When I was a kid, I worked at a local 9-hole golf course. It wasn't fancy. We didn't even take reservations, but we were the cheapest course in town and, inevitably, there would be times when we'd be so busy our customers would have to wait. On a few occasions like this, frustrated patrons would come up to us and ask how to get to the other golf courses around town.
The owner was proud of his course, and he hated hearing this particular question. So whenever he heard someone ask how to get to one of his competitors, he would boom out, "YOU CAN'T GET THERE FROM HERE!"
Clearly this was a preposterous answer. We weren't even located on a cul-de-sac, and anyone could simply retrace their steps and take a different road to a new destination. But that was part of my boss' little joke. His thunderous line didn't always work, but it did confuse and startle some customers into reconsidering and staying put to wait it out.
I think about this gambit a lot, because I often hear pension advocates use the same "YOU CAN'T GET THERE FROM HERE!" refrain. They'll often deploy it in response to data showing that teacher pension plans provide a secure retirement to only a small fraction of incoming teachers, and that those same plans have accumulated more than $500 billion in unfunded liabilities. The latest example comes from a paper written by Tyler Bond and Dan Doonan and published by the National Institute of Retirement Security (NIRS). The paper is ostensibly about what happened when states like Alaska, Kentucky, Michigan and West Virginia closed their defined benefit pension plans, but the paper's argument essentially boils down to saying that state-run pension plans are in such bad shape that they need new members to bail them out, and we must continue with the status quo forever.
On finances, Bond and Doonan argue that closing a poorly funded pension plan does nothing to pay down existing unfunded liabilities. This is true, but it ignores how the pension plans accumulated those unfunded liabilities in the first place. As Max Marchitello wrote in a recent report looking specifically at the case of West Virginia, when the state closed its defined benefit pension plan, it did little to address the plan's existing unfunded liabilities. The state still had to pay down those debts, and it largely failed to do so. But rather than blaming the state for accumulating those debts in the first place, or for failing to pay them off in a timely fashion, Bond and Doonan try to pin the blame on the shift to a defined contribution plan. That's a clever rhetorical trick designed to distract us from the question of whether the state's pension plan was good for the state or the majority of its teachers. (It wasn't.)
Similarly, Bond and Doonan attempt to argue that Alaska has trouble recruiting and retaining teachers because of its shift away from a defined benefit pension plan. But the authors neglect to mention that Alaska has always had trouble recruiting and retaining teachers. When we looked at the data on teacher turnover rates in Alaska over time, there was no visible effect of the retirement plan change. Teacher turnover was high in Alaska when they offered teachers a defined benefit pension plan, and it remains high now that new teachers are enrolled in a defined contribution plan. This finding squares up with what we've seen in other states as well; teachers just aren't that responsive to changes in their retirement plans.
Just like my former boss, pension advocates are using the "YOU CAN'T GET THERE FROM HERE!" refrain to scare legislators from asking tough questions about their state's pension plans. While responsible policymakers should be mindful of the effects of any potential change, and they should protect the promises made to current teachers and retirees, legislators should not be deterred from designing better solutions for the future. Ideally those plans would provide all new teachers with a path to a secure retirement.
Most people probably don’t realize not all workers are covered under Social Security. In particular, teachers constitute one of the largest groups of uncovered workers. Nationwide, approximately 1.2 million teachers (about 40 percent of all public K–12 teachers) are not covered under Social Security for their time in the classroom.
How did we get here?
Why are some teachers covered, and others aren’t?
As Leslie Kan and I write in our 2014 report, the exclusion of teachers from Social Security comes from decisions made decades ago. State workers were left out of the original Social Security Act in 1935, initially because of concerns whether the federal government could tax state and local governments. Later when states were given the opportunity to extend coverage to public sector workers in the 1950s, most states chose to extend coverage. A handful of states, however, chose not to. Instead, these states bet they could provide better benefits through their state pension plans alone than through the combination of a pension and Social Security. Indeed, pension benefits for full-career workers typically have a higher rate of investment return than Social Security. However, this arrangement works well only for the small percentage of teachers who stay 30 or more years in a single retirement system.
Today’s reality is that half of all new teachers will not stay long enough to qualify for a pension at all; and for those who do qualify, many will receive pensions worth less than their own contributions. For teachers in states without Social Security coverage, they're at risk of leaving their public service with very little in the way of retirement savings.
Today, the majority of uncovered teachers work in 15 states (Alaska, California, Colorado, Connecticut, Georgia, Illinois, Kentucky, Louisiana, Maine, Massachusetts, Missouri, Nevada, Ohio, Rhode Island, and Texas) and the District of Columbia. Additional states have varied coverage where some teachers remain left out.
Social Security Coverage for Teachers
Source: National Education Association, "Charteristics of Large Public Education Pension Plans."
Social Security coverage varies within states and sometimes even within districts. In California, almost all state government employees, state legislators, and judges are covered by CalPERS and Social Security. Teachers enrolled in the California Teachers State Retirement System (CalSTRS), however, are not covered by Social Security. Within California schools and districts—superintendents and district employees tend to be covered by Social Security, whereas classroom teachers tend to be uncovered.
States aren’t locked into keeping their teachers out of Social Security. When Social Security coverage was extended to the states in the 1950s, each state entered into what is called a Section 218 agreement with the Social Security Administration, detailing the extent of coverage. Today, federal law allows any state or local retirement system to modify their Section 218 agreement and join the program (states that opt into coverage can’t subsequently opt out). Certain states have state-level legislation that prohibit teachers from extending coverage, but most states do not have these barriers.
This provides an opportunity for states to reconsider their decades-old decisions. While not sufficient as a stand-alone benefit, Social Security could provide teachers with a floor of secure, inflation-protected, and portable benefits – something many teachers don’t have and genuinely need.
To learn more about teachers and Social Security coverage, read our report or watch the video below:
Earlier this month I wrote a piece showing how state pension formulas exacerbate teacher salary gaps across districts. Due to the way pension benefits are calculated – relying so heavily on salary and experience – a pay disparity between teachers with the same number of years of service will be compounded in retirement. Moreover, pensions are yet another way that more money is spent on educating students in low-poverty districts than in high poverty ones.
In this piece, I will tackle how experience in state teacher pension formulae can deepen existing school funding inequities.
Higher teacher turnover rates are not evenly distributed among districts or states. High poverty and rural districts tend to have higher teacher attrition. Indeed, high-poverty districts have been shown to lose teachers at nearly twice the rate as low poverty districts. Practically speaking, this means that less money is spent on both teacher salaries and pensions in low-income communities than in more affluent ones.
Deeper Dive into Turnover in West Virginia
Let’s look at how this plays out in West Virginia, a state pension system I’ve written about a lot recently. As shown in the graph below, as a county’s poverty rate increases, so does the average teacher attrition rate.
Of course some of that turnover is due to retirement. However, the fact remains: counties with higher attrition rates are likely to have greater concentrations of new, and inexperienced teachers. And importantly, newer teachers have much higher turnover rates than their more tenured colleagues.
Consider the following hypothetical example of three school districts with different turnover rates based on West Virginia’s average attrition rate described above. I set the yearly low-and high-turnover rates at a third above or below the average rate of 9 percent. While teacher turnover tends not to be this linear – with very high attrition in the first year – the idea is to show how different turnover rates, and thus levels of experience in a district, affect school finance. The table below shows my results. After five years, the low-turnover district will have lost about 30 percent of its new teachers, compared to the high-turnover distirct, which will have lost 57 percent of its teachers. In a state like West Virginia with a five-year vesting period, low-turnover districts could easily have about twice as many teachers qualify for a pension than in high-turnover districts.
High turnover districts in West Virginia, and indeed across the country, disproportionately serve low-income students and communities. Those vacancies are often filled with new teachers who themselves are more likely leave the profession after only a few years. As a result, high poverty districts have a disproportionate share of newer teachers who earn lower salaries and less valuable pensions. Ultimately, due to variable retention rates, less money is spent to educate students in high poverty districts than in more affluent ones.
How Teacher Turnover Interacts with State Pension Systems to Affect School Finance Equity
Teacher pensions are calculated based on two key factors: years of service and salary. The longer a teacher stays in a classroom the greater her pension wealth will be. This is because salaries and experience interact: teachers with more experience typically earn higher salaries.
School districts with higher teacher turnover will have less experienced teachers earning lower salaries and thus qualifying for less valuable pensions. In other words, less money will be spent in the aggregate on teacher compensation in districts with high teacher attrition. This pattern exacerbates existing school finance inequities since districts with high teacher turnover also tend to serve low-income communities.
The problem is likely worse than a first glance suggests. This is because many teachers in high-turnover districts will leave the profession before qualifying for pension benefits at all, not to mention serving long enough to reach the valuable back-end benefits. This adds to the problem of high-turnover, and often low-income districts spending less money on total teacher compensation and thereby less on educating their students. Often the largest source of teacher turnover in low-income districts is from teacher "movers," who go on to teach in a different district. Although these teachers keep their state pension benefits, they add churn to the districts they leave, which can negatively affect students.
There is no easy solution to this problem. Teachers leave the classroom for any number of reasons. However, state policy makers should understand the ways their policies interact and look for ways to minimize both the impact that can have on teachers’ ability to save for retirement, and funding disparities between high-and low-poverty schools.Taxonomy:
This may sound counter-intuitive, but here it is: Technically speaking, Ohio school districts do not contribute toward Ohio teacher pension benefits.
"How is this possible?" you might ask. After all, Ohio school districts are contributing 14 percent of each teacher's salary into the pension fund.
But wait, where is that contribution going? If you pull up the latest actuarial valuation report from the State Teachers Retirement System of Ohio, you can find out. Table I-1 shows that the plan estimates the "normal cost" of the benefits are worth 10.91 percent of salary. That is, across all individuals who enter the plan, after accounting for their age or how long they might stay, the plan thinks the promised pension benefits are worth an average of 10.91 percent of each teacher's salary.
You'll quickly start to screw up your face, especially if you know that every teacher is currently contributing 14 percent of their salary into the plan. Fourteen is more than 10.91 percent, how can that be?
This is due to the fact that Ohio STRS has accumulated unfunded liabilities of $24.8 billion. Every single STRS member is contributing 3.09 percent of their salary (14 percent - 10.91 percent) to pay off that debt.
That's not all. In addition to the employee contributions, school districts are also paying in 14 percent of each teacher's salary into STRS. That money is going into the plan, but none of that is going toward benefits. All of it is going to being used to pay down the unfunded liabilities.
In essence, Ohio has created a system where teachers, on average, are getting less out of their pension plan than they themselves put in. To be honest, it's hard to even call this a "retirement" system at all. The system is functioning like a debt accumulation tool and a tax on teachers, with retirement benefits on the side.
Again, this may be sort of hard to wrap your head around, but it's true. The figures above are all based on what the state's actuaries think the Ohio STRS plan will cost over time. Ohio is the only state in such a bad situation overall, but Illinois teachers hired as of 2011 are also paying more into the system, on average, than the state's pension plan thinks their benefits are worth. Other states may be in similar territory for new, less-generous benefit tiers, but they rarely report those data separately.
In contrast, Ohio also offers new teachers the option to join a defined contribution plan with a 9.53 percent employer match. For the vast majority of teachers, that's likely to be the better option.
In Georgia, K-12 school teachers are automatically enrolled in a defined benefit pension plan.
In contrast, at Georgia’s public colleges and universities workers are given a choice between the same pension plan or a defined contribution plan.
Why don’t K-12 teachers have the same choice as their peers in higher education?
Georgia's K-12 teachers deserve a better choice, because the current plan run by the Teachers Retirement System (TRS) of Georgia plan is poorly designed for the vast majority of its enrollees. The TRS pension plan requires a 10-year vesting period before workers qualify for any retirement benefits; this is so long that it would be illegal in the private sector. Georgia also does not provide Social Security coverage to all of its K-12 teachers, leaving some teachers entirely dependent on their TRS benefits. And yet the TRS formula delivers adequate benefits to only a fraction of members who remain teaching in Georgia for their entire career.
The vast majority of Georgia’s K-12 teachers would be better off in a plan offered to the state’s higher education employees. Unlike elementary and secondary teachers, workers at Georgia’s public colleges and universities are allowed to choose between either the TRS plan or the Optional Retirement Plan (ORP). Originally designed for university professors moving across higher education institutions, the ORP also has a number of features that would be attractive to K-12 teachers.
In contrast to the TRS plan, the ORP offers immediate vesting, which means that all workers begin accruing retirement benefits immediately, starting from their first day on the job. Georgia higher education employees are also covered by Social Security, meaning they receive a nationally portable, progressive benefit that many of Georgia’s K-12 teachers lack.
It may sound counter-intuitive, but the ORP is simultaneously more generous to employees and less costly to employers. It’s due to how each plan works. The TRS plan is based on a series of assumptions. It offers benefit through a formula, and the state estimates how much those benefits will be worth down the road and how much the plan needs to invest today in order to pay for those future benefits. When the state is wrong about any of those assumptions, or when it fails to save sufficiently, pension plans can accrue unfunded liabilities. Today, Georgia school districts are paying a total of 20.9 percent of each teacher’s salary into the TRS pension plan, but only about one-third of that (7.7 percent of each teacher’s salary) is going toward benefits for workers. The rest (13.13 percent of salary) is going to pay down those unfunded liabilities.
In contrast, the ORP has no unfunded liabilities, and it never will. Under the ORP plan, a worker's benefit is defined in terms of how much is contributed into the plan, and how fast those investments grow. There are no assumptions to worry about, and hence the plan will never accumulate unfunded liabilities. Georgia’s colleges and universities today are contributing 9.24 percent of each participating employee’s salary into the plan. That is both more generous to workers than the TRS plan (9.24 versus 7.77 percent of salary toward benefits) and less costly overall (the same 9.24 in ORP versus the total 20.9 percent in TRS).
Comparing Key Elements of Georgia Retirement Plans
Teachers Retirement System of Georgia (TRSGA)
Georgia Optional Retirement Plan (ORP)
Employer contributions for benefits
Employer contributions for unfunded liabilities
Total employer cost
Social Security coverage
Varies, depending on school district
There is one other element we haven’t discussed yet, and that is fairness. The ORP plan is more generous to participating workers on average, and it treats all workers the same. Every ORP plan member contributes the same percentage of their salary, and every ORP member receives the same percentage* contributed on their behalf.
In contrast, the TRS plan does not distribute benefits evenly. Its formula awards disproportionately large benefits to teachers who stay for 30 or more years, while leaving everyone else with substantially less. The 7.77 percent employer contribution that's going toward benefits in the TRS plan is an average across all members. Some teachers will eventually qualify for benefits worth much more than that, while many more will qualify for less.
The graph below shows how benefits would accumulate for teachers under the TRS pension plan versus the ORP. As mentioned above, the plans are not cost-neutral, and I’ve assumed a more conservative rate of return in the ORP plan than what TRS assumes. Still, the ORP would provide more value than the TRS plan for this hypothetical teacher’s first 30 years of service as a teacher in Georgia.
This graph also does not include Social Security; it’s purely a comparison of the two state-provided retirement plans. After factoring in Social Security, it's clear that participants in the ORP are receiving much more generous retirement benefits than teachers in the TRS plan alone. While some teachers might still prefer the pension plan, Georgia should at least give its K-12 teachers the same choices offered to the state's higher education employees.
*Note: The actual dollar amounts contributed will vary based on a worker’s salary, but that’s true in both the ORP and TRS plans.
Teacher pensions exacerbate differences in teacher salaries across districts. As we’ve written previously, this has significant consequences for school finance equity. The inequity in per pupil funding between high-and low-poverty districts is even worse when pension spending is factored in. We’ve also found that pensions can deepen gender-based pay gaps in education.
In each case, the problem begins with salary disparities. Since pension plans multiply salaries as a part of their formulas, any inequities in salary will be magnified by the pension plan. In this piece we will explore how differences in districts’ salary schedules – particularly between those serving high-and low-concentrations of students living in poverty – are compounded and grow through the pension system.
Salary differences, however, are not the only way that state pension funds exacerbate inequities. In a forthcoming piece we will examine how differences in district-level retention rates also can impact school funding equity.
Differences in salary across high-and low-poverty districts
More affluent school districts can afford to pay their teachers higher salaries. Often, districts serving a more affluent community offer higher starting salaries than other nearby districts, which can be an effective recruitment tool. The differences don’t stop there. Throughout a teacher’s career, these districts have the ability to offer more generous raises and set salary schedules with greater maximum salaries, potentially increasing retention. In short, the teacher pay disparity between districts often can increase throughout a teacher’s career.
A new data tool from EdSource helps to illustrate the problem by exploring school- level teachers’ salaries across six counties in the San Francisco Bay Area (Alameda, Contra Costa, Marin, San Francisco, San Mateo, and Santa Clara counties). For each district, the tool maps teachers’ lowest starting salary, their expected salary at 10 years of experience, and the maximum salary they could earn.
There is significant disparity across districts for starting, middle, and final salaries. Among unified school districts, Martinez has the lowest starting salary of $43,123, compared with $70,595 in Santa Clara. That is a $27,472 salary gap for a first-year teacher. The disparity is far greater among teachers earning the maximum salary. Palo Alto has the highest maximum salary for a unified school district, at $131,343, compared with $80,739 in Emery Unified. That is a salary gap of $50,604.
Consider Oakland and Santa Clara Unified school districts. As shown in the graph below, at the same level of experience, teachers in Santa Clara make considerably more than teachers in Oakland. Indeed, a 10-year veteran in Oakland will earn $6,471 less than a teacher in her first year working in Santa Clara. The maximum salary in Oakland is nearly $23,000 less than a teacher’s salary in her tenth year of service in Santa Clara. Teachers with the same level of experience can earn wildly different salaries depending on where they work. In short, the district a teacher teaches in matters significantly.
Pensions Increase Inequity Between Teachers in Oakland and Santa Clara
These disparities don’t stop while teachers are working; they continue into retirement due to the differing pensions they earn. Teachers in Santa Clara not only earn more than their colleagues in Oakland while they work, they also can expect much higher pension payments for the rest of their lives once they retire. In fact, the additional pension wealth paid to a teacher in Santa Clara is far greater than the higher contributions they paid to the pension fund due to their higher salaries while working.
If a 10-year veteran teacher in Oakland and Santa Clara both elect to retire, they will earn markedly different pensions. Indeed, the Oakland teacher’s annual pension benefit would only be about 60 percent of the pension provided to the Santa Clara teacher. Put another way, an Oakland teacher who retires after 10 years of service would earn an annual pension that is $8,519 less valuable than a teacher with the same level of experience in Santa Clara. If these two hypothetical teachers draw a pension for 20 years, the teacher from Santa Clara will receive an additional $170,380 over the course of her retirement even though she worked for the same number of years.
The disparity is even greater for those teachers who spend their entire career in the classroom and earn the maximum salary. When a 35-year veteran teacher earning the highest possible salary in the district retires in Oakland, she can expect an annual pension of $58,600. A teacher with the same number of years of experience and also earning the maximum salary in Santa Clara earned an annual pension of $88,500. That’s a yearly pension gap of $29,900 even though the teacher spent the same amount of time in the classroom and earned the highest salary offered by the district. If again these teachers draw a pension for 20 years, the teacher from Santa Clara would receive an additional $598,000 over the course of her retirement.
Since pension contributions are made as a share of a teacher’s salary, higher salaries do mean higher pension contributions. However, the slightly higher contributions made by a teacher from Santa Clara is vastly outweighed by a much larger pension windfall. Indeed, since teacher pensions are paid yearly and are determined by a formula based on years of experience and final salary, the additional pension wealth the teacher with the higher salary will receive far exceeds the value of the additional contributions she made. On net, a higher-paid teacher pays marginally more in pension contributions, but receives back considerably more than they invested in overall pension wealth.
High-Poverty Districts Subsidize Pensions in Wealthier Districts
Spending on retirement, which in California is shared between employees, districts, and the state, is yet another way that more funding is spent on the education of students from more affluent communities than on students from less financially well-off communities. In Oakland, 74 percent of students are eligible for free-and reduced-priced lunch (FRPL), a common measure of student poverty. In Santa Clara, on the other hand, 43 percent are eligible for FRPL.
Santa Clara has a greater capacity to pay teachers higher salaries, which can help the district recruit and retain effective educators, all while providing those teachers with more valuable retirement benefits which will be subsidized by every other district across the state. Pensions compound existing disparities between high-and low-poverty districts.
Although teacher pension systems are complicated, these examples illustrate how statewide pension plans can magnify school finance inequities. This is largely due to their structure and the fact that affluent districts can raise salaries without being fully responsible for the corresponding increase in retirement costs. Some states address this problem by pushing a greater share of pension costs back onto the districts themselves. While this does more closely align incentives, it nevertheless still puts higher-poverty districts at a disadvantage since they lack the necessary resources to provide higher and more competitive compensation packages to teachers.
While there are no easy solutions to this problem, policymakers should work on improving their pension systems to find ways to ensure that they do not contribute to and exacerbate inequities among districts in the state.