The Framework

An understanding of technology development and diffusion can contribute to the identification of some of the basic forces operating in a society. The interest of economists in this development of technology lies in its importance as an indicator of economic growth. Historically, technological progress has varied over time and space: the capacity of different societies to generate, adapt and disseminate appropriate technical innovations has shown considerable variation. The reasons for these differences reflect the socio-economic structure of these societies.

Most people in developing countries live in rural areas and work in agriculture. In India, after three decades of planning for socio-economic development, 76 per cent of the population lives in villages; agriculture accounts for about 42 per cent of the national income and provides a livelihood for more than 67 per cent of the population (India, Planning Commission, 1981; India, Registrar General, 1982). Rural development and the improvement of the low level of agricultural productivity have, therefore, been priority concerns of Indian planners since 1950. Development and diffusion of agricultural technology has been an important source of agricultural growth. Many studies have analysed the diffusion of agricultural technology and its impact on output, yield, employment and income distribution. There are, however, very few studies on the diffusion of agro-mechanical technology (Feder et al., 1982). Moreover, the available studies mostly concentrate on relatively expensive equipment such as tractors, threshers, oil engines and electric motors. Much less has been said about simpler and less expensive equipment whose contribution to agriculture may not be as visible but is still important, both in terms of lessening the drudgery of work and higher labour efficiency. Besides, the viability of mechanization can be questioned on many grounds in a country such as India where the majority of land holdings are small, 67 per cent of rural households owning less than one hectare of land (NSSO, 1986).

The agro-machinery used on various farms differs across regions in accordance with the nature of cultivation. Broadly speaking, there are two sources of farm-power: tractor-based agro-machinery and bullock- or man-based. In a given context, larger holdings use different sets of agro-machinery and, therefore, different types of farm power, than smaller ones. The ownership pattern of tractors and tractor-based equipment is heavily in favour of the large holdings. 1 The picture is similar in the case of electric motors and diesel engines (Dasgupta, 1977, Chapter 2). The growth of tractor-hiring facilities enables some of the smaller farmers to use tractors on rent, yet small farmers (with less than two hectares of land) are unable to make use of the facilities on any significant scale (Sharan and Krishna, 1976; Patel, 1980). Much of the same imbalance will naturally be seen in agro-machinery and equipment that is linked to the tractor as a power unit. Also, the system of hiring farm machinery, such as tractors and harvesters, has given rise to a new type of patron-client dependency between poor and rich farmers (Dasgupta, 1977, p.379).

There have been serious R&D efforts to develop small tractors. However, some technical problems make it difficult to develop a low horsepower tractor that would be sturdy and efficient and also affordable to small and marginal farmers. Besides, tractors are in fact not suitable in all conditions: bullocks perform better than tractors on clay soils, in ploughing comers and in isolated patches of highland. Moreover, tractors do not usually displace bullock labour. It has been found that tractor farms continue to keep at least one pair of bullocks (Dasgupta, 1977, pp. 108-9; Raj, 1974, p. 123). Thus, tractor-hire schemes notwithstanding, tractor-based agro-machinery remains limited to upper segments of the peasantry. The marginal and small farmers, who constitute a large proportion of the total cultivator population, rely mainly on bullock/man-operated implements. In fact, it has been shown that in the near future, available farm power will be dominated by manual and draft power; the tractor population, on the other hand, is small and only a modest increase is expected in the coming years (India, National Commission on Agriculture (NCA), 1976, Part X, Chapter 50). Furthermore, the impact of machines such as tractors, threshers, electric motors and diesel engines is similar to that of seed fertilizer technology. If these services are available only to a limited set of cultivators, the biases of seed fertilizer technology become a policy concern. The other major concern has been that mechanization may reduce employment. There is, however, no consensus regarding the impact of the use of above-mentioned machines on yields, output and employment (Binswanger, 1978; Basant, 1985).

Large scale co-operation, consolidation of small holdings, socialization and reorganization of land into holdings of optimal size can overcome structural constraints and pave the way for more rapid mechanization. It has been argued, however, that small farms which constitute a large chunk of the peasantry will continue to remain a very important part of the agricultural scene in India (Vyas, 1976; Joshi 1979). It is imperative, therefore, to think of ways in which technological upgrading of agro-machinery on bullock-operated farms can be accomplished. It is noteworthy in this context that some of the improved hand-operated and animal-drawn implements can be five to ten times more efficient in energy conversion as compared to primitive tools and implements commonly used by farmers (India, NCA, 1976, p.393). It has also been pointed out that improved implements can greatly accelerate bullock ploughing (Marsden, 1973, p. 7). It is in this context that the present study attempts to analyse the processes of agro-mechanical technology diffusion in a backward region (Panchmahal district of Gujarat State).

Focus of the study

Development and diffusion of agro-mechanical technology characterized by improved hand- and bullock-operated implements, therefore, is an important area of research. Few studies have, however, focused on this theme (for a detailed review see Feder et al., 1982). The present study is an attempt in the direction of understanding the constraints on the diffusion of agro-mechanical technology. The relevance of the study in the present context is brought out by an awareness that historically successful agricultural growth in different countries has capitalized on favourable production factors, e.g. land and mechanization in the US; labour and land improvements and biological technology in Japan. European countries also stressed biological technology before emphasis shifted to mechanical technology (Binswanger, 1984). Further, to rely mainly on mechanization for output growth may be misplaced since maximizing yields at the margin is not the same thing as maximizing output in a country as a whole. The two objectives can diverge widely if the former means concentrating scarce capital and skill resources on land with the greatest natural endowments, and on those holdings which can afford mechanization, while neglecting or discouraging more modest improvements which could be reproduced much more widely (Marsden, 1974, pp. 5-6).

In a broad sense, technology is more than just machinery and equipment or knowledge of methods of production; it is an integral part of the whole system of economic, institutional, social and even political arrangements that characterize the way a society functions. Looked at in this fashion, the role that technology can play will depend upon the socio-economic and political structure of the society and its transformation over time. In the context of agricultural technology, therefore, it is important to analyse not only its relation to output and inputs but also its inter-relationships with the institutional infrastructure of the region, the extent of commercialization of farming, the non-agricultural activities in the region and the values, attitudes, customs and behaviour patterns of rural society. Economic, social and other institutional adjustments thus become an integral part of technological change (Wang, 1981, p. 7). An exploration of these relationships with specific reference to agro-mechanical technology is the focus of the present study.

As mentioned earlier, the agro-machinery situation in India is characterized by two distinct technological worlds: a small segment consisting of large farmers increasingly employing tractor-powered equipment; and a large segment of small peasants using only hand- and bullock-operated equipment. Research efforts seek to improve and modernize this machinery but the pace of adoption of improved implements is extremely slow (India, NCA, 1976). Most tools are still designed, fabricated and supplied by local artisans and markets, particularly in backward areas.

Diffusion of technology is not, however, merely the pace of adoption or spread of a new technology, as normally referred to in economics. To the extent that technologies are difficult to imitate and require learning-by-doing and R&D, the technological development of a region should be viewed as its capacity to generate, transfer, adopt, adapt and incorporate into the economic system new techniques which displace or modify the older techniques, and/or to facilitate these processes in a smooth manner. Technology diffusion refers to this whole process. Defined in this sense, diffusion is a continuous process and is central to socio-economic development. The motivations and mechanics of diffusion are subjects that have received considerable attention. Analysis, however, has focused on changes in the structure of demand and on changing relative cost factors as the main inducement. Economists have viewed the process of diffusion as a response to profit expectation, as shaped by market size and structure. (For a detailed review of studies on agricultural technology diffusion, see Feder, et al., 1982).

While the demand side is important, diffusion of technology also depends upon the pace at which supply-side constraints are overcome. Relevant supply-side factors include the state of scientific knowledge, prevailing technology, changes in the number of skilled workers engaged in production and the upgrading of their skills, the organizational structure of technology generation, links between different agents involved in technology generation and the economic system. These factors determine the endogenous R&D and engineering efforts which may be called local technological capacity. Without an understanding of these supply-side forces, it is difficult to explain technology diffusion in a developing country, particularly in a backward region.

Methodology

This study is based on an intensive survey of four villages belonging to two different talukas of a backward district in Gujarat. It is essentially a micro-level study of the process of technology diffusion in agriculture. The possibilities of drawing generalized inferences are obviously limited. In so far as the selected villages represent different levels of development, one may be able to generate testable hypotheses regarding the processes of technology diffusion in transitional agrarian economies.

The district selected for study, Panchmahals, is regarded as one of the most backward districts of Gujarat in terms of agricultural development. Yield levels in the district have been low with limited growth of agricultural output. In the 1960s the compound annual rate of growth of agricultural output was less than 1.5 per cent (Bhalla and Alagh, 1979, Appendix 5}, without any significant improvement in recent years (Sharma, 1983).

Villages from two talukas, Kalol and Dahod, were chosen for the study. Indices of agricultural productivity show Kalol to be one of the better talukas and Dahod as one of the not so good ones (Government of Gujarat, 1979). While Dahod is a dry and agriculturally backward region, Kalol is relatively better irrigated and, agriculturally, better endowed. A comparative study of these contrasting talukas within the same district was expected to yield some useful insights into the specific processes of technology diffusion.

Selection of sample villages

Officially published sources did not yield any information about the number of artisans making agricultural implements in the various talukas and villages of Panchmahals district. Taluka Panchayat officials of Dahod and Kalol talukas were contacted to identify those artisans who make improved implements on a somewhat large scale and supply to the neighbouring villages. Jesawada and Garbada were identified as the two such centres in Dahod taluka, and Kalol town and Derol in Kalol taluka. It was then decided to concentrate efforts on the regions covered by the Kalol town artisans in Kalol taluka and Garbada artisans in Dahod taluka. Although Kalol and Garbada 'artisans' mainly produce agricultural implements only on order, their production process is highly mechanized and they also sub-contract for other manufacturers.

The artisans making improved implements in Garbada and Kalol town were contacted and were asked about the villages they cater to in the area. Two villages in each taluka, one backward and the other developed, were selected from this list. In the absence of any detailed village level information on the levels of agricultural development, the selection was based on the extent of irrigation using cultivated area per electric pump/oil engine as a proxy (Table 1).

Selection of sample households

Cultivators who use agricultural implements, artisans who fabricate them and registered agents of outside fabricators in the market were identified as the three main agents directly involved in the process of generation and diffusion of agro-mechanical technology. Three questionnaires were created, one each for cultivators, artisans and agents. As an example, the questionnaire for artisans is included as Appendix 1. All households in the four selected villages were listed. For selecting a sample of cultivators, all cultivator households were first divided into four strata of operational land holdings: marginal (less than one hectare); small (1-3 hectares); medium (3-7.5 hectares) and large (more than 7.5 hectares) for the four study villages. A stratified random sample of 20 cultivators was then chosen from each village. The number of sample households constituted more than 15 per cent of the total cultivator households in the village, except in Sureli, where it constituted only 6.5 per cent.

All the blacksmiths and carpenters in the selected villages were surveyed, along with the artisans making improved implements in Kalol town and Garbada. All the agents operating in the selected region were also interviewed.

Socio-economic Profile of the Study Villages

Panchmahals district is located on the eastern border of Gujarat State. Five of the eleven talukas and about 42 per cent of the district population are tribal. At the turn of the century, a large part of the district area was under forest or cultivable waste. In 1880-1, only 27 per cent of the total area was under cultivation, about 36 per cent under forest and 33 per cent cultivable waste. A large proportion of the uncultivated area was located in tribal talukas. Since then, however, the area under cultivation has steadily increased and, in 1974-75, the net sown area accounted for about 54 per cent of the total reported area of the district. Thus, agriculture in Panchmahals district, particularly in the tribal talukas, does not have a very long history. Of the two talukas chosen for this study Dahod is a tribal taluka while Kalol is non-tribal. According to the 1981 Census, the proportion of tribal population was 73 per cent in Dahod and only 8 per cent in Kalol (Government of Gujarat, Panchmahals Gazetteer, 1972, Statement IV.2, p.267; India, Registrar General, 1981, Paper 2; District Statistical Abstract, Panchmahals, 1974-75).

The major part of the district is 'hilly' and 'undulating' and Dahod taluka belongs to this part of the district. Kalol taluka, on the other hand, is flat (Government of Gujarat, Panchmahals Gazetteer, 1972). Both talukas have a dry climate, with the south-west monsoon providing rain during June-September. In terms of average annual rainfall, however, Kalol taluka is slightly better than Dahod. Panchmahals is a drought-prone district. Historically the eastern portion of the district of which Dahod is a part has been more vulnerable to scarcity conditions than the western portion where Kalol is located. In recent droughts also Dahod has been one of the worst attested talukas of the district. Kalol on the other hand is relatively less prone to droughts and famines. The villages selected from these two talukas provide a contrasting picture in terms of their socio-economic profile.