Sunday, June 01, 2008

Steven Horwitz and Post Keynesians

Steven Horwitz has kindly made his papers available online. I find that Horwitz has engaged in a couple of controversies with Post Keynesians, broadly defined. I refer to these interchanges:
  • Hill (1996a), Horwitz (1996c), Hill (1996b), and Horwitz (1998)
  • Horwitz (1996a), Cottrell (1996?), and Horwitz (1996b)
What worries me about these is whether I need to add to my already-too-long critique of Austrian Business Cycle Theory (ABCT) in yet another revision. (The latest publicly available version of my critique of ABCT is here. I have submitted an even more recent version to some journal.) Luckily, nobody in these interchanges seems to bring up my points from capital theory. I already reference his later book, Horwitz (2000). I hadn't known that he developed these ideas partly by arguing with Post Keynesians. In the remainder of this post, I illustrate some deficiencies of Horwitz's arguments.

Horwitz clearly states the mistaken Austrian view that, in a coordinated state, the length of production processes is inversely related to the interest rate:
"The capital structure ... reflects ... 'roundaboutness,' of production... Recent developments of Austrian capital theory have abandoned Böhm-Bawerk's emphasis on the 'average' period of production, the idea of the capital structure as reflecting roundaboutness remains. ... At higher rates of interest we would expect shorter processes of production ... Conversely, lower rates of interest should lengthen the capital structure, as the lower cost of time will make more roundabout, and hence more productive, processes more feasible... The capital structure needs to embody some notion of intertemporal coordination, i.e., the lengths of current processes of production should correspond to the willingness of consumers to wait for the availability of consumer goods. If the capital structure is not intertemporally coordinated, ...then there is avoidable economic waste." -- Horwitz (1996a)
Allin Cottrell takes Horwitz to task for ignoring established criticisms of ABCT and for simplifying Hayek to the point of misrepresentation. One of these criticisms is a Cottrell paper, which I do cover in my critique. Cottrell's comments are focused on monetary theory, not on the capital-theoretic analysis I write about.

Consider an imaginary coordinated state of the economy in which the expectations and plans of all agents are largely consistent and in which those expectations are being fulfilled. In such a coordinated state, must all markets clear? Must the labor market clear? Keynes answers these questions in the negative. He argues that if agents' plans and expectations were being more-or-less achieved, the level of unemployment would not need to be zero or whatever level is consistent with frictional unemployment. By the way, it is along these lines that I think a synthesis of Keynes and Sraffa can be achieved. Jan Kregel probably puts this better. Horwitz, in his responses to Hill, begs these questions. I refer to Horwitz discussion of savings, investment, interest rates, and banking. To me, Keynes and the Post Keynesians are more perceptive than Austrians both about what is seen and what is not seen in a capitalist economy. Since Horwitz does not understand the argument, he cannot credibly dispute me.

Here's another example of Horwitz's lack of understanding of Keynes: he attacks Keynes with a strawperson:
"Cash is not fundamentally different than other goods or services; if the public wishes to hold more of it, there is no reason that the laws of supply and demand should take a holiday. Keynes assumed that cash is 'barren' because it provides no yield to the holder, such that holding cash was socially wasteful. If one defines 'yield' to be some sort of financial return, then Keynes was right, but the same is true of any other good or service. Compact disc players do not yield a pecuniary return, yet people choose to hold 'stocks' of them because they provide a service - the playing of music. Cash is no different, in that it provides the service of being available to buy things. By Keynes's logic, fire trucks standing in fire stations are 'doing nothing' and yield no benefits. But they provide the benefit of being available for use when needed." -- Horwitz (1996c)
This is wrong for at least two reasons. First, for Keynes, money does not necessarily have no financial yield. In chapter 13 of the General Theory, Keynes states that one can
"draw the line between "'money' and 'debts' at whatever point is most convenient for handling a particular problem... It is often convenient to include in money time-deposits with banks and, occasionally, even such instruments as (e.g.) treasury bills." -- John Maynard Keynes (1936)
Second, and more seriously, Keynes is quite aware that money is liquid, that is, available to buy things. In chapter 17 of the General Theory, Keynes states that the expected return from the ownership of an asset is "equal to its yield minus its carrying cost plus its liquidity premium".

Horwitz seems to miss the point here too:
"Hill notes that 'to hold cash is to take refuge from uncertainty.' He then quotes G. L. S. Shackle, who argues that taking refuge from uncertainty (presumably by holding cash) means that we also take refuge 'from enterprise, from the giving of employment.' In the short run, my decision to hold more money in the form of bank deposits redistributes employment among different people, but does not reduce 'the giving of employment.' And in the long run, my uncertainty-induced decision to hold more bank deposits will generate increased employment.

In really existing capitalism with central banks, an increased relative demand for currency due to uncertainty can indeed touch off a decline in aggregate demand (assuming anything less than perfectly flexible prices) if the central bank does not react properly. However, in a banking system where individual banks were allowed to create their own currencies in the same manner as their deposits, this would not be the case. Where currency is a liability of the banking system, taking refuge from uncertainty in currency is no different from taking refuge in bank deposits and the effects are the same: an increase in wealth over time, due to the increase in saving... By making currency a bank liability rather than a reserve medium, a 'free banking' system would allow one less way for the public to withdraw bank reserves and cause a major contraction in credit and aggregate demand. The problem with contemporary central-bank-controlled capitalism is the very fact that when the public wants more of certain kinds of money, private enterprise cannot legally 'produce more of it.' Such a prohibition is not an irremediable fact of the world, but a particular institutional condition of modern capitalism that could be changed by ending government control over money." -- Horwitz (1998)
Whether or not the supply of money is endogenous, the demand for money does not directly give employment (excluding data entry for those changing some bits on a computer in some bank). This contrasts for the demand for an illiquid asset which must be manufactured to be supplied. As for the indirect effects, I find nothing but argued assertion in Horwitz.

I'm not even sure that Horwitz understands his own theory. Consider this assertion of the consistency of time preference and liquidity preference theories:
"Hill is implying that ... that cash trades off against such securities, ... that the demand for money is caused by interest rates and that, therefore, the interest rate is ... overwhelmingly a function of the supply and demand for money. However, the fact that I prefer the present to the future is intimately linked with the fact that the future is uncertain. ... I never denied that people hold money due to uncertainty. In fact, that is precisely why they hold it, as I have argued elsewhere ... The desire for liquidity or availability services is a manifestation of the more fundamental concept of time-preference. If I choose to sell bonds and acquire cash, this suggests that I am worried about the future and want cash so as to leave my options open. My time preferences have shifted toward the present. When I sell my bonds, bond prices fall and interest rates rise, as they should to reflect my concerns about the future. There is no contradiction between recognizing money's liquidity or availability services and a time preference theory of the interest rate." -- Horwitz (1998)
Time preference relates to time profile of ones desired consumption stream, to whether one wants to consume goods now or later. A change in what assets one holds for transferring purchasing power in the future is not a manifestation of a change in time preference. When the financial value of my savings and income stay unchanged, but I change the assets in which my savings are held to be more liquid, my time preferences have not not shifted, toward the present or otherwise.

By the way, Horwitz points out in his second reply to Hill that he is not arguing about theories that describe actually existing economies. He is explaining how the world would work if his utopian blueprint, especially for the banking system was adopted. I like how supposed fans of Hayek exhibit no appreciation for Popper.



Gabriel said...

>> "To me, Keynes and the Post Keynesians are more perceptive than Austrians both about what is seen and what is not seen in a capitalist economy."

You are now officially my hero!

Also, you pick too easy targets.

Anonymous said...

"To me, Keynes and the Post Keynesians are more perceptive than Austrians both about what is seen and what is not seen in a capitalist economy."

Paul Davidson's "THE ECONOMICS OF IGNORANCE OR IGNORANCE OF ECONOMICS?", a review of "The Economics of Time and Ignorance" by Gerald P. O’Driscoll, Jr. and Mario J. Rizzo, is well worth reading with regards to this.

"By the way, Horwitz points out in his second reply to Hill that he is not arguing about theories that describe actually existing economies. He is explaining how the world would work if his utopian blueprint, especially for the banking system was adopted."

This is hardly surprising, as Austrian economics seems to be little more than an ideological justification of capitalism rather than, say, a scientific analysis of how that system works.

After all, their account of the business cycle is rooted in contradiction. Firstly, it implies that the credit market (like the labour market) can be in equilibrium, a state which is denied for other markets. Secondly, it hopes that banks will not act like capitalists (i.e., seek to make a profit by manipulating interest rates and meeting consumer demand). Not to mention Rothbard's suggestion that businessmen, unlike bankers, can foresee the future...

Still, it has utility for the wealthy so I can see why certain of them would fund their think tanks and organisations.

An Anarchist FAQ

Robert Vienneau said...

Gabriel, don't think from my allusion that I have a high opinion of Bastiat. (My copy of his essays is an 1877 translation by David Wells. I mean the book I am holding was apparently printed 131 years ago.)

Iain, I reference that Davidson article in my SSRN paper. You apparently recall it better than I. I see that Davidson, for example, notes a confusion in O and R between liquidity preference and time preference which I also see in Horwitz.

Gabriel said...

You don't [have a high opinion], but the Austrians do. That's what makes it funny.