Pay no attention to weekly jobless claims, ignore the Manpower ( MAN) survey and don't bother scrutinizing layoff reports. When it comes to gauging interest rate moves, the only measure of employment that can offer any real insight is the monthly payroll survey. That was the message the Federal Reserve gave to investors Tuesday, as it left interest rates unchanged at a 46-year low of 1%. While nonfarm payrolls have been the mother of all employment statistics for a long time, several economists have recently complained that these data aren't painting an accurate picture of the job market. The payroll survey misses the formation of new businesses and is out of sync with other, more encouraging data, these pundits argue. "There are very good reasons to believe that the job market is better than the payroll data suggest," said Brian Wesbury, chief economist at Griffin, Kubik, Stephens & Thompson. In its policy statement at the end of January, the Fed seemed to be leaning toward agreement, saying that while hiring remains subdued, "other indicators suggest an improvement in the labor market." The jobless rate, for example, has fallen to 5.6% from 6.3% last June and unemployment claims have remained under 400,000 for several months. On Tuesday, however, the Fed shifted its focus back onto payrolls, saying that while companies aren't shedding jobs like they used to, "new hiring has lagged." "This brings people back to reality, the Fed is focused on the payroll survey," said Drew Matus, an economist at Lehman Brothers. "While the Manpower survey might be nice and the drop in claims might be nice, what they really want to see is hiring, and they're not seeing that." Given that payrolls have been so weak -- just 21,000 new jobs were created in February -- a rate hike doesn't appear to be in the cards this summer, and might not even materialize until sometime in 2005.