Human Capital Prices, Productivity, and Growth

Abstract

Separate identification of the price and quantity of human capital has important implications for understanding key issues in labor economics and macroeconomics. Price and quantity series are derived and subjected to robustness checks. The human capital price series associated with different education levels are highly correlated and exhibit a strong secular trend. Three resulting implications are explored: (1) using the derived quantities life-cycle profiles are re-examined; (2) the rising college premium is reinterpreted and found to be mainly driven by relative quantity changes, and (3) adjusting the labor input for quality increases dramatically reduces the contribution of MFP to growth. * The authors wish to thank Lance Lochner and Todd Stinebrickner for helpful comments and discussion.

The slowdown in American educational attainment

“…22 Because of the structure imposed on price growth, in particular the choice of ω, price growth between 1970 and 1980 also matches Bowlus and Robinson (2012) estimates for the period. See section 3.2.…”

“…Therefore, I use life-cycle earnings growth late in the lifecycle for the 1884 to 1958 cohorts to discipline price growth ( Figure 6, panel (c), solid lines). This methodology for recovering price growth was originally proposed by Heckman, Lochner, and Taber (1998) and more recently used by Bowlus and Robinson (2012) to measure the prices of human capital across various education levels.…”

Reassessing the trends in the relative supply of college-equivalent workers in the U.S.: a selection-correction approach

“…The idea that assuming fixed efficiency labor supply units in the construction of the relative supply series may lead to miscalculations is related to Carneiro andLee (2011) andBowlus andRobinson (2012). Carneiro and Lee (2011) argue that time variation in school quality may be used as a proxy to endogenize the fixed efficiency units.…”

“…As a result, the fixed efficiency units setup overestimates the growth rate of relative supply of college workers, which means that the slope of the SBTC trend has been overestimated in the literature. Bowlus and Robinson (2012) show that estimating the changes in the efficiency units over time poses a serious challenge of identification. 7 They find that both quality and quantity changes are important and should 6 See, e.g., Acemoglu (1998Acemoglu ( , 1999, Ciccone andPapaioannou (2009), andIranzo andPeri (2009).…”

“…7 Separating the changes in quality from those in quantity has been a difficult identification problem in the empirical human capital literature [Weiss and Lillard (1978)]. Bowlus and Robinson (2012) develop an identification strategy motivated by a be accounted for.…”