Monday, February 28, 2011

WI Public Sector Compensation: Examining the EPI Data

In my previous post on why the unions aren't winning the Wisconsin protests, Dean brought up a recent study showing that public sector workers in Wisconsin are undercompensated by 4.8%. The study was produced by EPI (Economic Policy Institute), which the New York Times describes as,"a nonprofit research organization supported by labor unions."  Predictably, some on the right, less sympathetic to organized labor, have denounced the study as flawed.
That both sides of this issue can look at the facts and draw opposing conclusions says more about statistics than anything else, but does this mean we should just throw up our hands and not even consider the question?  If I thought that was the right response, I wouldn't bother writing this blog.  So by giving EPI the benefit of the doubt and looking at the data, here is what I came away with regarding public sector worker compensation in Wisconsin.

1.  The 4.8% differential is not the same for all levels of education.  Workers with the least amount of education actually do better in the public sector than the private sector.  As education level increases, the wage discrepancy becomes greater.  I have seen this result reported elsewhere, and I believe it is relatively non-controversial.

2.  On average, public sector workers have more education than those in the private sector.

3.  This means that, on some level, the protesters in Madison are fighting for the wages of a workforce that is more highly educated on average.  Admittedly, calling the budget repair bill an, "assault on the middle class," is rhetorically more effective than, "these college graduates can't afford wage cuts."

4.  The percentage of teachers in the public sector work force have a dramatic impact on the results.  When you don't account for hours worked, EPI found that Wisconsin workers were, "undercompensated by 8.2%," but when they controlled for hours worked that number decreased to 4.8%.  EPI reports:

Full-time public employees work fewer annual hours, particularly employees with bachelor’s, master’s, and professional degrees (because many are teachers or university professors).

5.  To say that teachers work fewer hours than private sector workers is not a slight.  Having holidays and time off in the summer go with the job, the same way that working swing shift goes with some mill jobs.  Common sense would indicate that there is a value to having time off and that this value would show up as a discrepancy in compensation between public and private sector workers.  EPI downplays this with their control for hours worked.  My question on this, then, is whether all hours are treated equally.  Couldn't part of the compensation differential, even after adjusting for total hours worked, reflect the fact that teachers spend some of the most valuable hours away from work (i.e. at the holidays and in the summer)?

Finally, we should try to grapple with why it is that the compensation differential gets worse as education increases.  This reflects the fact that the rate of wage growth due to education is much higher in the private sector, but why should that be.  I see two possible reasons:

7.  First, for some jobs requiring high levels of education cultural and political norms are a barrier to paying compensation equal to the private sector.  As a society we may simply believe that these folks should be motivated, in part, by a desire to do public service.  For example, do you think we are ever going to pay SEC employees salaries comparable to what they could get on Wall Street?  I don't.  At the same time I suspect this is a relatively small factor in the compensation differential story.

8.  The other reason may, once again, come back to the presence of a large number of teachers in the public sector work force.  According to the EPI data 22% of state and local government workers had a master's degree, compared with only 5% of the private work force.  According to the US Department of Education, in 2007-2008 there were over 175,000 master's degrees conferred where the field of study was education.  Education has been the most popular field of study for master's degrees at least since 1970-1971.  The only other field that comes close is business.  Given that so many public sector workers work in the education field, the relative abundance of master's degrees in education could be contributing to the public sector wage differential.  It is true that EPI controls for level of education, but do they treat all master's degrees the same?  This would seem to be a mistake given education's popularity as a field for graduate level study.

While the discussion continues here in Wisconsin and beyond, you will no doubt continue to hear that public sector workers are undercompensated by 4.8%.  I hope that this discussion has shown you can be skeptical of that number, even without resorting to partisan attacks on the study's authors.  For me, the telling sing of the day when most public sector workers are undercompensated will come when we are no longer able to hire and retain workers.  In a timely bit of blogging from Eggster, we see that we are nowhere near that point yet


Dean Weichmann said...

Jeremy, thanks for taking a look. I admit I thought you would not comment.

As to the big picture must agree that the distribution of wealth has far too one sided.
The rich have become richer and the rest poorer including the middle class. This is a divide and conquer stratagy, summed up by a joke I just saw;
"A unionized public employee, a member of the Tea Party, and a CEO are sitting at a table. In the middle of the table there is a plate with a dozen cookies on it. The CEO reaches across and takes 11 cookies looks at the Tea Partier and says,"look out for that union guy, he wants a piece of your cookie."

You can argue the fine points of whether teachers are over compensated or not or how much but the truth is that is not the problem.

How can walker possibly argue with a straight face that union workers are overcompensted and still give out more tax breaks for business?

J. Strupp said...

"How can walker possibly argue with a straight face that union workers are overcompensted and still give out more tax breaks for business?"

Because tax cuts pay for themselves Dean.

d said...

J, get real, it is myth that tax cuts pay for themselves. Wishful thinking of the highest order.

This is just the Lafer curve with an assumption that we are on the end of the curve. We may have been when the highest marginal rates were 91% not now.

Anonymous said...

I think that eggster post is curious and wrong.

In the private sector, if I change employers within my field of expertise, my experience will be recognized and my compensation will match that. It is to my advantage to periodically change employers as the opportunity fits. Today I am learning the skills required for my next higher paying job 

As a teacher in the public sector in my state, experience is not recognized outside of a particular school district. Therefore it is a disadvantage to change school districts because of the resultant pay cut. Example: A teacher with ten years of experience in District A moves to District B. That teacher will be paid the same as a first year teacher. Why would a teacher want to change if the result is a pay cut?

It is not an apple to apples comparison.

Jeremy R. Shown said...


Struppster was being sarcastic. He doesn't really believe tax cuts pay for themselves.

J. Strupp said...

Maybe I changed my mind Jeremy ;)

Jeremy R. Shown said...


Why won't one school district recognize valuable experience earned in another?

Anonymous said...

I don't know (government....what do you expect?). It makes no sense to me. But it is a common problem that most teachers face.

Begonia said...

I agree with Anonymous that the Eggster graphic is problematic and an oversimplification. It is especially troubling to me that you would included it after your detailed analysis of the EPI data, which carefully tried to control for all sorts of different variables.

What is the education level of the "private sector" workers in the graphic? What is the average education level of the "public sector workers" in the graphic? If those were controlled, I would think the graphic would look different. Are we comparing low-wage employees in the private sector to salaried professionals in the public sector?

There is also an argument to be made that many public workers who aren't teachers find it difficult to be competitive in employment in private industry. A lot of the information that public workers have to master (arcane regulations, government processes) may make them a valuable public employee, but don't translate well to the skills needed in the private sector.

I am a public employee looking for employment in the private sector and I find that many of the qualifications that the private companies are looking for I could only have gained if I were in the private sector. It's very frustrating.