Primaries Expose Prediction Limits | Eastern North Carolina Now

    Publisher's note: This article appeared on John Hood's daily column in the Carolina Journal, which, because of Author / Publisher Hood, is linked to the John Locke Foundation.

    RALEIGH     The topsy-turvy campaign of 2016 is going to do wonders for many journalists, political practitioners, and scholars. They will produce a steady stream of essays, studies, books, and business plans designed to explain what happened, why others failed to predict it ahead of time, and how their insights can be used to predict or shape future elections.

    Their work will get a lot of attention. Each widely read argument will provoke a counterargument. Magazines and books will sell. Shows will draw audiences. Candidates, activists, consultants, and donors will mix and match into new political organizations. By then, the next set of elections will present themselves - and start the process all over again, probably by invalidating whatever hard-and-fast "rules" are gleaned from 2016.

    See what I did there? I issued a prediction of my own about how people will talk about and react to the lack of success in predicting past political outcomes. I think I'll be proven right. Isn't that what we always think?

    Whomever you're backing for president this year, I think you'll agree with me that many predictions, issued by careful analysts using the best-available information, have turned out to be mistaken. In some cases, candidates who trailed in the polls just before Election Day have actually triumphed, such as Ted Cruz in Kansas and Bernie Sanders in Michigan. In many other cases - think of Cruz in Louisiana, Marco Rubio in Virginia, and Sanders in Iowa - a challenger has dramatically closed the difference with the frontrunner at the end, producing delegate counts so close that "winning the state" didn't end up meaning much.

    Although some pollsters have better track records or sounder methodologies than others, even the best ones produce bad samples and poor predictions at least some of the time. The truth is, elections are complex events with lots of moving parts. Predicting them is inherently difficult, particularly when the election in question is a primary or caucus featuring multiple candidates, sparse polling, and lots of undecided voters.

    If elections are hard to predict, just imagine how hard it is to make accurate predictions about systems that are even more complicated, such as an economy. There is no single "election day" or "early voting period" when it comes to the continuous votes that consumers, workers, investors, and entrepreneurs cast virtually every waking hour of every day.

    The broader the scale, the less predictable these economic elections will be. Centuries ago, when most production and exchange occurred within small communities, the potential for surprising outcomes was lower. In today's regional, national, and international economies, all bets are off - and some bets turn out to be way, way off.

    Products get hot, then are not. People choose occupations, then change them. Companies seem to be at the top of their game, then suddenly collapse due to changes in leadership, technology, or consumer preferences.

    To say that predictions are often wrong is not to say that prediction is pointless. As a matter of fact, it is inevitable. Unless we have at least some sense of where things may be going, we can't act. Planning for the future is a basic, inescapable task. So is adjusting your plan as the facts change.

    The larger problem comes when a small number of "experts" gain the power to make predictions and issue plans on behalf of everyone else. These experts don't necessarily understand the system they purport to manage any better than the rest of us do. More to the point, they can't possibly possess all or even most of the relevant information, as it is widely dispersed among a population where individual preferences, predictions, and plans keep changing. It's what the economist Friedrich Hayek called "the fatal conceit" - the hubris that leads to disaster in centrally planned economies. This insight leads me to favor smaller government and greater individual freedom, here in North Carolina and beyond.

    By the way, I know some readers expected from the start of this column that I'd make a political point by the end of it. Congratulations - you predicted well.