(Note: This article initially contained a minor error in the wages and benefits calculations. In addition, the Bureau of Economic Analysis has released new data and revised previous data. The figures have been corrected and updated.)

A recent paper by the union-supported Economic Policy Institute purports to answer the question, “Are Michigan Public Employees Over-Compensated?” with a resounding “No.” The paper, however, uses a flawed approach relying on education (degree) groups without making occupational comparisons. The EPI paper also ignores valuable evidence of rich wage gains by workers who leave private-sector jobs to work for the government, and more.

On the first item, the author compares worker compensation in part by level of education, such as high school diploma, bachelor’s degree and master’s degree. But simply having a degree, regardless of what it is for, does not automatically demonstrate an equal comparison of occupational category.

For example, no one should be surprised that a Michigan Department of Education bureaucrat with a master’s degree in education may earn less than a person with an MA in engineering working for a private heavy construction firm. Yet under the EPI study’s methodology, the two positions should theoretically pay the same.

A more meaningful comparison is provided by Michael Van Beek, director of education policy at the Mackinac Center, who reports that in Kent County the average public school teacher enjoys a salary of $62,788, plus benefits equal to $26,486 annually. In contrast, Grand Rapids Christian Schools — which enroll about 3,000 students— pays teachers $46,576 on average, with benefits of around $15,553.

In other words, in a closer apples-to-apples comparison, being employed by the government comes with a $27,127 compensation premium, or 44 percent. Multiply that by 89,000 public school teachers in Michigan and it becomes clear how much EPI’s flawed method of “controlling” for education status skews their results.

Properly applied, education levels can provide useful explanations for differences in compensation levels, but what about comparisons of workers who are identical in skills, training and experience? A recent essay by American Enterprise Institute scholars Andrew Biggs and Jason Richwine described data that track individuals who move from private-sector to federal jobs. As Mackinac Center Senior Legislative Analyst Jack McHugh put it, “a worker who leaves a private-sector job on Tuesday finds himself making 8 to 28 percent more if he starts working for the government on Thursday,” with the latter figure referring to U.S. Post Office employment.

Those figures probably understate the wage differential, because they refer to wages only, not benefits. Although they cover federal workplaces only, it’s very likely that similar patterns exist at the state and local levels. Especially in Michigan, which over the past decade has witnessed extraordinary growth in state and local government compensation relative to private-sector levels.

Specifically, the Mackinac Center has calculated that Michigan’s total private-sector wages fell 25 percent between 2000 and through 2010, and non-salary benefits fell 6 percent. Nearly the opposite occurred in the public sector: Wages increased 1 percent and benefits increased 25 percent, based on federal Bureau of Economic Analysis statistics.

In addition, the EPI study makes no note of private vs. public sector “quit rate” differentials. There’s a reason the number of federal employees who quit their jobs is just one-third of private sector levels, and it’s probably not because government supervisors are such warm and caring human beings. A more likely explanation is that government employees understand very well just how good they have it. The pattern appears to hold in Michigan too, where a state Civil Service Workforce report indicated that only 1.5 percent of 50,000 state employees quit in fiscal 2010.

Moreover, Michigan’s annual public-sector salaries rank 12th in the nation, or 5.8 percent higher than the national average, even before adjusting for the cost of living. Not only are Michigan public-sector employees generally doing better than their private-sector counterparts, they are outperforming most other government employees across the nation.

Even if the author of the EPI paper could back up his claims that government employees are not overcompensated, he still would have to account for the issue of job security, which is ignored. The word “tenure” is wholly alien in private-sector workplaces, yet in the public sector it is extremely difficult to fire or lay-off even the most sub-standard employee. Surely this must be worth something to workers in dollar terms.

In addition, the EPI author presumably developed some statistical model to support his conclusions, but he failed to report any of its assumptions, limitations, equations or standard error levels. It is therefore impossible for scholars to replicate or repudiate the work, conveniently shielding it from criticism. The Mackinac Center has twice requested the model equation and dataset, but EPI has ignored the requests.

Despite the EPI author’s claims, there has yet to be a complete analysis of all public-sector employment compensation. But there is no denying that the total value of compensation in Michigan’s public sector grew substantially in the past decade while the means to pay for it has fallen. The private-sector is reforming while the public sector is coasting on increased benefits. That’s why benchmarking benefits alone can save the state $5.7 billion.

#####

Michael D. LaFaive is director of the Morey Fiscal Policy Initiative at the Mackinac Center for Public Policy, a research and educational institute headquartered in Midland, Mich. Permission to reprint in whole or in part is hereby granted, provided that the Center and the author are properly cited.