Baseless Criticism of Steven Yamarik's Research

by Andrew Gillen

My previous post highlighting the work of Steven Yamarik, who has recently noted that there do not appear to be positive economic externalities to education, is under attack. I was decidedly underwhelmed with these responses, both by the same individual, and was therefore going to ignore them, but based on the fact that both Mark Thoma and Arnold Kling link to them, I feel a rebuttal is in order.

The author of the responses is clearly in disagreement with the research findings of Yamarik, but does nothing other than to say that he is in disagreement. For instance, in the first response, he states

Think about how ridiculous this conclusion [Yamarik's finding that these is little evidence of economic externalities to education] sounds – and is. You need only make the most minor and realistic assumptions...

In other words, if you just assume that there are positive externalities, it's easy to to show that there are... positive externalities. He continues

I benefit from others getting more educated in a big material way.

To illustrate, he asks us to imagine a world in which no one is educated compared to a world in which everyone is educated. But the issue is not the either or categories of no one being educated vs. everyone being educated. In fact, such a binary choice completely misses the point. The point of taxing or subsidizing things with externalities is to get the optimal level of the activity, in this case education. Very rarely is the optimal level of activity all or nothing. The choice is not "do we educate everyone or no one" but rather "how many should be educated." The answer to latter, more appropriate question, depends on the marginal costs and benefits (both private and social) of education.

For a set of people, the private costs of getting educated are less than the private benefits, so they will attain an education without a subsidy. The subsidy is only needed when the private costs of education are higher than the private benefits, but lower than the social benefits. In such a case, the market by itself will fail to produce the optimal level of education. This failure is the economic rationale for a subsidy, and Yamarik notes that it is an invalid rationale based on the evidence that there is little external return to education.

The second response, is a little better (it doesn't include tangent attacks on the measurement of GDP or a certain political party), but not much:

there still are however enormous positive pecuniary externalities to education. The Vox EU article and studies it cites miss much and I have little doubt there are severe flaws with the models, econometric techniques, and assumptions behind them.

In other words, I don't like their result, so I'm going to assume that they're wrong without actually figuring out why.

Now, it is quite possible that Yamarik's and the other cited research have "severe flaws with the models, econometric techniques, and assumptions." (I have not yet reviewed the work in detail, so I can't say.) If there are flaws, find them and point them out. But you can't just assume that there are flaws whenever you see a result you don't like.