Wednesday, June 1, 2011

Noahpinion: The Libertarians and their classical theory of the business cycle

Noah Smith likes Arnold Klings "PSST".

Thiis is apparently an old post by Smith, but it's something I've been thinking about a lot lately too.  Like Noah, I too am looking for some new way of thinking about economics.  Begrudgingly, I have to admit I kind of like PSST, though its more the intellectual tradition that the gentlemen over at appeal to that I have an affection for, not the need to find a way to insist that nature should be left to take it's course.

The reason I find PSST oddly appealing is because it is an attempt to revive a branch of economics that has been completely ignored by the dominant paradigm and even by many on the left side of the heterodox spectrum.  The basic theory underpinning PSST--Ricardian comparative advantage--used to be the basis for a much wider discourse before it was forced into the ghetto of international trade theory.  Murry Milgate draws a distinction that may not be well articulated among economists today:
...[I]n order to isolate Keynes's point of departure from orthodoxy, we will need to know from which economic theories the conventional wisdom was derived.  As is well-known, Keynes did not help to clarify matters by grouping together under one heading "The Classical Economists", not only Ricardo, but also Marshall,. Edgeworth and Pigou (J.M.K. vol. VII. p. 3, n. I).[1]
Milgate goes on to draw a distinction between the "Classical Economists": Smith, Mathus, Ricardo and Marx and the "Marginalist" or neo-classical economics of Marshall , Walras and Pigou.  The modern economist--of course--worships at the alter of the marginalists.  Kling, who's denomination is prone to apostatize for the wrong reasons has touched a nerve for all of us who sit in the back pews.

Kling may be looking in the right place, but his world view dooms him to find the wrong answer.  He wants PSST and comparative advantage to save the free market.  The simple trade theory microeconomics of it dictate that everyone does what they are most efficient at doing and so there is no need to interfere.  The intellectual tradition of Ricardo is not so one dimensional.  The textbook model of comparative advantage is one that only has labor and a very simple, inert, returnless technology.  Once you add profits to the classical model things get a little less... natural.  Underlying the relations of capital and labor in the classical theory is a political economy.  Unlike the Marginalists' idea that labor and capital get paid their market determined marginal products for the classical theorists power relationships determine the rate of profit and the wage rate.  Notwithstanding the "natural" floor to the wage rate: Malthusian subsistence.  Milgate quotes an Adam Smith Kling has probably never heard of: "though in disputes [over wages] with their workmen, masters must generally have the advantage, there is however a certain rate below which it seems impossible to reduce for any considerable time, the ordinary wage of the lowest species of labour."[2]

Oh man, now we are nowhere near where Professor Kling wants to be, now we are in Marxian territory.  Marx comes out of the classical tradition and his theory of the business cycle in many ways is an attempt to fill in the distributional theory left by the earlier classicals.  Marx takes for granted that there is no wiggle room in which workers and capital can share in the surplus but rather offers a law of pure subsistence wages ground in power relationships not Malthusian population swings.

The classical theory of capital also does not anchor capital in its marginal product or in a long term full employment.  Employment is determined by the stock of capital but there is no mechanism that guarantees the full utilization of the capital stock will mean full employment.  Murry Milgate's book, which I have quoted heavily from is actually part of a strain of intellectual inquiry that has tried to ground Keynes' insights in a tradition other than the Marginalist school, whose market mechanisms cannot abide by a truly Keynesian longterm under-employment.  It is an intellectual movement touched off by Joan Robinson's simple challenge to the Marginalists of her generation to actually define what capital, the return to capital and the production function actually are.  A question to which Samuelson et al could not come up with a satisfactory answer.

I'm way ahead of myself here.  This is something I have only begun to explore myself.  This being a more complex issue than just appealing to an undergraduate sense of comparative advantage I fully admit this will take some time to sort out.  So let me slow down and propose a book club:

Robinson, Joan. The Production Function and the Theory of Capital. Review of Economic Studies Vol 21(2) 1953-1954

Walsh, Vivian and Harvey Gram Classical and Neoclassical Theories of General Equilibrium Oxford University Press, Oxford. 1980.

Sraffa. Piero. Production of Commodities by Means of Commodities. Cambridge University Press, Cambridge, 1960.

I'm calling it a book club because I have not fully digested these books though I do see the challenge to the marginalist orthodoxy which I feel has reached its natural conclusion in General Equilibrium models in which there is not even a difference between the long run and the short run.  I don't think a return to the classicals is a panacea or should replace marginalist thinking entirely  but I do think it is a legitimate way of looking at the economics we have all been taught from the outside.  As well, while most economists seem to find indeterminacy terrifying I think the lack of a clear theory of distribution of income is kind of attractive in a post industrial world where labor does not share in productivity gains.

The Blogosphere is faster than I am.

So there are a couple of things I want to address.  First, what modeling PSST means.

Arnold Kling:

Instead, I think that the right tools to use for macro are the two-country, two-good models of international trade. The "two countries" could be two sectors within an economy. The agricultural sector and the urban sector. The housing sector and the non-housing sector. And, of course, there are many more than two sectors. But just talking about two sectors is a big improvement over the GDP factory.

The problem here is that the two good two country model is absolutely not the right model for macro.  I mean, if you want to enshrine an uber efficiency of markets then yeah two goods/two countries will work.  Ignoring the question of what two goods and what two sectors, there is an additional sleight of hand here because if you try to model say three goods and three countries with Ricardian comparative advantage you get inefficient outcomes [3]  though you can get around this problem to a degree by proposing an infinite continuum of goods [4].  This actually brings me to my next point.  I disagree with Noah (and agree with Brad Delong--sort of--on this point) that you need an overly complex model to make this work.  You can probably model fairly effectively with an infinite amount of "i"s and "j"s.  Alternatively, a comparative advantage theory is not mutually exclusive with, say, sectors that produce C, G, I and EX.  Actually, come to think of it, I would be fine with a two sector comparative advantage model if Kling would concede gains from trade between a government sector and a not-government sector.

Anyway, I want to stress that I'm not overly concerned with reasserting a supply side theory of the business cycle.  The appeal of this classical line of thinking to me is that it gives us a different way of thinking about the distribution of output between capital and labor.  As well, it is an alternative way of thinking about the dynamics of investment spending beyond it being a passive response to interest rates.  And, the more modern direction of classical economics has been in the direction of trying to justify Keynes without the need for sticky wages and an inevitable return to a long run.

One more thing about Brad Delong's dismissal of PSST (which I largely agree with) in which he lumps in Marx and Hayek as liquidationists.  He also does this in his excellent Seven Sects of Macroeconomic Error  which is required reading now for my Macro 101 students.  While I appreciate the rhetorical genius of it I feel like it obscures the important difference between the two.  While both Marx and Hayek do assume that liquidation is inevitable, Hayek thinks it is desirable while Marx does not.  For Hayek, the divine hand of the market will solve its own problems.  Marx--in the classical tradition--recognizes that the solution to the problems of the market are political.  Delong is right to point out essentially Marx goes too far in assuming there is no political solutions[5] besides revolution but it is still an important fundamental difference in the "models" of the two economists.

[1] Milgate, Murry (1982) Capital and Employment Academic Press, New York  p36
[2] Milgate p37
[3]  McKenzie, Lionel W (1954). Specialization and Efficiency in the World Production, Review of Economic Studies21(3)
[4]Dornbusch, R., S. Fischer, and P. A. Samuelson (1977).  Comparative Advantage, Trade, Payments in a Ricardian Model with a Continuum of Goods, American Economic Review, 67(5)
[5] Since I'm doing footnotes anyway,  here I'm thinking of the work of Acemoglu and Robinson who have a great "Marxian" model in which you don't have a revolution or even much political disruption before "the bosses" make concessions : Acemoglu, Daron and James A. Robinson (2000) Why Did the West  Extend the Franchise?  Democracy, Inequality and Growth in Historical Perspective, The Quarterly Journal of Economics 115(4).