Understanding Income Dynamics: Identifying and Estimating the Changing Roles of Unobserved Ability, Permanent and Transitory Shocks∗

Sustained growth in earnings and wage inequality over the past few decades (most notably in the U.S. but also in many other developed countries) has generated widespread interest in both its causes and consequences, spurring large bodies of research in labor economics, macroeconomics, and growth economics.1 Perhaps, the greatest efforts have been devoted to understanding the role of skills, observed and unobserved, in accounting for the evolution of wage and earnings inequality. There is widespread agreement that the returns to observable measures of skill like education and labor market experience have increased dramatically since the early 1980s (Card, 1999; Katz and Autor, 1999; Heckman, Lochner, and Todd, 2006, 2008). There is greater disagreement about the evolution of returns to unobserved abilities and skills (e.g., see Card and DiNardo (2002); Lemieux (2006); Autor, Katz, and Kearney (2008)).2 More generally, the broader literature has yet to reach a consensus on the factors underlying changes in residual wage and earnings inequality (i.e. inequality conditional on observable measures of skill like educational attainment and age/experience). Figure 1 shows that the evolution of total inequality in log weekly wages and earnings for 30-59 year-old American men is closely mirrored by the evolution of residual inequality, and while the ∗We thank Youngmin Park for excellent research assistance, Peter Gottschalk and Robert Moffitt for sharing their PSID data creation programs, and Maria Ponomareva and Steve Haider for their useful suggestions. We also thank seminar and conference participants from Carnegie Mellon University, Ohio State University, University College of London, Uppsala University, and the MSU/UM/UWO Labor Day Conference. Lochner acknowledges financial research support from the Social Sciences and Humanities Research Council of Canada. For example, see surveys by Katz and Autor (1999), Acemoglu (2002), and Aghion (2002). Taber (2001) argues that increasing returns to unobserved skill in recent decades may be the main driver for the increase in measured returns to college, since individuals with higher unobserved skills are more likely to attend college.