Does anybody have a good explanation for this?
April 15 (Bloomberg) — The number of Americans filing claims for jobless benefits unexpectedly increased last week, indicating the improvement in the labor market will take time to unfold.
Initial jobless applications increased by 24,000 to 484,000 in the week ended April 10, the highest level since Feb. 20, Labor Department figures showed today in Washington. A Labor Department spokesman said the rise in claims was due more to administrative factors reflecting volatility around Easter than economic reasons.
Reluctance among some companies to hire is one of the challenges facing the economy as it recovers from the worst recession since the 1930s. Employment gains are needed to help spur consumer spending, which accounts for about 70 percent of the economy. “We’re not making rapid progress†in bringing down claims, Neal Soss, chief economist at Credit Suisse Holdings USA Inc. in New York, said in an interview with Bloomberg Radio, while also noting the holiday may have distorted the figures. “Job growth over the course of this year will be sufficient to bring the unemployment rate down.â€
Economists forecast claims would fall to 440,000 from a previously reported 460,000 the prior week, according to the median of 44 projections in a Bloomberg News survey. Estimates ranged from 415,000 to 451,000.
The emphasis is mine. What I’m puzzled by is not why the number of claims rose but why every economist surveyed underestimated the number of claims. Several possibilities come to mind:
- They’re surveying the wrong economists
- There were unforeseen eventualities. What?
- The data are wrong. How and why?
- The ceteris paribus mode of thinking frequently employed by economists doesn’t lend itself to the highly interactive real economy
- The economists are using similar models and their models are wrong.
A 5-15% underestimation is not negligible. What’s going on here?
Wishful thinking on the part of the economists?
Political gaming by the economists? (Unlikely, I’d judge)
Political gaming by the media? (More likely)
I suspect that it’s a problem with the models. Have you noticed how they always get the numbers wrong when they’re systemically down, but usually get them right on the upswing or with temporary downswings? I suspect that the models leave out some critical factor(s) that are present in long-term downturns but not in growth periods.
I know you’re not interested in the reason, but if the number of claims rose because people previously had left the job market had returned, then most of your suggestions would seem to fit. Specifically, the return of a disgruntled job seeker is almost purely a psychological phenomena that is difficult to model; I wonder if it is in the models (other than in disclaimer language).
Actually, I am interested in the reason. And, yes, job-seekers returning to the market is a possible explanation. Is there evidence for that other than the number of claims rising?
No, I don’t have any evidence. That’s what these figures look like to me; they confirm my bias that weak job improvements are goingt to be undercut for the foreseeable future by the return of discouraged workers to the labor force.
And I have no idea what Easter has to do with anything.
maybe has something to do with the census workers losing their job?
maybe economists have an interest in portraying current economic situation as rosier than it is (plugging in unrealistic numbers into the models), to beef up a positive psychology, to thereby encourage corps to hire?
Remember how Christine Romer, economic advisor to the Pres, predicted new job growth to begin lowering the employment rate below 8% in late 2009? What a joke. How was she able to do that?
5% is at the outer edge of acceptable error, anything beyond that is not.
I think one thing to keep in mind is that in the last two recessions the “lag” in unemployment has been substantial, and keeping in mind that 2 observations is not very large….a growing lag as well. In other words, we might not have bottomed out in terms of unemployment even though the economy might have bottomed out in terms of growth.
This w0uld suggest that the last reason is the explanation, the models are wrong.
There’s probably multiple causation working here but some of the disinclination to hire among companies is that new workers, in the present political climate, represent greater uncertainty in regard to future benefits and administrative costs than they did a year ago.
I’m not sure the models economists are are designed to measure that political perception or that most academic economists, who favor demand-side Keynesianism and robust government regulatory policies, are inclined to acknowledge that such costs exist for these policies.
Maybe . . . Just throwing out a crazy notion here . . . maybe attempting to model future activities involving humans is no more useful than consulting astrological tables. Maybe economists kind of know what happened 30 years ago but have no clue what’s happening next month.
Maybe, just maybe, they’re getting handsomely paid for what amounts to reading goat entrails.
There is a problem with this view.
Suppose you have a model, A, and it works well. It doesn’t tell you exactly when the recession starts, but it gives you some good information. Then something changes (new technology for example) and model A no longer works very well at all. Ideally we’d switch to B, but at the time of the change we don’t have enough information to determine the structure of B. So for a period of time the old models wont work and the new ones don’t have enough history to be developed. It looks alot like looking at goat entrails. But the basic underlying process is scientific. After all, that is how the “hard sciences” develop models. Model A works well. Whoops, these observations don’t fit, go back and see how we can re-work Model A to incorporate these new observations. The difference is that most people don’t care a whole lot about a model for biological evolution or planetary motion. Economics however touches much more on their lives.
I think we are in a period where something has changed. What has changded I don’t know, and trust me if I did I’d be working hard at developing new models based on that knowledge getting “famous” and possibly even making quite a bit of money. I hear being a tenured professor at a good school is a good gig. Good pay, summers off….