Can it be true? The economy actually grew 0.2% in the final quarter of 2001. That means the recession that officially began in March still hasn't produced two quarterly declines in real gross domestic product. So is it a recession or isn't it? And what does this surprising report mean for the outlook?
Preliminary GDP data are always subject to revision, but details of the Commerce Dept.'s Jan. 30 report make two key points that are unlikely to be revised away. First, if the U.S. was in recession, it's probably over and a gradual recovery has already begun. Second, the National Bureau of Economic Research (NBER) may need to rethink its methodology for determining recessions and recoveries. Those guidelines haven't kept up with a rapidly changing economy where productivity creates the bulk of output gains and services make up the lion's share of GDP.
Another inescapable conclusion from the report: overall demand in the U.S. remains amazingly resilient, even after September 11. Consumer spending last quarter, lifted by a surge in car sales, rose at a 5.4% annual rate -- a remarkable showing, recession or not. And the strength continues in 2002. "Everything is in place for a pickup in the economy," says retail analyst Linda T. Kristiansen at UBS Warburg, who thinks sales will improve into the fall.
END OF A CYCLE.
Many businesses also appear to be taking a second look at capital-spending plans. Outlays for high-tech equipment rose last quarter for the first time in a year, and orders for computer equipment jumped at a 20% annual rate. The war against terrorism also temporarily boosted government outlays at all levels. That will continue, albeit more slowly.
The big plus for the economy is that all this extra spending generated a record drawdown of inventories. Businesses liquidated their stockpiles at an annual rate of $120 billion last quarter, as great as the pace of the first three quarters of 2001 put together. That left store shelves and dealers' lots so bare that businesses will have little choice but to ramp up their output this quarter.
Auto makers are "coming around to the conclusion that 'hey, sales are better than we expected -- inventories are too low, let's increase our [production] schedules a bit,'" says auto analyst David Healy at Burnham Securities Inc. As a result, most economists are scrambling to put a plus sign on their first-quarter GDP forecasts.
Forecasters at the Federal Reserve are expecting real GDP to grow 1% to 2% this quarter. The Fed left interest rates unchanged at its Jan. 30 policy meeting and hinted that this cycle of policy easing is now finished. The Fed's statement implies that policymakers believe the worst is over and that long-run prospects for productivity growth remain favorable. Indeed, the GDP data suggest that productivity increased at a 3% or so annual rate in the fourth quarter, a stunning performance during a recession.
So much for the good news (for more, see BW Online, "From the New Economy, a New Recession"). The bad news is that the fourth-quarter strength raises some fundamental questions about whether the U.S. is determining its business cycle correctly. True, the economy was hurting in 2001. More than 1.7 million people have lost their jobs since March, and manufacturing saw its worst slump since the Depression. But it now appears the NBER's Business Cycle Dating Committee -- the academics who declare the cycle's peaks and troughs -- was too arbitrary in saying the economy hit a recession in March.
That's no small problem. An official declaration of a recession can have unintended consequences. Indeed, President Bush made a point of referring to the recession when he pushed for his stimulus plan during his State of the Union speech. With the recovery already under way, the government risks overstimulating the economy. Worse, companies might delay ordering if they think the economy is in the tank. The danger is that poor business decisions will be made if swings in economic activity are overstated. Ultimately, those mistakes could build upon themselves and cause a mild slump to spiral into a severe downturn.
The high stakes of declaring a recession is why economists are careful to employ a strict definition that does not include GDP. A recession, they say, is a "significant decline in activity spread across the economy, lasting more than a few months, visible in industrial production, employment, real income, and wholesale-retail trade." By that metric, the economy may not have weakened severely or broadly enough to qualify. That was clear prior to the September 11 attacks, a point that the NBER itself has admitted. And now, the GDP report raises questions about the depth of the weakness after September 11.
So why did the NBER go all the way back to March to mark the recession's start? That's when payrolls peaked. The NBER's other three measures of economic activity don't support a March start date. Industrial production has been falling since October, 2000. Except for an attack-related drop in September, business sales have been about flat since late 2000. And real incomes are still growing solidly.
When all that data is blended together -- as it is in the index of coincident indicators, which tracks the economy's current performance -- they suggest the economy held its own until August. That's when the index began a four-month drop that was exacerbated by the September 11 attacks. The coincident index suggests the recession didn't start until the third quarter. Nevertheless, the NBER committee remains comfortable with its decision. "Nothing that's happened would cause us to think we've made a mistake," says committee member Jeffrey A. Frankel of Harvard University.
Yet the NBER's reliance on job data misses the fact that U.S. business activity is shaped by two new powerful trends: the push to boost output by lifting productivity and a greater use of temporary workers to adjust to swings in demand. The combination means that as demand growth fluctuates, payrolls will experience wider swings than in the past.
Right now economists expect first-half real GDP to grow by 2%, for example, which means the labor markets will be weak until late summer. If job growth is the yardstick, this recession could be the longest in the postwar era. Indeed, committee member Ben S. Bernanke of Princeton University agrees that if productivity is strong, output could rise even while employment falls. "The question would be: Is that a recession?" he says.
Labor's new flexibility has caused a disconnect between job growth and consumer demand. Companies can quickly lay off temps when demand slows. But these same businesses can give real raises to their remaining workers commensurate with faster productivity. This divergence was evident in 2001. Unemployment has risen by about two percentage points in the past year, but real wages are growing at the fastest pace since the 1960s. The high-productivity lift to wages explains why demand has stayed strong even in a slack labor market.
The NBER would benefit enormously from higher-quality data that better reflect the New Economy. But until that happens, at a minimum the NBER should recognize the structural shift in the labor markets by giving more weight to the real-income numbers. It ought to deemphasize U.S. factory data further. Manufacturing is only 16% of the economy. With global producers satisfying a greater share of U.S. demand, industrial production is too narrow a gauge for the overall economy's health.
Despite the questions raised about the recession, the fourth-quarter GDP report was excellent news. Even better, growth is likely to remain in positive territory for all of 2002.
By James C. Cooper and Kathleen Madigan in New York, with Rich Miller in Washington and bureau reports