l g g- 2 E RESEARCH PAPER SERIES JutJ Ig 9 3 ENTERPRISE BEHAVIOR AND ECONOMIC REFORMS: A COMPARATiE STUDY IN CENTRAL AND EASTERN EUROPE AND INDUSTRIL REFORM AND PRODUcIvrIy iN CHINESE ENTERPRISES RESEARCH PROJECTS OF THE WORLD BANK CHINA NUMBER CH-RPS #3 JANUARY 1991 (REVISED JUNE 1993) Sources of Gains in Allocative Efficiency Gary Jefferson Brandeis University and Harvard Institute of International Development Harvard University Transition and Macro Adjustment Division Policy Research Department World Bank Washington, D.C. CONTENTS ACKNOWLEDGEMENT . 1 ABSTRACT ................................. ii INTRODUCTION ................................. 1 II. A CONVERGENCE ACCOUNTING FRAMEWORK ....... ............... 3 III. WORLD OIL PRICE SHOCK ................................. 5 IV. CHINA'S INDUSTRiAL REFORMS. 7 VI. CONCLUSIONS. 9 REFERENCES .. 13 APPENDIX ........................... 14 TABLES TABLE 1: SOURCES OF CONVEREGENCE/DIVERGENCE ................ 5 TABLE 2: INDEX OF RELATIVE MARGINAL PRODUCT OF LABOR IN PETROLEUM PRODUCTION AND TOTAL MANUFACTURING ...... 6 TABLE 3: U.S. MANUFACTURING: TEE OIL SHOCK LABOR'S MARGINAL REVENUE PRODUCT .................. 10 TABLE 4: NOMINAL MARGINAL REVENUE PRODUCTS: COMPARISON OF CHNA'S STATE AND COLLECTIVE INDUSTRY ..... 11 TABLE 5: SOURCES OF CONVERGENCE OF FACTOR RETURNS (ANNUAL RATES OF GROWTH) .................. 12 ACKNOwLEDGEMENT The research projects on 'Enterprise Behavior and Economic Reforms: A Comparative Study in Central and Eastern Europe", and "Industrial Reforms and Productivity in Chinese Enterprises" are research initiatives of the Transition and Macro Adjustment Division (PRDTM) of the World Bank's Policy Research Department and managed by I.J. Singh, Lead Economist. These projects are being undertaken in collaboration with the following institutions: The London Business School (LBS); Reforme et Ouvertures des Systemes Economiques (post) Socialistes (ROSES) at the University of Paris; Centro de Estudos Aplicados da Universidade Cat6lica Portuguesa (UCP) in Lisbon; The Czech Management Center (CMC) at Celakovice, Czech Republic; The Research Institute of Industrial Economics of the Janus Pannonius University, Peds (RILE) in Budapest, Hungary; and the Department of Economics at the University of L6dI, in Poland. The research projects are supported with funds generously provided by: The World Bank Research Committee; The Japanese Grant Facility; The Portuguese Ministry of Industry and Energy; The Ministry of Research and Space; The Ministry of Industry and Foreign Trade, and General Office of Planning in France; and the United States Agency for International Development. The Research Paper Series disseminates preliminary findings of work in progress and promotes the exchange of ideas among researchers and others interested in the area. The papers contain the views, conclusions, and interpretations of the author(s) and should not be attributed to the World Bank, its Board of Directors, its management or any of its member countries, or the sponsoring institutions or their affiliated agencies. Due to the informality of this series and to make the publication available with the least possible delay, the papers have not been fully edited, and the World Bank accepts no responsibility for errors. The authors welcome any comments and suggestions-. Request for permission to quote their contents should be addressed directly to the author(s). For additional copies, please contact the Transition and Macro Adjustment Division, room N-1 1065, World Bank, 1818 H Street, N.W., Washington, D.C. 20043, telephone (202) 473-1442, fax (202) 676-0083 or 676-0439. The series is also possible thanks to the contributions of Donna Schaller, Vesna Petrovic, Cecilia Guido-Spano and the leadership of Alan Gelb. i ABSTRACT This paper formulates a framework for analyzing the gains in allocative efficiency through the convergence of factor returns. Sources of convergence may include: (i) factor reallocations, (ii) differential rates of productivity growth, (iii) changes in production technologies, and (iv) changes in inter-sectoral prices, both product prices and factor input prices. Significant differences among factor retums may arise due to: (i) shocks that create temporary disequilibria, such as oil price shocks and regulatory changes, and (ii) uncompetitive markets, including, in the extreme, the allocation of factors through central planning rather than through the market. The paper applies its framework to an analysis of the convergence of factor returns following: (i) the oil shocks of the 1970s and (ii) the introduction of China's industrial reforms. Sources of Gains in Allocative Efficiency 1 I. INTRODUCTION In market economies, we find a persistent tendency for factor returns to equalize, an important property of market economies which contributes to productive efficiency. Significant differences among two or more sectors to factor returns may arise due to the following conditions: (i) shocks that create temporary disequilibria, such as oil price shocks, changes in trade regimes and new product or process innovations, and (ii) uncompetitive markets, including, in the extreme, the allocation of factors through central planning rather than through the market. In the first case, the passage of time, such as the years following the oil price shocks of the 1970s, should lead to a set of adjustments that reduce differences in returns to factors between the energy and non-energy sectors. In the case of central planning, economic system reform, such as that underway in China and E. Europe should lead to a convergence of factor returns. Following price or regulatory shocks or the introduction of new product or proceess innovation in market economies or following the creation of markets by reform-minded governments, factor returns may converge through a variety of channels. These include: (i) factor reallocations, (ii) differential rates of productivity growth, (iii) changes in production technologies, and (iv) changes in inter-sectoral prices, both product prices and factor input prices. This paper develops an accounting framework for measuring the contribution of each of these sources to changes in relative factor returns between two sectors. In market economies, profit-seeking behavior on the part of sellers and buyers tends to erode intersectoral and interregional differences in returns to labor, capital or materials. Studies conducted within industrial market economies concerning the costs of allocative inefficiency resulting from impediments to optimal resource allocation show that these costs are relatively small.' Nonetheless, transitory shocks can create disequilibria which require periods of time for adjustment. The oil price shocks of the 1970s, for example, by shifting back the energy supply curve, caused substantial increases in returns to petroleum products relative to other goods in the U.S. economy. Over time, however, these quasi-rents were eroded. By 1986, the initial increase in returns to labor employed in the petroleum products industry relative to returns in overall manufacturing had been eliminated. The question addressed by this paper is the nature of the adjustment mechanism. What combination of price adjustments, quantity adjustments or technological change led to the restoration of equilibrium in labor markets following the oil price shock? Prepared for the Fourth International Economics Conference on U.S.-PRC Cooperation in Beijing, November 10-12, 1992. Research support provided by the National Science Foundation and the Henry Luce Foundation is gratefully acknowledged. 1. See Leibenstein's (1976) review of this literature in which he concludes that in mature industrial economies, potential gains from allocative efficiency are trivial. 2 Sources of Gains in Allocauve Efficiency In China, Eastern Europe, and the former Soviet Union, economic reform initiatives seek to increase efficiency by introducing market-oriented policies and institutions into economies formerly dominated by state planning. Ecornomic system reform has also been an important policy thrust in many developing countries in which bureaucracy has dominated pricing and resource allocation decisions or within which product and factor markets are undeveloped. An important potential source of efficiency gain, in both socialist and developing economies, is price reform coupled with the introduction of factor markets to facilitate the deployment of labor, investment resoruces, and intermediate inputs toward activities where these inputs can generate their highest marginal social return and thereby improve allocative efficiency. At the outset of China's economic reforms in 1978, the industrial sector consisted of a core and a relatively small periphery. The core, consisting of state-owned enterprises, was relatively intensive in capital and materials, while the periphery, consisting of urban collectives and rural commune industry, was relatively intensive in labor. These patterns of factor intensity created and sustained substantial differences in factor returns. By the late 1980s, following a decade of reform, evidence was beginning to emerge that returns to capital, labor and materials were beginning to converge between the state-owned industrial core and the periphery. What combination of price adjustments, quantity adjustments or structural change was driving this converg;ence of factor returns? The restoration of market equilibrium among energy and non-energy sectors following oil price shocks in the 1970s and the convergence of factor returns within the industrial sector following the introduction of incentive and market reforms in China are but two examples of situations involving adjustment within a two-sector setting. Lewis (1954), Fei and Ranis (1964), and Yorgenson (1966) each modeled the dynamics of changing returns to labor between agriculture and industry within developing coluntries. Harris (1970) looked at the problem from a rural-urban perspective. The state and non-state, agriculture-industry, and rural-urban distinctions are but examples of a large number of dichotomous distinctions that warrant attention. Intuitively, changes in relative factor returns or marginal revenue products across two or more sectors result from two types of conditions. The first, affecting physical marginal products, consists of changes in relative factor intensities, differential rates of productivity growth, and technological changes that affect factor output elasticities. The second leaves physical products unaffected but, by changing relative prices, affects marginal revenue products. This 'latter source of change in factor returns includes both changes in relative product prices and changes in inter-sectoral factor input prices. At the national level, productivity growth may originate from any of three sources. These are: (a) productivity growth (technical change or realization of scale economies) within the firm, (b) increases in production efficiency resulting from the convergence of factor returns across sectors or firms, and (c) allocative efficiency in the general equilibrium sense in which prices reflect relative scarcities and consumer plus producer welfare are maximized. This paper Sources of Gains in Allocative Efficiency 3 evaluates the contribution of greater production efficiency to measured productivity growth and formulates an approach for decomposing the sources of growing (or declining) production efficiency. II. A CONVERGENCE AcCOUNTING FRAmEWORK As described in Section 1, changes in relative nominal factor returns between two or more sectors can originate from several sources. To see this formally for the general case, assume a twice-differentiable sectoral production function with scale economies of the general form: Q, = Atf(St, Xi), (1) where Q is real sectoral output, A is a measure of technical efficiency, S represents firm-level scale economies (e.g. average firm size, Q/N, where N is the number of firms in the sector),2 X is a vector of factor inputs, and t represents time. Equation (1) is homogeneous of degree one in the Xs,3 but the technology is constant-returns only if aQI&S = 0. Totally differentiating Equation (1) gives: dQt = dAtf( . ) + (aQIaS)4St + rj(aQ/aXI)^dXk,. (2) Dividing Equation (2) by Equation (1) and multiplying the second term on the right-hand side by S/S and the subsequent terms by X,/Xi yields: 0 0 0 q = a + j s + Eaj xj (3) where q, s and x represent rates of change in Q, S and X, aiga = 1, ai = (aQ/aXi)(XJIQ), B = (aQIaS)(S/Q)4 and ai and B are elasticities of output with respect to input i and scale. Total factor productivity growth (tfp) is represented by the sum of technical change (a) and scale effects (-Bs) in Equation (3). Equation (3) represents a relationship between physical inputs and prices. Now we add prices. The marginal revenue product of factor i can be represented as: MRPk = adRPqV0/(X-)d (4) 2. The scale variable could also be summarized by higher moments of Q/N. 3. That is, if firm scale (S) is unchanged, a uniform change in the Xs will result in a proportional change in output. For the Cobb-Douglas case used in this paper, the relationship between S and Q is shown in the Appendix. 4. In all fixed-weight calculations in this paper, the weights are averages of the initial and terminal values. 4 Sources of Gains in Allocaive Efficiency where P, and Pi are the prices of output and factor input i in period t. The coefficient a, is factor i's output elasticity, measured as a, = /(1-B).5 For the case in which the technology is constant returns to scale (i.e. B = 0), al = ui. Taking a logarithmic time derivative, Equation (4) can be transformed into a rate-of-change version as follows: 0 0 0 0 0 0 m rpi = a, + pq + q - P - xi (5) 0 where a, represents the growth, if any, of thie coefficient a,. Substituting Equation (3) into Equation (5) yields an expression for the rate of change of factox i's marginal revenue product in terms of its component changes: 0 0 0 0 0 0 0 m rp= + ( p -P.) + x + s s +a(x- x;). (6) The right-hand side of Equation (6) decomposes changes in factor i's marginal revenue product into tie folloning sources: (i) changes in the output price ( pq), (ii) changes in ie relevant input price, (pj),6 (iii) tectnicaI0change (ca), (iv) growth due scale economies (B s), (v) changes irfactor input ratios [aj( xj - x;)], and (vi) a shift in the own factor's output elasticity (ar). Equation (6) is the basic accounting relationship that can be used to decompose the sources of change in factor returns within a single sector. Within each sector there are i such relationships, corresponding to each factor. ][n order to investigate the sources of convergence or divergence for returns to factor i between two sectors, Equation (6) can be written for each of the sectors and appropriately differencecl to account for the sources of convergence or divergence. Designating the two sectors A and B and differencing yields the following terms: 00 0 0 5. This can be shown by substituting (q - n) for s in Equation (3) and solving for q, where n is the rate of growth in the number of firms. 6. Note that only changes in the own price of tlhe relevant fiLctcr affect its nominal return. Changes in the prices of other factor inputs have no direct effects on the marginal revenue product of the own factor. Sources of Gains in Alocative Efficiency 5 TABLE 1: SOURCES OF CONVERGENCE/DIVERGENCE 0 0 1. relative growth of marginal revenue products m rpi - m rp of which: 2. differential total factor productivity growth t' - t %pB of which: a. technical progress - aB 0 0 b. scale economies (B sA) - (B SB) 0 0 0 3. differential rates factor deepening a4( x - °x)A XB( ) 4. change in own factor output elasticity - B o_ o o 5. relative price change (PA- °P)-(P - Mp) of which: a. product price 0p _ b. input price 0 0p In Table 1 above, Item (1) measures the differential rate of growth of factor returns between sectors A and B. Items (2) through (5) represent a decomposition of the sources of differential rates of growth of each factor's marginal revenue product. III. WORLD OIL PRICE SHOCK During 1981, following the oil price shocks of 1973-74 and 1980-81, the price of goods producted in the petroleum products industry rose seven fold relative to 1972. Relative to the rise in overall manufacturing prices during the same period, prtroleum product prices rose three and a half fold. We use these two sectors -- petroleum production and total manufacturing - to compare changes in the relative returns to labor from 1972, prior to the price shocks, 1981, the year in which oil prices peaked, and 1986, the year in which relative returns to labor in 1972 were restored. 6 Sources of Gai,s in Alocative Efficiency Calculations of nominal retums to labor are obtained using the following expressions for labor's marginal revenue product (MRP): MRPLt = ceLt(Q/L)t, where a,,, labor's output elasticity, is computed as a three-year moving average of labor's income share. Using 1972 as a base year, we see that the marginal revenue product of an hour of labor input in the two sectors changed as follows: TABLE 2: INDEX OF RELATVE: MARGINAL PRoDucT OF LABOR IN PETROLEum PRoDuCoIN AND TOTAL MANuFAcTuRING Petroleum Total Mfg 1972 100.0 100.0 1981 3100.0 225.2 1986 268.2 268.5 The table shows that in 1981, compared with pre-oil price shock relative returns, returns to labor in petroleum production rose by 33 percent. Later, in 1986, returns to labor in the two sectors returned to the relative levelLs that had existed in 1972. This pattern of change would be sirnple to explain if the relative price of petroleum had risen and then fallen to its 1972 ratio, but the story is not that simple. Table 3 decomposes the adjustment of returns to labor into its key components, of which change in relative product prices is but one component. We can see from Table 3, that following the rise and fall of petroleum prices, in 1986, the price of petroleum products remained 53 percent above their relative level in 1972. Had other changes not occurred, returns to labor in petroleum products in 1986 would have exceeded those in total manufacturing by more than one-half. One source of convergence during 1972-86 was the slow growth of total factor productivity in the petroleum products industry relative to total manufacturing.7 From Table 3, we can also see that differential rates of factor deepening affected the relative returns to labor. Overall, the effect was to elevate further returns to labor in petroleum products relative to total manufacturing. The major change, however, compensating for the relative rise in the price of petroleum products was change in the respective production 7. The weights used to calculate total factor productivity are thre-year running averages with 1970-72 representing the 1972 set of weights, 1979-81 representing the 1981 set of weights and 1984-86 representing the 1986 set of weights. Sources of Gains in Allocative Efficiency 7 technolgies of the two sectors. Both sectors experienced a decline in labor's output elasticity, although the decline in labor's share in petroleum products from 0.101 in 1972 to 0.056 in 1986 in proportional terms was greater than the 0.461 to 0.397 decline in total manufacturing. In both sectors, the decline in labor's share was reflected in larger shares for materials and energy.8 IV. CmNA's INDUSTRIAL REFORMS Within China's industrial economy, two sectors -- the state- and collective-owned sectors -- represent 90 percent of gross industrial output, approximately 60 percent of which originates from the state sector with the balance coming from the collective sector. Important institutional and technological character-istics distinguish these two sectors. Production within the state sector is dominated by large and medium-size factories whose daily operations are extensively influenced by state regulation and government directives, although at the margin (i.e. outside the plan), market forces have increasingly entered into decision making. Contrasted with state-owned enterprise, many collective units, particularly those within rural areas, operate largely outside the orbit of state production quotas and material input allocations and typically enjoy greater autonomy than state-owned firms. In their study of the impact of reform on Chinese industry, Jefferson, Rawski and Zheng (1992) estimate the marginal revenue products of capital, labor and materials within the state and collective sectors. These marginal revenue products, shown in Table 4, display a convergence of returns to labor, capital and materials. During 1980-88, returns to labor in the collective sector rose from 48.5 percent to 55.4 percent of returns to labor in the state sector. Returns to capital within the two sectors, show a similar tendency to converge, as returns to capital within the collective sector declined from 168.7 to 153.7 of those for the state sector. During the reform period, returns to materials within the two sectors, already similar in 1980, converged from 1.05 to 0.99, as returns to materials in the state sector declined relative to the collective sector.9 Row (1) in Table 5 reports differences between the annual rates of growth of factor returns (mrp) between the state and collective sectors. These results show that returns to labor within the collective sector grew more rapidly than those within the state sector, while returns to capital within the state sector rose more rapidly than within the collective sector. These measurements are consistent with the observed tendency of the retums to labor and capital to 8. In both sector's capital's share also fell, while that for purchased services rose for total manufacturing and fell for petroleum products. 9. Because material inputs are purchased at administered, negotiated and market prices, the concept of "marginal returnm is particularly ambiguous for the case of material inputs. However, due to the principle of that enterprises and sectors that face relatively high material input prices also enjoy relatively high output prices ("gaojin-gaochu"), the average measure of Q/M used to calculate the marginal revenue product of materials (MRPM) in Table 5 may be a suitable approximation for the average productivity of material inputs purchased both within the plan and outside the plan. 8 Sources of Gains in Allocative Efficiency converge between the state and collective sectors shown in Table 4. Returns to materials show a slight decline within both sectors. By 1988, the more rapid rate of decline within the state sector had led to a virtual equality of material's intersectoral marginal revenue products. The contributions of the each of the determinants to the convergence of individual factor returns is examined below: a. Labor: In 1980, owing in part to the capital intensity of the state sector, returns to labor within the state sector substantially exceed that of the collective sector. Table 7 shows that differential rates of productivity change [Row (2)] and changes in relative factor intensities [Row (3)] contributed to the convergence of returns to labor between the state and colLective sectors. Specifically, higher TFP growth in the collective sector, combined with more rapid increases in the materials-labor ratio and the capital-labor ratio, motivated a more rapid increase in labor's physical marginal product in the collective sector relative to the state sector. By contrast changes in production technologies [Row (4)] and relative prices [Row (5)] retarded the convergence of labor's marginal revenue product. The decline in labor's output elasticity in the collective sector and the relatively rapid growth of product prices in the state sector considerably diminished the degree of convergence relative to what it would have been in the absence of these factors. b. Capital: At the beginning of the reforms, the return to capital within the collective sector exceeded that of the state sector. While retarding the convergence of labor's MRP, changes in the production technology of the collective sector and relative product prices between the two sectors were responsible for the decline of returns to capital within the collective sector relative to the state sector. At the same time, the higher rate of productivity growth within the collective sector tended to retard the convergence of returns to capital. The relatively modest diffeiences between the two sectors of labor and materials deepening were virtually mutually offsetting. c. Materials: As shown in Table 5, in :L980, by a modest margin, material's MRP within the state sector exceeded that of the collective sector. During 1980-88, the more rapid rise in collective sector productivity contributed to rnore equal returns as did an increase in the output elasticity of materials within the collective sector. Simultaneously, changes in relative factor intensities, particularly the relatively rapid decline in the labor-materials ratio in collective sector, slowed the rate of convergence between the two sectors, as did the tendency for product prices to rise more rapidly with the state sector and material input prices to rise more rapidly within the collective sector. These results reveal that while returns to labor, capital and materials showed a modest tendency to converge during 1980-88, the underlying factors that motive changes in the dispersion of factor returns changecl significamtly. With respect to the overall pattern of factor convergence, no single factor can be singled out as having been a key contributor or deterrent to convergence. This is not surprising since initially, returns to labor and materials were higher Oin the state sector while the return to capital was higher in the collective sector. Because, Sources of Gains in Allocative Efficiency 9 changes in relative productivity levels and relative product prices will uniformly affect relative returns to all factors, these changes cannot consistently motivate convergence or divergence of returns to all three factors. With the exception of labor, technological shifts contributed to convergence. Reallocations of labor and capital served to reduce large differences in returns to labor and capital, but, in doing so, retarded the convergence of returns to materials. The rapid growth of material inputs to the collective sector provided a large impetus to closing the gap between returns to labor while somewhat retarding the rate of convergence of returns to capital and materials. VI. CONCLUSIONS Shocks arising from unforeseen raw material shortages, product and process innovations, or regulatory changes in market economies give rise to processes of adjustment in which factor returns tend to converge through both price and quantity adjustments. Following the expansion of markets and incentives as part of a system reform program, gross disparities in factor returns should diminish thereby leading to greater allocative efficiency. The accounting framework developed in this paper develops an analytical framework that allows for identifying the sources of factor convergence and measuring their relative magnitudes. The example of the oil price shocks shows that price shocks lead not only to subsequent price adjustments, perhaps through changes in supply and demand, but also motivate changes in relative rates of productivity change, differential rates of factor deepening, and changes in production technologies. Each of these changes affects relative factor returns and overall productive efficiency. Likewise, in the transition of CPEs to market-oriented economies, we can see that price reform is only the first step in a subtle process of adjustment in which vast disparities in factor returns are reduced over time. Factor reallocations, differential rates of productivity change, and changing production technologies, as well as changes in relative product and factor prices, have all interacted to change absolute and relative factor returns in Chinese industry. Their cumulative impact has been to narrow differences in factor returns between China's state and collective industry. Application of the model to two very different situations -- adjustment to oil price shocks in the U.S. and adjustment to economic reform in China -- demonstrates the wide applicability of the analytical framework to a variety of changes in both market and quasi-market conditions. The framework is also applicable to other frequently investigated dichotomous situations, including the traded and non-traded sectors, rural-urban differences, agriculture and industrial disparities, and a wide range of other similar two-sector analyses. 10 Souirces of Gains in Allocative Efficiency TABLE 3: U.S. MANUFACTuRING: THE OIL SHOCK LABOR'S MARGINAL REVENE PRODUCr (% CHAN(GE 1972-1986) K L E M P (1) mnrp(P) - mrp(T) -32.5 % -0.1% -12.9% 1.1% -10.2% (2) differential TFP growth -15.4% (3) differential rates of factor deepening 1.4% 4.0% 8.6% 2.8% 27.8% of which: (a) capital deepening -0.6 (b) energy deepening -0.1 (c) material deepening 8.9 (d) services deepening -4.3 (4) change in own factor output elasticity 59.0% (5) relative price change 53.3 % of which: (a) product price (+) 53.3 (b) input price (-) 0.00 Sources of Gains in Allocauive Efficiency 11 TABLE 4: NOMINAL MARGINAL REVENU PRODUCTS: COMPARISON OF CEHNA'S STATE AND COLLECTIVE INDUSTRY A. State: MRPL MRPK MRPM level index level index level index 1980 1520.75 100.00 0.403 100.00 1.220 100.00 1988 2974.24 195.58 0.391 97.02 1.097 89.92 growth rate (%) 8.38 -0.38 -1.33 B. Collective: MRPL MP-PK MRPM level index* level index* level index* 1980 737.29 48.48 0.680 168.73 1.164 95.41 1988 1648.60 55.43 0.601 153.71 1.105 100.73 growth rate (%) 10.06 -1.55 -0.65 * Relative magnitude (comparable state sector figure for 1980 = 100.0). 12 Sources of Gains in AUocative Efficiency TABLE 5: SOURCES OF CONVERGENCE OF FACTOR RETURNS (ANNUAI, RATES OF GROWTH) L(a) K(c) M(c) 0 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ (1) InrpSoE mmocO -1.67% 1.17% 0.68% (2) differential TFP growth -2.23 % -2.23% -2.23% of which: (a) technical change -2.14 -2.14 -2.14 (b) scale economics -0.09 -0.09 -0.09 (3) dlifferential rates of factor deepening -4.75 -0.17 1.02 of which: (a) labor deepening n.a. 0.68 0.85 (b) capital deepening -0.51 n.a. 0.17 (c) material deepening -4.24 -0.85 n.a. (4) change in own factor output elasticity 3.76 2.60 -1.26 (5) relative price changes 1.64 1.08 1.88 of which: (a) product price (+) 1.64 1.64 1.64 (b) input price (-) 0.00 0.56 0.24 Sources of Gains in Allocative Efficiency 13 REFERENCES Desai, P. and Martin, R., "Efficiency Loss from Resource Misallocation in Soviet Industry," Ouarterly Journal of Economics. 98: 441-456, 1983. Fei, J. and Ranis, G., Development of the Labor Surplus Economy, Richmond and Irwin, Homewood, Ill., 1964. Harris, J.R. and Todaro, M.P., "Migration, Unemployment and Development: A Two-Sector Analysis," American Economic Review. 60:126-42, March 1970. Jefferson, G.H., "China's Iron and Steel Industry: Sources of Enterprise Efficiency and the Impact of Reform," Journal of Development Economics (forthcoming). , Rawski, T.G. and Zheng, Y.X., "Growth, Efficiency and Convergence in Chinese Industry: A Comparative Evaluation of the State and Collective Sectors," Economics Development and Culture Change. forthcoming. Jorgenson, Dale W., 'Testing Alternative Theories of the Development of a Dual Economy" in Theory and Design of Development. I. Adelman and E. Thorbecke, eds., Baltimore and London: Johns Hopkins University Press, 1966. Kuznets, Simon, "Economic Growth and Income Inequality," American Economic Review. 45,1, March 1955. Leibenstein, Harvey, Beyond Economic Man, Harvard University Press, Cambridge, 1976. Lewis, W. Arthur, "Economic Development with Unlimited Supplies of Labor, " The Manchester School, 22:155, 1954. SSB [State Statistical Bureau], Zhongguo tongii nianjian 1981 [Statistical Yearbook of China 1981], Beijing, 1981. SSB [State Statistical Bureau], Zhongguo tongji nianjian 1989 1989[Statistical Yearbook of China 1989], Beijing, 1989. Syrquin, Moshe, 'Resource Reallocation and Productivity Growth" in Economic Structure and Performance: Essays in Honor of Hollis B. Chenery. M.Syrquin. L. Taylor, and L.E. Westphal. eds., Academic Press, Inc. New York, 1984.