Jobs: How Hard a Hit?

Hurricane Katrina no doubt slammed payrolls in September. But thanks to rebuilding, they'll probably bounce back in the months ahead

By Rick MacDonald

Just how wide a swath did Katrina and Rita cut through the U.S. labor market in September? Most job indicators imply a sizable hurricane distortion to the nonfarm payrolls component of the September employment report, scheduled for release on Oct. 7.

We at Action Economics expect a decline in the headline payroll figure of 75,000, which includes a 250,000 subtraction attributable to the hurricanes and a 25,000 decline related to the Boeing (BA ) machinists' strike, alongside a 200,000 underlying growth rate. The unemployment rate is likely to spike to 5.1%, the workweek to hold at 33.7 hours, and hourly earnings to rise 0.2%.


  The Bureau of Labor Statistics (BLS) has issued a press release detailing how disruptions related to Hurricane Katrina will be treated in the September survey. A large distortion appears likely for the payroll survey, but the household survey should experience a smaller impact.

In the months that follow major hurricanes, payrolls generally fall short of the underlying growth trend in the survey period that captures the storm, but then rebound in the months afterward. Interestingly, the workweek does not tend to be affected.

We think the Katrina-related distortion will be, by far, the largest in the history of any natural disaster. This will be due both to "lost" payrolls in the affected region and the spike in energy prices and drop in volumes that rippled through the national economy.


  The initial jobless claims data since Katrina are consistent with our payroll forecast of a large storm-related distortion. In looking at the hurricane effect in each of the past four weeks for initial claims, the aggregated impact is an estimated 260,000 -- right in line with our 250,000 estimate for the distortion to payrolls.

Other data imply downside risk for payrolls as well. Both the University of Michigan's Consumer Sentiment and the Conference Board's Consumer Confidence surveys revealed a hefty drop on the month. In fact, the drop in the current series of the Michigan report is consistent with a payroll decline of 100,000 to 200,000.

The most recent Help-Wanted index dove to 35 in August, from 39, which leaves the figure declining 5.5% year-over-year and leaves the series at a 44-year low. But the monthly decline is not even noticeable on a long-term historical basis.


  One of the bigger September surprises was the strength in the Institute for Supply Management survey and its employment component. The September employment component rose to 53.1, from 52.6, which implies limited downside risk for both factory payrolls and the related temp-help aggregate.

We assume that this "sentiment" survey is capturing the "rebuilding" effect that we believe will dominate the payroll survey in the following month, rather than the disruption that was likely more prevalent through the BLS survey week of September.

Overall, payroll growth will clearly take a huge hit in September, thanks to Katrina. While our forecast is at -75,000, we would not be too surprised to see a drop greater than 200,000, given the magnitude of this event.


  But despite the enormous uncertainty surrounding the number, it's unlikely that the market will get much mileage out of the report. Any surprise will be quickly written off as a temporary distortion.

The market does not have much choice but to wait and see to what degree the rebuilding efforts over the coming months take hold. Given the solid tone in the economic data before the hurricanes struck, as well as the large reduction in inventories evident before Katrina, the outlook is arguably as strong as it has been all year -- no matter what the September payroll report reveals.

MacDonald is global director of investment research and analysis for Action Economics

Before it's here, it's on the Bloomberg Terminal.