Trust the data, not the vibes, on the teacher labor market

Actions speak louder than survey results

The “vibes” of the teaching profession are not good. This week, PDK reported that just four out of ten Americans would want their child to go into teaching, setting a new all-time low.

That sounds bad.

Except, maybe the PDK poll isn’t such a good predictor of where the teacher labor market is going. I contrasted their poll results with the number of people earning education degrees over time. The graph below shows the results. The green arrows represent the three highest readings from PDK on this question over time, whereas the red arrows show the three lowest readings.¹

As you can see in the graph, the highest PDK readings have occurred just before or right at peaks in the number of people going into education. When more young people are going into teaching training programs, adults are telling PDK pollsters that they want their children to go into teaching.

Meanwhile, the PDK lows are all on the opposite end of the spectrum. They have all happened either at or near troughs in the number of people getting education degrees.

Now, this could be a coincidence, and I wouldn’t want to put money on this as a counter-cyclical indicator. It’s possible that state and district leaders see poll results like this and respond accordingly, by making the teaching profession more attractive in some way.

It would be a nice economic story if we could line up this graph and show that average teacher salaries respond to changes in the teacher pipeline. That did happen in the 1970s and 80s. Average salaries fell in the 1970s as people surged into teachers colleges, and then salaries rose in the 1980s in response to the big decline in supply.

But that relationship seemed to die around 1990. In inflation-adjusted terms, teacher salaries have been remarkably flat since then, despite a big rise and then a subsequent fall in the supply of new teachers.

As I wrote last year, the stagnation in teacher salaries is caused by a few trends, including higher benefit costs, higher staffing ratios, a higher share of full-time workers, and big raises for part-time workers. It’s also worth noting that this is not a funding issue—school spending has outpaced inflation even as teacher salaries have not.

But even if the PDK data is not a perfectly counter-cyclical measure, it’s still important to note that it’s just a poll. And, fortunately, we have better measures of the teacher labor market than this type of survey data. According to those measures, schools have more teachers than ever, salaries are finally starting to rise, and the teacher pipeline is starting to grow again.

Want more reasons for optimism? I published a piece at The 74 this week making the economic arguments for why even more people should consider teaching.

1. I dropped the 2024 result since we don’t have the teacher preparation numbers yet.