Off-farm labour supply and employment : a case study of Fiji sugar cane farmers, 1970 and 1983

Abstract

Off-farm employment has become increasingly important as an aspect of resource adjustment and source of income in many developing countries. This study attempts to provide a better understanding of the contributions of off-farm income to total household income among a sample of sugar cane farmers in the Sigatoka area, Fiji. The role and extent of off-farm employment is examined using cross-sectional data for two time periods, 1970 and 1983. In addition, the study examines some of the factors that determine the supply of labour for off-farm employment among the surveyed farmers. A conceptual model is developed and applied to the 1983 data. The Tobit maximum likelihood procedure is utilised to test the influences of the life cycle, level of human capital, opportunities for off-farm work, non-wage income and net farm income on the off-farm labour supply of household members. The findings of the study indicate that income from off-farm work played an important role in raising the real standard of living of the 40 cane sample farm households. For many households, off-farm activities also act to reduce substantially the riskiness of their total portfolio of income earning assets. The study showed that while off-farm income was unequally distributed, overall it has improved the distribution of total income within the sample sugar cane farm households. This reflects a situation whereby the lowest income group in the sample have resorted to more off-farm work, resulting in an equal total household income distribution. The econometric study of the off-farm labour supply function suggested that age, level of human capital, the existence of opportunities and the level of non-wage income were significant determinants of off-farm labour supply. Net farm income was not found to be significant in determining off-farm labour supply. Due to data limitations and the sample size, discussion centred on the sign of the significant variables rather than absolute magnitudes, using an unrestricted reduced form equation. Some policy implications are presented in the final chapter.