Count 'Em Up, Folks: New Output Numbers

EENIE meenie minie moe. Catch a tiger by the toe.

Economists and others can chant this rhyme about making a choice when the Bureau of Economic Analysis (BEA) next week again estimates the growth of national output in the United States during the spring quarter. The federal agency will present the change in the real output of goods and services in two ways.

One is the familiar fixed-weight real-dollar gross domestic product. The other is a new measure using chain-weighted indices of real GDP and of inflation for the entire economy.

In a way, this is merely a technical change, just a curiosity to the nonexpert. But economists and statisticians welcome the shift. ''It gives a more accurate measure of what is going on in the economy at any time,'' says Suzanne Rizzo, an economist with HSBC Securities Inc., a New York brokerage firm.

In that second quarter, for instance, the first estimate shows the economy grew at a modest 0.5 percent annual rate after inflation if the old method is used. Using the new method, the real GDP declined at a 0.2 percent annual rate.

Next Wednesday's GDP revision may change those numbers. But they show that Federal Reserve policymakers were counting on a revival in the pace of the economy this quarter when they decided Tuesday not to lower interest rates. At their previous meeting in July they cut short-term interest rates 0.25 percent.

Most economists agree with the Fed. The consensus forecast of 50 economists surveyed by Blue Chip Economic Indicators is that real GDP (old measure) will grow 2.9 percent this year, picking up speed in the second half.

Ms. Rizzo, however, talks of a ''comparatively weak'' third quarter. Using the old measure, she figures growth will run at a 1 percent annual rate, or with the new measure, close to zero.

A popular rough measure says two quarters of a downturn in national output count as a recession. But the group of private economists that determine the official turning points in the business cycle look at far more statistical series than GDP alone. If this National Bureau of Economic Research group should eventually proclaim a recession, it could be politically embarrassing for the Clinton administration. Commerce Department statisticians plan to switch to the new measure when they make a comprehensive, or benchmark, revision of GDP data in December.

Using prerevision numbers, the chain-weighted measure shows more modest growth in real output in the current economic expansion that started in 1991. Growth is overstated in the old measure by 0.5 percentage points a year on average - a big change when the average growth in those years was a little under 3 percent. Difficulty in measuring the impact of the rapid drop in prices and sales boom in computers accounts for most of the overstatement. By contrast, the new measure indicates that real GDP growth in the five economic expansions between 1960 and 1990 was understated by an average of 0.5 percentage points.

Changing the measure doesn't change economic reality. But the new measure may explain why the recovery felt spineless to many Americans until it became more vigorous in 1994.

GDP is a sum of the economy's output of goods and services - cars and haircuts and thousands of other items - measured in dollars. This involves statistically weighing each item according to prices. But the relative prices of these items change over the years, stimulating changes in consumption and output of these items - what economists call a ''substitution'' effect. The BEA's old methodology uses prices from a single base year (currently 1987) to weigh production in each year or quarter. The new measure employs a set of moving-average (or ''chained'') price weights. Each year's real GDP estimates reflect annual price data for the current year and the prior year.

The new system will eliminate the dramatic revisions in the GDP data every five years or so when the BEA updates its real GDP base year for prices, Rizzo notes. But there will still be period revisions as new data about output becomes available. And, she maintains, the new GDP measure comes closer to reflecting the price structure that actually prevailed when each period's production took place.

Better statistics, economists and policymakers hope, will make their forecasts and policy decisions marginally better.