How Deep Is The Teacher Bench?

On most sports teams, coaches assess players in part by considering who is available to replace them. Teams with “deep benches” have more leeway in making personnel changes, because quality replacements are available.

The same goes for teaching. Those who aggressively wish to start firing larger numbers of teachers every year rely on an obvious but critical assumption (often unstated): that schools and districts can find better replacements.

In other words, it is both counterproductive (and very expensive) to fire teachers if you can’t replace them with a more effective alternative. Even those few commentators who have addressed this matter sometimes ignore another important fact: The teacher labor market is about to change dramatically, with a massive wave of retirements lasting 5-10 years. Thus, most current assumptions about the stability and quality of the applicant pool over this period may be unsupportable.

The numbers are a bit staggering.

Currently, including retirements, about nine percent of teachers leave the profession annually – that’s around 300,000 teachers every year. Many of them move to non-teaching jobs in education, and some return to the classroom at a later point, but most need to be replaced.

Then there is the upcoming wave of retiring baby boomers. This wave should (depending on the economy) soon begin in earnest (the first boomers turn 65 this year). It is difficult to predict the trend precisely, and its effects will vary by location, but when the flood gates open, annual teacher attrition could increase by up to 50 percent. That means we’ll need to replace around a half million teachers, perhaps more, every single year (retirees plus “normal” attrition). And this may last for several years.

If this scares you, it should. A half million teachers is roughly equivalent to one-third of the annual graduating class of every college and university in the U.S. combined. If every single Ivy League graduate in a given year decided to be a teacher, this would cover only a fraction of the annual demand. So, beginning very soon, there might be a pretty serious strain on the teacher “bench” – it’s a good bet that we’ll have a tough time replacing all these leavers/retirees without a decrease in the quality of the applicant pool, especially in low-performing schools and districts.

And this doesn’t include any possible uptick in the number of teachers fired based on performance.

Now, those who say that there are some teachers out there who just don’t belong in the classroom are undoubtedly correct. Of course they are. And we need to improve the mechanisms (and speed) by which such teachers are identified, given a chance to improve, and, failing that, dismissed.

But on the opposite end of the equation, “normal” attrition – that not from retirement – is abysmal. So, needless to say, if we really care about teacher quality, we must be equally concerned about keeping “good” teachers in the classroom. The best proven means of doing this, in my view, are increasing salaries and improving supports and working conditions (see here for a review of the retention literature).

But those who clamor for the systematic firing of a significant proportion of teachers every year have a responsibility to address the replacements issue. Some do, but most do not.  In the former category is economist Eric Hanushek, who regularly proposes that the “bottom” 6-10 percent of teachers be dismissed every year, based on their students’ test scores.  When he estimates how this would affect aggregate performance, he assumes that replacements will be of “average” quality (once again defined in terms of test scores).

First of all, many of the “bottom” 6-10 percent leave teaching on their own every year, especially in their first year, simply because they realize they aren’t good at it (teaching has a way of making that abundantly clear). That’s one of the reasons why attrition among newer teachers is so high.  I would also point out that a fair number of the "bottom" teachers fired under this proposal would be wrongly classified as such due to the imprecision of value-added models.

But, that said, I’m curious (and this is a real question): What makes Hanushek (and those making similar proposals) confident that the supply of quality applicants will, given the pending retirement wave and already-high levels of teacher attrition, be sufficient to justify the mistake-ridden and expensive dismissal of tens of thousands of additional teachers every year, relative to a focus on retention and improvement? 

Put differently, do they really think the bench is that deep?