Friday, December 23, 2011

The Real Negative Real Shock

Are all the problems in the U.S. economy nominal?  Robert Gordon implicitly says no in a new paper on the long-run outlook for U.S. productivity (hat tip Reihan Salam).  He makes the case that rapid productivity gains from 1995-2005 will not persist going forward:
The 20‐year period 1987‐2007 combines the inexplicably slow productivity growth of 1987‐95, the temporarily ebullient period 1995‐2000, and the interesting 2000‐07 period that in some dimensions looks like more normal behavior.  The seven years between 2000:Q4 and 2007:Q4 were neatly divided in half, with extremely rapid productivity growth between 2000:Q4 and 2004:Q2 (2.68 percent), and much slower growth from 2004:Q2 to 2007:Q4 (1.36 percent), averaging out to 2.02 percent for the seven‐year interval.   As argued above the productivity growth “explosion” of 2001‐04 rested on a combination of savage corporate cost cutting and delayed learning from the internet revolution.  Once profits had recovered the pressure for cost cutting disappeared, and eventually the delayed learning subsided as well.   
The paper approaches the task of forecasting 20 years into the future by extracting relevant precedents from the growth in labor productivity and in MFP over the last seven years, the last 20 years, and the last 116 years.  Its conclusion is that over the next 20 years (2007-2027) growth in real GDP will be 2.4 percent (the same as in 2000‐07), growth in total economy labor productivity will be 1.7 percent...
So over the next two decades Robert Gordon sees labor productivity growing at annual average rate of 1.7% compared to about 2.5% for 1995-2004.  If his view is widely held then that means firms will expect lower returns to investment and household will expect lower incomes.  Such lower expectations, in turn, would translate into lower investment and consumer demand today.  This, then, may account for some of the prolonged slump. 

So is Gordon's view widely held?  Is the forecast for productivity falling?  The Quarterly Survey of Professional Forecasters can answer these questions.  It asks forecasters what they expect the average annual productivity growth rate to be over the next 10 years.  The data starts in 1992 and is at an annual frequency.  Here is a figure of the data:

So yes, the consensus forecast is that productivity growth is expected to decline over the next 10 years.  It is hard not to look at this figure and conclude at least some of the ongoing slump can be attributed to it.  However, this does not necessarily mean it is the most important factor.  And I do not think it can be because we do not see a sustained uptick in the inflation rate, something that should be present if the permanently lower productivity growth rate were the main culprit.  Rather we see muted inflation since 2007 with the core inflation rate actually falling, something far more consistent with a large amount of insufficient aggregate demand.   And there is the negative output gap.  I still believe that the failure by the Fed to return nominal spending to its pre-crisis trend is the most important reason for shortage of aggregate demand.

Update:  Bill Woolsey notes that the lower expected productivity growth should only affect real variables but have no bearing on nominal expenditures if properly stabilized. 


  1. Lower natural interest rate. If the market interest rate falls with the natural interest rate, no effect on nominal expenditure. Ceteris paribus, maintaining nominal expenditure would require faster money growth. The slower real output growth and constant growth path of nominal expendiutre-- higher inflation.

    Employment? Ambiguous. Unemployment? Should be no effect.

    Growth of real standards of living? Lower.

  2. So yes, the consensus forecast is that productivity is expected to decline over the next 10 years.

    Productivity growth is expected to decline.

    How does one square these expectations (factor in population growth to boot) with the zero real yield on U.S. treasuries?

    I wonder whether these lower forecasts are also associated with more uncertainty? Confidence intervals would be helpful, if they were available.

  3. David,

    You are correct, I should have said productivity growth. I fixed that in the post.

    That is a great idea to check the variance around the mean forecast.

  4. better late than never...

  5. Why should productivity growth decline? I don't get this.

    It may be a rotten Fed policy stifles GDP growth, lessening corporate profits, lessening money to invest in R&D and capital equipment. Is that why?

    I sense global productivity will soar as modern equipment become more commonplace globally. The cell phone alone is a wonderful source of productivity improvement.

    It seems to me that every recession brings out doomsayers---and this has been one long and rotten recession.

  6. ECB,

    Glad to see you are still with us.

  7. Benjamin,

    Gordon explains the potential reason for the productivity growth slowdown are that the productivity gains of 1995-2004 are not sustainable ones. See his paper.

  8. David, very interesting post.

    There is no doubt that the USA long-term real growth outlook is worse than it used to be. However, in my view the present slump has very little to do with that. It is not exactly a "news" that RGDP growth is like to be lower in the coming decade than in the last couple of decades. To me it is still about overly tight monetary policy.

    However, in the long run it is obvious that productivity growth determines the general level of income and wealth in society. Monetary policy can not do anything about that - and Market Monetarists of course never argued that - even though some seem to think we did.

    Of graph with expectations for RGDP growth is, however, every interesting in regard to the discussion about whether US monetary policy was overly in the years prior to 2008. I don't think that there is strong evidence to show that that was the case. However, the relatively sharp drop in long-term RGDP growth might have been the trigger that pricked the "bubble" - if there indeed was a bubble due to overly loose monetary policy. This would confirm the kind of theory that you and George Selgin have put forward concerning relative inflation and productivity shocks. I am not sure that this is correct for the years just prior to 2008, but I think it is worth further examination. (I have just published a comment on boom-bust: - which could be relevant in the that regard.)

    However, whether or not US monetary policy was overly in 2004-8 is really not relevant for the present situation as it is clear that NGDP has fallen well below the pre-crisis trend and as such there is no doubt in my mind that US monetary policy is still overly tight.

    Btw this might actually be telling us something about why countries apparently growth slower in the decade after the bust than before. Maybe crisis coincides with a change in underlining RGDP growth. This would mean the balance sheet recession explanation is wrong.

  9. Lars,

    In my view the figure, which shows expected productivity growth, does shed some light on the stance of U.S.monetary policy in the early-to-mid 2000s. It shows expected productivity growth rates rising through about the mid-2000s. Robert Gordon's paper even talks about this 2002-2004 mini-productivity boom period.

    A surge in the productivity growth rate should, all else equal, lead to a higher natural interest rate. The Fed at this time, however, pushed the federal funds rate to then historically lows. This gap created an incentive for leverage and ultimately for bubble-like behavior.

    Along these lines here is an older post of mine making this point:

  10. David, I must admit I am a bit skeptical about the natural rate as a empirical concept and if you look at monetary aggregates the story is somewhat different (but I guess that is the Hummel/Selgin debate). Furthermore, if one looks at "demand inflation" then it is hard to argue that US monetary policy was overly loose after 2001. I don't mind bashing Greenspan, but I must say that I increasingly are coming to the conclusion that it is "too easy" to claim monetary policy was very expansionary after 2001 - despite interest rates being kept very low and interest rates are afterall not a very good measure of monetary policy tightness.

    I haven't made up my mind on this so it is not a critique, but rather an attempt to think out loud.

  11. By the way I would suggest that the best measure for monetary policy "tightness" is to look at NGDP growth minus your RGDP long-term growth expectations. If that number - which in my view is demand inflation - increases then monetary policy is becoming more loose. But of course we should be looking at levels of demand prices - or what I have termed the Quasi-Real Price Index (QRPI)

    I hope soon to do a post on the Great Recession through the lens of QRPI.


    Well, productivity is rising nicely, according to the BLS.

    Really remarkable is durable manufacturing.

    I assume we are getting a bit of a "free ride" that any manufacturing business gets as output goes up, you get greater output per fixed unit of input.

    Still, I think any theory that productivity gains are behind us is hooey.