We develop a human capital model with borrowing constraints explicitly derived from government student loan programs and private lending under limited commitment. Two key implications of our analysis are: (i) binding constraints may not depress investment; and (ii) a positive relationship between investment and ability is unlikely to arise in standard exogenous constraint models but arises more generally in our framework. Our model also helps explain a number of important empirical observations in the U.S. higher education sector since the early 1980s: (i) a strong and stable positive correlation between ability and college attendance for all income and wealth backgrounds; (ii) the rising importance of family income as a determinant of college attendance; (iii) the increase in the share of undergraduates borrowing the maximum from government student loan programs; and (iv) the dramatic rise in student borrowing from private lenders. In our framework, all of these are natural responses to the rising costs and returns to college (with stable real government loan limits) observed in recent decades. In contrast, the standard exogenous constraint model cannot simultaneously explain observations (i) and (ii) under standard assumptions about preferences; it is also silent on the rise in private lending. Finally, by incorporating both public and private lending, our framework offers new insights regarding the interaction of government and private student lending as well as the responsiveness of private student credit to economic and policy changes.