Learning by Doing and Ben-Porath: Life-cycle Predictions and Policy Implications

Abstract Many government policies affect incentives to acquire human capital. Two workhorse models dominate the literature analyzing these policies: Learning by Doing (LBD) and Ben-Porath (BP). This paper makes two novel findings related to these models. First, LBD and BP generate different predictions for life-cycle earnings growth. BP predicts that the heterogeneity in earnings growth rates across workers should disappear over the life-cycle because the workers stop investing as they age. The LBD model cannot replicate this feature because workers accumulate human capital automatically throughout the life-cycle. Second, the same model features that generate different life-cycle predictions between BP and LBD also generate different implications for policies that affect the payoff to human capital accumulation. To illustrate this quantitatively, I show that increasing marginal labor tax rates for top earners depresses human capital accumulation more under BP. As a result, the Laffer curve for top marginal income tax rates is flatter, and peaks 10 percentage points lower, with BP versus LBD.