Tuesday, December 23, 2014

For the record


 - by New Deal democrat

Over at Daily Kos, Meteor Blades has trotted out the canard that I made a big mistake in forecasting the kind of recovery we would have from the Great Recession.  So this post will serve to put that canard to bed.

Meteor Blades and I have not been on polite terms since I wrote a diary back in my DK days in May 2009, on that year's April jobs report, in which I said:
    The BLS reported this morning that in April the economy shed another 539,000 jobs.  The U3 unemployment rate also increased to 8.9%.  This is a .4% increase from March, and is what I expected.    This is one of those cases where "less awful" actually ought to give rise to some hope.
Meteor Blades  commented that:
One month may be the start of a new trend, but it is not of itself a new trend.
He followed that up with a front page post on that blog entitled  One Month of Slightly Less Terrible Jobless News does not make a Trend, in which he wrote:
Since the Bureau of Labor Statistics announced its monthly payroll survey numbers Friday, there’s been quite a bit of happy-talk – including from some progressives – about how we’re just around the corner from the end of the "Great Recession." ….Is the intent of the happy-talkers to use fact that unemployment is getting worse at a better rate to chill the widespread rage over what has - during the past three decades - brought us to the economic precipice? …. Beats me. 
….  Only the foolish can put a smiley face on Friday’s jobless report.
(my emphasis) 

Since then, Meteor Blades has steadfastly denied that his insult was aimed at either Bonddad or myself, but now that you can see the quotes side-by-side, the truth is pretty obvious.

Since, as it turned out, neither of us were foolish in the slightest, the canard that we predicted a V shaped jobs recovery is one of those things that the Doomers occasionally trot out to defend their abysmal record.  So let's set my record straight.

One of the things I am not afraid of doing is challenging the conventional wisdom, where there is data to support such a challenge.  In 2009, the conventional wisdom was that even after the recession officially ended, there would be a long period of increased unemployment and continuing job losses.  Since there was a contrary case to be made, I thought it was worth writing about, and did so in 3 parts. 

Here's the intro and conclusion from the first entry
Recently data has caused me to question the certainty of such a continuing "jobs recession."  While I'm not saying it will happen, there is a surprisingly decent case that contrary to the accepted wisdom, the recovery from this recession may not be "jobless" at all, but particularly in its earlier portion may feature quite robust job growth. There are several different reasons supporting such an outcome  
.... the recession is nearly over, but suggests that at least in the next few months, there could be actual job growth, more robustly than almost anyone suspects.
Notice that I challenged the "certainty" of a jobless recovery, and even highlighted the conclusion's "could be" in italics. 

And I repeated the qualification that I was only "making a case," not embracing a forecast, in part 2 and part 3 as well.  In the conclusion to part 2, I even posited a test to see if there would be a V-shaped jobs recovery or not, writing:
one way to tell if we [are] going to have a V-shaped recovery or not [is] "monthly manufacturing hours worked:"  an average rise of only 0.1 hours per month means a lackluster increase in plant use, over 0.15 supports a V-shaped recovery.
And here again is the into and conclusion in the last installment:
In this 3 part series, I am examining the mounting evidence that contrary to the accepted wisdom that we will have a "jobless recovery" where GDP turns up anemically but unemployment stubbornly rises, the recovery from this recession might not be "jobless" at all, but particularly in its earlier portion may feature quite robust job growth as the GDP starts to grow probably right now  
Finally, let me repeat that I am not saying that we will have a "V"-shaped jobs recovery -- only that a surprisingly substantial case can be made that it might indeed happen.
This is self-explanatory.  A "case can be made" for an alternative almost everybody seems to be dismissing, not "I forecast" X. I explicitly said to the contrary, and even noted a way to test the hypothesis.  In fact, a couple of months later,  I highlighted the contrary argument for a jobless recovery:
Since I recently hit the Rec list arguing that The Recovery may not be Jobless for Long, an admittedly minority view, it's only fair that I present the contrary conventional wisdom, which was ably set forth last week by Dr. David Altig
So what happened?  As to employment, that bottomed 8 months after the end of the 1991 recession, 22 months after the end of the 2001 recession, and 8 months after the 2008-09 recression.  The unemployment rate shows a more stark contrast,  peaking 15 months after the 1991 recession, 20 months after the 2001 recession, but only 4 months after the end of the 2008-09 recession. 

And by March 2010, I was able to write that there was a Bifurcated Recovery:
industry and associated economic metrics show a strong V-shaped recovery, the best since 1983, then once we look at that part of the economy most closely associated with average American consumers, another picture emerges entirely.
 I recommend that you check out the graphs in that last post. The recovery from the 2008-09 recession was indeed V-shaped for things associated with industry and production.  It also was relatively V-shaped for real GDP, and also, in the earlier part of the recovery, in terms of aggregate hours worked, which made up over half of their 22-month loss of 10% (to October 2009) over the next 26 months. For the number of jobs created, and certainly for wages, it was not.

Where I have made forecasts and been wrong, I have had no problem saying so.  In fact, earlier this year my forecast of a YoY decline of 100,000 housing permits didn't pan out.  What did I do?  I wrote about it publicly, and examined the data to see what might help my forecast - in the case of housing this  year, the fact that the huge Millennial generation created  upward pressure on household formation, and their presence was a boon to housing data just as that of the Boomer generation was half a century ago.

But the alleged forecast of a V shaped recovery isn't one of them, because that's explicitly what it was not.