WASHINGTON (Thomson Financial) - The new jobless claims number due out Thursday morning is as likely to raise as it is to reduce the confusion about what is going on in the US labour market.
The expectation is another 316,000 new first-time claims for unemployment insurance. But Ian Shepherdson of High Frequency Economics says "this is the first week of the annual auto retooling shutdowns and anything could happen".
For months now, claims and payroll reports have not behaved the way economists think they should.
As San Francisco Fed president Janet Yellen puts it: "Why has the labour market continued to be so strong, even while economic activity has moderated?"
Add to that a second question, how much of an inflation threat is the anomaly, and you have the Fed's latest economic conundrum. Is the labour market really as strong as the numbers suggest or is there something wrong with the numbers?
The most glaring gap is in housing. Over the past four quarters, real residential investment fell more than 16 pct while residential construction employment fell less than 4 pct.
Many economists are leaning toward the "wrong numbers" theory, with several explanations of why the labour market may be weaker than it appears.
First, there may be lags by employers in adjusting to new conditions. Yellen calls this a "benign explanation", with big declines in construction employment still to come.
Second, some analysts propose an "illegal worker" theory in which the relatively high number of construction workers who are in the US illegally can't file for unemployment insurance when they are laid off.
Economist James Hamilton who writes the Econobrowser.com blog, says that while "that claim is undoubtedly true, it's not correct to assume that these workers are not included in the [Labor Department] data." He says the IRS has issued millions of taxpayer identification numbers to people without a Social Security number and "government counts of the number of people working may be more accurate than one might have supposed".
Third, and growing more popular, is the "birth/death" theory. The Labor Department's so-called birth/death adjustment tries to estimate how many jobs are created and lost in the continual "birth" and "death" of small businesses that do not appear in the employer surveys.
Paul Kasriel at Northern Trust thinks the adjustment creates "phantom" workers, thus inflating the Labor Department's payroll figures. "In the 12 months ended June, the birth/death adjustment accounted for 57.3 pct of the 12-month increase in non-farm payrolls," he says.
That is 1.11 million of the 1.98 million total non-seasonally-adjusted payroll increase.
Daiwa Securities economist Mike Moran is sceptical because of the methodological problems. "While the effect cannot be isolated, the adjustment is probably not severe enough to lead to misguided assessments of the labour market and the economy," he says.
So, should the Fed worry about "high resource utilization", its term for tight labour markets, posing an inflation risk? Economists who think the labour market is weaker than the numbers suggest point out that the low unemployment rate has been accompanied by slowing growth in average hourly earnings.
However, HVB economist Harm Bandholz thinks "the reason for the slowdown in hourly earnings is the end of job shifts toward higher paid occupations". This does not change "true" inflation pressure, he says, "in fact, the 'true' labour cost pressure continues to point north."
dennis.moore@thomson.com
dem/wash/wj
