Most of the empirical literature in this area tends to analyze labor allocation decisions of economic agents using cross-sectional data. But such methods implicitly assume that model parameters are stable (constant) across firms and over time. The use of cross-sectional methods is therefore glaringly at odds with the firm-specific aspects of the theoretical models employed in labor economics. Using a large panel data this study investigates the simultaneous relationship between farming efficiency and the off-farm labor supply decisions of both farm operators and their spouses. We also account for unobserved heterogeneity and correcting for simultaneity bias in such estimation. Results reveal several interesting findings. First, farming efficiency (ratio of farm revenue to total variable cost) has a positive and negative impact on hours of off-farm work by farm operators and spouses, respectively. Second, agricultural subsidy has a negative and positive effect on off-farm work hours of farm operators and spouses, respectively. Finally, we find a dynamic relationship between off-farm labor supply and farming efficiency. Specifically, in the case of the farm operator, off-farm work first increases farming efficiency in the first period and then decreases it in the second period. On the other hand, we observe a positive correlation, for both periods, between off-farm hours worked by spouses and farming efficiency.