William Davidson Institute Working Paper 479
The main finding is striking: up to a point an increase in the industry specific minimum
wage relative to industry average wages, reduces wage inequality in the covered sector when
inequality is measured with the Gini and the Theil mean log deviation. However, it has no effect
on the covered sector when measured with the coefficient of variation. The toughness measure
has no impact on the earnings inequality of the self-employed in all three regressions. The effect
on the covered sector is non-linear and it is decreasing at a decreasing rate.6 As seen in Table 2,
at the lowest level of toughness we find in the data (30 percent), a one percentage point increase in
the toughness measure reduces inequality by 2 to 3 percent, depending on which measure is used. A
unit increase at the median level of toughness (55.5 percent) reduces inequality between 0.9 and 1.7
percent but at the highest level of toughness (89 percent) a one-percentage point increase actually
increases inequality (but only by 0.4 to 0.6 percent). Hence as long as the minimum wage is not set
“too high,” it reduces inequality.7 One explanation offered for this is that enforcement is even less
stringent when the minimum is very high.
3.3 Employment
Contrary to the traditional view, we find that up to a point minimum wage increases are
associated with increases in the level of total and covered sector employment (see Table 3). We
find no statistically significant correlation between the toughness of the minimum wage and the
number of self-employed. The relationship between the minimum wage toughness measure and the
level of employment is non-linear (for both total and covered sector): an increase in the toughness of
the minimum wage increases employment at a decreasing rate.8 As seen in Table 4, total
employment rises by 1.1 percent and covered sector employment rises by 1.3 percent when the
toughness measure rises from 30 to 31 percent. The employment effect is positive until the
toughness measure reaches between 68 and 75 percent (for total employment and covered sector