There is no relationship between per pupil spending increases and student performance.

The purpose of public spending on K–12 schools is to educate children. This is a surprisingly controversial stance among educators, who will point to all sorts of metrics other than student performance to justify ever greater spending. Part of the reason for that is probably that the student performance metrics look so poor.

Matthew Ladner of the Heritage Foundation has a new report that looks at the returns state governments have gotten for boosting K–12 education spending. And they have boosted it, by a lot, in real terms. Between the 1999–2000 and 2020–21 school years, the average state increased per pupil spending by 37 percent above inflation.

The largest increase was in Vermont, which doubled inflation-adjusted per pupil spending over that time span. The smallest was in Idaho, which increased spending by only 4.5 percent.

And yet, the average score on NAEP math, reading, and science tests for Vermont students declined by 1.4 percent between 2003 and 2019. The average score for Idaho students increased by 1.2 percent over the same time span. That means Vermont increased spending by 20 times as much as Idaho did and got worse results.

The best-performing state in terms of NAEP score improvement was Mississippi, where students’ average scores increased by 5.1 percent. Mississippi had a slightly below-average increase in per pupil spending, at 32.9 percent. Part of the reason for improvement is likely Mississippi’s “reading revolution,” whereby emphasis on phonics instruction has led to major improvements in one of the country’s poorest states.

What’s most striking about this chart in Ladner’s report is how there is basically no relationship at all between state per pupil spending increases (again, it is adjusted for inflation) and student performance:

This isn’t a function of Covid and school closures. Ladner intentionally stopped the data right before the pandemic to avoid that issue. The reason the NAEP scores begin in 2003 rather than 2000 is that 2003 was the first year all states were required to participate in the tests.

For an international comparison, Ladner looks at PISA math scores versus spending per pupil in 29 countries. The U.S. is one of the highest spenders, yet it has the second-lowest test scores:

It’s long past time to stop pretending that education spending has anything to do with student performance. That means state governments need to be more willing to stand up to teachers’ unions and say no to their endless demands for greater spending. The past 20 years have been an experiment to see whether enormous spending increases would improve student performance. The results show that they haven’t.