By John Cranford, CQ Columnist

With all the talk that the United States economy is headed for a recession and that the downturn may even be imminent, it’s worth wondering how we will know that the bad times are really upon us.

The short answer — and the honest one — is that the economy is likely to be on the rebound before it’s plainly evident that it was actually in a hole.

Dow Jones Industrial Average: Click here to view chart

At the moment, the recession outlook seems mostly like the wishful thinking of financial market bears and other pessimists. The numbers just aren’t there yet. For instance, gross domestic product surged ahead in the third quarter at a 4.9 percent annual rate, the fastest pace in four years. But there’s no more backward-looking indicator than GDP, so that solid growth figure for July, August and September is no guarantee that the economy will keep growing. In fact, the underlying support for the expansion that began in November 2001 could evaporate almost overnight, and it might not show up in GDP statistics for a long time.

Although there is no better measurement than GDP of the fullness of U.S. economic activity, it takes months and sometimes years of revisions for the data to settle down to a clear pattern. We now know that the economy contracted in four out of five quarters during the latter part of 2000 and 2001 — the last time the economy was in recession — but only one of those contractions showed up in GDP reports during 2001.

So, don’t look there for any sort of advance clue of a serious problem.

For now, the signs of broad trouble in the economy are mostly circumstantial, but that doesn’t mean they should be ignored. For instance, day-to-day movements of the stock market are a lousy indicator of the state or even the direction of the economy. But it is worth observing that in November alone the Dow Jones industrial average jumped up or down by more than a full percentage point on 10 out of 21 trading days. That’s an extraordinary degree of volatility, and the serious skittishness of stock investors tells us that they are worried.

And apart from doomsday forecasts from some economic analysts, there are anecdotal signs of trouble in the comments of Federal Reserve officials who almost daily are revising their outlook for the economy and hinting — finally, some would say — that the subprime mortgage problem has morphed into a full-blown credit collapse. They aren’t yet saying outright that the crisis could sink growth in the coming year, but that’s the way investors and analysts read their body language, and that’s why financial markets are expecting further Fed interest rate cuts in December and well into 2008.

It’s a Date

If the United States does experience its third recession in a quarter-century (evidence that periods of expansion have been much longer than average in recent years), it will be the National Bureau of Economic Research, a respected Cambridge, Mass., organization that will make the unofficial — but widely accepted — call.

A group at NBER, the Business Cycle Dating Committee, has been analyzing just this subject for almost 30 years. It meets irregularly, doesn’t talk much about its activities (or its thinking) and periodically issues reports. One thing that is clear is that this committee doesn’t adhere to the conventional rule of thumb that the economy is in recession when GDP contracts for two straight quarters. Instead, it looks at multiple indicators and tries to divine when a “significant decline in economic activity spread across the economy, lasting more than a few months,” has occurred.

The group particularly monitors four measurements of the economy that it believes tend to mark turning points: the total number of payroll jobs, the dollar amount of monthly personal income (less so-called transfer payments, such as Social Security benefits), the dollar amount of monthly business sales at all types of enterprises and a gauge of industrial production. Most of these indicators are relatively esoteric, and none are generally topics for dinner-table conversation, but all have a track record for predicting the course of the economy.

And at the moment, at least, none are clearly pointing to the possibility of recession. One, industrial production, may have reached a peak in September. But a one-month decline in October hardly shows a trend. And the other three measures have been rising more or less steadily for months.

Of course, these statistics are also subject to revision. And that means those who monitor them closely will also be alert to the possibility that whatever signal they send might turn out to be false.

That’s why the cycle-dating committee tends to be cautious. In November 2001, the panel reported that a recession had begun in March, eight months earlier. It wasn’t until July 2003 that the committee determined that the recession had ended the same month that its start date was announced.

Of course, these sorts of determinations are mostly of historical interest. If the economy does drop off a cliff, ordinary Americans will feel it, even if it doesn’t yet have a name.