Agricultural productivity growth in China: farm level versus aggregate measurement

China Economic Review 14 (2003) 53 – 71

Agricultural productivity growth in China: farm level versus aggregate measurement Colin A. CARTER a,b,*, Jing CHEN a, Baojin CHU c a

Agricultural and Resource Economics, University of California, Davis, CA, USA Giannini Foundation of Agricultural Economics, University of California, USA c Agricultural Economics, Nanjing Agricultural University, Nanjing, China

b

Accepted 26 August 2002

Abstract We measure agricultural productivity growth in China using alternative data sets: farm level data for Jiangsu province, national data, and provincial aggregate data for Jiangsu. For all three data sets, productivity growth was estimated to be strong during the immediate post-reform 1978 – 1987 period. According to the farm level data, productivity growth then slowed from 1988 to 1996. Alternatively, the national and provincial aggregate figures showed continued high productivity growth in the 1990s. These findings suggest that aggregate data may blur the true picture with regard to agricultural productivity growth in China. D 2002 Elsevier Science Inc. All rights reserved. JEL classification: O47; O53; Q12 Keywords: China; Productivity growth; Aggregation issues

1. Introduction It is generally accepted that the rapid output growth in China’s agriculture from 1979 to 1984 was due to significant productivity gain (Lin, 1992; Wen, 1993). In most developing countries such as China, agricultural productivity gain is central to the growth

* Corresponding author. Agricultural and Resource Economics, University of California, One Shields Avenue, Davis, CA 95616, USA. E-mail address: [email protected] (C.A. Carter). 1043-951X/02/$ - see front matter D 2002 Elsevier Science Inc. All rights reserved. doi:10.1016/S1043-951X(02)00086-X

54

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

of national wealth (Johnson, 1997). The continuation of agricultural productivity growth in China is particularly important, as more than 300 million workers remain in agriculture (nearly 50% of the country’s total labor force).1 An increase in rural incomes, through further agricultural productivity gain, would help close the relatively wide urban – rural income gap.2 The coastal-centered economic boom slowed down in the late 1990s and the reform of China’s state owned enterprises resulted in high urban unemployment (Saywell, 1998). This suggests that China’s overall development policy will continue to discourage labor movement from the countryside to the cities and therefore the rural economy itself will be viewed as a key to future national economic growth. Labor is an abundant resource in rural China and a large percentage of the labor force is used in grain production. However, grain cultivation is a relatively low-return activity, and the marginal labor productivity in grain is small. Further economic reform in the countryside would encourage farmers to withdraw from grain in favor of other crops or activities. However, agricultural reform is slow moving in China. From 1998, the central government reasserted its emphasis on ‘‘grain self-sufficiency’’ and introduced renewed government control over grain prices, by prohibiting private agents in the grain market (Crook, 1998). According to national aggregate data, total factor productivity (TFP) in China’s agriculture increased by 55% from 1979 to 1984 (Wen, 1993). This jump in productivity was unprecedented in the developing world, and most of the rapid change was attributed to the Household Responsibility System (HRS), which was a one-off institutional change.3 After the effects of the HRS petered out, a policy issue that surfaced in the late 1980s and early 1990s was a slowdown in the growth of investment in agriculture.4 Despite this apparent investment slowdown, we report a surprising result in this paper. Namely, the tremendous agricultural productivity gains enjoyed by China in the 1980s continued well into the 1990s. The total factor productivity index (TFPI) increased by almost 50% from 1988 to 1996, according to national aggregate statistics, and using Wen’s (1993) methodology. Are these strong

1 There is considerable variation in estimates of the percent of the labor force engaged in agriculture in China. For instance, Rawski and Mead (1998) argue that China’s official data overestimate the number of Chinese farm workers. 2 See Lardy (1994, p.24), who suggests that the gap in China’s urban – rural living standards is wider than anywhere else in Asia. 3 Stone (1993) suggests that the effects of the HRS may have been overstated. He indicates that several technological improvements were made prior to 1979. These included the adoption of new varieties of wheat, rice, and corn. For wheat and rice, it was new short-straw varieties and, for corn, it was hybrid varieties. In addition, Stone documents the significant improvement in irrigation facilities prior to institutional reform and the accelerated growth of fertilizer supplies. Stone (p. 352) notes that, ‘‘For staple crops, the increased supply of fertilizer nutrients was more significant than labour incentives fostered by the responsibility system reforms, which on balance led labour away from the previous overconcentration on staples. Food grain yields had been constrained not by inadequate labour application, but by insufficient soil nutrients.’’ 4 Total investment in agriculture slowed down between 1985 and 1990, and actually fell in real terms over this period. It then resumed growth at the beginning of the 1990s, but fell again in 1993 and 1994, in real terms. Only after 1995 did real investment in agriculture begin to rise again (Statistical Yearbook of China, 2000).

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

55

agricultural productivity gains in the 1990s plausible? This question is the motivation for our paper. We hypothesize that the continued high productivity gains in the 1990s seems unrealistic due to potential data aggregation biases5 and because of recent concerns over the reliability of China’s national agricultural production statistics.6 The purpose of this paper is to measure post-reform agricultural productivity growth in China using farm level (i.e., household) data and to compare the results to both national and provincial aggregate data, for the 1978 – 1996 time period. Measuring productivity growth is a complicated task, even in western economies where data are more reliable than is the case for China.7 Previous studies of China’s agricultural growth have all used very aggregate national or provincial data, even though the theory is based on microeconomic decision-making relationships at the individual firm level. Our disaggregate household level data are unique and they were obtained from farm cost surveys in Jiangsu Province, one of the most progressive agricultural provinces in China. Comparing the household results with those from aggregate data, we find that the productivity estimates tell a similar broad story (of high productivity growth) from 1978 to 1987, and then the estimates diverge from 1988 to 1996.8 The productivity growth rates estimated with national and provincial aggregate data remain quite high in the second time period but those obtained with household data show a significant slowing of the productivity growth rate (see Fig. 1). We suggest that the 1988 –1996 productivity results from the household data are more convincing and more consistent with expectations, given the slowdown in investment in China’s agriculture in the 1980s and the financial problems experienced by China’s farmers in the 1990s (Statistical Yearbook of China, 2000). According to the Jiangsu household data, the TFPI increased by 19% from 1990 to 1996, much less than the 44% growth implied by the national aggregate figures and the 61% implied by the Jiangsu provincial figures. The rest of this article is organized as follows. In Section 2, we describe the data and method used in this paper. In Section 3, we report results for national productivity indices. In Section 4, productivity results are reported for Jiangsu province, using both

5 The economics literature has debated the aggregation issue and, in particular, the practice of including intermediate inputs in output measures. This is discussed in Domar (1961, 1962). The issue of double counting with aggregate data is clearly a potential problem. Ball, Bureau, Nehring, and Agapi (1997) used aggregate data to estimate agricultural productivity growth for US agriculture. They report that ‘‘output is defined as gross production leaving the farm as opposed to real value added.’’ For livestock output, this means they net out onfarm use of feed, but otherwise include feed in the output measure. For the aggregate results reported below, we have followed more or less the same approach as Ball et al. (1997). 6 See Fuller, Hayes, and Smith (2000) who find that the growth of livestock production may be grossly overstated by China’s national statistics. This would clearly bias any measurement of agricultural productivity, because livestock accounts for about 30% of China’s agricultural GDP. 7 The overall reliability of China’s economic statistics is discussed in Rawski and Xiao (2001). 8 For the purposes of this study, we refer to the 1978 – 1987 time period as the first time period and to 1988 – 1996 as the second time period.

56

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

Fig. 1. National, provincial (Jiangsu), and farm level TFPI: 1978 – 1996. Sources: the farm level TFPI was calculated from farm cost survey data obtained from the Jiangsu Provincial Price Bureau, 1992 – 1997. The national and provincial aggregate TFPIs were calculated from data published in various issues of China Statistical Yearbook (State Statistical Bureau) and Jiangsu Statistical Yearbook (Jiangsu Statistical Bureau).

farm level household data and provincial aggregate data. Section 5 provides concluding comments.

2. Data and methodology We chose Jiangsu province for the purposes of our disaggregate analysis because we had access to a unique data set for this province and Jiangsu represents one of the most important agricultural provinces in China. In 1999, Jiangsu’s total gross value of agricultural output (GVAO), including farming, forestry, animal husbandry, and fisheries, was the third largest of any province in the country, behind Shandong and Henan. Compared to other provinces, the value of Jiangsu’s production from farming, animal husbandry, and fisheries all ranked in the top five in the nation in 1999 (Statistical Yearbook of China, 2000). Jiangsu was the third largest rice producing province and the fourth largest wheat producer in China. Jiangsu should have enjoyed as much, or even more, productivity growth than the national average, since 1978. In 1999, the per capita net income of rural households in Jiangsu was the third highest of any province in the country,9 and was more than one third higher than the national average. The more advanced economic status suggests that Jiangsu would have invested more in agricultural research and development than the 9

The municipal cities of Beijing, Tianjin, and Shanghai are excluded.

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

57

average province, which presumably would help give rise to higher productivity growth.10 Although the Statistics Bureau in China is the main source of data for economic studies of China, the farm level Jiangsu data used in this paper were collected using a different procedure, namely farm household surveys. The Jiangsu provincial Price Bureau conducts these surveys, and they have been carried out for decades. Many of the same households are included in the survey each year and the sample is sufficiently large to represent agriculture across the entire province. The number of included households varies across agricultural activities, according to the relative importance of each activity in each household. For example, the 1996 survey covered almost all counties in the province and more than 800 households were included in the survey. However, only 32 counties and 111 households were included in the portion of the survey dealing with cash crops, while 43 counties and 203 households were included in the section covering wheat production. For the purposes of our analysis, the household survey data were aggregated to county averages, and then to prefecture averages.11 The Jiangsu farm level data cover almost all-farming activities, including cropping and livestock. This includes wheat, indica rice, japonica rice, corn, soybeans, cotton, rapeseed, hogs, and fresh water fish. The sum of these activities accounts for over 90% of total agricultural output value and 85% of sown area in the province.12 The results of the household survey contain detailed information on both outputs and inputs for all crops, hogs, and fish. The crop data are based on unit-sown area (mu) and the livestock data are on an animal unit basis. The farming cost data is very detailed and is classified into nonlabor and labor costs, with several subcategories. National aggregate data published in various issues of China’s Statistical Yearbook were used for the national productivity estimates in this paper. Wen (1993) studied China’s GVAO, which consists of two main components: crop output (GVCO) and noncrop output (GVNCO). The latter is defined as production of animal husbandry, forestry, fishing, and sideline activities.13 Following Wen’s approach, we use four aggregate input categories: labor, land, capital, and current inputs. At the national level, agricultural labor is defined to include all workers engaged in farming, animal husbandry, sideline activities, fisheries, forestry, and water conservancy. The amount of arable land was adjusted for both multicropping and irrigation, following

10 For evidence of the linkage between advanced economic status and more investment in agricultural research, see Huang and Rozelle (1996). 11 There are 67 counties and 13 prefectures in Jiangsu Province. Officially, there were 11 prefectures prior to 1996. 12 Of the total value of provincial agricultural output in 1996, 63% was from crop farming, 21.8% was from livestock, and 14.1% was from fisheries. Vegetables were not included in our household data for two main reasons. First, vegetables were a relatively small share of the provincial GVAO in 1996, accounting for about 7% of provincial GVAO. Second, vegetable quality and price differences vary from season to season and from region to region and, therefore, there is no standard basis for statistical reporting of the value of vegetable production. 13 The gross output value from rural industry run by townships and villages was excluded from the GVAO calculation.

58

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

Wen. Farm capital is the sum of the value of draft animals, nondraft animals, poultry, and farm machinery. Current inputs include such items as seed and feed, organic and chemical fertilizer, electricity and insecticides. The Jiangsu Statistical Yearbook is directly comparable to China’s Statistical Yearbook and therefore all provincial aggregate variables used in this paper are defined exactly as for the national aggregate data. For the national and provincial aggregate data, the TFPI is defined as follows: TFPI ¼ f100  ðGVAO indexÞg=faðL indexÞ þ hðK indexÞ þ H ðS indexÞ þ yðC indexÞg

ð1Þ

where a, h, H , and y are weights, and L, K, S, and C are indexes of labor, capital, land, and current inputs. The numerator is the index of the GVAO, and the denominator is the weighted index of the four inputs. The weights proposed by Wiens (1982) to aggregate inputs were used to calculate the weighted input index, as in Wen (1993). The weights are as follows: labor (0.35), capital (0.09), land (0.36), and current inputs (0.20). All variables were deflated, following Wen’s technique. To compute the TFPI using household data in Jiangsu province, we first calculated an output and input index for each farming activity. The output value of each farming activity was measured using 1978 prices. In computing the input indexes, we first computed indexes for labor, nonlabor, and land inputs14 for each farming activity, and then we used the Wiens’ (1982) fixed weights to compute the overall input index. Using this fixedweight approach, results from the farm-level data can be directly compared with those derived from the national and provincial aggregate approach.15 Using the household data, we generated an overall TFPI estimate to enable us to compare household productivity results with the aggregate provincial and national data. To achieve this end, the input and output indexes for the entire cropping sector were computed as the weighted averages of the input and output indexes of each cropping activity, and the weights were set equal to the share of each crop in total Jiangsu acreage. The input and output indexes for the entire agricultural sector were then computed as the weighted averages of those for cropping and hogs,16 and the weights were equal to the share of output value in the GVAO for each year, for Jiangsu.

14 For the farm-level data, the land index was set at 100 throughout the entire period, because the survey was based on a unit area (mu). The weight assigned to nonlabor inputs was the sum of the capital and current input weights. 15 We also attempted an alternative approach to compute the input indexes. Instead of constructing the input indexes for labor, nonlabor, and land separately, we aggregated all input costs including nonlabor inputs (such as chemical fertilizer, seeds, and draft animals) and labor costs. Nonlabor costs and labor costs were then deflated by the general rural retail price index of industrial products and the general retail price index, respectively. Then, we constructed an input index from the aggregated input costs. A possible advantage of this alternative approach is that it does not rely on predetermined input weights. However, both approaches to measuring input growth gave similar results. 16 Fisheries were not included in the computation of the total input, output, and TFPI indexes due to the relatively small sample size for fisheries compared to crops and hogs. However, the inclusion of the fishery sector produces very similar TFP conclusions.

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

59

Although it is beyond the scope of this paper, further research should utilize household data to estimate production functions and measure TFP growth accordingly.17 Our objective was to compare the TFP results from three different data sets, and thus we chose a relatively simple methodology and held the input weights constant across the three data sets to allow us to compare the results. For a survey of alternative approaches to estimating China’s agricultural TFP, see Carter (1999).

3. Results using national data Using Wen’s (1993) approach and national data, the national productivity results are reported in Table 1 and Fig. 1. We find that the TFPI increased sharply in the 1990s, after relatively slow growth in the late 1980s. The estimated national TFPI increased from 221 in 1991 to 318 in 1996, growing at more than 7% per year, on average. Indexes for the GVAO and for the gross value of crop output (GVCO) are also reported in Table 1. The estimated GVAO increased by about 10% per year from 1991 to 1996, exactly twice the growth rate experienced during the 1985– 1990 time period. The difference between the GVAO and the GVCO represents the gross value of noncrop output. The gap between the two indexes grew after 1984, especially in the 1990s. Estimated labor productivity lagged behind TFP until the 1990s, and increased rapidly in the latter part of the study time period (Table 1), due to the swift growth in reported national agricultural output and a leveling off in reported national agricultural labor usage. The estimated number of farm workers increased from 283 million in 1978 to 350 million in 1991. In 1992, the absolute number of workers in agriculture began to decline for the first time since reform and, by 1996, there were an estimated 329 million workers engaged in agriculture, according to the official national data. Rapid rural industrial growth and migration has allowed for the movement of some labor out of China’s agriculture. Turning to the other national input indexes, we find that the aggregate effective sown area did not change significantly throughout the study’s time period. Capital was computed as the sum of the value of draft animals, nondraft animals, poultry, and farm machinery. The total value of machinery used in 1996 was more than three times that used in 1978, and the total value of animals increased by 50% during the same time period. Thus, the capital index in the mid-1990s was double the level in 1978, with an annual growth rate of 4.2%. For the purposes of calculating ‘‘current’’ inputs, Wen (1993) made an effort to include as many items as possible, and he assigned weights to each item. The same weights are used in this paper and the current input index is computed as the weighted index of nutrients (i.e., fertilizer), feed, seed, electricity, and insecticides.18

17

We thank one of the reviewers for this suggestion. The weights are follows: 0.646 (nutrient index), 0.155 (feed index), 0.155 (seed index), and 0.044 (index of electricity and insecticides). 18

60

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

Table 1 Agricultural productivity in China using national aggregate data, 1978 – 1996 Year

IGVAOa IGVCOb

Index of Farm Effective Capital Current labor sown area inputsa

1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996

100 119 127 141 158 172 195 202 212 230 243 230 264 273 289 309 363 409 444

100 111 120 124 139 152 167 175 183 194 190 180 213 215 222 232 265 299 321

100 101 103 105 109 110 109 110 111 112 114 118 120 124 123 120 120 117 116

Growth ratesc 1978 – 1996 7.3 1978 – 1987 9.0 1988 – 1996 8.2

5.7 7.4 6.9

1.0 1.3 0.0

100 97 96 96 96 95 95 95 94 95 95 96 96 97 97 97 97 97 98

0.0  0.5 0.3

Weighted Labor TFP index input index productivity index

100 108 110 112 116 120 128 134 145 150 159 164 167 170 176 183 194 208 219

100 106 109 109 114 104 93 103 112 122 127 133 139 151 161 174 195 215 219

100 101 102 103 106 105 103 105 108 111 114 117 120 124 126 128 133 138 140

100 117 123 134 144 156 179 183 191 205 213 195 219 221 235 257 303 351 382

100 117 124 137 149 164 191 192 196 207 214 197 221 221 229 241 273 297 318

4.2 4.3 4.0

4.3 0.8 7.4

1.9 0.9 2.6

6.3 7.8 8.2

5.4 8.1 5.6

Sources: Computed based on Eq. (1) and data from various issues of China Statistical Yearbook (1990 – 1997), State Statistical Bureau. a IGVAO and IGVCO are indexes of the gross value of agricultural output and the gross value of crop output, respectively; the general purchasing price index of farm products was used to convert each series into real terms. b Weights are distributed as labor (0.35), capital (0.09), land (0.36), and current inputs (0.20). c The growth rates were calculated with the following regression expression: ln( Y)=a + b * time.

The total nutrients index is a weighted sum of green fertilizer, human night soil, draft animal manure, hog manure, oil cake, compost, mud and pond manure, and chemical fertilizer. While chemical fertilizer application increased by more than three-fold from 1978 to 1996, the slow increase in the other components of the nutrient index meant that the total nutrient index only increased by about 118% from 1978 to 1996. The feed index increased by 62%, and the seed index, which is computed as a function of sown area, was almost constant during the entire time period. The value of electricity usage increased by more than five-fold from 1978 to 1996, but the usage of insecticides in 1996 was only 33% of the level in 1978. Thus, the index for electricity and insecticides decreased from 100 in 1978 to 91 in 1985, and then reached 219 by 1996. So, as a result of the developments in input usage reported in this paragraph, the overall index for ‘‘current’’ inputs increased 114% from 1978 to 1996.

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

61

Do the national TFP results in Table 1 accurately reflect the genuine effects of agricultural development in China in the 1990s? Or alternatively, is the accuracy of the national data and aggregate approach questionable? According to the official aggregate data, the gross value of the animal husbandry sector grew at a rate of 12% in the 1990s. However, it has been shown elsewhere that the figures for livestock production are exaggerated and the annual production growth for beef, pork, and poultry could be overestimated by at least 7% per year.19 This is perhaps reason enough to distrust the national data.

4. Results using household and provincial aggregate data Our detailed household level data enable us to compute productivity measures for each farming activity. In contrast, the national or provincial aggregate data only allow us to compute the TFPI for the total agricultural sector. The reason for this limitation associated with the aggregate data is the absence of national or provincial data on the proportions of each input used for individual crop or livestock activities. In other words, one main advantage of our household data set is that it contains very detailed input statistics and reports costs associated with each crop or livestock activity, at the farm level. In contrast, national or provincial aggregate data includes only an aggregate measure of inputs used for all agricultural activities lumped together. Using these household farm level data, we found that for Japonica rice, corn, soybeans, and fish, there was relatively high growth in unit output (measured by quantity) over the entire study period, 1978 –1996 (see Appendix A). The yield growth rates per mu for the major grain and cash crops in Jiangsu province were lower from 1988 to 1996 (i.e., the second period) compared to 1978 – 1987 (i.e., the first period), except for wheat. Table 2 reports the growth in output value, by farming activity, based on the Jiangsu household data. To a large extent, the growth in the value of output indexes in the second period was induced by increases in real prices in the late 1980s and the mid-1990s. For example, there was virtually no increase in the yields of either hybrid indica paddy rice or corn from 1988 to 1996. For all grain, the annual yield growth rate in Jiangsu for the entire 1978 –1996 time period was 2.7% (Appendix A), while the annual growth rate in unit value was 4.6% in real terms (Table 2). According to the national data, grain yields grew at an average annual rate of 2.9% across all provinces from 1978 to 1996. From 1978 to 1987, national grain yields grew

19 China’s State Statistical Bureau has recognized the problem of exaggerated meat production (see the 1998, 1999, and 2000 China Statistical Yearbook), and the Bureau shifted the estimated livestock production downward for post-1996 period. However, gross output value of animal husbandry was not adjusted in the yearbooks.

62

Table 2 Agricultural output in Jiangsu: household data, 1978 – 1996 Year

Growth rates 1978 – 1996 1978 – 1987 1988 – 1996

Japonica rice

Hybrid rice

Wheat

Corn

Yuan

Index Yuan

Index

Yuan

Index Yuan

65.9 85.6 75.7 81.1 84.8 91.4 100.1 105.5 118.7 104.1 104.9 111 97.9 96 112.3 138.7 178.5 185 181.9

100 129.9 115 123.2 128.8 138.7 152 160.2 180.3 158.1 159.2 168.5 148.6 145.8 170.6 210.5 271 280.9 276.1

100 104.3 104.3 103.4 110.8 119 132.1 128 135.6 133.3 126.2 136 124.5 124.9 115.5 133.2 212.7 204.7 178.8

87.3 88.8 98.9 96.5 102.9 123.4 141.5 135.9 160.9 149.3 146.2 147.8 136.5 173.6 167.9 192.9 264.1 259.5 240.4

100 68.6 100 48.5 101.7 97.3 141.8 65.4 113.2 78.6 114.6 59.4 110.5 82.7 120.6 77.7 117.8 85.2 124.2 76.3 141.3 80.5 117.3 75.6 162 89.7 130.7 82.7 155.6 98.5 143.5 95.8 184.2 98.4 143.4 111.8 170.9 88.1 128.4 102.1 167.4 82.6 120.5 114 169.2 71.6 104.3 109.8 156.3 76.2 111.1 87.9 198.7 65.4 95.3 100.9 192.2 96.2 140.2 112.3 220.9 84.1 122.6 111.4 302.4 101.4 147.8 149.4 297.1 118.1 172.2 168.4 275.2 134.6 196.2 122.4

4.6 5.3 8.8

104.5 109 109 108.1 115.8 124.4 138 133.7 141.7 139.3 131.9 142.2 130.1 130.6 120.7 139.2 222.2 213.9 186.9

3.1 3.8 6.3

5.7 7.2 8.5

Index Yuan

1.4 2.2 7.1

Soybeans Index Yuan

Rapeseed

Index Yuan

Cotton

Index Yuan

Hogs Index Yuan

Fish Index Yuan

Index

100 23.6 100 61.4 100 153.2 100 120.5 100 325.4 100 135 37.9 160.5 76 123.8 191.2 124.8 141.7 117.6 269.6 82.9 122.5 38.4 162.8 62.8 102.3 128.3 83.7 134.6 111.7 343 105.4 160.3 57.5 243.5 99.2 161.6 151.8 99 145.6 120.8 411 126.3 157.6 61.1 258.7 111.8 182.1 142.5 93 145.1 120.4 407.8 125.3 156 64.2 271.9 95.9 156.3 211.6 138.1 131.4 109.1 952.6 292.8 170.7 57 241.3 104.7 170.6 185.3 121 133.7 111 832.6 255.9 197.8 67.3 285 85.4 139.2 130 84.8 175.6 145.8 1399.1 430 230.8 86.5 366.3 92.8 151.3 165.8 108.2 146.8 121.9 1510.3 464.2 210.8 67.5 285.8 93.2 151.8 189.6 123.8 184.3 152.9 1891.8 581.4 235.3 82.5 349.2 72.2 117.6 196.8 128.4 254 210.8 2783.3 855.5 226.7 81.5 345.1 72 117.4 201 131.2 216.3 179.5 1026.6 315.5 181.4 71.4 302.3 94.3 153.7 227.1 148.2 188.7 156.6 1086.8 334 208.3 76.9 325.8 80.2 130.8 247.7 161.6 189.3 157.1 1129.9 347.3 231.9 88.6 375.1 83.6 136.3 177.2 115.7 174.3 144.6 1049.2 322.5 229.9 93 393.9 79.9 130.3 195.3 127.4 156.3 129.7 1455.1 447.2 308.4 98.4 416.7 99.9 162.9 309.9 202.3 240.1 199.3 2052 630.7 347.7 112.2 474.8 100.1 163.1 317.8 207.4 216.7 179.9 1506.8 463.1 252.6 149.1 631.2 92.3 150.4 298.2 194.6 193.2 160.4 1009.3 310.2

4.9 7.9 4.5

6.7 10.9 6.9

0.8 4.1 3.5

3.7 1.2 5.7

2.9 3.1 1.3

8.8 23.0 2.3

Source: Jiangsu Provincial Price Bureau, 1991 – 1997. 1. The output value for each crop is calculated on the basis of 1 mu sown area; for fish, it is calculated on the basis of 1 mu of water area; for hogs, it is calculated on the basis of one animal unit. 2. All values are based on 1978 prices and are deflated with the general purchasing price index. 3. The growth rates were calculated with the following regression equation: ln( Y)=a + b * time.

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996

Grain

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

63

Table 3 Unit agricultural input value indexes in Jiangsu: household data, 1978 – 1996a Year

Grain Indica rice Japonica rice Wheat Corn Soybeans Rapeseed Cotton Hogsb Fish

1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996

100 108 115 108 106 109 108 108 109 111 114 120 123 122 124 125 149 157 172

100 109 115 110 110 101 113 111 111 117 119 125 129 123 118 122 147 161 158

100 96 99 100 99 96 98 99 101 108 113 116 115 121 116 113 139 147 160

100 109 119 111 108 109 110 112 107 108 109 114 119 117 120 117 133 135 155

100 106 113 107 104 112 116 110 118 121 126 128 141 136 142 128 146 160 167

100 110 118 106 103 115 122 110 122 121 137 144 154 151 168 165 175 184 195

100 99 111 110 105 113 115 110 111 112 107 118 117 117 117 108 126 132 143

100 105 77 109 113 118 120 108 110 122 127 131 139 140 145 131 157 172 180

100 116 113 124 121 122 127 161 152 165 222 221 189 181 173 144 216 223 195

100 78 95 106 104 195 183 249 270 377 554 256 256 259 256 371 465 416 268

Growth rates 1978 – 1996 2.2 1978 – 1987 0.4 1988 – 1996 4.8

2.0 0.8 3.6

2.4 0.6 4.0

1.4 0.2 3.6

2.5 1.6 3.0

3.6 1.6 4.2

1.2 1.2 2.6

3.2 2.5 4.0

3.9 5.0  0.8

8.5 16.8 0.2

Source: Farm cost survey data, Jiangsu Provincial Price Bureau, 1991 – 1997. The growth rates were calculated with the following regression expression: ln(Y )=a + b * time. a Except for hogs, the input indexes were calculated as the weighted average of labor, nonlabor, and land indexes with the weights: labor (0.35), land (0.36), and nonlabor (0.29). b The input index for hogs was calculated as the weighted average of labor and nonlabor (including feed) inputs with the weights: labor (0.20) and nonlabor (0.80) because labor costs accounted for about 20% of the total cost throughout the study period.

by 4.1% and from 1988 to 1996 by 2.9% (China Statistical Yearbook, 1997). This suggests that measured grain output per mu at the national level was only slightly higher than at the (Jiangsu) farm-level from 1978 to 1996. Turning to Jiangsu’s farm-level input usage, labor inputs measured in man-days decreased sharply for all farming activities during the study period.20 For instance, annual labor usage in grain production decreased at a rate of 4.9% for the entire period, at a rate of 9.6% for the first period (i.e., 1978 –1988), and 2.4% for the second period

20

The labor input data in man-days is not reported here, but it can be provided upon request.

64

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

Table 4 Jiangsu’s TFPIs by activity, household data: 1978 – 1996 Year

Grain Indica rice Japonica rice Wheat Corn Soybeans Rapeseed Cotton Hogs Fish

1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996

100 120 100 114 122 128 141 148 165 142 140 141 121 119 137 168 182 179 160

100 96 91 94 101 118 117 115 122 114 106 109 96 101 98 110 145 128 113

100 107 114 111 119 147 166 158 182 159 149 146 135 164 166 195 217 202 172

100 130 97 109 115 108 119 128 134 119 110 92 94 82 117 105 111 128 126

100 127 109 150 152 140 147 179 195 174 186 177 128 153 164 180 212 218 152

100 146 138 229 251 236 198 258 301 237 256 240 197 216 224 238 239 259 324

100 125 93 146 174 138 148 127 137 135 111 99 131 112 117 120 130 123 105

100 119 77 91 82 117 101 78 99 101 101 100 107 115 80 98 129 120 108

100 101 99 98 99 90 88 91 80 93 95 81 83 87 84 90 92 81 82

100 107 112 119 120 151 140 173 172 154 155 123 130 134 126 121 136 111 116

Growth rates 1978 – 1996 2.4 1978 – 1987 4.9 1988 – 1996 4.0

1.1 3.0 2.7

3.3 6.6 4.5

0.0 2.1 3.5

2.5 6.3 1.6

3.2 9.4 2.8

 0.4 2.9 0.9

0.9  0.3 1.7

 1.0  1.9  0.6

0.3 6.1  2.5

Sources: Computed from farm cost survey data, Jiangsu Provincial Price Bureau, 1991 – 1997; computed from Tables 2 and 3. The growth rates were calculated with the following regression expression: ln( Y)=a + b * time.

(i.e., 1988– 1996).21 However, labor costs did not show any apparent decreasing trend due to an increase in the real wage rates (Appendix B). In fact, for most farming activities, we observe modest increases in labor costs over the study period. At the same time, there was a rapid increase in nonlabor costs associated with grain production in Jiangsu. In real terms, nonlabor costs increased much faster in the second period at a rate of 8.0%, compared to a rate of 3.3% in the first period (Appendix B). Table 3 reports unit input costs, based on the household data. From Table 3, we find that the total cost of grain production displayed an average 0.4% annual increase in the first period, and a much higher rate of 4.8% in the second

21 Taylor (1993) substantiates this significant decline in labor inputs. Taylor reports changes in labor use for selected crops from 1978 to 1985. Over this period, he reports that labor usage for rice declined by 22.1% per hectare, for wheat down by 52.7%, for soybeans down by 47.7%, and for corn down by 47.6%. Over the 1978 – 1985 time period, our household cost of production data show a 40% decline in labor usage in aggregate. For wheat, soybeans, and corn, our cost of production data record a smaller percentage decline than reported by Taylor, for the 1978 – 1985 time period.

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

65

period. This was due to the rapid increase in nonlabor costs, such as the cost of fertilizer. Unit fertilizer usage for grain maintained a growth rate of 5.2% annually during the 1990s, and for wheat in particular, this growth rate was as high as 11.8%. This indicates that the large increase in fertilizer usage pushed up grain production

Table 5 Total factor productivity of agriculture in Jiangsu, household data: 1978 – 1996 Year

Output index

1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 Growth rates 1978 – 1996 1978 – 1987 1988 – 1996

All agriculturea

Cropping sector b

Inputs

TFP index

Output index

Inputsc Weighted input index

Nonlabor input index

Labor input index

TFP index

Weighted input index

Nonlabor input index

Labor input index

100 121.0 110.6 123.9 130.0 137.3 145.1 148.2 165.8 153.3 150.5 145.8 140.0 142.4 156.5 162.9 216.5 226.5 221.1

100 105 111 108 106 107 111 109 109 113 116 121 125 123 123 121 141 149 160

100 109 116 113 117 123 129 134 133 144 156 164 173 173 174 173 198 229 250

100 106 119 111 104 100 106 97 98 100 99 106 110 105 104 99 135 131 146

100 116 99 115 122 129 131 137 153 136 130 121 112 116 127 135 154 153 138

100 120 111 123 128 137 143 163 177 185 224 169 161 165 169 188 258 242 220

100 110 115 115 118 122 128 141 138 150 183 186 180 178 179 166 205 232 239

100 105 110 107 105 105 108 109 108 112 120 124 123 121 118 114 135 141 147

100 106 116 109 100 96 99 93 91 94 90 96 98 94 87 84 113 107 120

100 115 101 115 122 126 128 135 144 137 140 126 118 122 130 134 156 150 140

3.4 4.9 6.4

2.0 0.8 3.6

4.4 3.5 5.3

1.0  1.1 4.3

1.4 4.1 2.9

4.2 6.5 3.5

4.5 4.0 3.2

1.7 0.8 2.2

 0.1  1.9 2.8

1.6 3.8 1.9

Sources: Computed from farm cost survey data, Jiangsu Provincial Price Bureau, 1991 – 1997; compiled from Tables 2 and 3. The growth rates were calculated with the following regression expression: ln(Y ) = a + b * time. a Fisheries were not included in the computation of the total input, output, and TFPI indexes. There are known difficulties associated with measuring fish output due to quality variation and the fishery data set was from a relatively small sample. We tried including the fishery sector and found similar results for the overall TFP. b Indexes of output and labor and nonlabor inputs for each crop are aggregated based on their shares in total sown areas; the weighted input index is computed with the weights: labor (0.35), land (0.36), and nonlabor (0.29). c The aggregated output and input indexes are computed on the basis of shares in gross value of agricultural output. For the aggregate input index, the cropping weights were used.

66

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

costs in the second period. This, in turn, implies that the observed output increases of farm products in the 1990s were largely caused by increased usage of nonlabor inputs. For the livestock sector, as represented by hogs and fisheries, production costs in the second period grew very slowly or declined (Table 3). The fall in production costs was accompanied by a decline in the value of output (Table 2). This means the estimated productivity gains for hogs and fisheries were both negative in the second period (see Table 4). Comparing TFPIs across all major farm activities in Jiangsu, we find that grain experienced one of the highest growth rates, especially during the first period. Except for wheat and cotton, the TFPI growth rates were all lower in the second period (Table 4). This led to a decrease of the estimated TFPI for the entire Jiangsu agricultural sector from an average annual growth rate of 3.8% in the first period to 1.9% in the second period (see Table 5 and Fig. 1). The estimated productivity of the cropping sector increased at a rate 4.1% per year for the first period and then slowed down to 2.9% for the second period. Estimated labor productivity did increase significantly and steadily over the study period. Grain output per labor day increased by an estimated 8.4% annually for whole period, with a 10.3% increase in the first period and a 6.2% increase during the second period. Similarly, the output of cotton per labor day enjoyed a growth rate of 4.0% annually during the whole period. The decline in labor usage led to a decrease in the share of labor in the total cost of grain production, from an average of 50.4% in the first period to 42.8% in the second period. At the same time, the nominal labor wage rate increased at a rate of 15.5% annually. This observation is consistent with the experience in other developing countries, in that labor productivity growth in agriculture has been greater than in other sectors of the economy during rapid economic development (Johnson, 1997). The measured improvement of labor productivity, using disaggregate data, indicates that the economy is following a normal growth pattern, and this is an important finding. The results discussed so far in this section were based on data from Jiangsu households. To serve as a check that Jiangsu may or may not be representative of Chinese agriculture, we compared Jiangsu aggregate provincial data with the national data. The provincial aggregate data were obtained from various issues of the Jiangsu Statistical Yearbook. The source of these data is the same as those published in the China Statistical Yearbook and used in our computation of the national TFPI. Using exactly the same approach as for the national approach, we computed the TFPI for Jiangsu based on the provincial aggregate data. Compared to the national results, our estimates show that Jiangsu experienced a similar TFP growth rate for both time periods (1978 – 1987 and 1988 – 1996)—see Table 6 and Fig. 1. In fact, the provincial aggregate data indicate that Jiangsu had a slightly higher TFP growth rate in the second period compared to the national figures. This is what we expected as Jiangsu is relatively advanced compared to other provinces in China and it is a leading agricultural producer. The estimated TFPI increased by 60% from 1990 to 1996 in Jiangsu (about 8% per year), higher than the 44% national average figure (about 6% per year), and much higher than the 19% computed from using household level data (about 3% per year). We found that the IGVAO and IGVCO

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

67

Table 6 Agricultural productivity in Jiangsu using provincial aggregate data, 1978 – 1996 Year

IGVAOa IGVCOb Input indexes

1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996

100 111 105 113 130 138 160 165 176 181 193 194 199 196 222 247 277 315 338

100 109 100 109 124 133 141 140 149 154 161 162 164 157 179 190 205 231 251

Growth rates 1978 – 1996 6.3 1978 – 1987 7.2 1988 – 1996 7.7

4.6 5.2 5.8

Weighted Labor TFP index c input index productivity Farm Effective Capital Current labor sown area inputs

100 104 103 104 98 98 94 90 87 85 85 89 90 91 89 85 83 81 80

100 99 98 100 100 100 100 100 99 99 98 98 97 96 97 95 93 94 94

100 111 111 112 118 121 126 135 136 146 154 161 151 149 155 158 163 171 175

100 105 103 103 110 109 110 109 113 115 121 124 130 136 143 146 157 170 174

100 103 102 103 103 103 102 101 101 102 103 106 106 107 109 107 109 111 112

100 107 102 108 133 141 171 184 203 212 226 219 220 215 250 290 332 390 422

100 108 103 109 126 134 157 163 174 178 187 183 187 183 204 230 255 283 301

 1.3  0.4  2.2 0.0  1.3  0.7

2.9 3.7 1.5

3.0 1.3 4.8

0.5 0.0 0.9

7.6 9.5 8.9

5.8 7.3 6.7

Sources: Computed based on Eq. (1) and data from various issues of Jiangsu Statistical Yearbook (1990 – 1997), Jiangsu Statistical Bureau. The growth rates were calculated with the following regression expression: ln(Y ) = a + b * time. a IGVAO is the index of the gross value of agricultural output. b IGVCO is the index of the gross value of crop output. c The weights are as follows: labor (0.35), capital (0.09), land (0.36), and current inputs (0.20).

for Jiangsu provincial data were similar to the national indexes. However, compared to the national figures, Jiangsu provincial data show a sharper decrease in farm labor usage and sown area. The provincial growth rates of capital and current inputs are also lower than the national average.

5. Conclusions Most previous studies of efficiency gains in China’s agriculture have used aggregate provincial or national data. The results of these past studies have not been called into question because they examined the immediate postreform period and it is generally accepted that this was a period of significant productivity gain that lasted

68

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

almost a decade. The largest sources of postreform efficiency gains were institutional change, and increases in chemical fertilizer application, increased mechanization (including electricity for pumping water and mechanized plowing), and technical change. In the subsequent decade, investment in China’s agriculture slowed considerably and actually fell in real terms over a large part of the 1985 –1995 period. When we take an aggregate approach to measuring productivity gain in this second decade (after reform) the results are striking. We find that estimates of annual efficiency gains continue to be very high, a somewhat implausible outcome. This puzzle provided motivation for us to measure China’s productivity growth using a disaggregate approach. By using national, provincial, and household data for China, we find that the three data sets result in agricultural productivity growth estimates that are relatively high for the 1978 –1987 time period, immediately following economic reform. The national data suggests productivity growth of 8.1% per year during this period compared to 7.3% for the Jiangsu provincial aggregate data and 3.8% for the Jiangsu household data. However, both aggregate data sets show much higher productivity growth from 1988 to 1996, compared to the results from the Jiangsu household level data. The estimated annual productivity growth rate for this period was 5.6% using national data and 6.7% using provincial data, versus 1.9% using the household data. The estimated TFP growth rate for the cropping sector was 2.9% per year from the 1988 to 1996 time period using Jiangsu’s household data. We believe the household results are more plausible because of reduced investment at the time and the fact that increases in farm product prices from the mid1980s were accompanied by rising farm input prices in China. In comparing the input, output, and TFP indexes, we find that the aggregate data likely overestimate labor usage, but at the same time, underestimate the large cost increases associated with nonlabor inputs. The relatively high growth rates of the aggregate national and provincial output indexes could be due to many factors. The exaggeration of livestock figures is clearly a problem with the aggregate data. A second major problem might relate to the inclusion of intermediate inputs in the aggregate GVAO. A third explanation is that the aggregate data overstate the total value of agricultural output, from the bottom up, due to political incentives to overestimate production. In addition, some important cost items might be left out of the aggregate data. These include indirect costs such as depreciation of fixed assets. Certainly, there are other reasons that might explain our findings and we cannot even rule out the possibility that the aggregate results are more accurate than the farm level results. Nevertheless, our results cast doubt on the efficacy of using an aggregate approach when studying agricultural productivity gains in China.

Acknowledgements The authors are grateful to two anonymous referees for their helpful comments and to Belton M. Fleisher for his valuable suggestions and editorial input.

Appendix A. Unit outputs for major farming activities in Jiangsu province: household data, 1978 – 1996

Year

Japonica rice

Hybrid rice

Wheat

Corn

Soybeans

Cotton

Rapeseed

Hogs (per head)

Fisha (per mu)

kg/mu Index kg/mu Index kg/mu Index kg/mu Index kg/mu Index kg/mu Index kg/mu Index kg/mu Index kg

Yuan

kg

Yuanb

251.5 275.5 264.0 271.0 244.0 319.0 331.0 324.5 343.0 319.0 325.5 330.5 322.0 344.0 373.5 395.2 393.4 405.9 431.8

90.0 94.3 92.8 99.8 102.0 98.2 98.0 114.1 100.9 102.9 113.9 110.3 107.4 108.3 104.9 91.3 95.2 92.5 89.2

435.0 521.7 523.4 551.9 543.8 485.3 478.5 657.7 566.6 700.9 890.7 822.2 694.8 696.3 675.5 613.7 945.3 771.8 673.4

369.8 232.7 289.0 329.9 327.4 705.4 556.0 720.5 789.0 787.0 1092 517.0 531.1 589.4 559.7 673.4 768.1 712.0 503.5

426.4 353.9 444.1 531.5 527.0 1224.2 1071.0 1792.2 1925.8 2411.5 3539.6 1303.4 1366.4 1413.7 1300.4 1779.8 2461.5 1771.4 1162.6

1.2 0.0 3.4 1.6 2.9  3.2

2.9 3.3  1.4

4.3 13.5  2.1

8.2 22.7  3.5

Growth ratesc 1978 – 1996 1978 – 1987 1988 – 1996

100.0 109.5 105.0 107.8 97.0 126.8 131.6 129.0 136.4 126.8 129.4 131.4 128.0 136.8 148.5 157.1 156.4 161.4 171.7

309.0 303.5 325.0 354.0 376.5 412.5 497.5 428.0 477.5 457.5 463.0 486.0 483.5 498.5 502.0 541.8 574.0 562.1 596.8

100.0 98.2 105.2 114.6 121.8 133.5 161.0 138.5 154.5 148.1 149.8 157.3 156.5 161.3 162.5 175.3 185.8 181.9 193.1

2.7 3.4 3.8

3.5 5.6 3.1

446.0 n/a 414.0 410.0 438.0 474.0 510.0 471.0 502.5 483.0 484.0 513.5 473.0 510.0 475.5 498.0 485.6 526.1 509.9

100.0 n/a 92.8 91.9 98.2 106.3 114.3 105.6 112.7 108.3 108.5 115.1 106.1 114.3 106.6 111.7 108.9 118.0 114.3

1.0 2.0 0.5

215.0 260.0 228.0 218.5 245.0 223.5 249.0 257.0 264.5 248.0 249.0 226.0 249.5 231.0 300.0 276.2 270.1 280.8 316.8

100.0 120.9 106.0 101.6 114.0 104.0 115.8 119.5 123.0 115.3 115.8 105.1 116.0 107.4 139.5 128.5 125.6 130.6 147.3

1.3 1.4 3.3

214.5 248.5 239.5 319.0 295.0 321.0 332.5 344.5 364.5 344.5 373.0 361.0 353.5 405.0 405.0 390.4 361.2 379.5 336.6

100.0 115.9 111.7 148.7 137.5 149.7 155.0 160.6 169.9 160.6 173.9 168.3 164.8 188.8 188.8 182.0 168.4 176.9 156.9

2.5 5.4  0.4

63.5 76.0 80.0 83.5 90.5 98.5 90.0 101.0 122.5 104.0 107.0 115.5 164.5 123.0 121.0 120.8 123.7 143.6 163.7

100.0 119.7 126.0 131.5 142.5 155.1 141.7 159.1 192.9 163.8 168.5 181.9 259.1 193.7 190.6 190.2 194.8 226.1 257.8

4.1 5.6 2.9

56.2 59.8 42.1 49.0 49.7 60.5 65.0 53.4 59.0 60.5 65.0 62.5 58.5 64.8 58.7 58.9 68.1 67.1 69.4

100.0 106.4 74.9 87.2 88.4 107.7 115.7 95.0 105.0 107.7 115.7 111.2 104.1 115.3 104.4 104.8 121.2 119.4 123.5

1.5 1.7 1.1

97.0 94.5 85.0 117.0 131.0 119.5 130.5 114.0 119.0 125.0 99.5 98.0 124.5 119.0 134.0 123.8 109.3 129.3 131.1

100.0 97.4 87.6 120.6 135.1 123.2 134.5 117.5 122.7 128.9 102.6 101.0 128.4 122.7 138.1 127.6 112.7 133.3 135.2

b

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996

Grain

Source: Jiangsu Provincial Price Bureau, 1992 – 1997. a

Calculated on the basis of 1 mu of water area. Real value in 1995 prices.

c

The growth rates were calculated with the following regression equation: ln(Y ) = a + b * time.

69

b

Year

Grain

Japonica rice

Non Labor Non labor cost labor cost cost 26 30 32 30 31 34 33 36 36 37 40 43 45 46 48 48 61 70 78

Growth rates 1978 – 1996 5.1 1978 – 1987 3.3 1988 – 1996 8.0

30 32 36 33 30 30 29 27 28 28 28 30 30 30 30 30 38 35 41

0.6  2.0 4.1

33 37 40 37 38 37 43 45 45 48 52 56 58 55 51 54 68 82 78

4.3 3.5 5.1

31 36 39 37 36 29 35 32 32 34 33 36 38 35 33 34 46 47 48

1.3  0.8 4.4

Hybrid rice

Wheat

Corn

Soybeans

Rapeseed

Cotton

Hogs

Fish

Non Labor Non Labor Non Labor Non Labor Non Labor Non Labor Non labor cost labor cost labor cost labor cost labor cost labor cost labor cost cost cost cost cost cost cost

Labor Non cost labor cost

Labor cost

36 34 36 37 37 39 38 40 43 47 55 56 57 61 55 61 80 92 98

24 28 32 32 29 30 30 35 31 35 32 38 38 35 25 27 47 39 36

190 93 70 183 160 332 272 363 300 446 601 113 147 183 149 307 345 350 107

5.5 2.9 7.5

37 34 36 36 34 30 32 31 32 35 34 36 35 37 36 29 40 38 46

0.7  1.3 2.3

25 30 32 30 31 33 34 38 36 37 39 40 44 44 46 45 48 53 65

3.9 3.7 5.1

27 30 36 31 28 26 27 24 23 22 22 24 24 23 23 23 32 29 34

 0.5  3.8 4.8

20 22 22 24 24 27 27 28 28 32 35 35 39 37 40 37 39 50 51

4.8 4.6 4.3

30 33 38 32 29 31 35 30 35 33 35 36 42 40 41 34 46 45 50

2.2 0.3 3.5

10 12 13 12 12 14 14 13 14 15 18 21 20 22 25 25 23 27 27

5.5 3.2 4.6

15 17 19 16 14 17 19 16 19 18 21 21 26 23 26 25 31 30 36

4.3 1.4 6.1

23 23 27 26 25 29 30 27 29 31 30 34 35 35 35 31 35 41 45

3.0 3.0 3.4

33 33 39 41 36 39 40 39 37 36 32 38 36 36 36 32 43 43 48

0.7 1.0 3.7

34 38 38 39 40 42 44 38 37 48 53 54 57 61 64 54 64 76 78

4.3 2.0 4.5

60 62 69 68 72 80 79 69 72 76 78 83 92 90 92 83 114 122 131

3.5 2.3 6.0

95 110 103 116 115 115 121 156 150 161 233 225 187 181 180 144 210 226 196

4.4 5.5  1.1

1.5 2.7 1.2

149 125 234 187 198 499 499 748 916 1325 2085 1023 990 969 985 1427 1872 1617 1090

13.9 26.2 0.7

2.4 15.8  2.1

Source: Jiangsu Provincial Price Bureau, 1991 – 1997. Note: All costs are in 1978 prices (wage rates). Nonlabor costs were deflated with the national general rural retail price index of industrial products; labor costs were deflated with the general retail price index. The growth rates were calculated with the following regression equation: ln( Y) = a + b * time.

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996

Labor cost

70

Appendix B. Agricultural labor and nonlabor input costs by activity in Jiangsu: household data, 1978 – 1996 (Yuan per mu or per head of livestock, 1978 prices)

C.A. Carter et al. / China Economic Review 14 (2003) 53–71

71

References Ball, V. E., Bureau, J. C., Nehring, R., & Somwaru, A. (1997). Agricultural productivity revisited. American Journal of Agricultural Economics, 79, 1045 – 1063. Carter, C. A. (1999). Research on institutional reform and agricultural productivity growth in China. (Chap. 11). In A. Watson, & C. Findlay (Eds.), Food security and economic reform: the challenges facing China’s grain marketing system ( pp. 202 – 219). London: Macmillan. Crook, F. W. (1998 July). Agricultural policies in 1998: stability and change. China: situation and outlook series WRS-98-3. Washington, DC: Economic Research Service, U.S. Department of Agriculture. Domar, E. (1961). On the measurement of technological change. Economic Journal, 71, 709 – 729. Domar, E. (1962). On total productivity and all that. Journal of Political Economy, 70, 597 – 608. Fuller, F., Hayes, D., & Smith, D. (2000). Reconciling Chinese meat production and consumption data. Economic Development and Cultural Change, 49, 23 – 43. Huang, J., & Rozelle, S. (1996). Technological change; rediscovering the engine of productivity growth in China’s rural economy. Journal of Development Economics, 49, 337 – 369. Jiangsu Statistical Bureau (1990 – 1997). Jiangsu Statistical Yearbook, Nanjing. Johnson, D. G. (1997). Richard T. Ely lecture: agriculture and the wealth of nations. American Economic Review, 97, 1 – 11. Lardy, N. R. (1994). China in the world economy. Washington, DC: Institute for International Economics. Lin, J. Y. (1992). Rural reforms and agricultural growth in China. American Economic Review, 82, 34 – 51. Rawski, T. G., & Mead, R. (1998). On the trail of China’s phantom farmers. World Development, 26, 767 – 781. Rawski, T. G., & Xiao, W. (2001). Roundtable on Chinese economic statistics: introduction. China Economic Review, 12, 298 – 302. Saywell, T. (1998, October 1). Steady as she goes. Far Eastern Economic Review, 70 – 72. State Statistical Bureau (2000). Statistical Yearbook of China. Beijing: China Statistical Publishing House. Stone, B. (1993). Basic agricultural technology under reform. (Chap. 9). In Y. Y. Kueh, & R. F. Ash (Eds.), Economic trends in Chinese agriculture: the impact of Post-Mao reforms ( pp. 311 – 359). New York: Oxford Univ. Press. Wen, G. J. (1993). Total factor productivity change in China’s farming sector: 1952 – 1989. Economic Development and Cultural Change, 42, 1 – 41. Wiens, T. B. (1982). Technological change. (Chap. 7). In R. Barker, & R. Sinha (Eds.), The Chinese agricultural economy ( pp. 99 – 120). Boulder, CO: Westview. Taylor, J. R. (1993). Rural employment trends and the legacy of surplus labor, 1978 – 1989. (Ch. 8). In Y. Y. Kueh, & R. F. Ash (Eds.), Economic trends in Chinese agriculture: the impact of Post-Mao reforms ( pp. 273 – 309). New York: Oxford Univ. Press.