Our earlier post on Exhausted Unemployment claims has provoked an interesting (if wonky) discussion on Claims Exhaustion.

Yes, it is a moving average, but that rise starting in December 2007 — when the recession began — which spiked vertically is unmistakable.

Here is the Labor Department data source, and their description of the data series:

“The monthly summary report is intended to provide the user with a quick overview of the status of the UI system at the national and state levels. This summary report contains monthly information on claims activities and on the number and amount of payments under State unemployment insurance laws. This data is used in budgetary and administrative planning, program evaluation, and reports to Congress and the public.”

The data goes back to 1972.

I looked at a simple 24 month series:



As was noted by Transor Z earlier:

Since Exhaustion Rate was first reported for the 6/72 data set, the average value has been 34% and the median value has been 33%.

In 37 years there have been 44 monthly values of 40% or higher. Only 7 of these were NOT in the period 2002-2009 and the highest of these was the 8/31/83 set with 40.83.

38.3 was the peak for the 1970s — 11/30/76.
40.83 was the high for the 1980s — 8/31/83.
40.11 was the high for the 1990s — 11/30/92

Until 3/31/09, the previous all-time high was set in 7/31/03 with 43.9. The March, April, and May figures have set successive all-time highs. You have to go back to 1/31/91 (29.87) to find a rate under 30%.

When you plot 12-month average benefits duration on the same graph, you see a clear correlation between the two. One question the data raise is that, during the early 80s recession, the average went over 17 weeks from 5/31/82 to 3/31/83. The 27 year average is 14.76 weeks. The latest figures put us only at 15.67 weeks. Given that even the “green shoots” crowd is predicting a jobless recovery at least in the short term, a 17+ week average would seem likely to follow, sending the exhaustion rate to further 27-year highs.


Here is another good chart showing the 2009 highs significantly above prior peaks:


Unemployment Benefit Exhaustions
Regular State UI Final Payments, Data from 1972 to 2009


Chart via BLS)

Exhaustions are the number of claimants who collect their full entitlement of unemployment insurance benefits. The data shown in the chart are for the regular state program.

(Hat tip, Kid Dynamite)

Category: Data Analysis, Economy, Employment

Please use the comments to demonstrate your own ignorance, unfamiliarity with empirical data and lack of respect for scientific knowledge. Be sure to create straw men and argue against things I have neither said nor implied. If you could repeat previously discredited memes or steer the conversation into irrelevant, off topic discussions, it would be appreciated. Lastly, kindly forgo all civility in your discourse . . . you are, after all, anonymous.

34 Responses to “Exhausted Claims part II”

  1. ben22 says:

    Given that even the “green shoots” crowd is predicting a jobless recovery at least in the short term, a 17+ week average would seem likely to follow, sending the exhaustion rate to further 27-year highs.

    That predication makes no sense to simple minds like mine. A credit money system and a consumer based economy can have a recovery on the backs of consumers without savings or jobs. Huh?

    I’d expect all the old measures such as this to show new records (the bad kind of records) given the size and scope of this downturn.

    Everything should be pushed to the extreme during credit deflation, which is already here.

  2. GB says:

    Quick we need a chart of U3 and U6 unemployment plus exhaustion

  3. DeDude says:

    The congress prolonged the maximum length of receiving benefits late last year. That has to have some effects on the number receiving and exhausting their benefits. I still think U6 is the best number to use and that the other numbers just provide a little extra granularity in the information.

    The “green shoots” debate is stupid. There is going to be something positive to be seen somewhere, even in the worst of all situations. What count is whether there are major trends or the big picture is “green”.

  4. Simon says:

    What happens to people after their benefits are exhausted and they are no longer recorded as unemployed? Commenting from the distant (socialist by comparison) Islands of New Zealand it seems totally bizaar that a benefit that might be keeping a family alive and together gets cut off after a certain time. Not recording them as unemployed seems like virtually culling them from society. Oh lets ignor those guys they are total losers.

  5. Simon says:

    I mean lets remember for a second where the blame for all this can likely be laid. Mostly NOT at the feet of those most effected by it. (Greenspan Banksters Insurance Fraudsters, They guys who get the biggest paychecks come what may)

  6. call me ahab says:


    re unemployment insurance- it does expire but was set up that way- a temporary hand to folks who are laid off- and is administered by the states and is paid for by a fund paid into by businesses-

    if they remain out of work then they will need to apply for other programs avaialbe from the states and federal government that are more long term such as welfare

  7. KidDynamite says:

    thanks barry… nitpickers will say that the problem with the chart you have (of absolute exhaustion totals) is that the labor force is now larger, so of course there will be more exhaustions. obviously, i think the recent spike thinks for itself.

    the simple fact that we’re seeing record exhaustion rates should be enough to explain to people why this time is not like the previous times in terms of recovery from recession…

  8. willid3 says:

    how much larger is the work force than it was in earlier times? and how does it compare to the estimated total population? and has the way its is measured be changed since those earlier dates?

  9. The Curmudgeon says:

    “(socialist by comparison)”


    wtf? Is that possible? The old Soviet Union wasn’t even socialist by comparison to the US today…

    …as to what happens when the unemployment checks run out? We’ll soon find out. The Grapes of Wrath comes to mind, except this time the Great Plains will be choked with economic refugees leaving California and heading East, to Washington, from whence all wealth emanates.

  10. dead hobo says:

    How can you not feel sorry for these people?

    If Uncle Stupid really wanted to make a difference, he would take steps to lower oil prices by regulating speculative derivatives that exist beyond the need for price discovery. This would decrease the Speculator Tax we all pay and goose spending. Jobs would result. Then Uncle Stupid should figure out how to stimulate R&D and Investment towards energy that replaces or significantly augments oil. More Jobs would result. This stimulus would be an excellent proxy for infrastructure spending. FDR had roads and dams, Uncle Stupid could have energy plentiful enough to waste if he tried.

    Life will suck for more and more people as long as the Speculator Tax continues and as long as it can mutate and grow whenever it wants.

    Hell of a day in the markets. Is it possible that the Fed decided to lower long term rates by removing special secret stock market stimulus from the playbook, causing a flight to safety in the 10 year? It will take a pretty big dip to keep people scared enough to stay in the 10 year. Then, kaboom in a few months. S&P 1000.

  11. pr1244883 says:

    First, everyone should realize that the exhaustion rate is a lagged statistic. They use the current 12 month total of final payments to the 12 month total of first payment 26 weeks ago. Therefore if they used both current data points the exhaustio rate would be 35%

    Second, does anyone know how the final payment figure is calculated? It says the last check paid under their original entitlement period. Since the UI benefits were extended to 40 weeks last year, are these figures describing the number of people that have been unemployed for 40 weeks in a row, or is it based on the original 26 weeks?

  12. dead hobo says:

    If the coming big dip is being engineered to lower the rates on the 10 year, then some pretty fancy dancing will have to be dreamed up so that whoever bought new issue bank stocks during the recent banking recapitalization won’t feel like they were screwed and told to call a cab.

  13. Transor Z says:

    @Barry, I’m just noticing now that you quoted me on this. I’m hugely honored.

  14. Its good stuff !

    Same for Kid Dynamite

  15. Pat G. says:

    “Only 7 of these were NOT in the period 2002-2009 ”

    I knew that there was going to be something special about this millennium. Folks who have exhausted their initial claim for unemployment benefits may be eligible for an extension of those benefits through monies appropriated in the Reinvestment and Recovery Act. So, tell anyone you know in that situation to refile. It might buy them another 20-26 weeks of benefits, depending on what state they live in. After that, if it doesn’t start to get better, I expect all hell to break loose.

  16. Mr. C. Cheese says:

    Goldman on track for record bonuses….Ouch !

  17. drollere says:

    this is the first posting of the data in which the exhaustion rate is clearly defined.

    the spate of commentary about whether this is a lagging indicator, or a moving average, or affected by laws or extended benefits or state regulations — all that seems to throw a lot of erudition around, but it comes a little late to disguise the basic imbecility.

    it’s a basic imbecility to interpret charts, diagrams, plots, schematics without a clear concept of what has been calculated or how the quantities behind the calculations were measured or whether they were measured in the same way over the entire time period or displayed on dimensions that make visual sense. those technical issues are preliminary to the even more obvious question of whether it matters, or to what, or by how much.

    to “whether it matters,” it should be obvious that the exhaustion rate is a moronic statistic to even look at, since it cannot tell you the number of total unemployed, or the number of new unemployed, or the number who were unemployed who find jobs. it apparently tells you the number of “last seen unemployed” now as a proportion of the number of new unemployed some months ago, minus some undeterminate number of people who found work, left town, died, struck it rich or were hit by lightning, expressed as a ratio — which is about the least stable *and* least interpretable mathematical form in which you can cast a statistic.

    what is wrong with total unemployed and new unemployed for commentary — and leaving exhaustion rate to the welfare budget office? those are numbers (and charts) you can use.

  18. wunsacon says:

    At what point do unemployed Palin/Joe-the-Plumber voters start filing for and then demand new-or-expanded welfare benefits and FDR-like workfare programs (e.g., in the form of the Pickens Plan)?

  19. Transor Z says:


    I posted this on the other thread in response to one of Kid Dynamite’s comments:

    You can back into something like the NSA continuing claims # (i.e., within 5%) by dividing total benefits paid by (avg wkly benefit amt*4). I get 5.7 million; the “official” NSA number is ~6 million.

    Working with the 5.7 million figure, that puts final payments at ~10% of total # payments — definitely not near the 50% mark I erroneously stated earlier.

    Now, check this out: If you work with the approximated NSA continuing claims figure using the quick-and-dirty calculation above, you can chart monthly expired claims as a % of NSA continuing claims. You see sustained peaks of >12% in 1975, 1983, and 2003. What don’t you see? You don’t see a >12% peak for 2009 . . . yet. Talk about bearish and ominous.

    I’m wondering if there may be some predictive value in looking at monthly expired claims as a % of the NSA total number of UI benefit recipients estimated using this DoL data. I was just playing with these numbers but the results were pretty interesting (I thought anyway) when I charted it.

  20. matt says:

    “We intended first to boost the stock and property markets. Supported by this safety net – rising markets – export-oriented industries were supposed to reshape themselves so they could adapt to a domestic-led economy. This step was supposed to bring about an enormous growth of assets over every economic sector. The wealth effect would in turn touch off personal consumption and residential investment, followed by an increase in investment in plant and equipment. In the end, loosened monetary policy would boost real economic growth.”

    It sounds plausible and like it might work. Except that it didn’t. The unnamed official was not an anonymous friend of mine at the PBoC. According to Tomohiko Taniguchi, in Japan’s Banks and the “Bubble economy” of the Late 1980s, the speaker was an official at the Bank of Japan and he made the comments in 1988, during a period when Japan was routinely referred to as a “creditor superpower” (and a country, by the way, with enormous foreign currency reserves, and whose currency would within one or two decades, everyone knew, become the world’s reserve currency).

    After the 1987 Crash in the US, many expected the Japanese markets also to crash. But they didn’t. After faltering briefly, the Ministry of Finance ordered the Big Four brokerages to support the market, and support it they did. Within a few months the Nikkei was testing new highs, leading a Ministry of Finance official to boast that manipulating the stock market was easier than controlling foreign exchange. Check Edward Chancellor’s Devil Take the Hindmost for an illuminating take on the Japanese bubble economy of the 1980s.

    From mpettis.com

    I thought some of you might find this interesting, given that the U.S. government seems to be taking the “attempt to inflate asset prices and all will be fine” approach.

  21. wunsacon says:

    I’ve been saying for a few years that “all jobs not nailed to the floor will go overseas”. Perot’s Giant Sucking Sound was only masked by the sequential booms of tech and housing. (Tech itself replaces workers, too. Ya need fewer real estate bubbleheads when services like redfin are around.) Along with the implosion in real personal income comes the implosion of the credit originally based on that income. “Poof”.

  22. wunsacon says:

    …Of course, that’s nothing new to this crowd. But, we’re finally seeing the scenario play out and show up in the data.

  23. Marcus Aurelius says:

    Green chutes.

  24. wunsacon says:

    Let’s give more money to the banks. Didn’t that “work” before??

    We won’t know we’re a “cargo cult” until we do it again and this time it *doesn’t* work.

    I’ll take drums. Who can dance? MEH?? Franklin??

  25. Todd in SM says:

    Barry – these are absolute numbers, correct? Wouldn’t a more historically accurate picture be gleaned from a ratio of these numbers against the total available workforce?

    Just wondering how much we are versus the 70s lull.


  26. Transor Z says:

    Here’s an April 2008 DoL summary of the findings of a pilot program to determine whether Exhaustion Rate really does correlate to reemployment rates.

    A couple of interesting take-aways:
    (1) acknowledgement by DoL that the 26-week period is higher than the average benefit exhaustion period, which is only 22 weeks.
    (2) acknowledgement that exhaustion rate does not correlate to reemployment in Q+1 but does roughly approximate reemployment in Q+2 (for current purposes you might wonder whether even that is true during a severe recession).

    Some historic data is available for extended benefits paid during past recessions and for all years from 1940 to 1998.

    Annual exhaustion data is also available:

  27. constantnormal says:

    What we really need is exhausted claims data going back to the 1930′s, so we can make proper comparisons to the Great Depression (a little TBP humor).

    Or at least appreciate the differences between then and now.

  28. constantnormal says:

    @wunsacon 6:33 pm

    — “Tech itself replaces workers, too.”

    This is something that has been ongoing for as long as there has been “technology”. The Pollyannas will tell you that something new always comes along as the economy expands and creates an ever-expanding mandala of employment choices.

    The Dismalists will point out that the rate of technological improvement is increasing, with the newly created jobs requiring ever-more-sophisticated skill sets, and that eventually robots will do everything.

    The Dismalists have a point. I suppose we could create new simple-enough-that even-a-human-can-do-them “make work” jobs, but people are smart enough that I doubt these positions would be very fulfilling — much the same as most government jobs today.

    But I wonder if anyone is looking at what kind of socioeconomic models would replace the pyramid of capitalism in such a system.

  29. mathman says:

    For the most part, newly unemployed workers find that there are no jobs paying the same rate they’ve accumulated over their years of service and that starting over means, in reality, taking a huge pay cut.
    Now that most of the jobs in our economy are service related, the starting pay rate is almost impossible to live on. We have a workforce being marginalized, underemployed, and replaced with machines to keep corporate profits up. This is what capitalism does – kills democracy (through influence) and makes life miserable for the citizenry (known as the workforce). When it comes to shoring up corporations on the backs of labor, then cutting labor to the bone, what result should any clear-thinking person expect?

  30. RangerTurtle says:

    When all the burger-flipping jobs are gone, unemployment exhausted, states broke from welfare, stim-one exhausted, we can then compare Depression II to Depression I. Maybe then it would be good to march on Washington and demand help for PEOPLE instead of help for CORPORATIONS/BANKS. Maybe that time is getting closer!

  31. Frank FB says:

    Since 1972 the labor force must have grown considerably. Would not the unemployment rate (and exhaustion rate) be better stated as percentages of the labor force, rather than raw figures as shown on the graph?

  32. [...] then it was taking money from our imaginary Uncle Sam. Now we get money from … ? (Source: The Big Picture) Chinese Electricity Output and GDP [...]