Friday, January 2, 2009

For now, stick with the non-seasonal numbers when it comes to the BLS employment report

Seasonal adjustments sometimes make an very weak labor market conditions less onerous; this was the case for this week's initial unemployment claims report, and will be for the employment reports in December and January. The statistical seasonal adjustment process doesn’t handle abnormalities in the data very well.

Next week, when the Bureau of Labor Statistics (BLS) releases the December employment report (and for January as well), I suggest following the non-seasonally adjusted data, rather than the seasonally adjusted data. The normal seasonal patterns are not present, and the seasonally adjusted data will paint a slightly more benign picture of the job loss that will be reported in the nonfarm payroll.

An example: seasonal adjustments cannot predict statistical anomalies in unemployment claims report

The Department of Labor released a shockingly strong weekly initial unemployment claims report last Thursday:
In the week ending Dec. 27, the advance figure for seasonally adjusted initial claims was 492,000, a decrease of 94,000 from the previous week's unrevised figure of 586,000. The 4-week moving average was 552,250, a decrease of 5,750 from the previous week's unrevised average of 558,000.

The advance seasonally adjusted insured unemployment rate was 3.4 percent for the week ending Dec. 20, an increase of 0.1 percentage point from the prior week's unrevised rate of 3.3 percent.
The labor market is improving – at least on a weekly basis, right? Wrong. The insured unemployment rate rose to 3.4% over the week, a signal that the national unemployment – currently at 6.7% - rate will rise.

The weekly claims fell simply because of the statistical process the BLS Department of Labor uses to rid the data of normal seasonal activity. Here is Bloomberg’s take on the report:
The number of Americans filing first-time claims for unemployment benefits tumbled last week, skewed by the shortened Christmas workweek, while total jobless rolls reached a 26-year high, signaling a worsening labor market as the economy heads into the second year of a recession.
I don't buy it. Christmas occurred on a workday (Mon. through Fri.) four out of the last five Christmas weeks, which should be (at least partially) accounted for in the seasonal adjustment process. Normally, the statistical process used to compensate for seasonal patterns works quite well. However, this process cannot compensate for anomalies in the data, or strange events that normally do not occur in December.

The chart illustrates the seasonally adjusted and non-seasonally adjusted data since Christmas of 2003, where I highlighted the comparable report for each of the preceding five years. Every year the number of claims filed in the last week of December rises, and on average, the NSA data rises by 75,233.

The seasonal adjustments extract this cyclical pattern (always rises in December) from the data, leaving only the trend that exceeds (or falls short of) the seasonalities. New claims filed beyond what normally happens in December (the seasonal adjustment) signals a weaker-than-normal labor market, while claims filed below what normally happens in December signals a stronger-than-normal labor market.

On December 27, 2008, the NSA number of claims filed rose by just 1,892, which is 73,331 below the average. The seasonal adjustment process sees this as a good thing, and reports a stronger-than-normal labor market, -94,000 new claims filed. But what happened on December 27, 2008 was clearly not normal, as every other indicator signals an contracting labor market.

It is more plausible that a negligible amount of seasonal hiring actually occurred in November and December. Therefore, the last week of December saw very few firings, and a negligible number of initial unemployment claims were filed. Seasonal adjustments would miss such a statistical anomaly.

Beware of the seasonal adjustments in this cycle; this applies to the Bureau of Labor Statistics employment report as well.

The chart illustrates nonfarm payroll since 2004. Like clockwork the seasonal pattern is this: firms hire in October and November for the holiday season; they start firing in December; and then they slash jobs in January.

This year, firms didn't hire in October nor in November - nonfarm payroll fell by 320,000 and 533,000, respectively. Therefore, the seasonal firing pattern through December and January will not exist and the seasonal adjustements will not apply.

In December and January, the seasonal adjusted series will show a stronger labor market than actually exists. Holiday workers will not be fired in December and January because they were never hired in October and November, and the seasonal job loss will be less than the non-seasonal job loss. The non-seasonally adjusted numbers will paint a more accurate picture of the labor market.

Rebecca Wilder


  1. I think DOL ETA produces the ui initial claims data rather than DOL BLS.


  2. Hi DD,

    Yes, you are correct, but how is this unclear in the article?


  3. "The weekly claims fell simply because of the statistical process the BLS uses to rid the data of normal seasonal activity."

    The above quote suggested to me the idea that BLS was producing the ui claims data.


  4. There is no reason to ever use the seasonally mal-adjusted data.

  5. Seasonal adjustments are an extension of the FED's "real bills doctrine".

    "But if the theory is applied under the assumption that labor and facilities are fully employed, the injection of new bank credit for financing inventories or any other phase of merchandising or processing is obviously inflationary"

  6. I note Obama stating in a Bloomberg report:
    [Jan. 3 (Bloomberg) -- President-elect Barack Obama said that Democrats and Republicans need to act with urgency to address the “great and growing” economic crisis, warning of double-digit unemployment if swift action isn’t taken. ]

    I tend to agree with his projection, vague as it is, but "double digit" is considerably greater than that predicted by the forecasting org that RW had posted some weeks back. I commented at the time that (gist) "those unemployment figures look too optimistic for such a marked downturn".

    There is a continuing disconnect as to how serious the present situation is, at least when expressing it against prior downturns. To be honest, I'm still sitting on the fence as to whether 'intervention' is actually helping or not. The UK is now about to try the initial US approach (buying up toxic assets), while the US is now doing what the Brits were doing initially!

    Omigod....I think the whole undertaking was best termed as "an experiment".

    Is this the deaf leading the blind, the blind leading the dumb, or the dumb talking to the deaf?

    I can't help but wonder....and this is echoed in the Sciences...sometimes, and I'm in the midst of a servicing dilemma as I write this...

    the reason the model isn't making sense is because the model was never correct to begin with! That's a very dark thing to write...but supposed models can be a huge disservice when you're looking for correct answers.

    If your premise is wrong to begin never will find the right answer except by sheer luck...

    Suddenly Kenynese is a cherished reference. In all deference to Keynes, he may have been wrong...but so may have been Friedman, Adam Smith, Greenspan and Bob Your Uncle.

    Perhaps when I get this piece of equipment I'm working on fixed (it is of my own design and development, and I am questioning my own judgement until I get answers) then I'm in pure cynic mode.

    At least Obama is projecting more believable unemployment numbers. Whether he comes up with any answers remains to be seen.

    He may not have any answers. Answers may not exist, Keynes et al besides.

  7. Hi Stephen,

    Actually, I think that the 8-8.5% unemployment rate seems about right. The rise in the unemployment rate is roughly equal (with a little algebra) to:

    – (growth rate of output) + (growth rate of productivity) + growth rate of the labor force).

    Simply put, in order for the unemployment rate to remain unchanged, output growth must be in line with the growth rate of productivity plus that of the labor force.

    Output and productivity tend to fall during recessions, and for argument’s sake let’s say by: -1.5% (-6% annualized) and -0.25% (-1% annualized) over the fourth quarter. It would not be unheard of to see the labor force decline by 0.5% in a recession quarter, so this this simplified example implies that the unemployment rate rises by 1.5% -0.25% - 0.5% = +0.75% in one quarter. That would put the average fourth quarter unemployment rate at around 6.75% (which is completely feasible given 6.5% and 6.7% Oct. and Nov. unemployment rates). Add another 0.75% for each quarter (grossly oversimplified) of the forecast, and one arrives at a 9% peak at the end of Q3 2009. This is just a simple example, and we could easily use viable numbers that would imply a peak unemployment rate in the 8-8.5% range, but to assume that kind of output drop (-6%) for several quarters to get into double digit unemployment is to me unbelievable.

    Each forecast (including the crude one I made) includes assumptions. I assumed that a government stimulus will occur, which gets me to 8-8.5% range. Obama’s assumption of double digits is likely no fiscal stimulus package and a failure of monetary policy (I imagine), which would certainly allow the economy to tumble quite far.


  8. Rebecca:

    Excellent answer! My mind is predisposed elsewhere, so I'm unable to comment on your methodology, albeit I might later when I can animate the equations. (I'm chronically dubious of Productivity claims, and an increase would increase GDP per capita, but *ostensibly* decrease employment)(your "assumptions" claim renders that moot, however, for the purpose of this example).

    So, assuming your methodology is sound...what then for Obama's projection? Is it a call to brace for the worst? (A standard trick of incoming politicians)

    The Great Depression showed unemployment reaching 25% in the US (considerably higher in some other advanced nations).

    To invert your logic, and as you allude to, then the Stimulus must be working!

    Again, I'm riven with skepticism and cynicism from my own work on the bench at this moment, but if unemployment in the US is able to remain in the 8% region...then I find other claims on the downturn matter to be fantastical.

    I'm not purposely deriding your claim, Rebecca, you may well be correct, but if so, other "assumptions" may prove totally false.

    I'd feel a whole lot better if those with the steering wheel could see through the windshield. I have serious doubts they can, or they wouldn't be all over the road and changing direction after assuring us that 'they' can see!

    Lol...I'll update this post when the equipment is fixed! I'm in the middle of a 'workaround' to isolate where I think the problem is, and then at least I can extrapolate your simple equation without the cynical bias I now have.

  9. Further to my previous post, best I update with this:
    [“Economists from across the political spectrum agree that if we don’t act swiftly and boldly, we could see a much deeper economic downturn that could lead to double digit unemployment and the American dream slipping further and further out of reach,” Mr. Obama said.

    Mr. Obama has vowed to “create or save” three million jobs over the next two years. In his address on Saturday, he omitted the word “save,” suggesting he would create three million jobs — a goal that many economists consider unattainable under current conditions.]

    Something is amiss....

  10. Nice expanation. First time I ever read thru an article on employment. Can't trust the #'s & can't use them in trading.