Monday, May 14, 2012

Does US male unemployment exaggerate gender gap?

Today Bloomberg discusses Barack Obama's attempts to win over female voters by taking a stand for gender equality. It points out, however, that the gender pay gap in the US actually increased under Obama's rule:
Meanwhile, the pay gap in 2010 showed that women earned 77.4 percent of men’s salaries, down from 77.8 percent in 2007 before the recession hit.
Why is that? Are American employers more sexist now than in 2007? 

It occurred to me that this period was one in which the construction sector collapsed, and with its collapse went a huge number of male jobs. The rapid rise in male unemployed in the US and UK even caused some commentators to talk about a 'mancession'

My guess was that the recession destroyed low-income manual work for American men, pushing those male workers in low-income jobs into unemployment. By wiping out a huge section of low-paid male workers, the remaining male workers disproportionately represented the higher-income bracket. The average or median income for males would seem to lurch forward, just because the poor men dropped out of employment. Since gender pay gaps are usually calculated on those individuals who are in employment, by cutting the poorer men out the gap would appear to widen. 

I'm not completely sure that this is the case; I don't have the time or data to explore it any further right now. I do see, however, this study by Northeastern University, 2010, showing that American unemployment rates for low-income workers did indeed rush forward when the recession hit:
Earlier work by the authors has shown that a disproportionate share of the losses in jobs and the increases in open unemployment were borne by males, the young (under 30, especially teenagers), the less well educated, blue collar workers especially those in the construction trades, and Black men.... The incidence of underemployment problems in the fourth quarter of 2009 was 13 times higher among those workers in the bottom household income decile as opposed to those residing in the top decile of the income distribution (20.6% vs. 1.6%).


  1. Seems quite plausible. The government has shed many jobs as well, especially at the state and local level in the US, presumably mostly low-paid workers. Perhaps the remaining better-paid workers of goverment are disproportionally male as well? Parliamentarians, mayors, governors and the like are certainly more likely to be male than female, I don't know if that applies to less visible major government positions as well?

  2. Seems plausible indeed Vidar.

    I'm interested in these cases where there is missing evidence as some groups of people are left out of the data. This is something Nassim Taleb discusses in The Black Swan. For example he points out that if we look at the traits of millionaires to find out the secret of success, we might be mistaken because we're not looking at the traits of all those who TRIED to be millionaires and failed, disappearing from our vision. Or it could be that risk-taking, for example, made some soldiers win wars - but also made millions of others die and vanish from the data.


Note: Only a member of this blog may post a comment.