Monday, February 24, 2014

ECRI now claims we're in the longest recession since 1929-32


 - by New Deal democrat

I haven't addressed the badly blown recession call by the formerly highly-regarded ECRI since its 2 year anniversary last September.  I'm updating my commentary now because, as of February, ECRI's fictitious recession, which they claim started in July 2012, is 1 year and 8 months old.  That makes it longer than 1973-74, longer than 1981-98, even longer than December 2007 - June 2009.  In fact if their claim is to be believed, this is the longest recession since the "great contraction" of 1929-32.

Forget government statistics, which ECRI notes can be revised significantly.  Private statistics which don't get revised like railroad traffic, truck tonnage, steel production, the ISM manufacturing and services indexes, temporary hiring and consumer purchasing as measured by the ICSC, Johnson Redbook, and Gallup, have all nearly relentlessly risen throughout this period.

ECRI's original mistake was a very human one:  the interpretation of a downturn in their indexes.  You may recall that they originally said that a recession was "unavoidable" even by concerted government action, and was imminent if it had not already begun, back in September 2011, just after the debt ceiling debacle.  During that time almost all economic statistics dropped like a rock, as consumers and businesses froze while they waited to see whether the US government would actually default on its debts, and if not, what altered landscape any proposed "grand bargain" might bring about.

As a result, the ECRI Weekly Leading Index and Weekly Coincident Index, which had each been as high as about +7 early in 2011, and had already dipped to 0 in the WLI and about +4 in the WCI at the time the debt ceiling debacle began,  dropped to about -7 and +3 by the time of their September recession call.  He's a graph of both of the them for the year 2011:





ECRI simply misinterpreted the fallout from a transitory political crisis as an organic economic development.

By stubbornly refusing to admit error, they have undercut if not outright contradicted  their own earlier thesis that the US economy was entering what they called "The Yo Yo Years." Their thesis was that GDP was not capable of meandering about 2% for very long.  It would either spike into expansion or downward into recession.  These Yo Yo expansions and recessions would happen with increasing frequency.  Instead GDP has generally meandered at around 2% +/-1.5% for 3 years now.

And instead of more frequent, if brief, expansions and recessions, they are stuck trying to defend a recession call going on 3 years old, with a seies of ad hoc references to the declining data du jour.

Their most recent missive, in fact, appears to be a flat-out contradiction to a position they took only 2 years ago.

Earlier this month, in an article entitled "Failure to Launch," ECRI touted their Weekly Coincident Index again, showing that the change in their coincident index had dropped precipitously from the end of November until the end of January:  



That's pretty straightforward.  Everybody who follows economic data knows that such a deterioration has taken place.

The part that appears disingenuous, however, is the preceding part of the index, which shows the growth rate climbing throughout the rest of the year, and spending the 8 months between April and November almost entirely above 4%.

Now take a look at this graph of ECRI's coincident index that was supplied to Mike Shedlock a/k/a Mish in March 2012 (annotations are Mish's):



Here's the very specific claim that ECRI made at the time:

The latest USCI growth rate is 1.94% (which can be rounded off to 1.9%). In January 1996, it had dropped only to 2.06% (which can be rounded off to 2.1%). This was certainly not below current readings. Of course, no recession followed. 

In 1998, the USCI growth came nowhere near current readings, so the question doesn’t arise. It wasn’t until January 2001 that it fell below 2%, and the recession began two months later. 

.... If you look at all the occasions in the last 50-plus years when USCI growth fell to 2.0% or below ..., it is clear that recessions began around those dates.... 

In sum, it is precisely accurate to claim that y-o-y USCI growth has never dropped to current readings in the past 50-plus years without a recession ensuing.

Shedlock was properly incredulous  at the claim that once the coincident index fell below 2 (to 1.94), the economy would inexorably slide into recession, whereas at 2.06 the economy would bounce back.

While ECRI has apparently taken its Weekly Coincident Index out of the public domain, we do have its monthly Coincident Index covering 2011 through 2013, showing that with the exception of 3 months in late 2012 and early 2013, the Coincident Index was above the allegedly critical "2" level:



And what the 2013 Weekly Coincident Index supplied by ECRI itself shows is that, to the contrary, all the time for the last year that ECRI has been claiming we are in recession, the Weekly Coincident Index was also well above that allegedly crucial "2" level.

In other words, ECRI's position now appears to be in blatant contradiction to the position they took in March 2012.  The Coincident Index did not continue to fall, but instead at virtually all times thereafter has been higher -- just like after January 1996.

One final thing.  It's pretty clear that one of the components of the Coincident Index is personal income. That's what explains the upward spike in December 2012, and the downward equivalent spike in January 2013.  At the time of ECRI's "Failure to Launch" commentary, the YoY comparison was with the December 2102 upward spike.  On March 3, January personal income will be reported, and the YoY comparison will now be with January 2013's downward spike.  I confidently predict that ECRI will not publicly update its Weekly Coincident Index once that data is reported.

In summary, ECRI appears to have descended to the level of Zero Hedge in trying to salvage their reputation.  Sad.