Seasonal adjustments have legitimate use only when there is a recognizable and relatively unbroken patterning in a data series. But most statistical agencies and their collection methods do not or cannot process structural changes. They adjust their benchmarks via the calendar, regardless of what happens in the intervening years.
That presents a large problem for seasonal adjustments and statistical estimates of data post-2008. What was reasonably expected (for volatility and beyond) of the economy and adjunct parameters before the housing bubble collapsed is not really relevant or applicable to how the data appears after that structural rupture. So these data series and estimations are using pre-crisis rules and adjustment processes and applying them to post-crisis estimations; therefore inappropriately “smoothing” data into what may not even exist.
In the New York Post yesterday the former head of the BLS confirmed as much. And we are not talking about the former chief from way back in 1987; Keith Hall was in charge of the data from 2008 until just last year. He knows this problem from up close:
“All parts of Washington’s data-collecting machine adjust to smooth out the bumps caused by the seasons of the year. But the recession that started five years ago was so severe and the recovery so anemic that the seasonal adjustments have been thrown off.”
When Chairman Bernanke mentioned that labor data “overstated” strength in the jobs market, there can be little doubt this was on his mind, as the article pretty much confirms:
“The former BLS bigwig is convinced the Federal Reserve understands the true state of the economy. But he says he’s never discussed any of this with the Fed, although he wouldn’t be surprised if the statistics gatherers at the central bank and at BLS have compared notes.”
That further bolsters the data we are seeing from outside the Establishment Survey from the BLS. While mainstream economists prefer it because it is so fully adjusted, as I have highlighted many times in the past it is a real weakness just for these reasons. The Household Survey may be more volatile, but it is far more “natural” or “pure” in terms of the data’s relation to the primary sources.
None of this is good news for the taper bulls/doves/hawks or whatever they are now. Not only is the economy slowing, even what statistical strength there appears to have been is in full danger of revisions that will surely come at some point. The calendar will eventually move far enough into the future so that this pre-crisis bias will become nothing more than historical curiosity.
Click here to sign up for our free weekly e-newsletter.
“Wealth preservation and accumulation through thoughtful investing.”
For information on Alhambra Investment Partners’ money management services and global portfolio approach to capital preservation, contact us at: email@example.com