Payroll Stats Become Even More Implausible by Jeffrey P. Snider Alhambra Investment Partners Since Q1 GDP was revised lower by almost 1% that meant estimates of productivity were going to be even more out of alignment than they were at the first release. Of course, in a less massaged environment productivity might have preserved some sense if there was less rigidity from the BLS on the employment side. In other words, when “output” estimates were reduced (and they were, by more than GDP) it would make sense that everything would be revised downward in a more cohesive process. Instead, output was reduced significantly, by 1.4%, while total hours worked was marked down by all of 0.1%. As a result, productivity is revised from a nonsensical -1.9% to an even more skeptical -3.1%. If this was just a one-quarter problem, then it would be easy to dismiss as random variation or expected variance in all these statistics trying to tie together across real economy lags and such. But that is not the case, as productivity, and by extension the estimates for how “expensive” marginal labor is and thus the primary reason businesses hire and fire in the first place, has been seriously “off” for some time. With these latest estimates and revisions, productivity is now -0.7% over the last 5 quarters dating back to last year’s polar vortex. That just doesn’t make any sense in a meaningful context of business operations in the real economy – especially when the BLS is saying that this has been the best run of hiring in decades. The productivity figures alone show us how the BLS is likely overstating labor gains. Simply substitute a more meaningful level of productivity, such as the average gain during the less robust hiring period of 2003-07, and, by the rigid method of calculation here, total hours gained almost disappear entirely. The last time the Establishment Survey was as robust as the past year or so was 1999; then, the average productivity was 3.7%! That number actually makes sense intuitively, since businesses would have good reason to go on a hiring spree. Porting that to the current period, as in the mathematical construction of productivity here, would mean, holding output constant, that total hours in the past five quarters would have been not +2.7% but -1.7%. These numbers literally do not add up. The deeper you go into the maze of calculations, the more that sentiment grows, not dissipates. The “next step” is to take productivity and turn it into what is called Unit Labor Costs. This is one estimate for how much it “costs” businesses to employ labor, not just in terms of compensation but in terms of what comes out of the capital transformation of that labor input. Where productivity is low or especially negative, unit labor costs are exceedingly high because you pay workers and (assumed) more workers and get less out of them per unit. That is why it makes no intuitive sense for businesses to be hiring rapidly during periods of high unit labor costs, even where revenue growth is robust. In the past two quarters alone, because the BLS will not budge on their employment estimates, we are supposed to believe that businesses in the US have been falling over themselves to hire when doing so is historically very expensive. With the Q1 revisions to “output” moving downward significantly, unit labor costs in Q1 were revised up from 5% to 6.7%! That makes for two consecutive quarters of about 6%, which is usually what ignites, historically, recessions. Unit labor costs surge prior to each and every recession, which, again, makes sense, but the problem with this historical comparison is that the calculations more recently no longer conform to any truly identifiable pattern. What used to be fairly stable and at least consistent has turned into a muddle of numbers all over the place. That process appears to have started at the outset of the dot-com recession, gained into the Great Recession and has never abated apart from the typical negative labor costs only in 2009 and Q1 2010 (what recoveries are actually made of; negative and low labor costs at the margins show businesses to hire more and start the recovery process, not in the opposite manner as the BLS has it now). This obvious lack of consistency more than suggests problems in the various statistical pieces flowing into the calculations here. We are left with two primary interpretations in terms of the macro view (and a slew of mathematical and statistical quiddities that I won’t get into here); either these numbers are correct and US businesses have suddenly lost all ability to profitably to project and conduct business or the BLS is being stubborn in over-estimating labor utilization across-the-board. The third possibility, that output itself should be revised much, much higher I simply reject out of sheer overwhelming evidence to the contrary. The first explanation, dumb businesses, seems almost exhaustively implausible except for the interference of QE and economists’ ridiculous projections. It is possible, how likely is debatable, that businesses have simply rejected what they see of the current environment and ramped up hiring in anticipation of what every single economist the world over told them to. I have said before that I think that is a very real problem and is likely causing economic imbalances in certain places (inventory comes to mind) but I don’t believe that would be sufficient in such a broad labor scope. No, I think all the evidence continues to point to trend-cycle over-estimation and the BLS refusing to bend toward what is becoming irrefutable reality. There is absolutely nothing to suggest this is the best employment environment in decades, and the fact that the payroll numbers keep making that comparison only has one effect – to so skew “downstream” statistics as to preclude any other interpretation. With Payroll Friday upon us yet again, it is useful to keep this in mind that those payroll estimates not for actual job counts but of chained variations are completely bogus.
Thx for the post, Yesterday when the jobs number came out i was very suspicious, markets on the edge of cratering, and all of a sudden miraculously, some jobs number comes out, with wage growth we havent seen in 10 years to boot, and the market still reacted negatively, when good numbers (even fake good numbers) make the market sell off, its usually a sign the market is done. Yesterdays jobs number didnt even make any sense, supposedly the economy is flying off the hook now when europe, China, and japan are all taking a bath and destroying their currency in an attempt to inflate their way out, and the USD is sky rocketing? I dont think so.... doesnt make any sense to me. And this upside surprise just so happens to coincide with when the fed wants to test the waters with a rate increase? Give me a break. Watch for these numbers to be revised down three months from now. Just to confirm im not just a biased idiot, i asked a couple buddies whose numbers dwarf mine in the market, and both of them agree, the jobs numbers dont make any sense. For the life of me I cant even think of a way yesterdays jobs number was possible, let alone plausible.
Here is my theory to solve the mystery. Please give me your feedback whether it makes sense or not. In the last few years corporates even government organizations are hiring more contracts compared to permanent positions. They are doing this to reduce their liability so that they can lay off contractors with no problem any time that economy slows down and also they offer no benefit. While this idea looks exciting and enticing, in long term it does not work and reduces the productivity. A good chunk of contractors pay is cut by the contracting agencies. So although the companies might pay a reasonable rate for a contracting position, the contractor does not get paid fairly. Sometimes 50% of salary is cut by the contracting agency. This way the contractors might accept the job but they don't have any liability to the company as the contract is "at will" so in many cases when they get a better rate they just leave the company. Companies also do not spend time and money on training the contractors and that reduces their efficiency. I believe efficiency goes down when the job market improves and contractors have more opportunities to hop around.
My take is that if it looks and smells like bullshit - then it probably is. Back in 98-99 there was actual hiring with plenty of people working. Then look at what happened when the music stopped - total annihilation. Now take the comparative overall equity index valuations and combine it with an arguably weaker economy and we can see that if/when the music stops the resulting sell-offs might even smoke both '00 and '08.
More horseshit from the BLS. Does the Sum of the Parts Equal the Whole? In a curious event, ZeroHedge reports The September Jobs Report Was Even Worse: U.S. States Lost A Total of 22,000 Jobs. Zero Hedge noted that the month-over-month total of nonfarm payrolls was -22,000 while the nonfarm total was well over 100,000 nationally. He posted this chart. Those numbers match my totals. And as ZeroHedge reports, 28 states lost jobs while 22 gained. But it's not that simple. From the same data, here is a table of nonfarm payrolls for the last three months. Nonfarm Payrolls (In Thousands) Jul States Sum 143,071 Aug States Sum 143,218 Sep States Sum 143,197 Jul National 142,093 Aug National 142,229 Sep National 142,341 Jul-Aug Nat 136 Aug-Sep Nat 112 Jul-Aug States 147 Aug-Sep States -22 Yes, month over month, states lost 22,000 jobs. But also notice that state totals overall beat national totals by 856,000 jobs. Which stat do you believe? I have been aware of this discrepancy for a long time, and have been working on this data with the Michael Lucci and Connor White at Illinois Policy Institute since July. Here is an email we received from Tyler Downing at the BLS on July 22. Hello Connor, Thank you for contacting the Current Employment Statistics (CES) program. CES independently develops national and state and area employment, hours, and earnings series. Both sets of estimates are based on the same establishment reports; however, CES uses the full establishment survey sample to produce monthly national employment estimates, while CES uses only the state-specific portion of the sample to develop state employment estimates. State and area estimates use smaller amounts of sample by industry than the national industry estimates. This increases the error component associated with state and metropolitan level estimates. For this reason, aggregating state data to the national level will also sum this error component, resulting in different estimates of U.S. employment, hours, and earnings. Summed state level CES estimates should not be compared to national CES estimates. Estimates for states and areas are produced using two methods. The majority of state and area estimates are produced using direct sample-based estimation. However, published area and industry combinations (domains) that do not have a large enough sample to support estimation using only sample responses have been estimated using modeling techniques. Random Results As noted above, the current state summation actually exceeds the national total, even though the latest month-over-month state total summation is negative. When I first went through this exercise many months ago, I thought there was a serious issue but concluded there wasn't. To verify, Michael Lucci and I went through the process of downloading year's worth of history but the results appeared random. Double Counting Part-Time Jobs? This is not a validation of the overall BLS process by any means. I remain firmly convinced the BLS is double counting part-time jobs. And in a recent phone conversation, a BLS analyst admitted it was possible. I asked a simple question: Why don't you sort out duplicate social security numbers? The answer I received was "we would like to but we do not have access to the data for privacy reasons". A decent sort-merge algorithm could hash this out easily, but only if the BLS had access to SS numbers. So here we are wondering why the sum of the parts exceeds the whole overall, while we frequently see the opposite effect month over month. The much-maligned BLS appears to blame, but in actuality, it appears as if the BLS does not have access to the data they need to produce valid numbers. Are major discrepancies like these better than no numbers at all? Mike "Mish" Shedlock