Much of the recent attention to higher minimum wages has focused on the well publicized report on the New Jersey fast food industry by David Card (Princeton University) and Alan Kreuger (Princeton University and the Department of Labor). Summarizing that paper briefly, the authors conducted a private survey of approximately 330 fast food restaurants in New Jersey and another 80 in neighboring Pennsylvania. Over the 7-8 months following New Jersey's increase in the minimum wage to $5.05 they failed to observe a decline in employment.
The publicity that this paper has received, however, should not be interpreted as a revolution in economic thinking. Although the White House referred to "a dozen" studies which found no job loss from higher minimum wages, four of them were by authors with connections to the fast food studies, one was an unpublished paper by the Economic Policy Institute, and two dealt with overseas experience (including a paper dating to 1970). The President's statement in the State of the Union that the "weight of evidence" shows no job loss is clearly not true if he wants to go back to the 1970s. The weight of the evidence over the last 25 years clearly supports the traditional view that higher minimum wages reduce employment. If we restrict ourselves to more recent literature, the weight of evidence continues to support the view that higher minimum wages reduce overall employment.
Although the New Jersey study was done by well respected economists, the limitations of this study cannot be glossed over. This is a study of one sub-sector of the retail trade industry. Employment in this sub-sector is by no means representative of the universe of minimum wage employment. Other studies which have looked at broader industry sectors have indeed found job losses.
Among the new research finding job loss from higher minimum wages is a paper by Dr. Lowell Taylor of Carnegie Mellon University (The Employment Effect in Retail Trade of a Minimum Wage: Evidence from California, the Employment Policies Institute and forthcoming in the Journal of Business and Economic Statistics). Using the changes in state-level minimum wages (which were also used in the fast food studies), Dr. Taylor found that significant employment lost jobs as well as declines in employment growth took place as a result of these wage increases. In contrast to the fast food studies which attempted to measure employment changes over a period of months in one small segment of minimum wage employment Dr. Taylor's work included all workers in all sectors of retail trade (including fast food), broken down into 64 industry sectors, analyzed on a county-by-county basis, over the course of a full year, in the nation's most populous state (California). Taylor's work showed a significantly depressing effect on job growth in retail trade following a minimum wage increase.
The easiest way to describe Dr. Taylor's findings is to describe the employment effect in a set of counties all of which experienced a 7 percent growth in retail sales in the year following California's increase to $4.25. In high-wage San Francisco County the wage effect was negligible, boosting average wages by less than 2 percent. In that year San Francisco's retail employment boomed, growing by 8 percent. In contrast, a low-wage county like Sutter saw wages in retail trade rise by over 9 percent. In that county retail employment fell by over 2 percent despite having the same retail sales growth as San Francisco. The more the minimum wage increase affected wages the more it reduced employment. This pattern persisted in all counties showing a 7 percent growth in retail sales volume.
Using data from all the counties in California, Dr. Taylor found that every 10 percent increase in average wages in retail trade caused by a minimum wage increase led to a 7 percent decline in employment.
Other recent research on minimum wages includes the work of David Neumark (Employment Effect of Minimum and Subminimum Wages, Employment Policies Institute and the Industrial and Labor Relations Review), now at Michigan State University. A key assertion of the new research finding no minimum wage effect on employment is that the effect of the minimum wage on labor demand has changed over time and that whatever effect it may once have had no longer applies. In contrast, Dr. Neumark found that, once corrections for mis-specifications in this new research were implemented, there continued to be strong negative effects on employment.
In summary, while there have been recent studies which have not found employment effects of higher minimum wages current research continues to find significant effects on employment. The President's "weight of evidence" is wishful thinking.
One additional area needs mentioning, the value of the minimum wage over time and its value relative to other wages in the economy. Administration officials have often noted that the minimum wage was at its highest inflation-adjusted value in 1967 and has declined ever since. What these statements ignore is that prior to 1966 most occupations which now employ large numbers of minimum wage workers were not covered by the minimum wage. This includes almost all employees engaged in retail, restaurant, hotel, laundry, food processing, agriculture and domestic services, where entry level work is concentrated. Including these workers invalidates simple time/wage comparisons. Only 56 percent of non-supervisory employees were covered by the minimum wage thirty years ago. Today, the comparable figure is 89 percent.
Whereas the minimum wage was formerly a minimum manufacturing wage it has now become a minimum service sector wage. Productivity growth in service occupations has always grown much more slowly than in manufacturing. In addition, the narrow focus on the minimum wage ignores other earnings of these workers. For example, hundreds of thousands of tipped employees are statistically minimum wage workers even though their tip income usually puts total earnings well above the minimum. The minimum wage has fallen because it now covers a much different employment sector than before. To link it to that earlier standard is to argue that low profit margin sectors of the economy must pay the same wages as the higher profit manufacturing sectors that formerly dominated the economy.
It is also frequently remarked that the minimum wage was 50 percent of average wages in the 1970s and has now fallen to only about 37 percent. Such a simple statement, however, fails to capture an important phenomena. Over that period of time the baby boomers have gone from being new entrants in the workforce and into their high wage earnings years. The large number of baby boomers skews any measure of average wages in the economy simply by force of numbers. Fully half of the change in the relationship between minimum and average wages is a statistical artifice from the aging of the baby boom (How the Baby Boom Skews Measures of Income Inequality, Dave Macpherson, The Employment Policies Institute (forthcoming)). If we had indexed the minimum wage to average wages we would have over-indexed it by 100%.
Raising the minimum wage is advanced as an anti-poverty measure, that it will provide significant amounts of income to low-income households. Analysis of the last minimum wage increase, however, showed exactly the opposite effect. Because many low-wage workers are found in higher-income households, 57 percent of the last minimum wage increase (from $3.35 to $4.25) went to families with incomes at least twice the poverty level. Only 17 percent went to families below the poverty line. Thirty-six percent of the benefits went to families whose income was at least three times the poverty level, or close to $45,000 for a family of four. (Public Policies for the Working Poor: The Earned Income Tax Credit vs. Minimum Wage Legislation, Richard V. Burkhauser and Andrew J. Glenn, The Employment Policies Institute.) Given the relatively constant distribution of minimum wage employment, future increases will be distributed in a similar manner.
A second study focused on the ability of higher minimum wages to reduce poverty. Looking specifically at groups likely to be affected by an increase in the minimum wage, it found no reduction in poverty rates from the increases in the minimum wage that took place in the late 1980s. (The Effect of Recent Increases in the U.S. Minimum Wage on the Distribution of Income, John T. Addison and McKinley Blackburn, The Employment Policies Institute.) While some members of these groups may have realized higher earnings, those higher earnings were insufficient to offset the negative employment impact.
Return Home