A recently developed quantile regression technique, which parsimoniously describes the entire conditional distribution, is applied to every March Current Population Survey since 1964. The study examines changes in the returns to schooling and experience at different points of the wage distribution and changes in within-group wage inequality. The results from the one-group and sixteen-group linear models show that the returns to schooling and experience differ across quantiles of the wage distribution but that their patterns of change are similar. Significant differences in wage inequality are also found across the various skill groups. Copyright 1994 by The Econometric Society.

We present a dynamic model of individuals’ educational investments that allows us to explore alternative modeling strategies for forecasting future wage distributions. The key innovation we propose is an approach to forecasting that relies only on the information that would be available at the actual time decisions are made and which incorporates the role of parameter uncertainty into the decision‐making process. We compare the performance of our method with alternative models of forecasting behavior, based on CPS data over the period 1964–2004.

This paper examines the mobility of individuals through the wage and earnings distributions, using 1979-1991 data from the National Longitudinal Survey of Youth. Lifetime wages will be more equally distributed than wages from any single year if individuals change position in the wage distribution over time. The results suggest that mobility is predominantly within group mobility, reducing wage inequality by 12%-26% over a four-year horizon. A detailed examination of within-group mobility, using year-to-year estimates of transition probabilities among quintiles of the distribution, reveals similar general patterns across all skill groups: mobility declined significantly over the years, especially at the lower end of the wage and earnings distributions.

This paper considers the problem of choosing the number of bootstrap repetitions B for bootstrap standard errors, confidence intervals, confidence regions, hypothesis tests, p-values, and bias correction. For each of these problems, the paper provides a three-step method for choosing B to achieve a desired level of accuracy. Accuracy is measured by the percentage deviation of the bootstrap standard error estimate, confidence interval length, test’s critical value, test’s p-value, or bias-corrected estimate based on B bootstrap simulations from the corresponding ideal bootstrap quantities for which B��. The results apply quite generally to parametric, semiparametric, and nonparametric models with independent and dependent data. The results apply to the standard nonparametric iid bootstrap, moving block bootstraps for time series data, parametric and semiparametric bootstraps, and bootstraps for regression models based on bootstrapping residuals. Monte Carlo simulations show that the proposed methods work very well.

In this paper, we expand on the seminal work of Altonji and Shakotko (1987) and Topel (1991) and reinvestigate the returns to seniority in the United States. We begin with the same wage equation as in previous studies. We extend the model of Hyslop (1999) and explicitly model the participation/employment and inter-firm mobility decisions, which, in turn, define the individual's experience and seniority. We introduce into the wage equation a summary of the workers' entire career path. The three-equation system is estimated simultaneously using data from the Panel Study of Income Dynamics (PSID). We find that for each of the three education groups studied the returns to seniority are larger than those previously found in the literature.

This paper introduces an alternative estimator for the linear censored quantile regression model. The estimator also applies to cases where the censoring point is unknown. Since the objective function is globally convex and the estimator is a solution to a linear programming problem, a global minimizer is obtained in a finite number of simplex iterations. The estimator has a square root of n-convergence rate and is asymptotically normal. A Monte Carlo study performed shows that the suggested estimator has very desirable small sample properties.

We develop and estimate a comprehensive dynamic programming (DP) model for the joint decisions of residential location, employment location, occupational choices, and labor market outcomes. We use data on immigrants from the former Soviet Union (FSU). We provide an extensive empirical evaluation of policies that have been designed to affect the residential and employment location decisions of the migrant population. The results shed new, and important, light on several issues regarding this group of immigrants. We find large regional differences in wages for the white-collar workers, but only little differences for the blue-collar workers. A careful examination of a number of policy measures indicate that a direct subsidy, in the form of a lump-sum transfer, is most effective in achieving the government stated goal of inducing people to reside in the northern region of the Galilee and southern region of the Negev. Other policies, such as rental and wage subsidies, can also be quite effective, but these are more difficult to administer.