My research focuses on the demand for and supply of skilled labor in the North American economy.

On the demand side, my main interest is how new technologies affect the labor market, in terms of employment, wages, and the returns to skill. I have also done research on new methods of measuring the extent to which the demand for skilled labor has risen over time. The relation between new technology and labor is not a one-way street. Traits of the labor market may also affect the rate of output growth and/or productivity growth of industry. Accordingly, in recent work with Cameron Odgers I have studied how the level of unionization influences the rate of investment. We have found a negative link between the rate of unionization and investment rates in Canadian manufacturing industries.

Broadly speaking, my main interest on the supply side of the market for skilled workers is examination of methods for increasing the supply of skilled workers.

The bulk of this work has examined the determinants of school quality. This research uses school-level and in some cases classroom-level data-sets to study how the various components of school spending such as class size and teachers' salaries influence student achievement. The main finding of this research has been that additional spending on its own has little impact on student outcomes, whether measured in terms of test scores or earnings of workers after they leave school. Other factors, in particular the incentives systems within schools, seem to play at least as important a role as spending per pupil in determining the quality of schools.

A second issue on the supply side of the market for skilled workers is measurement of the impact of business cycles on college enrollment. While previous work has suggested little link between business cycles and college attendance, I have found in joint work with Laurel McFarland that community college enrollment rises strongly during recessions. Unfortunately, current funding mechanisms for higher education fail to take this fact into account fully.

A third issue which I have studied on the supply side of the market for skilled labor is the accuracy of students' beliefs about the labor market. Traditional economic models of occupational choice assume that workers have full knowledge of the returns to education, and of wages in different fields. Using a survey of over 1000 undergraduates, I have found that on average students make errors of only about 6% when estimating salaries by field of specialization. Unfortunately, over half of student learning about the labor market occurs during the fourth year of undergraduate study, well after most students have chosen their major. Much of the variation in beliefs stems from students' field of study and family background. For instance, students from families with lower income tend to make lower estimates of the salaries of college-educated workers than do students from richer families. The study suggests that there is a need to improve the information which young people receive about the labor market.