Trade Adjustment Policies and Income Distribution in Three Archetype Developing Economies FUtE C0SW ~SWP44 World Bank Staff Working Paper No. 442 December 1980 Prepared by: Jairne de Melo (Consultant) Sherman Robinson Development Economics Department Copyright ® 1980 The World Bank 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. The views and interpretations in this document are those of the au and should not be attributed to the World Bank, to its affiliated organizations, or to any individual acting in their behalf. Trade Adjustment Policies and Income Distribution in Three Archetype Developing Economies by Jaime de Melo Georgetown University and Sherman Robinson World Bank December 1980 Presented at the Econometric Society World Congress, Aix-en-Provence, August 28 - September 2, 1980 de Melo wishes to acknowledge financial support from the Agency for International Development and from the World Bank. Our work has greatly benefitted from other work undertaken jointly with.Kemal Dervis, and we should also-like to thank Hollis Chenery, Martha de Melo, Joan Nelson, and.Larry ,Westphal for comments on an earlier draft. Neither,they, AID, nor the World .'Bank are.responsible.for any views and interpretations,.expressed in this. paper. The views and interpretations in this document are those of the authors and should not be attributed to the World Bank, to its affiliated organizations, or to any individual acting in their behalf. WORLD BANK Staff Working Paper No. 442 December 1980 TRADE ADJUSTMENT POLICIES AND INCOME DISTRIBUTION IN THREE ARCHETYPE DEVELOPING ECONOMIES This paper explores quantitatively the macroeconomic and distribu- tional impacts on non-oil producing, semi-industrial developing countries of external shocks originating in the world economy -- in particular, rising costs of imports and shrinking export markets. The empirical effects of the same external shock are modelled for three different archetype economies: a primary exporter, a manufacturing exporter, and a closed economy. Three different policy-adjustment regimes are considered: devaluation, premium rationing of imports (import licenses), and premium rationing in an environment with a fixed real wage for unskilled labor. The results show that differences in resource endowments and initial conditions strongly affect the impact of a given adjustment policy on both the economic structure and the distribution of income. By making simple assumptions about the way socioeconomic groups operate in the political process, the paper also examines how the struggle between the gainers and losers is likely to affect the policy regime to be chosen. An index of "political feasibility" is defined using -a variety of weighting schemes reflecting different views of the relative political power of different socioeconomic groups. The analysis provides a useful starting point for analyzing the political, as well as economic, implications of different economic policies. Prepared by: Jaime de Melo (Consultant) Sherman Robinson Development Economics Department Copyright ® 1980 The World Bank 1818 H Street, N.W. Washington, D.C. 20433 U.S.A. Table of Contents Page I. Introduction II. External Shocks and Macroeconomic Adjustment 5 III. The Three Archetype Economies 8 IV. The Macroeconomic Impact of an External Shock 18 V. Policy Choice and the Distribution of Income 27 VI. Class Conflict and Policy Choice 40 VII. Conclusions 51 References Appendix A: Constructing Three Consistent Social Accounting Matrices Appendix B: The Equations of the Model List of Tables Table 1: Socioeconomic Classification of Individuals . . . . . . . 11 Table 2: Three Archetype Economies: Structure of Production and Trade in the Base Run . . . . 14 Table 3: Three Archetype Economies: Sectoral Trade Ratios in the Base Run . . . . . . . . . 16 Table 4: Three Archetype Economies: Composition of the Labor Force . . . . . . . . . . . . . 17 Table 5: The Macroeconomic Impact of the External Shock .20 Table 6: Volume and Composition of Investment . . . . . . . . . . 23 Table 7: The Impact of the External Shock on Selected Relative Price Indices . . . . . . . . . . . . . 31 Table 8: Cost of Living Indices . . . . . . . . . . . . . . . . . 33 Table 9: Group Mean Real Incomes . . . . ... . . . . . . . . . . 34 Table 10: Group Shares in Total Income . . . . . . . . . . . . . . 36 Table 11: Analysis of Poverty . . . . . . . . . . . . . . . . . . 38 Table i2: Measures of the Distribution of Political Pzwer . . . . . 44 Table 13: Policy Comparisons . . . . . . . . . . . . . . . . . . . 47 Table 14: Policy Rankings . . . . . . . . . . . . . . . . . . . . 49 - ii - I. Introduction Changes in the international economy in the decade of the 1970's have resulted in great structural changes in developing and developed eco- nomies. In the two decades before 1973, world trade grew at an unprecedented rate of 8 percent a year. Since 1973, the rate of growth has halved. 1/ The reasons for this decline are clearly linked to the shocks that have occurred in the world economy since 1973, including the recession in the developed countries. These events, and the specter of growing restrictionist policies in the developed countries, have compounded the uncertainty about the future so that many developing countries have cut back investment in exporting industries and their governments have shown an increasing unwillingness to pursue outward-looking policies. If the trend growth in world trade is not resumed, either through increased demand by the developed countries or through increased trade among developing countries, then the role of international trade as an engine of growth will be seriously impaired. In addition to the relative decline in export markets, oil-importing countries have been seriously affected by the increase in energy prices. There has also been a general increase in the relative prices to developing countries of other essential inputs such as capital goods as a result of the general inflation experienced by developed countries. The shock of increased prices of crucial inputs for which substitutes are not really possible in the medium term has been especially severe for the developing countries. The continuation into the future of these effects, on both 1/ See Lewis (1980). - 2 - the export and import sides, will face developing countries with the need for further structural adjustments in their economies of the kind they have experienced since 1973. The nature of the structural adjustment required to restore external and internal equilibrium after a shock involves significant changes in rela- tive prices and incomes within the country. The first, and by no means least painful, adjustment requires that aggregate expenditure be cut to reflect the fall in the foreign terms of trade. Second, restoring equilibrium calls for a reallocation of resources from the more exportable sectors towards the sectors in which there is scope for import substitution. Finally, the real exchange rate has to adjust to attain external and internal balance by means of a shift of resources between sectors producing home goods and sectors producing tradables. In market economies, these resource shifts are achieved through changes in relative prices via the working of product and factor markets. In turn, the changes in product prices, wages and profits alter the distribution of income among socioeconomic groups. Finally, these distributional shifts take place in an environment where different socioeconomic groups attempt to maintain or improve their relative position in society thereby affecting the choice of government policy. This paper is inspired by the nature of the external shocks that have hit developing countries after 1973 and by our perception that required structural adjustments involve significant distributional shifts that have an important impact on the choice of trade adjustment policies. Our intent is to explore quantitatively the links between adjustment policies, economic structure and the distribution of income. The analysis is carried out with a - 3 - computable general equilibrium (CGE) model which provides a useful framework to analyze how the workings of the market mechanism, in conjunction with public policy, determine the returns to factors of production and the pur- chasing power of income. 1/ Rather than basing the study on the historical, political, and economic circumstances in a particular country, we have sought to investigate how a common external shock and trade policy response might have different effects in different environments. We create three representa- tive, or archetype, economies and subject them to the same set of policy reactions to an identical external shock. To keep the analysis and interpre- tation of results manageable, we emphasize distinguishing characteristics of our archetype economies and make them quite similar in other non-essential aspects. How the data sets are constructed is described in Appendix A. The approach followed here has some important implications for the analysis of development policy. First, it provides limited, but systematic, evidence of the extent to which differences in resource endowments and initial conditions reflected in differing economic structures affect the impact of a given economic policy. For instance, is an adverse movement in foreign terms of trade likely to affect more adversely a closed or an open economy and, if the economy is open, will it require a greater adjustment if it is a manufac- turing exporter or if it is a primary exporter? One can also investigate whether, in different economic environments, similar or different policies are preferable from the point of view of their impact on the distribution of income. 1/ The equations describing the model are given in Appendix B. For a further description of CGE models see Dervis, de Melo and Robinson (1981) and for an application to income distribution see Adelman and Robinson (1978) or de Melo and Robinson (1980). A second interesting set of questions relates to political-economy issues. By investigating the effects of policies on the distribution of income across a typology of countries, one can draw inferences about the likely course of action policy makers will take on the basis of pressures from the most influential groups in society. In a primary exporting economy, for example, how the income of large landowners is affected may be more important than in an economy that exports manufactures, in which the impact on industrialists may well determine the political feasibility of a particular policy choice. The remainder of this paper is organized as follows. The next section considers recent external shocks which have had, and will continue to have, an important impact on the growth prospects of developing coun- tries. The need to adjust to these shocks and the difficulty of accomplishing the adjustment due to structural rigidities are also discussed. Section III describes the archetype economies chosen for the analysis, along with the selected set of experiments. The following two sections describe how alterna- tive adjustment mechanisms affect the distribution of income in each economy and contrasts the effects of each policy response across the three representa- tive economies. Section VI discusses how the political struggle among socio- economic groups is likely to influence policy selection. Conclusions follow in Section VII. -5- II. External Shocks and Macroeconomic Adjustment Given the macroeconomic imbalance caused by recent external shocks, what determines the extent of disequilibrium and what determines the ease with which adjustment takes place? As we shall see below in sections V and VI, how adjustment takes place depends, above all, on how the various socio- economic groups will react to their perceived change in relative and absolute income. In addition, the nature and extent of disequilibrium caused by these shocks depends very much on the structure of the economy. Finally, the extent to which the decline in absorption can be cushioned by timely external borrow- ing and by the use of short term stabilization policies will affect the ease with which the transition can be achieved. To explore the range of possible adjustments, consider the standard model with tradables and home goods. Prior to the shock, the economy satis- fies its budget constraint and is in a state of external and internal balance. Let the shock consist of a rise in the world price of imports, holding the price of home goods constant. The shock reduces income measured in terms of tradables, leading to a decline in income as a result of the decline in the terms of trade, the size of which depends on the openness of the economy. In the very short run, prior to the cut in expenditures and to the reallocation of resources, consumers would attempt to move to substitute home goods for tradables which would create excess demand for home goods. Even- tually, resources would shift towards tradables in response to the rise in import prices. However, unless by pure coincidence, the economy would not yet have achieved macroeconomic balance as there would still be an excess demand for home goods at the unchanged relative price of home goods for tradables. - 6 - The difficult task facing policy makers is to achieve a permanent change in the real exchange rate so as eventually to reach macroeconomic equilibrium. However, this is not an easy task and, if the price adjustment cannot be maintained, perhaps because of sticky real wages, the economy may eventually be forced to deflate. One could, for instance, imagine that wages rise with the price of imports which in turn causes the price of home goods to rise, and so on. 1/ The outcome then is permanent structural disequilibrium with successive rounds of devaluation accompanied by domestic inflation. Assuming that equilibrium is eventually reached, it is clear that both a reduction and a switch in expenditures must be achieved if the economy is to avoid unemploy- ment of resources. However difficult it may be to carry out expenditure- switching policies, the costs of failure can be large in macroeconomic terms. This is not to say that the distribution of income, broadly defined, is only affected by the structural changes described above. The impact of inflation on real wealth is also important and should be kept in mind in discussing an adjustment of the magnitude required by the recent events in the world economy. However, the recent external shocks suffered by developing countries are structural rather than monetary and, in spite of their monetary consequences, they call for structural adjustments in the economy. It is the distributional consequences of these structural adjustments that we focus on in this analysis. Consider the following alternative structural adjustments. If employed labor is sufficiently well organized, then real wages are sticky, thereby reducing the role of expenditure-switching policies so that adjustment 1/ Findlay (1973), Chapter 12 presents a similar model where home goods prices are determined on a mark-up basis and real wages are fixed. - 7 - takes place mostly by expenditure reduction. Alternatively, if it is the capitalists' interests that prevail, real wages will fall and the question then is which capitalist group has the most political power. In an open economy, devaluation would be favored by capitalists who would look forward to the prospects of a greater volume of exports and to the possibility of import substituting activities. In a closed economy where import substitution is already at an advanced stage, capitalists would attempt to protect their present position. A system of import licenses issued directly to producers in the case of intermediate goods and perhaps to wholesalers in the service sector for consumer goods would be a likely policy outcome. Of course, these are extreme alternatives and there are other impor- tant considerations that will have an impact on the distribution of income. Clearly the poor will suffer greatly from devaluation, and even more so from a system of import licenses, in an economy which is a food importer (for example, Sri Lanka, Egypt and Chile). For different reasons, devalua- tion will benefit landowners in a primary exporting economy. It is therefore important in an analysis of the impact of adjustment policies on the distribu- tion of income to capture differences in the structure of production and trade and differences in the volume of trade as well as differences in the composi- tion of consumption expenditures among socioeconomic groups. - 8- III. The Three Archetype Economies As discussed by Chenery (1979), the pattern of growth across countries varies greatly. An important source of the variation is the de- gree of openness of the economy which is itself influenced by country size, factor endowments and past government policies. In their work analyzing the pattern of development across a large number of developing countries, Chenery and Syrquin constructed a typology of developing countries based on trade orientation and related to policy choLces of a country as well as on its natural endowments. Much of our inspiration in the construction of the three archetype economies described here comes from their work and more recent work based on comparative input-output data for a number of countries. 1/ The three economies are all assumed to be "semi-industrial" countries with an income per capita around $500 (1970 US$). Such-countries are in a transitional phase with their economic structure closer to that of the indus- trialized countries than to the very poor low income countries in which the overwhelming bulk of economic activity is in the primary sector. Thus, although the different patterns of international trade in the three countries result in considerable variation in economic structure, they all generate well over 20 percent of GDP in the manufacturing sector. While they have quite different structures of trade, production and employment, we have also assumed them to be similar in a number of important respects. First, they are identical in size with the same total physical output and size of the labor force. This simplification is particularly 1/ See Chener- and Syrquin (1975), Chenery (1979), Chapters 1 to 3, and Kubo and Robinson (1979). -9- useful when comparing the number of people in poverty and various other macroeconomic changes. Second, they share virtually the same technology. 1/ This assumption is common in cross-country comparisons where universal access to a common technology is often assumed. The input-output coefficients differ very little across the three countries, the capital structure (B) matrices are identical and capital/output ratios are the same. We do assume different labor endowments by skill categories which implies that labor/output and hence capital/labor ratios udffer. The final area in which we have imposed uniformity across the archetypes relates to the choice of parameters describing consumer and producer response to relative price changes. Trade elasticities, elas- ticities of substitution in production, and income and price elasticities of demand for consumption by socioeconomic groups are the same. Each socio- economic group makes expenditure decisions according to the Linear Expenditure System (LES). The parameters of the LES vary across socioeconomic groups, but are the same for a given type of group across countries. 2/ To simplify further, production functions are assumed to be Cobb-Douglas in capital and labor, with fixed input-output coefficients for intermediate inputs. Finally, the logvariances describing the distribution of income within socioeconomic groups are the same for a given socioeconomic group in all three archetype economies. The model has eight sectors. The primary (1) sector includes agricultural and mining while the food (2) sector includes processing and manufacturing of food, beverages and tobacco. Consumer Goods (3) might 1/ For a discussion of how this was done, see Appendix A. 2/ See the Appendix A for a description of the expenditure system. - 10 - also be called "light" industry and does involve some vertical aggrega- tions; for example, textiles and clothing, wood and wood products. Interme- diates (4) and capital goods (5) are demanded only by producers while con- struction (6) is a pure investment good. Social overhead (7) includes utilities, transportation and housing. Services (8) includes retail and wholesale trade, government services, banking, insurance, etc. Furthermore, we assume that (6) - (8) are pure non-traded sectors and that all government non-investment expenditure is concentrated in services. It is a difficult matter requiring compromise to come up with a satisfactory definition of socioeconomic groups which is useful both for economic and political analysis and is also closely related to the variables generated endogenously by a CGE model. The socioeconomic classi- fication described below represents one such compromise. Table 1 summarizes the socioeconomic classification of households, the sectors from which their income is derived and whether that income is wage or profit income. Finally, it provides the assumed within-group logvariance of the distribution of income within each socioeconomic group. 1/ By "farmers" we mean agricultural workers and small landowners or "minifundistas" who own a plot of land that provides enough income to support a family. What distinguishes them from marginal labor, besides their slightly higher income, is that they are tied to the land and earn, in addition to their wage income, a share of agricultural profits. Because 1/ The distribution of income within each socioeconomic group is given by a lognormal distribution with an exogenously specified logvariance. The overall distribution is derived by numerically aggregating the witlhin-group distributions. The technique is described in Adelman and Robinson (1978). - ii - Table 1 Socioeconomic Classification of Individuals Type Sectors of Income Within-group Activity Source Logvariance Farmers 1 Wages + profits a/ 0.50 Marginal Laborers 1-8 Wages 0.20 Industrial Laborers 2-7 Wages 0. 40 Service Sector Laborers 8 Wages 0.25 Agricultural Capitalists 1 Profits a/ 0.60 Industrial Capitalists 2-7 Profits 0.60 Service Sector 8 Profits 0.50 Capitalists a/ 25 percent of agricultural profits go to Farmers, the rest to Agricultural Capitalists. there is considerable variation in the size and quality of land owned by smaller farmers, we have assumed this group has a fairly high within-group logvariance. By contrast, marginal labor--the only group which is mobile across all sectors--is fairly homogeneous. With an income a shade above -half of the economy-wide mean income, they are the largest and poorest group in society and their income is fairly equally distributed. Industrial laborers, or organized labor, are employed in the manufacturing sector. The smallest wage-earning group in society, they are fairly disparate in composition and include white collar workers, clerical workers, technicians and engineers. Their income is about 1.5 times that of marginal labor. Service sector laborers include government workers and, finally, we have the three capitalist groups who get their income from profits. Their number is fixed exogenously with the income of the industrial capitalists about ten - 12 - times the economy-wide mean income and about 30 percent above that of agri- cultural capitalists. Service sector capitalists include a large group of self employed (such as retail traders, etc.) and have an income about five times the mean income. 1/ These groups whose incomes are determined by the model represent the minimum necessary to capture the diversity of socioeconomic groups found in semi-industrial countries. Clearly there is considerable range of varia- tion in mean inLome and, at this level of aggregation, a considerable degree of heterogeneity within socioeconomic groups. The countries are characterized by a fairly high degree of inequality, with Gini coefficients of approximately 0.50 (or overall logvariance of 0.60). 2/ About 45 percent of the overall logvariance is due to between-group variation and corresponds to that part of total inequality which is endogenous to the model. We now come to the distinguishing characteristics built into the production side of the three archetype economies. The distinctions introduced into the otherwise similar structures fall under three categories. First, the economies are distinguished by their volume of trade. Two of the economies are open in the sense that the volume of exports as a proportion of GDP is large (15 percent). Second, they are distinguished by their structures of production and trade. Of the two open economies, one is a primary exporter and the other is a manufacturing exporter. Accordingly, the structure of production is biased toward the primary sector in the first and towards manufacturing in 1/ Group and economy-wide mean incomes in the base run are given in Table 10 below0 2/ Gini coefficients of this magnitude are higher than those reported for Asian countries where the distribution of land is fairly equal but lower than those of most Latin American countries. See Jain (1975). - 13 - the second. The third distinguishing characteristic, which provides the basis for the structural differences described above, are variations in factor endowments, tariffs and subsidies that are assumed to account for the dif- ferences in production and trade structures between the three economies. There are thus three archetype economies: primary exporter (PE), manufacturing exporter (ME) and closed economy (CL). 1/ They differ in: sectoral structure of production, employment and capital; volume of trade; sectoral composition of eAports and imports; skill composition of the labor force; and past trade policies reflected in different sets of tariffs and subsidies. These differences are summarized in Tables 2 to 4. Table 2 compares the structure of production and trade in the Base Run (prior to any shock) across the three archetypes. All of the variation in structure comes from the relative sizes of the industrial and primary sectors, and from variations within the manufacturing sector itself. The PE economy, with a larger primary sector (and hence a smaller manufacturing sector), has a distinctively different gross output structure. The differences between the CL and ME structures are less pronounced, although the ME economy has a much larger consumer goods sector which provides its main source of foreign exchange while the CL economy has larger primary and food sectors to assure self sufficiency. The overall export and import ratios (see also Table 3) indicate the sharp difference in volume of trade between the closed and open economies. 1/ Note that the structure of imports describing the closed economy in Table 2 corresponds more closely to that found in countries like Mexico and Turkey than to that prevailing in larger countries like Brazil or India. Table 2 Tlree Archetype Economies: Structure of Production and Trade in the Base Rtm (percent) Gross Production _ Imnortœ Exports Sectors CL PE ME CL PE ME i CL PE tE Primary (1) -20.7 27.2 19.6 6.6 8.7 17.3 i 48.2 80.9 9.9 Food (2) 13.4 11.5 12.0 4.0 13.4 14.2 12.8 4.5 6.2 Consumer (3) 11.7 9.2 14.4 5.0 12.9 7.4 22.5 9.3 52.9 Intermediate (4) 11.6 9.1 11.1 51.2 34.4 36.8 13.9 4.5 23.2 Capital Goods (5) 5.4 3.9 4.9 33.2 30.6 24.3 2.6 0.8 7.8 Construction (6) 5.8 7.0 6.9 _ _ _ _ _ Social Overhead (7) 8.9 10.2 9.7 _ _ _ _ Services (8) 22.5 21.9 21.4 _ _ _ _ _ _ Primary (1) 20.7 27.2 19.6 6.6 8.7 17.3 48.2 80.9 9.9 Industry (2-7) 56.8 50.9 59.0 93.4 91.3 82.7 51.8 19.1 90.1 Services (8) 22.5 21.9 21.4 - - - - - - Tctal 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Ratio to GDP 1- 175.3 150.3 159.5 11.2 18.7 19.0 7.4 15.2 15.S Rai oG?.-.-..-. 7.41.2 15.5 f - 15 - Note that with world prices exogenous and equal to unity, the foreign capital inflow is the same for all three economies and is maintained at its Base Run level throughout the experiments. The variation in exports and imports is most obvious at the 3-sector level of aggregation. The PE and ME economies have opposite structures and the CL economy resembles the PE economy on the importing side, while on the export side it has a balanced structure with half of its exports originating from the primary sector. Within the manufacturing sector, note that the CL economy is almost as dependent on imports of interme- diates and capital goods as the other economies. 1/ TabLe 3 provides information on the sectoral trade orientation of the three archetype economies. It is crucial for understanding how the common external shock gets transmitted across sectors and across archetypes. The most tradable sectors in the sense that the domestic price system is closely tied to the exogenous world prices are the sectors which have high ratios of imports to domestic supply and/or high ratios of exports to domestic production. Note also that although there are three pure non-traded sectors. in the economy, these are among the most import-dependent sectors due to the high import content of total intermediate inputs. 1/ Accompanying, and responsible for, this foreign trade structure are the following trade incentives: tariffs of 25 percent, 35 percent, 20 percent for consumer goods, intermediates and capital goods for ME and PE; 15 percent across the board export subsidies for PE and 75 percent export subsidies for manufacturing in ME; no export sub- sidies for CL and the following tariffs: consumer goods (75 percent), intermediates (75 percent) and capital goods (20 percent). All three economies have the following trade substitution elasticities: Primary (3.0), Food (1.5), Consumer Goods (1.25), Intermediates (0.75), Capital Goods (0.25). Tab le 3 Three Archetype Economies: Sectoral Trade Ratios in the Base Run (Percent) Imported Imports/ a/ Inputs/Total LuortsI Domestic Suppl Intermediate Inputs Dommatli Produetls Sector CL PE ME CL PE ME CL P M Primary 2.3 5.7 11.0 9.7 13.7 14.7 9.6 30.3 4.9 Food 2.0 15.1 14.8 6.1 9.9 12.2 4.0 4.0 5.0 Consumer 3.0 19.2 9.5 12.0 19.2 15.7 8.1 10.2 35.7 Intermediates 29.8 49.3 49.2 17.1 25.1 23.5 5.0 5.1 20.2 Capital Goods 39.8 98.8 69.2 28.0 38.5 32.2 2.0- 2.0 15.4 Construction _ _ _ 24.8 30.9 27.3 _ _ _ Soc ' Overhead _ _ _ 25.1 31.6 27.4 _ _ Services _ _ _ 17.0 20.5 17.8 _ _ _ Average 6.7 13.8 13.2 15.8 20.7 19.5 4.2 10.2 9.7 a/ Domestic Supply - domestic production - exports. - 17 - The final distinction between the three archetypes relates to factor e-xdowments which are given in Table 4. All three economies have a total labor force of 10 million. As one would expect on the basis of the factor endowment theory of international trade, the pattern of trade is closely related to each archetype's factor endowments. Thus the PE economy has the largest supply of farmers while the ME economy has the greatest supply of marginal labor and especially of organized labor which are intensively used in the manufacturing sector. As is the case with most other indicators, the CL economy, which has the most balanced trade and production structure, has a pattern of factor endowments in between those of the two open economies. Table 4 Three Archetype Economies: Composition of the Labor Force (percent) CL PE HE Farmers 30 40 25 Marginal Laborers 30 27 33 Organized Labor 10 9 15 Service Sector Labor 30 24 27 Total 100 100 -100 - 18 - IV. The Macroeconomic Impact of an External Shock The external shock we impos'e on each of the archetype economies has two components. On the import side, the exogenous world prices of imports are raised by 25 percent across the board. On the export side, physical exports are lowered by 25 percent in all sectors and are essentially fixed to their new values. 1/ World prices of exports, however, remain fixed. This experiment is intended to capture the essence of the trends evident in the second half of the seventies; first, the increase in the purchase price of developing countries imports spurred by the rise in energy prices and compounded by the inflationary situation that developed in the industrialized countries; and second, the world recession that led to restrictionist policies in industrialized countries and the relative decline in world trade. We have deliberately chosen to ignore any additional terms-of-trade losses that occur because developing countries face downward sloping demand curves for their exports. In selecting the external shock, we have tried to facilitate the comparison across archetypes by the following simplifications. First, combining equal across-the-board rises in import prices with equal across- the-board reductions in export ratios yields an undifferentiated external shock for all three economies so that the impact effect depends only on the total volume of trade in the economy. Second, by fixing the world 1/ In the model, it is convenient to cut exports by lowering all the sectoral export supply ratios (see Table 3). This, however, does permit a slight export supply response through changes in the sectoral com- position and level of output. In the experiments, this effect is very small. - 19 - price of exports, the decline in the foreign terms of trade is the same for all three economies so that the resulting income loss depends on the initial volume of trade and not on its composition. To study the interactions between adjustment policies and the distribution of income, we compare the following three adjustment policies: devaluation, premium rationing and premium rationing with a fixed real wage for marginal labor. There are Important political as well as economic dif- ferences between the three adjustment mechanisms. A policy of devaluation, because of its fairly even spread across sectors, could be viewed as a poli- tical compromise reflecting some minimal degree of consensus among socio- economic groups. A policy of granting Import licenses giving rise to premia reflects a situation where political power is concentrated in the hands of the capitalists and would accentuate income concentration. Finally, a policy of premia on imports coupled with fixed real wages for marginal labor reflects a situation where no socioeconomic group is firmly in control of policymaking. Businessmen manage to appropriate the premia associated with a system of import control by licensing but, on the other hand, workers are sufficiently well organized to prevent a fall in real wages. What then is the macroeconomic impact of alternative adjustments to the external shock? Table 5 contrasts the magnitudes of changes in the major components of GDP at constant prices across the three countries for each of the alternative adjustment mechanisms. Consider first the case of devalua- tion. At constant prices, the fall in real GDP ranges between 3.7 percent for CL to 2.3 percent for PE. At first sight, this result seems counterintuitive since the economy which trades the least is the most severely affected. The result is due to the different structures of production and trade in the Table 5 The Macroeconomic Impact of the External Shock Experiment: Premium Rationing Devaluation Premium Rationing With Fixed Real Wage CL PHE C PE HE CL PE HE Change from Base Run (2) a/ GDP - 3.7 - 2.3 - 3.2 - 3.3 - 2.0 - 3.2 - 4.9 - 2.9 - 6.5 Private consumption - 7.7 - 4.2 - 7.5 - 4.4 - 4.0 - 6.6 - 5.7 - 5.0 -10.4 Government consumption 11.6 9.3 8.8 5.3 10.7 7.4 5.3 10.7 7.4 Investment -12.8 -16.7 - 7.1 -18.3 -16.7 - 9.5 -19.3 -17.4 -12.7 Exports -23.8 -22.8 -27.8 -23.8 -23.8 -28.9 -26.2 -23.8 -33.0 Imports -32.9 -35.5 -37.8 -32.8 -35.5 -39.5 -34.4 -36.3 -41.2 Exchange rate 62.0 37.4 33.9 0.0 0.0 0.0 0.0 0.0 0.0 Selected ratio. (2) Unemployment rate - - - - - - 3.6 1.6 7.1 Premium rate - - - 86.8 51.0 42.7 84.1 50.4 41.6 Total 1emia/GDP - - _ 9.2 8.3 6.6 9.0 8.2 6.5 a/ A1I variables are real, except the exchange rate. - 21 - three economies. The CL economy has the highest ratio of intermediate input requirements to total production and, moreover, its imports are concentrated in the intermediate sector with a low trade substitution elasticity. After the shock, the ratio of the value of intermediate inputs (both domestic and imported) to total absorption rises the most in the CL economy (up to 4.9 percentage points compared to 4.2 and 2.9 for the PE and ME economies). The result is that the adjustment is much more painful for CL than for the other two economies. The required devaluation in the open economies is around 35 percent while in the closed economy it is over 60 percent. With a real GDP lss of 3.2 percent, the ME economy is harder hit than the PE economy. The reason is that PE has a production structure biased toward the primary sector which, next to food, is the least import dependent sector in the economy. Although the ratio of imported to total intermediate inputs is about the same in the PE economy, the ratio of total intermediate inputs to total production is lower. The PE economy thus depends less on imported goods to maintain production levels. It is also interesting to compare how the major final demand aggre- gates are affected by the shock. Private consumption falls the least in the PE economy which is what one would expect since agricultural and food exports are now supplied to the domestic market, dampening the cut in aggregate consumption expenditure. With a 7.5 percent decline, the fall in private consumption expenditures is highest in the ME economy which has to bear the brunt of higher prices for food imports. Government consumption rises by approximately 10 percent across all three economies, largely because gov- erinent domestic currency revenues from tariffs increase both because they are levied on higher foreign prices and because the currency has devalued. - 22 - While, under devaluation, the differences in adjustments between countries are not great for exports and imports, they are quite significant for investment. In the model, fixed savings rates are assumed for the gov- ernment and for each socioeconomic group which implies that the model is "ssavings driven." Foreign capital inflows are assumed to be part of savings. To understand why real investment falls by more than 16 percent in PE but only by 7 percent in ME we need more information about the composition of investment and the sources of savings. First, note that one expects to find variations in relative prices and hence differences in the prices of capital goods. Even the same nominal investment would, after the shock, yield dif- ferent real investment. As can be seen from Table 6, the investment goods price deflator varies widely across the three economies, changing least for the ME economy (because of the increased availability of capital goods on the domestic market). There is also a second effect which can be seen by examining the composition of savings (in current prices) provided in Table 6. The trade deficit, measured in world prices, is identical and remains fixed across the three economies and also is assumed to be part of savings. 1/ But this saving takes place in domestic prices and hence varies with devaluation. The foreign component of domestic savings--22 percent in the base run--rises dramatically across all three economies with devaluation. And of course it rises most in CL where devaluation is greatest. Thus real investment falls less in CL than in PE, even though both economies experience essentially the same rise in capital goods prices. 2/ 1/ For further details see Appendix A. 2/ This effect, often referred to as the Hirschman (1948) effect, has been noted and investigated in the context of developing countries by, among others, Cooper (1971) and Krugman and Taylor (1978). Table 6 Volume and Composition of Investment Base Run Values i/ Ratios to Base Run (percent): Premium Rationing B " e Ri. Values a _Devaluation Premium Rationing With Fixed Real Wage CL PE ME CL PE ME CL PE ME CL PE ME Foreign Savings 22.0 22.0 22.0 159 141 136 100 100 100 100 100 100 Labor Savings 20.0 23.0 21.0 80 87 90 80 83 86 80 83 81 Capitalist Savings 38.0 49.0 30.0 82 88 87 103 102 97 100 102 93 Government 28.0 37.0 53.0 100 100 91 96 108 94 96 108 92 Total 108.0 132.0 126.0 103 99 98 96 100 94 95 99 92 Investment Goods Price Deflator 100.0 100.0 100.0 118 119 106 118 120 104 118 120 105 a/ Nominal savings. - 24 - Whether and to what extent foreign capital inflows are channelled into investment is an issue which has often been debated in the literature. After having been much discussed in the context of two-gap models, the issue has resurfaced in the debate on the role of alternative closure rules on the distribution of income. 1/ As we have shown elsewhere, how investment is determined may have a significant impact on the distribution of income. 2/ For example, had real investment been fixed, with increases in the prices of capital goods from 6 to 19 percent, there would have been significant forced savings which might have worsened the distribution of income, depending on the nature of the forced-saving mechanism assumed. Since exports are essentially fixed, the main difference between adjustment by devaluation and adjustment by premium rationing is the income transfer to capitalists. Also, devaluation has no inherent efficiency advan- tage over a uniform import premium. In other experiments not reported here, we allowed an export response to the policy shock by specifying logistic export supply functions (see de Melo and Robinson (1980)). When compared with adjustment by premium rationing, devaluation is more attractive in the case where some export response is allowed than in the version adopted here. Many of the differences in the pattern of adjustment between the three economies remains when adjustment takes place by premium rationing. It is interesting, however, to note that the GDP loss is slightly less than in the case of devaluation for two economies, CL and PE. This effect is largely 1/ See Papanek (1972) and Taylor (1980). 2/ See Dervis, de Melo and Robinson (1980), chapter 12. - 25 - due to the distortion-removing effect of adding premia to tariffs. Thus CL and PE, which have the most uneven tariff structures, end up with tariff structures that are much more uniform as a result of the imposition of pre- mia. 1/ Also note that the pattern of changes in real investment is now almost entirely accounted for by the differential impact of the shock on capital goods prices. Premium payments as a percent of GDP range from 6.6 percent for ME to 9.2 percent for CL. The potential rents accruing to the private sector from a system of licenses can therefore be enormous and it is no surprise that it is important what method of adjustment is used. Although more sensitivity analysis would be needed before one could establish a correspondence between devaluation and associated premia, these results are consistent with the often-made observation that huge rents are commonplace in economies which choose to maintain external balance with a system of import controls. 2/ Finally, we come to adjustment by premia combined with fixed real wages. The fall in GDP due to the contractionary impact of the external shock is,exacerbated by induced unemployment ranging from 1.6 percent for PE to 7.1 percent for ME. As a result, GDP falls substantially more, except in the primary exporter. The PE economy is least affected for two reasons. First, the cost of living rises less in the economies which export rather than import 1/ See Johnson (1960) for an analysis of the welfare effects of tariffs which takes into account the variance of tariffs across sectors. 2/ Note that the premium rate is higher than the devaluation rate even though the composition and volume of imports and exports are similar. This is entirely because of the normalization rule which maintains a weighted sum of composite prices constant. Thus the domestic cur- rency price of exports does not rise in the case of premia rationing whereas it does in the case of devaluation. It follows that to maintain the same price level, the domestic prices of imports do not rise as much. - 26 - food. This effect is important since the weight of food in the consumer basket of marginal labor is about 60 percent. Second, the PE economy has the smallest supply of marginal labor (see Table 4) so that, other things equal, aggregate employment will not fall as much. It is important to note that if we had instead made the short run assumption that nominal rather than real wages remain fixed, the impact of adjustment on the distribution of income would have been quite different. Indeed, with real wages falling, the demand for labor would have increased. Such an assumption is not justified in a medium-term model and, in any case, would require an analysis of inflationary mechanisms which is beyond the scope of the present analysis. - 27 - V. Policy Choice and the Distribution of Income The impact of the external shock results in a decline in real income and the extent of that decline depends on the choice of adjustment policy. In addition, the composition of imports determines to a large extent whether the burden of adjustment falls on consumption or on investment. Thus for the ME economy which is a food importer, the consumption loss tends to be high. For the CL and PE economies, which import capital goods, the decline in GDP falls most heavily on investment. The rate of real investment which is around 20 percent of GDP in the base run for the three economies falls to 17 percent in both the CL and PE economies. In addition to variations in real investment, the different policy adjustments have important variations in incidence among socioeconomic groups. Which groups bear the brunt of adjustment across the three economies and is it the same groups that lose the most under both devaluation and premium rationing? Can any group benefit in absolute terms and experience a rise in real income? We explore these questions in this section. Before examining the magnitudes of distributional shifts, it is useful to review the various mechanisms whereby the external shock affects the distribution of earned income. First, on the supply side the change in relative net prices affects the distribution of value added between sectors and, within sectors, the distribution of value added between wages and rents. These shifts determine, after taxes, the distribution of earned nominal income among the socioeconomic groups. Second, the distribution of income is affected by differential changes in the cost of living across groups. The relevant - 28 - prices are those of composite goods (which include imports). Third, the distribution of income is affected by transfers which here arise from the distribution of import licenses and hence premia. Consider first the distributional shifts associated with changes in net prices. As a first approximation, divide the economy into two sectors, agriculture and manufacturing, and assume that labor is perfectly mobile between them while capital is immobile. Furthermore, assume a fixed wage differential, CD, between agriculture (A) and manufacturing (M). 1/ In Figure 1, OOM represents the total labor force in the economy, the vertical axes measure the wage of labor in the two sectors, and the two curves indicate the marginal revenue product of labor. Prior to the shock, the equilibrium distribution of the labor force is given by L0 with labor income given by the two rectangles under the wage lines W and W0 , and capitalists- income given A M adcptlssicm ie by the triangular region between the wage lines and the respective marginal revenue product (MRP) curves. Now suppose that the external shock raises the net price agricul- tural terms of trade, thereby shifting the MRP curve in agriculture. Clearly capitalists in the manufacturing sector lose. On the other hand, the per unit rents of capitalists in agriculture rise by more than the percentage increase *in the agricultural net price. 2/ Finally,. the increase in the wage rate 0 1 (WA to WA) is less than the increase in the agricultural net price terms of *1/ 'This differential is-structural in that it is--assumed to be part of the.initial equilibrium.. 2/ Because-the'percent increase-in net price is the weighted sum of the- percent. increase in 'the:wage.rate (which is less than the increase -in 'the'price)'. and rate. of return to capital.- This effect is known as the magnification effect., - 29 - Figure 1 Tactor Mobility and the Distribution of Income Marginal Revenue Product Marginal Revenue Product in Agriculture in Manufacturing MRPA~~~~~~~~~~ W A _A C A AD 0~~~~~~~~ - 30 - trade, so that labor gains in terms of manufacturing but loses in terms of agriculture. Hence, to determine what happens to labor's real income we must consider its cost of living. As can be seen from Figure 1, the magnitude of income changes depends on the elasticity of substitution between labor and capital (a) in the two sectors and on the intensity with which the two factors are used in both sectors. A large value for aA combined with a small value forcaM and/or a high labor share in agriculture and a low labor share in manufacturing would lead to a flat MRPA and to a steep MRPM curve which in turn would imply that labor's income in terms of agriculture would rise. Since we assume the same (unitary) elasticities of substitution between capital and labor across all sectors, it is only differences in labor intensity between sectors which influence the distributional outcome. Introducing organized labor which is mobile within manufacturing does not alter the above mechanisms except that they share the manufacturing capitalists' losses, the extent of sharing being inversely related to the elasticity of substitution between the two labor catetgories (again assumed for simplicity to be everywhere unity). Table 7 compares the impact of the external shock on selected relative price indices for the three economies. Regardless of the selected adjustment mechanism, the foreign terms of trade rises by 25 percent. The impact on the net price agricultural terms of trade shows considerable variation across economies and adjustment mechanisms. Not surprisingly, it rises substantially in the ME economy (41 percent) which must replace agricultural products which were previously imported. There is no change for the PE economy while there is a 5 percent fall in the CL economy caused by the increased intermediate input costs in agriculture which have a high Table 7 The Impact of the External Shock on Selected Relative Price Indices - (Ratios to Base Run, percent) Experiment: Premium Rationing Devaluation Premium Rationing With Fixed Real Wages CL PE ME CL PE ME CL PE ME Foreign TOT 125 125 125 125 125 125 125 125 125 Net Price Agricultural TOT 95 100 141 87 78 137 83 76, 127 Composite Price Agricultural TOT 74 71 117 74 64 111 72 63 106 Note: TOT - Terms of Trade - 32 - import cost component. The composite price agricultural terms of trade, which is important in determining real income, follows the opposite course of the capital goods price deflator. This movement explains why real con- sumption falls so much in the ME economy. While there is much variation in shifts in price indices which contribute to shifts in the functional distribution of earned income, relative shifts in cost of living indices between socioeconomic groups are equally important in determining changes in real incomes across socioeconomic groups. Table 8 provides ratios of cost of living indices to their values in the Base Run for each socioeconomic group. The spread across groups is quite wide. Thus, for instance, the cost of living of farmers in the ME economy rises by 3 percentage points with devaluation while that of capitalists falls by 9 percentage points. Table 9 provides the group mean real incomes in the base run along with the percent changes from the values in the base run. It thus indicates which groups gain or lose in absolute terms. In turn, Table 10 provides information on the relative distribution of incomes by giving changes in group shares in total income. It is noteworthy that some groups gain in absolute terms after the external shock in spite of the sizable fall in the economy- wide mean income (see last row). Some striking patterns emerge: for the manufacturing exporter, regardless of the adjustment mechanism, farmers and agricultural capitalists always gain in absolute terms relative to the Base Run. In the other two economies, no group gains significantly when adjustment *is by devaluation. 1/ However, when there is adjustment by premium rationing, industrial capitalists gain absolutely along with service sector capitalists and, to a lesser extent, service sector labor. 1/ The only exception is the 0.7 percent increase for farmers in the PE economy. - 33 - Table 8 Cost of Living Indices (Ratios to the Base Run, percent) Experiment: Premium Premium Rationing, Devaluation Rationing Fixed Real Wages CL PE ME CL PE ME CL PE ME Farmers 84 86 103 85 83 102 85 83 100 Marg. Labor 86 89 100 87 87 99 87 87 99 Org. Labor 88 93 96 90 92 96 90 92 97 Serv. Labor 87 92 99 88 90 98 88 90 99 AG Capital 90 96 92 92 95 93 92 96 94 Ind. Capital 90 95 91 92 96 93 92 96 94 Serv. Capital 90 95 91 92 96 93 92 96 94 Total 88 91 97 89 90 97 89 90 97 Table 9 Group Mean Real Incomes Values: Change from Base Rum (percent) Premium Rationing Base Run Devaluation Premium Rationing With Fixed Wagest/ CL PE ME CL PE ME CL PE HE CL PE ME Farmers 2.5 2.9 2.9 -6.9 0.7 12.3 -13.5 -16.0 8.1 -16.7 -17.7 1.8 Marginal Labor 2.3 3.2 2.6 -8.2 -1.9 -16.4 - 9.9 - 4.4 -20.6 - 4.3 - 1.0 -13.0 Org. Labor 4.1 5.3 4.4 -7.2 -2.1 -17.0 -11.6 - 3.6 -26.2 -10.4 -3.2 -26.9 Serv. Labor 3.2 3.7 3.2 -2.2 -7.0 -21.6 0.6 4.1 -15.9 - 1.6 2.2 -20.0 Ag. Capital 31.7 40.5 37.1 -13.3 -7.9 26.8 -15.6 -21.2 22.8 -19.f -23.1 13.9 Ind. Capital 46.1 55.0 50.1 -11.8 -6.7 -13.3 23.7 24.4 - 2.1 23.3 24.3 - 3.8 Serv. Capital 23.1 28.3 25.9 - 5.8 -13.0 -14.7 7.1 9.8 -1.0 3.9 7.0 - 6.0 Mean Income 4.1 5.1 4.7 - 7.5 - 4.7 - 7.4 - 3.4 - 4.1 - 6.4 5.3 - 5.3 -10.0 a/ Unemployed marginal labor has mean incomes of 1.2, 1.6 and 1.3 for CL, FE and ME economies. Their mean income is assumed to be 0.5 that of employed marginal labor. - 35 - What emerges from these figures is that starting from a reasonable spread in mean incomes among groups in the base run (see Table 9, columns 1-3), the distribution of premia, amounting to close to 10 percent of GDP, raises the mean income of some recipient groups by up to 25 percent. No wonder that the choice of policy instruments is such a politically sensitive topic. Consider, for example, service sector capitalists. They are the largest and least well-off of the capitalist groups. With adjustment by devaluation, even their relative position in society is likely to deteriorate as a result of the increase in the relative price of traded goods. However, when they receive the premia associated with the licenses they are granted for importing consumer goods, their mean income rises by almost 10 percent in the PE economy and by 7 percent in the CL economy. But in the ME economy, where import substitution in the primary sector is assumed achievable in the medium run, the farmers and agricultural capitalists appropriate these rents through higher domestic agricultural prices. In that economy, service sector capital- ists gain little from a system of import licenses since imports of consumer goods become effectively replaced by domestic substitutes. Finally, one can take a more aggregate view of the distribution of income along broad functional lines and lump all capitalists together in one group. Table 10 indicates that, taken as a group, capitalists always improve their relative position when they get premia, but in the case of devaluation they only improve their position in the SM economy. It would thus appear that devaluation is not a pol.tically easy adjustment policy to implement. The relative position of marginal labor, the largest labor group in each of the economies, declines in virtually all instances. Again, the swing Table 10 Group Shares in Total Income Premium Rationing Shares: Change from Base Runa-/ With Base Run Devaluation Premium Rationing Fixed Real Wage CL PE ME CL PE ME CL PE ME CL PE ME Farmers 17.0 21.8 14.4 0.1 1.3 3.1 -1.8 -2.6 2.2 -2.1 -2.8 1.9 Marginal Labor 16.1 15.9 17.5 -0.2 0.4 -1.7 -1.1 -0.1 -2.6 -1.9 -0.5 -4.4 Unemployed -- -- -- -- -- -- -- -- 1.1 0.6 2.2 Org. Labor 9.4 8.9 13.2 2 0.0 0.2 -1.3 -0.8 0.0 -2.8 -0.5 0.2 -2.5 Serv. Labor 21.8 16.5 17.5 1.3 -0.4 -2.7 0.9 1.4 -1.8 0.8 1.3 -1.9 Ag.Capital 13.7 17.8 11.7 -0.9 -0.6 4.3 -1.7 -3.2 3.7 -2.1 -3.4 3.1 Ind. Cap. 12.1 11.5 17.7 -0.5 -0.2 -1.1 i 3.4 3.4 0.8 3.7 3.6 1.2 Serv. Cap. I 9.9 7.6 8.0 0.2 -0.7 -0.6 1.1 1.1 0.5 1.0 1.0 0.4 All Capitalists 35.7 36.9 37.4 -1.2 -1.5 2.6 2.8 1.3 5.0 2.6 1.2 4.7 Total 100.0 100.0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Gini Coeff. 0.51 0.52 0.52 -0.01 -0.01 0.02 0.02 0.01 0.03 0.02 0.01 0.03 Theil Index 0.59 0.62 0.63 -0.02 -0.03 0.07. 0.07 0.04 0.12 0.07 0.04 0.11 a/ In percentage points. - 37 - in its relative position is most pronounced in the ME economy. As is typical in developing countries, we assume that the employed support the unemployed by direct income transfers. 1/ Struggling to maintain real wages leads the marginal labor group as a whole (including the unemployed) to mitigate their relative income loss (4.4-2.2=2.2% loss) compared to premium rationing (2.6% loss). With or without fixed real wages, the alternatives facing the poorest are gloomy indeed. In these experiments, aggregate measures of income inequality such as the Gini coefficient hide more than they reveal. Even when the share of marginal labor falls over 4 percentage points relative to capita- lists, the Gini coefficient never changes by more than 3 percentage points. More revealing from a distributional standpoint is an analysis of the composition of poverty. The results are given in Table 11 which shows the proportion of total population in poverty, as well as the group composition of people in poverty. The cut-off point taken here is a mean income of 1.5 which, in the Base Run, implies that 14 to 21 percent of the population is in poverty. 2/ Consider the ME economy: it starts out with 15.6 percent of its population in poverty; after the adjustment, between 21.7 percent and 25.5 percent of its population is in poverty. Given a -population of 10 million in each one of the archetypes, it is easy to see that a large fraction of society can be adversely affected, even when aggregate measures of the relative distribution show little variation. An examination of 1/ In this model, this is done through transfers so that the unemployed receive half the mean income of the employed. 2/ This corresponds approximately to $150 per capita in 1970. Table 11 Analysis of Poverty (percentage shares) Premium Rationing Base Run Devaluation Premium Rationing With Fixed Real Wages CL PE ME CL PE HE CL PE ME CL PE HE Group Shares in Poverty: Farmers 36.6 27.6 29.0 40.6 27.1 23.6 44.7 36.3 25.4 46.9 37.3 28.2 Marginal Labor 22.5 7.5 15.3 28.9 8.1 26.6 30.3 9.0 30.6 25.7 7.7 23.7 Unemployed -- -- -- -- -- -- -- -- -- 80.9 45.9 70.1 Organized Labor 10.5 4.6 8.6 12.8 5.0 14.2 14.5 5.3 18.8 14.0 5.2 19.2 Service Labor 10.9 6.0 15.3 11.8 7.9 21.8 10.6 5.1 18.0 11.5 5.5 20.6 Overall 21.0 14.2 15.6 24.5 14.7 21.7 25.9 17.8 23.0 27.4 18.6 25.5 Group Composition of Poverty Population: Farmers 49.8 74.1 44.3 47.4 70.5 26.0 49.4 77.8 26.3 49.0 76.4 26.3 Marginal Labor 30.6 13.5 30.8 33.8 14.2 38.7 33.6 13.1 41.9 23.5 10.0 22.7 Unemployed -- -- -- -- -- -- -- -- -- 10.6 4.5 19.5 Organized Labor 4.8 2.8 7.9 5.0 2.9 9.4 5.3 2.5 11.7 4.8 2.4 10.7 Service Labor 14.8 9.6 17.0 13.8 12.4 25.9 11.7 6.6 20.1 12.1 6.7 20.8 Total | 100.0 100.0 100.0 j_100.0 100.0 100.0 100.0 100.0 100.0 1100.0 100.0 100.0 Note: Poverty is defined as an income of less than 1.5 in the base year. - 39 - the bottom half of that table indicates that there is a considerable change in the group composition of the poverty population as a result of the external shock. In the ME economy, in addition to a greater number of people in poverty there is a large shift in the composition of poverty away from the primary sector towards the urban sector. Although for all three economies there is an increase in the number of people in poverty as a result of the deterioration in the foreign terms-of-trade, the composition of poverty varies depending on the selection of adjustment mechanism. Thus, under devaluation, for both the CL and PE economies, the proportion of farmers among those that are in poverty declines, while the opposite occurs under premium rationing. - 40 - VI. Class Conflict and Policy Choice The picture that emerges from comparing the impact of alterna- tive adjustment mechanisms on the distribution of income in different economic environments is complex. There are certainly significant changes in the relative distribution among socioeconomic groups, with some groups gaining in absolute terms in spite of an overall decline in real income due to the external shock. Differences in economic environment alone suffice to make different groups gain or lose, even with the same adjustment policy. Different adjustment policies also have quite different effects on the distribution. The fact that the choice of policies has a significant impact on the distribution of income leads to a number of political questions that go beyond the usual economic analysis of policy choice. Given the changes in the distribution of income among socioeconomic groups induced both by the external shock and the policy response, how is the distribution of political power affected? How will the political struggle between the gainers and losers decide the final choice of policy? Finally, assuming that one can analyze how policies are selected when there are conflicting group interests, is it likely that the same policy will be selected in different economic environments? To explore these questions, we start from a number of simple assump- tions about the way that the social and political systems are structured. First, assume that each group realizes how its income would be affected under each policy regime. Second, that socioeconomic groups are distinct and care only about their own interests. We thus assume that we have partitioned the society into the important or politically relevant power groups, each pursuing - 41 - its own interests with perfect knowledge. Third, assume that the intensity with which any group cares about a given policy change is measured by the relative difference in its income between the two situations. Define Y (P) as the mean income of group i under policy regime P. In our experi- ments, P can be the following: B (Base Run), D (devaluation), P (premium rationing), and P,W (premium rationing with fixed real wage). For example, the intensity with which group i is for (+) or against (-) devaluation compared to premium rationing as a function of its relative change in income is defined by: Ri(D;P) = (Yi(D) - Y (P))/Y (B) (1) i i ~~~i i In order to evaluate the political feasibility of any policy choice, we must be able to compare the gains and losses between groups according to some measure of their relative political power. Thus, while a group may have a large gain-or loss, and so care a lot about a particular policy choice, it may have little political power and so be unable to affect the.final choice. We need.some weighting scheme reflecting relative political power to aggregate the gains and losses across different socioeconomic groups. There is certainly little agreement about how to derive such weights, so we have specified three different schemes which are consistent with different theories of the sources of political power and which should provide a reasonable range of values. .The analysis.is not concerned with the more fundamental and difficult issues relating to the theory of how the political process works. Rather, given a political system, we are trying to capture in a stylized.manner some of - 42 - the elements which are likely to determine the course of policy selection by those who are already in power. The mechanisms by which political power is exercised are not explicity considered. The first scheme assumes that relative political power is given by shares in total population - "one person, one vote." The second scheme assumes that relative political power is measured by relative shares in total income - "one dollar, one vote." The third scheme assumes that the political system is dominated by the economic elite and hence that the relative political power of different groups is measured by their share in the elite. In this third case, we assume that the elite is defined as the top 5 percent of the overall personal income distribution and that relative political power is measured by the socioeconomic composition of this group. The first scheme is rather naive but appealing to utilitarian welfare economists, if not to cynical political scientists. The second scheme is a kind of "effective demand" theory of the distribution of poli- tical power with groups able to buy influence in proportion to their shares in total income. Large groups are still important, even if they have a relatively low mean income. The third scheme is really a simple oligarchy that corresponds reasonably well to a naive Marxist view of how a capitalist society works. This last measure, though somewhat extreme, is intended to provide a first approximation of how policy decisions are made in countries where power is concentrated in the hands of the capitalist class. 1/ 1/ Note that the choice of the top 5 percent of the overall distribution is completely arbitrary. In general, the smaller the elite, the more relative power is ascribed to capitalists. Given the way we generate the overall distribution, it is a simple matter computationally to change the definition of the elite. For a description of the tecinique, see Dervis, de Melo and Robinson (1981), Chapter 12. - 43 - Table 12 gives the three measures of relative political power for the different groups in the Base Run. As one moves from group shares in total population to shares in total income to shares in the elite population, the relative power of capitalists increases. Of the other groups,.only organized labor essentially maintains its relative share of political power In all three measures. All others lose, with marginal labor and service sector labor having almost no representation in the elite in spite of their large shares of the total population. Under the three different policy regimes, population shares, of course, do not change but there are significant changes in the relative position of the different socioeconomic groups by the two other measures. Table 12 also gives the average of the absolute values of change in group shares for the different archetypes and different policy regimes. Except in the ME economy, devaluation leads to much less change in group shares -than either premium rationing or premium rationing with a fixed real wage. The,two premia policy regimes have a similar impact on the relative position of different groups. One might argue that the average absolute:change in group shares is a rough indicator of the,extent of social-and political disruption caused by a particular policy since..socioeconomic groups can be expected.to resist.any change in their relative.power in the polity. Thus in the CL and.-PE economies, devaluation is the policy regime that is-the least disruptive of.the'existing political order.:Whether,this,'fact makes devalua- tion more or less desirable is a question that cannot be answered without Table 12 Measures of the Distribution of Political Power Base Run Shares (Percent) in: Total Population Total Income Economic Elite-a/ Groups CL PE ME CL PE ME CL PE ME Farmers 28.6 38.1 23.8 17.0 21.8 14.4 6.5 7.1 5.8 Marg. Labor & Unempl. 28.6 25.7 31.5 16.1 15.9 17.5 0.2 0.3 0.2 Org. Labor 9.5 8.6 14.3 9.4 8.9 13.2 8.9 9.0 11.1 Serv. Labor 28.6 22.9 25.7 21.8 16.5 17.5 3.8 2.0 1.8 Ag. Capital 1.8 2.2 1.5 13.7 17.8 11.7 31.2 39.5 26.3 Ind. Capital 1.1 1.1 1.7 12.1 11.5 17.7 20.6 20.1 31.2 Serv. Capital 1.8 1.4 1.5 9.9 7.6 8.0 28.8 22.0 23.6 Sum 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Policy Regime Average Absolute Change in Group Shares Devaluation 0.0 0.0 0.0 0.4 0.6 2.1 0.3 0.6 2.0 Premium Rationing 0.0 0.0 0.0 1.5 1.7 2.1 1.2 1.3 2.0 Premia, Fixed Wage 0.0 0.0 0.0 1.6 1.8 1.9 1.2 1.4 1.8 a/ The economic elite is defined as the top 5 percent of income recipients in the overall distribution of personal income. - 45 - considering the desirability of the existing order, an analysis well beyond our scope. 1/ Given the "political power" shares in Table 12 and the informa- tion on changes in the mean incomes of the socioeconomic groups, we can undertake an analysis of the political feasibility of different policy regimes. Equation (1) defined Ri(Pi;P2) which measures the relative gain or loss for group i between policy regimes PI and P2. Expressions for the average gain of the gainers and the average loss of the losers, weighting by political power shares, are given by: G(P1;P2)=£i WiRi(PI;P2) iWi for all Ri > 0 (2) L(P 1;P 2) EiwiRi(PI;P2) / iWifor all Ri < 0 where Pl,P2 are policy regimes, W is the relative power weight of group i, and R is the relative gain of group i from equation (1) The average gain and loss functions reflect the intensity with which different groups support or reject one policy compared to another. The political feasibility of one policy compared to another depends, in addition, on the relative political power of the gainers compared to the 1/ There is an active literature in political science analyzing the political strains that accompany economic development. For an entry into this literature, see Huntington and Nelson (1976). - 46 - losers. Given that we used relative political power weights in defining the average gains and losses, G and L, the political feasibility of a particular policy choice depends on whether the weighted difference between G and L is positive or'negative. We can thus define an index of political feasibility, F, which is equal to this weighted difference. It is also just equal to the weighted average of the relative gains, R1, over all groups, gainers and losers: F(P1;P2) = EiWiRi(PI;P 2)) Note that since we have three different political power weighting schemes, there are three different feasibility indices for pairwise comparison of policy regimes. Table 13 presents the average gains and losses of the gainers and losers, G and L, as well as the index of political feasibility, F, for all pairwise comparisons of the three different policy regimes. In each pairwise comparison of policies, the first"column gives the relative political power of the gainers. We have already discussed which groups gain and lose and it is important to remember here that the group composi- tion of the gainers and losers does vary across archetype economies and policy regimes. However, within each archetype and pairwise comparison, the differences in G, L and F are due only to the different political weighting schemes used. Table 13 Policy Comparisons Devaluation vs. Premium Rationing Devaluation vs. Premia vith Fixed wage Premia vs. Premia with Fixed Wage Political Average Average Feasibilityl Political Average Average Feasibility Political Average Average Feasibility Archetype Weighting Weight of Gain Losa Index Weight of Gain Loss Index Weight of Gain Loss Index Econony Scheme Gainers G L F Gainers G L F Gainers G L F CL Population Shares 68.5 4.1 4.6 1.4 39.9 8.1 2.9 1.4 61.9 2.9. 4.5 0.1 CL Income Shares 56.1 3.8 14.2 -4.1 40.0 7.1 10.0 -3.0 74.6 2.7 4.0 1.0 CL Elite Shares 46.8 15.6 -14.0 -0.2 46.6 6.3 18.8 -7.2 90.9 2.8 1.3 2.5 PE Population Shares 74.7 10.0 12.6 4.2 49.0 15.2 5.8 4.6 65.7 1.8 2.7 0.3 PE Income Shares 64.4 10.2 20.0 -0.6 48.6 14.1 13.1 0.1 73.2 1.6 2.4 0.6 PE Elite Sh4res 55.9 11.8 26.0 -5.1 55.6 13.3 24.3 -3.4 90.6 1.7 0.5 1.5 ME Population Shares 71.1 5.2 6.3 1.9 39.6 10.4 2.9 2.3 68.5 4.2 7.6 0.5 ME Income Shares 56.8 5.3 9.4 -1.1 39.3 11.0 5.4 -1.0 82.5 4.2 7.6 2.2 ME Elite Shares 43.4 5.3 12.1 -4.5 43.2 11.8 8.9 0.0 99.8 4.6 7.6 4.6 Notes: See text for definition of G,L and F. - 48 - Consider, for example, the comparison of devaluation with preniun rationing. As the political weighting scheme goes from population shares to elite shares--democracy to oligarchy--the relative power of the gainers declines substantially. In the comparison of premium rationing with and without a fixed real wage (the last four columns), the effect is reversed, with the political weight of the gainers increasing with the more oligarchic weights. In the first comparison, the gainers tend to be the small farmers and laborers whose population share is high but whose representation in the top 5 percent is small. In the last comparison, the gainers are everybody but marginal and organized labor, so their weight is high under all three schemes, but especially so for elite shares. The average gains and losses, G and L, measure the intensity with which the gainers and losers care about the outcome. For the first two policy comparisons, with a few exceptions (six out of eighteen, given the different weighting schemes and archetypes), the feasibility index indicates that those who care the most will determine the policy choice. 1/ In the third pairwise comparison, between premium rationing with and without a fixed real wage, the opposite is the case in all three archetypes for the first two weighting schemes. Even though the losers lose more than the gainers gain, their weight is so low that the feasibility index is always positive. Table 14 summarizes the ranking of different policies based on the pairwise comparisons in Table 13. It is interesting to note that the various pairwise comparisons never violate transitivity. That is, if policy P1 is preferred to P and P is preferred to P3. it is always 1/ Note, however, that the measures of intensity also reflect the relative political power of different socioeconomic groups since they use the political power weights in defining the average gains and losses. - 49 - Table 14 POLICY RANKINGS Archetype Economv Weighting Scheme Policy Ranking CL Population Shares (D) > (P) > (P,y) CL Income Shares (P) > (P,W) > (D) CL Elite Shares (P) > (P,t.) > (D) PE Population Shares (D) > (P) > (P,W) PE Income Shares (P) > (D)_> (P,W) PE Elite Shares (P) ' (P,W) > (D) Population Shares (D) > (P) > (P,W) Income Shares (p) > (P,W) > (D) Elite Shares (P) > (P,') * (D) Notes: (D): Devaluation, (P): Premium rationing, (P,W): Premium rationing vith fixed real wage, > : preferred to, and a : indifferent to. - 50 - true that the pairwise comparison shows that P1 is preferred to P3. as would be predicted by assuming transitivity. Across all three economies, devaluation is the most preferred policy under the democratic weighting scheme while premium rationing is the most preferred under both of the other weighting schemes. Indeed, with one exception (PE economy, income share weights) devaluation is the least preferred policy under the other schemes. It is interesting that devaluation, which skews the distribution of income the least, always pre- vails in a more democratic political system. Noting that historically more democratic political processes tend to emerge in the more developed countries, one might speculate that the Kuznets U hypothesis (that the distribution of income first becomes worse and then improves as the process of development unfolds) is at least partly related to changing policy choices as political power shifts over time. There is remarkable agreement in policy rankings, given the choice of weights, across the three archetype economies. This is true in spite of the fact that the particular groups which gain or lose, and the magnitudes of their gains and losses, vary widely. Although our par- ticular choices of policy regimes to compare were purposely made to enhance their differences, it is tempting to conclude that there are generalizations about the political feasibility of different policies that are valid for a wide variety of economic environments0 The universal dominance of devalua- tion under a democratic weighting scheme and of premium rationing under the other schemes is certainly consistent with casual observation of a wide variety of countries, although our three simple weighting schemes do not come close to capturing adequately the wide variety of political systems repre- sented in the world. - 51 - VII. Conclusions This paper has attempted to capture the variety of ways in which semi-industrial developing countries have been affected by the recent external shocks. Caught between rising costs on the import side and a shrinking of markets on the export side, these countries have had to achieve a major adjustment to restore equilibrium. By constructing three archetype economies that cover a broad range of initial conditions--resource endowments, past trade policies, and economic structure--one can get a feel for the quantita- tive importance of initial conditions in adjusting to the external shock. While all three economies suffered, a somewhat surprising result is that for the particular external shock we simulate (a decline in export markets and an adverse movement in the international terms of trade), the closed, rather than the open, economy is the most severely affected. This result is due to the relative importance of imported intermediate inputs in the closed economy. These inputs are crucial and cannot be easily replaced by domestically produced intermediates, so the economy has to bear the full brunt of their rising costs. The composition of imports determines to a large extent whether the burden of adjustment falls on consumption or investment. For the manufacturing exporter (a food importer), the consumption loss tends to be high. On the other hand, for the closed economy and for the primary exporter (which both import capital goods), the decline in GDP falls most heavily on investment. The rate of real investment, which is around 20 percent of GDP prior to the shock, falls to 17 percent in the closed and primary exporting economies. - 52 - How the distribution of income among socioeconomic groups is affected by the external shocks depends both on initial conditions and on the choice of adjustment policy. In general,. the distributional shifts are most pronounced for the manufacturing exporter where farmers and agri- cultural capitalists gain regardless of the selected policy while all cate- gories of labor lose. In the other economies the response is more varied with industrial capitalist losing under devaluation but gaining under a policy of premium rationing of imports. Finally, when real wages are fixed, the contractionary effect of the external shock is more pronounced for the manufacturing exporter, which experiences a 10 percent fall in mean income along with an unemployment rate of 7.1 percent. The picture that emerges from comparing the impact of alter- native adjustment mechanisms on the distribution of income in different economic environments is complex. In particular, the distribution of premia, amounting to close to 10 percent of GDP, raises the mean income of some recipient groups by up to 25 percent, in spite of a sizable decline in overall mean income. With so much at stake, it is no surprise that the choice of policy instrument is such a politically sensitive topic. The fact that the choice of policies has a significant impact on the distribution of income leads to a set of political questions that go beyond the usual economic analysis of policy choice. The distribution of political power is also affected by policy choice and hence the political struggle between the gainers and the losers may well decide the final choice of policy. By making quite strong assumptions about the way socioeconomic groups operate in the political process, one can determine the policy most - 53 - likely to be chosen. The analysis works with three different measures of relative political power of socioeconomic groups, reflecting political regimes ranging from democratic to oligarchic. While the simple assumptions we have made do not do justice to the relevant theories of political and economic conflict, we hope this study represents a useful first step and indicates one important direction for future research in the application of planning models for policy analysis. - 54 - REFERENCES Adelman, I. and S. Robinson, (1978), Income Distribution Policy in Developing Countries: A Case Study of Korea, Stanford University Press, Stanford. Chenery, H., (1979), Structural Change and Development Policy, Oxford Univer- sity Press, London. and R. Syrquin, (1975), Patterns of Development: 1950-1970, Oxford University Press, London. Cooper, R. N., (1971), Currency Devaluation in Developing Countries, Prince- ton Essays in International Finance, no. 86, Princeton, New Jersey. Dervis, K.9J. de Melo and S. Robinson, (1981), Planning Models and Development Policy, Cambridge University Press, forthcoming. Findlay, R., (1973), International Trade and Development Theory, Columbia University Press, New York. Hirschman, A., (1949), "Devaluation and Trade Balance: A Note," Review of Economics and Statistics, vol. 31, pp. 50-53. Huntington, S., and J. Nelson (1976), No Easy Choice: Political Participation in Developing Countries, Harvard University Press, Boston, Massachusetts. Jain, S., (1975), Size Distribution of Income: A Compilation of Data, World Bank, Washington, D.C. Johnson,. H., (1960), "The Costs of Protection and the Scientific Tariff," Journal of Political Economy, vol. 68, no. 4, pp. 327-345. Krugman, P. and L. Taylor, (1978), "Contractionary Effects of Devaluation," Journal of International Economics, Vol. 8, pp. 445-456. Kubo, Y. and S. Robinson, (1979), "Sources of Industrial Growth and Structural Change: A Comparative Analysis of Eight Countries," mimeo, World Bank, ,Washington, D.C. Lewis, W. A., (1980), "The Slowing Down of the Engine of Growth," American Economic Review, vol. 70, no. 4, pp. 555, 564. Melo, J. de, and S. Robinson, (1980), "The Impact of Trade Policies on Income Distribution in a Planning Model for Colombia," Journal of Policy Modeling, vol. 2, no. 1, pp. 81-100. Papanek, G., (1972), "The Effect of Aid and Other Resource Transfers on Saving and Growth in Less Developed Countries," Economic Journal, vol. 82, pp. 934-50. Taylor, L., (1980), Macro Models for Developing Countries, McGraw Hill, New YorX. - 55 - Constructing Three Consistent Social Accounting Matrices Appendix A to An Economic and Political Analysis of Alternative Trade Adjustment Policies in Three Archetype Developing Economies by Jaime de Melo and Sherman Robinson December 1980 - 56 - - 57 - Appendix A Constructing Three Consistent Social Accounting Matrices I. Introduction This appendix describes the method used to generate the three archetype data sets used in the paper. Our approach is to work within the framework of the Social Accounting Matrix (SAM) and to use a variety of techniques to impose consistency at different stages of the gen- eration of the data.-/ While the method is applied to archetypal, made-up data sets based on "representative" countries, the approach would also be useful in dealing with disparate data for a single country. In assembling any economy-wide data base, whether real or representative, inconsistencies emerge which require one to make decisions about the relative reliability and accuracy of different data sources. For example, it is usually the case that the national income and product figures that emerge from input-output data are different from the official national accounts. The foreign trade statistics in the input-output accounts, the national product accounts, and the trade accounts are often quite different. Savings estimated from institutional flow-of-funds data are often different from investment estimated from production accounts. 1/ For a description of a SAM see Pyatt and Thorbecke (1976). - 58 - Finally, for example, the government flow-of-funds data are hard to reconcile among the different accounts. Such inconsistencies are sometimes due to variations in definitions among the different accounts but, more often, they arise from differences in coverage and accuracy. In practice, one is thus forced to assemble an economy-wide data base in sequence, starting with what are considered to be the more reliable data and forcing data assembled at later steps to be consistent with control totals from former steps. The method described below represents one such approach, with a definite sequence of steps in which consistency is forced, applied to constructing the data sets for the three archetype economies. There is, however, nothing sacred about the particular sequence we have chosen and one might well change it in particular applications. Also, in practice, one generally iterates informally across the steps rather than following a simple sequence. If data at a later step reveal glaring inconsistencies, one often questions the reliability of the data at earlier steps and goes back to reconsider them as well. The flowcharts in Figures A-1 and A-2 present the adjustment procedure in schematic form.-/ The steps are given by Roman numerals and the additional data required at each stage are described in boxes at the left of the figures. There are ten steps in all and forced reconciliation 1/ A computer program is available to do the procedure. - 59 - in which inpuNt parametersiare changed is-done at three steps (indicated by diamonds). The data which are distinctively different across the three archetype economies are marked by an asterisk. The adjustment is done in two distinct stages. First, in steps I to IV in Figure A-1, the production accounts based on an input-output table are reconciled. Second, in steps V to X in Figure A-2, the income and expenditure accounts are reconciled with the pro- duction accounts. Thus, in this procedure, the input-output accounts provide the starting point and are considered to be the controlling data source. 2. Production Accounts Starting from the input output accounts, steps I to III in Figure A-1 generate the final demand columns and the total sectoral value added row. Given total final demand and production by sectors, the available supply of goods by sector for intermediate use is de- termined as a residual. Given an initial set of input-output coefficients, sectoral demand for intermediate goods is determined. Unless we started with a consistent set of accounts, these demands will not equal supply and some reconciliation procedure is required. The procedure we use is called the "RAS" method and consists of iteratively adjusting the rows and columns of the matrix of intermediate inputs - 60 - Figure A-1 Reconciliation of the Production Accounts * Sectoral production *SImporal produci o n Composite good prices *Import ratios Net pie * Export ratios pie Exporif ratios Sectoral value added Exprifrte Sectoral exports Export subsidy rates Sectoral imports Indirect tax rates World prices . . Exchange rate Sectoral value-added ratios Ratios to total value added,: Macroeconomic aggregates: Private consumption, Value added Government consumption, Balance of trade Investment Private consumption Government consumption Investment Private sectoral consumption shares Government sectoral consumption shares ds By sectors: Private consumption Government consumption Investment by sector of Sectoral capital-output ratios origin Capital composition matrix Capital stock Caitl opoito mtrxIntermediate goods supply * Investment shares by sector of I destination ._,_,_ . - 61 - Figure A-1 (Cont'd) Input-output > Intermediate goods coefficients demands not converged RA Procedure converged Input-output coefficients Intermediate goods, IV supply = demand * Input variables which differ across the three archetype economies. - 62 - (the input-output coefficients) until the matrix converges on the specified row and column control totals.l/ The control row totals are given by the total supply of intermediate inputs by sectors, given the specified sectoral production and final demand. The column control totals are total purchases of intermediate inputs by sectors and are given by the total value of sectoral production minus sectoral value added. If one started with a consistent input-output table, steps I to IV would not be necessary. However, our intent was to build three distinctively different data sets for three archetype economies. The major differences in the three archetype economies are in their production, consumption, trade and employment structures. Tables 13-1 to 13-3 in Chapter 13 indicate the major differences. In addition to variations in the structure and volume of trade, we also assuimpd that tariffs and subsidies differed, reflecting historical differences in trade policies. We thus did not start with three sets of consistent accounts. Instead, we started with the three different sets of 1/ For a description of the algorithm and a discussion of its convergence properties, see Bacharach (1970). The sum of the control colunn and control row must be the same. In this case, the requirement is that total value added equal total final demand (i.e., that national income equal national product). - 63 - structural 4ata and a co =n set of input-output coefficients and used the procedure s:immarized in Figure A-1 to create three consistent data s;ets. The result is tha.. the input-output tables are slightly different between the three economies.-/ The input variables which are significantly different across the three archetype economies are marked by an asterisk. There was some adjustment of input data required to achieve similar input-output structures and also to achieve a similar balance of trade. Such adjustments were relatively minor, but did provide a check of the "reasonableness" and consistency of the quite different production and trade structures characteristic of the archetype economies. A.3. Income and Expenditure Accounts Steps I to IV yield a consistent set of production and aggregate income (value added) accounts. The next stage is to generate a consistent set of disaggregated income and expenditure accounts. The procedure we use is given in Figure A-2, Steps V to X and is essentially to generate the functional, institutional and household distributions by disaggregating value added. Receipt and expenditure balances are maintained for each account, with forced adjustment required at two steps to ensure consistency with the production accounts. The total labor force is assumed to be the same across the three 1/ The sectoral val,.e-added ratios are the same across the countries, having been specified to give the control column totals. - 64 - Figure A-2 Reconciliation of the Income and Expenditure Accounts 0 Value-added shares: ., Capital by sector, Labor by sector, and Functional income distribution: Skill category Capital V Labor by skill categories Tax rates: _ Indirect (as in Step I), Institutional income distribution Direct Enterorises Households VI Government Rest of the World Foreign exchange: Remittances Capital flows *Employment by skill catemporyentdby skilr Household income distribution category and sector x~atr ae * Non-wage income byafter taxes VIc earners: by recipient categories, farmers and per capita and aggregate capitalists . . - 65 - Figure A-2 (Cont'd) 0 Savings rates: e Institutional, | T otal savings VIII Household h c Adjust institutional __ No / Savings= \ .savings rates n-smn Yes- Aggregate consumption by household categories IX and government Sectoral consumption Average expenditure demand shares by consuming l groups L not converged RAS Sectoral no converconsumption Procedure supply (from Step III) converged - 66 - Figure A-2 (Cont'd) 0 Average expenditure shares Sectoral consumption, X demand = supply * Input variables which differ across the three archetype economies. - 67 - economies and the shares of sectoral value added by different skill categories of labor and by capital are also assumed to be the same. However, the skill composition of the labor force and the sectoral structures of employment differs across the economies, reflecting the very different structures of production. The sectoral structure of employment was adjusted informally so that wage differentials by skill categories and sectors would be similar across the three economies. The result, in Step V, is the functional distribution of income. From the functional distribution, we move to the institutional distribution in Step VI by generating the government and rest-of-the-world accounts, in addition to the household and enterprise accounts. Tax rates, both direct and indirect, are assumed to be the same. Given the different tax bases, the total size of the government sector differs, but not dramatically. Inputs for foreign remittances and capital inflows are shown to indicate where they would enter, but they are assumed to be zero in the three archetype economies. The household distribution by socioeconomic groups is really a reairanging of the "household" institutional distribution since, in both, households are classified by source ofincome. The only new data are the number of people in each household category. The labor households correspond to employment by skill category, the numbers of farmers and capitalists have been estimated so as to achieve sirmilar relative per capita mean group incomes across the three economies. In Step VIII, total savings are generated by applying fixed - 66s - -68 - FISzr== A-2 (Cont'd) savings rates (the same across the thuC economies) to institutional and household incomes. Tbtal savings So generated are compared to total investment from Step II in the pRoduction accounts. If they differ, then the savings rates of institutions (i.e., corporatelsaving) are Ave age spendRI-=ra srzss adjusted to achieve consi Otenc§. °er rn%lvi-(nr:g-s5 'tep IX generates' ,a . na SUPU)17 aggregate expenditure by ouseholds and government. Finally, in Step X, given average expenditure shares by consuming groups, total demand for products by sector is calculated. If this demand does not equal the sectoral supply of goods for consump- tion computed in Step III, an adjustment prodedure is required. A rec- tangular expenditure matrix is calculated with rows referring to sectors; ' Input variables which difftl/across the1_. t!irzs ntyp cz2:I.S and columns to household groups.- Each entry is the expenditure on good i (row) by household category j (c6lumn). The column control totals are aggregate expenditure by household groups and the row contrGi totals are total private consumption by sectors from Step III. The matrix of expenditures is then adjusted by the RAS procedure so that it satisfies the column and row control, totals.-/ The results yield ne-' average expenditure shares by household groups. The result of Step X is a set of average expenditure shares b/ 1/ Government expenditure shares are not adjusted, so the expenditurt matrix is defined to include any private consumption. 2/ Note that the procedure can be applied to a rectangular matrix. - 69 - household groups. The result of Step X is a set of average expenditure shares by household groups such that consistency is achieved between the sec- toral expenditures from the demand side and sectoral supplies of con- sumer goods from the production side. If the RAS procedure yielded dramatically different expenditure shares for the same socioeconomic group across the archetype economies, the specification of the structure of aggregate consumption in the production accounts would be suspect. In fact, we did do some informal adjustment of the structure of con- sumption in the production accounts so that there were no major dif- ferences in the structure of demand by the same socioeconomic group across the three economies. 4. Archetype Country Social Accounting Matrices Tables 1 to 3 give the SAM's for the three archetype economies (for the Base Runs described in the paper). Since production accounts are discussed in some detail in the analysis, no sectoral disaggregation is provided in the SAM's. Instead, the focus is on the institutional and household accounts. To make the presentation more compact, various accounts have been combined in the SAM's. Since capital income is all assumed to accrue to "enterprises," there is no separate institutional account for enterprises. Note that enterprise savings are shown as a Table A.1 Social Accounting Matrix: Closed Economy (CL) ~t u R e s 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 Activities 956 42 2 Commodities 461 69 71 37 86 47 42 37 389 93 108 Factors 3 Labor 265 4 Capital 229 5 Sum (3+4) 494 Households 6 Small Farmers 54 21 75 7 Marginal Labor 71 71 8 Organized Labor 45 45 9 Service Sector Labor 95 95 10 Agricultural Capital 62 62 11 Industrial Capital 87 87 12 Service Sector Capital 47 47 13 Sum (6 to 12) 265 217 482 14 Government 43 31 4 4 34 5 47 15 Cap4ta; Account 12 12 6 4 9 11 11 5 46 28 22 16 Rest of the World 64 17 Totals 998 .051 265 229 494 75 71 45 95 62 87 47 482 121 108 64 Table A.2 Social Accounting Matrix: Primary Exporter (PE) nditures 1 2 3 4 5 6 7 8. 9 10 11 12 13 14 15 16 Receipts 1 Activities 971 15 103 2 Commodities 438 110 86 43 81 78 48 35 481 72 132 Factors 3 Labor 316 4 Capital 292 5 Sum (3+4) 608 Households 6 Small Farmers 87 33 120 7 Marginal Labor 86 86 8 Organized Labor 53 53 9 Service Sector Labor 90 90 10 Agricultural Capital 100 100 11 Industrial Capital 101 101 12 Service Sector Capital 44 44 13 Sum (6 to 12) 316 278 594 14 Governmentt 43 27 5 5 39 5 54 15 Capital Account 14 14 10 5 9 17 14 4 59 37 22 16 Rest of the World 125 17 Totals 1089 1123 316 292 608 120 86 53 90 100 101 44 594 124 132 125 -__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __._ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __I_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __._ _ _ _ _ _ _ _ _ _ Table A.3 Social Accounting Matrix: Manufacturing Exporter (ME) Expenditures 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 Activities 973 22 98 2 Commodities 472 67 87 59 78 51 76 37 455 65 126 Factors 3 Labor 300 4 Capital 273 5 Sum (3+4) 573 Households 6 Small Farmers 53 20 73 7 Marginal Labor 87 87 8 Organized Labor 73 73 9 Service Labor Sector 87 87 10 Agricultural Capital 60 60 11 Industrial Capital 138 138 12 Service Sector Capital 44 44 13 Sum (6 to 12) 300 262 562 14 Government 49 24 7 3 52 5 67 15 Capital Account 11 11 6 7 9 6 10 2 40 53 22 16 Rest of the World 121 -Z't a', -'-- --Ie f- -s -'' - 7- 87- - I9 ' . - 73 - payment by the factor "capital" to the capital account (row 15, column 4). Also, in the factor account for "labor,", receipts by the four dif- ferent skill categories have been aggregated. Separate wage payments to different categories can be seen in the first four household cate- gories which correspond to the four skill categories of labor. Similarly, capital income in sectors has been aggregated in the factor accounts and somewhat disaggregated in the household accounts. The "small farmers" category of households receives a fraction (25 percent) of capital income from agriculture which reflects the fact that they are landowners. The "agricultural capital" category thus are larger farmers all of whose incomes are assumed to come from land ownership. Note that in the archetype economies, there are no remittances or foreign capital inflows which accrue directly to households. Instead, the balance of trade enters directly into the capital account and is thus assumed to be saved. This treatment is simple, but is justified for our purposes since we hold the balance of trade constant in the three econ- omies and across all experiments in the paper. A comparison of the three archetype SAM's indicates a number of interesting contrasts. First, note that the price normalization rule used in the CGE model is expressed in terms of composite good prices (see Appendix B). However, since tariffs and export subsidies differ among the three economies, so do domestic prices of imports and domestically produced goods. The result is that although gross production is virtually - 74 - the same across the three economies, the nominal receipts of "activities" are different, being higher in the more open economies which have signi- ficant export subsidies and hence higher domestic prices. The commodity accounts (total absorption) follow a similar pattern. Given their quite different production structures (and slightly different relative prices) the three economies also have different aggregate value-added ratios. The ratio is lowest for the closed economy (CL) and highest for the primary exporter (PE) -- 54 percent and 60 percent, with the manufacturing exporter (ME) having a ratio of 57 percent, in market prices. These differences, of course, lead to significant differences in GDP among the three economies. Columns 3 to 5 show the distributi6n of factor income to households (rows 6 to 13). Table 4 gives the percent distribution of nominal household income by socioeconomic groups. The quite different structures of the three economies are clearly evident, with small farmers and agricultural capitalists having small shares in the manufacturing exporter (ME), followed by the closed economy (CL) and primary exporter (PE). Industrial capitalists have by far the largest share in the ME economy. Columns 6 to 13 give the distribution of household income among consumption, savings, and government (taxes). The aggregate rates of direct taxation are similar across the three economies, being slightly higher in the ME economy (12 percent; compared to 10 and 9 in the CL and - 75 - Table A.4 Group Shares in Nominal Household Income (Percent) CL PE ME Small Farmers 15.6 20.2 13.0 Marginal Labor 14.7 14.5 15.5 Organized Labor 9.3 8.9 13.0 Service Sector Labor 19.7 15.2 15.5 Agricultural Capital 12.9 16.8 10.7 Industrial Capital 18.0 17.0 Service Sector Capital 9.8 7.4 7.P Total 100.0 100.0 100.0 Notes: CL = Closed economy PE = Primary exporter ME.- Manufacturing exporter - 76 - and PE economies). The aggregate household savings rates are essentially the same in the CL and PE economies (11 percent), but are much lower in the ME economy (8 percent). This result comes from the fact that in the ME economy, government saving is much more important and hence corporate saving and private saving, especially for capitalists, are lower. Government saving is 42 percent of investment in the ME economy compared to 26 and 28 in the CL and PE economies.-/ Note finally that the different volumes of trade are evident in the three SAM's by comparing the rest-of-the-world columns. The volume of trade in the CL economy is about half that in the other two. As noted above, the PE and ME economies have significant export subsidies (shown as an entry from Government to Activities; row 1, column 14). The average tariff rates are also quite different, more than double in the CL economy compared to the other two. 5. Consumption: The Linear Expenditure System In the construction of the three archetype SAM's, average expenditure shares by the different classes of consumers were given exogenously and then adjusted to achieve consistency with the production 1/ Government saving is higher because, given the larger capitalist income in the ME economy, the direct tax take is higher (given that the tax rate is highest on capital income). The effect is to shift savings from capitalists and enterprises to government. - 77 - accounts. The CGE model requires the.specification of a complete set of expenditure equations-and-A variety of expenditure systems have been used in various models. In the model used in this paper, we have specified the Stone-Geary linear expenditure system (LES). For each socioeconomic. group, consumer demand is given by (omitting a group subscript).: C= + pi[Y_ p y.] (1) where Y is total nominal expenditure for the group, yi are the committed expenditures of "subsistence minima" in physical terms and 61 are the marginal budget shares which determine the allocation of supernumerarv income (i.e., expenditure above that required for pur.:h:sing the sub- sistence minima).I/ Given the average budget shares, there are a variety of ways to esti,mate the parameters of the system depending on the exten. and quality of available data. For our archetype data sets, we have cho- sarn to compute the parameters of the LES for e-ach grc-.p --'en endogenous- ly specified average budget shares, income elasticities of demand. -nd a pa,-ar,e,e- measunring the elasticity of the marginal 'Atility of __come 1/ The e:penditure equatlk;ns are often gl4-en in per canita terms. See Brown and Deaton (1972) for a survey of differc-nit systems. - 78 - with respect to income (often called the "Frisch parameter") In the LES, the Frisch parameter is equal to the ratio of total expenditure to supernumerary expenditure: = -Y/(Y-S), S = Z P. *Y (2) Given the average budget shares and expenditure elasticities, the marginal budget shares are given by: Bi = Ei ai (3) where Li are the expenditure elasticities and cai are average budget shares. Note that the marginal budget shares must sum to one which is equivalent to imposing the condition, known as Engel aggregation, that the sum of the expenditure elasticities weighed by average budget shares must equal one. The subsistence minima, are related to the other parameters according to the following equation: 1 Se s (Y/P1)(an Bo ad o (4) 1/ See Fri.sch (1959) and Brown and Deaton (1972). - 79 - Our estimates of the average budget shares, income elasticities, and Frisch parameter are based on various cross-country studies, espe- cially Lluch, Powell and Williams (1977).-1 They estimate that the Frisch parameter rises from -7.5 to -2.0 as per capita income rises from $100 to $3,000 (in 1970 US dollars).-2 Our archetype countries are assumed to have a per capita GDP of about $500 which, given the various group mean incomes, implies that the Frisch parameter ranges from -5.0 to -1.6. The estimates we have used are given in Table 5, along with the average budget shares and income elasticities. Given the parameters of the LES, the own and cross price elasticities of demand can be computed from the following equations: nii = -Cit (PiYt/Y) ((5) =i ECi (Y iYi /)(6) In our eight-sector model, there is consumer demand for only five sectors. No distinction is made between durables and non-durables. The own price elasticities of demand for food range from -0.35 to -0.55 across the groups, while that for services range from -0.35 to -1.07. 1/ See also Chenery and Syrquin (1975) and Chenery (1980). 2/ See Lluch, Powell :tnd Williams, pp. 54-55. Table A.5 Expenditures Shares and Elasticities by Groups Average Expenditure Shares (%) Small Marginal Organized Service Agric. Industrial Service Farmers Labor Labor Labor Capital Capital Capital Primary 50.2 35.2 20.1 25.1 15.2 10.1 10.1 Food 20.4 25.5 30.7 35.7 15.5 20.6 20.6 Other consumer 9.8 9.8 14.8 9.8 19.8 19.8 19.8 Social overhead 9.8 14.8 14.8 14.7 19.9 19.9 19.9 Services 9.8 14.7 19.6 14.6 29.6 29.6 29.6 X Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Income Elasticities Primary .92 .90 .74 .82 .43 .43 .43 Food .92 .90 .74 .82 .43 .43 .43 Other consumer 1.19 1.16 1.19 1.20 1.20 1.20 1.20 Social overhead 1.19 1.16 1.19 1.31 1.12 1.12 1.12 Services 1.19 1.16 1.39 1.31 1.38 1.38 1.38 Frisch parameter -4.0 -5.0 -4.0 -4.0 -1.6 -1.6 -1.6 - 81 - These values are within the ranges reported by Lluch, Powell and Williams for households with mean incomes ranging from $500 to $1,500.1/ 1/ See Lluch, Powell and Williams, p. 248. - 82 - REFERENCES Bacharach, (1970), Biproportional Matrices and Input-Output Change, Cambridge University Press, Cambridge. Brown, A. and A. Deaton, (1972), "Surveys in Applied Economics: Models of Consumer Behavior," Economic Journal, vol. 82, pp. 1145-1236. Frisch, R.,(1959), "A Complete Scheme for Computing All Direct and Cross Price Elasticities in a Model with Many Sectors, Econometrica, vol. 27, pp. 177-196. Lluch, C., A. Powell and R. Williams, (1977), Patterns in Household Demand and Saving, Oxford University Press, London. Pyatt, G. and E. Thorbecke, (1976), Planning Techniques for a Better Future, ILO, Geneva. - 83 - The Equations of the Model Appendix B to An Economic and Political Analysis of Alternative Trade Adjustment Policies in Three Archetype Developing Economies by Jaime de Melo and Sherman Robinson December 1980 - 85 - Appendix B The Equations of the Model This appendix presents the complete set of equations describing the model. All endogenous variables are denoted by capital Roman letters; exogenous variables and parameters by capital Roman letters with a bar ot Greek letters; and policy variables by small Roman letters. Subscripts i, q, and g refer to sectors, labor categories, and socioeconomic groups. Superscripts m, d, and e are used to distinguish imported, domestically produced, and exported goods. An expanded version of this Appendix is available on request. 1. Trade Aggregation Functions The composite good in each sector is defined as a CES aggre- gation of domestically produced (D ) and imported (Mm) goods. I 1 Qi = Bi[ (Mi) i + (l-8.)(D.) i] i 1 ju1 1 1 2. Trading Price Equations m P, = r(l+t'i+PR)R P = 7.(l+t )R 1 1 1 - 86 - Where Tri, 7Ti are the exogenous world prices of exports and imports, tei and t the corresponding export subsidies and import tariffs. The exchange rate is denoted by R. When R is fixed, the uniform premium rate PR is positive. 3. Imported Demands Mm iai pi i D i Pm i Where U. is the elasticity of substitution associated with the CES aggregation function. Given (1), the ratios rd = D /Q can also be i i i determined. 4. Export Supply d ...e .d E = S X -e The export supply ratio, Si , is given exogenously. 5. Foreign Exchange Market In equilibrium the excess demand for foreign exchange must equal zero: - 87 - e 0 £ wt *i - Zir1 Et ' 0 The exchange rate, R, is the endogenous variable which adjusts so as to clear the market for foreign exchange. Imports are given by d e d (3) and exports from (4) i.e., Ei = SiXi. 6. Production Functions d -_--ct 1-ct iq X A iK i = 1 Li iq Technology is described by two-level Cobb-Douglas production functions. Sectoral capital stocks are fixed. Intermediate goods are required according to fixed input-output coefficients. 7. Net Prices Pi± (1ti)[P~ e + (1-Si) Pd p (1 MP i r 'a± where 1 ± ± ± ~ ~ ~ ~ dla - - 88 - These equations deduct the fixed intermediate input costs to provide the net receipts to the firm of selling a unit of output; Pi is the price of a unit of the composite good and ti is the value-added sub- d~~~~~~~~ sidy (or tax). Pi is the domestic price. The composite good price equation is the cost function dual to the trade aggregation function, equation (1). 8. Factor markets The demand for labor of each skill category in each sector is given by solving the first-order conditions for profit maximization: W mp . : iajLj Where Li is the labor aggregate defined in equation (6). The aggregate supply of labor in each skill category is assumed fixed. The average wages, Wq, are determined endogenously so as to clear the labor mar- kets; i.e., achieve zero excess demands: E L - L - 0 iq unless the real wage is fixed in which case the above equation is dropped and replaced by: _ W w - T v p S qi q where the Qi are weights in the price index that defines the real wage W - 89 - 9. Income Distribution and Macro Balances The average income of all groups -- different types of wage earners, farmers and recipients of non-wage income -- are determined endogenously. The distribution of income within each group is given by a two-parameter lognormal distribution function. The logvariance is exogenous and the logmean is a function of the endogenous group mean income. The equation is p = log(Y ) - 1/2 G2 where Y is group mean income and a2 is the logvariance. Aggregate investment is determined by applying fixed savings rates to the income of the various institutions in the economy: enterprises, households and government. Foreign cap'tal inflow is also assumed to be saved. The allocation of total investment to sectors is given exogenously. The demand for investment goods by sector of origin, Zi, is calculated by assuming a fixed composition of aggregate capital which differs across sectors. 10. Consumer Demand Equations Consumer demands for composite goods are given by a separate set 6f linear expenditure functions for each consuming group: pi Cig [Pi .ig + 6ig(Cg - 9 Pi Yjg)] Ng - 90 - Where C is aggregate group consumption and N is the number of people in each group. y and ( are parameters of the system. 11. Product Markets d Domestic prices, Pi, are determined endogenously so as to clear the product markets, making the excess demand for goods in each market equal zero: d d _d D +E - x = i i i d d d d d Where Di = F. + Ci + G + Zi The demands are solved behaviorally in terms of composite goods. The ratios of domestic to composite goods, d d ri =D.I/Q, are given by an equation analogous to (3). Thus: * d d - d d d ;d - d Z d - d. F1 s r e1i ij 1 ri ig; i ri i i i 12. Price Normalization The normalization rule maintains a constant price level measured by a wholesale price index: E ni Pi a P - 91 - Where Q iis base-year production weights. Table Al summarizes the endogenous variables and the associated equations. The exchange rate, wages and prices are determined so as to clear the foreign exchange, labor, and product mar~kets. The system can only determine relative prices, so an additional price normalization equation is required to complete the system. Table Al: Endogenous Variables and EqLuations Endogenous Variable Equation Reference Endogenous Variable Equation Reference Qi ()Pi, Pi (7) pm pe (2j W, Li (8) ,rd (3) Zi Gi Yg (9) Ed (4) ig (10) R (5) Pd, Did11 Xi, Li (6) HG3881.5 .W57 W67 no.442 c.3 de Mlelo, Jaime. Trade adjustment policies and income distribution in three archetype developing .