For white males born in the United States after 1950, there is a stagnation in the fraction of high school graduates that go on to complete a four-year college degree. At the same time, across successive cohorts, those with a four year-college degree achieve increasingly higher lifetime earnings than those with a high school degree. What caused this phenomenon? I formulate a life-cycle model of human capital accumulation `a la Ben-Porath, where successive cohorts decide whether or not to acquire a college degree. Individuals can accumulate human capital both in college and on the job. Human capital accumulation in college requires time and goods as inputs, while accumulation on the job requires only time. Cohorts differ by the sequence of rental price per unit of human capital they face. Price growth influences an individual’s college decision in two ways: as it declines, 1) the return to human capital investment decreases and 2) the opportunity cost of human capital accumulation in college increases relative to that on the job because of the lower relative price of time. I calibrate the structural parameters of my model to life-cycle earnings data for the NLSY79 cohorts and the price sequence to earnings data for the 1920 to 1970 birth cohorts. My model reproduces the observed pattern in college attainment for the 1920 to 1970 birth cohorts. I find that the stagnation in college attainment that occurs for those born after 1950 is due to the decrease in the rate of growth of the rental price per unit of human capital commencing in the 1970s. My model also generates 79% of the increase in earnings for college graduates relative to those for high school graduates. Part of this increase is reinforced by a stronger association between college and ability.