prices rise with the de-protection of cereals. The survey data set used here is the Enquête National sur le Niveau de Vie Ménages (ENNVM) for 1998 done by the government's Department of Statistics, which kindly provided the data set for the purpose of this study. This is a comprehensive multi-purpose survey 8 In addition to administering the tariffs on imported soft wheat, the Government of Morocco buys, mills and sells around one million tons of soft wheat in the form of low grade flour, which is sold on the open market to help consumers. 9 Rachid Doukkali kindly provided price predictions from the CGE model mapped into the categories of consumption and production identified in the survey. The production revenues were calculated from the survey data by matching these consumption categories to the variables containing information about household production of the corresponding goods. 12 following the practices of the World Bank's Living Standards Measurement Study.10 The ENNVM has a sample of 5,117 households (of which 2,154 are rural) spanning 14 of Morocco's 16 regions (the low density southernmost region -- the former Spanish Sahara -- was excluded). The sample is clustered and stratified by region and urban/rural areas. The survey did not include households without a fixed residence ("sans abris"). The survey allows calculation of a comprehensive consumption aggregate (including imputed values for consumption from own production). We used the consumption numbers calculated by the Department of Statistics. This is our money metric of welfare. Ideally this would be deflated by a geographic cost-of-living index, but no such index was available, given the aforementioned lack of geographic price data. 3.2 Implied welfare impacts at household level Tables 2a,b give the budget and income shares at mean points and the mean welfare impacts broken down by commodity based on the ENNVM; Table 2a is for consumption while 2b is for production. Notice how different consumption patterns are between urban and rural areas; for example, rural households have twice the budget share for cereals as urban households. Strikingly, while there is a 1.7% gain to urban consumers as a whole, this is largely offset by the general equilibrium effects through other price changes (Table 2a). Also notice that income obtained directly from production accounts for about one-quarter of consumption; the rest is labor earnings, transfers and savings. Of course in rural areas, the share is considerably higher, at 87%. And about one-third of this is from cereals.11 10 The survey's design and content are similar in most respects to the 1991 Living Standards Survey for Morocco documented in the LSMS web site: http://www.worldbank.org/lsms/ . 11 Notice that there is no income from meat recorded in the data. The most plausible explanation is that Moroccan farmers sell livestock to butchers or abattoirs rather than selling meat as such. Following conventional survey processing practices, livestock is treated as an asset, so that proceeds from the selling of livestock is not treated as income. This is questionable. As a test, we redid our main calculations using the survey data on the transaction in livestock, and adding net sales into income. This made negligible difference to the results. Details are available from the authors. 13 Table 3 summarizes the results on the implied welfare impacts. Our results indicate that the partial trade reforms have only a small positive impact on the national poverty rate, as given by the percentage of the population living below the official poverty lines for urban and rural areas used by the Government's statistics office.12 However, a larger impact emerges when we simulate complete de-protection (Policy 4). Then the national poverty rate rises from 20% to 22%. All four reforms entail a decrease in urban poverty (though less than 0.4% points) and an increase in rural poverty. (We will examine impacts over the whole distribution below.) Turning to the impacts on inequality in Table 3, we find that the trade reforms yield a small increase in inequality, with the Gini index rising from 0.385 in the base case to 0.395 with a complete de-protection of cereals (Policy 4). Impacts are smaller for the partial reforms (Policies 1-3). The overall per capita gain is positive for the smaller tariff reduction (Policy 1) but becomes negative for Policies 2, 3 and 4. As one would expect, there is a net gain to consumers and net loss to producers, though the amounts involved are small overall. There are small net gains in the urban sector for Policies 1-3. Larger impacts are found in rural areas, as we would expect. The mean percentage loss from complete de-protection is a (non-negligible) 5.7% in rural areas. Table 3 gave our results for the impact on poverty as estimated using the government's official poverty lines. It is important to test robustness to alternative poverty lines. For this purpose, we use the "poverty incidence curve," which is simply the cumulative distribution function up to a reasonable maximum poverty line. The results are given in Figure 1; to make the figure easier to read we focus on Policies 1 and 4. (The curves for Policies 2 and 3 are between these two.) 12 These have been updated using the CPI. The poverty lines were 3922 Dirham per year in urban areas and 3037 in rural areas. See World Bank (2001) for details. 14 We see that there is an increase in poverty overall from complete de-protection; this is robust to the poverty line and poverty measure used (within a broad class of measures; see Atkinson, 1987). The impact on poverty is almost entirely in rural areas; indeed, there is virtually no impact on urban poverty. However, in rural areas the results in Figure 1 suggest a sizeable impact on poverty from complete de-protection. The mean loss as a proportion of consumption for the poorest 15% in rural areas is about 10%. There is an increase in the proportion of the rural population living below 2000 Dirham per person per year from 6.2% to 9.9%; the proportion living below 3000 Dirham rises from 22.2% to 26.3%. (For the country as a whole, the poverty rate for the former poverty line rises from 2.8% to 4.4% under Policy 4, while it rises from 11.4% to 13.1% for the 3000 Dirham line.) Our finding of adverse impacts on the rural poor contradicts claims made by some observers who have argued that the rural poor tend to be net consumers of cereals, the commodity that incurs the largest price decrease with this trade reform (Table 1). We will return to this point when we study the welfare impacts further. Table 4 gives the mean impacts of Policy 4 by region, split urban and rural. Impacts in urban areas are small in all regions, with the highest net gain as a percentage of consumption being 1.3% in Tanger-Tetouan, closely followed by Tensift Al Haouz and Fes-Boulemane. The rural areas with largest mean losses from de-protection of cereals are Tasla Azilal, Meknes Tafil, Fes-Boulemane and Tanger-Tetouan. Table 4 also gives mean impacts for the poorest 15% in rural areas (in terms of consumption per person). When we focus on the rural poor defined this way, the region incurring the largest mean loss for rural households is Tanger-Tetouan, followed by Fes-Boulemane and Chaouia-Ouardigha. The contrast between the small net gains to the urban sector and net losses to the rural poor is most marked in Tanger-Tetouan. 15 To begin exploring the heterogeneity in welfare impacts, Figure 2 gives the cumulative frequency distributions of the gains and losses. To simplify the figure we again focus on Policies 1 and 4. We find that with complete de-protection (Policy 4) about 8.9% of the households incurred losses greater than 500 Dirhams per year (about 5% of overall mean consumption) while about 5% lose more than 1000 Dirhams per year. As one would expect, there is a "thicker tail" of negative gains for rural areas. About 16% of rural households lose more that 500 Dirhams and 10% lose more than 1000. In Figure 3 we plot the mean gains against percentiles of consumption per capita for Policies 1 and 4. We give both absolute gains/losses and gains as a percentage of the household's consumption. For policy 1, there is a tendency for the mean absolute gain to rise as one moves from the poorest percentile through to the richest, though the gradient is small. The mean proportionate gain is quite flat. For Policy 4, mean absolute impacts also rise up to the richest decile or so, but then fall. Proportionate gains follow the same pattern though (again) the gradient seems small. However, what is most striking from Figure 3 is the wide spread, particularly downwards (indicating losers from the reform). The variance in absolute impacts is particularly large at the upper end of the consumption distribution, though if anything the dispersion in proprotionate impacts tends to be greater at the other end of the distribution, amongst the poorest. In Figure 4 we provide a split between producers and consumers for Policy 4. As we would expect, to the extent that there is much impact on producers, they tend to lose, though not more so for poor producers than rich ones. For consumption we tend to see more gainers, and a higher variance in impact as one moves up the consumption distribution. However, we see that 16 the downward dispersion in total welfare impacts in Figure 3 is due more to the conditional variance in impacts through production than through consumption. There are two quite striking findings in these figures. Firstly, notice that there are sizeable losses on the production side amongst the poor. Granted, some large losses are evident for the high-income groups. But the claims that the poor do not lose as producers are clearly false. Furthermore, the poor are often not seeing compensatory gains as consumers. Secondly, it is notable that the results in Figures 3 and 4 indicate that the mean gains vary little with mean consumption. Focusing on the "poor" versus the "rich" is hardly of much interest in characterizing gainers and losers from this reform. The diversity in impacts tend to be "horizontal" in the distribution of income, meaning that there tend to be larger differences in impacts at given consumption than in mean impacts between different levels of consumption. Next we examine these two findings in greater detail. 3.3 Who are the net producers of cereals in Morocco? In the population as a whole, we find that 16% of households are net producers (value of cereals production exceeds consumption). These households are worse off from the fall in cereal prices due to de-protection. In rural areas, the proportion is 36%. However, the survey data do not support the claim that the rural poor in Morocco are on average net consumers of cereals. Figure 5 shows how producers and net producers are spread across the distribution of total household consumption per person in rural Morocco. We give both the scatter of points and the conditional means estimated using the local regression method.13 In the first (top left) panel we give the proportion of producers. Then we give the proportion of net producers (for whom production exceeds consumption of cereals in value 13 See Cleveland (1979). This is often referred to as LOWESS (Locally Weighted Scatter Plot Smoothing). We used the LOWESS program in STATA. 17 terms). Finally we give net production in value terms. In each case the horizontal axis gives the percentile of the distribution of consumption from poorest through to richest. We find that a majority of the rural poor produce cereals. Naturally much of this is for home consumption. However, even if we focus solely on net producers, we find that over one- third of the poorest quintile tend to produce more than they consume. Furthermore, the mean net production in value terms tends to be positive for the poor; in rural areas, the losses to poor producers from falling cereal prices outweigh the gains to poor consumers. More than any single feature of the survey data, it is this fact that lies at the heart of our finding that the rural poor lose from the reform. 3.4 Vertical versus horizontal impacts on inequality To measure the relative importance of the vertical versus horizontal differences in impact, we can use the decomposition method outlined in section 2. This decomposition requires an estimate of the conditional mean E(g y) , i.e., the regression function of g on y. We estimated this using the nonparametric local regression method of Cleveland (1979). Table 5 gives the results of this decomposition for each policy reform. For the small partial reform under Policy 1, the vertical component dominates, accounting for 73% of the impact on inequality. However, as one moves to the larger reforms, the horizontal component becomes relatively large. Indeed, we find that 119.8% of the impact of Policy 4 on inequality is attributable to the horizontal component, while -19.8% is due to the vertical component. So we find that the vertical component was inequality reducing for Policy 4, even though overall inequality rose (Table 5). There is clearly a high degree of horizontal inequality in measured impacts at given mean consumption. Some of this is undoubtedly measurement error, which may well become more 18 important for larger reforms. But some is attributable to observable covariates of consumption and production behavior, as discussed in section 2. In trying to explain this variance in welfare impacts, the characteristics we consider include region of residence, whether the household lives in an urban area, household size and demographic composition of the household, age and age- squared of the household head, education and dummy variables describing some key aspects of the occupation and principle sector of employment; Table 6 gives summary statistics on the variables to be used in the regressions. We recognize that there are endogeneity concerns about these variables, though we think those concerns are minor in this context, especially when weighed against the concerns about omitted variable bias in estimates that exclude these characteristics. Under the usual assumption that the error term is orthogonal to these regressors, we estimate equation (9) by ordinary least squares. The results are given in Table 7. Recall that these are averages across the impacts of these characteristics on the consumption and production choices that determine the welfare impact of given price and wage changes. This makes interpretation difficult. We view these regressions as being mainly of descriptive interest, to help isolate covariates of potential relevance in thinking about compensatory policy responses. Focusing first on the results for Policy 4, we find that larger losses from full de-protection of cereals are associated with families living in rural areas, that are relatively smaller (the turning point in the U-shaped relationship is at a household size of about one), have more wage earners, higher education, work in commerce, transport etc., and live in Chaouia-Ouardigha, Rabat, Tadla Azilal and Meknes Tafil. Recall that these effects stem from the way household characteristics influence net trading positions in terms of the commodities for which prices change. So, for example, it appears that larger families tend to consume more cereals, and so gain more from the 19 lower cereals prices. Results are similar for partial de-protection, though education becomes insignificant for Policy 1. In Table 8 we give an urban-rural breakdown of the regressions for Policies 1 and 4. There are a couple of notable differences. (Again we focus on Policy 4 in the interests of brevity.) We find significant positive effects of having more children and teenagers on the gains from trade reform in rural areas, presumably because such families are more likely to be cereal consumers. The education effect at higher levels of schooling is much more pronounced in urban areas. The effect of working in the transport and commerce sector is more statistically significant in urban areas, though this effect is still sizeable in rural areas. The regional effects are more statistically significant in urban areas than in rural areas. Of course there are still sizable regional differences in mean impacts in Table 8, though they are statistically less significant than we found in Table 7. In fact the quantitative magnitudes of the regional differences are just as large for rural areas in Table 8 as for urban plus rural areas in Table 7. It should not be forgotten that the results in Tables 7 and 8 are conditional geographic effects (conditional on the values taken by other covariates in the regressions). As we saw in Table 4, there are pronounced (unconditional) geographic differences in mean impacts in rural areas across different regions. Whether one draws policy lessons more from the conditional or unconditional effects depends on the type of policy one is using. If it is simply regional targeting, then of course the unconditional geographic effects in Table 4 will be more relevant. However, finer targeting by household characteristics, in combination with regional targeting, will call for the sorts of results presented in Tables 7 and 8. The share of the variance in gains that is accountable to these covariates is generally less than 10%. Values of R2 of this size are common in regressions run on large cross-sectional data 20 sets, though it remains true that a large share of the variance in impacts is not accountable to these covariates. (The exception to our low R2 is for Policy 1, for which almost half of the variance in gains across urban households is explained.) It must be expected that there is a sizable degree of measurement error in the gains, stemming from measurement error in the underlying consumption and production data. No doubt there are also important idiosyncratic factors in household-specific tastes or production choices. These regressions try to explain the variance in the gains from the reform. It is of interest to see if we can do any better in explaining the incidence of losses from reform amongst the poor. This is arguably of greater relevance to compensatory policies, which would presumably want to focus on poor losers. To test how well the same set of regressors could explain who was a poor loser from the reforms, we constructed a dummy variable taking the value unity if a rural household incurred a negative loss and was "poor"; to assure a sufficient number of observations taking the value unity we set the poverty line higher than the official line, namely at a consumption per person of 5,000 Dirham per year (rather than the official line of about 3,000). (We confined this to rural areas since that is where the losses are concentrated.) In the case of full de-protection (Policy 4), we find that about 14% of the variance in this measure can be explained by the set of regressors in Table 8, while for Policy 1 the share is 20%.14 While there are a number of identifiable covariates for identifying likely losers amongst the poor, it is also clear that there is a large share of the variance left unexplained. Another way to assess how effectively this set of covariates can explain the incidence of a net loss from reform amongst the poor is by comparing the actual value of the dummy variable described above with its predicted values from the model, using a cut-off probability of 0.5. For 14 The R2 for OLS regressions are 0.139 and 0.191 for Policy 4 and 1 respectively. Using instead a probit model to correct for the nonlinearity the pseudo R2's are 0.135 and 0.196. 21 Policy 4, there are 472 households out of 2,100 who were both poor and incurred a loss due to the reform. Of these the model could only correctly predict that this was the case for 18% (86 households). For Policy 1, the model prediction was correct for 27% of the 463 households who were both poor and were made worse off by the reform. Yet most forms of indicator targeting -- whereby transfers are contingent on readily observed variables, such as location -- would be based on similar variables to those we have used in our regressions; indeed, if anything targeted policies use fewer dimensions. This suggests that indicator targeting will be of only limited effectiveness in reaching those in greatest need. Self-targeting mechanisms that create incentives for people to correctly reveal their status (such as using work requirements) may be better able to do so. 3.5 Two caveats While the above results are suggestive, two limitations of our analysis should be noted. The first stems from the fact that the Doukkali (2003) model assumed fixed wage rates. While sensitivity to alternative labor market assumptions should be checked, we can speculate on the likely impacts of allowing real wages to adjust to the reforms. Here it can be argued that the export-oriented cash crops that will replace cereals will tend to be more labor intensive than cereals. Thus we would expect higher aggregate demand for the relatively unskilled labor used in agriculture, and hence higher real wages for relatively poorer groups. This will undoubtedly go some way toward compensating the rural poor, and may even tilt the vertical distributional impacts in favor of the poor. A second concern is that there may well be dynamic gains from greater trade openness that are not being captured by the model used to generate the relative price impacts; for example, trade may well facilitate learning about new agricultural technologies and innovation that brings 22 longer-term gains in farm productivity. These effects may be revealed better by studying time series evidence, combined with cross-country comparisons. 4. Conclusions The welfare impacts of de-protection in developing countries have been much debated. Some people have argued that external trade liberalizations are beneficial to the poor, while others argue that the benefits will be captured more by the non-poor. Expected impacts on domestic prices have figured prominently in these debates. The paper has studied the welfare impacts at household level of the changes in commodity prices attributed to a proposed trade reform, namely Morocco's de-protection of its cereals sector. This would entail a sharp reduction in tariffs, with implications for the domestic structure of prices and hence household welfare. The paper draws out the implications for household welfare of the previous estimates of the price impacts of reform done for a Joint Government of Morocco and World Bank Committee. Standard methods of first-order welfare analysis are used to measure the gains and losses at household level using a large sample survey. In a number of respects, our detailed household-level analysis throws into question past claims about the likely welfare impacts of this trade reform. In the aggregate, we find a small negative impact on mean household consumption and a small increase in inequality. There is a sizable, and at least partly explicable, variance in impacts across households. Rural families tend to lose; urban households tend to gain. There are larger impacts in some provinces than others, with the greatest negative impacts for rural households in Tasla Azilal, Meknes Tafil, Fes- Boulemane and Tanger-Tetouan. Mean impacts for rural households in these regions are 10% or more of consumption. There are clearly sizeable welfare losses amongst the poor in these specific regions. 23 The adverse impact on rural poverty stems in large part from the fact that the losses to the net producers of cereals outweigh the gains to the net consumers amongst the poor. Thus, on balance rural poverty rises. This contradicts the generalizations that have been made in the past that the rural poor in Morocco tend to be net consumers of grain, and hence gainers from trade reform. Yes, a majority are net consumers, but on balance the welfare impacts on the rural poor are negative. Our results lead us to question the high level of aggregation common in past claims about welfare impacts of trade reform. We find diverse impacts at given pre-reform consumption levels. This "horizontal" dispersion becomes more marked as the extent of reform (measured by the size of the tariff cut) increases. Indeed, we estimate that all of the impact of complete de- protection of cereals on inequality is horizontal rather than vertical; the vertical impact on inequality was actually inequality reducing. For a modest reform of a 10% cut in tariffs, the vertical component dominates, though there is still a large horizontal component. It is clear from our results that in understanding the social impacts of this reform, one should not look solely at income poverty and income inequality as conventionally measured; rather one needs to look at impacts along "horizontal" dimensions, at given income. We have been able to identify some specific types of households whose consumption and production behavior makes them particularly vulnerable. These results are suggestive of the targeting priorities for compensatory programs. The fact that we also find a large share of unexplained variance in impacts also points to the limitations of targeting based on readily observable indicators, suggesting that self-targeting mechanisms may also be needed. 24 Table 1: Predicted price changes due to agricultural trade reform in Morocco Sectors Consumers (% change in prices) Producers (% change in prices) Policy 1 Policy 2 Policy 3 Policy 4 Policy 1 Policy 2 Policy 3 Policy 4 Cereals and cereals products -3.062 -7.786 -12.811 -26.691 -2.858 -7.193 -11.744 -24.107 Fresh vegetables -0.714 -0.884 -1.051 -1.128 -0.580 -0.767 -0.871 -0.756 Fruits -0.637 -0.681 -0.683 -0.139 -0.429 -0.301 -0.104 0.843 Dairy products and eggs -0.472 -0.414 -0.257 0.751 -0.505 -0.487 -0.333 0.637 Meat (red and poultry) -0.320 -0.109 0.332 1.896 -0.306 -0.078 0.357 1.936 Sugar -0.200 0.100 0.400 1.300 -0.368 -0.378 -0.354 -0.094 Edible oils -0.671 -1.064 -1.405 -2.225 -0.632 -0.998 -1.336 -2.061 Fresh and processed fish 0.000 0.696 1.300 2.996 0.000 0.600 1.300 2.881 Other ag. and processed food -0.369 -0.402 -0.421 -0.635 0.268 1.294 2.475 5.388 Services 0.142 0.500 0.758 1.460 0.056 0.500 0.844 1.708 Energy, electricity and water -0.060 0.540 1.140 2.580 -0.051 0.549 1.149 2.597 Other industries 0.000 0.600 1.200 2.800 0.000 0.600 1.200 2.793 Note: The tariff cuts on imported cereals are 10%, 30%, 50% and 100% for Policies 1,2,3 and 4 respectively. Table 2a: Consumption shares and welfare impacts through consumption Consumption Policy 1 Policy 2 Policy 2 Policy 4 Shares National Cereals 0.084 0.2572 0.6540 1.0761 2.2420 Fresh vegetables 0.042 0.0297 0.0368 0.0437 0.0469 Fruits 0.022 0.0139 0.0148 0.0148 0.0030 Dairy products and eggs 0.032 0.0153 0.0134 0.0083 -0.0243 Meat (red and poultry) 0.112 0.0359 0.0122 -0.0373 -0.2129 Sugar 0.015 0.0030 -0.0015 -0.0060 -0.0195 Edible oils 0.032 0.0212 0.0336 0.0444 0.0703 Fresh and processed fish 0.013 0.0000 -0.0089 -0.0166 -0.0383 Ag. and processed food 0.101 0.0371 0.0405 0.0424 0.0640 Services 0.066 -0.0094 -0.0332 -0.0504 -0.0971 Energy, electricity, water 0.148 0.0089 -0.0799 -0.1688 -0.3819 Other industries 0.333 0.0000 -0.2000 -0.4001 -0.9335 Total 1.000 0.4127 0.4817 0.5506 0.7187 Urban Cereals 0.066 0.2034 0.5172 0.8510 1.7730 Fresh vegetables 0.037 0.0264 0.0327 0.0389 0.0417 Fruits 0.022 0.0139 0.0149 0.0149 0.0030 Dairy products and eggs 0.034 0.0160 0.0141 0.0087 -0.0255 Meat (red and poultry) 0.107 0.0342 0.0116 -0.0355 -0.2027 Sugar 0.011 0.0021 -0.0011 -0.0042 -0.0138 Edible oils 0.024 0.0163 0.0258 0.0341 0.0540 Fresh and processed fish 0.014 0.0000 -0.0096 -0.0180 -0.0414 Ag. and processed food 0.096 0.0354 0.0386 0.0404 0.0610 Services 0.067 -0.0095 -0.0333 -0.0505 -0.0973 Energy, electricity, water 0.155 0.0093 -0.0835 -0.1763 -0.3990 Other industries 0.368 0.0000 -0.2207 -0.4414 -1.0300 Total 1.000 0.3476 0.3067 0.2621 0.1231 Rural Cereals 0.136 0.4154 1.0565 1.7383 3.6217 Fresh vegetables 0.055 0.0394 0.0487 0.0579 0.0622 Fruits 0.021 0.0137 0.0146 0.0146 0.0030 Dairy products and eggs 0.028 0.0131 0.0114 0.0071 -0.0208 Meat (red and poultry) 0.128 0.0410 0.0139 -0.0425 -0.2427 Sugar 0.028 0.0056 -0.0028 -0.0112 -0.0364 Edible oils 0.053 0.0356 0.0564 0.0746 0.1181 Fresh and processed fish 0.010 0.0000 -0.0068 -0.0126 -0.0291 Ag. and processed food 0.115 0.0422 0.0461 0.0482 0.0728 Services 0.066 -0.0094 -0.0330 -0.0501 -0.0965 Energy, electricity, water 0.129 0.0077 -0.0694 -0.1466 -0.3317 Other industries 0.232 0.0000 -0.1392 -0.2785 -0.6498 Total 1.000 0.6042 0.9964 1.3993 2.4708 26 Table 2b: Percentage gains from each policy: Production component Production as a share of Policy 1 Policy 2 Policy 2 Policy 4 total consumption National Cereals 0.089 -0.2713 -0.6899 -1.1352 -2.3652 Fresh vegetables 0.053 -0.0381 -0.0471 -0.0560 -0.0601 Fruits 0.041 -0.0261 -0.0279 -0.0280 -0.0057 Dairy products and eggs 0.051 -0.0243 -0.0213 -0.0132 0.0386 Meat (red and poultry) 0.000 0.0000 0.0000 0.0000 0.0000 Sugar 0.000 0.0000 0.0000 0.0000 0.0000 Edible oils 0.025 -0.0169 -0.0268 -0.0354 -0.0560 Fresh and processed fish 0.000 0.0000 0.0000 0.0000 0.0000 Ag. and processed food 0.002 -0.0008 -0.0008 -0.0009 -0.0013 Services 0.000 0.0000 0.0000 0.0000 0.0000 Energy, electricity, water 0.000 0.0000 0.0000 0.0000 0.0000 Other industries 0.000 0.0000 0.0000 0.0000 0.0000 Total 0.262 -0.3774 -0.8139 -1.2687 -2.4498 Urban Cereals 0.010 -0.0311 -0.0792 -0.1303 -0.2716 Fresh vegetables 0.008 -0.0058 -0.0072 -0.0086 -0.0092 Fruits 0.016 -0.0105 -0.0112 -0.0112 -0.0023 Dairy products and eggs 0.007 -0.0031 -0.0027 -0.0017 0.0049 Meat (red and poultry) 0.000 0.0000 0.0000 0.0000 0.0000 Sugar 0.000 0.0000 0.0000 0.0000 0.0000 Edible oils 0.013 -0.0087 -0.0138 -0.0183 -0.0289 Fresh and processed fish 0.000 0.0000 0.0000 0.0000 0.0000 Ag. and processed food 0.000 0.0000 0.0000 0.0000 0.0000 Services 0.000 0.0000 0.0000 0.0000 0.0000 Energy, electricity, water 0.000 0.0000 0.0000 0.0000 0.0000 Other industries 0.000 0.0000 0.0000 0.0000 0.0000 Total 0.054 -0.0593 -0.1142 -0.1701 -0.3071 Rural Cereals 0.319 -0.9777 -2.4863 -4.0910 -8.5235 Fresh vegetables 0.186 -0.1329 -0.1645 -0.1955 -0.2099 Fruits 0.113 -0.0722 -0.0771 -0.0773 -0.0158 Dairy products and eggs 0.183 -0.0865 -0.0758 -0.0471 0.1375 Meat (red and poultry) 0.000 0.0000 0.0000 0.0000 0.0000 Sugar 0.000 0.0000 0.0000 0.0000 0.0000 Edible oils 0.061 -0.0409 -0.0649 -0.0857 -0.1357 Fresh and processed fish 0.000 0.0000 0.0000 0.0000 0.0000 Ag. and processed food 0.008 -0.0031 -0.0033 -0.0035 -0.0053 Services 0.000 0.0000 0.0000 0.0000 0.0000 Energy, electricity, water 0.000 0.0000 0.0000 0.0000 0.0000 Other industries 0.000 0.0000 0.0000 0.0000 0.0000 Total 0.870 -1.3131 -2.8719 -4.5000 -8.7527 27 Table 3: Household impacts of four trade reforms Baseline Policy 1 Policy 2 Policy 3 Policy 4 National Poverty rate (%) 19.61 20.01 20.33 21.04 22.13 Mean Log Deviation (x100) 28.50 28.92 29.00 29.14 29.17 Gini index 0.385 0.387 0.389 0.391 0.395 Per capita gain 0 6.519 -23.967 -54.816 -133.81 Mean % gain: price changes weighted by mean shares 0 -0.059 -0.513 -0.971 -2.141 Mean % gain: weighted by ratios of means (Tables 2a,b) 0 0.035 -0.332 -0.718 -1.731 Production gain 0 -32.078 -69.012 -106.308 -201.017 Consumption gain 0 38.598 45.046 51.492 67.207 Consumption per capita 9350.913 9357.433 9326.947 9296.097 9217.104 Urban Poverty rate (%) 12.19 12.05 11.96 12.05 11.76 Mean Log Deviation (x100) 25.49 25.41 25.32 25.23 24.93 Gini index 0.366 0.365 0.365 0.364 0.362 Per capita gain 0 35.518 24.8 13.747 -16.491 Mean % gain: price changes weighted by mean shares 0 0.357 0.374 0.394 0.442 Mean % gain: weighted by ratios of means (Tables 2a,b) 0 0.288 0.193 0.092 -0.184 Production gain 0 -6.308 -12.103 -17.793 -31.302 Consumption gain 0 41.826 36.903 31.54 14.811 Consumption per capita 12031.2 12066.72 12056 12044.95 12014.71 Rural Poverty rate (%) 28.28 29.31 30.10 31.54 34.25 Mean Log Deviation (x100) 17.47 17.82 17.82 17.93 17.76 Gini index 0.312 0.313 0.315 0.318 0.328 Per capita gain 0 -33.532 -91.321 -149.512 -295.845 Mean % gain: price changes weighted by mean shares 0 -0.634 -1.737 -2.855 -5.708 Mean % gain: weighted by ratios of means (Tables 2a,b) 0 -0.709 -1.875 -3.101 -6.282 Production gain 0 -67.671 -147.612 -228.562 -435.419 Consumption gain 0 34.139 56.291 79.049 139.574 Consumption per capita 5649.034 5615.502 5557.712 5499.522 5353.189 Note: All monetary units are Moroccan Dirham per year. MLD is only calculated over the set of households for whom consumption is positive. The mean % gains weighted by mean shares are simply the means across the sample of the % gains at household level. The second mean % gain is weighted by shares at the means points based on Tables 2a,b. 28 Table 4: Mean gains from Policy 4 by region Region Total Urban Rural Poorest 15% of rural households Oued Ed-Dahab-Lagouira -0.2 -0.2 . . Laayoune-Boujdour-Sakia El Hamra -0.34 -0.34 . . Guelmime Es-Semara -0.96 0.72 -3.47 -0.58 Souss-Massa-Daraa -1.31 0.42 -2.4 -3.09 Gharb-Chrarda-Beni Hssen -2.16 0.02 -3.86 0.1 Chaouia-Ouardigha -4.18 0.32 -8.31 -10.11 Tensift Al Haouz -0.87 1.12 -2.17 0.31 Oriental -0.87 0.38 -2.78 0.25 G.Casablanca 0.48 0.41 2.41 . Rabat-Salé-Zemmour-Zaer -0.59 0.33 -4.98 0.23 Doukala Abda -3.13 0.76 -5.92 -3.93 Tadla Azilal -6.93 -0.71 -11.04 -0.95 Meknes Tafil -4.89 -0.19 -11.35 -8.48 Fes-Boulemane -2.4 1.05 -11.52 -13.43 Taza-Al Hoceima-Taounate -4.47 -0.32 -5.78 -8.39 Tanger-Tetouan -2.94 1.31 -9.4 -22.03 Total -2.14 0.45 -5.71 -10.39 Note: Means formed over the household level % gains (equivalent to weighting proportionate price changes by mean shares). Table 5: Decomposition of the impact on inequality Policy 1 Policy 2 Policy 3 Policy 4 Vertical component 72.69 57.57 38.77 -19.77 Horizontal component 27.31 42.43 61.23 119.77 Total 100 100 100 100 Note: The decomposition is only implemented on the sample of households for whom both the baseline and post-reform consumption is positive. 29 Table 6: Summary statistics on explanatory variables in the regression analysis Mean Std. Dev Urban 0.580 binary Log household size 1.645 0.550 Log household size 2 3.009 1.621 Female headed household 0.170 binary If unemployed present 0.248 binary Number of wage earners 5.912 2.878 Share of children 0-6 0.140 0.162 Share of children 7-17 0.221 0.204 Share of elderly 60+ 0.120 binary Characteristics of the head Age of the head 0.505 0.143 Age of the head 2 0.275 0.155 Illiterate head 0.582 binary Incomplete primary school 0.100 binary Primary school completed 0.164 binary Low secondary school 0.058 binary Upper secondary school 0.059 binary University 0.036 binary Industry Not-employed 0.240 binary Industrie/B.T.P 0.004 binary Commerce/Transp./Commun./Admin. 0.273 binary Service Soci. 0.085 binary Autres services 0.064 binary Corps Exter. 0.125 binary Chomeur 0.012 binary Femme au foyeur/Eleve/Etudiant 0.037 binary Jeune enfant 0.009 binary Vielliard/Retraite/Rentiers 0.074 binary Infirme/malade 0.068 binary Autre inactifs 0.010 binary Regions Oued Ed-Dahab-Lagouira 0.012 binary Laayoune-Boujdour-Sakia El Hamra 0.014 binary Guelmime Es-Semara 0.023 binary Souss-Massa-Daraa 0.094 binary Gharb-Chrarda-Beni Hssen 0.058 binary Chaouia-Ouardigha 0.054 binary Tensift Al Haouz 0.100 binary Oriental 0.065 binary G.Casablanca 0.124 Binary Rabat-Salé-Zemmour-Zaer 0.081 Binary Doukala Abda 0.067 Binary Tadla Azilal 0.047 Binary 30 Meknes Tafil 0.072 binary Fes-Boulemane 0.051 binary Taza-Al Hoceima-Taounate 0.058 binary 31 Table 7: Regression of per capita gain/loss on selected household characteristics Policy 1 Policy 2 Policy 3 Policy 4 Coeff. s.e. Coeff. s.e. Coeff. s.e. Coeff. s.e. Urban 26.139*** 6.275 44.850*** 12.948 64.218** 20.068 113.714** 39.213 Log household size -57.242** 19.583 -78.454* 40.407 -100.548 62.626 -157.373 122.376 Log household size 2 77.337*** 16.806 167.523*** 34.678 260.865*** 53.746 508.026*** 105.023 Female headed household 2.502 7.431 4.072 15.333 5.605 23.765 9.161 46.438 If unemployed present 10.018* 5.909 23.344* 12.192 36.428* 18.896 67.997* 36.924 Number of wage earners -44.722*** 7.019 -101.428*** 14.484 -159.842*** 22.448 -313.541*** 43.865 Share of children 0-6 32.783* 17.72 89.774* 36.564 145.705* 56.67 277.637* 110.736 Share of children 7-17 25.070* 14.155 69.367* 29.206 113.738* 45.266 221.518* 88.453 Share of elderly 60+ -21.3 15.584 -23.551 32.155 -24.389 49.837 -24.334 97.385 Characteristics of the head Age of the head -38.511 108.759 -151.473 224.41 -272.681 347.809 -624.596 679.642 Age of the head 2 44.097 102.579 142.598 211.658 246.231 328.045 543.07 641.022 Household is literate only -8.871 7.983 -23.441 16.472 -38.257 25.53 -76.735 49.888 Incomplete primary education Reference Primary school completed -14.013* 6.757 -40.623** 13.942 -68.220** 21.608 -141.296*** 42.224 Low secondary school -12.98 10.4 -61.634** 21.458 -112.583*** 33.258 -250.335*** 64.989 Upper secondary school -12.462 10.775 -70.619** 22.233 -130.320*** 34.458 -286.333*** 67.333 University 2.575 13.527 -95.376*** 27.912 -197.887*** 43.26 -476.077*** 84.533 Industry Not-working/Agriculture Reference Industrie/B.T.P -3.71 36.465 -0.277 75.242 4.541 116.616 21.281 227.874 Commerce/Transport/ Communications/Admin. -59.926*** 8.198 -122.454*** 16.915 -185.113*** 26.216 -341.751*** 51.228 Service Soci. 4.424 10.036 17.18 20.707 30.536 32.094 66.804 62.714 Autres services -0.2 11.251 9.572 23.214 19.812 35.98 47.874 70.306 Corps Exter. 2.385 8.936 6.785 18.439 10.912 28.579 20.23 55.844 Chomeur 6.627 21.518 27.715 44.399 49.65 68.813 107.951 134.465 Femme au foyeur/Eleve/Etudiant 2.26 13.49 13.788 27.835 25.401 43.141 55.785 84.301 Jeune enfant 7.629 24.5 -3.891 50.553 -16.336 78.352 -51.207 153.104 Vielliard/Retraite/Rentiers 6.913 11.039 23.527 22.778 40.651 35.303 86.8 68.984 Infirme/malade 3.143 10.96 22.092 22.614 42.489 35.049 100.065 68.488 Autre inactifs -9.955 22.723 1.817 46.885 15.364 72.667 56.497 141.995 Regions Oued Ed-Dahab-Lagouira 19.216 22.51 -6.738 46.446 -34.818 71.986 -111.388 140.665 Laayoune-Boujdour-Sakia El Hamra -1.502 21.067 -20.145 43.47 -40.764 67.374 -98.323 131.652 Guelmime Es-Semara 9.666 16.639 11.901 34.333 12.774 53.212 12.391 103.979 Souss-Massa-Daraa -7.645 10.868 5.611 22.425 22.766 34.756 85.2 67.916 Gharb-Chrarda-Beni Hssen -10.087 12.229 -7.485 25.232 -3.592 39.107 10.494 76.418 Chaouia-Ouardigha -19.542 12.507 -49.255* 25.807 -81.319* 39.998 -169.114* 78.159 32 Tensift Al Haouz 2.964 10.696 14.527 22.071 27.258 34.207 65.274 66.842 Oriental -14.038 11.928 -19.198 24.612 -23.918 38.145 -31.056 74.539 G.Casablanca -3.322 10.429 -15.762 21.518 -28.418 33.35 -60.086 65.169 Rabat-Salé-Zemmour- Zaer -15.439 11.326 -33.817 23.371 -52.199 36.222 -97.061 70.78 Doukala Abda -13.169 11.76 -23.668 24.265 -34.315 37.607 -59.462 73.487 Tadla Azilal -55.774*** 13.093 -114.700*** 27.016 -174.099*** 41.872 -320.810*** 81.821 Meknes Tafil -37.594** 11.54 -74.192** 23.812 -111.929** 36.906 -209.391** 72.117 Fes-Boulemane -10.249 12.726 -15.356 26.259 -20.651 40.699 -33.326 79.528 Taza-Al Hoceima- Taounate 5.613 12.367 2.43 25.517 -2.415 39.549 -21.329 77.281 Tanger-Tetouan Reference Constant 144.096*** 34.638 247.104*** 71.472 354.469** 110.773 642.381** 216.458 R2 0.175 0.080 0.062 0.057 33 Table 8: Urban-rural split of regressions for per capita gains Urban Rural Policy 1 Policy 4 Policy 1 Policy 4 Coeff. s.e. Coeff. s.e. Coeff. s.e. Coeff. s.e. Log household size -32.840* 16.071 45.705 83.159 -89.255* 45.084 -527.017* 294.353 Log household size 2 40.492* 17.841 217.663* 92.32 79.415* 32.524 555.880** 212.348 Female headed household -2.696 6.018 -15.603 31.139 11.984 16.902 27.785 110.356 If unemployed present 2.138 4.668 25.238 24.154 11.086 14.482 35.299 94.551 Number of wage earners -23.972** 8.39 -143.745*** 43.414 -45.101*** 12.237 -321.182*** 79.894 Share of children 0-6 -15.648 15.206 25.903 78.686 95.815** 36.544 609.370* 238.601 Share of children 7-17 -10.44 11.986 -34.073 62.023 81.378** 29.771 622.563** 194.376 Share of elderly 60+ -17.696 13.328 4.67 68.967 -35.448 32.512 -167.42 212.274 Characteristics of the head Age of the head -26.02 96.18 -513.051 497.696 -82.081 216.7 -1.00E+03 1414.846 Age of the head 2 33.769 91.377 263.429 472.842 103.772 202.766 1129.226 1323.868 Household is literate only -10.567 6.965 -90.700* 36.042 -8.718 16.11 -75.293 105.182 Incomplete primary education Reference Primary school completed 0.157 5.566 -44.272 28.804 -31.613* 14.794 -270.881** 96.589 Low secondary school 6.416 7.632 -119.177** 39.494 -73.971* 31.399 -655.218** 205.005 Upper secondary school -5.731 7.551 -249.358*** 39.074 10.925 49.861 -46.655 325.547 University 9.241 9.282 -433.456*** 48.03 20.185 83.244 18.883 543.507 Industry Not-working/Agriculture Reference Industrie/B.T.P -4.779 25.641 7.254 132.684 56.769 124.939 366.598 815.737 Commerce/Transp./Commun./ Admin. -96.116*** 10.172 -444.047*** 52.634 -43.789** 15.445 -257.349* 100.843 Service Soci. -1.428 7.574 6.102 39.191 27.61 28.965 247.156 189.116 Autres services -4.7 9.133 6.023 47.259 21.228 25.434 161.257 166.061 Corps Exter. -2.611 6.884 -19.401 35.621 8.742 23.042 57.723 150.44 Chomeur -1.702 15.213 36.377 78.72 60.148 73.543 457.084 480.167 Femme au oyeur/Eleve/Etudiant -4.019 10.145 12.554 52.498 20.295 36.207 110.127 236.4 Jeune enfant -2.268 16.343 -129.322 84.567 107.247 152.23 720.704 993.92 Vielliard/Retraite/Rentiers 1.108 8.138 48.765 42.112 25.588 34.261 154.32 223.691 Infirme/malade 1.847 8.176 63.019 42.308 5.864 30.489 148.543 199.063 Autre inactifs -12.094 16.532 23.685 85.547 22.652 67.323 250.306 439.559 Regions Oued Ed-Dahab-Lagouira 21.2 15.068 -135.288* 77.973 Laayoune-Boujdour-Sakia El Hamra -2.496 14.153 -129.348* 73.236 Guelmime Es-Semara 7.558 13.813 -50.41 71.475 23.284 35.563 165.753 232.195 Souss-Massa-Daraa -1.425 10.023 -54.723 51.863 -8.417 21.371 211.302 139.535 Gharb-Chrarda-Beni Hssen -44.733*** 11.143 -204.020*** 57.663 17.31 23.762 208.808 155.141 Chaouia-Ouardigha -15.625 11.08 -89.734 57.333 -19.527 25.012 -201.804 163.304 Tensift Al Haouz -8.763 9.759 -37.2 50.5 8.732 21.097 147.015 137.74 Oriental -18.776* 9.806 -96.129* 50.74 -0.357 25.851 99.206 168.782 34 G.Casablanca -9.23 7.849 -112.350** 40.617 5.551 49.268 79.412 321.673 Rabat-Salé-Zemmour-Zaer -13.825 8.683 -118.444** 44.931 -36.873 30.677 -142.714 200.295 Doukala Abda -14.916 10.867 -80.126 56.232 -8.244 22.773 -3.679 148.687 Tadla Azilal -50.624*** 12.423 -213.855*** 64.285 -51.570* 24.832 -324.785* 162.13 Meknes Tafil -22.753* 9.622 -126.779* 49.79 -56.111* 24.782 -311.079* 161.8 Fes-Boulemane -11.946 9.954 -38.193 51.509 -2.002 30.661 -5.31 200.186 Taza-Al Hoceima-Taounate -20.264 13.982 -161.597* 72.352 16.747 22.229 80.917 145.137 Tanger-Tetouan Reference Constant 135.395*** 30.386 463.951** 157.234 162.613* 72.909 959.343* 476.029 R2 0.46 0.08 0.062 0.067 35 Figure 1: Impacts on poverty Total .6 Baseline Policy 1 Policy 4 enil .5 ytrevop .4 e th w .3 loeb lepoep .2 % .1 0 0 1000 2000 3000 4000 5000 Annual per capita consumption Urban Rural .6 .6 enil .5 enil .5 ytrevop .4 ytrevop .4 e e th th w .3 w .3 loeb loeb lepoep .2 lepoep .2 % .1 % .1 0 0 0 1000 2000 3000 4000 5000 0 1000 2000 3000 4000 5000 Annual per capita consumption Annual per capita consumption 36 Figure 2: Frequency distributions of gains/losses for Policies 1 and 4 Total Total 1 .02 Policy 1 .9 Policy 4 .8 .015 .7 itysned .6 itysned e tiv .5 .01 lau .4 mu ityilbabo C .3 Pr .005 .2 .1 0 0 -3000 -2000 -1000 -500 0 500 1000 -600 -400 -200 0 200 400 Urban Rural 1 1 .9 .9 .8 .8 .7 .7 itysned .6 itysned .6 e e tiv .5 tiv .5 lau .4 lau .4 mu mu C.3 C.3 .2 .2 .1 .1 0 0 -3000 -2000 -1000 -500 0 500 1000 -3000 -2000 -1000 -500 0 500 1000 Absolute gain pre capita Absolute gain per capita Figure 3: Absolute and proportionate gains for Policies 1 and 4 plotted against percentile of consumption Policy 1 Policy 4 1500 1500 1000 1000 500 500 atipacr 0 0 pessol -500 -500 in/ag -1000 -1000 teluosba -1500 -1500 -2000 -2000 -2500 -2500 -3000 -3000 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 25 25 10 10 5 5 0 0 tapiacrepss -5 -5 /loinag -20 -20 % -35 -35 -50 -50 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 Per capita consumption percentiles Per capita consumption percentiles 38 Figure 4: Production/consumption decomposition of the welfare impacts for Policy 4, plotted against percentile of consumption per person Policy 4, production Policy 4, consumption 1500 1500 1000 1000 500 500 0 0 itapacreps -500 -500 osl/niag -1000 -1000 luteosba-1500 -1500 -2000 -2000 -2500 -2500 -3000 -3000 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 Per capita consumption percentiles Per capita consumption percentiles 39 Figure 5: Net producers of cereals in the distribution of total consumption per person in rural areas 1 .75 sercudorplaer .5 ce % .25 0 0 10 20 30 40 50 60 70 80 90 100 1 1 slareecfosre .75 slareec noipt .75 ofnoipt um .5 .5 umsnoc totfo nsocla .25 producten nusim i on 0 %.25 onit proporta ducorP sa -.25 0 -.5 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 Consumption per capita percentiles Consumption per capita percentiles 40 References Abdelkhalek, Touhami, 2002, "De I'impact de la Libéralization du Marche Céréalier Marocain: Enseignements à Partir d'un Modèle de Comportement des Ménages Rureaux, Critique Economique 7 : 105-114. Atkinson, Anthony B., 1987, "On the Measurement of Poverty," Econometrica 55: 749-764. Auerbach, Alan J., and Kevin A. Hassett, 2002, "A New Measure of Horizontal Equity," American Economic Review 92(4): 1116-1125. Bourguignon, Francois, 1979, "Decomposable Inequality Measures," Econometrica, 47: 901- 920. Chen, Shaohua and Martin Ravallion, 2004, "Welfare Impacts of Morocco's Accession to the WTO," World Bank Economic Review, in press. Cleveland, William S., 1979, "Robust Locally Weighted Regression and Smoothing Scatter Plots," Journal of the American Statistical Association 74: 829-36. Cowell, Frank, 2000, "Measurement of Inequality," in A.B. Atkinson and F. Bourguignon (eds) Handbook of Income Distribution, Amsterdam: North-Holland. Doukkali, Rachid, 2003, "Etude de Effets de la Libéralisation des Céréales: Resultats des Simulaions à L'Aide d'un Modèle Equilibre Général Calculable," Joint Report of the Ministry of Agriculture and the World Bank. Hertel, Thomas W. and Jeffrey Reimer, 2004, "Predicting the Poverty Impacts of Trade Liberalization: A Survey," Department of Agricultural Economics, Purdue University. Ravallion, Martin, 2001, "Growth, Inequality and Poverty: Looking Beyond Averages," World Development, 29(11), 1803-1815. ______________, 2004, "Competing Concepts of Inequality in the Globalization Debate," Policy Research Working Paper, World Bank, Washington DC. 41 Ravallion, Martin and Dominique van de Walle, 1991, "The Impact of Food Pricing Reforms on Poverty: A Demand Consistent Welfare Analysis for Indonesia," Journal of Policy Modeling, 13: 281-300. World Bank, 2001, Kingdom of Morocco: Poverty Update. Washington DC, World Bank. __________, 2003, Kingdom of Morocco: Findings and Recommendations of the Cereals Working Group: A Critical Review, Washington DC, World Bank. 42