The Perils of Economic Revisionism

How can you tell where you’re going, if you don’t know where you’ve been? This is the problem confronting economists with increasing frequency these days because of Washington’s continuous revisions of its statistics.

To be sure, revisions are a fact of life--and have been practically since the government began issuing data. However, those changes rarely were major and did not alter the direction of the figures as often as they seem to be doing this year.

What makes these revisions so troublesome is that the U.S. economy is at a particularly vulnerable point now. It’s been nearly seven years since the last recession. That’s 2 1/2 times longer than the average peacetime expansion.

This long a stretch without a downturn has made economists especially sensitive to any numbers that appear to point toward a cessation of economic growth. Needless to say, policy-makers, executives and investors also have a lot riding on the ability to quickly discern a change in the business climate.

Policy-makers, in particular, need to have up-to-date--but reliable--information. For example, a couple of months ago it looked as though the economy were slowing quite abruptly. Practically all the numbers spewing from the government’s statistical mills depicted an economy that not only was slowing down--but also, in many cases, actually contracting.

What made this scenario particularly ominous was that it followed a rather rapid and palpable tightening of money by the Federal Reserve.

In little more than a year’s time, short-term interest rates shot up by more than three percentage points, putting them above long-term rates for the first time since the last recession. Not surprisingly, the money supply stopped growing, while total bank reserves contracted by the steepest rate in postwar history.

Needless to say, when confronted with these sudden and dramatic signs of slowing, the Federal Reserve changed course and eased credit. Amid warnings by Fed Chairman Alan Greenspan that “such softness conceivably could cumulate and deepen, resulting in a substantial downturn in activity,” interest rates declined, sending the stock market to a new record high almost exactly two years to the day that it had reached its previous peak.

Suddenly, Washington said “never mind.” Numbers that were originally reported as up a little were changed to up a lot--while others were even revised from down to up. As a consequence, the government’s first estimate of growth in the nation’s second-quarter gross national product was subsequently revised by nearly 50%, from an annual rate of 1.7% to 2.5%.

The financial markets, never all that much in agreement on the outlook since Wall Street’s nervous breakdown two years ago, became even more confused. Stock prices have been bouncing around since early September, while bonds and the dollar have been trendless as well.

Data Needed for Planning

This discord over the outlook could not have come at a worse time for business. Fall is the season when plans for the coming year are finalized and budgets are set.

Most companies must take into account the state of the economy--even if they sell such necessities as foods, or small-ticket items such as cosmetics, toiletries and tobacco.

While demand for these items may not vary in a downturn as much as sales of such big-ticket products as autos and houses, costs of production might--not to mention financing charges and how fast producers get paid.

Washington’s bean counters are not without a plethora of explanations for these seemingly unusual revisions to their statistics. Their first excuse is federal budget cutbacks. Government agencies have had to shrink staff, they say, and cannot pay enough to attract top-flight economists from the private sector to analyze the data.

For its part, the private sector must share some of the blame, as well. The wave of corporate takeovers, mergers and buyouts of the past few years has also produced staff cutbacks and has obsessed surviving managers with the need to cut costs. Filling out government surveys is usually the first activity to go because these questionnaires are voluntary. This obviously makes Washington’s statistics less reliable.

Deregulation and technological change have also contributed to data deficiencies. Industry lines have become blurred, regulatory agencies now receive less data and the government has yet to determine how to adequately measure many aspects of the increasingly important service sector.

For all their shortcomings, however, Washington’s data remains the only game in town when it comes to analyzing the state of business and predicting its future. So when today’s statistics are revised--as they inevitably will be--economists will no doubt change their forecasts once again. After all, the first thing that they learned in Econ 101 is: If you have to forecast, do it frequently!