Look at the data (and make sure it’s the right data). Then put it in perspective.

by Russ Roberts on February 1, 2013

in Data, Education

Over the years I’ve become increasingly skeptical of the ability of multivariate econometric techniques to tease out causality in a complex world. Some people misinterpret this to mean that data are irrelevant or that facts don’t matter. On the contrary. Facts matter. Evidence matters. Empirical exploration is very important in both the physical and the social sciences.

But the truth is elusive and even facts are sometimes hard to nail down or misleading.

I was reminded of this by a recent op-ed by Nicholas Kristof. Kristof is talking about this wonderful success story–the triumph of an inner-city school in national chess competitions, chronicled by Paul Tough in his book, How Children Succeed, and discussed in this EconTalk episode.

Then Kristof says this:

I wish the column could end on this triumphant note. But if these extraordinary kids are a reminder of what can happen when we invest in creating opportunity, they are also a reminder that budget cuts fall disproportionately on the needy.

“Funding for extracurricular activities has dried up,” said John Galvin, an assistant principal who oversees the 95-member chess team. The kids run bake sales, candy sales and walkathons to raise the $50,000 needed to attend tournaments each year, but on trips they sometimes survive on peanut butter.

Galvin has tried approaching corporations and hedge funds for donations but has had little luck. Budget cuts have already trimmed the after-school chess club to three days a week from five.

A moving documentary about the team, “Brooklyn Castle,” is scheduled to air on PBS later this year, and that may help with fund-raising.

But similar cutbacks are playing out all across America. In 35 states, inflation-adjusted school financing is below 2008 levels, according to the Center on Budget and Policy Priorities. As of July, school districts have slashed 328,000 jobs since 2008, and budget cuts have devastated early childhood education that lays the foundation for children’s lives.

Affluent kids continue to enjoy nursery school and chess tutors, even as programs for poor kids are eliminated. Education is the best escalator out of poverty, but for too many kids it’s creaking to a standstill.

As we make historic fiscal decisions in the coming months, let’s not balance budgets by slashing investments in our future. That would be like economizing on heating bills by feeding the front door into the fire.

The key “fact” in this sad story:

In 35 states, inflation-adjusted school financing is below 2008 levels, according to the Center on Budget and Policy Priorities.

The first fact is a bit uninformative–that school funding is below 2008 levels could be a bad thing or it could be relatively unimportant. You would want to know the size of the decrease. But there is a bigger problem. The link to the Center for Budget and Policy Priorities that Kristof provides isn’t about school financing. It’s about the level of state aid to schools. So it excludes changes in local or federal spending. The link does provide some magnitudes. The median change is a reduction of $216 in per-pupil spending. That’s not a very big cut. But again, it’s not an actual cut in school spending. It’s a cut in state aid to schools. So Kristof’s fact, while potentially interesting, is not the right fact to make the point he is making.

Where did the Center for Budget and Policy Priorities get their data? In their charts they say the source is “CPBB Budget Analysis and National Center for Educational Statistics enrollment estimates.

The National Center for Educational Statistics (part of the Department of Education) actually publishes total public school spending data, not just aid coming from state governments. These data are a better measure of what is going on with school expenditures. The latest data they provide is 2010. Not as good as 2013 but at least it’s actual spending, not just state aid. Here is what they find on the national level:

Click on so you can read it better. Yes, real per-pupil expenditures grew more slowly during the recession and the recovery. But they didn’t go down at the national level. They’re up 3.7% in real terms. What about individual states? Go to Table 5 in the report. Only eleven states are spending less per-pupil in 2010 than they did in 2007. The biggest cut is in Hawaii–$301 which was a 2.5% reduction.

New York, the state where the middle-school chess team is from (Brooklyn) and where Mr. Kristof’s paper is located, increased per-pupil spending from 16,506 to 18,167. That’s a 10% increase. Yes, tough times in the state of New York. Of course that’s for the whole state. Some school districts are surely doing worse than others. And that brings me to the central point that is missing from Kristof’s column. Yes, sometimes you have less money to spend. When that happens you cut the least essential items. Is there any evidence that schools actually do this? My impression is that school budgets have grown through smaller class size and increases in the numbers of administrators. And surely higher salaries for staff, generally. Maybe there should be fewer administrators. Maybe there should be fewer teachers (and yes some schools have cut teachers.) Maybe there should be less money spent on computers and more on the chess team.

After the recession and a year into the recovery–2007-2010, schools spent more, not less, on their students. Of course it is possible that when the 2013 data come out, those numbers will actually show cuts. And yes, some states and some schools surely have fewer resources. For those who are spending less, isn’t it possible that there are things to cut before we cut the life-transforming chess team? Who in a school or on the school board has the incentive to make spending cuts wisely? If anything, isn’t their incentive to do it poorly making the case that the school simply needs more money to do its job? And people like Nicholas Kristof make that case for them without looking more closely at the data. It’s a shame.

Finally, you have to put facts in perspective. Only 3 states spent less per-pupil in real terms in 2010 than they did in 2006. Was 2006 a year of great educational hardship? Look at the trend in the data. For the states that have actually cut spending, the cuts come after many years of increases. What do we have to show for those increases? Not so much, I suspect. Are cuts really going to devastate our children’s future? Almost certainly not if those cuts were made by someone who actually had an incentive to do it wisely. But who in the system has that incentive?

Here is what Kristof tweeted about his column:

In my last column before book leave, I argue that education is the most important entitlement

He may be right. Alas, education and spending on education are not the same thing.

Be Sociable, Share!

Comments

comments

Add a Comment    Share Share    Print    Email

Previous post:

Next post: