Initial Jobless Claims Plunge

Initial Claims for Unemployment Insurance fell by 34,000 last week to 407,000 (last week was also revised up by 2,000, so one could see it as a 32,000 decline). This was much better than the expected level of 442,000.

We can give thanks that we have be able to get out of the “trading range that initial claims have been in for the last year. Initial claims have been generally trending down since they hit a secondary peak of 504,000 (after revisions) on 8/14. The path has, however, been erratic. Since claims can be volatile from week to week, it is better to track the four-week moving average to get a better sense of the trend.

The Four-Week Moving Average

It fell by 7,500 to 436,000. After declining sharply in the second half of 2009, the four-week moving average has been stuck in a tight trading range. We seem to be stuck in a pseudo recovery, though if the current downward trend can be sustained, there is real hope.

The economy is growing, but not at the sort of rate needed to add a significant number of jobs and to put a dent in the huge army of the unemployed. In hindsight, the run-up to 500,000 seems to be mostly a function of the Census workers being laid off (they are almost all gone now, and the Census was completed quicker, and at less cost than anyone had expected). As that effect waned, we returned to the previous baseline.

Relative to a year ago, the four week moving average is down by 67,750 or 13.5%. The graph below (from http://www.calculatedriskblog.com/) charts the path of the four-week average since the start of the century. Clearly we are breaking out of the trading range to the downside. That is, after all, what makes a trading range a trading range. While I would like to see the number confirmed next week, and not reversed or revised away, it looks like we have something to be thankful for this Thanksgiving.

Continuing Claims

The data on regular continuing claims was also positive. Regular continuing claims for unemployment insurance fell by 142,000 to 4.182 million. They are down by 1.324 million or 24.0% from a year ago.

Regular claims are paid by the state governments, and run out after just 26 weeks. However, in October, half of all the unemployed had been out of work for 21.2 weeks (down from 25.5 weeks in June, but up from 20.4 weeks in September), and 41.8% had been out of work for more than 26 weeks. Just for a point of perspective, prior to the Great Recession, the highest the median duration of unemployment had ever reached was 12.3 weeks near the bottom of the 82-83 downturn.

A week from Friday we will find out if there has been any improvement in the duration of unemployment numbers. Clearly a measure of unemployment that by definition excludes 41.8% of the unemployed paints a very incomplete picture.

After the 26 weeks are up, people move over to extended benefits, which are paid for by the Federal government. These benefits can increase the total amount of time people get benefits to up to 99 weeks (depending on the unemployment rate in your state). The map below (from http://www.cbpp.org/cms/index.cfm?fa=view&id=3164) shows the maximum length of benefits by state. While regular claims are down, it is in large part due to people aging out of the regular benefits and “graduating to extended benefits.

However, recently even the extended claims have started to trend down, but in an irregular fashion. This week they (the two largest programs combined) fell by 262,000 this week to 4.665 million and are up by 672,000, or 11.9% over the past year. A much better measure is the total number of people getting benefits, regardless of which level of government pays for them. Including a few other minor programs, claims fell by 316,000 in the last week, and are down by 672,000 or 7.3% over the last year.