Using a dynamic programming model of schooling decisions, we investigate the relationship between subjective discount rates and the labor market ability (the discount rate bias) on a panel taken from the National Longitudinal Survey of Youth (NLSY). Given household human capital and Armed Forces Qualification test scores (AFQT), subjective discount rates, which vary between 1% and 5% per year, are found to be negatively correlated with labor market ability. The true return to schooling is estimated around 6% per year. Estimates obtained from a model where neither the ability bias nor the discount rate bias are considered, indicate that one percentage point can be imputed to the correlation between discount rates and labor market ability and at least another percentage point can be imputed to the positive correlation between the per-period utility of attending school and labor market ability. The model is used to simulate the effects of an increase in the level of human capital of one generation on both schooling attainments and labor market productivity of the next generation. We find the true intergenerational education correlation to be relatively low; an increase of 1 year in the average level of schooling will raise the level of human capital of the next generation by approximately 0.15 year of schooling and translates into a 1% productivity (wage) growth.