The Leading Indicators report is, for the most part, a compendium of previously announced economic indicators: new orders, jobless claims, money supply, average workweek, building permits, and stock prices. Therefore, the report is extremely predictable and of very little interest to the market. Though this series does have some predictive qualities, it is a common criticism that it has predicted "nine of the last six" recessions.
The Commerce Department previously published the leading indicators series. The collection and publishing of these data is now done by the non-profit Conference Board, which also produces the Consumer Confidence index.In Depth
The purpose of the leading index is straightforward: It is designed to signal turning points in the business cycle.
The index of leading indicators includes the ten economic statistics listed below.
- The interest rate spread between 10-year Treasury notes and the federal funds rate.
- The inflation-adjusted, M2 measure of the money supply.
- The average manufacturing workweek.
- Manufacturers' new orders for consumer goods and materials.
- The S&P 500 measure of stock prices.
- The vendor performance component of the NAPM index.
- The average level of weekly initial claims for unemployment insurance.
- Building permits.
- The University of Michigan index of consumer expectations.
- Manufacturers' new orders for nondefense capital goods.
The Conference Board, the organization that produces the leading index, standardizes these variables according to their individual weights in order to construct a composite leading index. Note that we have listed the components in order of importance. The difference between 10-year Treasuries and the fed funds rate carries the most weight; historically, this approximation of the slope of the yield curve has proven relatively more successful than other components at predicting future economic activity. Along those same lines, orders for nondefense capital goods carry the smallest weight because they have typically proven relatively poorer at pointing to changes in the direction of economic growth at large.
The leading index receives plenty of criticism. Indeed, skeptics often joke that it has correctly signalled nine of the last six recessions. Meanwhile, in its literature, The Conference Board cites the lead times with which the leading index has correctly predicted economic downturns. It is thus fair to ask whether the leading index is useless or priceless.
The answer lies somewhere in between. The charge that the index predicts recessions that do not come to fruition--and fails to warn of those that do--is hardly a fair criticism. No forecaster, even armed with an arsenal of economic statistics, has a perfect track record when it comes to predicting recessions. It is therefore unreasonable to assume that a ten-component index can do any better. That said, the index does have some reliability problems. For example, it failed to turn down prior to the 1990-91 recession, and in 1995 it signalled a downturn that never came to pass.
The leading index is more useful now that The Conference Board has taken control of it (the Department of Commerce stopped producing it at the end of 1996). Conference Board researchers quickly scrapped two of the old components--the change in sensitive materials prices and unfilled orders for durable goods--and added the interest-rate spread that appears in our list above. The index now lacks a wholesale price term, which some see as critical to determining future demand and inflation trends, but on net the new index emits less pronounced false signals and does a better job than it used to.
Briefing finds the leading index most helpful when we can make a statement
like this: The leading index has decreased only once during the past year. Of
course, even a strong trend like that does not guarantee that a recession will
not form over the coming six to nine months. But we can get additional help from
looking at the leading index with the coincident index, which is also published
by The Conference Board, and alongside a couple of other leading indices
published by Columbia University. Indeed, there exists much research that deals
with the criteria for determining recession warnings (i.e., the leading index
must fall during four of seven months and the coincident index must fall for
three straight months).