Employment Effects of the 2009 Minimum Wage Increase: New Evidence from State-Based Comparisons of Workers by Skill Level

A frequently-cited 1979 survey of professional and business economists found that 90% “generally agreed” or “agreed with provisions” that “a minimum wage increases unemployment among young and unskilled workers” (Kearl et al. 1979). This survey precedes the less-settled empirical literature of the past two decades.

We use “at-risk” here to refer to demographic groups that have a relatively high proportion of workers earning wages between the old and new minimum wage.

Card writes “The imposition of a national minimum wage standard provides a natural experiment in which the ‘treatment effect’ varies across states depending on the fraction of workers initially earning less than the new minimum” (Card 1992, 22).

Deere, Murphy, and Welch adopt the general approach but not the vocabulary or full method in their analysis.

The coding here differs slightly from that used in an EPI study by Even and Macpherson (2010), which finds that 19 states were unaffected by the July 24, 2009 increase. The EPI classification includes DC, IL, and KY, all of whom increased their minimum wage July 1, as states unaffected by the increase in the Federal minimum. Since our coding focuses on changes in the effective minimum wage between February and November, 2009, rather than its level on July 24, these three states are appropriately included among states with an increase. For more details, see Appendix Table 1.

Typical examples include Vermont, Washington, and Montana, all of whom by statute increase the state minimum wage at the same rate as the increase in the CPI from August to August for the preceding 12 months. Florida’s increase is based on the percentage change in the CPI for urban wage earners and clerical workers in the South Region for the 12-month period prior to September 1.

If groups j and k are related factors of production, the demand for k could be affected when the wage for j changes. Given the definition of groups we use and the magnitudes of impacts that we find, this impact is likely to be negligible.

The DIDID model includes additional interaction terms to capture the treatment effect.

This education restriction eliminates some 18- and 19-year olds who are college students and thus results in a less-skilled and younger group of teen workers that are more likely to be affected by a minimum wage increase.

About 0.5% of potential sample members are excluded because they are in the Armed Forces, including about 1.0–1.5% of the sample at ages 22–31.

Arizona, California, and New Mexico were, respectively, 30.1%, 36.6%, and 44.9% Hispanic and 4.2%, 6.7%, and 3.0% black in 2009. The national average is 15.1% and 12.3%.

That is why we do not use this for our employment analysis.

The tipped employee minimum was $2.13 in both 2008 and 2009, in spite of the increase in the Federal minimum.

There may be some exceptions to this depending on the extent of the tip credit in a state. In a state with a small tip credit, the proportion earning less than $6.55 might decline when the minimum wage increased to $7.25.

For ease of presentation, we do not present results for males, age 20–59 without a high school degree. Results are very similar to those shown for all age 20–59 without a high school degree.

This DIDID effect (0.0041) is, equivalently, the difference in the between-state effects for teens and prime-age males (=–0.0062–(–0.0103)) or the difference in the within-state effects (–0.0146–(–0.0188)). All other DIDID results can be calculated in the same way.

Standard errors are clustered at the state level.

The full results are available upon request.

The Hispanic coefficient in the between-state model measures the impact across all states for the sampled group (e.g. teens), while the corresponding within-state estimate measures the impact for a different age and geographic sample. Thus, the estimates could well be different.