New Unemployment Claims Down

New Claims Down, Existing Ones Up

There was more evidence today that the economy is stabilizing in the initial unemployment claims data. This week, initial claims dropped by 24,000 to 601,000. That brought the four-week moving average down to 621,750, a decline of 10,500.

The four-week average is now down by 37,000 from its peak set two months ago. This is significant since historically, the four-week average has peaked out right around the time the NBER has dated the end of previous recessions.

Still, as the chart below (from http://www.calculatedriskblog.com/) shows, the level of new claims is still extraordinarily high. Unfortunately, the chart does not have the recession bars in to show the point about being an end of recession signal. This data does tell us that the economy is no longer cliff-diving. Now it is stagnating at a very low level.

However, in the past, the decline in initial claims was confirmed shortly thereafter by a decline, or at least a leveling-out of continuing claims. It looked like that might have happened last week, but unfortunately that was revised away, so we now have continuing claims setting a record for 19 weeks in a row.

This week they reached 6.816 million, 59,000 above the revised level of last week. In absolute terms, it simply blows away any past peak. However, the population and the workforce grow over time. As a percentage of covered employment, the current situation is bad, but not as bad as in either the mid 1970's or the early eighties downturns. Currently 5.1% of covered employees are getting unemployment checks vs. 5.4% at the peak in 1982 and 7.0% in 1975.

The divergence between the initial and continuing claims series suggests that while the pace of new layoffs has slowed, there has been no real pick-up in hiring. A somewhat more benign reason is that for most of the country, unemployment benefits have been extended as part of the stimulus package (though this depends on the state).

However, this almost always happens in a recession, so while it might influence the level of claims relative to a few years ago, it is not a factor compared to previous high levels. As I have written about repeatedly, the length of time people are unemployed once they lose their jobs is extraordinarily large this time around ("off the charts" is probably a better way of putting it). This means that even with the extended benefits, many people are dropping out of the continuing claims number because their benefits have been exhausted. Down that road leads to serious poverty.

Despite the good news on initial claims, it is highly likely that unemployment will continue to rise.

Note that during the previous two recessions, after continuing claims peaked they remained elevated for an extremely long time -- a very different pattern than in earlier downturns. We have not yet reached the top of the "mesa" this time, but when we do, I fear that it could be even wider than the one associated with the 2001 recession.

How do people cope with being out of work for an extended period of time, especially after the benefits have run out? Well, first they tap their savings -- but given the extremely low savings rates over the past several years, those are likely depleted very fast. To the extent they were held in the market, the recent rebound has helped, but the S&P 500 is still 31% lower than it was a year ago. In the past, people might have tapped the equity in their houses to get by, but the housing ATM long ago shut down.

The only thing left is to run up the credit card balance. However, when they get maxed-out, the only resort is to default and go into bankruptcy. Rapidly increasing credit card defaults are going to be a serious problem for banks with big exposures to credit cards like Capitol One (COF) and American Express (AXP).

On the plus side, it means we have a very large gap between actual output and potential output, which will for the time being prevent inflation from getting out of hand -- despite the flood of liquidity being provided by the Fed. Core inflation will not be a problem for awhile, but headline inflation will spike due to rising energy prices. It will not cause a wage price spiral, just lower real standards of living as prices go up and incomes stagnate.

Read the full analyst report on COF

Read the full analyst report on AXP