Following the release last Friday of the Bureau of Labor Statistics' employment report for December, in which the reported job gains for the month were 292,000 versus the consensus expectation of 200,000, the financial media championed, challenged or ignored the report.
Of these three, the easiest to do is champion it. It requires no critical analysis and it's what most people want to hear anyway. There are numerous examples of writers expressing this attitude, so I won't bother with a URL as a reference point.
The next easiest response is to ignore it. The ignore responses provided by credible sources are best exhibited, in my opinion, by The Economic Cycle Research Institute (ECRI) and The Levy Economics Institute.
Rather than challenge the BLS report directly, ECRI notes that the data is skewed by the increase in the number of multiple job holders. As for Levy, while it notes that the data is skewed by an increase in temporary and low-wage jobs, it mostly passes on challenging the report by stating, "The U.S. economy is showing weakness in Nearly Everything But Employment (NEBE)."
Both organizations appear to have decided that the BLS data is so bad that spending the time and space necessary to refute it directly diverts attention from more important issues.
There is value in making those priorities, but it's also indicative of the trend toward irrelevance of government data as the source of inputs for substantive analysis of economic activity. And that trend should worry everyone.
The last option is to challenge the veracity of the BLS data. This can be done in numerous ways.
David Stockman chose to challenge the logic of the seasonal adjustment the BLS used.
He raises good points.
As technology increasingly has allowed for the collection and wide disbursement of data, the ability of the economic and political clergy to be able to manipulate subjectively has been diminished.
The response has been to augment government data collection systems increasingly with subjective inputs.
This is a subjective I've written numerous columns on, such as when the subjective of fraudulent census bureau data became an issue in 2013.
Coming back to the immediate issue of the most recent employment report by the BLS, if we assume the BLS report is accurate, there are some implications of it that can be gleaned.
A few of them are that payrolls and employment (including self-employment) are growing at 1.8% and 1.6%, respectively; wage and salary disbursements are growing at 4.8%; the "labor share" of cost of goods sold is 43.7%; and personal consumption expenditures (PCE) as a percentage of GDP are 69.6%.
In order for this to all be true, nominal GDP would be more than 4% and real GDP about 3.3%.
However, we know the nominal GDP through the third quarter was below 3% and real GDP was 2%.
In order for the BLS figures to be accurate, there would have had to have been an extraordinarily large increase in both the production and consumption components of fourth-quarter GDP.
We also know that not only did that not happen, but also that the exact opposite did. The latter is most evident in the GDPNow estimation of fourth-quarter real growth in GDP, which according to GDPNow decelerated to .8%, with downward revisions to the estimate occurring throughout the quarter.
The consistent downward revisions are the result of the continuous flow of incoming data on things such as personal consumption expenditures, real final sales, health care expenditures, after-tax income, price changes, population growth, real private nonresidential investment, private employment, payroll tax receipts, and many others that are aggregated into the singular measure of GDP.
As I noted yesterday, all these measures of activity are mathematically bound to each other as they are simply measuring capital at different points in the continuum.
A 0.8% real GDP growth rate implies a growth rate in nominal final sales, minus government deficit spending, of below 3%, which traditionally has been recessionary since the 1940s.
If the Bureau of Economic Analysis reports as much when the advance estimate for fourth-quarter GDP is announced Jan. 29, the financial media's response almost assuredly will be to blame it on warm weather reducing consumer spending in the quarter, and that as such it is not indicative of a trend.
However, that also begs the question as to why the BLS did not alter the seasonal adjustment used for December downward and how it arrived at upward revisions for the previous months in the quarter as well.
The bottom line is that it is mathematically impossible for the employment data reported by the BLS to be accurate, and further that it is highly illogical that BLS personnel are not aware of this fact.