Guest post: Both initial claims and continuing claims now pointing to recovery

Submitted by Edward Harrison of the site Credit Writedowns.

I’m back after a brief hiatus. I would like to present you with some data on jobless claims, an economic data series that will become increasingly important in the weeks and months ahead. The data suggest that a recovery is imminent. This should come as no surprise as everyone is jumping on the recovery bandwagon (Joseph Stiglitz and Nouriel Roubini are just two examples). How robust a recovery we see and whether this recovery is sustainable or leads to a double dip are wholly different questions.

Here are the data. In the week ended 16 May 2009, there were a seasonally-adjusted (SA) 631,000 initial claims for unemployment in the U.S. While this was 12,000 fewer than the previous week, continuing claims continued an upward trajectory. A SA record 6.66 million continuing claims for unemployment were filed in the week ended 9 May 2009. What should be clear from these two data points is that the employment market in the U.S. is weak and those being laid off are finding it hard to find jobs.

jobless-claims-2009-05-21

However, this week’s data reveal a more positive message underneath the gloom. Back in June of last year, I mentioned that jobless claims had a very good track record of predicting recession and recovery (see my post “Another Perfect Recession Indicator”. I mentioned in that post that a particular series I tracked (the yearly change in continuing claims) gave an indication that recession began in December 2007. Now that we have been in recession for nearly a year and half, both initial jobless claims and continuing claims are suggesting that the recession will end later this year.

With initial claims, I am now looking at a 4-week average of the unadjusted numbers and comparing them to levels one year ago. If these comparisons start declining, that is a good sign. I am also doing the same exercise with continuing claims. And in the past, this has been a fairly good indicator of an end to recession. Look at the charts below.

jobless-claims-2009-05-21-history
continuing-claims-2009-05-21

What you should notice in the charts above is that the yearly change in both initial and continuing claims has peaked right at the time the recession ends in all recessions since 1967 when the data series began.

In this particular recession, the initial claims comparison peaked in January and the continuing claims comparison reached a (temporary?) peak last week. To be more specific, the year-on-year change in 4-week average unadjusted initial jobless claims reached a high of 327,590 on 31 Jan 2009 and the year-on-year change in 4-week average unadjusted continuing claims reached its current peak of 3,361,267 on 2 May 2009, declining slightly for the last week that is available. Whether the continuing claims peak holds is an open question. But both initial claims and continuing claims are now pointing to a recovery. And, by the way, you should notice that actual continuing claims have been declining for three weeks. Seasonal adjustments have sent them higher.

Assuming the continuing claims series change has peaked or will soon peak, the obvious question is this: “Why is this recession different?” Before I answer that question, I should point out what recession is and is not. “A recession is a significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in real GDP, real income, employment, industrial production, and wholesale-retail sales.” (see my post “Economic recovery and the perverse math of GDP reporting”). So a recession basically measures the first derivative. That means a recession is over when the first derivative or the change in employment, income, GDP, and sales turn up. An end of a recession does not mean the economy is firing on all cylinders. It does not mean that the economy has returned to its previous level of output. Hardly, when a recession ends, output is always less than when it began – this is axiomatic. By definition, if recession lasts only as long as the change in economic variables is negative, output is going to be lower when it is over than when it began.

So, I see the end of recession as a much less important event than the media certainly seems to. And for the record, I have said I see a recovery happening probably in Q4 2009 or Q1 2010 (see my post “The Fake Recovery”).

The real question is how robust a recovery are we going to have and this is directly related to why the jobless claims series has been sending a false signal. Now, initial claims has been sending a recovery signal since January. Yet, continuing claims continued to rise more quickly until last week. In the past, one had seen these two series as harbingers of imminent recovery. But, I am talking Q4 here. Why? Deleveraging.

In the end, consumers are going to be forced to reduce debt and save more in this more cautious financial environment. Team Obama does seem intent on re-kindling animal spirits but the personal savings rate has gone up nonetheless. This will be a drag on GDP growth going forward and means that the economy’s rebound will be more tenuous and slower to develop. In my view, this means recovery will be delayed and once it gets going it will be weak. The potential for a double dip is very high.

So, to be clear, first derivatives are starting to turn up and since recession is a first derivative event, we are probably going to see an end to this recession soon enough. But, with structural problems still remaining, the U.S. economy will be weak for a long time to come.

Source
Unemployment Insurance Weekly Claims Report – U.S. Department of Labor

Print Friendly
Tweet about this on TwitterDigg thisShare on Reddit0Share on StumbleUpon0Share on Facebook0Share on LinkedIn0Share on Google+0Buffer this pageEmail this to someone
This entry was posted in Guest Post on by .

About Edward Harrison

I am a banking and finance specialist at the economic consultancy Global Macro Advisors. Previously, I worked at Deutsche Bank, Bain, the Corporate Executive Board and Yahoo. I have a BA in Economics from Dartmouth College and an MBA in Finance from Columbia University. As to ideology, I would call myself a libertarian realist - believer in the primacy of markets over a statist approach. However, I am no ideologue who believes that markets can solve all problems. Having lived in a lot of different places, I tend to take a global approach to economics and politics. I started my career as a diplomat in the foreign service and speak German, Dutch, Swedish, Spanish and French as well as English and can read a number of other European languages. I enjoy a good debate on these issues and I hope you enjoy my blogs. Please do sign up for the Email and RSS feeds on my blog pages. Cheers. Edward http://www.creditwritedowns.com

19 comments

  1. AgentGenev

    what? this is total nonsense. first of all roubini and stiglitz were exactly the opposite of the recovery bandwagon in their latest statement, roubini wrote a massive paper about why there is no recovery coming any time soon.

    second, we’re not even seeing a slow down in claims, how does that point to recovery??

    Even the Fed which is idiotically optimistic about their forecasts just said that recovery may take 5-6 years.

    i really don’t know what the hell you’re on, but it definitely is you on the recovery bandwagon, not Roubini and Stiglitz

  2. spanwars

    Your model doesn’t seem to fit all your data. The second leg of the 1981-2 recession has the initial claims comparison peaking early, then dropping off slightly for a few quarters, before the actual big drop down that coincided with the end of the recession. So maybe we shouldn’t get too excited about “peaks”, but rather look for a big dropoff.

    I also suspect the “peak” of layoffs in January was partly due to “pent up” layoffs. I.e. the early stage of this recession was _very_ mild, due to the early intervention of the Fed, and setting expectations for the fabled 2nd half 2008 recovery, so companies held off on layoffs, and then when the recovery didn’t materialize, had an “excess” supply of employees to fire starting Q4. You can see this as the rather shallow slope of continuing claims in the early part of this recession… before it goes vertical.

  3. lineup32

    My brother’s company finally shut the doors last week now 75 employee’s without jobs. Anderson Lithograph in LA a high end printer shut its doors recently along with Graphic Press another LA printer combined they employed around 400.
    Last night I got a call from the owner of a small printer in Hayward, crying on the phone it was all over for him. The difference between real life and data points gets bigger every day.

  4. Edward Harrison

    spanwars,

    you are incorrect about 1982. In 1982, the one year comparison peaked at 1.2 million in Oct 1982, one month after recession. The numbers went up without retracing and fell after that point.

    lineup32,

    You imply the stats are not the real world. There is no difference between what the stats are saying and what is happening on the ground. Recession means an end to declines in the broader economy, it does not mean an end to employment weakness. The 1990-91 and 2001 recessions are examples of both (I know because I was directly affected both times -graduating into a recession the first time and working in the employment industry the second).

    Expect 10% unemployment in the fall. Recovery and rising unemployment are the usual trend when recession ends.

  5. econoreader

    Before making comparisons to previous recessions (and their recovery patterns), we should note that this recession differs from previous ones (post GD) in a number of significant ways:

    * Substanial drop off in consumer spending
    * Households with massive debt
    * Huge drop in wealth
    * Real estate prices down nationally
    * Monstrous government deficits
    * Financial institutions close to insolvent, credit scarce
    * Global trade down
    * Many states/cities revenues far short of spending

    This being a fundamentally different kind of economic contraction, perhaps we should use a different nomenclature.

    “Recovery” itself implies more than “recession/contraction over”, the word implies continued growth. And W-shaped recovery also implies growth (that middle slant looks up at least 30 degrees to me to be a W shape), otherwise if we went flat and then down… well that doesn’t look a like a W.

    Instead of looking forward to “recovery” now, we should just consider future anemic growth to be a “pause” before assuming which way it will go afterwards (up, down, or flat). I don’t know anyone who calls Japan’s last 18 years a “recovery” after their contraction was over, even though they logged many quarters of GDP growth (however small).

    Let’s call it recovery once the banks are being weaned off federal dollars, household debt is back down to a manageable level, and growth is organic. That will actually feel like recovery.

  6. spanwars

    Ed,

    My statement was about the initial claims comparison, not continuing, for 1982. Look at your graph. It peaks 1/4 of the way at about 200k into the 2nd leg of the 81-82 recession, then seems to vary between 110k-160k for the rest of the recession. Quite an early peak. Do you disagree?

  7. B. Mull

    There needs to be a better measure than unemployment claims. My company has a proud tradition of not laying people off.

    However, they aggressively demote and terminate people for poor evaluations (usually offering them the chance to quit honorably first). Naturally these people are ineligible for unemployment.

    Despite the fact that people are trying harder than ever to hold on to their jobs, I notice a definite up-tick recently in these forced resignations and terminations for cause. Go figure.

  8. tg

    John Mauldin has another opinion, which is much more logical for me:

    “The typical pundit keeps telling us unemployment is a lagging indicator, and that the recovery will be well under way before it shows up in the job numbers. Therefore, you should buy what they are selling, because the recovery is on its way. But that may not be the case this time. One of my favorite reads, when I get to see it, is the economic analysis from Bridgewater. They are among the best thinkers anywhere, and everyone who follows them gives them a great deal of credence. This is what they wrote about unemployment being a lagging indicator last month:

    “Normally, labor markets lag the economy because incremental spending transactions are financed via debt, stimulated by interest rate cuts. But as long as credit remains frozen, spending will require income, and income comes from jobs. And debt service payments are made out of income. Therefore, in a deleveraging environment job growth becomes an important leading, causal indicator of demand and other economic conditions.

    “… The bounce in the economy and the stabilization in markets reflect government actions that are big enough to impact near-term growth rates, but are not sufficiently directed at the root problem of excessive indebtedness to produce permanent healing. The deterioration in employment markets will continue because companies’ profit margins are so deeply damaged that a little bounce in growth won’t do much to alter their need to cut costs. This deterioration in labor markets will undermine demand and continue to pressure loan losses, which will keep the pressure on the banks and elevate the cost of capital for tentative borrowers, inhibiting credit expansion.”

    This again illustrates the problem of using past performance to project future results. You have to look at the underlying conditions in order to get a real comparison, and we have not seen a deleveraging recession in the US for 80 years. Using the past data in today’s world is statistical masturbation: it may make you feel good, but it is not producing anything really useful, and may be harmful to your portfolio.”

  9. lineup32

    Your past work life in the employment field is valuable but whether it gives you some special insight into unemployment and how it will impact the economy I don’t know. My own experience of owning a small manufacturing firm,borrowing money,hiring,firing,making payroll and paying taxes for 25 years has shaped my idea of what a recovery might look like and I don’t see one bit of evidence that recovery for the American worker is anywhere on the horizon.
    Best of luck to you and thanks for posting.

  10. Neal

    So when the slope reaches 0, the ultimate question is if the recovery will look like a V, or U or W or even a L. Or perhaps even like the edge of the continental shelf–a more rapid drop to follow.

    Rignt now there is much talk about “green shoots”.

    Unfortunately, the “green shoots” depend much on pretending that the banks are OK, that the auto industry will not essentially shut down completely this summer, that more foreclosures are not in the offing, that the vaunted stability of the dollar isn’t threatened, that the state and local governments haven’t entered into a time of vigorous cutbacks.

    Other than that, I would say it is probable that we have entered a period of stability and perhaps the economy will not fall further for a month or so.

  11. Edward Harrison

    tg,

    that’s the right analysis. Perhaps you think I’m bullish here because I am talking about a recovery. I am not. I agree 100% with what Ray Dalio at Bridgewater has to say. In fact, I thought of using a quote from that Mauldin piece in this post, so I am glad you brought that to everyone’s attention.

    as for the term ‘recovery,’ many of the comments feel it is a misnomer. I would tend to agree with that because the middle class is getting squeezed and this will not feel like a recovery in the least.

    spanwars,

    yes, you’re right. there was a peak in initial clams in 1982 way before the recession ended. It was Jan 1982. That’s a good catch because it brings up something I think Davd Rosenberg is on to, that you can have growth in a recession. April – June 1982 saw 1.7% growth in GDP, but Jul-Sep was down and Oct-Dec barely grew.

    My thinking here is that we could see a repeat: an inventory induced (shedding less inventory adds to GDP growth) uptick in Q2 or Q3 followed by a downtick before a sustainable recovery comes onboard.

    The key being what tg pointed out: deleveraging. That’s what’s different in this recession.

    lineup32,

    Your narrow view of the economy may cloud your broader view of the whole economy. Remember, recovery means things have stopped going down in general in the U.S. as a whole. If Q1 2010 comes and we are 5% lower in output than we were 2 years earlier, it’s a lot easier to see growth. And that doesn’t mean everyone is benefitting. Did everyone benefit in 2001 when that recession ended? Recovery doesn’t mean all is right with the world, it just means the statistics have stopped getting worse.

  12. Brian's Blog

    I think your analysis is rather timid and not particularly insightful. All recessions ended with a recovery; the fact that we are in a recession now implies there will be a recovery. As others have pointed out, this isn’t a typical recession; unemployment claims – new and continuing – could rise. Your limited model does not account for this. Hence, your prognostication is conjecture.

  13. Brian's Blog

    I think your analysis is rather timid and not particularly insightful. All recessions ended with a recovery; the fact that we are in a recession now implies there will be a recovery. As others have pointed out, this isn’t a typical recession; unemployment claims – new and continuing – could rise. Your limited model does not account for this. Hence, your prognostication is conjecture.

  14. Edward Harrison

    Brian,

    I am a labor economist and I have models supporting the view that their is a strong statistical correlation between the 6-month and 1-year change in jobless claims and recessions ends and beginnings.

    But, perhaps you are right that the post is not fleshed out enough in presenting the statistical evidence for the connection between jobless claim and recession.

    A good attempt is here:
    http://www.voxeu.org/index.php?q=node/3524

    I am not as bullish as Gordon here. What I am saying however is that this statistical correlation is due to the relationship between consumer spending and GDP growth and that this relationship has broken down in this particular recession, the reason being the unusually low savings rate and high debt in the U.S.

    In the period from the mid-1980s, recovery was achieve through consumers decreasing savings rates and increasing debt. As the savings rate is already very low and debt financing is more difficult, I anticipate the relationship between a peak in claims and GDP will not hold. The recession will be delayed and the recovery will be weak.

  15. tg

    “I am a labor economist and I have models supporting the view that their is a strong statistical correlation between the 6-month and 1-year change in jobless claims and recessions ends and beginnings.”

    I would not use any models. All the models failed already in this recessions and on the way to it.

    We have so much factors in the economy and this crises is so different from all of the previous ones, that I will use my common sense instead of statistical future telling.

    Nevertheless is nice to hear other opinions as well. Who knows, perhaps you are right.

  16. Elizabeth Ericson

    What we are seeing here is free trade in reverse. An economy largely devoid of labor standards imploding. Anyone over the age of 30 is going to face a rude awakening in mass numbers. Maybe if they are not traumatized, they will obtain a minimum wage part-time “JOB” within 3 years to pay for their divorce.

  17. Hugh

    A recovery that is not sustainable is not a recovery.

    I tend to look at employement numbers. I take jobs lost and add them to jobs that would normally be needed to compensate for population growth.

    When I do look at unemployment I tend to look at the standard U3 number and the U6 which is a broader measure of underemployment and those marginally attached to the workforce.

    What these two things tell me is that unemployment remains high and job creation is still negative.

    When I factor in how many families are facing foreclosure or are underwater on their mortgages, I just don’t see any cause for your brand of optimism.

    I think we all agree that the Obama stimulus, small as it is in comparison to what is needed, will have some moderating effect mostly next year. But that is one shot and done by early 2011.

    As for the money being funneled to the banksters, we are already seeing it being used in economically negative ways as with the speculative spike in oil prices. The curious thing is that this attempt to re-inflate the paper bubble will actually worsen conditions in the real economy because it is a misallocation of resources away from productive sectors of the economy and through speculative plays and ventures will serve to funnel money out of the real economy (the one in which most Americans live).

Comments are closed.