The prominent economists who determine the dates when U.S. recessions begin and end have yet to agree on what seems to be a foregone conclusion on Wall Street: the recession is over.

The National Bureau of Economic Research's business cycle dating committee said on Monday it was too soon to say precisely when the recession ended. At least one of its members wasn't convinced it has, underscoring how fragile the recovery remains even though the economy probably just wrapped up its third straight quarter of growth.

NBER is notorious for waiting so long to announce recession dates that its decisions are usually a non-event for investors. Indeed, even Monday's cautious non-decision did not stop the Dow Jones industrial average from hitting an 18-month high.

The group said in a statement that although most economic indicators had turned up, the committee thought it premature to pinpoint the month that the recession officially ended.

U.S. economic data is subject to revisions, sometimes major ones that turn positive readings on gross domestic product or employment negative, making it hard for NBER to make a call.

Comments from committee members suggested the reluctance to declare the downturn over was based on more than revisions.

I think there's a risk -- less of a risk than a month ago, nevertheless a risk -- that this economy could turn back down again, committee member and Harvard University economist Martin Feldstein said on CNBC.

If it did that sometime soon, I don't think we'd want to call the increase that we've seen in the last six months a recovery. I think we'd want to say that that was just a temporary rise in what was otherwise a longer economic downturn, he said.

However, fellow Harvard professor and committee member Jeffrey Frankel declared on his blog last week the recession was over, pointing to the Labor Department's report that employment rose in March.

WHY THE DOUBT?

Most Wall Street economists think the downturn probably ended in June or perhaps a couple of months later. The economy resumed growth in the third quarter of 2009, although employment continued to contract for several more months. Payrolls have recorded only two monthly gains since the recession started -- in November 2009 and March 2010.

U.S. economic growth has been more robust than Europe's, which slackened at the end of 2009, heightening concerns that the recovery might falter. Meanwhile, China has led a powerful resurgence among major emerging economies, helping to lift global output.

The NBER defines a recession as a significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in real GDP, real income, employment, industrial production, and wholesale-retail sales.

That means it does not follow rules of thumb such as two consecutive quarters of GDP to signal a recession, nor does it declare a downturn over as soon as GDP turns positive.

A closer look at indicators the NBER watches helps explain its reticence. For instance, while real GDP has been growing since the middle of 2009, personal income flattened in February and industrial production growth has been inconsistent.

Harvard's Frankel said another problem for the NBER is that it must determine the month in which the recession ends, yet economic output is reported on a quarterly basis.

In response to a query from Reuters, he said the committee looks at both GDP and national income, a measure that gauges output based on income rather than spending. Gross domestic income did not turn positive until the fourth quarter of 2009, one period after GDP resumed growing.

The difference between GDP and GDI has been a hot topic of discussion among data watchers recently, particularly after a Federal Reserve economist released a paper suggesting that GDI gave a more accurate reading of the business cycle.

Of the data that is reported monthly, Frankel said labor market indicators were the most important. His favorite measure, total hours worked, troughed in October.

Total hours worked tends not to lag behind economic activity as much as employment does, because firms start lengthening the workweek of their existing workers before they start hiring new workers, Frankel said.

The NBER said its committee reaffirmed that the recession began in December 2007.

(Reporting by Emily Kaiser; Editing by Chizu Nomiyama and Andrew Hay)

China Unemployment