Former GE CEO Jack Welch recently ignited a firestorm over his ‘tweet’ that the US September labour market data appeared to have been manipulated for political reasons. Indeed, there were some curiosities in that particular report, but they hardly end there. The fact is, much official US data are misleading in some way, if not manipulated. Don’t believe me? Well, rather than label me a ‘conspiracy theorist’, why don’t you take a look at the evidence? What you see might surprise you.
The Use, Misuse And Abuse Of Statistics
“Unbelievable jobs numbers…these Chicago guys will do anything…can’t debate so change numbers.” So read the ‘tweet’ that the media immediately labeled a ‘conspiracy theory’, leading in short order to former GE CEO Jack Welch’s resignation as business correspondent from Reuters News.
Publicily accusing the US president or his advisers of deliberately fudging economic data for political purposes is a serious act, one certain to draw a response from the media. But as is so often the case, the response this time was primarily to remark at how outrageous the claim supposedly was, rather than to examine the evidence in any detail. This was as true on the ‘right’ side of the media spectrum as on the ‘left’. Take, for example, this response from the prominent conservative ‘think-tank’, the American Enterprise Institute (AEI):
There’s no interference with the Bureau of Labor Statistics numbers. The White House didn’t lean on the BLS and influence them to kick the unemployment rate below 8 percent. That talk should be confined to crazytown.[1]
OK, let’s suppose the AEI knows how things work in Washington. But then consider the following excerpt from the same article:
This month’s household survey is puzzling when viewed in light of the payroll survey and indicators for the broader economy. The headline unemployment rate fell from 8.1 percent to 7.8 percent, despite the fact that the labor force increased by 418,000. There were a whopping 456,000 fewer unemployed this month than last, and an incredible 873,000 more employed. The employment-to-population ration edged up by a solid 0.4 percentage points.
Ah, so the data were ‘puzzling’ and ‘incredible’—not believable. So perhaps the only way to reconcile the two excerpts above is to draw the conclusion that, while the AEI apparently knows that the data were not deliberately manipulated, they are, sadly, some combination of inaccurate and unreliable rubbish.
Am I being too harsh? Well, much US labour market data are not particularly ‘hard’ but based on surveys, estimates and models rather than on real numbers. Most of the data are published in incomplete form and subject to large revisions, both in the following months and in the annual, so-called ‘benchmark’ revisions. Indeed, the final estimates of labour market data, when they become available over a year later, frequently show quite a different picture than the original flash estimates.
In this instance, I doubt the AEI or other keen observers would be surprised in the event that the September 2012 headline unemployment rate is eventually revised back up to 8% or higher. It makes one wonder just what all the fuss is really about or why financial markets tend to respond with such vigour to flash labour market data headlines. Perhaps the market-markers who benefit from meaningless volatility actively encourage their casino clients to roll the dice on the first Friday of the month. They do, after all, employ legions of expensive economists to forecast these somewhat random numbers.
Leaving data veracity or reliability aside for the moment, lost in most of the discussion around Mr Welch’s tweet is the background of the current labour market situation in the US. It is not good. Keep in mind that the ‘headline’ unemployment rate, even if accurate, does not give a remotely full picture. This is because it is a narrow measure of unemployment, excluding ‘underemployed’ part-time workers, so-called ‘discouraged workers’ not seeking work, or those who have left the workforce entirely. Common sense tells us that some portion of part-time workers would in fact prefer full-time work; that some ‘discouraged workers’ would jump at a job if offered up to them; that some who leave the workforce do so because they don’t consider themselves employable, but would seek jobs if they changed their minds.
Well, the ‘U6’ unemployment rate includes part-time workers looking for full-time work and discouraged workers. In September, this rate was 14.7%, nearly double the headline rate and one representing a far larger portion of the workforce. And what of the workforce itself? Well, according to the BLS, the workforce increased by 418,000 last month.
That sounds a rather large number. But let’s place it in context. Over the past two years, rather than growing with the population, the workforce has in fact shrunk as an unprecedented number of workers have stopped seeking work. Since 2008, the participation rate has declined from over 66% to under 64%, the lowest level since the early 1980s, when a far smaller number of women worked full-time. (It is possible that some portion of these folks do in fact continue to work in an informal capacity, say for cash-in-hand, a simple way to avoid payroll and other taxes that make low-paid work uneconomic.)
Finally, let’s consider some ‘hard’ economic data that should have a strong relationship with the state of the labour market. Regardless of whether people are unemployed, or only employed part-time, or simply work odd-jobs for cash in hand, presumably the usage of food stamps—government-supplied coupons redeemable only for food—should correlate well to the general health of the labour market. Food stamp usage data are not estimated and subsequently substantially revised the way labour market data are. It is a real figure representing a real budget item. When jobs are scarce, food stamp usage should rise, and vice-versa.
Well, guess what? Even though the headline unemployment rate has declined somewhat over the past two years, from 10% to under 8%, food stamp (SNAP) usage has soared, from 32mn to 47mn, a record high, and over 15% of the US population! Yes, the official headline unemployment rate might get more attention, but does it really tell us much about the general availability of living wages, or rather the lack thereof, for US households? Have a look:
Does Headline Unemployment Show A Misleading Picture Of The Economy?
Source: US Bureau of Labor Statistics; US Department of Agriculture
Let’s return for a moment to Mr Welch’s tweet: Whether such a narrow, unreliable statistic is ‘manipulated’ or not is borderline irrelevant. A look at a broad data set shows that labour market conditions are far worse than the headline unemployment rate implies. But what if, in fact, the bulk of headline US economic data were, in fact, also somewhat misleading, if not necessarily manipulated? Should we be confident that we are getting a full, accurate picture of the real state of the economy?
Some educated, experienced observers think not. I have previously cited the work of economist John Williams, who at www.shadowstats.com maintains a database of alternative methodologies for calculating not just US unemployment, but also consumer price inflation (CPI) and gross domestic product (GDP). He is not just making these data up. His methodologies are in fact based on how various US government agencies used to calculate the data in years past, before these methodologies were modified in some way. Here, for example, is what the CPI has looked like in recent years based on the old methodology:
Is Price Inflation Actually Nearly 10%?
And Is Unemployment Over 20%?
And here is a look at how things in the labour market look today using the old methodology for calculating the broad unemployment rate:
If official inflation is understated, this implies that real GDP growth is overstated, that is, much of the ‘growth’ being measured is merely ‘nominal’—price inflation—rather than ‘real’. (I elaborate on why I think GDP is a poor measure of growth in my book, The Golden Revolution. Follow it—and me—on Twitter.)
Readers take note: The way in which the government has changed the calculation of official economic data through the years has always, in each and every instance, made the US economic situation appear healthier than when using the previous, ‘inferior’ methodology. Coincidence? Decide for yourself. But be careful what you tweet, lest you be labelled a statistical ‘conspiracy theorist’.
John Butler is co-founder of Atom Capital, an FSA-regulated, London-based fund manager. In addition to managing the Amphora Commodities Alpha Fund, Atom Capital oversees a diversified, multistrategy investment platform and provides associated wealth management and consulting services for professional investors in the UK, Europe, and internationally.
Did one large state not submit its unemployment claims in time to get counted or not? The BOLS stated last week that one large state missed the deadline.
No rational person would dare use the same statistics to compare the US economy of the 1970s/1980s with the US economy of today.
Given technology/outsourcing, national borders are no impediment to finding workers. The ‘US labor force’ includes hundreds of millions in China, India, Mexico and elsewhere. In my area of expertise, banking, it is routine to consider using employees in either Bangalore or Charlotte.
A bank can increase profits, its share price, and GDP by cutting jobs in Boston and transferring the work to Bangalore. Locating the job in Boston may actually reduce the profit margin to a level that bank no longer considers attractive.
Yet, our brain trust of the likes of Jack Welch (Professional GOOPER WHORE), can’t see this distinction.
First, why do they (the Government) have to use survey’s instead of hard numbers? I am self employed, and have to pay my taxes monthly. Why not incorporate those tax receipts. The companies filing bankruptcies, likewise are publicly known, easy to incorporate their data. Then the folks who elect to retire before social security eligibility (they have enough resources). Then there are “tax collections” which are accumulated monthly. I am reasonably certain that the numbers are indeed “manipulated”, and that things are far worse then what is reported. The stats might just as well tie in to all the welfare programs, as well as “government hires” for distribution of “welfare”, as they might just as well be unemployed.
GOOD WORK HERE AND RIGHT ON
Here is my take on what is going on —
ISSUED REPORTS:
On September 27, 2012 the Bureau of Economic Analysis (BEA) issued its final estimate of 2012 second quarter growth in GDP, which was a disappointing 1.3% down from the previous estimate of 1.7% which also wasn’t good.
On October 5, 2012 the Bureau of Labor Statistics (BLS) issued a jobs report for September that was unbelievable since it showed that there were 114,000 jobs created but employment when up by 873,000 and unemployment went down by only 456,000.
On October 4, 2012 the Department of the Treasury issued the Monthly Statement of the Public Debt of the United States showing a very surprisingly small increase of only $50.5 billion when the past 12 months had averaged over $100.0 billion per month.
On October 12, 2012 the Department of the Treasury issued the Final Monthly Treasury Statement of receipts and outlays of the United States Government for fiscal year 2012 showing a surprising low increase in the National Debt of only $1.09 trillion.
On October 15, 20012 in only 15 days, the Treasury shows that the National Debt is now $124.7 billion more then it was on September 30, 2012 at a staggering $16,192 billion or $16.2 trillion very close to the public debt limit.
On October 16, 2012 the Bureau of Labor Statistics (BLS) issued the Consumer Price Index (CPI) for September showing a surprising two month trend of the CPI increasing at an annual rate of 7.2% — the QE inflation may be here now!
REPORTS TO BE ISSUED PRIOR TO November 6th Election Day:
On October 26, 2912 the BEA will issue the 2012 third quarter growth in GDP estimate.
On November 2, 2012 the BLS will issue the jobs report for October 2012.
From a financial analysis point of view based on the latest series of government reports it would appear to me that someone is cooking the books. When I was a corporate trouble shooter I used the financial statements to find where problems were. That skill can be used in looking at the government statements to find what is really going on verses what may be claimed. Therefore it is my professional opinion that the reports that have been issued during October have been slanted to show that economic conditions are better then they are in an attempt to influence the election. If I am correct in this opinion the following will be true before the election:
On October 26, 2912 the BEA will issue its first estimate of 2012 third quarter growth in GDP probably showing a growth of over 2.0%
On November 2, 2012 the BLS will issue the jobs report for October 2012 probably showing 200,000 jobs created and unemployment dropping to 7.5%.
Clearly the administration is setting the stage for a major media blitz proclaiming the Obama agenda has worked — so give me, Obama, 4 more years to finish my work.
PRESIDENT HOUDINI THE EMPLOYMENT MAGICIAN
(http://www.chrissstreetandcompany.com/2012/10/president-houdini-employment-magician/)
All great magicians employ misdirection to create miraculous illusions, leaving their audiences stunned and confused. President “Houdini’s” miraculous creation of 873,000 jobs in the month of September magically drove down the national unemployment rate from 8.1% to 7.8%, just as absentee voting begins. The trick left Republicans and economists equally befuddled, since the last time a similar monthly jobs increase appeared was June of 1983 when the economy was growing at an astronomical 9.3% annual rate. But looking behind Barack the Great’s smoke and mirrors reveals that the President’s highly controversial July suspension of the “workfare” requirements that welfare recipients must actually do real work to be counted as employed seems to have dramatically reduced the U.S. Labor Department’s unemployment rate.
The Bureau of Labor Statistics publishes two monthly surveys that measure employment levels and trends: the Current Population Survey (CPS), also known as the “household survey”, and the Current Employment Statistics (CES) survey, also known as the “payroll survey”. With most economists estimating the current U.S. economic growth at an anemic 1.5%, it seemed ludicrous that the Obama Administration could report a monthly gain of 873,000 jobs in the household survey, just short of the all-time record 900,000 jobs gain in June of 1983, when under President Ronald Reagan the economy was growing 6 times faster at 9.3%. In fact, the same household survey report showed a recessionary slide of 195,000 jobs in July and another 119,000 decline in August. Adding to the suspicions regarding the credibility of the household survey, the more reliable “payroll survey” that tracks the rate of jobs growth through IRS withholding data was unchanged from last month.
After the report, the internet “blew-up” with conspiracy theories that the employment numbers must have been consciously manipulated by the U.S. Labor Department to help the President’s reelection odds. Even the highly-respected former CEO of General Electric, Jack Welch, tweeted: “Unbelievable jobs numbers..these Chicago guys will do anything..can’t debate so change the numbers.”
Like all spell-binding illusions, the real set-up for this phenomenally great employment report was engineered back on July 12, 2012, when the Obama Administration announced an Executive Order that eliminated President Bill Clinton’s highly-praised workfare reform that required welfare beneficiaries to get real job in order to continue to receive payments. According to former Clinton Advisor Dick Morris, in 1996 Senate Republican Majority Leader Trent Lott stated: “I don’t want anyone going to a truck drivers’ school that advertises on a matchbook cover and avoiding work.” The Republicans included 42 U.S.C. § 615(a)(2)(B) in the Temporary Assistance for Needy Families (TANF) reform legislation to make sure every states’ welfare recipients were required to work in a real job and also inserted section 607 to prevent future secretaries of Health & Human Services (HHS) from waiving the real workfare requirement.
With 1.4 million of the two million families receiving TANF payments not actually in real jobs, Obama took heat for changing the work rules. The Heritage Foundation warned:
“in the past, state bureaucrats have attempted to define activities such as hula dancing, attending Weight Watchers, and bed rest as ‘work.’ These dodges were blocked by the federal work standards. Now that the Obama administration has abolished those standards, we can expect ‘work’ in the TANF [welfare] program to mean anything but work.”
Presidential candidate Mitt Romney, a former governor, howled “the linkage of work and welfare is essential to prevent welfare from becoming a way of life.” Republican Congressional leaders screamed the waiver was a “blatant violation of the law” by allowing states to substitute “vocational educational training or job search/readiness programs” to “count as well” in meeting the work requirements.
But as the heat dissipated and the campaign news cycle moved on, President Houdini was positioned to triumph. No one knows just how many of the up to 1.4 million TANF welfare recipients have now been re-designated by the states government as “employed”, but isn’t it just magical how 873,000 people started working last month. Is Barack Obama really the new Ronald Reagan?