Home Address:
  5503 Copeland Court
  San Jose, CA 95124

Telephone: (408) 358-3697

Labor Economics
Game Theory and Microeconomics
Financial Economics

Microeconomics
Labor Economics
Game Theory

2000 (Oral) Microeconomic Theory, Financial Economics
1999 (Written) Macroeconomic and Microeconomic Theory

The Structure of U.S. Wages: Occupational Choice and Policy Implications

Professor Michael Keane
Professor John Roemer
Professor Giuseppe Moscarini

May 2006

Ph.D., Economics, Yale University, May 2006 (expected)
M.Phil., Economics, Yale University, 2001
M.A., Economics, Yale University, 1999
B.A., Economics, University of California, Berkeley, with High Distinction, 1995

Economics Department Fellowship, Yale University, summer 2002
Yale University Dissertation Fellowship, spring 2002
Yale University Graduate Student Fellowship, 1998–2002
Phi Beta Kappa, University of California, Berkeley, 1995
Edward Frank Kraft Scholarship
UC Berkeley Honorary Scholarship
UC Berkeley Class of 1906 Scholarship
San Francisco Archdiocese Scholarship

European Economic History, Fall 2003, Spring 2001
American Economic History, Fall 2001
Introductory Microeconomics, Fall 2000

Research Associate: RAND, Summer 2001
Investigated wage growth among science and technical workers relative to the defense industrial base.
    Results presented at the RAND Pittsburgh workshop.
Research Assistant: Professor Michael Keane, 2001–2002
Research Assistant: Professor Ann Huff-Stevens, 2000
Research Intern: AFL-CIO, California Labor Federation, 1994

Senior Consultant, KPMG LLP
Litigation and Forensic Investigative Services Group (San Francisco), 1995-1998
Conducted research in support of assessments of economic damage in complex civil litigation, negotiation, and arbitration. Areas of specific project experience include valuation of economic damages arising from intellectual property infringement, breach of contract, business interruption, anti-competitive and unfair business practices, and investigation of investment management fraud.

Professor Michael Keane
Department of Economics
Yale University
PO Box 208264
New Haven, CT 06520-8264
Phone: (203) 432-3548
Fax: (203) 432-6323
Email: michael.keane@yale.edu

Professor Giuseppe Moscarini
Department of Economics
Yale University
PO Box 208268
New Haven, CT 06520-8268
Phone: (203) 432-3596
Fax: (203) 432-2128
Email: giuseppe.moscarini@yale.edu

In this dissertation, I develop a tractable equilibrium model of the evolution of schooling choice, employment decisions and the distribution of earnings in the U.S. over the 1968-1996 period. In order to address several of the broad phenomena in the modern U.S. labor market, a framework for the structure of wages is often required. However, most frameworks employed to date, and particularly those estimated in an equilibrium setting, are able to capture only narrowly focused aspects of employment and wages. Prior studies have typically modeled the structure of wages in a partial equilibrium setting with an eye toward investigating or isolating the roots of a specific phenomenon. In most recent cases, the focus lay on the rise in education-based wage differentials. This obscures from view several important and interesting developments. For example, across occupations, education skill growth and wage growth typically act in concert, though there are several exceptions. While craft workers and laborers are among the most rapidly educating occupations, their wage declines are worse than all but transport operatives. Furthermore, there is often no clear link between average education skill levels and occupation wage behavior. While some occupation input groups converging in education also exhibit converging wage paths (e.g. male service and clerical workers) others have wage paths that diverge considerably (e.g. managers and technicians).

While partial equilibrium models are adept at capturing more detailed behavior of wages such as those described above, they are unable to reliably predict implications of policies on the wage structure since skill supplies and prices are taken as exogenous. On the other hand, equilibrium models that endogenize the determination of skill supplies and prices lack the detail necessary to investigate policy implications in a comprehensive manner. The model developed in this dissertation improves upon each in its ability to capture detailed evolution of the wage structure with 160 labor inputs differentiated by education, occupation, age and gender. Hence, policies may be evaluated by examining their effects on the overall wage structure and the allocation of workers across occupations in the alternate equilibrium.

The first chapter develops and estimates an overlapping generations static equilibrium model of schooling and occupation choice that allows for a detailed and surprisingly precise fit of the evolution of wages and employment from 1968 to 1996 along several dimensions: occupation, education, age and gender. The model incorporates education choice for each entering cohort as well as labor supply/occupation decisions (over ten occupation alternatives and home) for sixteen gender-education-age groups in each period. Therefore, supplies of each of the 160 distinct labor inputs are determined endogenously, with the equilibrium in each period described by a vector of wages such that markets for each labor input clear. An alternative model is also developed where productive capital input is also determined as a product of the equilibrium. The model is underpinned by a flexible CES production technology that allows for estimation of the degree of substitutability between each input. Estimation of both models relies on a method of moments estimator to find production function and preference parameters that best fit wage, occupation choice probability and education attainment data from the Panel Study of Income Dynamics (PSID) spanning 1968 to 1996. The fit of the model to wages and employment relies crucially on capital-skill complementarity embedded in the production technology, with skill defined along occupation (rather than education) classification as well as imperfect substitution in production between education levels, genders and age groups.

The second chapter evaluates education tax and subsidy policy experiments utilizing the equilibrium model of schooling, occupation choice, and wages developed and estimated in the preceding chapter. The equilibrium setting of the model allows for the evaluation of the effects of several counterfactuals or current policy choices when extended to future periods beyond the initially estimated 1968 to 1996 span. Unlike partial equilibrium analysis, equilibrium feedback effects on wages, employment and education choices are accounted for in evaluating the impact of policies on the welfare of individuals. The focus of the experiments performed here lies on education subsidy experiments aimed at equating incentives to attend college across individuals with differing parental education backgrounds. Initial types distinguishing individuals as they enter the model are based on gender and four levels of parental education (highest attained among both parents). For each type, a fixed non-pecuniary cost to college attendance based on parental education background is incorporated in the model and estimated in the prior paper. Differences in these costs are converted to dollar amounts, which inform subsidy levels that would equate incentives across types – with subsidies targeting those types with lowest parental education attainment. Not surprisingly, model estimates indicate effects for parental education figure prominently in the value to college among entrants. As differing predicted benefits from similar education subsidies using other equilibrium models of schooling and wages (e.g. Lee 2002 and Heckman, Lochner, and Taber 1998) highlight, the degree of feedback effects depend critically on model structure. This model yields substantive feedback effects on skill prices and education choices for later cohorts, though benefits to targeted types remain significant.

The third chapter estimates the impact of minimum wage legislation in the 1990s at both the state aggregate and individual agent panel levels. First, controlling for interstate differences and focusing on teenage employment, I explore employment effects of rising wage floors using Current Population Survey data. The sample covers the 1990s in which several changes in both the federal and state minimum wage policy took place. Second, I exploit the more detailed nature of individual panel data from the PSID to estimate the changes in the likelihood of those bound by the minimum wage increases to be employed following the change in policy. Here I investigate the substantial activity in minimum wage policy at both the state and federal level primarily in the years 1990 and 1991. Overall, I find no detrimental employment effects from the rising wage floors. In addition, the results provide evidence that those with higher individual-specific "propensity for turnover" effects are more likely to keep their jobs under higher minimum wages. This lends evidence to the positive predicted effect of higher minimums via reduced labor turnover mitigating gaps between marginal product and the minimum wage.