Teacher Pensions Blog

  • Public sector unions praise Social Security. Except they don’t want it for all of their workers.

    The National Education Association describes Social Security as the “cornerstone of economic security,” and Randi Weingarten, President of the American Federation of Teachers, describes it as “the healthiest part of our retirement system, keep[ing] tens of millions of seniors out of poverty [which] could help even more if it were expanded.” Last year, the Alliance of Retired Workers, an affiliate of the American Federation of State, County and Municipal Employees (AFSCME) even made the Social Security Administration a blue and white frosted birthday cake for its 78th anniversary. 

    But not all local government workers have Social Security. Over 6 million public sector workers are not covered by Social Security, including about 1.2 million public school teachers; in 15 states, public sector workers do not pay into or receive benefits from the system. If you were to ask, however, whether all state and local workers should have Social Security, most public sector unions would adamantly reply, no

    Why do unions hold such conflicting views on Social Security? The primary reason—pensions. Unions fear that extending Social Security coverage will signficantly cut into existing pensions, which are more generous to full-career workers in states that do not offer Social Security coverage.

    However, public pensions in states without Social Security coverage offer more generous benefits because they were designed as a stand alone benefit. Coordinating Social Security with state pension plans would likely result in equal or better retirement benefits overall for more teachers, especially those who do not qualify or receive much of a pension. What’s more, unlike pensions, Social Security is portable and does not penalize workers for moving across state lines. While the politics around teachers and Social Security coverage are at odds, Social Security could be a core part of improving teacher retirement plans. In particular, Social Security could provide a floor of retirement security for early career teachers who often leave the system with nothing.

    We’ll have more to say on this topic in the coming months.

  • My colleague Jason Weeby asked me what I thought of a Michigan legislator's plan to close the state's current retirement plan and instead offer all new teachers a 401(k)-style plan. I had two: 

    1. "401(k)" is occasionally thrown out by pension advocates as a scary alternative to traditional pension plans. But not all 401(k)s are the same, and they could be structured in a way that would offer teachers greater portability than current pension systems while placing more teachers on a path to a secure retirement.

    The term "401(k)" comes from the federal tax code, but it's often used to describe a broad category of defined contribution retirement plans, where the employer defines how much they plan to contribute in a given year and individual employees have their own accounts that they can take with them. But beyond that basic structure, they vary considerably in terms of how much the employer contributes, where the money is invested, or how much guidance they offer employees. This stuff matters immensely. A recent Center for Retirement Research brief shows just how much. The chart below from their brief starts with an employee making smart choices from the time he was 29 in 1982 until he turned 60 in 2013. He started young, contributed regularly, balanced his investment portfolio, and never cashed out. Those actions would net him $373,000 in retirement savings. But if he paid high fees on his investments that would cut his total down to $314,000. If, rather than transferring the entire balance when he changed jobs, he cashed out a little bit, that "leakage" would bring his retirement savings down to $236,000. Last, if he didn't contribute every year ("Intermittent contributions") and he waited to start saving well into his 30s (labeled "Immature System" in the graph), his total savings would drop to only $100,000. That's $273,000 in retirement savings in lost retirement savings. 

    Luckily, 401(k) plans are taking steps to address these problems. Due to federal rules requiring plans to improve their offerings and research showing that "nudges" and default options are powerful influences, plans have adopted smarter rules that protect individuals or at least encourage them into better savings habits. More plans automatically enroll workers, they recommend employees gradually increase their contributions over time, and they offer low-fee and life-cycle funds that help individuals make smarter investment decisions. And rather than being indifferent to what employees do with their savings when they leave, more plans reccommend individuals roll over their balance into a new retirement account or convert the entire balance into an annuity that will pay them a guaranteed amount throughout their retirement years.

    By my read, the Michigan 401(k) proposal appears to do none only some of these things. It would offer only a 4 percent employer matching contribution (with no minimum) and include none only some of the protections mentioned above (*Correction: The state's defined contribution plan does offer low-fee and life-cycle investments, but it does not discourage leakage or establish sufficient savings levels as the default for new teachers.) Based on what we know about individual decision-making, it would not be a very good plan for teachers. Contrast that with Oklahoma, which recently passed a law shifting new state employees into a 401(k)-style plan with minimum employee and employer contributions and a matching contribution of up to 7 percent. Oklahoma would also support individuals to make smart investment decisions and encourage them to convert their savings to annuities rather than letting it "leak" out. An Oregon task force just released recommendations for a plan with many of these same protections. These examples suggest it's possible to design a 401(k)-style plan that supports workers in meeting their retirement savings needs. 

    Teachers are the largest class of professional workers in the country so it's of national interest that they have a strong retirement plan. Today's pensions systems aren't meeting that goal, but change from traditional defined benefit plans need not be unworkable or unfavorable to teachers.

    2. Michigan currently offers teachers a choice between a hybrid plan (which combines a traditional pension component with a very small defined contribution element) or a standalone, slightly larger defined contribution plan. The hybrid plan is the default option for teachers and isn't serving them well, but at least they have a choice. The proposed legislation would eliminate that choice altogether. If Michigan teachers truly believe that they're going to stay in their job for 25 or 30 years, the current hybrid plan might be their best option. This applies to only about 27 percent of Michigan teachers (or about 20 percent nationwide). But for the large majority of teachers--who don't know how long they'll teach, or don't know how long they'll teach in a given state, or just aren't confident that life won't get in the way--may want to choose a retirement plan that's more portable. Giving teachers a choice is a good thing, but with the odds and percentages at play, the default should be a well-structured, portable option. A well-structured 401(k) plan could meet that criteria; this one in particular would not. 

  • Peter Greene and Neerav Kingsland have been debating the financial efficiency of public charter schools. Rather than wade into that debate, I'll take on one element of it that's rarely mentioned: teacher pensions. 

    To begin with, Kingsland is absolutely right to point out that states have engaged in fantastic accounting practices when it comes to their pension plans. States have not paid for pension costs on an honest accounting basis, and they have accrued billions of dollars in pension debt that avoids so-called "balanced budget" requirements. 

    Most importantly, Greene makes a big mistake when he writes that charters can avoid pension or other benefit costs through high turnover rates. In fact, the majority of states with charter schools require them to participate in the state pension plan. In those places, Greene's argument is exactly backward: Charter schools and their teachers pay the same high employer and employee contribution rates as all other schools, but higher turnover rates mean their teachers will get much less in return. Charter schools (and other urban schools with high turnover) subsidize the retirements of everybody else! 

    In fact, much of this story is broadly applicable beyond charter schools. Charters serve higher concentrations of low-income students, on average, than other schools and turnover tends to be higher in schools like that. But even within public school districts, some schools have much higher turnover rates than others. State pension plans treat them all the same, and we end up in a situation where there are some big winners at the expense of lots of small losers. Charter school teachers tend to be on the losing end. 

    Even in the places where charter schools are not required to participate, state pension plans impose rules that disadvantage teachers who move into or out of the system. Pension plans impose a retirement savings penalty on teachers who move across state lines or who leave teaching. Those same rules punish any teacher or principal who may wish to transfer between a traditional public school and a charter school. 

    Finally, there are those who see the current state of affairs--where new workers are propping up a system that has over-promised what it can actually deliver--and believe we can't afford to stop putting new workers into it. What they are essentially saying is the pension system is in such bad shape that we need to keep forcing new people into it, and that certain groups of teachers--especially ones who are newly entering the profession or ones who work in urban schools or charter schools--should sacrifice their own retirement savings for the good of the pension system. That's a terrible argument, and at some point policymakers will have to realize that placing new teachers into a bad pension system doesn't solve the problem, it just delays the inevitable.  


  • According to national data, four out of ten teachers will leave the classroom within five years. But turnover isn’t evenly distributed. The highest turnover happens in high poverty urban and rural public schools. In 2004-05, close to half of all public school teacher turnover happened in just one quarter of all public schools.  

    What’s more, turnover varies even within a district. A recent report from the National Council on Teacher Quality (NCTQ) tracked teacher retention across the Miami-Dade County Public Schools as one of several indicators correlated with teacher quality. NCTQ found that a sizeable portion of the county’s turnover took place in two particular low-income voting districts, and that many of the teachers who resigned left after only teaching a few years in the classroom. In the geographic region with the most turnover (voting district 2), over a third of all teachers who resigned had two or fewer years of experience, and over half had fewer than five years in the classroom. Florida makes similar statewide turnover assumptions about its teachers and assumes that over 60 percent of teachers will leave after 5 years. The graph below shows how teacher turnover differs within the Miami-Dade Public schools.

    Teacher Resignations in the Miami-Dade Public Schools (by voting district)

    Source: National Council on Teacher Quality, "Unequal Access, Unequal Results," 2014. 


    In terms of retirement, the Miami-Dade County Public Schools teachers in voting districts 1 and 2 are particularly vulnerable if they remain in the traditional state pension system. Any teacher that leaves the classroom before eight years will not qualify for a pension, and Florida assumes that roughly three-quarters of teachers won’t qualify. However, since 2002, teachers can choose to opt into a 401k-style retirement plan instead of a traditional pension. Teachers who choose the 401k-style retirement plan will automatically qualify for their account savings after just one year, and many teachers seem to prefer the plan according to one study. The 401k-style plan benefits are portable and better for shorter-term and mobile teachers, while the traditional pension plan are better for teachers who stay beyond eight years. Teachers in voting districts 1 and 2 would be better off choosing a 401k-style plan where they could leave with retirement account savings rather than nothing.

    Having flexible plan options can give mobile teachers, especially in urban and rural public schools where turnover is high, more secure retirement benefits. School districts should additionally take a role in clearly explaining benefits options to teachers.  

  • Last night Gina Raimondo, the Treasurer of Rhode Island and a champion of penison reform, won the Democratic nomination for governor of Rhode Island. A few quick thoughts: 

    1. Pension reform can be good politics. Despite being a Democrat in state where President Obama took 63 percent of the vote, and despite heavy union opposition and spending against her, public polls showed strong support for Raimondo's leadership on pension issues. She brought facts and data to a knife fight, and she won. 
    2. The pension reform legislation is still tied up in lawsuits, but it would be a good thing for both taxpayers and teachers. Once enacted, it's estimated to save the state $4 billion over two decades and would put the state's pension funds on a much more sound footing. Just as importantly, it would benefit the vast majority of teachers. 
    3. There's still a lot of work to do. Even under the most generous assumptions and assuming the reforms pass legal muster, Rhode Island's teacher pension plan faces an unfunded liability of nearly $3 billion (it has saved only about 57 percent of what it has promised to pay out in benefits). We need more politicians like Raimondo who are willing to tackle pension issues and seek out solutions that are good for their state and good for their workers. Rhode Island just proved it's possible. 


  • The Empire Center and several other organizations have published a database of New York teacher and administrator pensions that lists the pensions and service years of every member. Contrary to its name, the highest pensioners in the New York Teacher Retirement System are not teachers. Fourteen out of fifteen of the top pensioners are former superintendents and one a research professor. None retired as public school teachers.

    In New York, as in most other states, pensions are based on an employee’s years of service and final average salary, and teachers, principals, and superintendents all participate in the same retirement system. While not all administrators are former educators and only serve an administrative position for a few years, many have come into the profession as a former principal or teacher or other administrative pathway, often with years within the same state and local system. Long tenure in a single retirement system, paired with a high superintendent salary, equates to a very lucrative pension. In Missouri, a teacher who stays for a full career accrues $250,000 in lifetime pension wealth, a principal accrues over $360,000, and a superintendent $450,000.

    In New York, one superintendent had close to 40 years of service in the Sewanhaka Central Schools, with only five years as superintendent. His annual pension totals $223,000, significantly more than a teacher with the same amount of experience would get. The highest paid retiree was the superintendent of the Commack school district, who has an annual pension of $325,000 (based on a final salary of over $655,000), the lifetime equivalent of over $5.5 million (a true pension millionaire). These are extreme examples that can occur because traditional pension formulas rely so heavily on final salary and total service years, and many superintendents have accumulated prior service years as teachers or mid-level administrators to count toward a full career. A career educator can work and pay into the retirement system with lower teacher or principal contribution rates for the majority of their working years and still qualify for a pension for the rest of their life based on their much higher superintendent’s salary.

    What is the impact of these top officials on student achievement? One would hope that such sizeable pay and benefits would mean a significant boost in student outcomes. According to a new Brookings report, however, superintendents in Florida and North Carolina had only a minimal effect on student achievement. Compared to teachers, who have a larger impact on student achievement, superintendents had little. Administrators of course play a vital role in other aspects of managing a district, but they benefit disproportionately from the heavily back loaded nature of the traditional pension system.