Part-time farming trends in China: a comparison with the Japanese and Korean experience (1).

Zhou, Zhang-Yue ; Sumner, Daniel A. ; Lee, Hyunok 等

A better understanding of the development of part-time farming in China would be invaluable both for academic inquiry and policy formulation. This study, using farm-level survey data, examines the development of part-time farming in China. The two key questions examined are (1) how part-time farming evolves over time and (2) how part-time farming may be related to regional economic development. This study also highlights part-time farming in Japan and Korea and compares China's with them. Given the many similarities in agrarian arrangements between these three countries, it should be valuable to examine part-time farming and related policy issues in China with a reference to Japan and Korea.

1. Introduction

Since the Chinese government instituted a strict policy restricting rural labour movement in the early 1960s, there has been a huge labour force retained within the rural area that is "surplus" to farming (Mu 1989). (2) Currently, this labour force is estimated in the order of over 200 million. To sustain China's economic development, it is argued that this labour force needs to be absorbed into non-farming activities (Zhou, Dillon and Wan 1992; Johnson 1999).

Broadly, there are two major paths through which labour can be shifted out of farming activities: (1) leaving the farm completely and moving to other industries and (2) being farming based and where possible engaging in nonfarming activities. The latter has been often referred to as "part-time farming". Even though a large-scale relocation of farm labour may not be expected in any near future, China has already been experiencing part-time farming as a trend.

Hence, an understanding of the development of part-time farming in China would be invaluable both for academic inquiry and policy formulation. Unfortunately, despite the fact that there is a rich literature that addresses the agricultural labour issue in general, studies that examine the part-time farming issue in China remain scarce. A few such examples include Yang and Cai (1991), Tian and Shi (1991), Ye (1992), and Zhang, Chen and Wu (1996).

Yang and Cai (1991) suggest a model of decision-making on part-time farming behaviour and examine factors that affect part-time farming behaviour using macro-level data. Tian and Shi (1991) offer a broad discussion on the social and economic background of part-time transfer of agricultural labour and the relationships between part-time transfer, industrialisation, urbanisation, and agricultural development. They point out that part-time transfer can be used as a practical choice for transferring agricultural labour in China. Using cross-sectional farm survey data, Zhang et al. (1996) describe the part-time farming situation and the impact of part-time farming on farm households in ten provinces in 1994. Using 1986 survey data, Wan (1995) assesses rural-rural, rural-urban and urban-rural migration in China. Rozelle et al. (1999) examine rural-urban migration using cross-sectional data collected in 1996 from eight provinces in China.

Few have attempted to empirically assess the historical trend of part-time farming in China. This study, using ten-year farm-level survey data, examines the development of part-time farming in China. The two key research questions we try to answer are (1) how part-time farming evolves over time and (2) whether and how part-time farming may be related to regional economic development.

In addition, this study will also highlight part-time farming situations in Japan and Korea and compare them with China's. There are many similarities in agrarian arrangements, especially in land distribution methods, between Japan, Korea and China. Such agrarian arrangements have primarily contributed to the emergence of part-time farming in both Japan and Korea, and both governments have explored various policy measures to deal with part-time farming development. Thus, we may learn some lessons by examining these countries' experience and comparing their experience with the situation in China.

We discuss the factors that affect part-time farming in the next section. We then provide conceptual considerations and data description in Section 3. Section 4 reports and discusses the empirical results of part-time farming trends in China. In Section 5, we present cross-country comparisons based on historical data. We conclude with policy implications.

2. Factors Affecting Part-time Farming

Part-time farming can be found in many countries in the world (for examples, see Kada 1980; Sumner 1982; Jussaume 1991; Loyns and Kraut 1992; Tweeten et al. 1993; Ali 1995; Weiss 1997). Part-time farming is an important way for farmers to divert family labour to non-farming activities. There are some major factors that affect the allocation of on-farm and off-farm labour. (3)

Technical change. Capital input using technical change will substitute capital with labour. Output effect induced by the increase in the marginal product of labour input would positively affect demand for labour but with a quasi-fixed input, in this case, land, making the negative substitution effect greater. As a result, part-time transfer will take place when farmers try to best utilise their labour but cannot completely leave their farm due to farming needs.

Land system. In countries where land is limited and equally distributed and land transfer is restricted, part-time farming will evolve.

Relative return. It is common that farming has a relatively lower return, as often observed in disparity of rural and urban incomes. The desire to obtain a higher return and maximise family income is an important force driving farmers to take part in non-farming activities.

Traditional land ideology. In some countries, farmers are so attached to their land that they do not want to give it up. Where social security programs are lacking, some perceive their land as security or family insurance. This leads them to holding on to their land even if their non-farming income is very high.

Urbanisation and industrialisation. Rapid urbanisation and industrialisation has a higher capacity to absorb agricultural labour.

Dual social structure. Where rural and urban areas are separated by administrative forces, there is little chance for agricultural labour to leave farming completely and be engaged in urban industrial activities. They can only take part in non-farming activities within rural areas.

A combination of some of the above factors has induced a rapid development of part-time farming in both Japan and Korea. In Japan, around 50% of farm households in 1950 were full-time farm operators. Since then, there has been a marked shift away from full-time farming in line with Japan's period of rapid economic growth and industrial development. Increasing employment opportunities and higher incomes in non-farm sectors attracted a large number of the agricultural labour force to urban centres. Many farm households found that it was possible to earn a high wage from non-farm employment while continuing to hold onto their land and farm it on a part-time basis (Hayami 1975, p. 75; OECD 1995, pp. 11-15). Part-time farming now accounts for a substantial part of the Japanese agriculture. Of the 2.6 million commercial farm households in 1996, only 17% are classified as full-time farms, and the remaining 83% are part-time farms (4) (GOJ, 1998, p. 10).

Like Japan, Korea has also experienced a rapid emergence of part-time farming. The factors contributing to this rapid emergence are similar to those in Japan. Rapid industrialisation and improvement in productivity, regulations governing the sale and transfer of land and the traditional view of land as a family asset, all have led to small average farm size and income disparity between farming and other sectors (OECD 1999). In 1965 part-time farms accounted for only 9% of the farm households; in 1998 37% of the 1.4 million farm households were classified as part-time farms (5) (Ministry of Agriculture and Forestry 1999b, p. 41).

There are many similarities in agrarian and social arrangements between China, Japan and Korea, such as small but relatively evenly distributed landholdings, restrictions on land transfers, and the traditional view of land as a family asset. In addition, the strict rural-urban separation designed by policy has also been relaxed to some extent since the early 1980s. However, income disparity between agriculture and other sectors is still large, particularly in recent years as a result of rapid development in non-farming sectors and in urban areas following the economic reforms. All such factors favoured the emergence of part-time farming. While this trend of part-time farming has important implications in government policy formation, this issue has received little attention. This study attempts to fill this gap by using farm-level survey data to assess the part-time farming status in China and to compare it with that in Japan and Korea.

3. Conceptual Considerations and Data

3.1 What is Part-time Farming?

Part-time farming refers to "the practice whereby a farm worker or farm family earns income from both agricultural and non-agricultural employment or self-employment" (Jussaume 1991, p. 3). Part-time farming is not unique to modern times nor to the developed nations. "It is an option that agriculturally dependent households around the world have exercised for generations. Only the recognition of the importance of part-time farming is new" (Jussaume 1991, p. 3). While there is a consensus about the notion of part-time farming, the technical definition differs slightly by country. This study adopts the classification of part-time farming that has been used in Japan. (6)

In Japan, data collection on part-time farm households traces back to as early as 1906 (Institute of Developing Economies 1969, p. 114). In Japan part-time farming is defined as "very small farms and any farm in which one or more household members are engaged in jobs other than farming" (OECD 1995, pp. 14-15). (Korea also follows this definition; see OECD 1999, p. 35). More detailed classification includes (OECD 1995, p. 11): (7)

* Part-time farm household: farming household which has one or more household members engaged in jobs other than farming and where they have been employed for 30 days or more, or engaged in their own non-farm business from which they have earned 100,000 yen in 1990 (70,000 yen in 1980) or more, during a year.

* Part-time farm household Type I: farm household where less than 50% of income was earned in non-farm jobs.

* Part-time farm household Type II: farm household where more than 50% of income was earned from non-farm jobs.

* Full-time farm household: farm household which has no household member engaged in non-farm employment.

OECD also adopts this classification of part-time farming (Krasovec 1982, recited from Jussaume 1991, p. 48).

3.2 Data

Farm-level survey data are regularly collected by the Research Centre for Rural Economy (RCRE) of the Ministry of Agriculture in China. The survey first began in 1986 and was carried out annually except in 1992 and 1994. In each village, some 100 households were surveyed. The survey instruments have evolved over the years. Those used for 1986-91 were the same (with 312 variables), but they were expanded for the 1993 survey (with 394 variables) and further expanded for the 1995-97 surveys (with 439 variables). Data used in this study cover the period from 1986 to 1997.

For this study, three provinces that represent the different stages of economic development were selected. They include Guangdong which represents most developed, Hubei which represents medium developed and Yunnan which represents least developed. Three villages were then selected from each province according to the level of economic development as measured by per capita income. Brief background information on the selected provinces and villages is presented in Table A1.

In the farm household surveys by the RCRE, there are nine categories of major family business activities: l) cropping, 2) forestry, 3) animal husbandry, 4) fishery, 5) industry, 6) construction, 7) transportation, 8) tertiary services, and 9) others. Since the rural economic reforms, farms have been allowed to diversify their farm activities in any of these categories.

In this study, activities belonging to categories 1-4 are considered as farming and the rest as non-farming (i.e., categories 5-9). (8) Detailed data on labour-days used for each of the nine business activities are available from the household survey. To ensure the data accuracy, we used a number of variables to cross check the data. Whenever doubt arose on any observation, that observation was eliminated.

4. Empirical Results

Figure 1 shows the historical trend of part-time farming for the data combining all three provinces. Three observations stand out. First, for the studied period, part-time farming (both Type I and Type II) is significant, close to 50%. Second, the majority of part-time farming households falls in Type II. Third, part-time farming peaked in 1993 and it then slowed down.

[FIGURE 1 OMITTED]

The part-time farming trend broken down at the provincial level is shown in Figure 2. As expected, Guangdong province shows the highest part-timing farming trend, consistent with its economic development. This also is the region where the post-1993 downturn of part-time farming did not occur. Further, while the bulk of part-time farm households are Type II, the share of Type I part-time farm households has been steadily declining, reaching 2% in 1997. In the economically least developed Yunnan province, part-time farming development fluctuates most and has declined sharply (from 50% in 1993 to 32% in 1997). It also has a higher proportion of Type I part-time farm households. In the economically medium developed Hubei province, the part-time farming level is below that of Yunnan, the economically least developed region. This is due to the fact that one of the villages included in the analysis for Yunnan (Village 1) is in a wealthy area with a very high part-time farming level (see Table 1).

[FIGURE 2 OMITTED]

Using the data presented in Table 1, we generated a correlation coefficient matrix between the three measures of part-time farming, part-time, Type I and Type II to verify which kind of labour actually governs the pattern in part-time farming. Correlation coefficients are presented in Table 2. According to Table 2, there seems to be no distinct pattern between Type I and Type II part-time farming and between part-time and Type I part-time farming. At the village level, we obtained mixed - positive and negative - correlations between these two sets of relationships. However, the correlation between part-time and Type II tends to be very strong. The positive and strong correlation between Type II and part-time, along with mixed signs of correlation between Type I and part-time, further indicates that the major factor that governs the pattern in part-time farming is Type II part-time labour, not Type I labour.

Further examining the data in Table 1, it can be found that, in general, in economically more developed areas, e.g., in all three villages in Guangdong province and in Yunnan's Village 1, part-time farming is taking place at a much higher level (50% or higher) and is increasing. On the other hand, the level of part-time farming in medium or less developed areas is accordingly lower. This indicates that the level of local economic development is an important factor for the emergence of part-time farming in a region. It is interesting to note that the poorest village (Village 2 in Yunnan) once had a very high part-time farming level (65% in 1993) but dropped sharply in recent years (17% in 1997). This phenomenon may have been dictated by the availability of non-farm employment. It is possible that many farmers from this poor village are always looking for employment opportunities outside farming. Therefore, their non-farm activities also fluctuate with outside opportunities. This is consistent with the situation that prevailed in China during that time period.

In the early 1990s, there were increased numbers of farmers who travelled around the country to look for employment opportunities, and there were abundant reports about this (see, for example, Chen and Xu 1993; Zhu 1993; Wang 1994; Research Group 1995). As reflected in the results from this study (Figure 1), the part-time farming level increased and peaked in 1993. Then, as urban economic reforms went further, employment for urban dwellers became a problem. Many city governments imposed restrictions on employment for rural labourers (Cai and Chan 2000). Soon many of the rural labourers had to return to their land, and, as seen from Figure 1, part-time farming dropped. The reverse may also be true; reduced restrictions on farmers' employment in urban areas would result in increases in part-time farming. It is likely that government policy has an important bearing on the ups and downs of part-time farming, particularly, in the case of Village 2 of Yunnan.

In addition to the level of local economic development and government policy, there are a number of other factors that affect part-time farming. Figure 3 shows per capita arable land. The average land holding over the three provinces is about one mu (1/15 hectare; 1 hectare = 15 mu). Not unexpectedly, per capita arable land exerts a force on the development of part-time farming. Guangdong's higher level of part-time farming is in part related to the low per capita arable land. Figure 3 shows that per capita arable land declined over time; faster in more developed regions. If this trend continues, we will observe more part-time farming as per capita arable land further declines.

[FIGURE 3 OMITTED]

Figure 4 presents the ratio of non-farm earnings to farm earnings per labour day. In all provinces and years studied, non-farm earning has been higher than farm earning. This encourages employment away from the farm. Note that this ratio was largest in 1993 and the part-time farming level was also the peak during the same year (see Figure 1). Johnson (1999) noted that, "The margin between the return to farm and nonfarm labour must be large enough to induce a rapid transfer to nonfarm jobs". The results from this study tend to support Johnson's proposition. Figure 4 indicates that while the provinces differ in ratio by some margin these differences became pronounced in 1993. It is interesting to note that these differences are more distinct at the village level (Table A2 in the Appendix). However, by comparing Table A2 with the part-time trend at the village level (Table 1), it is not clear that higher part-time farming is always related to higher income ratios.

[FIGURE 4 OMITTED]

Besides these exogenous factors such as earning ratios, part-time farm decisions are also affected by the individual's qualification (suitability) and transaction costs. Whether an individual can start up his non-farming activities, knows which non-farming activities to work on, and can market his nonfarm products (or equivalently, if he looks for a job elsewhere, whether he dares to go beyond his village to look for a job, know where to look for a job, or, may be offered a job), would depend on his education level/labouring skills, and his ability to obtain information. Figures 5-8 show that in Guangdong there is a higher family expenditure on media items, number of radios, recorders and TVs and a lower percentage of rural labourers who are illiterate or semi-illiterate. (Also, the percentage of rural labourers who completed senior high school or above is highest (Table 3)). This indicates that the rural labourers in Guangdong would be better informed about the opportunities of non-farm work, which is consistent with high part-time farming in Guangdong.

[FIGURES 5-8 OMITTED]

Besides TVs and radios, telephones and computers have an important role in providing information. Ownership of telephones by rural households was not recorded in the survey until 1993, and the ownership of computers was not recorded until 1997. As expected, the ownership of telephones and computers is higher in a more economically affluent province. Village 3 in Guangdong has a high level of computer ownership (Table 4). Interestingly, it is in this village that the part-time farming level has increased most (Table 1).

To further investigate the relationships between the household characteristics and the level of part-time farming, we carried out a regression analysis to determine the factors that affect the level of part-time farming within a household. In this regression analysis, the dependent variable measures the level of household part-time farming, which is constructed as a ratio of non-farm work days to total work days. Based on our earlier discussion, we chose the following as our independent variables that explain the level of part-time farming. These variables are: 1) number of labourers in the family, 2) total area of arable land owned by the family, 3) number of televisions owned by the family, 4) urban per capita income of the province, 5) the wage disparity between nonfarm and farm earning, and 6) the level of education.

Urban per capita income at the provincial level measures the level of regional economic development. The annual income data are obtained from SS[B.sub.1] and SS[B.sub.2] (1996). The wage disparity measures at the village level and is calculated as the ratio of the daily wage of non-farm work to farm work (Table A2). Three binary variables are used to represent the level of education (there are four levels of education and the category of "illiterate and semi-illiterate" is used as a base). In addition to these variables, time binary variables are used to separate out any time-related effects (1986 is used as a base).

The regression results are presented in Table 5. As expected, number of labourers, number of TVs owned by a family and the urban per capita income in the province they live, all contribute positively to the family's participation in non-farm jobs. On the other hand, a family engages less in non-farm jobs with larger land holding. However, the sign of the wage disparity variable (the ratio of the daily wage of non-farm work to farm work) is contrary to expectations. (1) It may be noted that such conventional expectations are based on some assumptions such as complete labour mobility, availability of non-farming employment, and possession of capability to move. Not all such conditions are prevalent in China. Indeed, some situations are very peculiar to China, e.g., restrictions on labour mobility. (2) Further, Table A2 shows that higher income ratios are not necessarily associated with higher absolute earnings. That is, with very small absolute earnings in both farming and non-farming employment, a small increase in non-farming employment earning can result in a significant increase in the income ratio. This typically happens in those poorer villages as shown in Table A2. For labourers in poorer regions, constrained to a great extent by subsistence nature of farming activities -- e.g., foodgrain production first for family consumption, engaging in farming is relatively less risky, and labour may be allocated to first satisfy subsistence production. (3) It is also likely that non-farming employment opportunities are limited in such poorer regions. Thus, the existence of some peculiar situations prevailing in China, especially, the subsistence nature of production and lack of non-farming employment opportunities in some poorer regions, may offer some explanations about the contradictory result for the wage disparity variable. Nonetheless, rural wage disparity and its impact on other aspects of rural economy is a complicated but very important issue and further research should be rewarding.

Compared to the illiterate group, the regression results show that more education tends to have a negative effect on participation in non-farm jobs. These education results are less intuitive. The coefficients are not significant except for the category of junior high graduates. Conventional expectations would mean that a higher education level would have a positive effect on participation in non-farm jobs. Coincidentally, some other studies (Wan 1995; Meng 1996; and Rozelle et al. 1999), using various survey data, also find that the effect of education on migration is insignificant or even negative in China. While Rozelle et al. (1999) argue that, "Although descriptive statistics show migration rising with education level, it may be that education is correlated with other factors ... and is not significant in itself". Wan (1995) proposes that Western migration models should be modified substantially before they can be applied to China.

All the coefficients for the time dummies have negative signs. It is interesting to note that these negative effects are statistically very significant from 1993. This may be in part due to the shift in government policies related to further restricted urban employment opportunities of rural residents, which lends support to our earlier observations that policy environment affects part-time fanning development.

To sum up, our analyses based on household survey data suggest that for farmers to take part in non-farm activities, first, they must have the internal desire to do so, second, they must have the intellectual and financial abilities to do so, and third, there must be suitable external opportunities for them to do so. This indicates that farmers' intellectual and financial abilities are crucial. These enable farmers to gather and digest information, to have the courage to look for non-farming job opportunities and to have the minimal labouring skills to be engaged/employed in non-farm work. Local economic development status, conditions of infrastructure, availability of markets for non-farm products, and government policies are important external factors. Not all of these external factors were addressed in this study due to the lack of data. Nonetheless, local economic development and the policy environment are two important factors that affect part-time farming development in China.

5. Part-time Farming in China, Japan and Korea: Cross-country Comparison

In this section, we turn our attention to part-time farming in two other northeast Asian countries, including Japan and Korea. Although they are in different stages of economic development, Japan, Korea, and China have in common a number of issues related to the early stage of part-time farming development. They all share similar agrarian characteristics -- a small per capita land holding, income disparity between farm and non-farm sectors, industrialisation and rapid urbanisation. It will be thus instructive to examine their experience of part-time farming development and draw some lessons for China.

5.1 Economic Development and Part-time Farming

The experience from these three countries indicates that part-time farming is closely related to economic development. Table 6 shows the relationship between the relative contribution of agriculture to total GDP and part-time farming development. (9) While there are ups and downs in the level of part-time farming in all of the three countries, the general tendency is that part-time farming is on the increase over time.

Consider the times, the mid 1950s in Japan, the mid 1970s in Korea, and the late 1980s in China, when these three countries had a similar share of agricultural GDP out of total GDP (e.g., at around 24%). China has a higher level of part-time farming than does Korea but lower than Japan. However, China's Type II part-time farm households are at a higher level than both Japan and Korea, with Korea being the lowest. Note that Japan's part-time farming level has plateaued since 1975 (Table 6). Compared to Japan's part-time farming level at a similar percentage of agricultural GDP out of total GDP (e.g., at about 6%), Korea's is much lower and it would be interesting to keep an eye on the development of the part-time farming level in Korea.

5.2 Part-time Farming and Income Implications

Implications of part-time farming development on farm income seem to be significant. Table 7 clearly indicates that part-time farm households on average earn a much higher income than do full-time farm households. Total income differences between Types I and II part-time farms have been minimal in Japan, whereas these income differences are large in China (Table 7). Part-time Type I households in China have been earning only about one half or one third of those part-time Type II households and even less than those of full-time households. The relatively higher proportion of part-time Type I households but with very low income earning capacity in Yunnan's Villages 2 and 3 (see Table 1) may have led to this abnormal result. In addition, observations from these two villages accounted for as high as 28% of total observations out of the nine villages. Data on income for full-time and part-time households for Korea were not available. However, at the aggregate level, it is shown that non-farming income has increased rapidly during the past three decades and non-farming employment has been important for farmer income.

5.3 Factors Contributing to the Development of Part-time Farming

Given the similarities in many aspects between China, Japan and Korea, several factors that contributed to the development of part-time farming in these countries are also very similar. Common to these countries is the small per capita arable land area. In 1997, per capita arable land was 0.077 hectare for China, 0.0395 hectare for Japan, and 0.042 for Korea. (10) As seen from our data, in China regions with low per capita arable land are associated with higher part-time farming.

Income differential between farming and non-farming employment is another key factor. This factor played a crucial role in "enticing" rural workers to be engaged in non-farming activities in Japan and Korea (Nakayasu 1991; Chung 1993). The same was observed for China as seen in this study. Rapid industrialisation and economic development (since the 1950s in Japan and since the 1960s in Korea) have created non-farming opportunities for rural labourers. In China, rapid economic development in the past two decades (since 1980) also generated many more employment opportunities. However, as discussed earlier, the possibility of rural residents taking these employment opportunities depends on government rural/urban policies.

In both Japan and Korea, government policies have encouraged farmers to be engaged in non-farm employment. We will further elaborate upon this in the next section. In China, however, government policy on this front has generally been restrictive. Before the economic reforms took place in the early 1980s, farmers in China certainly had the desire to do non-farm work. Their ability to do so was, however, severely restricted by government policies. At most, farmers in some relatively developed areas (e.g., in southern Jiangsu, coastal Guangdong) could run some township and village enterprises.

5.4 Government Policy Responses

In the latter half of the 1950s, Japan's economy began to develop rapidly, expanding the labour market and offering farmers opportunities for non-farm employment. Post World War II economic growth saw further development in the heavy and chemical industries, thus concentrating the employed population in certain regions. Over-concentration and depopulation due to regional imbalances in the general population and the employed population became a social issue in the 1970s. The Japanese government responded to this by devising policies that induce industry to disperse to local regions. Some manufacturing companies moved to areas where labour and land were cheaper. This expanded off-farm employment opportunities within commuting distance (Nakayasu 1991). Table 6 shows that the part-time farming level increased by 18 percentage points between 1970 and 1980.

In Korea, rural to urban migration has been encouraged by the government. By 1968, farm population began to decrease as a result of this migration and industrialisation and economic growth. During 1965-90, farm population decreased slowly, by 3.5% annually. Korea's major industries have been built near large urban centres and port cities, thus concentrating new jobs in only a few areas. Rural young people migrated to these areas for jobs, leaving behind the elders and those who have less ability for non-farm work. There has been concern that this drained too many able labourers away from farm production. In addition, the over-concentration of industrialisation also became a social issue. In 1985, the Korean government introduced a rural industrialisation program. This increased opportunities for non-farm jobs in rural areas and has led to an increase in the part-time farming level (see Table 6).

While there are no restrictions for rural people to migrate to urban areas, the Korean government has in recent years implemented various incentive systems to encourage people to stay in rural areas in order to avoid overcrowding in a few metropolitan areas. For example, many universities allocate a special quota for applicants from rural areas, thus making their admittance easier. As a result, this weakens their parents' desire to move to the urban area for the purpose of their children's education.

In China's earlier years of the communist government's rule, 1949-57, there was little restriction on rural-urban migration. In fact, during the Great Leap Forward period (1958-59), many rural labourers moved from the rural to urban areas. During 1960-62 when the country experienced extreme economic difficulties, many urban residents were sent to the rural areas to ease the government's supply burden to the urban systems. Since then, until the late 1970s, a strict rural-urban separation policy was used. During this time period, farmers were largely confined to the limited land. Since the early 1980s, nationwide economic reforms have been under way. Large-scale transfer of farm labour to non-farming activities began in the mid-1980s when rural township enterprises developed rapidly (Zhou, Dillon and Wan 1992). Interregional movement of rural labour started in the late 1980s. Following Deng Xiao Ping's visit to China's south in early 1992, large-scale interregional movement of rural labour took place and peaked in 1993. It was reported that in 1993 there were over 60 million rural labourers wandering around the country looking for employment (Zhu 1993, Research Group 1995).

Not all rural labourers could find a job in the urban system, nor did the urban system have sufficient employment opportunities to often With such a large number of unemployed rural workers in urban areas, there were problems in passenger transportation and social problems in the cities. The government responded to this situation by restricting rural workers' employment in the urban areas. As a result, the part-time farming level has dropped since 1995 (see Figure 1). Compared to the Japanese and Korean experiences, in China there were few government initiatives to create non-fanning employment within the rural areas. At times, the government has relied on farmers' own initiatives to set up their township and village enterprises to create non-farming jobs.

5. S Substitution between Farm Inputs

Our discussion on farm labour indicates that labour input use is declining in farming. This raises an interesting question that relates to farm input substitution and output production. In all three countries, the total farm labour force was reduced, by the largest percentage in Korea (-28%) and the smallest (-3%) in China (Table 8). Table 8 also shows that agricultural production index and food-stuff production index were increased in both China and Korea, whereas in Japan they dropped by 6 per cent compared to 1990. Data in Table 8 clearly suggest the possibility of input substitution. While fertiliser use was reduced in both Japan and Korea, there was an increase in the use of farm machinery. Japan has the highest per hectare machinery. Japan's arable land is 5% of China's, yet Japan's total number of agricultural tractors is three times and harvester-threshers is ten times China's. However, China's higher production index seems to be due largely to the rapidly increased use of fertilisers. Given the very low mechanisation level compared to Japan and Korea, if China were to take a similar path, it is likely that there will be a further reduction in Chinese farm labour in the future.

6. Concluding Remarks

In this study we assess the part-time farming status in China using farm-level survey data. Given that our sample contains only three provinces, it may be hard to draw from our study some broad conclusions that may apply to the whole country. Nonetheless, our study may still provide useful information since our analysis is based on more than 8000 individual observations that vary widely in economic classes and labour situations.

Our analyses indicate that part-time farming is significant in the regions of China examined. Overall, the part-time farming level was increasing over the period 1986-97. It peaked in 1993 but has slowed since then. Part-time farm households now account for more than 40% of total farm households in the studied regions.

It seems that the local economic development level is a major factor for the development of part-time farming in a region. In economically more developed regions, part-time farming is developing fast, at a much higher level, and increasing relatively steadily. In economically less developed regions, part-time farming development is relatively slower and fluctuates.

Government policy also has an important bearing on the development of part-time farming in China. With reduced restrictions on farmers' employment in non-farming activities or in urban areas, part-time farming increases; otherwise, many farm labourers have to return to or stay on their land. Other major factors that affect the development of part-time farming in China include the very small per capita arable land, income differences between farming and non-farming employment, farmers' education level/labouring skills and their ability to obtain and digest information.

The development of part-time farming in our data for China seems to have halted since the mid-1990s, especially in poor areas. It has been widely recognised that this slowdown in part-time farming development has resulted in lower income in the rural areas, and this has partly contributed to the sluggish market in recent years. However, following other countries' examples, such as Japan, there will be a further shift in the labour force from farming to non-farming. How to make this shift is an important area that calls for further investigation. Given the relatively more important role of public policy in China, this transition can be significantly aided by prudent policy planning.

Future policy can be directed toward carrying out needed institutional reforms, especially removing restrictions on rural people's employment in the urban areas. Flexibility in rural part-time jobs will reduce the number of people adding to urban congestion while also increasing rural incomes. Carefully conceived policy measures are needed to increase the opportunities of non-farming employment for rural people within the rural areas. Such policy measures will help China to alleviate/avoid over-concentration of industrialisation and population and other associated social problems as happened in Japan in the 1970s and in Korea in the 1980s. The availability and improvement of basic infrastructure in rural areas (e.g., road, water, electricity, and telecommunication facilities) are also essential.

Part-time farming is a complex issue. This study is only a modest attempt to address a very small number of questions. Many more questions remain to be explored. Some may include:

* To what extent has traditional land ideology affected the development of part-time farming in China?

* How fast is China's part-time farming likely to develop in the near future?

* How has part-time farming affected farmers' income in China?

* How has part-time farming development affected the life of rural people and the development of the rural community?

* How has part-time farming affected agricultural production, and how will it affect agricultural production in the future?

* What are the likely impacts of part-time farming on the preservation of rural traditions and institutions?

* What are the key institutional reforms that are needed to facilitate the development of part-time farming in China?

Jussaume (1991, p. xiv) notes, "A great deal about part-time farming remains unknown to the scientific community". This is especially true for China. This fact encourages increased research attention to the trends and development of part-time farming in China. This is an issue that China will have to face and that will affect China's long-term sustainable economic development.

Table A1:
Per Capita Rural Income at the Provincial, City/County,
and Village Level in 1997
 Brief Description of the
Province City/County Village Region

Hubei Hanyang Village 1 A relatively developed
([yen] 2102) ([yen] 3189) ([yen] 3304) region within the
 province with good
 transportation
 Hanchuan Village 2 Similar to Hanyang but
 ([yen] 2424) ([yen] 2938) slightly less developed
 Changyang Village 3 A hilly area with a less
 ([yen] 1636) ([yen] 1784) developed economy and
 market

Guangdong Dongguan Village 1 Highly industrialised
([yen] 3468) ([yen] 5021) ([yen] 10667) area close to Guangzhou
 and Hong Kong
 Dianbai Village 2 Relatively less developed
 (n.a.) ([yen] 2411) area within Guangdong
 province
 Wuchuan Village 3 Similar to Dianbai but in
 (n.a.) ([yen] 2948) a coastal area

Yunnan Yuxi Village 1 A wealthier area with
([yen] 1375) ([yen] 3248) ([yen] 6442) much injection from a
 highly developed
 tobacco industry
 Lanchang Village 2 Very remote and
 ([yen] 578) ([yen] 719) mountainous area, one
 of the poorest regions
 Dali Village 3 Very remote and
 ([yen] 2279) ([yen] 2114) mountainous area, one
 of the poorer regions
 with tourist resources

Note: National level per capita income of rural population in 1997 was
[yen] 2090. At the city/county level, there are no systematic data
available for per capita income of rural population. Figures given here
are from various sources. Village level per capita income was calculated
from the village sample survey data.

Sources: Ministry of Agriculture 1997, p. 433.
Hubei Statistical Bureau 1999, pp. 391-392.
Yunnan Statistical Bureau 1998, pp. 489-490.
[SSB.sub.1], 1998, p. 346.
Table A2:
Ratio of Per Labour-day Income between Non-farming
and Farming Employment

 Hubei Guangdong

Year Village Village Village Village Village Village
 1 2 3 1 2 3

Per labour-day income from farming employment (yuan)

1986 5 5 3 5 4 7
1987 6 6 3 6 5 6
1988 7 6 3 9 7 8
1989 8 7 3 7 10 9
1990 7 8 4 15 10 8
1991 6 7 3 15 10 7
1993 7 8 5 13 17 11
1995 19 15 7 70 18 18
1996 14 14 6 66 16 18
1997 18 13 8 44 14 14

Per labour-day income from non-farming employment (yuan)

1986 9 5 4 7 18 7
1987 14 5 2 7 19 8
1988 17 6 6 10 24 8
1989 18 7 7 18 15 8
1990 15 7 7 19 16 9
1991 12 7 8 28 18 9
1993 17 7 7 72 37 24
1995 32 17 13 60 49 24
1996 32 18 16 66 50 20
1997 20 26 13 66 41 21

Ratio of per labour-day income between non-farming and farming
employment

1986 1.8 1.0 1.5 1.4 4.3 1.1
1987 2.2 0.9 0.6 1.2 3.9 1.4
1988 2.6 1.0 2.0 1.1 3.5 1.0
1989 2.3 1.0 2.3 2.6 1.5 0.9
1990 2.1 0.9 1.8 1.3 1.6 1.1
1991 2.0 1.0 2.7 1.9 1.8 1.3
1993 2.4 0.9 1.4 5.5 2.2 2.2
1995 1.7 1.1 1.9 0.9 2.7 1.3
1996 2.3 1.3 2.7 1.0 3.1 1.1
1997 1.1 2.0 1.6 1.5 2.9 1.5

 Yunnan

Year Village Village Village
 1 2 3

Per labour-day income from farming employment (yuan)

1986 2 1 2
1987 5 1 3
1988 8 1 3
1989 10 2 2
1990 11 2 3
1991 10 2 3
1993 9 3 4
1995 16 5 8
1996 22 4 8
1997 16 4 9

Per labour-day income from non-farming employment (yuan)

1986 5 3 3
1987 7 2 3
1988 8 6 8
1989 6 6 6
1990 11 5 4
1991 17 6 5
1993 26 11 6
1995 29 8 19
1996 40 7 14
1997 46 3 19

Ratio of per labour-day income between non-farming and farming
employment

1986 2.3 3.3 1.5
1987 1.4 1.8 1.2
1988 1.1 4.0 3.0
1989 0.6 3.0 3.0
1990 1.0 2.5 1.3
1991 1.7 3.0 1.7
1993 2.9 3.7 1.5
1995 1.8 1.6 2.4
1996 1.8 1.8 1.8
1997 2.9 0.8 2.1

Source: calculated from the farm survey data.
Table 1:
Part-time Farming Trends in China at the Village Level

 1986 1987 1988 1989 1991 1990 1993 1995

 Hubei Province, Village 1

Full-time 82 83 76 77 . 73 57 62
Part-time 18 17 24 23 . 27 43 38
 Type I 4 1 4 1 . 1 0 4
 Type II 14 16 20 22 . 26 43 35

 Hubei Province, Village 2

Full-time 44 41 58 50 49 27 45 69
Part-time 56 59 42 50 51 73 55 31
 Type I 12 13 4 10 12 4 12 5
 Type II 45 46 38 40 39 69 43 26

 Hubei Province, Village 3

Full-time 93 84 92 92 89 89 85 79
Part-time 7 16 8 8 11 11 15 21
 Type I 2 10 3 2 2 3 3 8
 Type II 5 6 6 7 9 8 12 13

 Guangdong Province Village 1

Full-time 70 63 54 49 40 42 51 51
Part-time 30 38 46 51 60 58 49 49
 Type I 8 10 17 10 11 9 0 0
 Type II 23 27 29 41 49 49 49 49

 Guangdong Province, Village 2

Full-time 42 35 40 45 41 39 45 34
Part-time 58 65 60 55 59 61 55 66
 Type I 3 2 2 4 4 1 3 4
 Type II 55 63 57 51 55 60 52 62

 Guangdong Province, Village 3

Full-time 58 54 43 38 28 26 24 13
Part-time 42 46 57 62 72 74 76 87
 Type I 2 5 3 9 6 2 3 7
 Type II 39 41 53 53 66 72 73 80

 Yunnan Province, Village 1

Full-time 50 57 41 49 46 48 38 41
Part-time 50 43 59 51 54 53 62 59
 Type I 7 6 6 4 4 1 0 0
 Type II 43 37 53 47 51 51 62 59

 Yunnan Province, Village 2

Full-time 79 85 78 73 67 55 35 49
Part-time 21 15 22 27 33 45 65 51
 Type I 14 10 14 11 11 26 22 41
 Type II 7 5 8 16 21 19 43 10

 Yunnan Province, Village 3

Full-time 65 67 74 75 84 80 76 78
Part-time 35 33 26 25 16 20 24 22
 Type I 17 15 13 9 7 8 13 14
 Type II 17 18 14 16 10 12 11 8

 1996 1997

 Hubei
 Province,
 Village 1

Full-time 78 75
Part-time 22 25
 Type I 2 6
 Type II 20 19

 Hubei
 Province,
 Village 2

Full-time 63 56
Part-time 37 44
 Type I 2 8
 Type II 36 36

 Hubei
 Province,
 Village 3

Full-time 78 82
Part-time 22 18
 Type I 2 6
 Type II 20 12

 Guangdong
 Province
 Village 1

Full-time 50 51
Part-time 50 49
 Type I 1 1
 Type II 49 48

 Guangdong
 Province,
 Village 2

Full-time 35 46
Part-time 65 54
 Type I 7 3
 Type II 58 51

 Guangdong
 Province,
 Village 3

Full-time 16 10
Part-time 84 90
 Type I 4 2
 Type II 80 87

 Yunnan
 Province,
 Village 1

Full-time 43 35
Part-time 57 65
 Type I 1 4
 Type II 56 61

 Yunnan
 Province,
 Village 2

Full-time 61 83
Part-time 39 17
 Type I 31 14
 Type II 8 3

 Yunnan
 Province,
 Village 3

Full-time 80 84
Part-time 20 16
 Type I 4 4
 Type II 15 12

Source: calculated from the farm survey data.
Table 2:
Correlation Coefficient Matrix between Three Measures of
Part-Time Farming

 Part-Time Type I Type II

Hubei
 Village 1 Part-Time 1
 Type I -0.18 1
 Type II 0.98 -0.37 1
 Village 2 Part-Time 1
 Type I 0.38 1
 Type II 0.94 0.03 1
 Village 3 Part-Time 1
 Type I 0.51 1
 Type II 0.83 -0.05 1

Guangdong
 Village 1 Part-Time 1
 Type I -0.05 1
 Type II 0.83 -0.59 1
 Village 2 Part-Time 1
 Type I 0.21 1
 Type II 0.94 -0.14 1
 Village 3 Part-Time 1
 Type I 0.02 1
 Type II 0.99 -0.10 1

Yunnan
 Village 1 Part-Time 1
 Type I -0.48 1
 Type II 0.96 -0.70 1
 Village 2 Part-Time 1
 Type I 0.69 1
 Type II 0.78 0.08 1
 Village 3 Part-Time 1
 Type I 0.84 1
 Type II 0.71 0.23 1

Source: calculated from the farm survey data.
Table 3:
Education of Rural Labourers at the Village Level

Province Village 1986 1987 1988 1989 1990

 Illiterate and Semi-Illiterate
 Rural Labourers (%)

 1 12 10 11 9 10
Hubei 2 25 25 22 22 20
 3 15 14 12 11 13
 1 10 10 11 13 13
Guang- 2 40 38 37 36 34
 dong 3 35 33 32 30 30
 1 16 14 11 14 12
Yunnan 2 90 92 89 87 88
 3 43 29 35 30 32

 Rural Labourers Who Completed
 Primary School (%)

 1 47 48 44 47 45
Hubei 2 32 31 31 33 32
 3 44 43 45 43 47
 1 65 66 65 63 61
Guang- 2 33 35 36 37 36
 dong 3 34 37 39 40 38
 1 56 55 55 53 56
Yunnan 2 9 7 10 10 10
 3 35 47 45 48 47

 Rural Labourers Who Completed
 Junior High School (%)

 1 31 32 33 32 33
Hubei 2 32 33 35 33 37
 3 34 37 36 40 37
 1 21 22 22 22 24
Guang- 2 23 24 24 25 25
 dong 3 25 25 23 24 26
 1 27 27 30 29 27
Yunnan 2 1 1 2 2 2
 3 20 23 18 20 19

 Rural Labourers Who Completed
 Senior High School or above (%)

 1 10 9 12 12 13
Hubei 2 11 11 12 12 11
 3 7 6 6 6 3
 1 3 1 2 3 3
Guang- 2 4 4 3 3 4
 dong 3 6 6 6 7 6
 1 1 3 4 5 5
Yunnan 2 0 0 0 0 0
 3 2 2 2 2 2

Province Village 1991 1993 1995 1996 1997

 Illiterate and Semi-Illiterate
 Rural Labourers (%)

 1 11 11 8 6 10
Hubei 2 20 12 18 18 16
 3 10 17 23 24 23
 1 11 5 4 3 3
Guang- 2 34 24 25 29 28
 dong 3 27 26 21 18 16
 1 11 10 13 9 9
Yunnan 2 87 85 84 85 81
 3 31 30 26 24 16

 Rural Labourers Who Completed
 Primary School (%)

 1 43 42 43 47 47
Hubei 2 33 30 34 30 32
 3 51 45 39 35 38
 1 58 61 60 56 52
Guang- 2 36 39 41 37 40
 dong 3 38 39 41 42 43
 1 56 49 46 49 47
Yunnan 2 10 14 13 13 17
 3 46 45 50 50 58

 Rural Labourers Who Completed
 Junior High School (%)

 1 33 41 43 41 37
Hubei 2 36 49 40 43 41
 3 35 33 33 36 35
 1 26 28 29 35 37
Guang- 2 26 30 28 28 26
 dong 3 28 29 30 32 30
 1 28 37 36 33 35
Yunnan 2 3 1 2 2 2
 3 21 23 23 25 24

 Rural Labourers Who Completed
 Senior High School or above (%)

 1 13 6 6 6 6
Hubei 2 11 9 8 8 11
 3 3 5 5 4 4
 1 5 7 7 7 8
Guang- 2 4 7 5 6 6
 dong 3 7 7 8 8 12
 1 5 3 5 8 8
Yunnan 2 0 0 0 0 0
 3 2 2 1 1 2

Source: calculated from the farm survey data.
Table 4: Possession of Telephones and Computers at the Village Level

 Hubei Guangdong

 Village Village Village Village Village
 1 2 3 1 2

Telephone
1993
No. of households 60 60 60 100 100
No. of telephones 1 0 0 31 2
Average per household 0.02 0 0 0.31 0.02

1997
No. of households 60 60 60 100 80
No. of telephones 18 9 5 97 7
Average per household 0.30 0.15 0.08 0.97 0.09

Computer
1997
No. of households 60 60 60 100 80
No. of computers 0 0 0 3 0
Average per household 0 0 0 .03 0

 Guang- Yunnan
 dong

 Village Village Village Village
 3 1 2 3

Telephone
1993
No. of households 100 100 100 100
No. of telephones 11 15 0 0
Average per household 0.11 0.15 0 0

1997
No. of households 100 100 100 100
No. of telephones 33 54 0 0
Average per household 0.33 0.54 0 0

Computer
1997
No. of households 100 100 100 100
No. of computers 20 0 0 0
Average per household 0.20 0 0 0

Source: calculated from the farm survey data.
Table 5:
Regression Results

Independent Variable Coefficient T ratio

Intercept 0.214 11.04
Number of Labourers 0.009 4.00
Land Holding (mu) -0.010 -16.04
Number of TVs 0.049 7.72
Urban Per Capita Income (000 yuan) 0.085 22.52
Wage Disparity -0.014 -3.50
Education Dummies
 Primary School -0.009 -0.59
 Junior High -0.063 -4.54
 Senior High and Above -0.020 -1.35
Year Dummies
 1987 -0.009 -0.61
 1988 -0.010 -0.65
 1989 -0.042 -2.84
 1990 -0.027 -1.71
 1991 -0.029 -1.86
 1993 -0.122 -6.70
 1995 -0.297 -12.98
 1996 -0.375 -15.07
 1997 -0.417 -15.83

Note: Dependent variable = days of non-farm work/total work days; R
square=0.151; sample size=8366. A linear regression model is adopted
and the estimates uses the ordinary least squares method.

Source: calculated from the farm survey data.
Table 6:
Agricultural GDP and Part-time Farming

Year Agricultural Full-time Part-time Farming Households
 GDP out of Farming Total Type I Type II
 Total GDP Households % % %
 % %

 China

1986 27.1 65 35 8 27
1987 26.8 63 37 8 29
1988 25.7 63 37 7 30
1989 25.0 62 38 7 31
1990 27.1 55 45 7 37
1991 24.5 53 47 6 41
1993 19.9 49 51 6 45
1995 20.5 51 49 10 39
1996 20.4 54 46 7 39
1997 19.1 57 43 5 38

 Japan

1946 n.a. 54 46 29 17
1950 n.a. 50 50 28 22
1955 22.8 35 65 38 28
1960 14.6 34 66 34 32
1965 11.2 21 79 37 42
1970 6.1 16 84 34 51
1975 5.5 12 88 25 62
1980 3.7 13 87 21 65
1985 3.2 14 86 18 68
1990 2.5 12 88(65) 14 52
1995 1.9 12 88(65) 14 50
1996 1.9 13 87(64) 13 51

 Korea

1965 n.a. 91 9 n.a. n.a.
1970 26.4 69 33 20 13
1975 24.7 81 19 13 7
1980 14.4 76 24 14 10
1985 12.8 79 21 9 12
1990 9.0 60 40 22 18
1995 6.5 57 43 18 25
1996 6.3 56 44 16 27
1997 5.7 59 41 14 27
1998 4.9 63 37 13 24

Sources: Ministry of Agriculture and Forestry 1999a, pp. 41,426.
Government of Japan, Japan Statistical Yearbook, various issues.
SS[B.sub.1], China Statistical Yearbook, various issues.
Calculated from the farm survey data.
Table 7:
Changes in Income Structure of Farm Households in
China, Japan and Korea

 China (yuan) **

 1986 1989 1993 1997

Total farm household income 4059 6775 13896 20811
 Farming income 1114 1685 1805 3045
 % of total income 27 25 13 15
 Non-farming income * 2945 5090 12091 17766
 % of total income 73 75 87 86

Full-time: Total household income 3464 5981 8761 14888
 Farming income 1190 1703 2128 3185
 % of total income 34 28 24 21
 Non-farming income 2274 4278 6633 11703
 % of total income 66 72 75 79

Part-time I: Total household income 2445 4073 4168 10626
 Farming income 1531 2487 2652 8877
 % of total income 63 61 64 84
 Non-farming income 914 1586 1516 1749
 % of total income 37 39 37 17

Part-time II: Total household income 5949 8955 20911 30982
 Farming income 813 1473 1331 2044
 % of total income 14 16 6 7
 Non-farming income 5136 7482 19580 28938
 % of total income 87 84 93 93

 Japan (1000 yen)

 1970 1980 1991

Total farm household income 1592 5594 8738
 Farming income 508 952 1120
 % of total income 32 17 13
 Non-farming income * 1084 4642 7618
 % of total income 68 83 87

Full-time: Total household income 1374 4123 5543
 Farming income 982 2371 2771
 % of total income 71 58 50
 Non-farming income 392 1752 2772
 % of total income 29 42 50

Part-time I: Total household income 1619 5827 9582
 Farming income 974 3166 5444
 % of total income 60 54 57
 Non-farming income 645 2661 4138
 % of total income 40 46 43

Part-time II: Total household income 1665 5926 9462
 Farming income 232 437 475
 % of total income 14 7 5
 Non-farming income 1433 5489 8987
 % of total income 86 93 95

 Korea (1000 won)

 1970 1980 1990 1997

Total farm household income 256 2693 11026 23488
 Farming income 94 1755 6264 10204
 % of total income 76 65 57 43
 Non-farming income * 62 938 4762 13284
 % of total income 24 35 43 57

Full-time: Total household income
 Farming income
 % of total income
 Non-farming income
 % of total income

Part-time I: Total household income
 Farming income
 % of total income
 Non-farming income
 % of total income

Part-time II: Total household income
 Farming income
 % of total income
 Non-farming income
 % of total income

* Non-farming income refers to any income other than from farming,
including transfer payments.

** Data for China are aggregates of the three provinces included in
this study. Corresponding data for the country as a whole are not
available.

Sources: China, calculated from the survey data.

Japan, OECD (1995), Agricultural Policy Reform and Adjustment in Japan,
p. 20.

Korea, Ministry of Agriculture and Forestry (1999a), Major Statistics
of Agriculture and Forestry, p. 104.
Table 8:
Agricultural Production Index and Farm Inputs Use in
China, Japan and Korea

 China Japan

 1990 1997 1997/ 1990
 1990
 (%)

Total population (000 1155305 1244202 7.7 123537
 persons)
Farm population (000 834636 853439 2.3 8893
 persons)
Farm labour (000 persons) 333364 324349 -2.7 4270
Arable land per farm 0.287 0.293 2.0 1.23
 labour (ha)

Agricultural production 153
 index (1989-91 = 100)
Food stuff production 156
 index (1989-91 = 100)

Total fertiliser 25903 39809 54 1839
 consumption (000 tonnes)
 Per hectare (tonne) 0.27 0.42 55 0.35
Total agricultural 824106 703117 -15 2142210
 tractors (unit)
 Per hectare (unit) 0.0086 0.0074 -14 0.41
Total harvester-threshers 38719 114000 194 1214900
 (unit)
 Per hectare (unit) 0.0004 0.0012 197 0.23

 Japan Korea

 1997 1997/ 1990 1997 1997/
 1990 1990
 (%) (%)

Total population (000 126038 2.0 42869 45731 6.7
 persons)
Farm population (000 6017 -32.3 6917 4817 -30.4
 persons)
Farm labour (000 persons) 3490 -18.3 3237 2324 -28.2
Arable land per farm 1.42 15.5 0.652 0.828 27.1
 labour (ha)

Agricultural production 96 124
 index (1989-91 = 100)
Food stuff production 96 126
 index (1989-91 = 100)

Total fertiliser 1510 -18 1104 882 -20
 consumption (000 tonnes)
 Per hectare (tonne) 0.31 -13 0.46 0.42 -8
Total agricultural 2210000 3 41203 131358 219
 tractors (unit)
 Per hectare (unit) 0.45 9 0.02 0.07 249
Total harvester-threshers 1208000 -1 43594 74258 70
 (unit)
 Per hectare (unit) 0.24 5 0.02 0.04 87

Sources: Ministry of Agriculture and Forestry 1999b, pp. 290-291,297,
299,301.
Ministry of Agriculture and Forestry 1999a, pp. 21,42, 218,427-428.
Ministry of Agriculture 1990, pp. 18, 26, 314.
Ministry of Agriculture 1997, pp. 6, 8,297.
FAO 2000, FAOSTAT Statistics Database.

Notes

(1.) The authors wish to thank a number of individuals who helped in this study, especially, Jung-Sup Choi, JooHo Song, De-Wen Wang, Wei-Ming Tian. We are grateful to Jock Anderson and Douglas Forno of the World Bank for reading an earlier version of the paper and for their constructive suggestions in revising the paper, and to the Journal's referee for critical but very constructive comments. We also wish to thank Ji-Hong Li for her assistance in data analysis and Marjorie Wilson for her editorial assistance.

(2.) Strictly speaking, no part of the agricultural labour force is redundant or surplus, although their marginal product of labour may be very low. By "surplus", here we really refer to agricultural workers with a relatively lower marginal product of labour than that of comparable labour in other sectors of the economy (for further elaboration, see Schultz 1964, pp. 53-57).

(3.) The discussion on factors affecting part-time farming is partly based on Tian and Shi (1991).

(4.) A new classification, "commercial farm household" and "non-commercial farm household", was introduced in 1990 in Japan. The former includes farms with an area of cultivated land of 0.3 hectare or more, or whose annual sales of agricultural products amount to 500000 yen or more. Those other than commercial farm households are classified as non-commercial farm households. Of the 3.39 million farm households in existence in 1997, 23% (0.78 million) were classified as non-commercial farms and 77% (2.61 million) as commercial farms.

(5.) The percentage of part-time farms accounted for around 44% in 1996 but dropped to 41% in 1997 and again to 37% in 1998. This is likely due to the 1997 financial crisis which led to reduced employment opportunities in non-agricultural sectors.

(6.) The definition of part-time farming used in this study may understate the degree of full time farming. This concern was also shared by the paper's reviewer. However, this definition has been widely adopted in the previous studies and official data construction. The main reason we adopted this definition is to maintain the comparability with other studies and data. However, in order to minimise the concern indicated above, we kept the distinction of Type I and Type II classification in our part-time farming analysis as much as possible. Further, in our econometric analysis, we used the ratio of non-farm to total work days as the part-time farming variable (see Section 4). This is consistent with the definition that was suggested by the reviewer. In addition to studies contained in this study that address the definition of part-time farming, other studies that discuss part-time farming definition issue include, for example, Lund (1991), Gasson (1991), and Bryden, Fuller, MacKinnon (1992).

(7.) For more details about the evolution of the part-time farming definition, see Institute of Developing Economies (1969).

(8.) In 1996, one further category was added: no family business operations. Currently, the number of families without any business operations is small.

(9.) In Japan, part-time farming took place long ago. As noted earleir, official data collection on part-time farm households traces back to as early as 1906. However, the year of 1946 was chosen for Japan as the starting point for comparison. This was dictated by the data availability for China. In this paper, we intend to compare the level of part-time farming development in relation to the share of agricultural GDP out of total GDP. For China, 1986 was the first year for which we can calculate the level of part-time farming development. Subsequently, the share of agricultural GDP out of total GDP for China for 1986 was used to select the starting year for Japan and Korea when each of them had a similar agricultural GDP share to China's.

(10.) In early November 1999, the Chinese government officially admitted that the total arable land was 130 million hectares (1.951 billion mu) (Li and Zou 1999). According to this new figure, China's per capita arable land is 0.105 hectare. Earlier, the officially acknowledged arable land area was about 1.5 billion mu or 100 million hectares.

References

Ali, A.M.S. 1995, `Part-time farming in Bangladesh: A village case study of its causes', Contemporary South Asia, Vol. 4, pp. 309-330.

Bryden, J., Fuller, A. and MacKinnon, N. 1992, `Part-time farming: A note on definitions -- A further comment', Journal of Agricultural Economics, Vol. 43, pp. 109-110.

Cai, F. and Chan, K.W. 2000, `The political economy of urban protectionist employment policies in China', paper presented at the International Conference "China: Growth Sustainability in the 21st Century" held at the Australian National University, Canberra, Australia, 9-10 September 2000.

Chen, J.G. and Xu, Y.J. 1993, `A reflection on the large-scale mobile rural labour force around the country', Problems of Agricultural Economics, No. 6, pp. 32-34.

Chung, K.W. 1993, `Migration impacts on the change of family farm structure in Korea', Journal of Rural Development, Vol. 16, pp. 15-40.

FAO 2000, FAOSTAT Statistics Database, Rome, Italy.

Gasson, R. 1991, `Part-time farming: A note on definitions -- Comment', Journal of Agricultural Economics, Vol. 42, pp. 200-201.

Government of Japan, Japan Statistical Yearbook, various issues.

Hayami, Y. 1975, A Century of Agricultural Growth in Japan, University of Minnesota Press, Minneapolis.

Hubei Statistical Bureau 1999, Hubei Rural Statistical Yearbook 1999, China Statistical Press, Beijing.

Institute of Developing Economies 1969, One Hundred Years of Agricultural Statistics in Japan, Institute of Developing Economies, Tokyo.

Johnson, D.G. 1999, `Agricultural adjustment in China: Problems and prospects', Office of Agricultural Economics Research, the University of Chicago, Paper No. 99.01.

Jussaume, R.A. 1991, Japanese Part-Time Farming: Evolution and Impacts, Iowa State University Press, Ames.

Kada, R. 1980, Part-Time Family Farming: Off-Farm Employment and Farm Adjustments in the United States and Japan, Centre for Academic Publications, Tokyo.

Krasovec, S. (ed.) 1982, Part-time Farmers and Their Adjustment to Pluriacity, proceedings of seminar, Slovene Academy of Sciences and Arts, Ljubljana, Yugoslavia.

Li, A.D. and Zou, Q.L. 1999, `China's total arable land is 1951 million mu', People's Daily, 4 November 1999, p. 1.

Loyns, R.M.A. and Kraut, M. 1992, `The family farm in the next decade: The positive role of part-time farming', Canadian Journal of Agricultural Economics, Vol. 40, pp. 591-604.

Lund, P.J. 1991, `Part-time farming: A note on definitions', Journal of Agricultural Economics, Vol. 42, pp. 196-199.

Meng, X. 1996, `Regional wage gap, information flow, and rural-urban migration', paper presented at the Conference on Rural-Urban Migration, 25-27 June, Beijing.

Ministry of Agriculture, China's Agricultural Statistics, various issues, ChinaAgricultural Press, Beijing.

Ministry of Agriculture and Forestry 1999a, Major Statistics of Agriculture and Forestry,

Ministry of Agriculture and Forestry, Republic of Korea, Seoul.

Ministry of Agriculture and Forestry 1999b, Agriculture and Forestry Statistical Yearbook, Ministry of Agriculture and Forestry, Republic of Korea, Seoul.

Mu, G.Z. 1989, `A historic examination of farm surplus labour force in China', Chinese Rural Economy, No. 3, pp. 40-51; No. 4, pp. 41-46.

Nakayasu, S. 1991, `Japan's agricultural structure: Characteristics and changes', Agriculture and Agricultural Policy in Japan, (edited by the Committee for the Japanese Agricultural Session, XXI IAAE Conference) pp. 143-158.

OECD 1995, Agricultural Policy Reform and Adjustment in Japan, Organisation for Economic Cooperation and Development, Paris.

OECD 1999, Review of Agricultural Policies in Korea, Organisation for Economic Cooperation and Development, Paris.

Research Group of the Ministry of Agriculture of China 1995, `A survey of the status of mobile rural labour force', Chinese Rural Economy, No. 1, pp. 43-50.

Rozelle, S., Guo, L., Shen, M.G., Hughart, A. and Giles, J. 1999, `Leaving China's farms: Survey results of new paths and remaining hurdles to rural migration', China Quarterly, No. 158, pp. 367-393.

Schultz, TW. 1964, Transforming Traditional Agriculture, The University of Chicago Press, Chicago.

SS[B.sub.1] (State Statistical Bureau), China Statistical Yearbook, various issues, China Statistical Press, Beijing.

SS[B.sub.2] (State Statistical Bureau) 1996, China's Regional Economy: A Profile of 17 Years of Reform and Opening-Up, China Statistical Press, Beijing.

Sumner, D.A. 1982, `The Off-Farm Labour Supply of Farmers', American Journal of Agricultural Economics Vol. 64, pp. 499-509.

Tian, Y. and Shi, F.Z. 1991, `On part-time transfer of agricultural labour', Problems of Agricultural Economy, No. 6, pp. 54-59.

Tweeten, L., Dishon, C.L., Chern, W.S., and Imamura, N. (eds) 1993, Japanese and American Agriculture: Tradition and Progress in Conflict, Westview Press, Boulder, Colorado.

Wan, G.H. 1995, `Peasant flood in China: internal migration and its policy determinants', Third World Quarterly, Vol. 16, pp. 173-196.

Wang, X.G. 1994, `Rural economic structural transition and choice of new models: An economic analysis of the phenomenon of large-scale mobile rural labour force', Problems of Agricultural Economics, No. 7, pp. 47-52.

Weiss, C.R. 1997, `Do they come back again? The symmetry and reversibility of off-farm employment', European Review of Agricultural Economics, Vol. 24, pp. 65-84.

Yang, T and Cai, F. 1991, `Part-time fanning households and agricultural labour transfer', Chinese Rural Economy, No. 11, pp. 43-50.

Ye, X.Q. 1992, `Industrialisation, part-time farming and agricultural development', Rural Economy Forum, No. 10, pp. 26-33.

Yunnan Statistical Bureau 1998, Yunnan Statistical Yearbook 1998, China Statistical Press, Beijing.

Zhang, X.H., Chen, H.S., and Wu, Z.G. 1996, `Part-time farming and income situations in traditional agricultural regions in China: A comparison', Agricultural Economics Research Materials, No. 4, pp. 40-49.

Zhou, Z.Y., Dillon, J.L. and Wan, G.H. 1992, `Development of township enterprise and alleviation of the employment problem in rural China', Agricultural Economics, Vol. 6, pp. 201-215.

Zhu, Z. 1993, `The problem of the large-scale mobile rural labour force around the country: Current status, causes, and countermeasures', Chinese Rural Economy, No. 12, pp. 33-36.

Zhang-Yue Zhou Asian Agribusiness Research Centre, The University of Sydney, Orange

Daniel A. Sumner Department of Agricultural and Resource Economics, The University of California, Davis

Hyunok Lee Department of Agricultural and Resource Economics, The University of California, Davis