In one potential scenario, state government would mandate and pay for a pilot merit-pay bonus program. In Michigan, however, the state government’s cost in passing such a mandate would be higher than in many states.

The Headlee amendment to the Michigan Constitution prohibits state government from placing unfunded mandates on local government.[49] To the extent that state government mandated that a local school district (or districts) implement a merit-pay bonus program, it would have to pay the entire cost of implementing the mandate. Hence, the state could not simply provide money for the bonus awards themselves: It would also need to cover district costs associated with disbursing the higher pay,[*] including the districts’ mandatory additional tax and premium payments for workers’ compensation, Social Security, Medicare, state and federal unemployment insurance, and school-employee pension and retiree health care. The pension and retiree health care premiums alone totaled nearly 18 percent of school employees’ pay in fiscal 2007.[50] And if the state mandated that school districts use new exams to measure student achievement, the state would have to pay for the school districts’ costs in buying and administering the tests.

Thus, a state mandate might prove prohibitively expensive, especially for an experimental pilot program in an era when slower-than-expected annual revenue growth and continued state spending growth has repeatedly left the state’s budget tight.[†] Nor in this climate are state policymakers likely to set aside seed money for a controversial merit-pay pilot program without mandating that some district use it, since the money could remain unclaimed and leave legislators open to accusations of misplaced priorities during a period of "budget austerity." Michigan remains, after all, a powerful union state.

In a second potential scenario, a local district might try to implement merit pay within its boundaries. Many local school districts have faced tight budgets in recent years, however. In some districts, the collective bargaining agreement also includes a clause that prohibits a district from using standardized test scores in evaluating teachers — a barrier to merit-pay compensation as long as the clause remains in effect.[‡],[51] Given the potential cost, obstacles and controversy, many school districts will hesitate to initiate merit-pay programs without assistance.

The challenges involved in state or school district financing of a pilot merit-pay bonus program lead us to recommend that private foundations consider choosing a Michigan school district and financing a merit-pay bonus program there, covering both the cost of testing and the bonuses themselves over the span of a few years. Private foundations are already playing a role in merit-pay pilot programs, providing all of the award money in the first year of the ACPP program in Little Rock[52] and some of the funding for the Denver ProComp and Chicago REAL programs. Moreover, private money would not be subject to claims from special-interest groups soliciting for other programs, and concerns over the Headlee amendment and local contract prohibitions would no longer apply.

That said, a private foundation would still find it difficult to be an entirely free agent. The foundation would almost certainly want to work with a school district’s officials before trying to finance a merit-pay pilot program in that district, if for no other reason than to gain access to student achievement data. The federal Family Educational Rights and Privacy Act of 1974 prohibits the release of educational records without parental consent.[53] Obtaining this consent from scores of parents individually would be problematic, especially compared to forging a working relationship with the district,[§] which has legal access to the results of tests taken in district schools. The district could protect student privacy by replacing the students’ names with codes that nevertheless allow the foundation to link the scores with the appropriate teachers.

For the school district’s sake, the foundation should probably pay teachers directly. As noted earlier, if the school district acted as a middleman by accepting foundation money for the bonuses and transferring it to the teachers, the district would be on the hook for a wide range of state- and federally mandated taxes and insurance premiums. In contrast, merit-pay bonuses sent directly to teachers by the foundation would not generate this burden, although the foundation would need to file IRS-1099 forms for each recipient and obtain the recipients’ W-9 forms (basically Social Security numbers).[¶]

Whether direct payment would also prevent the bonuses from being subject to disclosure under the state’s Freedom of Information Act would depend. If the district were informed in writing of the size of individual teachers’ bonuses, the information would be subject to FOIA, even if the bonuses were based partly on personal information like principals’ evaluations.[**] If, however, the district were not informed in writing, individual teachers’ bonuses would probably not be subject to FOIA, though it is difficult to be certain, since the issue has not been litigated in Michigan.

There are several advantages to having a private foundation, rather than a public institution, finance a pilot merit-pay program. Some of the legal and financial complexities involved in state or district financing would disappear; the program would be less subject to political compromise; and the program would be more cost-effective. These advantages should make it easier to launch the program, and after a few years of promising results, state and local policymakers should be more willing to adopt and finance broader reforms. Hence, although a foundation would have to endure a few years of costs and negotiations in implementing a temporary pilot program, the foundation would also know that its work was strategic. A well-designed pilot program could easily serve as a catalyst for badly needed, long-term and demonstrable improvements in children’s learning — and do so in a key union state.

[*]MCL 38.103a(2)(h) of the Public School Employment Retirement Act explicitly includes merit pay in “compensation.”

[†] It is also worth pondering whether a state mandate would prove wise as a policy matter. A determined state Legislature might fare better than local school boards at designing a merit-pay program that required real student achievement gains, but a state program could preclude useful experiments at the local level.

[‡]Such clauses could be excised during the bargaining of a new contract (or in impasse). The factors involved in teachers’ evaluations are likely a “permissive” subject of bargaining, meaning that the district and the union are permitted to bargain on the issue, but are not required to. A district might therefore have an easier time removing the clause during contract negotiations than if the topic were mandatory.

[§]In theory, of course, a foundation could approach parents independent of the district in order to obtain access to students’ test scores. Such an approach might seem appealing if a foundation believed a district’s officials were inefficient or uncooperative.

But this tactic would pose numerous problems. It is unlikely that a private foundation would succeed in convincing all (or very nearly all) of a school’s parents to release their children’s test scores, especially without the school district’s support. Basing merit-pay bonuses on the test scores of only some students, however, would undermine the validity of the bonuses and the program itself. The results of the project would be seen as — and almost certainly would be — skewed, since participating children could easily differ in important ways from children whose parents were unwilling to participate.

Moreover, a foundation approaching parents independent of the district could generate ill will with school district officials and the teachers themselves. This ill will might also be shared by parents who refused to release their children’s test results, since they could rationally fear that their children, whose test scores would not count in the privately funded teacher bonuses, would not receive the same attention in the public school classroom.

[¶]A prize or award is taxable income and not a gift even if provided by a "donor [who] derives no economic benefit from it" (see Robertson v United States, 343 US 711, 714 (1952)).

[**]The Michigan Supreme Court has held that teacher evaluations may be obtained through FOIA (see Bradley v Saranac Community Schools Bd of Educ, 455 Mich 285 (1997)). The court explicitly ruled that the privacy exemption in MCL 15.243(1)(a) did not prevent disclosure of performance evaluations, and the court also ruled that a collective bargaining agreement stating that only administrators would see the evaluation did not prevent its disclosure under FOIA.