Force decides
How do we know what's really going on with teacher pensions? It's a big problem.
Andrew G. Biggs, writing in 2022 for a special issue of the research journal Educational Researcher, says that "public-sector pensions are not covered by the Federal Employee Retirement Income Security Act of 1973, meaning that there is no centralized collection of public pension financial disclosures."
So if there's no central place to analyze public pension numbers then you're going to have competing claims using competing datasets. Its also notable that there's no information on public pensions before 2001, so you can't do any long-term analysis that would show current trends as relatively normal or reasonable.
Probably my favorite line of Marx is that between equal rights, force decides. He's talking about how workers own their labor and sell it on the market to capitalists who purchase it, but because each has an equal right as buyer and sellers of that strange commodity, you get class struggle. This struggle manifests everywhere in society since it's how we make our material lives. We could say too that where equal numbers exist, force decides. If you've got a dataset that you think is reasonable and you've got backing, then it's the force of class struggle that decides the truth of your claims rather than some other criteria. The same is true, it turns out, for teacher pension research.
Biggs has been publishing on teacher pensions for the last twenty years. He's up front about where his numbers come from and their specific limitations. He gets his data from the Public Plans Database (PPD), kept by the Center for Retirement Research (CRC). A quick look at the CRC website shows that:
This website is developed and maintained through a collaboration of the Government Finance Officers Association (GFOA), MissionSquare Research Institute, the National Association of State Retirement Administrators (NASRA), and the Center for Retirement Research at Boston College. A variety of foundations and corporate partners provide additional support for initiative projects.
Maybe for another post I'll dig more into the revenue sources for GFOA, MissionSquare, NASRA, and the CRC at Boston College. On this page, the PPD don't reveal the "variety" of foundations and corporate partners so we don't know who benefits from research on this dataset, but the vibe is public retirement finance.
Interstellar
In any case, Biggs tells us that his numbers (it's an essay about pension solvency, btw) come from 31 teacher pension plans in the PPD dataset, which itself has limitations. Due to the lack of centralized data reporting generally not every plan has information for every year available. Another shortcoming of the PPD is that
its figures are as reported by teacher pensions in actuarial valuations, comprehensive annual financial reports, and other filings. These figures are calculated using standards promulgated by the Governmental Accounting Standards Board (GASB). Most notably, the GASB allows the present value of future pension benefit liabilities to be calculated using a discount rate equal to the expected return on the pension's investments. Generally, this discount rate has been in the range of 7-8%, based on the high proportion of risky assets held in teacher pension plans. Other figures, such as annual required contribution rates (ARCs) and funded ratios, are based on these present values.
I'm going to get into this stuff more in a future post, particularly the category of pension liability, the discount rate, and its relationship to the annual required contribution rate. Basically, contributions to pensions by states and districts get calculated by thinking about what the pension will be able to pay out when people retire in the future. It's kind of an interstellar braintwister where you have to make all kinds of assumptions about what today's money will be worth when teachers retire in the future. (Yes, I'm referring to the movie Interstellar there. I'm thinking of that scene where the father looks back on the present and past from the future.)
But this all means that, when we dig into the actuarial approach to teacher pension costs, there can be even more ambiguity in the reported numbers we use when we think about them. Who is making these interstellar calculations, for instance, and how? Biggs hauntingly notes that "the most detailed data on teacher retirement system liabilities are calculated by using higher discount rates than most independent analysts believe are appropriate." This has important implications for the pension reform debate, since pension liability is basically what states and districts owe to keep them solvent. One wonders why the "make teacher pensions defined contribution plans" people don't talk more about how liabilities are calculated?
Like, for instance, the Bureau of Economic Analysis's National Income and Product Accounts (NIPA) and the Federal Reserve's financial accounts of the US "publish pension liability figures that are calculated by using a corporate bond yield as the discount rate." Is a corporate bond rate better for a discount rate calculation when thinking about teacher pension contribution costs? Or should we assume 7%-8%, which is higher than analysts think is appropriate?
I'm thinking of the fluctuation in inflation over the last couple years, how we're in a new regime of rates after a couple decades of quantitative easing that funded the neoliberal non-recovery after 2008. Wouldn't all that impact our understanding of what pensions have for future retirees, and thus what school districts and states owe? How does market or solidarity ideology influence the actuarial decisions here? And then how does that shape the pension debate itself? It's up the researchers and analysts behind the datasets, and the decisions they make come to bear on the numbers we read in reporting and research about the pensions themselves.
Force decides
It's an epistemological problem in teacher pension policy, which quickly becomes a political one. How can we know the truth about teacher pensions? To what extent should bond yields or average risky assessments over time, or regimes of interest rates determine teacher pension contributions? It's not clear. Between equal numbers, force decides: if the analysts are pegging things to investments or risk or rates or the Fed, and thus have an interest in those kinds of entities, they'll make that calculation.
To finish this post on Biggs, I'll just note that he's studying pension solvency--whether they have enough money to cover their obligations--and thus the difference between discount rates, for instance, comes to bear on the problem of what we should establish as the contribution rates that school districts and states should be making to pensions so they remain solvent. This is different than the question that Chad Aldeman was asking about what districts and states actually are spending on those contributions. But as we'll see next week, that number is also debatable.