Beware Economists Bearing Gifts

Emily Oster thinks that screen time for kids is probably fine.  Her analysis is both original and well-considered—just not in the same places.


Economists, more than people from any other discipline it seems to me, love to write outside their area of expertise. You can find economists confidently telling you about how marriages work, about the epidemiology of child injury, or about why obesity has increased. Sometimes there are valuable insights in this work, but usually the economist just ignores all the other really important factors that he or she doesn’t understand. This problem of economist-overreach is a growing and really annoying problem. It isn’t just that economists pretend to an authority they don’t actually have, it’s that people seem to actually believe them.


Karl Marx and Adam Smith swap ideas, courtesy of Pedro Reyes
Karl Marx and Adam Smith swap ideas, courtesy of Pedro Reyes


As an economist myself, I’m sensitive to this issue. I try to get around it by collaborating with real experts in a field and by publishing in journals in the field I’m writing for, not in economics journals. When I’m writing about child health, for example, I expect to get a much more challenging and insightful review from a pediatrics journal than from an economics journal. Not all economists seek out careful, expert scrutiny of their work.


Oster is right that how children use the medium and what is on it are just as important as the amount of time that is spent with a screen. That is certainly true and not at all new.   But mostly she’s out of her depth.


Oster cites several papers finding an association of high rates of television viewing in childhood with outcomes like obesity, poor cognitive development and poor executive function, and she finds them wanting.


But all these studies have an obvious problem: the amount of TV children watch is not randomly assigned. In the general population, kids who watch a lot of TV — especially at young ages — tend to be poorer, are more likely to be members of minority groups and are more likely to have parents with less education.


This is a dull person’s idea of what a clever critique sounds like. But if an economist says something with sufficient smugness it can pass as wisdom, when in fact it’s plain wrong. In many studies these confounders are indeed controlled (often very well) including in some that she cites.


OK, we can’t expect economists to know everything, but reading the papers that one critiques is surely not too much to ask.


A solid grasp of this field would reveal a very different picture than the one Oster paints. We have good evidence about the effects of television viewing from short-term randomized-control experiments of food choice, obesity, aggressive behavior and imaginative play. We also have results about the long-term, longitudinal association—yes, controlling for many important confounding factors—of television with executive function, cognitive development, obesity, sleep, aggression, and other outcomes.


So far no one has been able to convince any funders to do a multi-site, multi-year randomization of television reduction to see what its effects would be. But the literature taken as a whole clearly suggests that excessive viewing, particularly of violent or even just non-educational content, probably causes poor health outcomes.


Oster seems to draw the strength of her convictions from one paper published in a top economics journal that found that the introduction of television in the 1940s and 1950s had virtually no impact on test scores in middle school. That paper is indeed well done. But it is only one paper, and it’s more than hasty to conclude that screen time today is “probably just fine” because TV in the 1940s had little impact on standardized test scores.  TV in the 1940s is different than media today. And a finding of no impact on standardized test scores in middle school may have very little to do with other important outcomes (or even of academic progress—but that’s a whole different issue). Indeed, TV got to many parts of Canada later, and in a study with a very similar design found that the expansion of TV in the 1970s had significant adverse effects in many areas.


Economists are quick to pat each other on the back, telling each other that they’re very clever. But just because economists think other economists know what they’re talking about, doesn’t mean the rest of us have to agree. The next time you see an economist confidently weighing in a matter outside of economics, take it with a huge grain of salt.



Leave a Reply

Your email address will not be published. Required fields are marked *