Which Dead Economist Must I Follow?
I contributed recently to the Politics at Warwick blog, which I thank for its hospitality. My post elicited a comment to which I'd like to respond; my response is longer than my original post so I decided to include it here. First I'll put up my original post, dated 13 November 2013. Then I'll quote the comment and respond to it.
How should the economics curriculum respond to the global financial crisis and ensuing recession? Community activists and students have become vocal in this discussion, as recently described by journalist Aditya Chakrabortty and Matthew Watson.
Events have prompted questions about economists’ understanding of financial markets; the same events have generated a deluge of new data. How should economists respond? Economic research has already responded; hundreds of new articles have analysed global imbalances, market efficiency, corporate behaviour, regulation and deregulation, policy rules, the politics and economics of past crises, and the relative fragility of economic and political institutions in history.
The core curriculum has been slower to change. Here are two reasons. The first is that we no longer teach from handwritten notes and a chalkboard; students and teachers demand comprehensive textbooks with instructor manuals, PowerPoint slides, and websites. These take years to develop (and revise). Although slow, change is already visible. Because change is slow, there is more to be done. Change will probably accelerate through initiatives like the CORE (Curriculum Open-access Resources in Economics) project.
A better reason for inertia in the curriculum is our foreknowledge that the full meaning of recent events will take decades to establish – although many people believe that they are already obvious. To illustrate, today we continue to make new findings about the last Great Depression, which began in 1929, although many who lived through the 1930s were so certain of the answers that they were willing to kill and die on that basis.
How should the core curriculum change? A common complaint is that economics is dominated by a single school of “neoliberalism” or “market fundamentalism.” There are calls for more diversity in economics; some students want more access, specifically, to Keynes and Marx.
It is simply untrue that mainstream textbooks reflect principles of market fundamentalism. I can’t think of a principles text that doesn’t follow the initial explanation of market equilibrium with an immediate, detailed discussion of the varied sources of market failure and the regulatory interventions that might follow.
While one may learn from both Keynes and Marx, what is to be gained from taking them outside their historical settings? A Keynesian model focuses on flows (of income and employment) while neglecting stocks (of capital and debt). Yet capital and debt are very important! Keynesian principles are linked to a model of household behaviour (the “marginal propensity to consume”) that half a century of applied research has comprehensively invalidated. A Marxian model simplifies the continuum of capital ownership into a two-class society; additionally it throws out efficiency and substitution, so distribution is all that remains. In the context of today’s mainstream, each of these is now a stagnant, oxygen-starved backwater.
The importance of competing traditions is much overrated. Those that wish to organize the curriculum around them seem to believe that the major decision each economist must make is “Which dead economist must I follow?” and after that her research findings and policy recommendations will follow. This may be a natural reaction to the fact that mainstream economists are often unenthusiastic about policies that gather widespread popular support, for example rigid immigration controls, employment protection, and double taxation of corporate income. It might be easier for the supporters to say “Oh, those economists are all neoliberals who are ignorant of Keynes and Marx” than work patiently back through the evidence that fails to confirm their biases.
“Economics ought to be a magpie discipline,” writes Chakrabortty. But Economics is a magpie discipline. Most non-specialists – and most journalists – think public and private finance are all we do. They are amazed when I describe the sheer diversity of research that is done just in my department (here and here). We suck up topics and data from any time and place; we don’t care what discipline claims to own them. Then they backtrack and say, “Of course, I didn’t mean to criticize you (or Warwick), I meant Friedman (or Chicago).” The fact is there are no clear intellectual boundaries among schools of thought; we should all mingle in the same fluid mainstream, which is broader, deeper, and faster than you think.
Concluding, Chakrabortty reports a lament for the good old days. Tony Lawson recalls the Cambridge economics faculty in the time of Nicky Kaldor and Joan Robinson: “There were big debates and students would study politics, the history of economic thought.” I remember; I was there too, as a student. The big debates were an exercise in identity politics, not economics. Hostile clashes between intolerant armed camps ended in a war of attrition that benefited no one, least of all students. There is a warning here: be careful what you wish for.
On the day that my post appeared, the anonymous blogger Unlearning Economics posted a comment which you can read in full here after scrolling to the bottom. Here's my response, with excerpts from the comment inset.
Unlearning Economics quotes me and comments:
“A Keynesian model focuses on flows (of income and employment) while neglecting stocks (of capital and debt).” First, Keynes didn’t ignore capital or debt at all; that is simply false ... Keynes carried over some silly marginalist concepts like the efficiency of capital (clearly he mentioned capital once or twice).
My response: It is useful to distinguish between “things Keynes wrote about at various times” and “things that are core principles of the Keynesian model.” Of course we could argue about the division, but it seems to me capital and debt belong to the former, not the latter. The point is exemplified by Keynes’s model of consumer behaviour (more below), in which capital and debt play no role, although they were fundamental to other models available at the time. As for the marginal efficiency of capital, Keynes introduced this to rationalize his discussion of investment (a flow), not to understand the behaviour of the capital stock.
Again, Unlearning Economics quotes me and comments:
“Keynesian principles are linked to a model of household behaviour (the “marginal propensity to consume”) that half a century of applied research has comprehensively invalidated.” Which research would this be? “People don’t consume all of their income” is hardly a false statement.
My response: Yes, it's true that “people don’t consume all their income,” but that isn’t the issue. The issue is whether the main thing in consumption is a stable proportion between household consumption and current income at the margin, as Keynes believed. I should add that he not only advanced this idea in theory but also applied it in practice, for example in his writing about how to pay for World War II. Franco Modigliani, Milton Friedman, and others investigated this idea after the war and failed to find it in the data. They did identify a stable relationship between consumption and wealth, or lifetime (or "permanent") income, to which changes in current income make a small or negligible contribution. Because permanent income is uncertain, the future (including expectations of inflation and the real interest rate) are fundamental. Modern new-Keynesian models drop the marginal propensity to consume (and the multiplier) and focus exclusively on intertemporal substitution intended to smooth consumption over time. Which takes me on to our next point.
Unlearning Economics adds:
Second modern post-Keynesian is *all about* stock-flow consistent models ... people like Godley, Keen etc have updated [Keynes’s] work.
My response: Yes, certainly. Here another distinction arises, between Keynes and the post-Keynesians (or Marx and the post-Marxists). Naturally, there is development in the Marxian and Keynesian traditions. I could not dispute that, when a great economist produces an insight that turns out partial or incomplete or defective in some respect, it may be fixable. There is evolution. Evolution has produced neo-Keynesians and post-Keynesians (and Marxists). Evolution is better than stagnation. At some point it generates new species. Every year I help teach a “new-Keynesian” model of the macroeconomy to Warwick undergraduates, and every year I (and I hope they) learn something new. Yet the label “new-Keynesian,” like George Box’s economic models (more below), although useful, is also wrong. Would Keynes recognize it as his? It’s debatable. Does it matter? Only if you want to claim ownership over Keynes’s legacy.
Unlearning Economics quotes me and comments:
“A Marxian model simplifies the continuum of capital ownership into a two-class society; additionally it throws out efficiency and substitution, so distribution is all that remains.” Marx spent a lot of time talking about the difference between financial capital and industrial capital; Lenin updated his work in the context of ‘globalisation’ (imperialism). I also have no idea how you got that Marx “throws out efficiency and substitution”: he continually praised the efficiency of capitalism, and substituting capital for labour is the main driving force behind the tendency of the rate of profit to fall.
My response: I considered proposing that “efficiency of capitalism” belongs to “things Marx wrote about at various times” rather than “things that are core principles of the Marxian model.” On reflection, however, I am not convinced that Marx had a concept of efficiency at all, not in the sense that economists use it today (when you can’t reallocate resources without using more of some input or consuming less of some output). Marx did write a lot about the productiveness of capitalism, but I do not think he was thinking about total factor productivity.
Similarly, I do not think Marx had a concept of substitution in the sense of a choice between alternatives that varies with their relative price. Yes, it’s possible to rewrite Marx’s idea of the organic composition of capital (the constant/variable capital ratio) as capital/labour substitution, but that is not at all how Marx described it. His idea (Capital III, chapter 13) was that over time the ratio of constant and variable capital in value terms will tend to rise, but that seems to be driven by the accumulation of capital, not by relative factor prices. What is described is a process that piles capital up faster than labour; there is no process that allows substitution between them along a frontier. With labour the only source of surplus value, the rate of profit on capital must tend to fall.
Since Unlearning Economics charges me at the outset with “complete ignorance of Keynes and Marx,” I thought I had better come up with a test of Marx’s understanding of substitution. When I teach the subject of economic warfare I put a lot of emphasis on substitution. People who are not mainstream economists (military commanders, for example) often make biased predictions about the effectiveness of blockades and sanctions, and this is because they lack a concept of substitution. They expect a blockade to curtail production and so to cause the rapid downfall of the blockaded economy. In history, the curtailment of trade has generally brought about relative price changes that stimulate substitution and in turn this will make the blockade less effective than expected. So I looked to see how Marx wrote about blockades.
In 1861 the American Civil War led to a blockade of the Southern ports. The Confederacy responded with a cotton embargo; they thought this would trigger such a crisis in Britain’s textile industry that London would have no choice but to intervene on the Confederacy’s behalf. In the autumn of 1861 Marx wrote (in several places including here, for example) predicting that the stratagem would succeed: “England is to be driven to break through the blockade by force, to declare war on the Union, and thus to throw her sword on the scales in favour of the slave-owning states.”
It did not happen. In Britain the price of raw cotton shot up (which Marx could see), and this stimulated a search for alternative supplies, soon found in Egypt and India (which Marx entirely discounted). He had made the characteristic error of someone who does not get substitution.
Back to George Box, who wrote that “all models are wrong, but some are useful.” The models you can find in Keynes and Marx are no different; they are all wrong in the Boxian sense. Are they useful? Sometimes, yes. Marx’s idea of surplus extraction is useful for understanding societies with closed elites and extractive economies, although not modern capitalism. Keynes’s idea of animal spirits gives insight into modern capitalism as a nice corrective to the idea of rational expectations. But, why should I, or you, or anyone confine themselves to the limits of any “wrong” model by declaring that I am a Keynesian or a Marxist? When scholars do that, it surely tells us more about the politics of identity than about their scholarship.
In fact, I fear that the tone of this discussion may exemplify my final point: when social science is polarized into schools of thought, s/he who is not with me is against me and personal rancour is the likely product.