"The Retrenchment of the State Employee Workforce in Michigan," a paper by Charles Ballard and Nicole Funari, is basically a summary of a lengthier report by the Michigan House Fiscal Agency.1 As such, it provides some useful information. However, the conclusions drawn by the authors and attributed to the paper in the media are not substantiated by the data presented, and the use of the data is in some cases misleading.
For instance, despite Ballard and Funari's assertions to the contrary, one cannot conclude from evidence presented in the Ballard and Funari paper (or the original House Fiscal Agency report) that Michigan state employees are paid less than their counterparts in the private sector. A closer examination of the data suggests that the opposite could be the case. Nor, as another example, do the data show that state services are in jeopardy due to reduction in state employment. While it is clear that the number of classified state employees — that is, civil service employees — has declined from its peak in 2001, there are a number of conclusions one can draw from this other than that state services are in jeopardy.[*] It may be that productivity of state employees has risen or that there were more than an efficient number of state employees at the employment peak. In addition, total state employment, including legislative and administrative unclassified workers and state university and agencies personnel, fell by only 1 percent over this same period.
The authors admit that the report is not designed to come to the conclusions attributed to it. For example, on the first page of the report, the authors write, "No attempt will be made to assess the optimal number of state employees, or the optimal compensation." However, in the same paragraph, they write with no supporting documentation, "Further cuts will run an increasing risk of leaving state government unable to perform its vital functions."[†] Later in the paper, the authors write: "A complete comparison of the earnings of state workers with the earnings of private-sector workers would require a sophisticated econometric analysis, controlling for a host of variables. Such an analysis is beyond the scope of this report." The authors are indeed correct that an econometric analysis would be needed to do a proper comparison and that this was obviously beyond the scope of the paper, yet they conclude that state employees are on average paid less than their private-sector counterparts. This paper will demonstrate that the conclusions of Ballard and Funari are disputable by pointing out numerous limitations in their analysis. These include:
- Using data on civil service classified employment to draw conclusions about growth in state employment, while not discussing total state employment
- Reaching unsubstantiated conclusions regarding the connection between state classified employment and state services
- Failing to account for differentials in non-wage employee compensation when comparing state employees with private-sector employees
- Underestimating state employee compensation by ignoring pension and other retiree benefits, such as health care coverage
- Using data and surveys in a confusing and misleading way when discussing educational background differentials between state and private-sector employees
- Drawing conclusions about state employee real wage growth from only a subset of state employees
- Not comparing state employee real wage growth with Michigan private-sector real wage growth
- Failing to account for the benefits that state employees received in compensatory leave time in the wage-reduction agreement of fiscal 2004
- Overestimating the role of current state employees in addressing the state's budgetary problems.
[*] As described by the state House Fiscal Agency, "The Constitution of the State of Michigan of 1963, Article XI, Section 5, establishes the classified state civil service and authorizes the Civil Service Commission (CSC) to classify all state positions and establish state employee compensation rates." (Wild, Stansell, and Bean, “Civil Service Salary and Benefit Comparisons,” 3.) Hence, the term "classified state employees” is often used to mean “civil service employees."
[†] There are a number of other assumptions that are reasonable. For instance, the substitution of technology and capital for labor could lead to improved delivery of key services.