This research utilizes new data to track the earnings effect of the National Supported Work experiment (NSW) on the youth and Aid to Families with Dependent Children (AFDC) target groups for 8 years following training. The research indicates that the NSW's effect on the AFDC recipients ranged from $375 to $525 in 1978 dollars during the years 1982-86. The sum of the estimated earnings effects for the AFDC treatments over the observed posttraining period more than offset the costs of the training program. The NSW was found to have no discernible effect on the earnings of the youth target group.

We estimate the employment effects of federal minimum wage increases using monthly Current Population Survey (CPS) data from 1979 through 1997. We find that the empirical differences in the new minimum wage literature based on CPS data primarily can be traced to alternative methods of controlling for macroeconomic conditions. We argue that the macroeconomic controls commonly included in models where no employment impact is found are inappropriate. We consistently find a significant but modest negative relationship between minimum wage increases and teenage employment using alternative controls or allowing employer responses to the policy to occur with some delay.