Public school teachers deserve a compensation system that puts them on a secure path toward retirement. Right now, that’s not what they have.

By Marcus A. Winters

Teachers unions often defend defined benefit (DB) retirement plans on the grounds that they ensure retirement security. For teachers, it might be comforting to know that upon retirement you will receive a fixed (and generous) check each month for the rest of your life regardless of what the stock market does between now and then. But the security from investment risk inherent in DB plans masks another sort of risk to which teachers are greatly exposed because of how their DB plans are designed. We might call it attrition risk — the chance (likelihood) that an entering teacher is not employed within a school covered by the same pension plan for her entire working career.

I suspect that many teachers don’t realize that how much they benefit from their pension plan depends on how long they work for a particular employer. Those who stay employed by public schools covered by the same plan (most often that includes all public schools within a state) for their entire career make out great. Teachers who leave participating public schools earlier (i.e. most entering teachers) do not. Of course, at the time they’re hired, no one knows which type of teacher they’ll turn out to be.

In this article, based largely on several studies conducted with Joshua McGee (McGee & Winters 2013a, 2013b, 2016; Winters & McGee, 2014), I illustrate how the severely backloaded structure of today’s public school teacher pension systems benefit only a small proportion of entering teachers while putting the rest on an insecure retirement path. But there is a cost-neutral solution to this problem that would benefit most teachers entering public school classrooms today without removing any of the protections from the stock market with which teachers have become accustomed.

Retirement (in)security

According to the Bureau of Labor Statistics (2009), about 89% of public school teachers participate in DB plans that provide them with fixed payment for life at the time of retirement (often adjusted for inflation). Many people prefer the traditional DB model because their guaranteed retirement amount suggests a far safer — though potentially less lucrative — retirement wealth than offered by 401(k)-style defined contribution systems that are subject to the whims of the stock market over time.

But the DB structure of teacher pensions has produced a common misperception that today’s teacher pensions ensure a safe retirement path for all. They don’t. These pension plans are structured such that teachers earn very little retirement benefit for their first several years of teaching and then rapidly accrue pension wealth as they near the “normal” retirement age. Unfortunately, the large majority of teachers who begin their working lives teaching in a particular state’s public school system won’t last there long enough to accrue their pension reward.

This problem is common but not inherent to DB retirement plans. It is driven by the fact that the benefit is a function of their final average salary. A retired teacher’s monthly pension payment is based on a calculation that takes into account the number of years that the teacher worked within a school covered by the plan (usually all public schools in the state) and the teacher’s final average salary (often the average salary over the last three years of employment). For each year of service within a covered school, the plan increases the fixed payment by an additional percentage (often about 2%) of their final average salary. A penalty is imposed for each year a teacher exits before what is known as the time of “normal” retirement. Once the teacher no longer works within a school covered by the plan, he stops accruing years of service and thus does not increase pension wealth within the plan.


Because it comes in the form of a fixed payment for life, the total amount of a teacher’s pension benefit depends both on the amount of the monthly payment and also how long the teacher lives past retirement age. Nonetheless, we can use conventional actuarial techniques to put a dollar value on the teacher’s annuity. We calculate the present value of the teacher’s annuity, net of the teacher’s contributions, under the rules of her pension plan were she to exit after any number of years after being hired.

The design of teacher pension programs is such that the value of a teacher’s retirement wealth (present value of the lifetime annuity) can change suddenly at various points across a career. The annuity’s value increases linearly as a teacher earns more service, but it also increases in big jumps at particular points as a teacher approaches specified retirement thresholds. Conversely, a teacher’s annuity generally loses value each year after reaching the plan’s normal retirement eligibility threshold because, with each additional year of work, that teacher is foregoing a year of retirement in which a payment would have been received.

Allowing teachers to acquire retirement wealth smoothly throughout their careers would lessen the blow of layoffs considerably.

To illustrate, the blue line in Figure 1 shows the accumulation of pension wealth across the career of a 25-year-old entrant into the New York City teaching workforce. The line represents the present value of the teacher’s accumulated employer-provided pension wealth (that is, excluding the teacher’s contribution) at any given age.

As is typical in other systems, New York’s teachers earn very little employer-provided retirement wealth in the early and middle portions of their careers, followed by steep accrual during each year of service in late career, and negative accrual each year after reaching the plan’s normal retirement age. A teacher who began working in the New York City public school system at age 25 and exited the system 38 years later would retire with an employer-funded lifetime annuity worth about $610,250 — an average of $16,059 per year of service (McGee & Winters, 2013a, 2013b). However, had that same teacher exited the state’s public school system after 20 years of service (perhaps, as pointed out by Goldhaber, Grout, & Holden in a February 2017 Kappan article, to take a teaching job in another state), the employer contribution to her retirement would be a lifetime annuity worth only about $59,572 — an average of $2,979 per year of service.

Winners and losers under the current system

Teachers can’t take home their retirement money until they retire. The backloaded nature of the pension plan, then, is not particularly concerning for entering teachers who are certain to remain employed by a public school covered by their plan until retiring at or around age 63 — to the extent that such people actually exist. But for those who don’t fit that profile, the backloaded structure of today’s pension plans has serious consequences. Unfortunately, most teachers leave well before they might benefit from the structure of the retirement plan.

The dashed black line in Figure 1 illustrates, at each age, the percentage of an entering cohort of 25-year-old teachers whom the pension plan assumes will remain in the system, using the scale at the right of the figure. Only one-third of teachers from the cohort in New York City are expected to remain long enough to receive the maximum pension payout at age 63. It’s worth noting that this result from New York’s plan likely downplays problems faced by those teaching under other public teacher pension plans across the nation because New York’s assumed turnover is far lower, and thus more conservative, than those used by most other pension plans. According to our calculations, based on figures reported by the Institute for Educational Sciences, only about 28% of American public school teachers nationwide remain in the profession for even 20 years.

Consider three hypothetical teachers — Emily, Julie, and Sarah — all of whom were 25 when they began teaching during the same school year. Each remained in teaching for their entire career until retiring 40 years later. Emily taught her entire career in New York City and then retired with an employer-sponsored annuity worth about $592,158. Julie worked in the Philadelphia school system for her entire career and retired with an employer-sponsored annuity which we calculate to be worth $404,433.

Sarah, on the other hand, worked in the New York City school system for the first 15 years. Then, when her partner was transferred to another job, Sarah moved to Philly and worked in the public school system for the remaining 20 years of her career. Sarah taught in the same school systems for the same amount of time as did Emily and Julie. But she didn’t remain in either system long enough to benefit from the pension backloading. Consequently, Sarah retired with employer-sponsored pension wealth worth only about $62,089.

It could be worse. According to a report from Bellwether Education Partners, teachers covered by 17 statewide pension plans don’t fully vest for 10 years (Aldeman & Rotherham, 2014). So, someone who teaches nine years in Illinois, nine years in Indiana, and nine years in Michigan before leaving the profession would have earned zero employer-sponsored pension wealth for her 27 years of service to students.

It’s perverse, but today’s pension systems don’t just anticipate that most teachers won’t last long enough to collect meaningful pension wealth; they count on it. The relatively high pension payoffs provided to those who leave the plan at or near the age of normal retirement (Emily and Julie in our example above) are funded by redistributing dollars away from the pension wealth of those teachers who exit earlier in their careers (our unfortunate Sarah).

A smooth accrual pension plan

There is nothing inherent in the market protections of DB plans that requires retirement wealth to be accrued so late in a teacher’s career. Teacher pensions could be restructured in such a way that teachers earn retirement wealth in relatively equal intervals throughout their careers. Simply moving compensation around in this way would cost taxpayers nothing and would not expose teachers to any additional market investment risk.

We call our alternative structure a Smooth Accrual Defined Benefit plan (SA DB). By smooth we mean that the benefits earned by teachers are a constant percentage of their cumulative earnings. Like other DB plans, the SA DB would offer teachers a lifetime annuity at retirement.

Today’s pension systems don’t just anticipate that most teachers won’t last long enough to collect meaningful pension wealth — they count on it.

To be clear, this plan is in every way a DB plan. I am not describing a 401(k) or other defined contribution plan. Further, the plan is structured to have the same expected value as existing plans, and thus does not represent an aggregate benefit reduction. Consequently, this proposed change would not itself address the pressing issue of large unfunded liabilities tied to pension plans that is pressuring many state budgets. The only difference between a current plan and its respective cost-equivalent SA DB plan is the rate at which teachers accrue pension wealth during each year of employment. The SA DB plan pays meaningful retirement benefits for each year of a teacher’s service.

The red line in the figure represents the wealth acquired at each period for our 25-year-old entrant in New York City under a plan that allows teachers to smoothly accrue retirement compensation across their careers. The line is calculated to represent a cost-neutral plan for taxpayers relative to the current plan (Costrell & Podgursky, 2010).

Teachers acquire significantly more retirement wealth early in their careers under the smooth accrual plan than under the current plan. For example, a teacher who exits the district at age 45 with 20 years in the classroom receives the equivalent of $59,572 in retirement wealth under the current plan but would leave with $151,120 under a smooth accrual plan. Because the current pension plans redistribute foregone retirement wealth from those who exit earlier to them, those teachers who remain employed in the New York City plan from age 25 until the normal retirement age do better under the current backloaded system. A teacher who retires at 63 would receive the equivalent of $610,250 in an annuity under the DB but would have earned $415,107 under the smooth accrual plan.

In short, the new smooth-accrual plan would benefit some teachers while offering lower retirement wealth to others. The question is, what proportion of entering teachers would benefit from each plan?

We can see from the figure the age of retirement at which teachers would benefit most from each of the two plans, along with the percentage of teachers who are actually still teaching at that age. Those who leave early in their careers would receive more from the smooth-accrual plan, as represented by the red line, which lies above the (blue) line representing wealth under the current plan. The advantage shifts to the current plan at the point at which the lines cross, which in the case of New York City happens at age 56. At present, only about 42% remain employed in the schools past this age, which is to say that fewer than half of teachers benefit from the current plan.

Again, New York’s plan is a conservative representation of the backloading phenomenon facing public school teachers across the nation. In Philadelphia, for instance, only about 18% of entering teachers are expected to remain employed in the state system long enough to benefit from the current plan relative to how they would fare under a cost-equivalent SA DB plan.

A more fair system

Proponents of the current plan structure sometimes argue that its design incentivizes teachers to remain teaching in their schools for a sustained period of time, thus limiting turnover. In fact, it can be seen from the figures and from empirical research (Costrell & McGee, 2010) that teacher exit decisions are sensitive to the incentives produced by the accrual pattern of their pension plans.

In recent research (McGee & Winters, 2016; Winters & Cowen, 2013), we have presented simulations showing that given what research shows about teacher experience and classroom effectiveness, moving from the current system to a SA DB plan would be expected to have minimal effect on teacher quality. But even if we were to concede that keeping teachers in their school is a desirable goal, is it right to do so by jeopardizing the retirement security of the majority of entering teachers? There has to be a better way.

Perhaps backloaded pension plans were justifiable in the age in which they were developed, when populations were less mobile. But few young people today enter any job or career knowing for certain that they will remain there for their entire working life. Further, if implemented with fidelity, policies recently enacted across the nation that weaken the employment protection offered to public school teachers might add considerable additional attrition risk. Several districts and states now strip tenure protections from teachers who receive below-standard performance evaluations over a period of time. Experienced teachers who were once protected from termination by their tenure status now might face the real possibility of losing their jobs. This prospect is particularly chilling for a teacher with about 20 years in the classroom. Were such a teacher to lose her job due to poor performance she would not only lose her salary but also leave with very little retirement wealth for the first two decades of her working life. Allowing teachers to acquire retirement wealth smoothly throughout their careers would lessen the blow of such layoffs considerably.

Public school teachers deserve a compensation system that puts them on a secure path toward retirement. Many teachers have been lulled by the defined benefit structure of their pension plan to believe that their futures have been cared for. Unfortunately, for most of those entering classrooms today, that is just not the case. Unless states move to a more smoothly accruing plan, the situation will only worsen in the future as teachers, like other professionals, become more mobile.


Aldeman, C. & Rotherham, A. (2014). Friends without benefits: How states systematically shortchange teachers’ retirement and threaten their retirement security. Washington, DC: Bellwether Education Partners.

Bureau of Labor Statistics. (2009). State and local government workers, national compensation survey. Table 2. Retirement benefits: Access, participation, and take-up rates. Washington, DC: Author. www.bls.gov/ncs/ebs/benefits/2009/ownership/govt/table02a.pdf.

Costrell, R. & McGee, J. (2010). Teacher pension incentives, retirement behavior, and potential for reform in Arkansas. Education Finance and Policy, 5 (4), 492-518.

Costrell, R. & Podgursky, M. (2010). Golden handcuffs. Education Next, 10 (1).

McGee, J. & Winters, M. (2013a). Better pay, fairer pensions: Reforming teacher compensation. New York, NY: The Manhattan Institute.

McGee, J. & Winters, M. (2013b). Rewarding experienced teachers: How much do schools really pay? New York, NY: The Manhattan Institute.

McGee, J. & Winters, M. (2016). Better pay, fairer pensions III: The impact of cash-balance pensions on teacher retention and quality. New York, NY: The Manhattan Institute.

Winters, M. & Cowen, J. (2013). Would a value-added system of retention improve the distribution of teacher quality? A simulation of alternative policies. Journal of Policy Analysis and Management, 32 (3), 634-654.

Winters, M. & McGee, J. (2014). Better pay, fairer pensions II: Modeling preferences between defined-benefit teacher compensation plans. New York, NY: The Manhattan Institute.
MARCUS A. WINTERS (marcusw@bu.edu) is an associate professor, Boston University School of Education and a Manhattan Institute senior fellow.

Originally published in October 2017 Phi Delta Kappan 99 (2), 32-36. © 2017 Phi Delta Kappa International. All rights reserved.