Privatization Review a Mixed Bag

Last July, the Mackinac Center for Public Policy released its 107-page primer on the privatization of services in Michigan public schools. In February 2008, the Great Lakes Center for Education Research & Practice, founded by the Michigan Education Association school employees union, offered a seven-page, largely flattering quasi-rebuttal.

While it is pleasantly surprising that an organ of the MEA would offer praise for "A School Privatization Primer for Michigan School Officials, Media and Residents," the review also included several erroneous or hyper-generalized criticisms. At times the GLC author contradicts himself while confirming Mackinac Center arguments. It appears that the review’s strongest criticism against the school privatization primer is that it did not sufficiently make the MEA’s case against privatization.

Stay Engaged

Receive our weekly emails!

The review of the privatization primer is comprised of six sections and concludes with a section titled, "Usefulness of the Report for Guidance and Policy and Practice," which reads in part:

This report is useful. It presents credible surveys of current policies across states showing that contracting out of food, transportation, and custodial services is widespread, although public provision is still more common. The report describes practical steps for issuing and monitoring contracts; these rules of thumb may help districts avoid making costly mistakes or getting locked into unfavorable contracts. Some districts may find useful the discussion of how unions sometimes oppose contracting out and the report’s catalog of service providers in Michigan. These practicalities may have deterred some districts from investigating contracting out. For districts very dissatisfied with their in-house services, this document may help them move forward with contracting out.

The second and final paragraph of this section is shorter than the one above and attempts to minimize the value of the primer. Several of its arguments, among others, are addressed below to some degree.

The GLC outsourced authorship of the report to an out-of-state professor, Clive Belfield of Queens College, City University of New York. Accompanying the paper was a press release summarizing Belfield’s findings and providing media contact information for Teri Battaglieri, the GLC’s director.

The Great Lakes Center for Education Research & Practice is a non-profit organization that conducts reviews of education-related research, particularly works produced by market-oriented institutions. In September 2001, then-MEA President Luigi Battaglieri announced the creation of the Great Lakes Center at a press conference in Lansing. The Michigan Education Association would later file a lawsuit against the Mackinac Center for accurately quoting the MEA president at this GLC press conference. The Michigan Court of Appeals ruled unanimously against the MEA in 2004.

According to the Great Lakes Center Web site, Luigi Battaglieri is now chairman of the Great Lake Center’s board of directors and executive director of the MEA.

Following is an analysis of the GLC’s review by section:

Summary of Review

Belfield’s paper falsely contends that "the report presupposes that [contracting out] is beneficial." Page 41 of the primer explicitly states that "experience has shown that contracting can sometimes fail and can also be controversial." Also on page 61, it states that "privatization, like any other human enterprise, can fail." That hardly qualifies as a presupposition of benefits from contracting.

The MEA has contracted out at its headquarters for food, janitorial, security and mailing services and in three out of four cases with non-union firms. Should we deduce from this that the MEA presupposes that contracting out is beneficial? Indeed, the GLC has apparently outsourced the work of this review to a New York-based scholar. Is there not a revealed preference for contracting out work by the very institutions that presume to question its usefulness?

The GLC review also claims in its summary that the primer relies on testimonies from individuals as opposed to "direct data or research." This may be the most laughable charge considering other statements in the GLC’s review. On the one hand the review states, "there is almost no published literature in academic journals on potential cost-savings from contracting … by school districts." On the other, the GLC press release said Belfield "credits the report for providing credible surveys of the current breadth of contracting-out practices," one of which is from an academic journal. So which is it?

Academic journals contain little on school privatization (outside the voucher and tax credit movements, which are often considered a form of privatization). In fact, on page 19, under the heading "Empirical Studies on Cost Reduction," the primer indicates, "as noted above, systematic, nationwide data on school bus contracting are scarce."

Second, the Mackinac Center conducted an exhaustive review of contracting data, including information from all 50 state education departments, the U.S. General Accountability Office, the U.S. Department of Agriculture and the Centers for Disease Control and Prevention, the latter in a peer-reviewed academic journal that was duly cited in the privatization primer.


"If education or a service associated with education can be provided at a genuinely equivalent quality but at less expense by a private company instead of a public enterprise, then a very compelling case can be made that the private company should be hired." So begins the GLC review of the Mackinac Center’s privatization primer. It is hard to disagree with such logic. The Center has been saying exactly that since it opened its doors in 1988.

Findings and Conclusions of the Report

This section offers little more than a three-paragraph summary of the privatization primer.

Rationales Supporting Findings and Conclusions of the Report

The GLC report erroneously charges that the privatization primer’s "suggested rationale for supporting privatization largely rests on the argument that if districts do it, it must be beneficial for them. Since almost 40 percent of districts in Michigan do contract out the report assumes the practice must be beneficial in these districts."

This may be the most disturbing charge laid out by GLC. The primer includes a section titled, "Responses to Anti-Privatization Claims." Within that section is a sub-section addressing "The Argument that Privatization Has Failed Everywhere."

This appears on page 61, nearly half way through the book, and a little late to develop a rationale for school support service privatization if that were the primer’s intent. The argument was simply that "the overall success of contracting is borne out by the fact that it is increasing, rather than decreasing, among Michigan school districts." The GLC took dramatic liberties with the primer’s text in order to make its arguments.

An even harder claim to substantiate is that "it is implied" by the privatization primer that districts that have not contracted out have refrained from doing so due to opposition or an ignorance of the benefits. The term implied suggests a certain psychological liberty on the part of the author as it is so open to interpretation as to be little more than a critic’s opinion. The GLC does not offer a whisper of evidence, however anecdotal, to back up this claim. The Mackinac Center primer does explicitly discuss union opposition to contracting out on pages 55-58 and 61, but it in no way argues that the remaining 60 percent of Michigan school districts refrain from contracting out due to opposition or ignorance, nor was that ever the primer author's intention.

The only really intriguing argument advanced by the GLC review’s author is that of "transaction costs" of competitive contracting, but this is easily refutable. In another example of employing words that are so generalized they mean little, the GLC paper states, "It is possible that the resources needed to secure a contract exceed any potential cost-savings from hiring a private firm." Well, of course it is possible. Hypothetically, many things are. The question should be, is it probable?

In the Mackinac Center’s last school support privatization survey, only about 4 percent of Michigan school districts that had privatized one of the three major noninstructional services reported no savings. It is impossible to tell how many of the nearly 78 percent of districts that did report savings didn’t save enough to cover the cost of the transaction, which includes overcoming vociferous union and employee opposition. For that matter, there also may be considerable expenses inherent in not privatizing noninstructional services.

As an economist, Belfield should recognize that there are "opportunity costs" to not exploring competitive contracting. That is, by keeping services in-house a district may continue to pay 5 percent to 20 percent more for equivalent services, plus the cost of handling union grievances, than is otherwise necessary.

Another generalization advanced by the GLC review, under the heading of "transaction costs," is the idea that "in the long run there may be a cost of being locked in to [sic] a single contractor for provision of a service." But the privatization primer very clearly recognizes that "a district with a well-written contract can also exercise a cancellation clause …" These "out" clauses are actually quite common in Michigan. In fact, Chartwells Dining Services reports that its standard contract contains a 60-day cancellation clause on either party’s part and bills it as "without cause." That is, the decision to exit the deal is left up to the discretion of the parties. If Chartwells fails to please its customers, it can be dismissed with relative ease.

Continuing the transaction cost theme, the GLC paper argues that a private contractor may be inclined to impose a "risk premium" on the vendor. But if such a risk premium were baked into any proposal, it would raise the cost of providing the service. That such a premium may be included in a bid accepted by a school board and still save the district money is a testament to just how much money there is to be saved through the competitive bidding process.

The GLC author ends his review with a flurry of different arguments that are no stronger than the erroneous and often sweeping generalizations explicitly addressed above. The Mackinac Center’s privatization primer may be the most exhaustive work done on the subject anywhere in the United States. The Center advances this argument because in writing the primer I spent countless hours surveying the government, academic, business and non-profit universes for hard data on everything from estimates of savings (or a lack thereof) to the most serious pitfalls districts face in the contracting process. This includes sourcing union documents used to help opponents thwart privatization.

While we applaud the GLC’s willingness to praise the Center’s research in print, we take serious issue with its complaints. The Mackinac Center does not exist to make the MEA’s case against privatization. It is a research and education institute and provides a wide range of school-reform publications to the public and others to help them make informed decisions. We stand by the primer’s findings and its recommendations.

The Mackinac Center encourages people to draw its own conclusions about the school privatization primer. It can be read online at The GLC report can also be found online. Interested parties may also contact me with their own perspective on the primer and/or the GLC response at

The Mackinac Center’s privatization primer was the third in a series of school reform primers. "A Collective Bargaining Primer for Michigan School Board Members" and "A Michigan School Money Primer for Policymakers, School Officials, Media and Residents" have apparently not been reviewed by the Great Lakes Center.


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 author and the Center are properly cited.