Turn, Turn, Turn: Predicting Turning Points in Economic Activity

strength and duration, that have, so far, characterized economic activity. January 31, 2001: Preliminary real GDP growth for the fourth quarter is barely above 1 percent, more than 4 percentage points below what is was only two quarters before and about 7 percentage points below its level in the third quarter of 1999. Much of the press, and some forecasters, predict that the first quarter of 2001 may be the beginning of a recession. Whether this is the case or not is on the minds of many policymakers while this article is being written. The current state of the economy, not to mention the stock market, is certainly a far cry from what it was a few months ago. The recent gyrations in the economy and in the stock market remind us that the business cycle may not be dead—yet. They also remind us that economic conditions may change fast and somewhat unpredictably. This article focuses on providing some evidence on econometric models' ability to forecast these sudden changes in the business cycle, also called turning points. A model that can correctly predict turning points would clearly be useful to the business community and the general public. Investment decisions are made with an eye toward future economic conditions. The clearer the crystal ball, the wiser the decision. Policymakers would also benefit from the ability to forecast turning points. As late as May 2000 the Fed raised interest rates by 50 basis points to 6.5 percent, the last of a sequence of federal funds rate increases, totaling 1.75 percent, that started in June 1999. The increase in target rates at that time was justified by the strength of the economy and the dangers posed by a potential comeback in inflation. 1 Without in any way implying that such a policy move has " caused " the current slowdown in activity, one could reasonably argue that policymakers might have behaved differently then had they known what was to come. These suppositions bring us to the main question of this article: How good is the state of the art in turning point forecasting?