Abstract
This paper investigates the impact of farm-nonfarm diversification (FND) on household income and food expenditure in urban Ghana using propensity score matching (PSM) technique to account for potential selection bias. We find diversified households to be statistically different from undiversified households in terms of household characteristics. Age, gender, educational attainment of the household head, household size, ownership of livestock and agricultural land, and receipt of miscellaneous and rent incomes are positive and significant determinants of FND in urban Ghana. In addition, we find that participation in both farm and nonfarm activities positively and significantly impacts household income and food expenditure. In the light of growing urbanization, with its implications for unemployment, poverty and food insecurity, we recommend diversification among urban households as a means of smoothing income and consumption.