Predicting Recessions

Or, how to tell a snake oil salesman

Photo by petr sidorov on Unsplash

”So, do you think there will be a recession?”

It’s the most common question I am asked when people find out I’m an economist, just ahead of “What stocks should I buy?” The answer to the latter question is always “index funds”. The answer to the former is never definitive. Both answers do a great job of frustrating the person who asked the question.

There are a few old saws in economics, sayings that have been repeated countless times. One of my favorites is from Harry S. Truman, who asked for “a one-handed economist. All my economists say ‘on the one hand . . .’, then ‘but on the other . . .’” Another, unattributed saying is that “economists have correctly predicted nine of the last five recessions.” Both these sayings are funny. But they also hold a grain of truth. There are always economists, and non-economists, predicting recessions.

Every time the Dow stock market index reaches a new 1,000-point high, some pundit somewhere is writing an article about how this is the peak, and a recession is right around the corner. There are always Cassandras, people warning about the perils of the future. And guess what? Sometimes these Cassandras are right. After all, a broken clock is right twice a day. When a recession happens there will always be someone who “predicted” it. Never mind the legions of people who also predicted recessions incorrectly in the years leading up to it.

To that end, do not trust someone who says there will definitely be a recession. This is either someone who doesn’t know what they are talking about or someone who does but wants to generate clicks or make a splash in the media. Having economists who are always hedging and referring to one hand versus the other may be frustrating, but it’s also the truth. The American economy is made up of millions of businesses and hundreds of millions of consumers. It’s influenced by global trade patterns, oil prices, Washington politics, the freaking weather, consumer sentiment, and hundreds of other variables. Predicting the economy is like predicting the butterfly effect. It’s impossible to do so with certainty. If the 2014 polar vortex can cause a significant drop in GDP, how can anyone hope to predict the future of an economy? Being an economist is like being a weatherman, except you get paid more. Just like giant predicted snowstorms turn out to be little more than flurries, predicted recessions often prove to be mere ripples in the world economy.

One of the reasons I became an economist in the first place is because of an incorrectly predicted recession. In Spring 2008, Steven Levitt gave a talk at the University of Illinois, where I was a student. This was when Levitt’s book, Freakonomics, was the talk of the intellectual community. Hundreds of people were in attendance. During the Q&A I asked him whether the current sub-prime mortgage crisis would cause a recession. He said it would not and gave a substantial rationale for why. With the benefit of hindsight, this is a spectacularly wrong prediction. The subprime mortgage crisis not only led to a recession but to the greatest recession since the Great Depression.

How did he get it so wrong? First, this was an unscreened Q&A. Off-the-cuff remarks are not official pronouncements, and they should be taken with a grain of salt. Second, I’ve since learned that Levitt was flipping houses as a side gig at the time, so his exposure to a housing market crisis may have clouded his objectivity. Levitt is a great economist who helped open the field to the type of research I do today. But in this case he could not have been further from the truth.

If a recession is impossible to predict with certainty, are there any harbingers of one? Yes. Lack of certainty aside, there are several indicators that might help predict recessions. These are far from perfect and some are still disputed entirely, but they are worth considering. The first is the CBOE Volatility Index (VIX). The VIX tracks how large of swings are occurring in the S&P 500 stock market index. The higher the volatility, the more nervous investors are. More nervous investors mean a recession is (theoretically) more likely. The second is the presence of an inverted yield curve for bonds. The details of an inverted yield curve are beyond the scope of this post, but when one occurs it often presages a recession.

Then there are measures that indicate a recession is either about to start or is already occurring. The clearest of these are the unemployment rate and GDP. If the unemployment rate is increasing rapidly or GDP is decreasing rapidly the economy is likely already in a recession. In fact, the general-but-technically-incorrect definition of a recession is two consecutive quarters of GDP decline, so after one quarter of GDP decline the economy is halfway to a recession. Of course, there are many instances where unemployment rises or GDP drops for one quarter and a recession doesn’t occur, so even these are not sure signs of an impending crisis.

In short, recessions cannot be predicted. There are too many unknowns. Crucially, recessions are somewhat of a self-fulfilling prophecy. If a majority of consumers think a recession is coming so they cut back on spending, that decrease in consumption will then cause a recession! Then, when a majority of consumers think a recession is over and start spending money again, the recession ends. That’s simple enough, but its impossible to say how hundreds of millions of people are going to behave.

One thing I’d like to see, just once, is a journalist get past the cheap talk and have a recession predictor put their money where their mouth is. It wouldn’t be that hard. If someone confidently predicts a recession with certainty, simply ask, “You say there will definitely be a recession in the next year. This being the case, have you liquidated all of your stock holdings and put every penny into shorting the market?” If the answer is anything but yes, that tells you everything you need to know.