Thursday, August 29, 2013

Murphy Responds to Krugman, DeLong and Me

Bob Murphy has responded to the firestorm he started when he attempted to refute Brad DeLong within DeLong's own framework.

I hasten to emphasize, once again, that in doing this Murphy has entered the devil's den and adopted the tools of the economic devils, not only does Murphy work within the non-Austrian framework of GDP but he also adopts a type of bizarre GDP forecasting result, something called "potential GDP," that is some kind of outside real markets measure of GDP "capacity".

Here's Murray Rothbard on GNP, the econometricians' precursor measure to GDP, and forecasting:
The mathematical laws in the physical sciences contain numerous constants; but the imitative method in the social sciences is proven vain by the fact that not a single constant has ever emerged. Moreover, despite the use of sophisticated econometric models and high-speed computers, the success rate of forecasting economic quantities has been dismal, even for the simplest of aggregates such as Gross National Product, let alone for more difficult quantities; the record of GNP forecasting by economists has been poorer than a simple layman’s extrapolation of recent trends. In fact, the federal government has had notably poor success even in forecasting the one variable under its own absolute control—its own expenditure in the near future.
And I , again, point to the many problems with GDP identified by Frank Shostak, in an article, What Is Up With GDP?, where he correctly concludes:
[T]he GDP framework is an empty abstraction devoid of any link to the real world.
So we have Murphy using the non-Austrian concept, GDP, in combination with  forecasting tools that have never shown practical success in forecasting GDP, never mind the Alice-in Wonderland "potential GDP." This is the doozy of a position from where Murphy starts his defense. He goes on to ditch his initial method of attack on DeLong (since as Krguman pointed out, he failed to correctly understand how the CBO calculates "potential GDP")

I repeat. He has ditched his initial argument and takes up on the clue provided to him by Krugman and uses Krugman's approach to go after DeLong. He (after Krugman's correction) publishes a chart  to show that, using CBO data, he is correct (and, well, forget about his initial errors.)

But the problem with this new Murphy defense is that he assumes correctness of CBO forecasting (which has NEVER been correct) and he further assumes, without foundation, that the CBO's entire decline in "potential" GDP growth is caused in CBO calculations by a decline in investment. But this is far from necessarily the case. As Krugman initially pointed out, the CBO not only takes into consideration changes in investment but growth of labor and technological progress. Murphy ignores labor and technological factors and simply assumes the entire decline in GDP capacity is the result of investment. This is just poor, sloppy argument. He may go back and attempt to examine what numbers the CBO used for labor and technological progress to buttress his case, but this is like introducing the tooth fairy, Santa Claus and Saddam's weapons of mass destruction into the Alice in Wonderland story. Government has been terrible about measuring technological growth. As Alan Greenspan pointed out in his book, The Age of  Turbulence, government missed the growth in productivity caused by the desktop computer!:
The data we were getting from the Commerce and Labor departments showed that productivity (measured as output per hour worked) was virtually flat in spite of the long-running trend toward computerization. Icould not imagine how that could be. Year in and year out, business had been pouring vast amounts of money into desktop computers, servers,networks. software, and other high-tech gear...This became evident as early as 1993 when new orders for high-tech capital began to accelerate after a protracted period of sluggish growth. The surge continued into 1994 suggesting that the early profit experience with the new equipment had been positive.
There were other, even more persuasive indications that the official productivity numbers were awry. Most companies were reporting rising profit margins. Yet few had raised prices. That meant their costs per unit of output were contained or even falling. Most consolidated costs (that is, for business considered as a whole) are labor costs. So if labor costs per unit of output were flat or declining, and the rate of growth of average hourly compensation was rising, it was an arithmetical certainty, that if these data were accurate, the growth of output per hour must be on the rise; productivity was truly accelerating.

This is the nutty framework that Murphy wants to play in. I stand with my original assessment: OMG.


  1. Even a little thinking should lead one to conclude that GDP and other similar measures are meaningless. Here's a thought experiment. What does GDP even mean in the case of a reasonably good gold standard? With the supply of money basically fixed, GDP doesn't go up in any meaningful sense. In fact, further derivatives like GDP per capita might actually decline, in the face of robust population growth (another variable to add to the thought experiment). But productivity does. It's probably also a bit clearer, when looking at it from that perspective, that government employees aren't "productive" in any meaningful sense, and actually counter-productive, quite literally.

    1. First, let me say I agree with the spirit of your post re: GDP.


      "What does GDP even mean in the case of a reasonably good gold standard? With the supply of money basically fixed, GDP doesn't go up in any meaningful sense"

      I don't think this is correct. GDP can (and would) still rise in such an economy. As productivity rises, prices fall and therefore the GDP deflator (inflator under the GS) would cause GDP to adjust upward. This is how GDP rose during the phase of America's history that was under the GS. Although I will agree that FRB and the lack of a full-reserve system muddies the water a bit.

    2. Of course GDP can rise with a fixed money supply. Real GDP growth is the sum of productivity growth, population growth and change in the employment to population ratio. If you are seeing productivity growth and population growth without GDP growth, then you are seeing a ton of layoffs because the only way that can happen is if the employment to population ratio declines. But of course if you buy into the Austrian anti-math gibberish, then you don't believe in measuring anything including population growth. The strategy is to discredit all the economic indicators and then make up whatever you want. So long as the conclusion makes the govt looks oppresive, then you are good to go.

  2. "He goes on to ditch his initial method of attack on DeLong (since as Krguman pointed out, he failed to correctly understand how the CBO calculates "potential GDP")

    I repeat. He has ditched his initial argument and takes up on the clue provided to him by Krugman and uses Krugman's approach to go after DeLong. He (after Krugman's correction) publishes a chart to show that, using CBO data, he is correct (and, well, forget about his initial errors.)"

    Unless I'm mistaken, I think you misunderstand Murphy. It's not that Krugman -corrected- Murphy, it's that Krugman's point is perfectly in line with what Murphy was originally saying. Krugman's point is only further evidence of Murphy's original point.


    1. Yes, this would be correct. Also, Murphy never assumed correctness of CBO forecasting. "...I’m doing this whole post within DeLong’s framework, just to show he’s making a non sequitur even on his own terms."

    2. @Bharat

      LOL Krugman wasn't supporting Murphy. He was pointing out further errors in Murphy's analysis based on his initial attack on DeLong.In that first attack, Murphy was wrong from all frameworks. Murphy then ditched his original analysis and incorrectly used Krugman's point in order to support his own claim. (See below)

      @ammonymous You write:Murphy "never assumed correctness of CBO forecasting," precisely one of my points, it's a nutty position from where to start a debate, where you don't even assume the data is correct, but the further problem is that even giving Murphy this point, he states that the entire "potential GDP" is based on business investment, but within Krugman and DeLong's "frameworks", they both mention labor and technical progress.

      The only way Murphy might get out of this nutty hole is to dig further into the the hole and possibly use the labor and productivity data projections of the CBO, which are even less grounded in reality.

      It's all nuts.

    3. Apologize for the late reply since it may not be noticed:

      Mr. Wenzel,

      Krugman's intention was not to support Murphy, I realize that. Nevertheless, his actual argument did support him.

      Murphy's point is not that investment is the sole factor in determining potential GDP. DeLong originally claimed that there are zero signs of lower potential GDP: this means every single factor DeLong looked at showed no indication of lower potential GDP. Thus, Murphy's response that "well, investment is lower" is a perfectly adequate response.

      Krugman's response that "the CBO already takes investment into account" was besides the point. Precisely because the CBO's measure of potential GDP is also down, after investment being down! Therefore, this is just further evidence that Murphy was right, even within Krugman and Delong's framework.


    4. Other than stating "investment is lower," Murphy shows us no data to back this up and gives the impression that the entire decline in "potential GDP" is because of investment.

      What Murphy did was dump his original argument, once he realized it was wrong, Wenzel correctly points this out. Then Murphy starts from Krugman's argument but fails here by failing to take into consideration changes in labor and productivity---Wenzel then points out that if Murphy attempted to do so, he would be going deeper into a bizarre Alice-in-Wonderland analysis that has nothing to do with reality?

      In other words, do you seriously think any of these measures have any value? It's a fools game which Murphy played poorly first in his argument against DeLong, where he misunderstood how the CBO made it's calculations and second by not acknowledging that there were other impacts on "potential GDP" beside investment.

    5. @Bharat,

      Why are you defending Murphy, when it is clear that he didn't understand the CBO modeling in his first go round and then distorted what the results meant during his second go round?

      It is just terrible argument and a very strange topic for an Austrian to be debating in the first place.

    6. Why is it clear Murphy didn't understand CBO modeling in his first post?

      Saying "investment is lower" is a perfectly adequate response to "there are no signs in all factors involved in potential GDP, including investment, that would lower potential GDP".

  3. Stop embarrassing us with this nonsense, Murphy.

  4. Maybe this is Murphy's way of baiting Krugman into the Murphy/Krugman debate.

  5. Murphy is still trying to argue from a Keynesian point of view and gets scolded by DKuehn:

    I talk basics to Kuehn, and his head explodes: