Linkage Between Rural Non-Farm Income and Agricultural Productivity in Nigeria: A Tobit-Two-Stage Least Square Regression Approach

Multiple motive prompt agricultural households to diversify their income activities. Some of these activities were due to the "push" and "Pull" factors. The consequence(s) of these factors are the widespread households' income diversification. One of such is Non-farm activity which seems to offer a pathway out of poverty. This study therefore examines the effect of rural non-farm income on agricultural productivity in Nigeria. Data from Nigeria General Household Survey–Panel 2010 was used to generate information on households' access to non-farm income, agricultural production, investment and socio-economic characteristics. Data were analyzed using descriptive statistic (mean, frequency and percentage) and inferential statistics such as Tobit regression and Two Stage Least Square (2SLS). Tobit regression was used to analyze factors influencing non-farm income in the study area. While 2SLS model was further employed to analyze the effect of non-farm earnings on agricultural productivity of the respondents. In this study, the descriptive result shows that the mean age of households' head in Nigeria was 52 years, with most households being married and having mean households' size of about 12 persons. Also, the majority (88.3%) of the households were male headed. The mean non-farm income among the participating households was inline graphic 251, 723 ($1,678.18) per annum. In addition, the mean value of surplus crop produced was inline graphic 34, 274 ($228.49) while mean area cultivated was 1.99 Ha per respondent. Also, of the six regions in the country, South west was the most favourable to earnings in both cropping and non-farm activities. The households participating in non-farm activities were more productive agriculturally than their non-participating counterparts. Tobit regression result of the factors influencing non-farm income across rural areas in Nigeria indicates that estimates of equation of the model are jointly significant at 1% level of significance. The pseudo R square was 38% and from the thirteen included variables only four (Educational attainment, non-farm enterprise investment, sex of households' head and marital status) were statistically significant at different levels. Furthermore, the effect of nonfarm income on agricultural productivity is positive and significant at 5% in both the 2SLS and Ordinary Least Square Regression (OLS). Educational attainment of the household and capital investment significantly increase the ability of a typical rural household in Nigeria to earn non-farm income. The important linkage between farm and non-farm activities among rural households in Nigeria therefore suggests that attention needs to be given to non-farm sector for rural development as non-farm activity was not only a source of income for the participating household but a source of investment fund to boost agricultural productivity.