Shift happens indeed



The Chronichle of Higher Education has a very good piece on the 50 year anniversary of Thomas Kuhn's classic The Structure of Scientific Revolutions (SSR). It is worth remembering that the official methodology in economics remains Popperian.

The conventional history of ideas in economics, as represented by Schumpeter’s monumental History of Economic Analysis or by Mark Blaug’s work is one that emphasis the continuity and growth of knowledge. Hence, the frontier contains the set of accepted truths and the past is a history of how true knowledge was achieved. Knowledge progresses in a twofold process, first conjectures are made. Conjectures are not the ultimate truth, but are accepted while not proven incorrect. Then, tests are designed to falsify the conjectures. Hence, in Popperian terms theories can never be proven true or verified, but they can be accepted while they are not refuted. In other words, there is rational progress in science, including economics. Falsification is, then, the instrument to measure progress, since it provides the demarcation criteria to define what is and what is not scientific knowledge, and also to determine whether a theory is superior to the other.

Thomas Kuhn provided and alternative to the conventional approach. The evolution of scientific knowledge is tied to history and to the sociology of the scientific community. A paradigm can be described as the set of rules and methods accepted by a scientific community. Scientific knowledge is for the most part the result of normal science, that is, the work of scientists within the confines of their paradigm. However, normal scientists are sometimes confronted with critical anomalies that put in doubt the validity of the whole paradigm (Kuhn’s model of paradigmatic change is the Copernican revolution). An accumulation of anomalies leads eventually to a paradigmatic crisis, and to the emergence of new paradigms.

New paradigms explain the anomalies unexplained by the old one. More importantly, new paradigms often provide a new vision of the object of study. Hence, new paradigms imply that hypotheses are incommensurable in terms of the old paradigm. The methods of testing hypotheses are no longer accepted. Progress is then only possible within the paradigm. Comparisons between paradigms involve the values of the scientific community and the persuasive ability of scientists.

The Kuhnian model was developed to discuss scientific progress in the hard sciences, so a reasonable question is whether it applies to soft sciences too, admitting that economics is a soft science. In economics several paradigms co-exist for long periods, something that should happen only in periods of crisis. Also, the Marginalist Revolution took place even though there was no evident anomaly unexplained by the dominant paradigm (classical political economy; in all fairness Ricardian economics had been abandoned in the 1830s, and vulgar economics dominated the profession). Ideological reasons seem more important in that case. Further, the Keynesian Revolution, that occurred as a result of the high levels of unemployment (an anomaly if there is one for the normal marginalist science) that could not be explained by the dominant paradigm, was to a great extent aborted, and the idea of a natural rate of unemployment was reinstated.

A. W. Coats (1969; subscription required) notes that anomalies and critical experiments are not common, but still the concept of scientific revolutions may serve in economics as an ideal type to clarify the sociological elements in the development of economic ideas. SSR remains vital for economic methodology, in particular after the last crisis has shown, once again, the limitations of the self-adjusting marginalist paradigm.

If you’re surprised, that means that you were part of the problem

Back in July, in the midst of the debt-ceiling debate, Paul Krugman argued that those that were surprised by the GOP tactic of blackmailing the administration, threatening a default in exchange for cuts on social spending and the maintenance of the tax cuts for the rich were part of the problem. Normal was already not part of the GOP. I agree.

Now Krugman tells us that in this crisis a "lot of the blame goes to the economists, by the way, who abandoned what they used to know." But the thing is that the mainstream of the profession has been dominated by the academic equivalent of the Tea Party for a very long time. My point is that if you didn't know that economists forgot certain things about recessions, and never learned a few other things, you have not been paying attention and/or you must be part of the problem too.

Krugman knows this well, since he argued that:

“By the early 1980s it was already common knowledge among people I hung out with that the only way to get non-crazy macroeconomics published was to wrap sensible assumptions about output and employment in something else, something that involved rational expectations and intertemporal stuff and made the paper respectable. And yes, that was conscious knowledge, which shaped the kinds of papers we wrote. So you could do exchange rate models that actually had realistic assumptions about prices and employment, but put the focus on rational expectations in the currency market, so that people really didn’t notice. Or you could model optimal investment choices, with the underlying framework fairly Keynesian, but hidden in the background. And so on.”

That is, in order to publish (in 'respectable' journals) you had to wrap your reasonable assumptions in crazy models. So it should have been clear back then that rational expectations, real business cycles, supply siders, and their political counterparts in the Reagan administration were more dangerous that Old Keynesians and New Old Keynesians (or Old New Keynesians for that matter) were willing to admit.

The problem is not just that New Keynesians of all sorts and political affiliations (Ben Bernanke, Brad DeLong, Paul Krugman, Greg Mankiw, Christina Romer or Larry Summers) can be seen as equivalents to the old Neoclassical Synthesis, the modern equivalents of John Hicks and Alvin Hansen, trying to incorporate the Keynesian insights that lack of effective demand was behind the Great Depression (now our Great Recession), and that fiscal stimulus is necessary, while maintaining the contradictory argument that the price and quantity of all "factors of production", including labor, can be determined by the equilibrium in the labor market. [If this is true lower real wages should equilibrate the labor market and involuntary unemployment should vanish].

From a policy point of view this is certainly important, but it misses the more essential question that Keynes theory was not (at least was not intended to be) about imperfections, and arguably the inability of the Neoclassical Synthesis of overcoming that original contradiction is part of the reason of the rise of New Classical economics, and the acceptance by New Keynesians of the Friedmanian notion of a natural rate.  Can you blame the profession that believes in the self-adjusting nature of the system towards the natural rate (included in all New Keynesian models) that fiscal stimulus is only needed in the short run and that the economy is on its path to recovery?

Hansen (1938, p. 34), in the book depicted above, said that the profession was: "living in a time when economics stands in danger of a sterile orthodoxy." [The time, by the way, was the 1937-38 recession]. We are in that position again, and people like DeLong and Krugman, as I said before, the best within the mainstream, would miss the opportunity of providing a more solid foundation for economic theory if they do not recognize the importance of the heterodox contributions of the more radical disciples of Keynes and Kalecki. We do not need another Neoclassical Synthesis, and we should try not to miss this new opportunity to complete the Keynesian Revolution.

Further, although we have our Hansens, so to speak, we do not have our Lauchlin Currie or our Marriner Eccles.  That is, the real heterodox Keynesians within the administration. Currie, by the way, wrote an unpublished review of the General Theory, for the eyes of the Board only, that is far better than most responses in academia, which did not rely in either interest rate (liquidity trap) or real wage rigidity. In fact, Currie argues correctly that (following chapter 19 of the General Theory) falling wages would make things worse. If respectable economists in the mainstream, like Krugman and DeLong, miss this opportunity this period will be remembered as 'the years of low theory.'

PS: For a discussion of Eccles and Currie see here. The classic book on Currie is by Roger Sandilands here.

What ended the Great Depression?

Conventional wisdom contends that fiscal policy was of secondary importance to the economic recovery in the 1930s. The recovery is then connected to monetary policy that allowed non-sterilized gold inflows to increase the money supply. Often, this is shown by measuring the fiscal multipliers, and demonstrating that they were relatively small. This working paper shows that problems with the conventional measures of fiscal multipliers in the 1930s may have created an incorrect consensus on the irrelevance of fiscal policy. The rehabilitation of fiscal policy is seen as a necessary step in the reinterpretation of the positive role of New Deal policies for the recovery.

Dr. Krugman and the natural rate of interest



So I have said a few times here that, while Krugman has been extremely useful for Keynesians in recent policy debates, as a New Keynesian (neo-Wicksellian would be a better term for this school of thought), he is not properly a real Keynesian.  In a recent post he shows exactly my point.  He says:

"There is still a sufficiently low real interest rate that would produce recovery, but it’s a rate that’s hard to achieve."

In other words, there is a rate of interest that would increase investment and bring about the full employment level of savings.  In this post he surprisingly seems to say that liquidity traps or lower zero bound limits (rigidities) for nominal rates do not matter.

The reason seems to be connected to the fact that creditors must have a positive effect on their net wealth in a deflationary balance sheet recession, and their spending should go up.  Hence, creditors should spend more with a slightly lower interest rate.  Wealth effects have been the traditional neoclassical argument for a self-equilibrating economy since Pigou.  If this were true no fiscal policy would be actually necessary.

It's hard to believe that Wall Street bankers would spend sufficiently more for a recovery to follow.  And I doubt that Krugman believes that this effect is sufficiently strong in the real world.  But he does believe in some sort of natural rate, like Wicksell did, which is compatible with Friedman's natural rate of unemployment.

Keynes, on the other hand, thought that the very concept of a natural rate should be discarded.  In chapter 17 of the General Theory Keynes states that:

"In my Treatise on Money I defined what purported to be a unique rate of interest, which I called the natural rate of interest — namely, the rate of interest which, in the terminology of my Treatise, preserved equality between the rate of saving (as there defined) and the rate of investment. I believed this to be a development and clarification of Wicksell’s “natural rate of interest”, which was, according to him, the rate which would preserve the stability if some, not quite clearly specified, price-level. ... I had not then understood that, in certain conditions, the system could be in equilibrium with less than full employment.
I am now no longer of the opinion that the concept of a “natural” rate of interest, which previously seemed to me a most promising idea, has anything very useful or significant to contribute to our analysis."

So, in fact, there might be the case that NO long term rate of interest, a highly conventional one according to Keynes, would be compatible with full employment.  The socialization of investment, in Keynes' terms, then would be necessary.  In other words, effective demand matters in the long run, not just the short run, because there is no tendency for self-adjustment.

Since Friedman's infamous Presidential address to the American Economic Association the neo-Wicksellian approach has dominated macroeconomics.  In this view, the central bank pins the short run policy rate to the long run natural rate, and the economy (save for rigidities and imperfections) moves automatically to full employment.  No fiscal policy is necessary, again with the exception of short run imperfections.  That's why long term considerations about deficits and debt are important. 

My question is: should we be surprised that with this theoretical model as the dominant one, we are in a situation in which the administration is unable to understand and unwilling to promote the fiscal expansion necessary to get us to full employment (or at least lower levels of unemployment)?

PS: Keynes developed the ideas in chapter 17 on the basis of Sraff'a's critique of Hayek's theory of capital in 1932 (here; subscription required).  Note that while Keynes understood that the idea of a natural rate of interest that equalized investment to full employment savings had to be discarded, he did not get that his negatively sloped marginal efficiency of capital actually provided the basis for such a rate.

Mr. Krugman and the Ancients



The preliminary paper posted by Krugman for the Cambridge conference on the 75th anniversary of the publication of the General Theory (GT) is an interesting piece. Not fundamentally for what it says, which is nothing new if you read his blog. Something along the lines liquidity traps imply that you need fiscal policy, and we are at one right now. But it is revealing piece about what mainstream Keynesians understand about the evolution of macroeconomics, and how much knowledge has been lost with the rise to dominance of the neoclassical/marginalist approach.

First, it is important to note that by the time of the New Deal and the Keynesian Revolution American academia was dominated by institutionalism. Mitchell was the head of the National Bureau of Economic Research, John Maurice Clark at Columbia was one of the leading figures of the profession and was the president of the American Economic Association in 1935, and so on. Yes, neoclassical economics dominated in England, with Marshallian traditions in Cambridge, and there were several neoclassical economists in the US, like Irving Fisher, but in America they were still not dominant.

Also, Keynesians or proto-Keynesians like Marriner Eccles and Lauchlin Currie, and institutionalists like Adolph Berle and Rexford Tugwell, were instrumental in bringing a whole generation of economists that where like Clark a mix of institutionalists with Keynesians into the New Deal administration. John Kenneth Galbraith would be the most prominent example. It is important to note that none of these economists thought that Keynes’ ideas were related to wage or interest rate rigidities, or that the problem with the Depression was that wages were too high.

These ideas only became dominant after Hicks and Modigliani, and were popularized in Samuelson’s neoclassical synthesis. In fact, it was the neoclassical synthesis, and not General Equilibrium, that made neoclassical economics the dominant approach in the US. The new economists were trained basically in Marshallian micro (partial equilibrium consumer and production theory) and Keynesian macro (Keynesian cross and ISLM). Old institutionalism started to vanish. [And by the way that is fundamentally, with the addition of natural rate ideas, and a Phillips curve in the macro part, what is essentially taught to undergrads; General Equilibrium is a graduate thing].

Krugman correctly criticizes Barro for the interpretation that the problem for Keynes was high wages, and that monetary policy was the solution. But he says: “if that’s all that it was about, the General Theory would have been no big deal.” Note that neither is a correct interpretation of the GT. Krugman having been trained as an old Keynesian, in the neoclassical synthesis (even if it is very likely that Rational Expectations were already important in his graduate training), is okay with the notion that Keynes believed in some sort of rigidity. For him, contrary to Barro, the main rigidity is not in the labor market, but in the capital market. Namely: a rate of interest that is too low and cannot be reduced further.

I should say here that it is a bit amazing that the discussion does not even include a footnote on Keynes and Pigou effects, and how wealth effects according to the neoclassical authors reestablished the notion of full employment equilibrium, and Keynes’ own views in chapter 19, and Kalecki’s famous reply to Pigou. Not even a footnote on Patinkin’s work on the topic. And that in a nutshell is the problem with Krugman’s paper. He gets stuck with two interpretations the uncertainty version of chapters 12 and 17, and the neoclassical synthesis of chapter 18, but he never bothers with chapter 19, the first in the whole book in which flexible wages and prices are allowed, and still no full employment is reached.

That’s why is weird that he thinks that Keynes didn’t say anything about debt. In chapter 19 Keynes says:

“the depressing influence on entrepreneurs of their greater burden of debt may partly offset any cheerful reactions from the reduction of wages. Indeed if the fall of wages and prices goes far, the embarrassment of those entrepreneurs who are heavily indebted may soon reach the point of insolvency, — with severely adverse effects on investment.”

Debt deflation is integral to Keynes' analysis. And that’s why Krugman has to reinvent what was already known (redundant originality one could call it), but put it into a New Keynesian model (his paper with Eggertsson cited in p. 18), which further complicates the issues, since New Keynesian models assume a natural rate and a tendency to it, that is not reached because of some sort of rigidity. Krugman never learnt the ancients (and I’m not even talking about the surplus approach, but just the old Keynesian tradition).

Don’t get me wrong, as I said about the DeLong in another post, Krugman has been essential to debunk a lot of crazy ideas, and support adequate policies. But it is a problem when the reasonable people in the mainstream still use a model that is basically self-adjusting to full employment.

I would not venture a full explanation of why this happened. But my hunch is that the defeat in the capital debates of the 1960s, admitted by Paul Samuelson, which among other things showed the impossibility of having a natural rate of interest (the existence of a rate of interest low enough that would provide full utilization of capital, something that Keynes knew it was important, saying so in the GT, but was unable to obtain because of his insistence on the concept of a marginal efficiency of capital), the dominant approach turned to eclecticism. This meant that Arrow-Debreu General Equilibrium (a short run equilibrium solution) provided the totem for the idea that markets work and are efficient (freeing the radical market fundamentalists to unleash the Rational Expectations revolution, for which markets always clear), while more reasonable authors (still within the mainstream, like Krugman) developed several research agendas on the basis of finding imperfections.

While one might agree with several policy propositions by the more reasonable mainstream authors (we need fiscal stimulus and the debt ceiling is not a real problem being two of those), it is important not to forget that their theoretical stance is deeply flawed. For more on this see my paper on a forthcoming edited book on the 75th anniversary of the GT.

The age of uncertainty

This video of the old series by John Kenneth Galbraith is fantastic.  The other parts are also available. A little nugget in the begging is the video of John Maynard Keynes.  In an era of gold bugs, a good reminder of why the Gold Standard did not work.