Thursday, September 24, 2009

Unemployment insurance rate: still a leading indicator of the national unemployment rate

Every morning I give my economic spiel to the bond group – this morning, the Department of Labor reported that the number of weekly initial claimants fell 21k to 530k, dragging the 4-wk moving average down 11k to 553.5k. Also in the release, the insured unemployment rate (number of employees claiming unemployment insurance divided by the stock of employees that qualify for unemployment insurance under the regular 26-week (generally) state programs), which is seen as a leading indicator of the unemployment rate, dipped 0.1% to 4.6% in the week ending September 12. A downward trend here normally leads the national unemployment rate.

But in times like these, when the actual number of insured rests around 9 million and exceeds the 6.1 million in the regular state programs, does the insured unemployment rate still indicate trends in the national unemployment rate? (The chart to the left illustrates the total unemployed claiming insurance benefits under the regular state programs (the calculation of the aforementioned unemployment insurance rate) + claimants under the emergency programs, EUC 2008 and Extended Benefits (see the release here).)

It looks like the relationship remains rather strong. The chart below illustrates the estimated relationship between the monthly average of the insured unemployment rate and the national unemployment rate (currently 9.7%) since 1981.

The simple equation has an R2 = 0.858, which is respectable. And the most recent data points, July and August in green and red, respectively, rest very close to the fitted line – August is right on the fitted line. If the insured unemployment rate continues to decline, the relationship suggests that so, too, will the unemployment rate.

However, the initial claims numbers will be dropping as well, and initial claims are the most current information out there (besides the daily Treasury receipts). Initial claims remain well above any level that would suggest a decline in the unemployment rate (around 350k-400k).

I don't believe that the economy will see a jobless recovery - i.e., the job loss that occurred for almost two years following the end of the 2001 recession (November 2001).

The 4-wk average to date is starting to look that way, but there is just so much spare capacity - August 2009 capacity utilization rate was just 69.6% compared to 73.5% in November 2001. I just don't see why a firm would opt to buy new capital before it uses its excess capacity - that means hiring workers.

Rebecca Wilder

7 comments:

  1. "I just don't see why a firm would opt to buy new capital before it uses its excess capacity - that means hiring workers."

    I just don't see why a firm would raise production period, unless government stimulus is going to be perpetual.

    ReplyDelete
  2. Rebecca,
    A very thoughtful, well argued, well modelled thesis. Can I ask what production will need to ramp up though? After the last recession (minor) production of most things only recovered due to the home ATM machine being used to buy and furnish homes as well as plenty of consumer non essentials. That capital is now gone. I am a pessimist, but I am still looking for the sector to lead a way out here.

    ReplyDelete
  3. I don't see why a firm would hire new workers before it gets its existing staff back to full time. I'm seeing a lot of 4-day schedules and 1-week-per-month shutdowns from the suppliers for the firm I work for.

    You can have capacity numbers so low that you have to first get your part-time staff up to full time before bothering to hire new staff.

    Particularly when (thanks to things like health care reform) you can't predict what a new worker would cost, why hire?

    ReplyDelete
  4. HI getyourself connected,

    I don't know - anything service-oriented. The service sector (anything from delivery pizza to fiber-optic cable installation to dentistry) needs labor. When demand re-emerges, it is very likely that this sector will lead the way.

    It was really nice in Boston today, wasn't it?

    Hell Who Struck John!

    It is always nice to hear a fresh voice around here. You bring up a very good point - hours have been slashed pretty much in line with employment, so there will be some output that can be gained simply by adding work hours. However, that can only go so far (see this point see this post that I wrote a month or so ago). Look for leading indicators like employment services to get a feel for hiring trends (virtually non-existent at this point).

    Rebecca

    ReplyDelete
  5. rebecca,
    I thought it was a bit humid for September, but who can argue with the great sunshine!

    We are indeed a service based econonmy, but I think overbuilt, overplayed areas (think phoenix, etc) will never see growth in service like they estimated. I think there are 2 or 3 too many Starbucks, Dunkin Donuts (if you are a true Bostonian!!), or Hi-Fi Pizzas as is.

    Along these lines, you may find Bill Gorss of PIMCO (who is a shill, but whatever) take ona new normal at odds with a huge recovery story. The problem with economics is the glacial pace at which things move. You have to wait so long to see things.

    ReplyDelete
  6. Hi getyourselfconnected,

    "new normal at odds with a huge recovery story"

    To be sure, 2.5-3% growth in 2010, which is the consensus forecast, is not a huge recovery by any means. That is why the unemployment rate is expected to remain elevated above the natural rate well into 2011 and beyond (depending on the growth rate).

    The service sector is dropping quickly; but like you said, if I had to guess on which sector will start adding jobs first, it would be there. I work in one of the towers off the Pru - just saw a "help wanted, full-time" sign in the window of Papyrus (a luxury card store).

    Rebecca

    ReplyDelete
  7. I have no idea what to think about unemployment.

    Gasoline prices are falling.

    The rate-of-change in monetary flows (MVt) (the proxy for real growth) has started it's projected decline (the first since Sept 08).

    ReplyDelete