Off-farm income, risk, and agricultural production: a case study of smallholders in India's semi-arid tropics

The purpose of the current study was to determine to what extent development of the nonfarm sector stimulates crop production and agricultural employment in India's semi-arid tropics (SAT). Two characteristics of India's SAT agriculture formed the basis for the study: i) that the major agricultural producers, smallholders, earn a significant part of their income outside their own farm, especially in the daily labour market; and ii) that crop production risk is high. As the farmers are risk averse and they have limited options to smooth consumption through changes in assets and liabilities, they try to stabilise income. This could result in the choice of crop production technologies that render relatively stable, but low yields. On the other hand, off-farm income may provide an alternative for stabilising income and ensure minimum consumption in all years. Hence, because they can earn off-farm income, farm households may be willing and able to increase crop incomes, even if this involves higher crop income variability from year to year.The study shows that the labour market is extremely important in preventing farm households in India's SAT from experiencing very low incomes. Households actively adapt family labour supply to production risk. Much more important for income stability, however, is the flow of labour income that households earn independent of their risk preferences. As expected, production risk affects not only farm-household labour supply, but also the use of the major inputs in crop production: male and female labour and fertiliser use. Depending on the specific input and the production environment, risk appears to increase or decrease the use of a specific input.The above suggests that the household labour endowment will affect crop input use and income from crop production. The first is true, given that there is sufficient year-round off-farm employment and not just employment during the agricultural growing season. In this case, households with a larger labour endowment feel less need to adapt the use of crop labour and fertilisers to production risk. However, these small changes in input use hardly seem to affect the magnitude and stability of crop income: the estimated input elasticities of the mean and variability of crop yields are low. Surprisingly, the household labour endowment thus has little effect on crop income.Actual labour income may, however, still have a significant impact on net crop income: some farmers work longer hours than others do. Moreover, income from minor sources of off-farm employment, like nonfarm self-employment, may also affect net crop income. Empirical estimates show that the size and nature of the effect of off-farm income on net crop income depends on the characteristics of the income flow. Activities that require labour simultaneous to crop production and that have low returns on labour decrease net crop income. If the returns on the off-farm activity are certain and high relative to the casual farm wage, households will, however, simply substitute hired labour or capital for family labour. Moreover, they will use part of the off-farm income to intensify crop production. Finally, if the nonfarm activity becomes the main source of interest of the farm household, income from this activity will no longer affect crop production.