Tuesday, May 01, 2018

How Has Economics Failed?

The Financial Times is having a debate about whether economics has failed. The first interchange is here, with some followups here. (I happen to have two tabs about neoliberalism open at the moment as well.)

Mainstream economics is a failure in so many dimensions that its failure cannot be characterized shortly in any comprehensive way. For example, I am not going to discuss funding sources and economics role as a system justification. Even so, you might find this post long and wandering.

As you can see, I do not take the claim to be that non-mainstream, heterodox economics has failed. In the comment section, for one FT page, Percy Pavilion writes, "So Marshall, Keynes, Kahn, Kaldor and Harcourt failed did they?" I think any opinion on whether Keynes, Kahn, Kaldor, and Harcourt were failures to be irrelevant to the topic. Was it Joan Robinson, commenting on the claim that it is all in Marshall, said something like, "The problem is that the opposite is also in Marshall, too"?

Anyways, I want to focus on the failure of economic theory. I think any well-trained economist in some field knows how to tweak assumptions to get any results that they want. Start with a model of perfect competition. Add an information asymmetry, a transaction cost, a search cost, sticky prices and a Calvo fairy, some monopoly or monopsony, whatever. Then you can argue that some policy will lead back to efficient markets. Or, if so inclined, that government failure prevents any some implementation.

Given this openness to such tweaks in mainstream economics, why is Sraffian economics or some other variety of heterodox economics not more widely accepted? Most of the time, when I or others put forward a model with Sraffa effects, we are not arguing about the implications of an imperfection. Rather, the argument is that the model of perfect competition does not work the way that mainstream economists teach. And this is usually in an open model with no way to resolve class interests in some utopian world. To understand such models, it is helpful to have some understanding of the history of political economy and alternative theories of value and distribution. As Mariana Mazzucato notes at the FT, mainstream economists are trained to be ignorant of such topics.

What to make of empirical research? Some of it, say, on malarial nets, is by researchers striving to be useful. One can raise questions about external validity and whether atheoretical research is possible or desirable. But this is a different direction from my comments above.

But how does the theory relate to the supposed empirical turn? Are textbooks updated to state that some theories (e.g., skills-biased technical change) have been rejected by the evidence? Can you point to an introductory textbook that includes a section on behavioral economics and empirical evidence? (I expect "Yes" to be an answer to this question.) But how many experiments decide between contrasting theories? It is my feeling that the ability to tweak a theory includes an ability to maintain it as consistent with any empirical evidence that shows up. And much empirical work takes a particular approach for granted, without testing it. I like reading Marshall Steinbaum, but will cite him as somebody that does not realize how many effects he takes as a consequence of monopsony might be consistent with Sraffian price theory, before adding in failures of competition.

I am unsure how to take a lot of applied research. I do not think tariffs on steel and aluminum should be imposed on the whim of an ignoramus. I suppose one can use Leontief matrices to trace through the effects of such tariffs on the automotive industry, aluminum cans and the beverage industry, etc. And I suppose that one can apply game theory after the fact to rationalize, say, the Chinese reaction. But can one say beforehand whether soy bean farmers in the mid west or Boeing executives trying to sell Dreamliners should have more to worry about? And is not the question of the long term effects of undermining the World Trade Organization more a qualitative question for historians that data-driven economists? I want to take analysis of the possible effects of the exhaustion of North Sea oil reserves on sustainable social spending in Norway as useful. But I suspect that if I look carefully, I will find theoretical errors embedded in many of the equations in supposedly applied research.

I am amused that Maurice Obstfeld shows up in the FT interchanges, when I have just demonstrated the failure of what Krugman and Obstfeld teach in one edition of their textbook on international trade. I am aware that the boundary between mainstream and non-mainstream economic theory is not necessarily well-defined. Nevertheless, I do not expect most mainstream economists to be able to conduct a reasonable conversation about the topics in this post. At the FT, Diane Coyle, Tony Yates, and Tim Harford exemplify my expectation. And much of what I am saying is old hat. You can see some of what I am saying as echoing Christian Arnsperger and Yanis Varoufakis. Or even Robert Solow.


Blissex said...

First of all my usual quote that illustrates why "internal consistency" is much better for the career of an Economist than "external consistency":

«I found myself sitting next to a very likable young middle-aged academic tenured at an elite British university, whom henceforth I will refer to as Doctor X and whose field is closely associated with this blog. ... Every year I publish papers in the top journals and they’re pure shit.” Doctor X, who by now had had a glass or two, felt bad about this, not least because “students these days are so idealistic and eager to learn; they’re really wonderful.” Furthermore Doctor X could and would like “to write serious papers but what would be the point?” ... The amount of funding Doctor X’s department receives depends not on how many papers or their quality its members publish, but instead on in which journals they are published. The journals in Doctor X’s field in which publication results in substantial funding will not publish “serious papers” but instead only “pure shit” papers, meaning ones that merely elaborate old theories that nearly everyone knows are false. Moreover, even to publish a “serious paper” in addition to the “pure shit” ones could taint the department’s reputation, resulting in a reduction of its funding. In any case, no one at a top university would read a “serious paper” because they only read “top journals.”»

That's totally realistic.

Blissex said...

«helpful to have some understanding of the history of political economy and alternative theories of value and distribution. As Mariana Mazzucato notes at the FT, mainstream economists are trained to be ignorant of such topics.»

As usual I point to "internal consistency" with JB Clark's "three fables" as the goal of Economists.

Secondarily, there is the Second Theorem of Welfare that is used by those Economists to claim that distribution and growth or the cycle are independent, so Economists can just consider growth or the cycle, and Economics need not be "political economy", and leave distribution to politicians. But then of course the Second Theorem of Welfare is based on ridiculous assumptions and is subject to the usual Theorem of the Second Best, so it is just a ridiculous excuse.

Blissex said...

«And much of what I am saying is old hat. You can see some of what I am saying as echoing Christian Arnsperger and Yanis Varoufakis. Or even Robert Solow.»

Please leave Y Varoufakis out of this, as while interesting and intelligent, many of his arguments are thinly veiled propaganda or warmed up "vulgar" keynesianism.

For alternative approaches R Solow is good, but also S Keen and less recently H Minsky and R Kindleberger, and of course JK Galbraith in a different way. For older works that bearded 19th revolutionary had interesting insights, but more recently J Robinson whom you mention, R Khan, Kalecki, whom you also mention, plus some random italian and french authors.
P Sraffa did not really provide a theory, he just illustrated that even the simplest static models don't work as the neoclassicals claim, when thoroughly ridiculous misrepresentations like the "production function" are removed.

Emil Bakkum said...

Hello. If you want to promote a certain school in economics, it makes sense to downgrade the other ones. However, if you want to understand reality, it is best to keep an open mind. It is wise to learn from all schools, and to assume that other people are not ignorant or perverse fools. I think that an understanding of a single model does not help in understanding reality. You need the whole collection of models in order to guess what may be possible. Sraffa may have shown that the neoclassical paradigm has some faults at the macro level. But I still think that the neoclassical theory excels in giving a coherent picture of economics as a whole. She is great at the introductory level. Incidentally, my few books about Sraffa are theoretical. I would be nice to see statistical estimates about the importance of reswitching etc.

Robert Vienneau said...

Emil, you might notice that Blissex, for example, is arguing for a plurality of models. I may come across as harsh on contemporary neoclassical economists. But when I read historical figures who laid down this mistaken approach, I try to first take them as on good faith.

I am even willing to work out an exercise for a neoclassical model. I will be intrigued to find out if Soon Ryoo has a better approach to Piketty, in a tradition I think more worthwhile.

I have previous catalogued empirical evidence on Sraffa effects. Stefano Zambelli has a contribution published earlier this year.