|CERES No. 059 (FAO Ceres, 1977, 50 p.)|
any rural development strategy must be an integrated effort to promote growth and employment in all sectors while taking into account the uniqueness of each country
by Derek Byerlee, Carl K. Eicher, Carl Liedholm and Dunstan S.C. Spencer
In the early 1970s, the economic growth trickle-down theory of development was pushed aside by many economists. African countries were urged to devote more attention to equity objectives, such as employment, income distribution and access to social services. Much was written about conflicts between output and employment, the need to "dethrone" GNP and adopt employment-oriented development strategies. Specific interventions were advocated-intermediate technology, labour-intensive industries-to support the new development economics, which stressed equity and basic needs. Unfortunately, the new development economics was "unencumbered" by empirical evidence in most African countries. There was little microeconomic data in the early 1970s on output, incomes and labour inputs and productivity for different regions and farming systems. Since 50 to 90 percent of the people in Africa live in rural areas, there was an obvious and urgent need for reliable empirical research on production processes, employment and income distribution in all sectors of the rural economy, including farming, fishing, processing, marketing and rural small-scale industries
Beginning in 1971, a research network was formed to carry out microlevel data collection in rural areas of Ethiopia, Nigeria, Ghana and Sierra Leone. This comparative research was designed to generate detailed information through multiple interviews with farmers, small-scale industries and processing and marketing firms. The research methodology used an integrated approach to collect data simultaneously from the same rural households on daily labour inputs, costs and returns to traditional and improved systems of farming, marketing, processing and small-scale industries. Studies of rural consumption and rural-urban migration were integral parts of the research.
Seasonal labour shortages
A major part of the research was centred in Sierra Leone, where 500 randomly selected rural households were interviewed twice weekly to collect daily data over a twelve-to-fifteen-month period. Also, 120 fishing families, 75 rice-processing firms and 350 small-scale industries in rural and urban areas were interviewed continuously over the year, and a nationwide study of migration was conducted to trace migrants from rural areas to towns. Similar but subnational studies were conducted in Ghana, Nigeria and Ethiopia. This article summarizes the results and insights we gained during the course of this research.!
The relevance of land surplus and labour surplus development models to African conditions has been widely debated. We observed that labour inputs by African farmers are quite low, especially by Asian standards, averaging 1 000 to 1 500 hours per year in many African countries. However, there is now overwhelming evidence of seasonal labour shortages during planting and weeding periods in most African countries. For example, in Sierra Leone, male adult labour inputs were 200 hours per month in the peak season on over half the farms in our sample. Female labour inputs on farming were somewhat lower but, when allowance is made for household tasks, these are comparable to male inputs. On the other hand, there is substantial "surplus" labour in rural areas during the slack agricultural season, although Sierra Leone farmers worked 70 hours per month, even in the slackest month.
These results show that labour shortages and surpluses are both present in rural Africa. Seasonal shortages should be a central consideration in the design of development projects for rural areas. At the same time, slack season surpluses provide the opportunity for expanding output and employment through irrigation, road construction, rural housing and rural small-scale industry
Lack of employment opportunities and low rural incomes, of course, stimulate rapid rural-urban migration and contribute to open urban unemployment. As many as 1.5 percent of the rural population may leave for urban areas each year, resulting in unemployment rates of young migrants in excess of 30 percent. Nonetheless, in Sierra Leone, where urban migrants retain access to land, two out of three migrants returned to rural areas, usually as a result of economic hardship. Other migrants, mostly school-leavers, who stayed often remain unemployed for extended periods waiting for the "right" job with the support of urban relatives.
As a general guideline
Given the seasonal labour shortages and the relatively high land/man ratio in most of tropical Africa, government-subsidized mechanization programmes have been established in many countries. In most cases, these schemes have not been successful largely because of their high costs of operation. In some cases, such as in Ghana's rice-farming areas, privately operated tractor cultivation has been shown to reduce the cost of production, even after allowing for subsidized fuel and overvalued foreign exchange. Although mechanization is often criticized-even where profitable-for its labour displacement effects, our research on tractor cultivation in Ghana and Sierra Leone indicates that such labour displacement has not occurred because the area cultivated has been expanded, thereby increasing the demand for labour in weeding and harvesting. Nonetheless, because of high costs of production, we recommend as a general guideline that subsidies on mechanization be reduced.
At the other extreme, biological technologies have often been promoted: these are difficult for farmers to adopt because they sharply increase the labour demands during the peak season. For example, improved cotton-growing methods in the north of Nigeria increased labour requirements by 50 percent during the peak season at the very time when farmers were planting food crops.
Research and promotion of new agricultural technologies in tropical Africa should always assess the nature of the seasonal labour profile of farming systems into which that technology is to be introduced. Past agricultural development efforts have too often emphasized "extremes" of technologies, such as large tractors, on the one hand, or intensive cultivation practices involving new seeds, fertilizer and water control, on the other. There is a striking lack of research in Africa compared to Asia on generating appropriate technologies to break these seasonal labour bottlenecks. Research is urgently needed to improve the productivity of hand tools and promote the transition to animal power or small tractors. By the same token, agronomists and plant breeders should pursue research on earlier and later maturing varieties, which shift planting and harvesting dates and reduce labour requirements during these bottleneck periods. Finally, planners should concentrate on promoting technologies that are appropriate to labour and other resource endowments of small farmers. Fertilizer use is an example of a technology that often provides substantial potential for increasing output and incomes, but whose additional requirements are concentrated during harvesting-a time that is generally not a labour bottleneck in tropical Africa.
In rougher weather
African development plans have generally underplayed the potential of fisheries - both freshwater and offshore - as sources of income, employment and protein. At the same time, economists have conducted virtually no microlevel research to help in realizing this potential.
A survey of small-scale and large-scale fisheries on Sierra Leone's coastline provides some of the first estimates of income, output and employment in the sector. Results indicated that large-scale trawlers had the highest production cost of any of the nine different technology sets (i.e., combination of boats, nets and propulsion equipment) found in the survey. However, present policies and institutions favour large-scale trawlers by allowing equipment to be imported at 10 percent duty and providing loans at 11 percent interest rates, in contrast to small-scale fishermen who pay a 35-percent duty on imported inputs and borrow money on the informal credit market at interest rates exceeding 40 percent. Despite these adverse factor prices, many small-scale fishermen had invested substantially in improved technologies, including outboard motors. Significantly, these improved technologies were often quite labour intensive. For example, mechanized boats were able to travel further out and to operate in rougher weather so that labour was more evenly distributed throughout the year.
Given the substantial and neglected potential of fisheries, African governments should allocate more resources to promote small-scale technologies, which are cost-efficient and labour intensive. A combination of pricing policies to equalize import duties and interest rates and research and extension programmes should be used to promote small-scale fisheries.
Industrial processing of agricultural products will play an increasingly important role in income and employment generation in African countries in the future. Various technologies are employed in these industries, ranging from traditional hand methods up to large mills (e.g., stork major mills for oil palm and disc sheller mills for rice). The capital intensity of these production processes varies widely and affects production costs and employment. Two other factors also have an important bearing on the choice of technique. First, the technical efficiency (e.g., percentage of clean rice to hulled rice or palm oil per unit of fruit) generally increases with large mills. Secondly, the assembly costs of transporting agricultural products to processing plants increases with the size of plant, tending to offset the higher technical efficiency. In Sierra Leone, a linear programming model incorporating these various interactions was used to estimate the optimal size and location of rice-processing facilities and the effects of alternative policies on production costs, employment and income. Conceivably, if high rice prices, which favour technically efficient mills, continue and capital is available at a 10-percent interest rate, some 40 000 man/years of employment in hand pounding could be replaced by larger rubber roller mills in Sierra Leone. However, with more realistic rice prices and with investors charged a 20-percent cost of capital, small rice mills combined with hand pounding will minimize costs as well as promote rural incomes and employment.
These results are also interesting in that they show that no one processing technique but rather a combination of techniques is optimal, depending on the specific transportation and marketing characteristics of a region.
Rural nonfarm activities, including small-scale industries and trading, are an important source of employment and incomes in rural areas. In tropical Africa, rural households allocate up to one third of their labour to these nonfarm activities. Small-scale industries, such as tailoring, blacksmithing and carpentry, are one of the most important components of this nonfarm activity. When these small industries are defined as firms employing less than 50 people, we find that over 70 percent of all industrial employment in Nigeria and Sierra Leone is located in rural areas-a statistic usually not available in conventional national accounts, which underestimate these activities.
In-depth studies of small-scale industries reveal that they are generally efficient, profitable and largely self-financed. Moreover, they operate without the active encouragement of governments and produce their own skills through the apprenticeship system.
Among the surveyed small-scale industry firms in Nigeria and Sierra Leone, there was a range of alternative techniques indicating scope for labour/capital substitution. Moreover, many small-scale industries compete with large-scale industries using much more capital-intensive techniques. For example, the cost of capital per job in small firms in Sierra Leone was $418 compared to over $7 000 per job in large firms. As a result, the labour-intensive small-scale industries generated both the highest employment and largest output per unit of the scarce factor-capital. That is, in the short run there is no conflict between output and employment in choice of technique in the small-scale industrial sector.
The major policy recommendation from this research is to shift emphasis from promoting urban industrial estates to a concentrated attack on problems facing small-scale industries, such as low skills, poor access to markets (particularly export markets), and limited access to investment resources. Moreover, efforts should be made to change the present pricing system, which favours large industries through duty-free imports of equipment and low-interest loans from formal credit institutions.
Part of our research investigated rural-urban sector linkages. In particular, we investigated the demand patterns generated by rural consumers to determine the demand outlook for labour-intensive commodities. We found that in Sierra Leone, about 60 percent of any increase in rural income was spent on food-largely locally produced under labour-intensive methods. Also, the demand for labour-intensive goods from small-scale industries was shown to be surprisingly high, and in fact had a faster projected increase in demand than goods from the urban-based large-scale industries. The demand outlook for labour-intensive goods, therefore, seems more favourable than previously assumed, although this effect was shown to be weakened if rural incomes become more unequal.
From better-off households
Another important rural-urban linkage occurs through migration. An important question is to what extent efforts to raise rural incomes will reduce the outflow of migrants to urban areas. In fact, evidence from both Tanzania and Sierra Leone indicates that educated migrants (i.e., secondary schooling) are at least ten times more likely to migrate than persons with no education. In both studies, the estimated reduction in migration that would be achieved through an increase in rural incomes was much lower for educated than uneducated persons. Moreover, the Sierra Leone survey showed that uneducated migrants were originating in poorer households and poorer regions of the country while educated migrants largely come from better-off households, which can afford to educate their children. Planners should-recognize this basic dichotomy in designing strategies for reducing migration. Raising rural incomes particularly in poorer regions will be effective in reducing the migration of uneducated migrants. More far-reaching programmes are required to slow down the migration of educated persons. These programmes include decentralization of government projects and industry to rural areas and small towns, revising school curriculums to meet rural needs, aggressive development action programmes and efforts to hold down urban wages and social services.
Our research provides solid empirical support for a development strategy for tropical African countries that focuses on small-scale family-operated producers-farming, fishing, processing and rural industries- because these small firms can efficiently generate both output and employment. Since there is no conflict between employment and output in a small-producer strategy, there is no need to de-emphasize economic growth objectives.
Major attention should be given to the development of a rural strategy that includes mobilization of resources and particularly human resources in all sectors of the rural economy, not just farming. The components of a rural mobilization strategy will vary from country to country depending on its stage of economic history, its problems and its opportunities. In some countries, land reform and institutional restructuring might be far more important than correcting factor price distortions. In other countries, priority might have to be given to correcting the "cheap" food pricing policies that exploit farmers by undervaluing producer prices in order to placate urban consumers. However, while we stress the uniqueness of problems on a country-by-country basis, our research does not support a piecemeal development strategy. A rural strategy must be an integrated effort to promote economic growth and employment in all rural economic sectors. It is obviously far easier to plan than to implement such a strategy because implementation will require political commitment by elites as well as institutional restructuring and a large amount of trial-and-error projects, which must be tested at the farm, village and regional level.