Simon Wren-Lewis, Robert Waldmann, and Paul Krugman have all recently devoted additional space to explaining – ruefully, for the most part – how it came about that New Classical Economics took over mainstream macroeconomics just about half a century after the Keynesian Revolution. And Mark Thoma got them all started by a complaint about the sorry state of modern macroeconomics and its failure to prevent or to cure the Little Depression.
Wren-Lewis believes that the main problem with modern macro is too much of a good thing, the good thing being microfoundations. Those microfoundations, in Wren-Lewis’s rendering, filled certain gaps in the ad hoc Keynesian expenditure functions. Although the gaps were not as serious as the New Classical School believed, adding an explicit model of intertemporal expenditure plans derived from optimization conditions and rational expectations, was, in Wren-Lewis’s estimation, an improvement on the old Keynesian theory. The improvements could have been easily assimilated into the old Keynesian theory, but weren’t because New Classicals wanted to junk, not improve, the received Keynesian theory.
Wren-Lewis believes that it is actually possible for the progeny of Keynes and the progeny of Fisher to coexist harmoniously, and despite his discomfort with the anti-Keynesian bias of modern macroeconomics, he views the current macroeconomic research program as progressive. By progressive, I interpret him to mean that macroeconomics is still generating new theoretical problems to investigate, and that attempts to solve those problems are producing a stream of interesting and useful publications – interesting and useful, that is, to other economists doing macroeconomic research. Whether the problems and their solutions are useful to anyone else is perhaps not quite so clear. But even if interest in modern macroeconomics is largely confined to practitioners of modern macroeconomics, that fact alone would not conclusively show that the research program in which they are engaged is not progressive, the progressiveness of the research program requiring no more than a sufficient number of self-selecting econ grad students, and a willingness of university departments and sources of research funding to cater to the idiosyncratic tastes of modern macroeconomists.
Robert Waldmann, unsurprisingly, takes a rather less charitable view of modern macroeconomics, focusing on its failure to discover any new, previously unknown, empirical facts about macroeconomic, or to better explain known facts than do alternative models, e.g., by more accurately predicting observed macro time-series data. By that, admittedly, demanding criterion, Waldmann finds nothing progressive in the modern macroeconomics research program.
Paul Krugman weighed in by emphasizing not only the ideological agenda behind the New Classical Revolution, but the self-interest of those involved:
Well, while the explicit message of such manifestos is intellectual – this is the only valid way to do macroeconomics – there’s also an implicit message: from now on, only my students and disciples will get jobs at good schools and publish in major journals/ And that, to an important extent, is exactly what happened; Ken Rogoff wrote about the “scars of not being able to publish stick-price papers during the years of new classical repression.” As time went on and members of the clique made up an ever-growing share of senior faculty and journal editors, the clique’s dominance became self-perpetuating – and impervious to intellectual failure.
I don’t disagree that there has been intellectual repression, and that this has made professional advancement difficult for those who don’t subscribe to the reigning macroeconomic orthodoxy, but I think that the story is more complicated than Krugman suggests. The reason I say that is because I cannot believe that the top-ranking economics departments at schools like MIT, Harvard, UC Berkeley, Princeton, and Penn, and other supposed bastions of saltwater thinking have bought into the underlying New Classical ideology. Nevertheless, microfounded DSGE models have become de rigueur for any serious academic macroeconomic theorizing, not only in the Journal of Political Economy (Chicago), but in the Quarterly Journal of Economics (Harvard), the Review of Economics and Statistics (MIT), and the American Economic Review. New Keynesians, like Simon Wren-Lewis, have made their peace with the new order, and old Keynesians have been relegated to the periphery, unable to publish in the journals that matter without observing the generally accepted (even by those who don’t subscribe to New Classical ideology) conventions of proper macroeconomic discourse.
So I don’t think that Krugman’s ideology plus self-interest story fully explains how the New Classical hegemony was achieved. What I think is missing from his story is the spurious methodological requirement of microfoundations foisted on macroeconomists in the course of the 1970s. I have discussed microfoundations in a number of earlier posts (here, here, here, here, and here) so I will try, possibly in vain, not to repeat myself too much.
The importance and desirability of microfoundations were never questioned. What, after all, was the neoclassical synthesis, if not an attempt, partly successful and partly unsuccessful, to integrate monetary theory with value theory, or macroeconomics with microeconomics? But in the early 1970s the focus of attempts, notably in the 1970 Phelps volume, to provide microfoundations changed from embedding the Keynesian system in a general-equilibrium framework, as Patinkin had done, to providing an explicit microeconomic rationale for the Keynesian idea that the labor market could not be cleared via wage adjustments.
In chapter 19 of the General Theory, Keynes struggled to come up with a convincing general explanation for the failure of nominal-wage reductions to clear the labor market. Instead, he offered an assortment of seemingly ad hoc arguments about why nominal-wage adjustments would not succeed in reducing unemployment, enabling all workers willing to work at the prevailing wage to find employment at that wage. This forced Keynesians into the awkward position of relying on an argument — wages tend to be sticky, especially in the downward direction — that was not really different from one used by the “Classical Economists” excoriated by Keynes to explain high unemployment: that rigidities in the price system – often politically imposed rigidities – prevented wage and price adjustments from equilibrating demand with supply in the textbook fashion.
These early attempts at providing microfoundations were largely exercises in applied price theory, explaining why self-interested behavior by rational workers and employers lacking perfect information about all potential jobs and all potential workers would not result in immediate price adjustments that would enable all workers to find employment at a uniform market-clearing wage. Although these largely search-theoretic models led to a more sophisticated and nuanced understanding of labor-market dynamics than economists had previously had, the models ultimately did not provide a fully satisfactory account of cyclical unemployment. But the goal of microfoundations was to explain a certain set of phenomena in the labor market that had not been seriously investigated, in the hope that price and wage stickiness could be analyzed as an economic phenomenon rather than being arbitrarily introduced into models as an ad hoc, albeit seemingly plausible, assumption.
But instead of pursuing microfoundations as an explanatory strategy, the New Classicals chose to impose it as a methodological prerequisite. A macroeconomic model was inadmissible unless it could be explicitly and formally derived from the optimizing choices of fully rational agents. Instead of trying to enrich and potentially transform the Keynesian model with a deeper analysis and understanding of the incentives and constraints under which workers and employers make decisions, the New Classicals used microfoundations as a methodological tool by which to delegitimize Keynesian models, those models being insufficiently or improperly microfounded. Instead of using microfoundations as a method by which to make macroeconomic models conform more closely to the imperfect and limited informational resources available to actual employers deciding to hire or fire employees, and actual workers deciding to accept or reject employment opportunities, the New Classicals chose to use microfoundations as a methodological justification for the extreme unrealism of the rational-expectations assumption, portraying it as nothing more than the consistent application of the rationality postulate underlying standard neoclassical price theory.
For the New Classicals, microfoundations became a reductionist crusade. There is only one kind of economics, and it is not macroeconomics. Even the idea that there could be a conceptual distinction between micro and macroeconomics was unacceptable to Robert Lucas, just as the idea that there is, or could be, a mind not reducible to the brain is unacceptable to some deranged neuroscientists. No science, not even chemistry, has been reduced to physics. Were it ever to be accomplished, the reduction of chemistry to physics would be a great scientific achievement. Some parts of chemistry have been reduced to physics, which is a good thing, especially when doing so actually enhances our understanding of the chemical process and results in an improved, or more exact, restatement of the relevant chemical laws. But it would be absurd and preposterous simply to reject, on supposed methodological principle, those parts of chemistry that have not been reduced to physics. And how much more absurd would it be to reject higher-level sciences, like biology and ecology, for no other reason than that they have not been reduced to physics.
But reductionism is what modern macroeconomics, under the New Classical hegemony, insists on. No exceptions allowed; don’t even ask. Meekly and unreflectively, modern macroeconomics has succumbed to the absurd and arrogant methodological authoritarianism of the New Classical Revolution. What an embarrassment.
UPDATE (11:43 AM EDST): I made some minor editorial revisions to eliminate some grammatical errors and misplaced or superfluous words.
Uneasy Money 
David: “The importance and desirability of microfoundations were never questioned. What, after all, was the neoclassical synthesis if not an attempt, partly successful and partly unsuccessful, to integrate monetary theory with value theory, or macroeconomics with microeconomics? But in the early 1970s the focus of attempts, notably in the 1970 Phelps volume, to provide microfoundations changed from embedding the Keynesian system in a general-equilibrium framework, as Patinkin had done, was to provide an explicit microeconomic rationale for the Keynesian idea that the labor market could not be cleared via wage adjustments.”
YES! That point needed making. And YES! to your following paragraphs too. That recognition of the importance and desirability of microfoundations was there in the GT too.
Slightly off-topic, but I think you might be the best blogger to write something on this question: “I don’t disagree that there has been intellectual repression, and that this has made professional advancement difficult for those who don’t subscribe to the reigning macroeconomic orthodoxy,…”. Was that equally true in the 1940’s, 1950’s, and 1960’s, when Keynesianism was the reigning macroeconomic orthodoxy? My sense is that it was.
LikeLike
Thank you, I like your approach. While there is clearly a problem, trying to explain the source of the problem, particularly as Krugman does with ideological agendas is a distraction, and more or less irrelevant.
New Classical econ is either advancing the general understanding of the economy or it is not, (net); it has profound flaws or it does not. Causation need not be explained, it certainly doesn’t need to be speculated over – that’s a sideshow.
It is always far more devastating and effective to tackle the actual shortcomings of a flawed argument or approach than it is to be sidelined into speculative debates about why such a flaw exists and is pursued by smart people.
I have a simple question. I think it is related. How do microfoundation include ‘government’ in their models – on a microfounded basis?
Who is the representative agent for determining the needed allocation of resources to those functions of government seemingly everyone agrees are government necessary functions? (not to mention who gets to decide which functions are necessary? on what microfounded basis?
In general, I have never seen a argument or explanation for how underlying structural levels of government should be optimized. I see lots of opinions, particularly on marginal participation, or cyclical participation.
I suppose the basic question is, taking just education as an example, something like, how much government spending today on education maximizes the future potential economy? How exactly is a rational agent supposed to figure that out? How is a rational agent supposed to determine what his optimal level of taxation is? government benefits are?
Everyone seems to either hate the government or like the government or tangentially accept the government’s role in ‘institutions’. I haven’t ever seen anyone just rigorously define the beast’s role in the economy.
Thanks for another fantastic post,
Dan
LikeLike
is it microfoundations per se that is the problem ? Recall John Bryant (“Keynes-type RE model, QJE 1983) who has a micro-founded model that gives recessions. As he says, almost anything can be modeled as optimal rational behavior. The New Classicals were allowed to get away with imposing a particular approach to microfoundations. People like Cooper and Diamond and Bryant and Roger Farmer showed microfoundations is really not the issue
LikeLike
Microfoundations is not analogous to physics, it is analogous to alchemy.
LikeLike
Dan has a good point. All the microfoundations I have seen assumes government benefits are limited to transfer payments. If you build a libertarian counter-factual assumption into the model, you are not going to get anything sensible out of it.
LikeLike
Nick, Glad that you approve!
About intellectual repression, I agree that there was also intellectual repression in the Keynesian Revolution. I think that is partly the result of the self-interest of those in the new reigning orthodoxy. What seems special about the New Classical Revolution is the extent to which the losing side actually collaborated in their own repression by buying into the methodological precepts promulgated by the winning side. Also even during the heyday of the Keynesian Revolution, there were dissenting voices at major outposts, especially Chicago, but even in places like Harvard and Princeton who resisted the Keynesian tide. I think that there is less of that now than there was then, and my conjecture is that anti-New Classical faction was cowed by the spurious methodology precept of microfoundations. It would be worthwhile to investigate the similarities and differences in the reactions to the Keynesian and New Classical Revolutions, but that would require more reading and research than I have time for right now. Why don’t you ask David Laidler what he thinks about all this next time you talk to him.
Dan, Thanks for your comment. I think the public choice people have tried to do work along the lines that you suggest. In addition, Gary Becker wrote a paper in the 1958 “Competition and Democracy.” Anthony Downs wrote a famous book An Economic Theory of Democracy and Mancur Olson wrote The Logic of Collective Action and Buchanan and Tullock wrote The Calculus of Consent. So there is a considerable literature on that subject. Earl Thompson probably went the farthest in arguing that democratic decision making is very likely to be efficient and came up with ingenious arguments to explain why apparent violations of economic optimality by democratic legislatures (until the legislatures are overtaken by ideological activists) are almost always efficient responses to some form of externality not taken into account in private competitive markets. His book Ideology and the Evolution of Vital Institutions (with Charles Hickson) brings together the many strands of his work into a single volume.
Charles, I am not against providing microfoundations. I think that is a good strategy, and it has produced some good results, as you point out. What I am opposed to is making New Classical style microfoundations the only methodologically acceptable way of doing macroeconomics.
root_e, I agree microfoundations is not physics. I don’t agree that it’s alchemy, but I sympathize with your frustration.
LikeLike
Thank you for another great post.
Reductionism is a seductive force in and of itself and may explain, as you say, why most economists conspired with microfoundations regardless of ideology.
As an aside. Your article caused me to re-think my purely reductionist position on the mind-brain question. Can’t have it both ways; reductionist in one field but not in another .
LikeLike
Economic is independent of state in his origins as Adam Smith said. And this is the way walked for both ricardian and marxists to resolve economics. But if you see the new paradigm in Biology by Lynn Margulis, there is a feedback by totality on individual processes. Evolution to form a biological cell was for symbiotic group of several individual very specialized and this groups form the cell base for modern cells. States and groups that have influence in states are not considered in microfoundations. If you see economy till middle XIX century, bondage and groups specialized in bondage were determining economic behaviour and not the individual preferences. Paralleling, groups specialized in gross exploitation of indian, chinese, latin american,asiatic poverty are determining politic of USA, England and absorbing individual preferences of this part of world population with local military dictators or democrats intimidated by USA power. Actual economic science can’t predict economy behaviour.
LikeLike
Off topic. Do you agree with Gross?
http://www.pimco.com/EN/Insights/Pages/For-Wonks-Only.aspx
Maybe some p[ost about this? Thanks
LikeLike
Digital Cosmology,
From the article you reference:
“Assuming that the interest rate on outstanding debt in the U.S. is approximately 4.5% (admittedly a slight stab in the dark because of shadow debt obligations), a Fed governor using this template would want credit to expand by at least 4.5% per year in order to prevent the necessary sale of existing assets (debt and equity) to cover annual interest costs.”
The presumption is that the amount of equity remains fixed over time. That is why existing equity must be sold to cover annual interest costs if credit fails to expand or contracts. But what happens when equity expands? What happens when equity expands enough to overcome credit contraction?
LikeLike
You inspired me to write up a radical thesis for the lack of nominal wage adjustments: there is no microeconomic reason. It’s simply infinitely unlikely to have wage growth distribution incorporate wage adjustments without massive coordination of the economy.
http://informationtransfereconomics.blogspot.com/2014/10/wage-stickiness-is-entropic-force.html
LikeLike
thank you
LikeLike
i just skimmed the post, but i’ve read phelps, lucas etc. and i don’t have much use for krugman (he talks out of both sides of his mouth, a bit like cornel west). but i do see u live in dc (where i grew up, and actually still live—-i read most of this stuff at MLK library in Dc which carries AER, JPE, QJE, etc. and more) and i see you are an ‘antitrust lawyer’. that seems to be an upcoming paradigm—rather than trust, we should promote antitrust (i’ve been partly practicing this since i was a preteen—i don’t trust anyone really). i went to some event a couple weeks ago and went to the wrong place. an FCC lawyer said he was going to call the police on me so i said i’m leaving which i did. he is involved in ‘net equality’ (or neutrality if you’re agnostic) and i actually was in the protest which took over the streets downtown in ancient history (nov 6 2014). maybe 1 person got arrested—but it wasn’t a protestor—it was someone in a car who was pissed off for being obstructed in traffic and hit someone. s/he had 5 cop cars immediately .
LikeLike
ps. my main point which i forgot was i think hegemoney of new classical econ is probably the same dynamic as makes say, christianity, islam, hinduism, or in biology ‘oak trees’ or maples or spruce , or various animals hegemonic. i think this actually comes out of chaos theory (math/physics)—the gaussian distribution/CLT is the easiest, but actually there is a large hierarchy of limit theorems( see s kleene in math logic, ‘degrees of complexity’, ‘posts’s problem’ all on wikipedia, etc. also, jens bjorneboe (friend of arnae naess who i have no use for even if i was in EF! but i guess we a product of his times)
LikeLike