This approach is necessary for developing effective empirical strategies to produce credible estimates of the causal effect of drug use on labor market outcomes. Economic theory is well suited to this effort, because at its core it is a behavioral model of individual choice. The analysis of economic theory indicates that drug use is often related to labor market outcomes in ways that may not be obvious or consistent with the intuition that motivated study of the problem. Information on causal relationships is necessary to designing effective public policy and employer policy, the previous research has not provided credible evidence of a causal relationship between drug use and labor market outcomes. In addition, future research needs to use more of an experimental approach than has been the case to date. A final point to consider is the effect of existing policies on the relationship between drug use and labor market outcomes and the need to test the effects of alternative policies on the relationship between drug use and labor market outcomes. Table, notes, and 27 references (Author abstract modified)