fore for building small local plants. However, the investment cost per ton of cement may still be lower than for big plants if idle equipment with low opportunity cost can be used. Aside from transportation considerations, the impact on employment and industrial training may have been considered to be very important in the regional

context.

The scanty statistical information available indicates that the production of cement in small-scale plants in China has increased from roughly 5 million tons in 1965 to approximately 19 million tons in 1973. Consequently, the total increase in the small-scale sector amounts to 14 million tons. A small plant producing 20,000 tons per year employs around 200 persons. The total increase in industrial employment in small-scale cement plants would then correspondingly be at least 140,000 persons. A substantial number of rural plants are smaller and more labor-intensive than in the example cited here.

Thus, it can be safely assumed that the total direct employment effect from rural cement plants (manufacturing only) would amount to at least 250,000. This number is at least 10 times higher than employment in a small number of modern large-scale cement plants producing the same quantity of cement using a rotating kiln with an annual capacities in the range 150,000-600,000 tons.

There are other important labor force considerations aside from the employment effect. First, a large number of people are being trained in industrial process technology. Second, a sizable number of people have, inside production units, received training in organizational skills. A smaller but still sizable number have been trained in administrative skills related to the procurement of machinery and raw materials, distribution of products and coordination with other industrial units.

THE INTEGRATED APPROACH

Rural industrialization is not pursued as a policy in isolation. It should be seen as the outcome of a combination of strategies-for various industrial sectors on one hand and for integrated rural development on the other hand. Industrial sector strategies aimed at expanding the lower end of the spectrum can be nothing more than part of an overall policy to develop rural areas. The sector strategies usually cannot operate in isolation, as they are dependent on demand and resources which are influenced through rural development. However, the industrial sector strategies may lend a strong support to integrated rural development programs.

The development of rural industries is a function of both demand. and availability of local resources, where it is generally easier to influence the former than the latter. Rural industry in China can be categorized as backward linkage industries meeting a demand for agricultural inputs, and consumer goods, and forward linkage industries being mainly based on locally available physical and human resources. See figure 3. The backward linkage industries, of coures, require local human resource but are initially very dependent on external technological and financial resources.

Rural industries with backward linkages to agriculture usually cannot be introduced until changes in agricultural technology create the demand for industrial input. Forward linkage industries are often dependent on supplies arising from increased agricultural production.

Furthermore, the demand for many industrial products will be limited until there is a general increase in the purchasing power of the locality. This is in most places-at least initially-almost totally dependent on increases in agricultural productivity. In sum, rural industrialization can progress only gradually and must be closely integrated with the overall planning of the localities.

Figure 3 illustrates the flow chart for a county or a locality which is well endowed with natural resources including coal, iron ore, and limestone. In such a locality the industrial activities can be divided into five different components. The first is the heavy small-scale industry which includes what the Chinese often term "five small industries". These supply energy, cement, chemical fertilizer, iron and steel, and machinery, which directly or indirectly provide agriculture with the inputs which are necessary for raising the productivity. These plants are usually run by the county.

17

The second component is various resource-specific industries which may provide raw materials such as coal, iron ore, limestone, and other minerals to the "five small industries" or to larger national enterprises. It is evident that the resource base-location, size, and quality decides to what extent small-scale industries can be set up. Économies of scale and the development of the transportation system are other important factors—as exemplified in the earlier section discussing local manufacture of nitrogen fertilizer and cement.

The third and fourth components of industrial activity in a county both belong to light industry. The main responsibility of this sector is to process agricultural and sideline produce and to provide the locality with some of the needed consumer goods. This may include flour milling, oil pressing, cotton ginning, et cetera, as well as manufacture of textiles and shoes, household goods of porcelain and metal, canned fruit, and so on. Light industrial enterprises are found at county, commune, and brigade levels.

The fifth component, the farm machinery repair and manufacture network, produces simple farm implements, tools, and also heavier equipment. Of primary importance is the repair and maintenance of farm implements and machinery in order to sustain a high rate of machinery utilization. Therefore, the repair and manufacture network has a three-tier structure where, in principle, each brigade and each commune within a county should have its own unit. Naturally, the smallest units are run by the brigades, medium-sized ones by the communes, and the relatively large units by the county. The goal is to develop a clear division of labor among units at different levels. The brigade units should engage in simple repairs and manufacturing. The commune units should be able to carry out more complex repairs and also manufacture heavier equipment. The county stations should be able to carry out the repairs of any complex for machinery being used within the county and also manufacture a substantial proportion of heavy equipment to be used in agriculture or in the units of the repair and manufacture network.

The Chinese planners have attempted to use the demand arising from a modernizing agriculture to create backward linkages—in a

17 These and many others were the mainstay of the Great Leap Forward when China attempted to mobilize her abundant manpower resources and utilize local raw materials to simultaneously boost production in almost all industrial sectors.

rural context for almost all of the important inputs required by agriculture.

Some of the backward linkage industries, such as cement and chemical fertilizer, are justified because of the existence of local raw material resources. However, their development has not been triggered because of existing local resources but because of a demand caused mainly by changes in agricultural technology, and they almost completely lack justification without local demand.

Mechanical inputs are manufactured in county-, commune-, and brigade-level units where the activity at brigade level usually con sists only of a couple of blacksmiths using a few pieces of relatively simple machinery. Chemical inputs are usually produced in countylevel enterprises even if mixing and simple phosphate fertilizer production may be found in commune-level enterprises. The manufacture of consumer goods is carried out at all three levels but more complex products (e.g., molding of plastic sandals) only in county-level units. Processing of agricultural and sideline produce is mainly carried out by brigades and communes. Forestry-based and mineral-based industry is generally carried out by brigades and communes, depending on the resource base, the manpower situation and the technology utilized. Handicraft industry is found at commune, brigade, and household level. All this relates to industry in rural areas proper, and the situation may be quite different in near-city locations.

Rural industries generally serve a small market, the size of which varies with the level-brigade, commune, county, or region--where the enterprise is found. Through choice of enterprise size and manufacturing technology it has been possible in many industrial sectors to transfer industry into rural areas and manufacture economically to meet a local demand, if all externalities are considered. Further, size of enterprise and technology is closely related to the controlinitiative level and many localities communes and counties-are already being drawn into subcontracting and manufacturing export items, thus also meeting an external demand. This is particularly true for rural industry in near-city locations.

It appears that the Chinese planners are now guided by a conviction that most of the facilities for economic and social activities in rural areas can be provided most effectively and economically when they are clustered in space so as to take advantage of interdependence in their functions and use.

To fully understand the financial implications of an integrated approach to agricultural development and rural industrialization, it may be advantageous once more to look at the diagram (figure 3), which shows the commodity flow between industrial components and agriculture. The figure also shows which economic units at county level control the various parts of the economic system. The county agricultural-industrial system should be seen as one functional organization with a number of subdivisions. All the different units must not necessarily operate with a surplus. The farm machinery repair and manufacture network usually operates with little or no surplus. The command or control of the operations of the enterprises does not lie within the single units but is vested in the production command of the county (left top of the figure). Under its direct control are the various economic bureaus-industry bureau, farm machinery bureau,

agriculture and forestry bureau, water and electricity bureau, et cetera. The overall productivity of the system depends, aside from capital accumulation, on technology improvements in agriculture and industry. So it is natural to find that both industrial and agricultural extension are under the science and technology group, which is a subgroup of the production command.

CONCLUDING REMARKS

The development of the lower end of the industrial spectrum in various sectors has made it possible to locate industries in rural areas— in near-city locations as well as in rural areas proper. The sector strategies are naturally undergoing constant changes due to changing internal and external conditions. Consequently, the lower end of the industrial spectrum is expanding rapidly in some sectors, while it is contracting in others, thus influencing the rural industrial system. Integrated rural development, on the other hand, includes rural industry only as one component of many instruments where improved public health, education, introduction, of agricultural technology all contribute to achieving policy objectives, such as increased gainful employment, increasing productivity and reducing differentials between urban and rural areas.

The justification of rural industrialization is economic as well as political. Economic growth may, however, in the short run have been sacrificed because of the need to transfer technical, financial, and planning resources to rural areas in order to start rural industrialization programs. In the long run, however, rural industrialization is likely to contribute to a more rapid economic growth than would otherwise have been possible. First, a decentralized pattern of urban development is likely to lead to a less capital-intensive expansion of industrial growth, which is better adapted to prevailing factor availabilities and relative factor prices. Second, if a majority of the rural population, through integrated rural development, might be persuaded to remain in villages and expanding county capitals, less investment for expanding large urban centers will be necessary. Third, rural industrialization provides the opportunity of simultaneously promoting agricultural and nonagricultural elements in the same local areas. Nonagricultural elements are just as dependent on a thriving increase in farm output and income as the latter are dependent on them. Fourth, a rural industrialization which has a strong core of local engineering enterprises is likely to play an important role in any decentralized industrialization policy.

An attempt is made to summarize the economic reasons into three main categories-to reduce short- and medium-term capital requirements, to incorporate the economy of many locations into the longterm planning, and to achieve resource mobilization.

The capital requirements are reduced in a number of ways. First, capital expenditures needed for housing and infrastructure are likely to be considerably lower in rural areas than in urban areas. Second, rural industries will in many cases reduce transportation costs, the high transportation costs being partly a reflection of capital scarcity affecting the transportation sector. Third, as almost all equipment and machinery in small plants can generally be produced within the coun