Economic Policy Institute Being Economical With The Truth Again

3
672

The Economic Policy Institute is terribly concerned that the Supreme Court might stop unions from being able to charge money to people who aren’t members of the union. This is the so called “fair share” argument in the Janus case. The EPI is always and everywhen against anything that would diminish the income or power of unions. Largely upon the reasonable basis that the unions pay for the existence of the EPI.

However, we still cannot let this pass:

The claim that government workers are overpaid is a legislative ploy used to cut pay and curb bargaining rights. State and local government workers already earn less than similar private-sector workers. In particular, comparing the hourly wages of state and local government workers with those of private-sector workers, after controlling for education, age, gender, race, ethnicity, state, and other factors known to affect pay, we find that workers in state and local government make between 3.7 percent and 8.2 percent less on average than their private-sector counterparts.

That’s nonsense, so much so as to be entirely scrotal. The way they’ve worked it out is, from their report:

To estimate wage differences, we use ordinary least squares to estimate a wage equation using
pooled Current Population Survey Outgoing Rotation Group (CPS-ORG) data from 2013–2017,
restricting the sample to wage and salary earners ages 18–64 who are either state/local
government workers or private-sector workers. For the base specification, we regress log hourly
wages on dummies for being a state/local government worker, level of education, race/ethnicity,
foreign-born status, gender, and state of residence, along with age and age-squared. The values
given are coefficients on the state/local government worker dummy; the range is the result of the
inclusion or exclusion of industry and occupation dummies in the specification. Note that by using
hourly wages in the regression, we have accounted for the fact that some workers—for example,
teachers—often don’t work all year.
6. To estimate wage differences, we use ordinary least squares to estimate a wage equation using
pooled CPS-ORG data from 2013–2017, restricting the sample to wage and salary earners ages
18–64 who are state/local government workers. For the base specification, we regress log hourly
wages on dummies for being in a union, level of education, race/ethnicity, foreign-born status,
gender, and state of residence, along with age and age squared. The values given are coefficients
on the union dummy; the range is the result of the inclusion or exclusion of industry and
occupation dummies in the specification. Note that by using hourly wages in the regression, we
account for the fact that some workers—for example, teachers—often don’t work all year.

“Pay” is everything that you get for going to work. Sure, we can play little games about the differences between pay, wages, and compensation, but we really shouldn’t have to do that with a think tank attempting to educate us upon public policy. We should be able to trust people when they make a statement like the above, that public sector workers earn less than private.

The sadness being that we cannot trust the EPI. For by far the largest difference between public and private sector pay is pensions. Which are just deferred wages after all. Varied reports talking about various such pension schemes pointing out that the public sector variety are equivalent to perhaps as much as a 30% pay rise.

Sure, we can quibble again about the extra value of those defined benefit pensions. But one thing we really do know is that the two sectors gain different pensions. Pensions are part of what you get for going to work, they’re “pay.” And the EPI doesn’t even mention pensions in its comparison of pay rates.

That is, they’re lying. Or if that’s too strong with you the EPI is being sufficiently economical with the truth that we cannot trust anything they say, can we?