当前位置:首页 >> >>

Rural-Urban Disparity and Sectoral Labor Allocation in China


Rural-Urban Disparity and Sectoral Labor Allocation in China Dennis Tao Yang and Hao Zhou

August 1996 Revised: September 1996

Abstract This paper examines China’s rural-urban segmentation and its causes in the context of economic reforms. Household survey and aggregate data indicate a V-shaped process in which the rural-urban consumption and income differentials decreased between 1978-85, but then have continually increased historically high levels. This sectoral division is consistent with production function estimates based on provincial data that reveal higher labor productivity in urban/state-owned industries than in rural industries and agriculture. To explain the V-shaped change, we argue that the precedent of successful rural reforms raised farmers’ relative earnings, but the remaining obstacles for an efficient sectoral allocation of labor have prevented China from eliminating dualism. The recent financial policies consisting of urban price subsidies and increased investment credits have also had influential distribution effects biased against the rural sector.

*Department of Economics, Duke University. The authors would like to thank Belton Fleisher, Dong Furen, John Allen McNaughton, Thomas Rawski, Peter Schran, and especially Sarah Cook for valuable comments and discussions.

Rural-Urban Disparity and Sectoral Labor Allocation in China Dennis Tao Yang and Hao Zhou Department of Economics Duke University

August 1996 Revised: September 1996

Prepared for presentation at the annual meetings of the Association for Asian Studies in Honolulu, Hawaii, April 1996 and an ASSA/CES joint session at New Orleans, 1997; under consideration for a symposium issue of Journal of Development Studies

*Department of Economics, Duke University. The authors would like to thank Dong Furen, John Allen McNaughton, Thomas Rawski, Peter Schran, and especially Sarah Cook for valuable comments and discussions.

Rural-Urban Disparity and Sectoral Labor Allocation in China Dennis Tao Yang and Hao Zhou Department of Economics Duke University

September 1996

A working paper submitted to the CCER-World Bank visiting program China Center for Economic Research, Peking University

*Department of Economics, Duke University. The authors would like to thank Dong Furen, John Allen McNaughton, Thomas Rawski, Peter Schran, and especially Sarah Cook for valuable comments and discussions.

Abstract This paper examines China’s rural-urban segmentation and its causes in the context of economic reforms. Household survey and aggregate data indicate a V-shaped process in which the rural-urban consumption and income differentials decreased between 1978-85, but then have continually increased to historically high levels. This sectoral division is consistent with production function estimates based on provincial data that reveal higher labor productivity in urban/state-owned industries than in rural industries and agriculture. To explain the V-shaped change, we argue that the precedent of successful rural reforms raised farmers’ relative earnings, but the remaining obstacles for an efficient sectoral allocation of labor have prevented China from eliminating dualism. The recent financial policies consisting of urban price subsidies and increased investment credits have also had influential distribution effects biased against the rural sector.

Rural-Urban Disparity and Sectoral Labor Allocation in China

I. Introduction The separation of rural and urban sectors has been a major feature of the Chinese economy which has been studied extensively by economists.1 Shortly after establishing the socialist regime in 1949, China started a development strategy that emphasized urban industries with capital intensive technology. Extracting agricultural surplus and retaining profits in industries were the key sources of capital accumulation. The centrally planned system, which had urban/state-owned enterprises and rural communes as its administrative foundations, was effective in ensuring such an industrialization process. Among the policy measures, restrictions on rural-to-urban migration made it possible to maintain low urban consumption and to increase industrial investments. Prior to the start of economic reforms in 1978, capital goods were excessively concentrated in the urban sector and a large fraction of the labor force was restrained from leaving agriculture.2 As a result, urban workers’ productivity and earnings exceeded that of their rural counterparts. As part of the reforms, a series of policies have been introduced to reduce the rural-urban division. Such policies include increases in procurement prices for agricultural products, the adoption of household responsibility systems, liberalizing local markets, the relaxation of restrictions on labor mobility to cities, and capital investments in rural industries. There have been improvements in sectoral factor markets and rapid increases in farmers’ earnings. One remarkable success is the development of rural industries which have

For instance, Rawski (1979), Lardy (1983), Perkins and Yusuf (1984), Nolan and White (1984), and Riskin (1987) have documented the origin and the features of the Chinese dualistic economic structure. Johnson (1990) and Putterman (1992) discussed the sectoral economic development in light of recent economic reforms. In 1978, the urban sector employed 95 million workers while the rural sector had approximately 306.4 million labor force (see Table 6). In contrast, the total value of fixed assets in the state-owned enterprises (primarily urban) counted for 448.82 billion yuan while the value of the fixed assets in agriculture was only about 94.98 billion yuan (SSB 1993; Perkins and Yusuf 1984). These numbers indicate a ratio of 3.2:1 in labor and 1:4.7 in capital between the rural and urban sectors. 1
2

1

become a powerful source of contribution to China’s economic growth.3 Because of these improvements in the rural regions, economists and other China observers have speculated that the rural-urban gaps have begun to narrow (Oi 1993; Zhang et al. 1994a).4,5 This paper systematically examines China’s rural-urban disparity from several angles using different types of data. First, we use national average per capita information to compare consumption levels of rural and urban residents. Next, we use comprehensive household survey data released by the State Statistical Bureau of China to examine differences in family disposable incomes. Contrary to the belief in a persistent, narrowing rural-urban gap since the reforms, the data indicate a V-shaped process in which the consumption and income differentials decreased between 1978-85, but then increased to historically high levels in the early 1990s. Dualism did still prevail in China after one and a half decades of economic reforms. Using 1987-92 Chinese provincial data, we also conduct production function analyses to examine labor productivity differences in state enterprises, rural industries, and the staple agriculture. The results reveal substantially higher productivity in urban/state-owned industries than in the rural sectors, confirming the sectoral division found in consumption and income comparisons. We then proceed to investigate sectoral labor allocation as a determinant of disparity. We argue that the high concentration of capital in cities and labor in the countryside, a result of the heavy-industry oriented development strategy, was the basis of sectoral insulation. Furthermore, labor mobility restrictions were

Rural industries consist of township and village level enterprises (TVE) and private enterprises. Since 1984, the gross value of rural industries has increased by an average of 20 percent per year, and in 1994, TVEs produced about 42% of the nation’s total industrial product and employed 120.18 million workers (SSB 1995). Exceptions include Johnson (1995) and Knight and Song (1995) who emphasized the continuing sectoral gaps in consumption and incomes. Researchers have also studied other aspects of disparities. For references on provincial distribution of consumption and production, see Lyons (1991) and Tsui (1991); for convergence of factor productivity across state and collective industries, see Jefferson et al. (1992); for changes in the distribution of household incomes, see Khan et al. (1993); and for convergence of per capita production across China’s provinces, see Chen and Fleisher (1996). This paper focuses on rural-urban disparities. 2
5 4

3

instruments in achieving the development strategy in the centrally planned regime. Consequently, the relaxation of rural labor mobility restrictions since the reforms and the precedent of successful rural transformations did raise farmers’ relative earnings, reducing disparity between 1978-85. However, there are remaining

institutional obstacles, both in urban and rural sectors, that still prevent an efficient labor allocation and therefore the elimination of dualism. In addition, the contractionary policies in the late 1980s and the recent financial policies consisting of urban price subsidies and increased investment credits have also had influential distribution effects biased against the rural sector. These new urban-biased policies were powerful enough to cause an upward movement in the V-shaped trend since 1985.

II. Differences in Consumption and Income An analysis of consumption and income provides one kind of evidence on the rural-urban linkages of an economy. In the classical models of dualistic economic development, wages of workers in a modern sector are higher than the workers earnings in a traditional sector.6 The development process involves continuous labor transfers from the low to the high income sector, and dualism will end when factors of production are rewarded with competitive prices. For labor, comparable earnings and consumption in the two sectors are often used as indicators for rural-urban integration. If severe gaps exist, they indicate sectoral segmentation. However, one should be cautious in comparing rural-urban consumption and incomes. First, labor quality, including schooling, training, and experience, has to be adjusted when considering earnings in alternative sectors. Second, the comparison should be made in real, not monetary, terms. Third, any difference in the cost of living between urban and rural areas should be taken into account. Furthermore, the comparison

See especially Lewis (1954) and Ranis and Fei (1961) for the basic analytical framework of dualistic development. Although Lewis did not define the traditional (subsistence) and modern (capitalist) sectors as specific occupations, the former is often identified empirically with agriculture and the latter with industries. According to Lewis, the industrial wage could be significantly higher than compensations in the subsistence sector because of government regulations, a situation which mirrors the Chinese experience. 3

6

should also reflect the provision of subsidized public services, such as health and housing. In practice, it is difficult to adjust for all these factors and we have to rely on the available information. In what follows, we will first compare rural-urban per capita consumption expenditures. Then, we will utilize newly published information by the State Statistical Bureau to examine sectoral income differences and contrast them with the trends in consumption expenditures. These analyses will cover an extended time period so that, if the above biases are systematic and time invariant, then the assessment will reflect relative changes before and after the reforms. Table 1 presents nominal per capita consumption expenditures between 1952 and 1992. The figures, taken from the 1993 Statistical Yearbook of China, include the national average, separate accounts of rural and urban levels, and the ratio of sectoral expenditures. It is helpful to analyze the data according to four historical periods: (1) 1952-57 were the years of socialist restructuring; (2) 1958-60, the Great Leap Forward (GLF) movement; (3) 1961-1977, the period of economic stagnation and the Cultural Revolution; and (4) 1978-1992 covered the era of economic reforms. From 1952-57, the urban-rural consumption ratios were high, ranging from 2.4 to 2.6. During the Great Leap Forward, China suffered from a major decline in food supply and the protective government policies toward cities enlarged the urban-rural consumption differences to the highest levels of 3.1 in 1959 and 3.2 in 1960.7 In the subsequent years of economic stagnation, the consumption ratios ranged between 2.3 and 2.9, revealing an upward trend in the late 1970s. These figures show a significant gap between rural and urban consumption levels prior to the reforms. The emphasis on heavy industry development and the restrictions on rural-to-urban migration were largely responsible for these consequences.8

Lin and Yang (1995) indicate that the food entitlement to urban residents was a main reason for the great famine during GLF that resulted in 30 millions of excess deaths, primarily in rural areas. During the period of national emergency, urban bias more severely reduced the welfare of the farm population. See, for instance, Perkins and Yusuf (1984) on the rural-urban connections in light of government expenditures and revenues between 1949-78. Deliberate policies of restricting urbanization was an important 4
8

7

Have economic reforms reduced the sectoral consumption differences? The data reveal a V-shaped trend in which the urban-rural ratio declined in the beginning years of the reforms, reaching its lowest level of 2.2 in 1985, but since had steadily increased to 3.1 in 1992, comparable with the highest levels during the Great Leap Forward. These figures reject the notion of narrowing rural-urban disparity in the post-reform era. There are at least two difficulties of comparing rural-urban consumption expenditures in China. The first is the lack of regional commodity prices, especially for self-retained products by farm households. This problem of relative prices may cause systematic bias in evaluating the sectoral consumption ratios. However, if the regional price biases are time invariant, the above conclusions on the changes of rural-urban disparity will not be affected. The second concern is that the use of “rural/urban” dichotomy has been changing with time (Chan and Xu, 1985; Goldstein, 1990). In 1984 particularly, the central government relaxed the criteria for population size and the proportion of nonagricultural residents to define a town, and consequently, the number of townships increased fourfold. At the same time, a large number of households were reclassified as urban. Such changes in household status would affect the aggregate rural-urban consumption ratios, but the direction was not clear because the reclassification likely included households with above-average rural status, but who were comparatively poor among urban households. The result was a reduction in the average consumption expenditures for both urban and rural families. From a different angle, such a massive reclassification could only cause a level shift in rural-urban ratios, but not changes in the ratios over time. In short, these two concerns do not alter the V-shaped observation. The analysis of consumption expenditures does not reflect real purchasing power, however, because savings are omitted. For this reason, incomes are usually better indicators of rural-urban disparity. To calculate personal incomes in China, it is necessary to be aware of the institutions that determine the sources of earnings. In cities, wages only represent a fraction of the total income, which also includes welfare

cause of rural poverty. 5

provisions such as housing, health services, in-kind-transfers, and price subsidies. In contrast, the earnings of rural households are primarily value-added from self-employed activities, including agriculture.9 In a recent study, researchers at the State Statistical Bureau (SSB) of China (Zhang et al. 1994b) made systematic comparisons of personal disposable incomes of rural and urban residents based on household survey information. Their study was specially designed to deal with institutional ambiguities in estimating

comparable, full personal incomes. Table 2 presents per capita disposable income (PDI) of urban residents between 1980-92. The SSB defined the urban PDI with two components, pecuniary and non-pecuniary earnings, where the former was the wage income and the latter consisted of housing subsidies, medical services, in-kind compensations, and price subsidies. The figures show that non-pecuniary earnings counted for a large share of urban disposable incomes of 30.8 percent in 1980, and despite a steady decline, it still counted for about 19 percent of total incomes in 1992. These non-pecuniary incomes are given by specific institutions to their employees and therefore are not available to rural migrants into cities. Later, we will discuss these urban welfare provisions as barriers to permanent rural-to-urban migration. Table 3 reports price deflated urban and rural PDIs and their ratios. These figures confirm a V-shaped trend from previous consumption analyses in which the urban-rural ratio declined in the first half of the 1980s, reaching the lowest level of 2.26 in 1985, but since then, the ratio has steadily increased to 3.05 in 1992. To visualize the changes in rural-urban disparity, we have plotted the consumption ratios from Table 1 and income ratios from Table 3 in Figure 1. Two conclusions become immediately apparent: (1) urban residents have had much higher levels of consumption and incomes in contemporary China; and (2) the recent economic reforms have not reduced the structural division.

More specifically, their earnings equal the gross value of total produce minus operation costs of various productions, the depreciation of productive assets, taxes, the value of procurement quota, and other production expenses. 6

9

In most countries, especially those in their early stages of industrial development, the welfare of urban residents exceeds their rural counterparts. To understand the relative magnitude of the gaps in China, Table A1 presents urban-rural income ratios for 36 countries. The ratios for the majority of the countries were below 1.5. In 1985, there were only four countries in which the average urban earnings were more than two times the rural earnings. There were five countries in 1990 and three countries in 1995 which had ratios of 2 or more. Although caution must be given to cross-country comparisons, the rural-urban division in China is very serious indeed. Comparisons of income and consumption levels are indicative of the economic welfare of rural and urban residents. But perhaps a more important implication is on the efficiency of resource allocation. If people in certain regions earn much more than comparable labor in other regions, inefficient allocation of labor exists. And, for that economy, there is potential to raise total output through the reallocation of the labor force. In what follows, we complement the above consumption and income analysis with an investigation of sectoral labor productivity.

III. Sectoral Labor Productivity The efficient allocation of the labor force refers to a situation where workers are employed in places of their highest productivity. When workers are compensated by their marginal value product, differentials in sectoral income and consumption for comparable labor would reflect the inefficiency of labor allocation. However, if wages are not determined through competitive mechanisms, income differences can no longer serve as good indicators of efficiency. Direct measurements of labor productivity are necessary. In China, there is a long history of wage intervention by the government (Zax 1994). Even today, wages of state enterprise employees often consist of five sources: basic compensation, seniority, schooling or training, administrative rank, and bonus. In many situations, the amount of compensation to workers’ attributes are arbitrarily set, so that earnings may not reflect productivity. Due to wage setting practices in 7

China, labor productivity estimates can provide direct information on the sectoral misallocation of labor. Previous researchers have studied sectoral productivity differences in China. For instance, Jefferson et al. (1992) estimated the marginal productivity of labor in state-owned and collective enterprises for the period 1980-88. They found that labor productivity was much higher in the state sector. However, the collective enterprises they defined included urban collectives and township-village enterprises (TVEs), and therefore the estimates were not rural-urban comparisons. Nolan and White (1984) calculated average output per worker in agriculture and state industries for the period 1952-1981, concluding that per worker output was substantially higher in these industries. Their study, however, did not estimate marginal productivity. In what follows, we utilize the 1987-92 Chinese provincial data to analyze labor productivity in state industries, rural industries, and agriculture.10 We treat rural industries as a sector because they have grown rapidly in recent years. Before the start of economic reforms in 1978, China had a dual economy which consisted of state-owned, capital-intensive industries in cities and collectivized, labor-intensive agriculture in rural areas. The rural decollectivization gave the freedom of occupational choice to the households, and farm labor started to move into industries that had higher labor productivity. In the mid-1980s, there were still restrictions on labor movements into large cities, but the control over migration into small towns was loosened. At the same time, farmers were allowed to invest capital in non-agricultural activities. Such a policy environment resulted in the rapid growth of rural industries. To compare sectoral marginal productivity of labor, we estimate a Cobb-Douglas production function for each of the three sectors. The basic form of the empirical function is

The decision to utilize this time period primarily reflects data availability and consistency. The State Statistical Bureau of China has released input-output data for all three sectors since 1986, but starting in 1993, the statistical yearbooks have changed the reports of several economic variables for the rural enterprises, such as replacing gross sales information with value-added measures. Therefore, we use data between 1987-92 for statistical analyses. Throughout the paper, we examine China’s economic situations up to 1992 for the sake of data consistency. 8

10

yi
c ili iki imi ie

 




tDt  Ji

,

(1)

where I indexes the sectors, y is a measure of output value, c is a neutral displacement parameter, (l,k,m) are sector-specific production inputs, D represents year dummies, and  is a sector-specific stochastic term. Taking natural logarithms of this function, the equation for estimation is lnyi
lnc iilnliilnki ilnmi

t
88

M D  .
92 t t i

(2)

The data used for estimation are obtained exclusively from the Statistical Yearbooks of China (SSB 1988-93), thereby ensuring consistency and comparability of computational results. For each sector, we use y to measure the gross output value. For the state and rural industries, we use total labor force, net value of fixed capital, and average value of circulating capital to measure labor input (l), capital input (k), and intermediate input (m). For agriculture, we use total labor force, total sowing area, total number of small tractors, and disaster areas to measure labor inputs, land inputs, mechanization, and weather adversity. The panel data include information of 30 provinces for six years.11 The gross output value, net value of fixed capital, and average value of circulating capital are deflated to the base year 1987. We use the industrial output price index to deflate the output of state industries, the index of rural industrial output price to deflate the output of rural industries, and the index of agricultural produce price to deflate agricultural output. For capital values, the deflating index is the industrial capital goods price. All price indices are taken from SSB (1988-93). Table A2 provides average provincial information for all the main variables in the regressions. It shows that rural industries had an annual growth rate of 17.6 percent while state industries grew at 6.1 percent

11

Except that there were 29 provinces in 1987 because Hainan province was established in 1988. 9

and agriculture grew at 6.8 percent. Clearly rural industries were the major sources of economic growth in that period. The growth rates of labor inputs were about the same in rural industries, state industries, and agriculture, at 2.0%, 1.8%, and 2.0%, respectively. The average productivity of labor (APL) in state industry rose slightly from 20 thousand yuan/person in 1987 to 24 thousand yuan/person in 1991. In rural industries, APL rose sharply from 6 to 13 thousand yuan/person, and in agriculture, APL remained almost the same, around 2 thousand yuan/person. Systematic differences in average productivity of labor existed among the three industries. The OLS estimates of (2) are presented in Table 4. All coefficients, except for the number of small tractors in agriculture, are statistically significant with reasonable signs and magnitudes. For state industries, the intermediate inputs constitute the largest share of output value, which reflects the fact that state industries are concentrated on producing production goods which heavily utilize intermediate inputs. Note that the share of capital (0.20) is less than the share of labor (0.382) in state industries with a high capital-labor ratio. A possible explanation for this is that the efficiency of capital utilization is still low despite years of reforms. For rural industries, we note that the share of intermediate inputs (0.282) is relatively low, which is consistent with the fact that rural industries primarily produce consumption goods. In addition, the capital share (0.736) is high, reflecting the fast growth of capital stocks and their efficient utilization. The share of labor (0.094) is low, consistent with the fact that the competitive wage rates in rural areas are lower than the protected wage rates in state industries. For agriculture, disaster areas have a negative effect on output. We note that the share for sow area (0.752) is significantly higher than the contribution of labor (0.314). From the production estimates, we can compute the economy of scale for state industries, rural industries, and agriculture. They are 1.123, 1.112, and 0.849 respectively, which are all significantly different from one12.

The estimated scale parameters for state industries and rural industries are both larger than one, consistent with the results of Jefferson et al. (1992), although the magnitudes of their estimates were slightly smaller (1.07 and 1.04 respectively). The economy of scale in agriculture is consistent with micro production studies, such as Putterman and Chiacu (1994) and Yang (1995). 10

12

Taking the partial derivative of (1) with respect to the labor inputs, an expression for sector-specific marginal productivity of labor can be expressed as

MPL i
c i

ili i
li



ki im i



i




iyi
li

.

(3)

The marginal productivity of labor around the sample means may be obtained by substituting the estimated labor coefficients i and the sample means of output and labor inputs in this equation. Table 5 reports the estimated marginal productivity of labor for the three sectors. Although data limitations made it impossible to examine the productivity changes in the entire post-reform period, the estimates for 1987-92 can still be compared with the previous consumption and income analyses. Several noticeable features have emerged. First, the labor productivity in urban/state industries is substantially higher than in agriculture and rural industries. The long practice of China’s development strategy that emphasized a capital-intensive state industry and restrictions on rural-urban migration is the primary cause of such an imbalance. The recent reforms have not yet resulted in efficient resource allocation. Second, labor productivity has been increasing in all three sectors. And third, in absolute terms, the productivity gap between the urban (state industries) sector and the rural (agriculture and rural industries) sector has also been increasing at a moderate pace. This result is consistent with the increasing rural-urban disparity in consumption and incomes. While caution must be taken in interpreting these aggregate estimates, they nevertheless provide references on the fundamental forces that influence sectoral labor flows. The larger the differences in productivity levels, the stronger the “pull” factors are to lure workers to migrate and change employment. In fact, since the reforms in which the government has started to permit worker mobility, farmers have gradually found jobs in rural industries and rural people have moved into cities for temporary employment. In regard to the above consumption, income, and productivity analyses, important economic questions remain. What are the key factors that have segmented the rural-urban sectors? What reform measures have contributed to the 11

reduction in disparity in the early period of reforms? Why has inequality been increasing in recent years? These are the questions to which we now turn.

IV. Labor Mobility Restrictions From the viewpoint of neoclassical economics, differences in the marginal productivity of comparable labor between two places induce labor flow to the higher productivity place. Such a movement tends to reduce the productivity differences and to equilibrate the spacial labor market. These tendencies remain when there is cost of migration. The driving forces are the motives of individual utility maximization and the profit maximization by firms. Similarly, capital investments are also guided to places of the highest returns. We would like to argue that a series of policies implemented during the centrally planned regime (194978) have insulated China’s sectoral factor markets. More specifically, we examine labor mobility restrictions that have trackable records and observable impacts on the patterns of allocation.13 The basic argument is that, during the old regime, the combination of capital accumulation in cities and restrictions on rural-to-urban migration resulted in high capital-labor ratios in cities and low ratios in rural areas. This policy mix was the root of sectoral disparity because input proportions determine factor returns. The recent economic reforms have gradually loosened mobility restrictions and introduced sectoral market linkages. As a result, rural people have found temporary employment in cities and rural industries have grown rapidly, reducing the gaps in consumption and incomes. However, many institutions established in the pre-reform era have remained as obstacles for rural-urban integration. This section elaborates these arguments based on the observed patterns of labor allocation.

Sectoral capital investments are also crucial in determining the extent of rural-urban disparity. However, capital formation is closely related to the structure of interest rates, price setting practices, accounting systems, and the measurements of capital are less directly comparable than labor across sectors. In this paper we concentrate on labor allocation. 12

13

Policies in the Pre-Reform Era (1949-78) In this section we provide a brief overview of policies towards labor in this period.14 We emphasize that restrictions on rural-urban migration were policy tools to implement the heavy-industry-oriented development strategy, which centered on capital formation in industries at the expense of agriculture. The effective implementation of the plan caused an unbalanced allocation of productive factors. The main mechanisms of controlling labor migration in China were created between 1949-57. By 1953, the government had established the household registration (hukou) system that clearly distinguished the agricultural and non-agricultural status. In the same year, the State brought all grain procurement and distribution under its direct control with a system of Unified Procurement and Unified Sale for grain and oilbearing crops as a way to suppress the agricultural procurement prices (Sicular 1988). To control urban food demand and to facilitate food distribution, the government also established a food rationing system for urban residents in 1956 that specified ration standards according to individuals’ age, employment, and other demographic characteristics (Chen 1982). The rationed commodities included grain, soy beans, cotton cloth, edible oil, pork, and other necessities (Walker 1984). These policy schemes formed the basis of labor mobility restrictions in China. Although the government established labor allocation mechanisms, its control on sectoral mobility was rather loose in this initial period of socialism. In fact, there was large demand for labor by the newly established urban enterprises which recruited rural workers under the permission of labor bureaus. As Table 6 shows, the urban labor force was nearly doubled between 1952-57, and in a slightly longer period (1949-57), the percentage of urban population increased from 10.6 to 15.4 percent (Chan and Xu 1984). The economic situations during the Great Leap Forward (GLF) (1958-61) were chaotic, marked with large swings in sectoral labor allocation. To support industrialization, about 41 million workers exited

14

See Chan and Xu (1984), Riskin (1987), and Chang (1994) for additional references. 13

agriculture between 1957 and 1958, which was a 21 percent decline (Riskin 1987). Among these workers, approximately 17 million worked in the iron, steel and other heavy industrial undertakings in the countryside, while close to 16 million migrated into cities working in state industrial enterprises.15 Because of reductions in agricultural labor, sudden institutional changes, natural calamities, and a series of policy mistakes, Chinese agricultural production collapsed (e.g. Lin 1990). The grain output dropped by 15 percent in 1959 and reached only about 70 percent of the 1958 level in 1960 and 1961. Ashton et al. (1984) revealed that this crisis resulted in widespread famine in rural areas and caused about 30 million excess deaths. To release the pressure of urban food demand and to increase labor inputs for agricultural production, the government sent 10 million workers back to their rural homes in 1961 and thus reversed the labor flow during the GLF (Bernstein 1984). At the same time, rural-to-urban migration was strictly prohibited. Restrictions on rural-urban migration remained effective during the period of Cultural Revolution (1966-76). In contrast to the urbanization experience of most countries, China engaged in a massive campaign of rustication movements (xiaxiang and xiafang) in which city youths and intellectuals were sent to live and work in the countryside. Between 1968 and 1975, a total of 12 million youth were persuaded and coerced to leave their urban homes (Bernstein 1977).17 In the opposite direction, it was difficult for rural residents to convert their status to urban status and migrate into cities. The few ways that a rural resident could convert household registration were through enrollments in colleges or universities (shangxue), civilian job recruitments (zhaogong), and army recruitments (canjun). The civilian job recruitments were the largest source of agricultural to non-agricultural conversions, but they were strictly controlled by labor bureaus because the state had to provide the individuals with food rationing, housing, medical care, fuel, and other urban privileges.

15

Ashton et al. (1984), Walker (1984), and Bernstein (1984) provided similar estimates.

Cadres and intellectuals were sent to “May Seventh Cadre Schools” in rural areas. Different from the rusticated youth, who expected to live permanently in the countryside, cadres were on programs of variable length and many of them returned to cities after the training. 14

17

Because of these institutional barriers, the government had effectively implemented its development strategies and achieved the corresponding sectoral allocation of labor. The moderate urban growth from 10.6 percent urbanization level in 1949 to 17.9 percent in 1978 reflects the effect of counter-urban growth policies. As footnote 2 explained, there were high concentrations of capital in the urban sector and labor in the rural sector. These unbalanced factor proportions dictated higher productivity for city workers, which was the basis for the sectoral disparity in consumption and incomes.

Deregulation and Mobility in the Post-Reform Era (1978-present) Due to these biased development strategies, the urban per capita income was 2.4 times the rural per capita income in 1978 and the urban-rural consumption ratio was 2.9.18 Because of these gaps, the government gradually adopted a series of deregulation policies that opened access to urban employment and opportunities in local rural industries. Farm households responded to these policy changes and the newly created employments significantly increased their relative earnings. However, serious barriers still exist to prevent an elimination of the rural-urban disparity. The adoption of the household responsibility system (HRS), between 1979-84, was the important first step to give households the freedom of allocating productive resources, including labor. Regulations governing occupational choice and internal migration were also gradually relaxed. In 1983, the government began to allow farmers to engage in long distance transport and marketing of their products in cities (CAY 1987). In 1984, the State started to encourage farmers to leave agricultural production and, where appropriate, to work in nearby small towns (FBIS 1984). A major policy reform took place in 1988 when the central government officially relaxed the controls over residential change. It was announced that farmers could move to cities if they could provide their own staples and were financially capable of running a business (Forbes and Linge

The income ratio is based on urban disposable incomes and rural net earnings from household survey data (SSB 1988, p799). 15

18

1990). This was a landmark deregulation which removed the legal restrictions on rural-to-urban migration. These policy changes allowed farm people to pursue better economic opportunities. First, the number of workers who temporarily worked in cities exploded. According to studies reported in BRLMS (1995), about 15.47 million rural people worked in cities between 1982-1987. In 1992, the daily population of temporary or “floating” rural workers increased to 35.75 million, and in 1993, increased to 38.66 million.19 Their average stay in cities was about 195 days (Shen and Tong 1992). At the national level, temporary employments have contributed greatly to the income of Chinese farm households. According to a recent study by Rural Research Group of Annual Analysis (1995), the total earnings of an estimated 41.4 million rural laborers from working outside of their familial localities were 151.08 billion yuan, or 18 percent of the total rural income in 1994.20

Second, nonagricultural activities within rural regions have absorbed more rural labor. Table 6 shows that, during the years of Cultural Revolution 1966-76, less than 5 percent of the rural workers were engaged in nonagricultural activities. The national labor force was mostly concentrated in urban industries and agriculture. The relaxation of controls on farm household activities resulted in a jump in the nonfarm employment to 31.5 million in 1979, or 10.2 percent of the rural labor force. Since then the nonfarm employment increased steadily into the early 1980s, grew faster since 1984, and reached 97.8 million in 1992. From 1977 to 1992 the number of workers in rural areas increased by 135.5 million, while the majority of the increase, 80.5 million, found employment in local industries, construction, transportation, commerce, and trade. In 1992, income from non-agricultural sources accounted for an average of 38.6 percent of farm household earnings (SSB 1995). Massive labor transfers to rural nonagricultural activities and urban employment have been the

The term, "floating population,” is commonly used in China, referring to those who do not have permanent household registration in the places they live or work.
20

19

For reference of household studies, see Cook (1996) and Hare (1996). 16

responses of rural people to better economic opportunities made available by these reforms. These sectoral reallocations of labor have raised the farm household earnings and helped to reduce the rural-urban disparity. However, to the present day, the economic environment for labor mobility is far from perfect. In what follows, we discuss briefly the barriers that still prevent the elimination of dualism.

Existing Institutional Barriers In most developing countries, permanent and family migration of rural residents into urban areas has played a central role in long-term economic growth. In contrast, the patterns of China’s rural labor mobility are rather unique because the massive floating population are primarily engaged in temporary jobs and rural industries have absorbed most workers. We would like to argue that these special patterns of labor allocation reflect the existing policy interventions of the government. In particular, the urban welfare systems and the rural land arrangements still influence the sectoral integration. Some constraints on labor mobility associated with the household registration system, such as the legal rights of residency and commodity rationing, are already abolished. But urban residents still receive welfare privileges largely unavailable to rural migrants (Yang 1996). State enterprises and other government agencies manage approximately 78 percent of urban housing that is exclusively allocated among urban employees (Cai 1991). In addition, government work units provide health insurance/services and pensions primarily to permanent workers.21 These special welfare provisions prevented the rural workers from migrating with other family members and forced them to have a short planning horizon. High costs in child care and education at elementary and middle school levels also hinder family migration into cities. Inherited from the centrally planned system, urban work units have continued to run child

Based on a survey of 1021 temporary rural workers conducted in the Zhujiang Delta in 1994 (BRLMS 1996), 82.3 percent of them lived in temporary dormitories, 15.8 percent rented housing, and 1.3 percent lived with relatives. Their employers provided health insurance to only 11.9 percent of the workers and pensions to 3.9 percent of them. 17

21

care centers and kindergartens that only admit the children of their own permanent employees. The operational expenses of elementary and middle schools come primarily from the education bureaus of urban districts, and in turn, the schools only admit students who have household registrations within the districts. It is not uncommon for schools to enroll students from elsewhere with admission fees and donations, but the charges are high. 22 Partly because of these potential expenses in child care and education, most rural temporary workers are males and they do not come with other family members (Shen and Tong 1992). Rural institutional arrangements also influence rural labor mobility decisions (Yang 1997), an area that has received less attention. Chinese farm families under the household responsibility system have the landuse rights but not the rights of alienation. If permanently leaving agriculture, they have to return the land to local authorities and consequently give up a stream of future land earnings. As a result, Chinese farmers seldom engage in family migration and have incentives to split familial labor supply to farm and nonfarm employment. This division of time is a second best solution under the existing land arrangements that takes advantage of higher nonagricultural wages and avoids the loss in land values. But this solution drives a wedge between urban and rural labor earnings, and some households have left their land uncultivated (BRLMS 1995). The development of land markets, particularly land rentals and sales, has at least two advantages. One is to increase the efficiency of land utilization, and the other is to make farmers financially prepared to migrate into cities.

V. Understanding the V-Shaped Change

Systematic reports on admission fees and donations are rare because schools do not want to release this information to the public. Based on personal experience in 1996, I learned that the donation made to send a student to a top elementary school from an adjacent district was 20,000 yuan in Beijing. For a high school student from Shanxi who tried to attend a school in Beijing, the admission fees ranged from 5,000 to 40,000 yuan among the few possible alternatives. These fees are very high because the annual per capita disposable income of Beijing residents was only about 5,084 yuan in 1994 (SSB 1995; figure unavailable for 1996).

22

18

The preceding section has illustrated that the heavy-industry oriented development strategy and the restrictions on rural-to-urban migration during the centrally planned regime have driven a wedge between sectoral consumption and incomes. The relaxation of institutional barriers since the reforms and the permission of more economic freedom to the farm people have resulted in massive labor transfers to rural industries and temporary urban employment. These new work opportunities have raised farm household incomes and have tended to reduce rural-urban disparity. However, there remain institutions created in the past which still prevent an efficient sectoral allocation of labor. Consistent with these changes, the increased labor mobility should effect a gradual decline in rural-urban gaps, but then the disparity will stay at a certain level, reflecting the remaining institutional barriers. In addition, to understand the V-shaped change, complementary analyses to labor mobility are needed. A straightforward explanation for the declining disparity between 1978-85 is that the rural transformation preceded the urban one, which formally started in 1985.23 As early as 1978 or 1979, major rural reforms started to take place, including substantial increases in the purchase prices of eighteen farm products by an average of 22.1 percent, the experiments of household responsibility systems (HRS), and the permission of village trade fairs.24 These reform measures had induced positive supply responses and created production incentives, resulting in a 9.2 percent annual growth in real agricultural gross output between 197884 (Lin 1992; McMillan 1989). During the same period, total industrial output grew at an annual rate of 5.5

Limited urban reforms had already started prior to 1984 (Johnson, 1990). For instance, the stateowned enterprises experimented with various financial systems, including profit-contracting and a schedule of four taxes, to replace the old profit retention program. A comprehensive reform package, that included reducing the role of government agencies, reforming the planning system, the adoption of a double-tier price system, the separation of government from enterprise functions, and the responsibility system to urban enterprises, was not formally introduced until 1984. More specifically, the price increases included 20 percent for grain, 25 percent for fats and oils, 15 percent for cotton, 26 percent for pigs, and 20-50 percent for fourteen other products (Johnson, 1990). And, due to the great success of HRS in the poorest areas, the system was supported by the government and adopted in the nation in a sweeping fashion. By the end of 1980, 14.4 percent of all households had adopted the system; by the end of 1981, 45.1 percent; by the end of 1982, 80.4 percent; and by 1984, about 99 percent (Lin, 1992). 19
24

23

percent in real terms. The faster growth in agriculture, combined with labor mobility effects, contributed to the reduction in income and consumption ratios. What factors have caused the increasing disparity since 1985? One possible explanation is the major urban reforms started in 1985 may have greatly improved the efficiency of state enterprises. In particular, the use of wage bonuses and the adoption of responsibility systems may have tapped the potentials of workers and managers and resulted in large output and wage increases. However, a careful comparison of the sectoral output growth between 1985-91 was inconsistent with this explanation. During this period, the real gross domestic product (GDP) of the rural sector increased at a remarkable annual rate of 13.7 percent while the real GDP of the urban sector grew at 7.6 percent (Wu 1994). Since real output growth came primarily from rural industries during this period, the catching-up of urban reforms may not explain the faster wage growth in the urban sector. We would like to argue that the increased rural-urban differential since 1985 has been caused primarily by the government’s financial transfer programs in favor of the urban sector. As Table 7 illustrates, the government investments in state-owned enterprises increased by an astonishing four-fold in nominal terms between 1984-92, reaching 301.265 billion yuan in 1992, which equaled 11.3 percent of China’s GNP. In the same period, the “price/inflation subsidies” to the urban residents were kept at high levels ranging from 21.834 to 32.149 billion yuan annually. In contrast, the state investments in the rural economy had much fluctuation and were much less as compared to the urban investments and price subsidies. In 1991, when rural investment reached its peak at 13.08 billion yuan, the amount was only 34.8 percent of the urban price subsidies in that year! According to Brandt and Zhu (1995), the state-owned enterprises had successfully used the increased investment credits to subsidize wages for their employees, a direct income transfer to urban residents.25

Brandt and Zhu developed a positive, general equilibrium model to explain the cyclical behavior of output growth and inflation in post-reform China. They used an assumption for the model that the government wants to equalize the benefits of the growth process between the state-sector and the more rapidly growing nonstate sector. See their study for more factual descriptions on output growth, investment credits, and wage 20

25

Between 1985-92, inflation became an concern in China (Table 7). Since the wages of rural workers were supported with output growth while the wages of urban workers came in part from money creation, there had been consistently higher inflationary taxes imposed on rural earnings. These monetary and financial transfer mechanisms in favor of urban residents have had opposite effects to the labor mobility improvements on rural-urban disparity. We believe that the effects of inflation and income redistribution between 1985-92 were powerful enough to produce the upward portion of the Vshaped trend. This argument is consistent with the evidence that real output growth in the urban sector lagged behind the real output growth in the rural sector during the entire time period. Discretionary policies of the central government may also have a powerful influence on the welfare of the people. After two years of double-digit inflation in 1988 (18.5%) and 1989 (17.8%), the government launched a series of contractionary polices that tightened investment credits and restricted the growth of new firms. Blamed for competing for raw materials with the state enterprises and producing low quality products, the government increased its control over rural industries. For two consecutive years 1989-90, the total number of TVEs and their employment dropped, and in 1990-91, the price indices of agricultural products showed reductions of 2.6 and 2 percent (SSB 1993). These policy consequences would either directly or indirectly lower the earnings of the rural people, resulting in an upward shift in the rural-urban gap.

VI. Summary and Concluding Remarks The main objective of this paper has been to examine whether economic reforms have reduced China’s ruralurban segmentation, a main feature of the centrally planned system. We have found that per capita consumption and income levels are still much higher in cities and, in fact, these gaps have been increasing in recent years after a brief decline between 1978-85. This sectoral division is consistent with production function

changes in the state and non-state sectors. 21

estimates based on 1987-92 provincial data that revealed higher labor productivity in urban/state-owned industries than in rural industries and agriculture. Although the economic reforms have raised the nation’s standard of living, they have not corrected the urban bias. This paper has emphasized the sectoral allocation of labor as an equilibrating mechanism that can improve rural-urban integration. However, the pursuit of a heavy-industry oriented development strategy before 1978 had insulated China’s factor markets, causing much higher labor incomes in cities than in the countryside. Although economic reforms have relaxed mobility restrictions, we argue that the existing institutions, such as urban housing, other welfare provisions, and the rural land arrangements, are still responsible for dividing the two sectors and distorting labor allocation. Farmers still do not receive the full benefits of migration deregulations, and rural-urban differences continue to exist. Our analysis also suggests a disturbing fact that, although the rural sector has been the engine of China’s economic growth since the reforms, the urban sector has received fruits of the reforms disproportionate to its contributions. The distributive mechanism has been a combination of increased urban investments and subsidies and a relatively higher inflationary tax on rural earnings. This urban biased policy mix has gradually replaced the more visible, physical restrictions on rural-to-urban migration. What are the efficiency losses caused by these redistribution schemes? Are the transfer programs viable in the long run? What institutional changes are necessary to reduce the incentives of the Chinese government in pursuing the urban-biased policies? These questions are worthy of further investigation because they directly affect China’s future political stability and economic efficiency.

22

References Ashton, Basil, Kenneth Hill, Alan Piazza and Robin Zeitz. "Famine in China, 1958-61," Population and Development Review 10 (December 1984): 613-45. Bernstein, Thomas P. Up to the Mountains and Down to the Villages: The Transfer of Youth from Urban to Rural China. New Haven, Connecticut: Yale University Press, 1977. ______. "Stalinism, Famine, and Chinese Peasants." Theory and Society 13 (May 1984): 339-377. Brandt, Loren and Xiaodong Zhu. “Soft Budget Constraints and Inflation Cycles: A Positive Model of the Post Reform Chinese Economy,” Department of Economics, University of Toronto (October 1995). BRLMS, Bulletin of Rural Labor Mobility Studies. Beijing: Eight Study Groups on Labor Mobility, 1995-96. Cai, De-Rong. China’s Urban Housing: Theory, Practice, and Reform Strategies. Beijing: China Statistical Publisher, 1991. CAY, China Agricultural Yearbook. He Kang, ed., English edition, Beijing: Agriculture Publishing House, 1987. Chan, K. W. and Xu, X. "Urban Population Growth and Urbanization in China since 1949: Reconstructing a Baseline," The China Quarterly (1984): 583-613. Chang, Kyung-Sup. "Chinese Urbanization and Development Before and After Economic Reform: A Comparative Reappraisal," World Development 22, No.4 (1994): 601-14. Chen, Chu-yuan. China’s Economic Development: Growth and Structural Change. Boulder, Colorado: Westview Press, 1982. Chen, Jian and Belton Fleisher. “Regional Income Inequality and Economic Growth in China,” Journal of Comparative Economics 22 (1996). China, State Statistic Bureau (SSB), China Statistical Yearbook, 1986-1993. Beijing: Statistical Publisher of China, 1987-1994. Cook, Sarah B.. “Agricultural Labor Productivity and Household Incomes during Market Transition: Evidence from A Chinese County.” Presented at the Annual Meeting of Association for Asian Studies, Honolulu, April 1996. FBIS, Foreign Broadcast Information Service, Daily Report: China. Springfield, Virginia: National Technical Information Service, U.S. Department of Commerce, 1984. Forbes, D. and Linge, G. "China's Spatial Development: Issues and Prospects," in G. Linge and D. Forbes eds.: China's Spatial Economy - Recent Development and Reforms. Hong Kong: Panther Press, 1990. Goldstein, Sidney "Urbanization in China, 1982-87: Effects of Migration and Reclassification," Population 23

and Development Review, December 1990, Vol.16, No.4, 673-701. Hare, Denise. “Efficiency Considerations of Out Migration from Rural China.” Presented at the Annual Meeting of Association for Asian Studies, Honolulu, April 1996. Jefferson, Gary H., Rawski, Thomas G., and Zheng, Yuxin. "Growth, Efficiency, and Convergence in China's State and Collective Industry," Economic Development and Cultural Change (1992). Johnson, D. Gale. The People's Republic of China: 1978-1990, San Francisco: ICS Press, 1990. ______. “Is Agriculture A Threat to China’s Growth?” Office of Agricultural Economics Research, The University of Chicago, Paper No. 95:04 (April 1995). Khan, Azizur R., K. Griffin, C. Riskin, and Zhao R. W. “Sources of Income Inequality in Post Reform China,” China Economic Review 4 (1993), 1:19-25. Knight, John and Lina Song. “China’s Labor Markets” (in Chinese), Bulletin of Rural Labor Mobility Studies (Beijing: August 1995). Lardy, Nicholas. Agriculture in China’s Modern Economic Development. Cambridge: Cambridge University Press, 1983. Lewis, Arthur "Development with Unlimited Supplies of Labor," The Manchester School of Economics 22 (May 1954): 139-91. Lin, Justin Yifu. "Collectivization and China's Agricultural Crisis in 1959-61." Journal of Political Economy 98, No.6 (1990). ______. “Rural Reforms and Agricultural Productivity Growth in China,” American Economic Review (1992). ______ and Yang, Dennis Tao. "Food Availability, Entitlement, and the Chinese Famine of 1959-1961,” Duke University, working paper, 1995. Lyons, Thomas P. “Interprovincial Disparities in China: Output and Consumption, 1952-1987,” Economic Development and Cultural Change 39 (1992), 3: 471-506. McMillan, J., J. Whalley, and L. Zhu. “The Impact of China’s Economic Reform on Agricultural Productivity Growth,” Journal of Political Economy 97, 4 (1989): 781-807. Nolan, P. and White, G. "Urban Bias, Rural Bias or State Bias? Urban-Rural Relations in Post-Revolutionary China,"Journal of Development Studies 20, (1984): 52-81. Oi, Jean. “Reform and Urban Bias in China,” Journal of Development Studies, Vol.29, No.4 (July 1994). Perkins, Dwight and Shahid Yusuf, Rural Development in China. Baltimore: John’s Hopkins University Press, 1984. 24

Putterman, Louis. “Dualism and Reform in China,” Economic Development and Cultural Change 40 (1992): 467-93. ______ and Ana Chiacu, “Elasticities and Factor Weights for Agricultural Growth Accounting: A Look at the Data for China,” China Economic Review 5, no.2 (1994): 191-204. Ranis, Gustav and Fei, John C.H. "A Theory of Economic Development," The American Economic Review, No.4 (September 1961): 533-65. Rawski, Thomas. Economic Growth and Employment in China. New York: OUP, 1979. Riskin, Carl. China’s Political Economy: The Quest for Development since 1949. Hongkong: Oxford University Press, 1987. Rural Research Group of Annual Analysis. The 1994 China’s Rural Economy Development Report and Development Trends in 1995. Beijing: China Social Science Publisher, 1995. Shen, Y. and C. Tong. China’s Population Migration (in Chinese). Beijing: China Statistical Publisher, 1992. Sicular, Terry. “Grain Pricing: A Key Link in Chinese Economic Policy,” Modern China 14, no.4 (1988): 451-486. SSB, State Statistical Bureau. Statistical Yearbook of China 1987-93 and 1995 (Beijing: China Statistics Publisher, 1987-93, 1995). Tsu, Kai-Yuen. "Decomposition of China's Regional Inequalities," Journal of Comparative Economics 17, (1993): 600-27. Walker, Kenneth R. Food Grain Procurement and Consumption in China. New York: Cambridge University Press, 1984. Wiemer, Calla. "State Policy and Rural Resource Allocation in China as Seen Through a Hebei Province Township 1970-85," World Development 22, No.6 (1994): 935-47. Wu, Harry X. “Rural Enterprise Contributions to Growth and Structural Change.” In Christopher Findlay, Andrew Watson and Harry Wu, editors, Rural Enterprises in China, New York: St. Martin’s Press: 39-68. Yang, Dennis Tao. “Education and Off-Farm Work,” Economic Development and Cultural Change, forthcoming (1995). _____. "China's Land Arrangements and Rural Labor Mobility," China Economic Review, February 1997. ______, “The Effects of Institutions on Worker Mobility and Labor Market Efficiency,” in The Reformability of China’s State Sector, eds. James Wen, New Jersey: World Scientific Publisher, forthcoming, 1996. Zax, Jeffrey S. “Human Capital in A Worker’s Paradise: Returns to Education in Urban China,” Working 25

Paper, Department of Economics, University of Colorado at Boulder (1994). Zhang, Xiaohe, C. Findlay, and A. Watson. “Growth of China’s Rural Enterprises: Impacts on Urban-Rural Relations,” Journal of Development Studies 31, No.4 (1994a): 567-84. Zhang, Xinmin et al. “The Analysis of Rural-Urban Income Disparity,” State Statistical Bureau of China, Working Paper (1994b).

26

Table 1. Per Capita Consumption Levels of Rural And Urban Residents
Year 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 Note: National average 76 87 89 94 99 102 105 96 102 114 117 116 120 125 132 136 132 134 140 142 147 155 155 158 161 165 175 197 227 249 266 289 327 403 447 508 635 694 723 803 935 Rural residents 62 69 70 76 78 79 83 65 68 82 88 89 95 100 106 110 106 108 114 116 116 123 123 124 125 124 132 152 173 192 210 232 265 324 351 389 473 513 524 570 648 Urban residents 149 181 183 188 197 205 195 206 214 225 226 222 234 237 244 251 250 255 260 267 295 306 313 324 340 360 383 406 468 520 526 547 598 727 833 991 1281 1394 1477 1676 1983 2.4 2.6 2.6 2.5 2.5 2.6 2.3 3.2 3.1 2.7 2.6 2.5 2.5 2.4 2.3 2.3 2.4 2.4 2.3 2.3 2.5 2.5 2.5 2.6 2.7 2.9 2.9 2.7 2.7 2.7 2.5 2.4 2.3 2.2 2.4 2.5 2.7 2.7 2.8 2.9 3.1 Urban/Rural

The units are yuan except for the urban/rural ratio.

Data Source: SSB (1993).

27

Table 2. Urban Per Capita Disposable Income (PDI): Yuan
Urban PDI1 620.2 681.7 721.5 770.4 870.3 967.2 1143.5 1268.7 1477.7 1704.7 1922.0 2148.5 2484.2 Pecuniary PDI 429.4 490.4 525.3 562.9 650.1 738.9 900.0 1002.2 1182.1 1375.8 1512.8 1700.6 2013.3 Housing subsidy 74.3 74.7 74.5 75.0 81.0 91.0 101.8 106.7 115.3 121.9 146.6 159.8 171.8 Medical subsidy 39.9 21.4 22.0 23.0 21.6 22.2 34.6 41.6 51.8 54.7 103.6 123.0 136.5 In-kind transfer3 22.0 22.9 24.7 26.3 30.4 34.4 41.4 45.8 56.0 63.0 69.4 77.2 91.3 Price subsidy4 54.6 72.4 75.0 83.3 87.1 80.9 65.8 72.3 72.4 89.2 89.6 87.9 71.3 Non-pecuniary income2 190.8 191.3 196.2 207.5 220.2 228.3 243.5 266.5 295.6 328.9 409.2 447.9 470.9

Year 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 Note:

1. Urban PDI = Pecuniary PDI + Non-Pecuniary PDI 2. Non-pecuniary PDI = Housing subsidy + Medical subsidy + In-kind transfer + Price subsidy 3. In-kind transfer = 5% of living expenses of the residents, based on urban household surveys 4. Price Subsidy = Total government price subsidies over the total urban population

Data Source: Zhang et al. (1994b).

Table 3. Per Capita Income Disparity between Rural and Urban Residents Unit: Yuan
Urban PDI 620.19 681.71 721.54 770.36 870.31 967.20 1143.49 1268.66 1477.70 1704.68 1921.96 2148.52 2484.24 Deflated urban PDI 576.92 665.08 707.39 755.25 847.43 864.34 1068.69 1174.68 1224.27 1465.76 1897.29 2044.26 2287.52 Rural PDI 191.33 223.44 270.11 309.77 355.33 397.60 423.76 462.55 544.94 601.51 686.31 708.55 783.99 Deflated rural PDI 186.66 219.92 257.98 309.15 353.91 383.05 410.32 445.79 491.69 536.22 667.62 700.04 750.35 Urban/Rural Deflated PDI 3.09 3.02 2.74 2.44 2.39 2.26 2.60 2.64 2.49 2.73 2.84 2.92 3.05

Year 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992

Data Source: Zhang et al. (1994b). 28

Table 4. OLS Production Function Estimates
Independent Variables Constant Labor Capital Intermediate Inputs Sowing Acreage Small Tractor Disaster Area Coefficients Rural Industry 0.726 (0.059)*** 0.094 (0.031)*** 0.736 (0.052)*** 0.282 (0.032)*** ... ... ...

State Industry 0.023 (0.097)*** 0.382 (0.083)** 0.200 (0.082)*** 0.541 (0.060)*** ... ... ...

Agriculture -2.383 (0.356)*** 0.314 (0.062)*** ... ... 0.752 (0.098)*** 0.018 (0.049) -0.217 (0.043)*** 0.867 130 178

Adjusted-R2 F-Statistics Observation Note:

0.974 838 178

0.991 2294 172

1. The numbers in parentheses are standard errors. *, **, and *** represent 10%, 5%, and 1% level of significance. 2. Five year dummies were also used in each equation.

Table 5. Sectoral Marginal Productivity of Labor: Yuan/Person
Year 1987 1988 1989 1990 1991 1992 Agriculture 476.2 476.7 447.8 524.5 555.7 601.2 State Industry 7708.5 8125.9 8086.4 8048.6 8467.1 9346.2 Rural Industry 588.9 708.4 705.6 652.8 880.6 1211.2

29

Table 6. Sectoral Allocation of the Labor Force 1952-1992: millions
Urban1 Labor 16.0 18.6 20.0 21.2 29.8 31.0 51.9 52.8 59.7 51.7 43.2 43.7 46.0 49.7 52.0 53.1 55.0 57.1 62.2 67.9 71.3 73.4 76.5 82.0 86.7 91.1 95.0 99.7 104.4 109.4 112.8 115.2 118.9 123.6 128.1 132.1 136.1 137.4 140.6 145.1 147.9 Rural Labor 182.4 186.1 190.9 195.3 200.3 205.7 213.0 207.8 197.6 202.5 213.8 220.4 229.1 235.3 244.5 253.7 262.9 274.0 281.2 287.5 286.5 292.6 296.8 299.5 301.4 302.5 306.4 310.3 318.4 326.7 338.7 346.9 359.7 370.7 379.9 390.0 400.7 409.4 420.1 430.9 438.0 Rural Agriculture ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 278.1 283.4 282.3 288.6 292.2 294.6 294.4 293.4 283.8 286.9 291.8 298.4 309.2 312.1 309.3 311.9 313.0 317.2 323.1 332.8 341.8 350.2 348.6 Rural Nonagriculture2 ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... 8.8 10.2 10.5 10.5 11.3 12.1 14.7 17.3 31.5 31.9 35.1 36.9 38.1 43.4 58.9 67.0 75.3 81.3 86.1 85.0 86.7 89.0 97.8

Year 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 Note:

1. Urban Labor Force does not include the individual self-employers, which mainly worked outside the industrial sector. 2. Rural Non-agriculture Labor Force mainly consists of the employment in Township and Village Enterprises (known as TVEs). Data Source: Urban labor force and rural labor force were taken form China Statistical Yearbook (SSB, 1993). Rural agricultural and non-agricultural labor force from 1970-1987 were taken from Johnson (1990) p39, and the figures for 1988-1992 were taken from SSB (1989-1993).

30

Table 7. Inflation, Sector-Specific Investments, and Urban Price Subsidies
Year National Retail Price 100.7 102.0 106.0 102.4 101.9 101.5 102.8 108.8 106.0 107.3 118.5 117.8 102.1 102.9 105.4 Urban Consumer Price 100.7 101.9 107.5 102.5 102.0 102.0 102.7 111.9 107.0 108.8 120.7 116.3 101.3 105.1 108.6 Rural Consumer Price -----------------------------------107.6 106.1 106.2 117.5 119.3 104.5 102.3 104.7 SOE Investment 500.99 523.48 558.89 442.91 555.53 594.13 743.15 1074.37 1176.11 1343.10 1574.31 1551.74 1703.81 2115.80 3012.65 Rural Investment ----------------------------30.00 61.00 113.28 128.30 71.82 92.18 124.01 97.88 99.33 130.08 68.00 Urban Price Subsidies 11.14 79.20 117.71 159.41 172.22 197.37 218.34 261.79 257.48 294.60 316.82 370.34 380.80 373.77 321.49

1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992

Notes: All inflation rates are in percentages; investments and subsidies are in nominal prices in 100 million yuan. Data Source: SSB (1993).

31

Figure 1. Rural-Urban Disparity in Income and Consumption: Rural=1

32

Data Appendix Table A1. The Ratio of Non-Agricultural Income to Agricultural Income: An International Comparison (US dollars)
Country Switzerland Denmark Netherlands Canada U. K. Australia Singapore Israel Cyprus Bahrain Seychelles Belarus Hungary Botswana Mauritius Turkey Russia Ukraine Fiji Slovak R. Costa Rica Poland Kazakhstan Lithuania Paraguay Bulgaria Peru Moldova Swaziland Romania Egypt Sri Lanka Ghana Kenya Burundi Malawi Income 36320 26390 20850 20290 18150 17790 17440 13720 10300 8130 6020 3210 3170 2980 2860 2850 2750 2430 2090 2070 2020 1940 1740 1730 1390 1360 1350 1350 1190 1190 660 550 450 330 210 190 1985 1.25 1.50 1.16 0.76 1.15 1.46 ----1.69 1.31 ----1.46 1.11 1.23 4.08 0.92 ----1.05 1.09 1.44 0.99 1.75 1.01 0.96 1.01 2.04 1.16 ----1.09 0.82 1.04 1.28 1.73 1.32 3.18 1.76 3.76 1990 1.25 1.59 1.12 0.77 1.14 1.50 2.08 1.66 1.34 1.69 1.22 1.06 1.40 3.09 1.08 0.99 1.01 ----1.22 0.97 1.83 1.09 0.91 1.03 2.68 0.88 ----0.99 ----0.94 1.62 1.55 1.43 2.86 ----4.33 1995 1.26 ----1.11 ----1.16 --------1.85 1.29 1.72 1.19 1.34 1.66 3.02 1.09 1.39 ----------------1.59 1.19 1.01 1.30 2.01 2.01 ----1.03 ----1.22 1.56 1.51 -----------------

Data Source: The GNP per capita series was obtained from the “World Bank Data and Text on CDROM”, the International Bank for Reconstruction and Development/The World Bank, 1995. The agricultural and non-agricultural compensation was taken from the “Yearbook of Labor Statistics,” the 54th issue of 1995, International Labor Office, Geneva.

33

Table A2. Provincial Averages of Key Regression Variables
Year Variables1 Agriculture 1987 1988 1989 1990 1991 1992 Output 4675.7 4769.4 4620.5 5562.1 6042.7 6509.7 Labor 30870.0 31455.7 32440.5 33336.4 34186.3 34037.0 Sowing Area 217434.4 217303.6 219830.9 222543.1 224378.7 223510.6 Rural Industry Output 1987 1988 1989 1990 1991 1992 2934.1 3673.4 3523.3 3647.4 4448.1 6607.9 Labor 4702.5 4893.9 4712.5 5455.1 4767.0 5148.8 State Industry Output 1987 1988 1989 1990 1991 1992 8250.1 9001.1 9049.7 9201.0 9917.8 11068.3 Labor 4086.0 4229.0 4272.6 4364.4 4471.9 4521.2 Capital Input Inputs2 5242.4 5312.5 5202.5 5730.8 6237.1 6592.7 2215.0 2254.2 2430.5 2833.9 2865.7 6060.1 Intermediate Capital Input 999.8 1085.7 1099.3 1182.3 1285.4 1552.1 Intermediate Inputs 1134.6 1355.8 1398.1 1590.4 1919.0 2439.2 Small Tractor 4713.0 5319.0 5848.0 6231.4 6528.6 6604.0 Disaster Area 2039.0 2394.0 2443.0 1752.2 2751.4 2589.3

Note:

1. Output unit = 100 million yuan, labor unit = 10 thousand persons, sowing area unit = 10 thousand mu, small tractor unit = 10 million walt, disaster area unit = 10 thousand acre, capital input unit = 100 million yuan, intermediate input unit = 100 million yuan. 2. The definition of intermediate inputs became more inclusive in 1992. In regression analysis, we assume that the relative changes are proportional across different provinces so that the year dummy variables would capture the effects of the change in variable definition. The estimated regression coefficients are therefore not affected.

Data Source: SSB (1987-1993). 34


相关文章:
更多相关标签: