Download this report (pdf)
Download the introduction and summary (pdf)
Read the full report in your web browser (Scribd)
Teachers are the most important school-based resource affecting student achievement. Few would argue with this statement, or that the future competitiveness of the U.S. economy requires improved academic results from public schools, including those serving predominantly low-income students and students of color. These facts should inform an array of changes to policies around the licensing of teachers and their performance evaluations, tenure, and compensation.
As a facet of compensation, teacher pension policy should be subject to the outcome- and equity-oriented workforce goals of broader teacher reform programs. Teacher pension policy should help attract to teaching especially promising collegegraduates and career-changers—candidates with combinations of cognitive and noncognitive characteristics known to predict the future effectiveness of teachers. Pension policy should also attract candidates with expertise in shortage subject areas, such as math and science, as well as encourage especially effective teachers to remain in the profession and to work in the schools with the greatest needs.
Yet pension policy is not a potent lever, on its own, for serving these workforce goals. Pensions, a matter of deferred compensation, represent a relatively small slice of total teacher compensation. The trick, then, lies in identifying pension policies that enable, catalyze, reinforce, or complement other policies sharing the goals of improving the overall quality of the teacher labor force and creating greater equity in the distribution of teaching talent.
Traditional defined-benefit pension plans, in which 89 percent of public school teachers participate (see text box for a primer on pension plans on page 6), serve these goals poorly. One reason is that the financial condition of existing definedbenefit plans is vulnerable to “pension holidays,” when employers—school districts—fail to make contributions to their employees’ pension plans. Another culprit is imprudent benefit enhancements. Several consecutive years of unusually high returns on a pension plan’s investments have been used in the past to justify benefit enhancements or to free up employer contributions for other uses.
But the long-term ability of a plan’s assets to cover its liabilities generally requires that high returns in some years make up for low returns in others. State policymakers have short-term political incentives to ignore this fact. Yet the financial vulnerability of defined-benefit pension plans matters a great deal because current teachers and retirees participating in these plans are legally guaranteed their anticipated benefits, by and large.
Effective compensation policies, including ones meant to address the goals above, require sound financial footing. Accordingly, this paper begins with two recommendations:
- Amend state constitutions subjecting any benefit-enhancing legislation to protracted, rigorous scrutiny
- Amend the Elementary and Secondary Education Act so that a school district’s allocation of funds under Title I, Part A—a program that provides supplementary funds to school districts serving concentrations of low-income children— is penalized in proportion to failure to make actuarially required contributions to defined-benefit pension plans
These two measures are aligned with recommendations published by the American Federation of Teachers, and they go some distance toward assuring retirees, currently active teachers, or prospective teachers that their pensions are safe and secure.
Policymakers can take comfort in the fact that two states, Georgia and Oklahoma, have already implemented the first recommendation, which does not entail immediate pain to stakeholders. The second recommendation involves adding a tool to the suite of fiscal requirements for receipt of Title I funds. The idea is appropriate because a substantial fraction of Title I funds wind up as contributions to pension plans, and embracing this recommendation offers federal policymakers a chance to respond concretely to popular concern around the sustainability of teachers’ defined-benefit pension plans.
This paper goes much further, however, in redefining teacher pensions. We make a third recommendation that addresses the mismatch between traditionally defined-benefit pensions and the workforce goals our nation must embrace to improve the productivity of our education system. The problem, simply, is that traditional retirement benefits are back-loaded. Basically, pension wealth as a fraction of cumulative earnings is much greater for teachers who spend multiple decades with a single employer than for those who teach for less than one decade.
Back-loaded benefits have four interrelated shortcomings. First, back-loaded benefits buttress the traditional salary schedule in which pay is pegged to longevity. With defined benefits upon retirement keyed to final average salary (most often the highest average annual salary over any three consecutive years), teachers closest to retirement are understandably wary of changes in salary policy. That makes it difficult to accelerate salary growth for teachers in their first 10 years of service—a move that would promote retention of early-career teachers and arguably increase the supply of teaching candidates. This would allow employers to select, on average, candidates with greater promise. Accelerated salary growth has become a tenable policy option because of widespread efforts to implement meaningful performance evaluation policies.
Back-loaded defined benefits based on final average salary mean that teachers nearing retirement are motivated to oppose performance bonuses or incentives to teach in high-poverty schools or shortage subject areas if they cut into across-the-board salary increases. And because long-term veterans tend to be highest on the political pecking order among teachers, back-loaded pension benefits represent a formidable obstacle to the adoption of salary policies attuned to workforce goals. More bluntly, back-loaded pension benefits are the linchpin of the status quo in teacher compensation.
Second, back-loaded pension benefits may work well for prospective teachers planning careers of 30-to-40 years in the same state, but such plans are rare among prospective teachers in the 21st century. Research using the Department of Education’s School and Staffing Survey data shows that by 2004 over 40 percent of new teachers were career-changers, and a more recent survey conducted by the Pew Research Center shows that Millennials, today’s teens, and twenty-somethings, are more than twice as likely as Baby Boomers to say they will switch careers. Thus, back-loaded benefits are poorly matched with the anticipated career trajectories of today’s potential teachers.
Third, the back-loaded benefits as offered by existing defined-benefit pension plans are not fully portable. Teachers who change states suffer serious losses of pension wealth, despite provisions designed to facilitate portability. Little is known about interstate migration of teachers, but if barriers to migration frustrate compensation policy goals, then the dynamic career-plans of Millennials are likely to compound the problem.
Changing districts within states can also cost teachers substantial pension wealth because of contractual caps on salary-schedule placement in a new district. Such caps, which can lower final average salary, and thus back-loaded pension benefits, are an artifact of traditional salary policies, and inimical to an incentive environment concerned with equity and productivity.
Fourth, back-loaded benefits get the incentives for remaining in the profession wrong. A compelling and consistent finding from 40 years of research on educational productivity, recently synthesized by Jennifer Rice King in a report for the Urban Institute, holds that teachers are as effective on average in promoting student achievement gains after 5 to 10 years as they ever will be. Back-loading means that the pension-based incentive to teach for an additional year is much greater for teachers a couple of decades into the profession than for those with fewer than 10 years of experience. It is plausible that a more even distribution of retention incentives along the continuum of experience could increase the rate of retention among teachers still on the learning curve without significantly lowering the retention of effective teachers with more than 10 years of experience.
In fact, under the stipulation that policy changes only affect new teachers, there would be no downside in terms of teacher quality to embracing policies in which pension wealth grows steadily with cumulative earnings. And for most prospective teachers, such a pension wealth accrual pattern would better serve the primary purpose of pension policy—ensuring secure post-retirement income.
There is, fortunately, an established way to define pension benefits that fits this bill. In a so-called cash-balance arrangement, teachers’ defined benefits are represented by notional account balances that reflect teacher and employer contributions plus interest credited at some indexed rate. Participants receive their benefits upon retirement either as a lump-sum or as an annuity—a lifetime stream of annual payments that is the socially responsible default option.
By definition, in a cash balance plan the ratio of pension wealth to cumulative earnings is constant over years of service, and pension wealth accrues steadily such that the pension-based incentives to teach for an additional year are distributed quite evenly over years of service. In other words, cash balance defined benefits put retention incentives in play for early-career teachers and temper the exaggerated late-career incentives characteristic of back-loaded benefits.
Cash-balance defined benefits also resonate with the career expectations of today’s prospective teachers. The approach allows teachers that have completed a vesting period to leave the profession or to begin teaching in another state with pension wealth proportional to cumulative earnings. The approach is also amenable to graduated vesting schedules in lieu of the traditional cliff vesting, whereby teachers suddenly become eligible for benefits after five years, typically, but sometimes as many as 10 years. And while cash balance defined benefits are closely related to salary, they shift attention from final average salary to career-average salary, thereby increasing the salience to new teachers of compensation policies serving outcome- and equity-oriented workforce goals.
This paper’s third recommendation follows from these findings: Existing defined-benefit pension plans accommodate new teachers in dedicated tiers in which all of their benefits are defined using a cash-balance approach.
While their benefits would be defined differently, new teachers would participate in existing defined-benefit plans alongside teachers that expect back-loaded benefits. This arrangement for new teachers to participate in existing plans is important because the fund managers for these plans count on an annual influx of new participants to elevate risk-tolerance and, therefore, long-term investment yield.
The alternative—shunting new teachers into separate pension plans—would undermine the financial condition of existing plans. Furthermore, if separate plans for new teachers were defined-contribution plans, similar to the 401(k) retirement savings vehicles so prominent in the private sector, then the new teachers would forego the low management costs enjoyed by defined-benefit plans, and they would shoulder greater investment risks—increasing the chances that public investments in teacher retirement do not yield secure post-retirement income.
The remainder of this paper presents background knowledge on the shortcomings of back-loaded pension plans in light of the need to attract top talent to the teaching profession in an era of multiple-mobile careers, and then unpacks the logic behind its three recommendations. The third recommendation is admittedly bold, especially in a climate of fiscal retrenchment and heightened political rhetoric. But the potential new costs associated with the cash-balance recommendation are manageable, and the paper shows that such costs represent crucial investments in our education system. The first two recommendations, in contrast, carry virtually no financial cost, embodying the ethos of “the new normal: doing more with less.”
Together, these three steps point the way toward a transition in the way that the teaching profession arranges for pensions, from one where the incentives for entering and remaining in the classroom are hopelessly mismatched with career expectations of 21st century teaching candidates, to a way in which pensions complement other compensation policies to improve the quality of the teaching workforce and the distribution of its talent.
Raegen T. Miller is the Associate Director for Education Research at the Center for American Progress.
Download this report (pdf)
Download the introduction and summary (pdf)
Read the full report in your web browser (Scribd)