Submitted by Jesse of Le Café Américain
Orwellian use of government economic statistics, par excellence.
The moving average of the Non-farm Payrolls marked the downturn in the economic expansion with amazing clarity by a steep drop in late 2007. It will also mark the bottom and a sustained upturn when it arrives.
30 comments
Comments are closed.
Along with Naked Capitalism, Jesse’s also got a great site. Shared traits of thoughtful, relevant, thought provoking, careful analysis… could go on and on. Kudos to both.
The 3.7% cumulative job loss so far in this cycle makes this recession the second worst in 50 years for U.S. labour markets. Interestingly, there were no revisions to the February numbers, however, January’s estimate was revised significantly lower to –741,000 from –655,000. While unemployment is a lagging indicator, don’t expect a strong recovery. This recovery will be long and it will give new meaning to the phrase the “jobless recovery”.
Thanks Jesse, I added your blog to my blog roll.
cheers,
Leo
“The problem is that, until reforms are made in the system, all this accomplishes is to encourage financial speculation.”
That’s not so much of a problem, as it is a goal. Who wants a bull market more than life itself? Who has captured the Congress, the President, and the Treasury?
Great stuff Jesse. Yves, Ed Harrison and your self are undoubtedly the best, most consistent objective opinions and discussions in the media – of any form. Ed Harrison had an interesting piece on government statistical manipulation where the 40 000 new home starts magically turned into 500 000+. Imagine what manipulation will now occur with the change in the mark-to-market rules?
Is this a serious post?
Colin Powell, and his son, Michael, both harmed themselves by going with the status quo.
At least Hank Paulson had the guts to state that, what he was he was about to do (buying failed assets at above market prices) was illegal and he wanted immunity.
His protege has no compunction.
It is time for the bondholders to hold the steaming bag.
I too enjoy Jesse’s site (and the food pictures).
former bank regulator William Black is stone cold smokin’ on Bill Moyers Journal tonight. He calls Geithner, et al, out. A must see. I suggest a link to this incredible interview.
Agreed, Laborboy, I just watched it… William Black doesn't pull any punches and essentially accuses Geithner of being part of the "coverup" to hide the problem assets and all the fraud.
Here's the transcript & video:
http://www.pbs.org/moyers/journal/04032009/watch.html
Bill Moyers interview with William K. Black
http://www.pbs.org/moyers/journal/04032009/watch.html
Agreed, W. Black laid it all out.
For all the taxpayer monies being paid out, you would think we could at least receive justice in return but the con job continues.
Not happy with the reporting here just regurgitated numbers the government puts out. It’s a disservice treating us like we are ignorant.
No offence, but this is retarded. Why on earth you’d plot seasonally adjusted series along with not seasonally adjusted one, and make some long-stretched conclusions?
Since allegations have been put out here recently according to which the data releases by the government statisticians are deliberately forged to manipulate the public. What do I have to imagine how these evil things are actually done?
Is there a secret government committee that regularly meets before the data releases to decide how the forgery will be done this time?
Or is there a secret order by the president to the heads of the government apparatus to manipulate the data in a certain way?
Does anyone who makes such accusations here that there is a conspiracy, care do explain how this conspiracy is actually carried out?
Perhaps it’s not much more than some dull government apparatus that uses standardized statistical models because they have done it in this way always (or at least for a long time). And the inertia of the apparatus is very large before any change is made, when the models are not appropriate for reality.
Hanlon’s razor: Never attribute to malice that which can be adequately explained by stupidity.
rc
rootless,
With all due respect, in every company I’ve been in, rules and procedures were tweaked in questionable ways to yield what the higher ups wanted. That’s hardly conspiracy, that’s the way most organizations work. So in my view, the issue is not whether this sort of thing happens, but whether it is becoming more clumsy or brazen.
This via e-mail from economist Gonzalo:
Yves, remember a few weeks ago, an e-conversation we had about U-3 and U-6 unemployment numbers, and the perennial “revisions” to previous months’ reports that were quietly chalked up, after the initial number proved to be off the mark by something like 25%?
Well check out the Bureau of Lying with Statistics’ summary for this past month:
http://www.bls.gov/news.release/empsit.nr0.htm
Near the bottom, the BLS is revising January numbers, from 655,000 to 741,000. Now remember, the January number, when reported in early February, had been 598,000 new non-farm unemployed. Scary, sure, but less than the December ’08 number; even when revised in early March, the January number was still less than the December number, which would lead to the reasonable inference that job losses were slowing down.
But the under-reported 741,000 revision means that January unemployment accelerated.
Why is this relevant? Two reasons:
Number one, historically, unemployment growth seems to slow and then peaks out as the economy touches bottom of a recession. In past recessions, unemployment growth stops and reverses course just as the economy picks up again.
Number two, the BLS is seemingly (and deliberately ) hiding the fact that unemployment growth is INCREASING month to month. Forget about the business birth/death model the BLS uses, which is a Madoff-like black box: Even if we take their numbers at their word, non-farm unemployment is accelerating rapidly. Check out that Floyd Norris blog post I sent you, which has a handy listing of the previous BLS figures, and then the later adjustments:
http://norris.blogs.nytimes.com/2009/03/06/will-job-numbers-keep-being-revised-down/
Scratch out the January number and insert the 741,000 figure: What does that tell you? Apart from the likely fact that the BLS is buckling under pressure to spritz up the monthly report, unemployment growth is accelerating rapidly, and there is clearly no end in sight.
I believe there is no way the U-3 number for June ’09 will be less than 10%. The U-6 number will be 18% at least, 20% by year’s end—assuming, of course, that the Bureau of Lying with Statistics actually reports honest numbers.
Sorry for the downer,
Yves – I have followed 4 blogs above all others, including NC, Credit Writedowns & Jesse, where it's nice to see responses for a change. I'm flattered beyond belief that my taste in blogs is reflected in your choice of contributors. The 4th blog that I find increasingly thought-provoking is Mish – so contibutions from him would not just make my day, they'd make my decade.
And what exactly is the proof now for the allegation that the data, like the payroll data as stated in the posting, are deliberately manipulated? To state almost each organization works in the way that the employees do what the bosses want is a bit vague.
The downward revision of the payroll data in recent months isn’t any proof for deliberate forgery either, except, of course, in the eyes of the ones who look for confirmation of what they already have known, since it is obvious anyway and they were open minded. The thinking seems to be a bit circular here: Since there is an evil intention, the data are manipulated. And the downward revision shows, in turn, that the data were manipulated and there was an evil intention behind it. Well, it’s not obvious to me. Apparently, I am not open mined enough for this way of thinking.
For instance, it could be a sampling problem during the sharp economic downturn. How do you exclude this possibility? Why are these later revisions done at all, if all of this is manipulated anyway? Wouldn’t the evil doers want to maintain the manipulation of the public until the end then?
And why did they want to manipulate the employment change downward in the first data release for Nov 2005, or for Aug and Sep 2006, or on average from the year 2003 to 2007 so that they had to revise the data upward later?
http://www.bls.gov/web/cesnaicsrev.htm
rc
rootless cosmopolitan said – “Does anyone who makes such accusations here that there is a conspiracy, care do explain how this conspiracy is actually carried out?
Perhaps it’s not much more than some dull government apparatus that uses standardized statistical models because they have done it in this way always (or at least for a long time). And the inertia of the apparatus is very large before any change is made, when the models are not appropriate for reality.
Hanlon’s razor: Never attribute to malice that which can be adequately explained by stupidity.”
Rootless, excellent comment, and it is well past time your good questions were given a fair hearing and answered. Not just about labor statistics, but about the entire ‘financial’ crisis which I believe is just another intentional manipulation of the common folks by the puppet’s of the wealthy ruling elite, all meant to consolidate power and eliminate the expensive and high resource consuming middle class.
Why must it be broadened beyond just labor statistics? Because, context is pertinent to those labor statistics and they were produced in a greater context of a nation that has had its government bought and paid for by the ruling elite so many times over so many generations that the ‘rule of law’ and government now stand as an unresponsive to the people scam.
Unfortunately, given that scamerican system of corrupt ‘rule of law’, your questions would never even be allowed into a court room. You would first be denied standing, and then if you were successful in gaining standing, most all of the information germane to your case would be disallowed during the discovery period. Tsk. Tsk, Sooooo sorry rootless …
But there is a court in this world that still functions with some degree of integrity. It is called the court of public opinion. I suggest your questions be used to stimulate the organization of that people’s world court of public opinion on the internet and try the individuals responsible for causing this current financial crisis. Using the common criminal law criteria of means, motive, and opportunity, as a guide, I would start with putting pivotal figure Alan Greenspan on trial.
Let the blog owners and the regular readers of the good financial blogs mentioned in this thread draw up the allegations and indictment and present the evidence. Invite the comments of people with political persuasions excluded by the scam electoral process. Exclude all of the wealthy elite’s mainstream media. After all, fairness is the goal here …
Sure Mr. Greenspan won’t show up, but he can be tried in absentia. With the wealthy elite’s mainstream media excluded it would be interesting to see who would speak in his defense.
Start now, feel the power — organize the ‘People’s World Court of Public Opinion’ … let the fair and open internet trial of Alan Greenspan begin.
Let’s answer the questions and apply the appropriate penalty. Does he get a slap on the wrist for being a simple stupid man? Or does he get death by hanging for the treason of conspiring with cronies to consolidate their power and thereby cause the deaths and hardship of millions of the nation’s and world’s citizens?
i on the ball’s razor: Deception is the strongest political force on the planet.
i on the ball patriot
There is an interesting Air Force brief that details their Internet/digital/cyber strategy that I believe was covered by Noah Shachtmann in his Wired’s Danger Room posts.
Why is this relevant? In it the Air Force suggests it must do a better job of responding to adverse opinion and ideas especially in pertinent communities and mil bloggers. The rootless comments, while constructive and well reasoned, resonate in this light. It is perfectly consistent with the White house who dismissed blogs then holding a call with them to discuss PPIP. So when comments such as this appear seem a little too well reasoned.lawyerly, useful to remember that a comprehensive media strategy includes defense too.
http://blog.wired.com/defense/2009/01/usaf-blog-respo.html
RC meet Orwell; Orwell meet RC
RC
You smoke some good stuff and I applaud your faith in the Government. After all the government always works for the good of the people hence they repealed Glass Steagel – who needed the leverage limits? Why we should all be allowed to lever up 40:1, Robert Rubin earned 115 million from citigroup – an outstanding and profitable institution after finishing as treasury sec under clinton and ofcourse Geithner’s plan favors the taxpayer over the banks for sure.
Ofcourse the BLS can’t fudge the data deliberately, just like Bear and Lehman’s balance sheets were OH so healthy and AIG is still AAA, heck i’ll add one more A to how stable it still is.
Carry on living in utopia sir. Denial is pleasant and Ignorance bliss.
I’m more inclined to see this as government incompetence. The continued use of the birth/death model, which adds jobs at the same rate as in a boom time, even when we are losing 600,000 jobs a month, is pretty convincing evidence. (No one in their right mind would do this, but the model says it must be, so let’s use the model, not our right mind)
Of course they have to adjust the numbers down. The birth/model sets them too high, so when more complete data comes in what is obvious to us in advance only becomes obvious to them then.
No one in government is going to do things differently than in the past, unless forced to do so, especially if that means painting a darker picture. So, initial numbers will continue to be meaningless in anything other than a steady state situation.
Yves, would you please get someone who actually understands the birth/death model to comment? I have seen so much misinterpretation of this adjustment, not only here but at The Big Picture.
If you actually bother to read the BLS methodology, you would discover that the B/D adjustment has two components: trend and seasonal. The trend component gets picked up in the sample. The seasonal component must be done separately. The graph shows the seasonal adjustment, which historical analysis shows is pretty consistent whether we’re in a recession or expansion.
Last year’s nonfarm estimates (the revised monthly) were pretty accurate. Without the B/D adjustment, they would have been substantially off.
Second point, for all you paranoids: the monthly numbers get revised because there are businesses in the sample that report late. In these times, late reporters are probably biased downward (didn’t have time to get the report in because we were too busy laying people off). But there’s no way to model that kind of thing. BLS has two choices: stick with its procedures, or introduce analyst judgement into the process. You really don’t want the latter (which is apparently what you think you have).
BLS, by the way, is beating up on state employment departments because the state estimates, as bad as they are, are too rosy. The sum of the states is nowhere near as bad as the national total. Why? Because the states allows substantially more analyst judgement to enter into the process (a smaller sample means more judgement calls about whether a large change in an individual firm’s employment is “typical” or “atypical”, with the latter pulled out of the overall trend).
I could go on. I’ve been estimating employment at the county level for a state employment department for 20 years. We have no reason to make the numbers look better or worse, we want them to be accurate.
Ironically, discussion like this may be responsible for the problem. Official statisticians are so cautious about methodological complexity and change that they tend to use established, standard techniques until the case for change is overwhelming. I once produced seasonal adjustments for official statistics, and was often frustrated at the difficulty of modifying the methods. I used to say that I had a better idea of the true picture of the data than any policy maker, because I knew the weaknesses of the seasonal adjustment methods, and in some cases, even had my own research versions that I considered did a better job than the methods that I was oliged to use. Official statisticians deserve more respect.
It’s intellectually dishonest for people to still complain about the B/D model without offering any analysis. We don’t have to speculate on the impact of the B/D mode. It’s been in effect long enough that we can now gauge it’s effectiveness.
The Birth Death model IMPROVES the accuracy of the monthly employment numbers. See here for the results.
You should be suspicious of anyone that continues to rail on the B/D model and question their motives for doing so. Or at least their intelligence.
Steve,
Your last paragraph is pure ad hominem, which raises questions about your integrity and motives.
In addition, there were many instances in 2007 and 2008 (noted on this blog) where the BD model claimed large numbers of establishments added in construction, when there was ample evidence construction was contracting across the board. Issues like that do raise questions about the accuracy of that exercise.
rootless,
An only partial list of games playing I have seen:
Just about every merger model massaged ad naseum to make the deal “work” no matter how far short it fell when first constructed using realistic assumptions
Revenues and expenses fooled with to make official reports look better (all sorts of games here, expenses received literally held back from being input, to more sophisticated methods. I have seen accountants put their names on flagrantly phony numbers)
Client reports showing less than credible combinations of market growth and price trends to support investment/business line additions that management was keen to pursue
Is that specific enough? I am not about to name names (n terms of specific deals and individuals), but I have seen this sort of behavior in every company I have worked in, as well as in their interactions with clients
The Birth Death Model used to be a minor adjustment in BLS NFP data. It has been around for many years. Since ’06/07, it has become a major statistical nuisance.
The theory behind it was that small businesses are under represented in the Establishment survey. Much of the job creation comes from these smaller firms. The 2001 changes (effective 2003) BLS significantly modified the B/D adjustment. They went from a minor statistical adj to overweighted.
There were two methodological problems with this adjustment: 1) It took a primarily MEASURED data series and introduced a significant amount of extrapolated/modeled data into it (deducing new NFP job from new state Incs).
2) The model improved accuracy somewhat in the beginning of the economic cycle, but at the cost of making BLS data much less accurate at the end of the economic cycle.
Example: In 2007, about 75% of the new jobs created were due to the B/D adjustment.
~~~
Since someone challenged my integrity about the motivations for my B/D adjustment criticism, allow me to point this out: I run money, and don’t care about the ideological or political battles between economic theories or politicians. I only want to protect my clients capital and generate positive returns when opportunities arise.
I cannot say the same of the people who were the staunchest defenders of the BD adjustment.
Indeed, many of the people who took the BLS data at face value, who did not delve beneath the headlines, and approached such data interpretation in a no-critical fashion got mangled in the markets.
Those “analysts” who did not skeptically approach BLS inflation data, NFP jobs info, B/D adjustment — were way too bullish going into 2008. They believed the official data as well as the spin, if you followed the advice of less critical thinkers, you lost tons of money . . .
I’m with Yves. The idea that this can all be attributed to stupidity rather than a system evolving to one in which incentives spur “creative” use of statistics does not coexist well with my 20 years of experience in real estate finance. If these people are stupid, then over 90% of the earth’s population is stupid. Smart and stupid lose their power as relative adjectives in that scenario.
Hanlon’s razor is cute, but ultimately facile.
Viv, you have put forward some nice strawman arguments as well as some good ad hominem against my person. You also seem to have jumped quickly to assumptions about me w/o even knowing me. Unfortunately for you, none of this rebuts any of the points I made regarding the actual subject.
S, every good conspiracy theorist knows that facts or counter-arguments that are in contradiction to the conspiracy theory are part of the deception by the conspirators. I am afraid you have exposed me and any denial by me will be futile. Additional well-reasoned arguments from my side will only provide more evidence against me being with the dark forces.
Yves, you may have experienced a lot of bad stuff in the corporate world. I don’t question that. However, it’s a logical fallacy to take this as prove for the truth of the claims made in Jesse’s posting.
Please, could someone finally explain to me what the hard evidence is supposed to be for the claim made by Jesse according to which the January number of the Monthly-Non-Farm Payroll changes were an “Orwellian use of government statistics” and revised downward, because “the BLS statistics are being heavily manipulated by the government to try and manage public perceptions”? I really don’t know how you guys find out these things, besides from your gut-feelings.
Anonymous, April 6, 2009, 12:46PM, true, Hanlon’s razor is not the deep analysis. I didn’t bring it for that reason and I have made some more concrete points. But neither is the claim made in Jesse’s posting according to which something were “obvious” that isn’t. And certainly, it never has been my intention to say that “all be attributed” to stupidity. However, I see the tendency in these kind of blogs, even more so in the comments, to jump to explanations that heavily rely on evil manipulation, collusion or even conspiracy theories based on accusations w/o hard evidence, rumors and emotion, even when other explanations are analytically reasonable and plausible. It’s a capitulation of analytical thinking in my eyes. Many of the posting in Yves’ blog are quite analytical and I like that. There is a lot I have learned from Yves’ postings. This guest posting wasn’t, though. This one fits more in the other category.
rc
@rc,
Hay it’s hard to reconcile with the possibility of gross fraud in the elected government. To name names and places could bring damage to individuals unintentionally (whoops un-detonated cluster mine) sorry kids, my bad.
Join the military and get your self a cryptic security clearance, then you’ll see after a few years of observation all kinds of silly stuff.
Trends in bad info have a meaning, you just have to decide which way the wind blows.
skippy