Why Some Indicators Lead and Others Lag — and What That Means

Why an Indicator Leads

• It measures a decision, not a result: Building permits, new orders, and hiring plans record commitments made now that will become output months from now.
• It is cheap to reverse: Employers cut overtime hours before cutting jobs, because hours can be restored instantly and a laid-off worker cannot.
• It reflects expectations: Stock prices and consumer expectation surveys aggregate forward-looking judgments about conditions that have not arrived yet.
• It reflects credit availability: Lending standards and interest rate spreads determine what firms and households will be able to spend later.
• It is fast to respond: Initial claims for unemployment insurance are filed within days of a layoff, long before that layoff appears in monthly payroll or quarterly output data.

Why an Indicator Lags

• It measures the accumulated consequence of something already over: The average duration of unemployment keeps rising after a recovery has begun, because the pool of long-term unemployed clears slowly.
• It is contractually sticky: Labor costs per unit of output, the prime lending rate, and posted service prices change only when contracts and policies are revisited.
• It is a stock adjusting to a flow: Inventory-to-sales ratios and outstanding loan balances take time to work down after demand turns.
• It responds to policy that responds to the cycle: Interest rates set in response to inflation follow the cycle rather than anticipating it.

Lagging indicators are not useless. Because they confirm what has happened, they are the natural check on a leading signal that may have been noise.

What the Three Composite Indexes Contain

• Leading Economic Index: The Conference Board's composite includes average weekly manufacturing hours, initial unemployment claims, manufacturers' new orders for consumer goods and materials and for nondefense capital goods excluding aircraft, an ISM new orders measure, building permits, stock prices, a leading credit index, the interest rate spread between long and short Treasury rates, and average consumer expectations for business conditions.
• Coincident Economic Index: Four series that move with the cycle — nonfarm payroll employment, personal income less transfer payments, industrial production, and manufacturing and trade sales.
• Lagging Economic Index: Series including the average duration of unemployment, the inventories-to-sales ratio, labor cost per unit of output, the average prime rate, commercial and industrial loans outstanding, the ratio of consumer installment credit to personal income, and the consumer price index for services.

How the Classification Is Used — and Misused

The honest way to describe these indicators is as a summary of historical timing regularities, not as a forecasting machine. A leading index turns down before recessions have started in the past, but it has also turned down without a recession following, and the lead time has varied widely from cycle to cycle — which means the signal cannot be read as a schedule. Composites exist precisely because single series are noisy, and analysts conventionally look for a decline that is sustained over several months, meaningful in size, and broad across components before treating it as significant. It also matters that indicators are revised: the numbers available in real time are not the numbers that appear in the history books, which is one reason the NBER's Business Cycle Dating Committee waits for revised data and typically announces turning points long after they occurred. The classification is a way of organizing evidence about where the economy has been, not a way of knowing where it is going.