December 21, 2015
The American Enterprise Institute posted a peculiar blog post last week, echoing a similar, older post from the Cato Institute. The author attempted to make a connection between unemployment and minimum wage laws in Organization for Economic Co-operation and Development (OECD) European countries, creating a table that purported to show that those countries with legal minimum wages had higher youth and total unemployment than those that did not.
There are some problems with this data and its presentation. The first thing to note about the table, which is accompanied by a question asking whether one would rather live in a country with a minimum wage or not, is that it does not include every European country nor does it even include every European OECD-member country. The second issue is that much of the data is incorrect or not directly comparable. Some of the errors include: Portugal’s harmonized unemployment rate (via the OECD) is 14.13 percent in 2014, not 13.9 percent; Ireland’s youth unemployment rate is 23.93 percent not 26.9 percent; Finland’s youth unemployment rate is 20.42 percent not 19.3 percent. In addition, the minimum wage values appear to be exchange rate values that fluctuate somewhat arbitrarily, rather than the OECD minimum wage values adjusted for purchasing power parity (PPP) for the year 2014. (Adjusting for PPP takes into account the different costs of living.)
Another issue (pointed out to author of the post) is that all of the countries that are listed as having “no minimum wage” have other mechanisms to create wage floors, such as national collective bargaining agreements or collective bargaining agreements by sector. The latter put a floor on the wages that can be paid by the bulk of employers in a sector. These effectively create minimum wages, and reasonable equivalents can be determined from those agreements. For example, for Iceland, a national collective bargaining agreement sets a clear minimum wage; for Norway, the lowest-paid sector under collective bargaining agreements can be used to determine a de facto minimum wage.
A more accurate replication of AEI’s table is below, including all European OECD countries with data available. (Notes are included on how all equivalent minimum wages are determined.)
As shown, the countries without legal minimum wages actually have, on average, higher equivalent minimum wages and lower youth and total unemployment than those with legal minimum wages. This suggests that opposite of AEI’s narrative that minimum wages correlate with (or even lead to) higher unemployment. It also suggests that there are probably historical and other reasons for favoring collective bargaining to set wage floors over legal minimum wages.
An update of the post acknowledges that countries have collective bargaining agreements, but argues that these agreements are not equivalent to minimum wages, since younger workers are less likely to be under these collective bargaining agreements.
In fact, these collective bargaining agreements create equivalent minimum wages that apply to the vast majority of the population — whether or not workers are union members. It is also important to note that the U.S. and other places with legal minimum wages often have exceptions, for example for tipped workers or those employed by very small businesses, so there are even fewer differences between legal minimum wages and wage levels that are collectively bargained.
It is more likely that the difference in average youth unemployment between these groupings of countries (only 7.53 percentage points when calculated accurately) is due to factors other than those introduced by wage floors. For example, Greece and Spain have very high youth unemployment. This is quite obviously due to the austerity that both countries have had to implement following the recession, not their relatively modest legal minimum wages. There may also be good reasons for a high youth unemployment rate in some countries. In France, and perhaps in other European countries with free or low-cost education, youth are more likely to be in school and not working at all. This causes the unemployment rate to appear high because the labor force is smaller. In the U.S., many students work part-time and are thus counted in the labor force, leading to a lower rate.
An OECD-wide comparison may also be helpful at discerning any relationship between minimum wage and unemployment. The two graphs below show youth or total unemployment versus legal minimum wages or equivalents.
Static images of the above charts are here and here. Data for both charts (all OECD countries) is available here.
The results are similar: the trend lines indicate a weak correlation between higher minimum wages and lower unemployment for OECD countries around the world. The result is not statistically significant, but in any case, clearly does not support the AEI story.
So what can be said about minimum wages? Unlike the conclusions from AEI, there is considerable research showing that modest changes to minimum wages have no discernible effect on employment. They also lead to reductions in poverty. Contrary to the claim in the AEI post, workers might prefer to live in a place that has either has a minimum wage or that allows unions to collectively bargain either nationally or sector-by-sector.