Edgar Fiedler, Assistant Secretary of the Treasury for Economic Policy under Nixon and Ford said, “If you have to forecast, forecast often.” Fiedler’s words are particularly relevant during volatile economic times.
Consider the case of two prominent Colorado forecasts. Both USA Today/Moody’s Colorado and the Leeds School of Business at the University of Colorado prepare composite and sector forecasts for the state.
The two employment forecasts are at opposite ends of the spectrum. Moody’s is most likely too high (2.6%) and CU is hopefully too low (.5%).
The most basic test for measuring forecast accuracy is to determine whether the forecast correctly predicts the direction of the forecast (is it positive or is it negative). CU projects 5 of the 14 sectors will post job losses, while Moody’s says 1 of 14 will shed jobs. Said differently, there is a difference in opinion about the direction of the sectors on these 6 sectors.
There is also great disparity in the magnitude of the forecasts. Projections for only two sectors are remotely similar, with a difference of less than 1% points.
Through Q1, the Office of Labor Market Information reports employment gains of 0.7%.
The country would have to experience a mild double dip to achieve the CU forecast. On the other hand, Colorado would have to add jobs at a rate of at least 3% for the remainder of the year to achieve the Moody’s forecast.
Expect a flurry of forecasts from CU, Moody’s, and others as they try to understand the strength of the recovery.
©Copyright 2011 by CBER.