Absurdly Large BLS Revisions, part II

Our previous look at this Absurdly Large BLS Revision simply wasn’t satisfying; So I had to spend some additional time reviewing this data . 

Here is the official BLS statement:

"Each year, the Current Employment Statistics (CES) survey data are
benchmarked to comprehensive counts of employment for the month of March derived
from state unemployment insurance (UI) tax records that nearly all employers are
required to file. For national CES series, the annual benchmark revisions over
the last 10 years have averaged plus or minus two-tenths of one percent. The
preliminary estimate of the benchmark revision for March 2006 is +810,000 (0.6

BLS currently is researching possible sources for this larger-than-normal
expected benchmark revision. On initial review, the difference between the CES
sample-based estimates and the UI employment counts does not appear to be
concentrated in any one industry or geographic region."  –BLS Benchmarking Revision


"Researching possible sources" — let’s call that the bureaucratic understatement of the year. The BLS has a benchmark restatement that is 300% of its 10 year average, and they announce: "We are looking into this."

I am waiting to speak to a statistician at BLS (they appear to be closed on Columbus Day), but if we were to take this data at face value, than the economy is MUCH STRONGER than previously data would have us believe. From March 2005 through March 2006 NFP growth averaged 236,000 a month versus the past 2Qs of ~119,000.

If the Fed is data driven, and if they consider this BLS Benchmarking reliable, then the issue at hand is this: This would make the Goldilocks scenario far less likely, and it
would undercut the entire basis for the recent Bond market rally. Given this newfound economic strength, they should be back on their tightening campaign sometime soon.

Bill King is even blunter:

"For months the Fed has professed, some say groveled, that it is ‘data drive’. We, as you all know, have been questioning the veracity of government data for over a decade. We now know that some very important data that has been driving the Fed is wrong. The Fed has transacted policy off faulty data. Ergo, Bomber Ben’s decision to pause could be based on faulty data. And interest rates and US productivity could be based on erroneous data. Therefore investor and trading decisions for much of the past year and one half are based partially on faulty US economic data."

I am curious as to how BLS missed nearly a million jobs; Did the businesses who did all this hiring forget to report back to the CES ?

John Williams, of Shadowstats.com, reports that

"the September
employment report showed severe deterioration, despite a number of reporting
This is despite the comparative annual boosts starting
to show up in data from the effects of last year’s terrible hurricane season.
The heavily touted annual gain in September retail stores sales is a prime
example of such an effect."

I suspect that this revision may accidentally reveal a few economic and market truths:

1) Is the Fed really data-driven, or do they merely say that because it is more palatable than admitting this is all a seat of the pants exercise?

2) Has the equity market rally truly been based on a Goldilocks scenario, or has it been correctly anticipating a bold and robust economy?

3) Was the bond market anticipating a slow down? Given all this hiring and economic strength, we have to wonder how real their fears and concerns actually are.

 Something strange is afoot at the Circle K: Either the economy is much stronger than we previously believed, ending the Goldilocks scenario — or the data driven Fed is reliant on what we now know to be bad data.

Neither case presents an attractive outcome.

Print Friendly, PDF & Email

What's been said:

Discussions found on the web:
  1. Anon commented on Oct 9

    Not taking sides. Just wondering… has the much-maligned ADP survey been right all along?

  2. oo commented on Oct 9

    How do these revisions square with the Household Survey ???



    You mean the survey that includes FARM Payroll (as opposed to Non-Farm), family workers, uncompensated employees, and freelancers?

    Before assking that, you should read this: Redux: Household versus Establishment Surveys

  3. Lo commented on Oct 9

    Fed Funds pricing in 30% chance of cut now rather than 20%

  4. lurker commented on Oct 9

    Thank goodness the market is always right, and NEVER data driven. Shouldn’t fight the Fed, but worshipping their every utterance seems a mistake as well. I can think of several folks profiled in the Market Wizards books whose opinions on the market right now I would value at many multiples of anyone at the FED.

  5. mike commented on Oct 9

    OT – just a quick off-topic tidbit for Barry: that La Jolla house you posted several months ago has dropped it’s asking price from $24MM to $18MM, fwiw.

  6. cornerkick commented on Oct 9


    I’m a bit surprised by your statement that this would “end the Goldilocks scenario”. Correct me if I am wrong, but I thought your view was that we wouldn’t see a soft landing because the popping of the housing bubble would cause more damage than expected, sending the economy into a serious recession (and realization of this is what would send the Dow to 6800).

    If this data is correct and the economy is much stronger than anticipated, it would seem to indicate that the economy is in better shape to weather the ending of the housing bubble and that there is a better chance that the economy will either not fall into recession or that it will be much less severe if it does.

    Did you make that statement assuming that the higher growth would lead to more Fed hikes, which would then lead to a worse economy or do you think this increases the inflation risk? Since an economy in a deep recession would not seem to have much inflation risk, I don’t see how a fear of higher inflation (due to this data) fits with the housing bubble driven deep recession model.

    If one felt that the economy wasn’t going to go into recession, I could see how this new revision would drive nervousness about higher inflation, however. For that reason, I suspect the gold bugs will welcome it.

  7. S commented on Oct 9

    GDP declined from 5.6% in 1Q to 2.6% in 2Q and yet the labor market is tight??

    I’m confused.

  8. mm commented on Oct 9

    that 5.6% was due to Katrina crushing the 4Q GDP , so 1Q was pushed upwards …. average 4Q 2005 and 1Q 2006 for more normalized #

  9. wcw commented on Oct 9

    Yes, this could end the Goldilocks scenario, by making things too hot. You’re confusing what BR thinks with what the market is pricing in. If the market were pricing in a hard landing, prices would be behaving very differently than they are.

    The real question for me is whether the bond market and Bill Gross are right, or whether the stock market and James J. Cramer are instead. The former is trading as if Goldilocks and the soft landing were the blue-sky upside, the latter as if that same scenario were the gloomy-stormcloud downside. They may yet coincide, but it seems highly unlikely.

    For now, put me down in the bond-market camp, though given those revisions, I have been revisiting my models and expectations.

  10. TJR commented on Oct 9

    Of course the data is flawed, because the BLS methodology is flawed. They are measuring the economy based on how it has been in the (distant) past rather than how it is structured today. As they report on their site “The production and nonsupervisory worker hours and payroll data have become increasingly difficult to collect, because these categorizations are not meaningful to survey respondents. Many survey respondents report that it is not possible to tabulate their payroll records based on the production/nonsupervisory definitions.” Small samples and poor data collection methods result in faulty reports. This is classic bureaucratic GIGO (Garbage in, garbage out).
    BLS needs to reconcile the CES methodology to reflect real employment and pay in today’s economy. This includes incorporating data that is now reflected in the household survey, ADP etc.

  11. wcw commented on Oct 9

    Oh, and PS — seen the homebuilders today? Up 3%, with bonds also up (if only slightly).

    It’s improbable that both are right.

  12. Paul Jones commented on Oct 9

    Is it possible that they are just cooking the books before a big election?

  13. fred hooper commented on Oct 9

    “BLS needs to reconcile the CES methodology to reflect real employment and pay in today’s economy”
    Doing so might open a political can of worms if questions such as the following were to be answered honestly: Do employment and productivity stats include the wildcard of 12-20 million illegal immigrants in the workforce? How much has American wage inflation been subdued due to the deflationary effect of cheap labor from South America? Does the trade deficit include $20B cash sent by laborers to Mexico each year? How much money leaves the country in exchange for imported illegal drugs? Any trust in government statistics is misplaced.

  14. Cherry commented on Oct 9

    You have to remember, just announcing such a huge revision is as much political as data driven.

    I doubt that large of revision stands, even ADP wasn’t that high.

  15. Michael C. commented on Oct 9

    Where are Cramer’s pompoms?

    Need to see them to trigger that sell signal.



    Here: A chest pounding editorial in the WSJ

    The Consumer Lives

  16. KirkH commented on Oct 9

    “GDP declined from 5.6% in 1Q to 2.6% in 2Q and yet the labor market is tight??”

    Here’s a theory. We’re transitioning from an economy based on consumer spending to a meritocracy. You now actually have to be skilled to be productive (and highly paid) and there is a shortage of talent. This transition also means that consumers, who mainly do repetative tasks ripe for automation, are starting to suffer (see median wages). Productivity is now a threat to wealth equality and it’s accelerating. I don’t agree with much Jeremy Rifkin writes about in “The End of Work” but the data is a bit scary.

    Throw in global wage arbitrage and the deflationary effects of accelerating technology and you have one weird bowl of soup.

  17. David Yaseen commented on Oct 9

    Not buying the revision for a second until I hear where the numbers came from. The admin has been desperately seeking good jobs numbers forever, and suddenly they find nearly a million less than a month before the election? I mean, really.

  18. TDM commented on Oct 9

    I think we need to start looking at better ways to do surveys. When we count all the jobs from all the unemployment insurance tax records we count 810K more jobs than the employment surveys predict. Also, wage surveys are showing that wage growth has been stagnant but when we count all the income from all the tax receipts, counting all monetary compensation up to the 90k SS cutoff and only the Medicare part above the cutoff, “Contributions for government social insurance” are up over 9% for the year (BEA, through Q2). Payroll taxes have been increasing faster than GDP over the past 3 years and is pushing the SS surplus and the on-budget deficit up significantly. We need to look at why wage survey data is not showing the massive increase in compensation found from counting all tax returns.

  19. KirkH commented on Oct 9

    From Bloomberg:

    Oct. 9 (Bloomberg) — Columbia University’s Edmund S. Phelps won the Nobel Prize in economic sciences for his theories on keeping inflation stable in an economy by limiting the expectations of price increases…
    The winner gets 10 million Swedish kronor ($1.37 million) which is roughly enough for a down payment on a modest one bedroom condo in downtown San Diego.

    OK, that last bit is mine. I imagine the aptly named Fed would love to keep that guy fat and happy considering BR’s last post.

  20. wcw commented on Oct 9

    In re meritocracy, pleasee see this. The comments thread is long, but here is a good one I didn’t write: “1. The competition for talent leads to increased inequality. 2. We are experiencing increased inequality. 3. Therefore, we are seeing a competition for talent, and 4. Therefore, we live in a meritocracy. – which is obviously full of holes.”

    As for consumption, please see my chart of consumption as proportion of GDP and explain how that congrues with your hypothesis.

  21. RP commented on Oct 9

    huh? “but if we were to take this data at face value, than the economy is MUCH STRONGER than previously data would have us believe”

    This doesn’t change GDP, does it?
    If not, it sounds to me like more people were working,
    but for less money than originally thought and with
    less productivity. Why does that imply tightening?

Posted Under