Does Non-Farm Income Raise Farm Productivity? New Evidence from India
利用印度八邦农户面板数据，发现非农收入与农业生产力呈U型关系，且影响因土地规模而异，为理解非农收入对农业的复杂作用提供了新证据。
On the face of various agro-climatic shocks, farmers often resort to non-farm activities as a coping mechanism for attaining sustainable livelihood. While the impact of non-farm income on farm productivity is well documented in the literature, we contribute to the literature by exploring the possibility of a non-linear impact of non-farm income on farm productivity - differential impact across different types of farmers (small, medium and large) based on their landholding size. We first develop a small theoretical model and establish various channels which could result in non-linearity. For empirical estimation, we use household-level information from eight Indian states based on Village Dynamics in South Asia panel dataset for a period of five years (2010–2014). We adopt an Instrumental Variable Tobit Model to correct for endogeneity of non-farm income arising from unobserved heterogeneity and simultaneity between non-farm income and farm productivity. Our empirical results unveil a U-shaped relationship between non-farm income and farm productivity. This confirms our hypothesis about non-linear impact of non-farm income on farm productivity. Our heterogeneity analysis shows that the impact of non-farm earnings on farm productivity varies with size of landholdings. Policy implications of our empirical results are also discussed.