There is no doubt that economic forecasting is a tough business. Aside from the fact that a "black swan" event can undermine even the most analytically sound of predictions, the process is invariably tainted by mistakes, bad data, noise, inherent biases, and other influences that can make it hard to see the forest for the trees. What's more, external pressures -- such as an employer who doesn't like to see "bad news" -- can force those who are putting together the estimates to shade them one way or another.
Even so, it is interesting to see just how far off the so-called experts were in their mid-December predictions about how U.S. Gross Domestic Product would turn out in the first quarter of the year.
Back in early January, in a post entitled "Cognitive Dissonance," I noted that "according to the Wall Street Journal's Semiannual Economic Forecasting Survey of 60 leading economists, only one, James F. Smith of Western Carolina University and Parsec Financial Management, [expected] to see a recession -- two or more quarters of declining gross domestic product -- in the U.S. this year....That [was] despite the fact that three separate indicators [had] emitted strong warnings of an impending economic downturn."
While the jury is still out on whether growth will turn negative during 2007, it is definitely seeing a major slowdown, as yesterday's report of a weaker-than-expected gain of 0.6 percent gain in first quarter GDP makes clear. Yet only weeks before the year actually started, the professionals' consensus estimate was for an increase of 2.2 percent, with 58 out of 60 of those polled predicting a rise of one percent or more. The closest to the mark was Susan M. Sterne of Economic Analysis, with a forecast of 0.8 percent.
Given all the negative economic influences that were known at the time, not least of which was the rapidly deflating housing bubble, the obvious question is why so many of the forecasters' estimates ended up on the high side of the actual data. A charitable answer might be that it is difficult to get a handle on an economy in transition. Alternatively, it could mean that many of these paid professionals are deluding themselves or others.
I guess we'll know better when we compare the next set of estimates, due in about a month's time, to forthcoming reports.