BETA
This is a BETA experience. You may opt-out by clicking here

More From Forbes

Edit Story

Leading Economic Indicators Are Misleading

Following
This article is more than 4 years old.

“What’s a reliable leading indicator of recession that I can easily use in running my business?” I’m often asked that question and seldom give a satisfactory answer. In truth, there are none that are reliable, though a few come close. Understanding the statistical problem—in common-sense terms, of course—helps business leaders see their challenge more clearly.

A leading indicator indicates where the economy is going before it goes there. It’s not like a leading lady of the theater, the most popular of all actresses. In economics, a leading indicator resembles the dancer who leads his partner with gentle pressure to let her know where the pair are headed.

Some indicators attempt to tell us where the economy currently is. These are called coincident indicators. For example, the Wall Street Journal recently reported on a new measure that gives an alert when the economy actually goes into recession. That’s more useful than it appears, because time lags in economic data delay announcements of a recession. By the time we’re sure, companies have been losing sales for months.

Finding a leading indicator is straightforward. Gather a bunch of economic statistics and see which ones turn down before the economy turns down. Most of our data cover the period after World War II, so we are comparing economic statistics to the business cycles of this era.

One significant problem with this approach is that we’re only looking at 11 recessions. That is a pretty small sample. I routinely look at about 200 statistical series, but I can create more by calculating 12-month changes, 3-month changes, ratios of one series to another, and other arithmetic operations. We can consider a thousand or more candidates for leading indicator status. Econometrics professors often say that if we torture the data long enough, they will confess to something. We may find some series or combination that reliably works on the last 11 recessions. That does not mean the series will work on the next one.

The United States economy has had more recessions than those 11 that occurred since 1945, but a lot has changed in the economy over the years. Go far enough back and we were primarily an agricultural economy, with transportation by wagon or barge. Then manufacturing grew, along with railroads and heavy industry. Then services expanded, and the computer revolution, and the Internet. How useful will analysis of long-ago recessions be when we’re wondering where the economy will go in 2020?

A good example of underlying change in the economy casting doubt on leading indicators is the yield curve, the spread between long-term interest rates and short-term rates. This has been one of the best leading indicators, often calculated as the difference between interest rates on 10-year treasury bonds and 2-year treasury bonds (though sometimes other maturities are used.) When short-term rates are higher than long-term rates, we say the yield curve is inverted. But some economists point to a trend over the past decade of long rates having a smaller premium over short rates. It used to take a big change in short- and long-term interest rates to get a yield inversion, but now a small move in rates will trigger the inversion. In addition, critics of this leading indicator point to globalization as making one country’s long-term rates less connected to that country’s economy and more connected to worldwide financial markets. Economists joke about the danger of saying, “This time it’s different,” but we also know it’s dangerous to say, “Nothing has changed in the economy or financial markets.”

The most popular index of leading indicators for the U.S. is calculated by the Conference Board. Their work is based on a methodology developed by the U.S. Bureau of Economic Analysis, though they have modified the details. Underlying data and updates are available by subscription.

The Conference Board’s current list of leading indictors is shown below, with links to underlying data for those series that are produced by government agencies and thus in the public domain.

Average weekly hours, manufacturing

Average weekly initial claims for unemployment insurance

Manufacturers’ new orders, consumer goods and materials

ISM® Index of New Orders

Manufacturers' new orders, nondefense capital goods excluding aircraft orders

Building permits, new private housing units

Stock prices, 500 common stocks

Leading Credit Index™

Interest rate spread, 10-year Treasury bonds less federal funds

Average consumer expectations for business conditions


In general, the overall index of leading indicators does a better job than any one indicator. That said, some are better than others. My favorites are the interest rate spread and initial claims for unemployment insurance.

I have argued in other articles that professional economic forecasters have failed to predict most past recessions. We economists look at the leading indicators, as well as a lot more data, analyzed with complex models. It’s hard to believe that one or two simple indicators would do a better job. As a result, business leaders cannot expect to find a simple, foolproof way to know when the next recession is coming before it gets here. The best companies maintain flexibility and keep a close eye on their sales and pipeline.

Follow me on Twitter or LinkedInCheck out my website