A Different View Says Employment May Be Picking Up

(MoneyWatch)  Paul Kasriel is the insightful and outspoken chief economist at Northern Trust in Chicago. During this crisis and recession he has offered commonsense answers to complicated questions, and seems to have done so again with this week's analysis of unemployment, from a slightly different viewpoint. Rather than rely on macro unemployment statistics, which are extrapolated to the overall economy from small surveys, he looks at the bottom-up aggregation of state unemployment statistics, and concludes that new hiring may have already started, or that it may begin soon. I'm not so sure.

Consider the graph below, from Kasriel's newsletter of November 25. The red line, the four-week moving average of initial unemployment claims, peaked in April at 659,000 and has retreated to 479,000 as of November 21. Make no mistake, Kasriel says: plenty of people are still being fired.


But now look at the blue line, which shows the number of continuing unemployment claims, including those who are extended and emergency benefits. That number, Kasriel reckons, has also peaked, and even decreased slightly in the November 7 tally. He adds:

Again, nothing to brag about but something to be thankful for. The combination of a decline in the number of firings per week and a slight drop in the total number of unemployment insurance beneficiaries suggests either that hiring has commenced in a small way or that it is about to.

However, he told me, the initial hiring will not be strong enough to keep the overall rate of unemployment from climbing further.

The Bureau of Labor Statistics publishes a less well-known report, on hiring and firing behind the overall unemployment numbers, called JOLTS, for Job Openings and Labor Turnover Survey. It is released late -- the November 10 report covered activity in September -- but here's what it says:


The blue line shows the rate of hiring. It's low, and still below the rate of firing, but maybe it has bottomed out.

One more graph, the one that makes me unsure about Kasriel's forecast of an upturn. It compares the number of continuing unemployed people to the total non-farm payroll, going back to 1970. It reminds us that after the last two recessions, employment stalled for a while before getting back to a growth trajectory. Accompanying them was a sustained peak in continuing unemployment (in the black circles); prior recoveries had a sharp drop.


I exchanged follow-up e-mails with Paul K, and I may have read too much enthusiasm into his research note. He writes:

Even though I believe labor market conditions are improving..., the first year of this recovery is unlikely to yield a significant improvement in the labor market for three reasons. Firstly, because the financial market will not be able to create a "normal" amount of credit for the private sector because of capital constraints, real GDP will not grow fast enough to bring down the unemployment rate until the second half of 2010, and then, only marginally. Secondly, with the average workweek at only 33.0 hours, businesses can accommodate increased demand by extending the working hours of their existing staff rather than having to hire additional staff members. Thirdly, the U.S. economy is going through structural changes which mean that some jobs in some industries will come back only slowly, if at all. Industries undergoing structural changes for the worse are motor vehicles, residential real estate and financial services.