The Long-Term Consequences of the University Early Retirement Program

Posted by Jeffrey Brown on Aug 30, 2010

Filed Under (Retirement Policy, Uncategorized)

Earlier this year, the University of Illinois instituted a “Voluntary Separation Incentive Program” for academic professionals and civil service employees.  It also instituted a Voluntary Retirement Program for tenured faculty and some other academic employees.  Both of these programs were implemented to help downsize the university in order to cope with both temporary and longer-term budget problems.  In an email to faculty, the Chancellor of the Urbana-Champaign campus noted on 8/27/10 that:

“more than 500 employees took advantage of our Voluntary Separation Incentive Program, and some 90 tenured faculty members and 16 adjunct professors and lecturers will leave the University under the Voluntary Retirement Incentive Program. Achieving this reduction in staff through voluntary programs minimized the dislocation and disruption that would have resulted from more drastic measures.”

Over the past 6 months, I have heard a lot of discussion among university colleagues and members of the community as to whether these incentive programs are really a good thing or not.  Concerns seems to focus on two aspects.

1. Are these programs really cost effective?  For example, how do we know that we are not just subsidizing retirement that would have happened anyway?  If 50% of the people who retired under this program were planning to retire this year anyway, this would essentially doubles the cost per-incremental-retirement.

2. How do we know that the “right” people are retiring?  For example, might it be the case that the most productive employees are the ones accepting these packages, because they are the ones with the best non-University alternatives for producing income?

I do not have the administrative data from Illinois that one would ideally like to have to answer these questions in our case.  But I did find a wonderful 2004 paper by John Pencavel of Stanford University, entitled “Faculty Retirement Incentives by Colleges and Universities” in which he discusses these and other topics.  I encourage you to read the whole paper if you have the time.  But here is my brief summary of some of his findings (based on his own research and that of other economists):

He examines the early retirement programs used by the University of California for tenured faculty in the first half of the 1990s and finds that:

1. Faculty are indeed responsive to monetary incentives.  He finds that an incentive equal to a one percent increase in a pension replacement rate is associated with a 3.7 percent increase in the acceptance of the early retirement incentive.

2. He shows that it is exceedingly difficult to accurately forecast who accepts such a package, which in turn makes it very difficult to accurately predict behavior in advance.  In the paper (around pages 30 and 31, he explains the reasons in detail).
3. Contrary to the concern raised in my question #2 above about the “best” faculty being most likely to retire, he finds that after controlling for age, length of service, academic discipline, and campus, one can treat salary as a proxy for the “quality” of a faculty member (i.e., if we compare two 60 year old professors in the same department who have been on the same campus for the same number of years, the more highly paid one is, on average, “better.”)  He finds that “those individuals with higher salaries were less inclined to accept the buyout program, a finding that is commonly found in studies of the relationship between pay and quit rates.”  In other words – on average – the better employees (as proxied by salary) were less, not more, likely to accept early retirement.

4. For those that object to the use of salary (conditional on age, campus, dept and years of service) as a proxy for quality, he also reports results of another study that found that in the years before the early retirement programs, “those faculty with lower research output were more inclined to accept the severance incentives than other faculty.”  Again, this suggests that the most productive faculty are less likely to retire under an incentive program than are less productive faculty.

On net, I confess that reading this research makes me feel better about the program that the University implemented.  However, Pencavel concludes by cautioning against declaring an early retirement program a success simply based on the reduction in payrolls and the absence of adverse selection in who retires.  A full cost-benefit analysis would require an analysis of a much broader range of implications – including the overall impact on the quality of teaching and scholarly output.

In short, only time will tell …

2 Responses to “The Long-Term Consequences of the University Early Retirement Program”

  • Amanda N says:

    This post was extremely interesting to read as a student here at the univeristy. I actually had no idea that the university offered an early retirement program. Now I even question if I would know any of these professors that accepted this offer. Mr. Pencavel does provide some reassuring statistics; however it does not mean that his statistics apply in our case since the economy has changed a great deal since 2004. Some professors may have accepted this offer due to the financial instability of the economy and we could have potentially lost more qualified professors than we think.

  • Bryan M says:

    I agree with Amanda that this post was interesting to read. I have spoken to employees (not at the University of Illinois) who were offered generous early retirement packages just as they were about to retire anyway. I believe that the folks who come up with these policies often do not think them all the way through, not that I can create a better plan off the top of my head.