The behavior of weekly unemployment claims seems to be signaling the end of the current recession. Claims were relatively flat at high levels for the past 6-7 weeks, and this week they dropped unexpectedly by a fairly large amount (610K instead of the expected 660K). This chart (its first appearance here) shows the rate of change, year over year, of the 4-week moving average of claims. That growth rate peaked 5 weeks ago, and has since dropped meaningfully. I would note that the reduction in claims was foreshadowed in my post last weekend. Obviously we'll have to see more follow-through to be sure, but this could well mark an inflection point in the economy, with the rate of decline now slowing instead of picking up speed. Similar behavior can be seen at the end of the last two recessions. The list of "green shoots" is getting longer.
Scott-
ReplyDeleteThis is consistent with what I hear anecdotally from companies. While not rocking and rolling, they say that business is stabilizing and becoming more predictable. Lack of visibility was a key factor in driving layoffs to extremes. Also, we may see a surprising level of profitability in the near future. Downturns are great occasions to clear out some dead wood, and unsurprisingly some CEOs have told me they've done just that. It would take only a moderate pickup in business to see some operating leverage kick given the much leaner cost base.
Thanks, that's very encouraging.
ReplyDeleteScott,
ReplyDeletePrevious recession in 2002 showed a slight dip then went up much higher.
How do we know its not a repeat?
No way to know, of course, until more facts come in. Meanwhile, I would note the following which would support the guess that this time we have seen the peak in claims: 1) the recession has been underway for over 15 months, 2) online job postings are rising, 3) implied volatility in the markets is way down, meaning fears have declined and that is what bottoms are all about, 4) commodity prices are rebounding, 5) equities are rebounding, etc.
ReplyDeleteMiller-
ReplyDeleteRemember that in 2002, there was an extension of UI benefits that came out of Congress. A few analysts got it right and saw that claims would spike up again as more people signed on for a few more weeks of the gov't cheese, but that the spike would not be indicateive.
I think the dip on the chart that Miller refers to is in 2001, not in 2002.
ReplyDeleteHmm. Did something else happen in 2H '01...? I seem to remember something. What was it...?
ReplyDelete