This election, can we really trust the ‘experts’?

Your Facebook feed is likely cluttered with predictions about the election. Filmmaker Michael Moore seems sure that GOP nominee Donald Trump will be our next president. CBS, meanwhile, speculated on a blowout for Democratic nominee Hillary Clinton. The prediction silliness was prevalent during the primaries with such classics as this prediction of a Clinton-Sen. Marco Rubio (R-Fla.) general election (I — ahem — may or may not have made this same prediction to my friends) and this one of a Trump-Sen. Bernie Sanders (I-Vt.) tilt.

{mosads}Political predictions are one thing, but what about predictions about the impact of actual policy changes? Surely those must be grounded in less biased, more reasoned analysis. Not necessarily. New York Times columnist David Brooks recently pointed out that both critics and supporters of the Affordable Care Act overstated their predicted estimates of its impacts. In my area of research regulatory policy, it is common to see predictions that regulations will destroy the economy or save millions of lives. As with the ACA, most of these predictions tend to be overstated. Most of the predictions also tend to be made by individuals who have very clear theoretical views about the role of regulation in the economy.

When should we trust predictions, about politics or about policy? The simple answer is “not often.” The more complicated answer can be found in the work of Philip Tetlock, whose 2006 study of experts was groundbreaking. Tetlock found that experts seldom outperform non-experts in making predictions and that the types of experts we see on television are among those least likely to be correct in their prognostications. He also discovered that experts who know a little about a lot of things are more accurate than those who have one big theory and use it to predict everything.

The proliferation of experts on 24-hour news networks, on blogs of every political stripe and in a Facebook post near you has led to increased exposure to those experts least likely to be right. I would argue that this exposure has increased an already natural skepticism of all expertise. When we see so many predictions, and in particular so many loudly wrong predictions, we become skeptical of everything, including predictions that are highly likely to be true — such as those of a warming climate or a crumbling infrastructure. When some of those experts are being paid to espouse particular views, it deepens cynicism further. 

So whom should you believe? Here are a few hints:

• Does the expert tell you why he or she might be wrong? Does he or she tell you what assumptions might be incorrect? If not, be suspicious.
• Does the expert always reach the same conclusion? Is an event always good for a particular candidate, in the expert’s view? Does he or she think every regulation is bad or every regulation is good? This type of consistency is probably evidence of rigidity and a high likelihood of being incorrect.
• Does the expert stand to gain from espousing a particular point of view? If so, then it should surprise no one that the expert holds this view and it should not be given any deference.
• Is the expert on TV? Bad sign.
• Does the expert reinforce what you already think about an issue? This doesn’t mean that the expert is wrong, but it does mean you should force yourself to exercise extra scrutiny because you are more inclined to accept the expert’s point of view without thinking.

Expertise is important in politics and in policy. But it must be accompanied by a healthy dose of skepticism. But don’t take my word for it — I’m just another expert.

Shapiro is an associate professor and director of the Public Policy Program at Rutgers University and a member of the Scholars Strategy Network.

The views expressed by contributors are their own and not the views of The Hill.