Gender Dis-Aggregated Off-Farm Earnings And Effect On Farm Expenses: Evidence From Rural Ghana

This study looks at the gender dimension of the effect of off-farm incomes on farm expenses. Previous analyses of the effect of off-farm earnings have used the unitary household approach and ignored the intra-household gender and bargaining features. Using cross-sectional data from rural Ghana, we estimate censored regression models to explain the effect of gender dis-aggregated off-farm earnings on farm expenditures. Separate models are estimated for male and female non-farm earnings using instrumental variable Tobit models and two-part models. The dependent variable in the regressions is the logarithm of chemical input expenditure. To account for the endogeneity of off-farm incomes for both spouses, the literacy level of each spouse and the average off-farm earnings of the neighbours are used as instruments. Empirical results show that while males' off-farm earnings lead to increased farm expenditures for both commercial and non-commercial households, female non-farm earnings positively affects farm input expenses for commercial households only. The results indicate that when male heads in farm households access off-farm income, higher farm input expenditures occur. Apart from confirming income non-pooling among the sampled households, the analysis also supports the rationale that the spouse that controls income dictates the patterns of spending and this is potentially detrimental for farm productivity particularly for non-commercial households. A key takeaway is that, promoting rural off-farm income projects, though they may increase overall household income, may not have the effect of increasing agricultural productivity if households are non-commercial and women dominate off-farm activities. However, if non-commercial households can be nudged toward greater production, the off-farm incomes of both spouses would contribute to greater farm input expenditure and hence greater productivity in agriculture.