Who's clapping?
At his blog teacherpensions.org, the neoliberal economist Chad Aldeman has made some disturbing claims about teacher pensions. I've been writing about the politics of teacher pensions for organizers, focusing on Aldeman because his blog comes up frequently in searches for information about teacher pensions and he's a big name in the arena.
In June 2023, in a post called "How Much Do School Districts Spend on Teacher Pensions?" he argued that these pensions are inefficient, noting they cost districts as much as they spend on tutoring, about $63 billion a year. He says districts spend $1,549 per student on teacher retirement costs on average. New York City apparently spends more than $4,700 per student on these costs. The implication is that districts can and should be spending this money on students rather than its retirees.
But pensions worsen the lives of teachers too by preventing them from getting paid more, he says. In Chicago, for every $1,000 that goes to teacher salaries, the district pays $520 towards teacher pension costs. In New York City it's $340; $330 in Philadelphia. Finally, Aldeman includes a shocking statistic: that retirement costs for state and local educational agencies have risen 322% since 2004, outpacing healthcare and salary costs.
It's hard to come away from reading the blog thinking anything other than that teacher pensions are in a terrible crisis, and things should be changed quickly and dramatically. The change he recommends is moving towards a defined contribution model rather than defined benefit, or at least a hybrid model. Instead of using solidarity to provide defined amounts of income for teachers in their old age, we should use a market-oriented approach that leaves teachers out in the cold to handle the risk of aging alone as individuals facing the market.
He makes these claims by estimating pension costs for school districts--what districts spend on pensions--and comparing that to district's per-pupil enrollments and costs and per-employee costs in terms of wages. (He also tracks pension debt, which is when school districts owe back payments on their pension obligations.)
I was curious about where all these numbers were coming from and how Aldeman got to them, so I read the methods section, which provides some...illuminating perspective. Kudos that he even included this section, as it undermines much of what he claims in the post. He admits that "The methodology I came up with here is certainly not perfect, and there are some sources of error that I’ll discuss." Let's take a look.
Plausible?
Aldeman begins the methods section by writing: "I set out to estimate how much school districts spend on teacher and other employee pension costs." Three things of note about this seemingly innocuous sentence. First, states make big contributions to teacher pension plans as well as local school districts. So it's reasonable to ask what role state financing plays here, and whether he's not really measuring district expenditure (more on this later).
Second, he includes "other employees" in his estimates. This includes people like superintendents and other highly paid administrators who aren't teachers but are covered by teacher pensions. Should superintendent pension costs be lumped with teacher pension costs? This is a question I've developed throughout my reading. To what extent are the highly paid management positions responsible for increases in retirement cost? Even the disgraced superintendent who embezzled millions of dollars from his district in Long Island, the subject of the movie Bad Education, gets a pension of nearly $180,000 a year.
Third, it's a small thing, but Aldeman uses the word "estimate" with respect to his goals viz. school district expenditures on retirement costs. Later in the methodology section, he says "the numbers here should be treated as plausible estimates rather than precise budget amounts" (my emphasis). What's the difference between a plausible estimate and a precise budget amount?
Basically, he's saying that the numbers he's coming up with can be used but aren't precisely what we'd find in budgets if we actually look at them. Spot the ideology here. What counts as plausible? Who determines what's plausible and what's implausible? The word plausible comes from the same etymological root as applause and plaudit, that is, what's worthy of applause. So who's clapping here and why? If Aldeman has a market-oriented project, rather than a solidaristic one, then data supporting market-oriented conclusions will likely be more plausible, particularly in the absence of precision. In other words, these numbers can be used if you're pushing a market project in pension policy. (He even says "Anyone looking for a precise number for a given district should consult with the district finance office." Right, that is, anyone not doing a market project.)
Attributable?
Let's see an example of how this all works in terms of who makes the contributions to pensions. When it comes to the question about state contributions versus school districts, he says plainly that
I’m attributing all “employer” contribution rates to a given school district, but in some places, the state pays some or all of the “employer” contribution rate. In those cases, I’m attributing spending to a district that comes from the state, so the estimates here should be interpreted more as “spending attributable to a district” rather than “spending that directly comes out of a district’s budget.”
Okay, so Aldeman doesn't know how much a state contributes to a pension for any given district he's talking about...in a post about district expenditure. When he says that New York City or Chicago or Philadelphia is "paying" a certain amount of money towards teacher retirement, he actually doesn't know if it's the district or New York State, Illinois, or Pennsylvania. That makes a big difference, because state funds work differently than local district funds. The money can't just go towards any old thing, like teacher salaries or curriculum or facilities.
Flabbergasting
The turn of phrase "spending attributable to a district" is a wondrous slight of hand consistent with his distinction between precision and plausibility. You can see the ideology still at work. Spending in this sense is 'attributable'--that is, can be attributed-- to a district without really knowing whether or how we can attribute that spending to the district. It's plausible rather than precise. When you break it all down it's flabbergasting.
He's claiming in the main text of the blog post that spending coming out of a district's budget is going to teacher retirement when it should be going to other things, but in the methods, buried further below in a place that people probably won't read, he says he actually doesn't know if the money is coming out of the district's budget or not. The spending is 'attributable', meaning that he feels confident that he can attribute the spending to the districts in the way he's attributing it. So we just have to trust him!
Even though he's asking about school district spending on teacher pensions and making claims about district spending on teacher pensions, comparing that with student enrollment and teacher wages, he admits he doesn't really know if it's the district or the state thats doing this spending. This skillful not-knowing suffuses the post. Going back to the beginning of the post, he writes "It can be hard to get a handle on teacher retirement costs." Indeed! He also says at the beginning that he'll talk about why "we don’t have better answers to these questions in the first place." He admits he doesn't really know and that his answers to the questions aren't that great, but he goes ahead to make his strident claims anyway.
This struck me, particularly as I've been reading recent peer-reviewed research on teacher pensions. There's a lot of ambiguity in teacher pension data, which I'll focus on next week. But just remember to dig into the methods of these things and think critically, raise an eyebrow, to wonder who's clapping when it comes to the methodological magic such 'analysis' and 'research' claim to be doing.