To assess the desirability of higher minimum wages we typically focus on aggregate employment effects — how much a particular increase would lower overall employment. The current views on this question range from no job loss (according to the Clinton Administration) to a loss perhaps as high as 3 percent for teens (the workers most affected by the minimum wage) for every 10 percent increase in the minimum wage.
As David Neumark shows in this paper, however, a focus on the employment effect of a minimum wage increase on broad population groups presents a distorted view of its impact. Subsumed in that aggregate is a shift in the composition of the work force. Less skilled teens are displaced from the job market, while more highly skilled teens are lured in by higher wages (even at the expense of cutbacks in their educational attainment). The conventional measures of job loss, by ignoring flows of young workers in and out of employment, can result in a measure of net job changes that markedly understates the true magnitude of the employment effect for the lowest-wage workers whom minimum wages are intended to help.
Dr. Neumark goes beyond simple aggregate measures by focusing on the changes in the work and education status of individual teenagers aged 16-19 following an increase in the minimum wage. He finds that higher minimum wages have significant negative effects on the employment prospects of less skilled teens, losses which are masked by their replacement in the work force by more highly skilled teens. In addition, increases in the minimum wage are associated with an earlier age for leaving school. The well-documented earnings premium associated with continued education makes this a potentially troublesome source of long-run earnings loss.
Employment and Schooling Effects
Using state-level data spanning 1979-1992, Dr. Neumark is able to estimate the impact of a higher minimum wage — $5.15 an hour, as proposed by the President — on today’s young workers. He finds that the least skilled of these workers would suffer employment losses while the better skilled members of the cohort could enjoy employment gains.
Overall, the increase in the minimum wage proposed by President Clinton would increase, by about 2 percentage points, the number of youths aged 16-19 who are neither in school nor at work. This 2 percentage point increase in the probability of idleness in the entire teen population generates an approximately 20 percent increase in the number of idle youths. At the same time, the probability of being employed while in school falls by as much as 4 percentage points, decreasing the overall number of teens working while in school by 8 to 15 percent.
These employment changes are not distributed evenly across all youths, but rather are concentrated among those youths with the worst employment prospects. For example, youths already out of school but not employed — whether they left school prior or subsequent to graduation — experience a 4.4 percentage point increase in the probability that they will continue to be idle (neither in school nor at work) after a minimum wage increase. In contrast, young people initially working while in school are more likely to be employed upon leaving school after a minimum wage increase. In fact, the evidence shows they may actually be leaving school earlier in response to the higher wage.
Dividing teens into younger (16-17) and older (18-19) groups provides further evidence on the sorting process in the labor market. Younger idle youths have an almost 6 percentage point increase in their chances of continued idleness. In contrast, no significant effects are found for older teens. Idle youths aged 16-17 are highly likely to have left school prior to graduation and hence to have lower skills than either the larger universe of teens or even of older teens who are neither in school nor at work. The reduction in favorable labor market outcomes for this group is consistent with an increased demand for higher skilled workers following a minimum wage increase.
The effect of higher minimum wages is even stronger for minority youths. If these youths are idle — neither at work nor in school — before the minimum wage is raised, they have a 7.3 percentage point increase in their chance of continuing to be idle afterwards. If they had a job before the minimum wage rose, they face a 4.6 percentage point higher probability of becoming idle. A higher minimum wage also increases the probability, by 2.6 percentage points, that these youths will stop mixing work with education; they are more likely to leave school only to find themselves without employment.
Finally, Neumark finds that the relationship between a teen worker’s wage and the new minimum markedly affects the employment outcome. Those teens who were out of school and employed with wages below the new minimum showed a 4.5 percentage point increase in the probability that they would be subsequently non-employed. Notably, teens who already earned above the new minimum (and who probably hold higher skills) showed no increase in the probability that they would become non-employed.
Changes in the minimum wage, often thought to affect only aggregate employment levels, are now known to have impacts both in and outside the labor market. Inside the labor market, higher minimum wages affect the composition of the minimum wage work force, reducing the employability of less skilled workers. At the same time, higher minimum wages may accelerate the rate at which youths terminate their formal schooling. Limitations of the data preclude this study from following individual workers for more than two years, so we do not know if they ultimately acquire additional education. We do know, however, that declines in the level of educational attainment are associated with declines in lifetime earnings.