The Shakeup In Economic Statistics
In the arcane world of government statistics, major changes are about to take place that could deeply affect Corporate America. Revisions in the way the Bureau of Economic Analysis computes real gross domestic product may change not only Federal Reserve monetary policy, but company investment decisions as well.
On the face of it, the new statistics, which are due out in December, are distressing. Both economic growth and productivity are portrayed as much lower than previously estimated. The numbers will show that the U.S. economy did not grow at a 3.1% annual rate during the past four years but only at 2.6%, making it one of the weakest expansions on record.
It gets worse. Productivity gains disappear. Instead of rising at a 1.9% yearly rate in the 1990s, nearly double that of the '70s and '80s, the revised statistics will show productivity going up only 1% annually, no more than in previous decades.
Finally, the government will deemphasize the dollar figure for GDP, replacing it with an index. So the first-quarter GDP of $5,480 billion in 1987 dollars, will appear as 118.3, meaning that it is 18.3% above its index value of 100 in 1987.
If the GDP and productivity numbers are right, the policy implications are tremendous. The U.S. is a lot less competitive internationally, inflation is much more of a threat domestically, wages have little chance of rising, etc.
But don't panic just yet. GDP and productivity growth are being revised downward precisely because BEA is changing some component numbers it has learned to measure more precisely. The government is still awaiting better measurements for other components of the Information Revolution that have raised efficiencies, lowered prices, generated innovations, and improved quality. Business purchases of software ($51 billion in 1994), for example, are still counted with paper clips and other office supplies instead of being included under investment. While BEA does adjust for price and quality changes in computers in its statistics, it doesn't for telecom gear or for semiconductors. Its inflation indexes are missing the huge price drops in high-tech goods for consumers and businesses, therefore understating GDP.
So stay tuned for more revisions. Once the stat folks figure out how to measure the cool stuff of the Information Revolution, the GDP recount will go up, and with it, productivity. Maybe some private econometric outfits can come up with this before BEA. In the meantime, we hope BEA continues to report GDP in dollars, not in some abstract index, if only to let us see how the U.S. stacks up against Japan and the rest of the world.