Initial Claims Fall - Analyst Blog

The economy got a Christmas present today in the form of a decline in initial claims for unemployment insurance of 28,000. We are now down to 452,000, almost 200,000 below the peak set back in April. The decline also reverses increases in the last two weeks. The four-week average (see graph from http://www.calculatedriskblog.com/ below), which is generally considered a better measure given the week-to-week volatility in the numbers, has been in a very steep downtrend for most of the year and so far is showing no sign of a plateau as it did following the last two recessions. It is acting much more like it did following recessions of an earlier era, when it would pretty much go straight up and then straight down.

However, keep in mind that initial claims only capture one side of the employment picture – the number of people losing their jobs – and it says nothing about the number of people finding jobs. Even in the best of economic times about 300,000 people lose their job every week. However in good economic times, nobody cares, since they can find another job and often at better pay, immediately. The employment numbers measure the net between job losses and new jobs. We probably need to get down to the 400,000 to 425,000 level to indicate that the economy is gaining jobs on balance. We are not there yet, but we are getting much closer. The four-week average is now down to 465,250, a decline of 2,750 over the last week.

We also saw some good news on the continuing claims front. Regular state continuing claims, which run out after 26 weeks, dropped by 127,000 to 5.076 million this week. That brings them to their lowest level since early February, and well off the peak of 6.904 million set in June. However one of the biggest problems in this recession is that the duration of unemployment is unusually long. Once people lose their jobs, they have a much more difficult time finding a new job than they did in prior economic downturns.

A key piece of evidence for this is the ratio of long-term unemployed (out of work more than 26 weeks) to short-term unemployed (out of work less than 5 weeks). Prior to this downturn, it had never risen above 0.72 (late in the 1982-83 recession). As of November, it was 2.10. On average, someone who is out of work has been out of work for more than 26 weeks. OK, the mean is a poor measure since you cannot be unemployed for fewer than zero weeks, but even the median is at a record high, and 38% of the unemployed have been out of work for more than 26 weeks. Clearly, then, a measure that drops people after 26 weeks is not a very good one under these circumstances. After the state benefits run out, extended claims, which are paid by the Federal government mostly from ARRA funds (the stimulus package), kick in. Those now total 4.732 million (two largest programs combined). There was a little bit of good news there as extended claims fell by 2,900 on the week. The 2,900 drop is not that significant in and of itself, but it means that the 127,000 decline in regular continuing claims is real, and not just people having their benefits expire.

It is possible that the decline in initial claims might just be a case of businesses not wanting to look too much like Scrooge by laying off people just before the holidays. It will be important to continue watching the numbers as we move into the New Year. However, it looks like we will start to see job growth in 2010, which will help to solve a multitude of problems. With a job, people will be able to pay their mortgages and credit card bills. That would certainly be good news for firms like Bank of America (BAC) and J.P. Morgan (JPM). People will be able to go out to the mall again, helping out retailers like Macy’s (M) and J.C. Penney (JCP). With a job they have an income, and they will be paying income taxes (and not drawing unemployment benefits), which could actually start to make a serious dent in the deficit. That would make it a very Happy New Year.