ADP Employment Report Still Weak

According to the ADP employment report, payrolls declined 203,000 in October compared to a decline of 227,000 in September. The consensus predicted payrolls to decline by 198,000 for the month which was in-line with the actual report.

While the pace of payroll declines has been steady over the last few months, an overall drop of over 200,000 jobs is still very depressing.

In order for unemployment to decline, payrolls need to increase by a little over 100,000 a month. Given that a 100,000 a month gain is still 300,000 away from October's report, it seems very likely that the unemployment rate has not peaked.

Over the last two recessions we've seen "jobless" recoveries where the peak unemployment comes well after the overall turnaround in the U.S. economy. We don't expect the unemployment rate to peak until sometime in the second half of 2010.

Details of the ADP report look a little promising in the near term but the future may not be so bright. Small businesses, defined as firms with less than 50 workers, saw their lowest level of declines (-75,000 workers) since July 2008. Medium-size businesses also shed 75,000 workers. Unfortunately, payrolls at firms of this size may come under attack due to lack of available funding. The bankruptcy of CIT wiped away a source of funds for many firms and could force small to medium size firms with volatile cash flows to layoff workers until the flows become more stable.

Construction employment continues to be under siege as payrolls declined 51,000 and brought the total decline in construction jobs to 1.675 million workers. Manufacturing payrolls declined by 65,000 workers.

The ADP report was originally designed as a leading indicator of the Bureau of Labor Statistics national unemployment report. The data from the ADP report and the BLS report used to be highly correlated, but during the latest recession the numbers became decoupled. In early 2009 the ADP report was reintroduced as a statistical estimator of the final BLS unemployment report. As a result, the ADP does a better job of projecting the final BLS payroll data than the original data.