I believe in nonlinear dynamics. [Laughter] I think I have even experienced them, and probably you have as well on occasion, in terms of the difficulty that we have in forecasting recessions. Our forecast isn’t just some sort of “push a button on a linear model and here is the result.” But I do think the current situation illustrates to me why it is, in fact, so hard for us and why we don’t forecast recessions very often. As I indicated, I think there is a configuration of a number of indicators that make it easy for me to imagine you looking back on next June and saying, “Indeed, what you saw back there in December—in terms of the jump in the unemployment rate, the drop in the manufacturing ISM, and the uptick in initial claims— were all precursors of a recession.” The point of my remarks is that I am pretty darn worried about that possibility. But it is hard at this point to make that call—we are coming off the data in the fourth quarter, which have exceeded our expectations considerably.
As I noted, not all the things that you might expect to be highly sensitive to a business cycle downturn, such as motor vehicle sales, have moved in that direction. I know this is unfortunate, and we really cannot be as helpful to you as I would like. We are saying, in effect, “Yes, we’ll call the recession when we see the weak spending data.” But the weak spending data will already be lagged one or two months, and that is the reason that we will be looking back, if we’re lucky to be able to do it in March, saying, “The spending data indicate that a recession may have started in December.” The current forecast is our best judgment. But I think it wouldn’t take much more in the way of negative news for us at this point to regime-shift, as you said, into recession mode. We are not quite there yet, but we are certainly worried about that possibility at this point.