This makes me step back and think about the whole econ blogosphere over the past several years. The thing that inspired me to start blogging was the famous Paul Krugman article, "How Did Economists Get It So Wrong", published in September 2009 (that, and John Cochrane's dismissive response). Before that article, I had seen the econ blogosphere as mostly about micro, and mostly about "everyday economics" - pointing out cute little observations about how daily life reflects economic behavior and incentives. It was kind of interesting, but not really. Then, suddenly, it became something very different. A war was on. The blogosphere was Ground Zero for a very deep and fundamental argument about the purpose and practice of macroeconomics. And suddenly, the policy consequences couldn't be more important. After spending a year sitting on the sidelines of that argument, I decided to turn Noahpinion - which had just been a personal bullshit diary - into an economics blog.
Now it seems that the war is winding down. Arguments still flare up over the proper use of microfoundations, or over monetary policy. But - and this is just a feeling I get - the frequency of titanic clashes seems to have peaked.
Undoubtedly, this is just because of the economic situation. We're not in a boom by any means, but we're no longer in a crisis, or even obviously stagnating. Policymakers have emerged from their foxholes; they no longer feel a need to deviate from the comfortable pre-2008 consensus that monetary policy is the only necessary tool of demand management. I believe that this is the reason for the "fall of Keynesianism" described by Farrell and Quiggin.
So does that mean Krugman's insurgency failed? Seen narrowly as a push for countercyclical fiscal policy, I'd have to conclude "yes." Even Obama's "stimulus" bill was mostly just temporary tax breaks, not spending increases as Keynesian theory would recommend. All around the developed world, instinctive fear of debt made the policy recommendations of Krugman and other Old Keynesians a political non-starter regardless of their theoretical justification.
What about in academia? The crisis provoked a mini-boom in macro papers focusing either on Keynesian theory (e.g. Eggertsson & Krugman 2010) or empirics (e.g. Nakamura & Steinsson 2011). But I don't see this as a big deal. The theorizing is just DSGE with a couple frictions; the profession has not yet felt a collective need to overturn the basic methods and philosophy of macroeconomics, as Robert Lucas and others did in the 1970s. The analytical framework that emerged from that 1970s revolution, which remains absolutely dominant in macro to this day, has again and again proven itself flexible enough to build models to suit any passing fad or consensus opinion. This is not to say that DSGE can accurately describe any phenomenon - merely that it is capable of telling a mathematical story that delivers basically any desired policy conclusion. So after fiscal policy goes out of vogue, Old Keynesian models may go back to near where they were pre-crisis - subjects of study for smart people, but not accepted by the main body of the profession. The consensus may shuffle toward the ideas of Hall and Eggertsson and Krugman, but the change will be marginal.
Like the wars of Louis XIV, the push for a rehabilitation of Old Keynesianism has resulted in a lot of sound and fury, but only modest territorial gains.
Nor do I think Krugman's push against microfoundations will provoke a return to aggregate-only models. First of all, that assault was always a bit half-hearted - in principle, microfoundations are highly desirable for any macro model. The problem is more with the poor (but convenient) choice of microfoundations that macroeconomists have been willing to accept since DSGE came into vogue.
So a layperson might conclude that the insurgency that Krugman launched in September 2009 - and which has consumed and defined the econ blogosphere since then - was ultimately defeated. But I do not believe that this is the case. Although the battle for New Old Keynesianism is mostly over, the Krugman insurgency launched a much deeper, more profound, and more long-lasting war. It shook the philosophical foundation of macroeconomics, and that foundation is still shaking.
Since 2008, everyone had been asking: "Why did macroeconomists miss the crisis?" Even the Queen of England asked it! Shouldn't we expect macro theory to help us avoid macroeconomic disasters? Of course, the profession at first closed ranks against the criticism. Economists protested that Rational Expectations or the Efficient Markets Hypothesis made it impossible to predict a crisis. These protests mostly fell on deaf ears (and rightly so, because they are logically fallacious). Still, the reflexive wagon-circling made it hard to pin down exactly where economists had gone wrong - after all, if all the experts insist that experts have value, who are non-experts to disagree?
But when Krugman, a Nobel Prize winner, came out and said publicly that the macro profession had allowed itself to be satisfied with uselessness and irrelevance, it broke the facade of unity. Like Greg Smith departing Goldman Sachs, here was an insider who was willing to stand up and say that the whole system was rotten. And then Krugman went further. After revealing that top economists were dissatisfied with macro's ability to predict crises, Krugman revealed that they also couldn't agree on how to deal with crises. That's where the push for Old Keynesianism came in. It may not have resulted in a permanent sea change in macroeconomists' modeling consensus, but it told the public that there were deep divisions within the profession on the question of how to fight recessions.
And that, really, was all the public needed to know. If macroeconomists hadn't conclusively discovered how to avert crises and also hadn't conclusively discovered how to recover from crises, what good had they done for society? Why were we paying professors hundreds of thousands of dollars to study this subject if nothing usable had emerged?
Of course, it would be wrong to paint the challenge to macro as a one-man Krugman Show. It isn't. Even the stalwarts of the profession have been questioning how much macroeconomists really understand - see John Cochrane and Greg Mankiw, for a couple of examples. But it was Krugman who took this argument public, who took the case to the wider educated lay populace, and aired macro's dirty laundry to millions of engineers, scientists, financiers, businesspeople, politicians, lawyers, and journalists. What Krugman (and Brad DeLong) did to macro was similar, in some ways, to what Lee Smolin and Peter Woit did to string theory - except on a much bigger stage, since Krugman is such a huge name in his field, and macroeconomics has a lot more important policy ramifications than the theory of black holes.
So the battle over Keynesianism may be over, with the New Old Keynesians fought to a bloody standstill, but the wider Macro Wars have only begun. As for how this affects the blogosphere and the rest of econ's public face, one thing is for sure - we're not going back to talking about how abortion affects crime rates.
(Side note: Since my PhD will be done soon, I think I may turn this post into a book this summer, if I have the time...)
Update: Krugman comments, and notes that A) were it not for the New Old Keynesians, damaging austerity might have been implemented to a greater degree, further harming the world economy. That's a good point; policy is all about the balance of forces. Krugman also says - and I absolutely agree with this - that tere is a deep rumbling of dissatisfaction and disillusionment with the current paradigm among grad students and assistant profs, which will eventually build into a large wave of change. The Krugman insurgency was the beginning of the new Macro Wars, not the end.
"in principle, microfoundations are highly desirable for any macro model."
ReplyDeleteWhy is that? And I mean that as a serious question. My training is in organic chemistry, and chemists are highly promiscuous about models, using ones like Lewis dot structures which are absolutely preposterous from a micro--i.e., quantum mechanics--level. Only when the model fails do they bring in some stripped-down version of the relevant physics (like the explanation for the Diels-Alder reaction, if you're curious).
Real scientists seem to have a much healthier view of models. Getting fundamental Theories of Everything is all well and good, but working scientists throw them over the side the minute they're can be swapped out for something simpler that gives good predictions.
Economics seems to view self-consistent, mathematically-rigorous, microfounded models an end in themselves for little reason other than aesthetic appreciation. Am I off base there?
"In principle" is the operative term here. In a perfect world, microfoundations would capture everything. But the world is not perfect and too much insistence on microfoundations might ruin it for everyone.
DeleteMaybe since I am just not in the field I don’t get it, but why are you saying Keynesianism is dead?
ReplyDeleteSeems to me like the austerity approach is pretty much not working anywhere. I suppose we are recovering, but ridiculously slowly.
So I am not sure what you are referring to when you say the “Krugman Insurgency” failed? Was there a meeting of economists somewhere that came to this conclusion? Or has it just become obvious that Keynesianism is wrong? Or do you just mean in terms of policy making that Keynesianism isn’t making headway?
Given the relative ineffectiveness of monetary policy you cannot say that Keynesianism is wrong.
ReplyDeleteI'm with Krugman.
The parade may have passed but the Emperor is still naked.
"in principle, microfoundations are highly desirable for any macro model."
ReplyDeleteIs there a single example in physics for where a micro model was successfully developed before a decent phenomenological macro description was available?
It would be much more confidence-inspiring if you were in a position to say something like "the partial macro models XYZ have done rather well in this crisis, now what I would like is to find appropriate micro underpinnings for which there is independent evidence".
Maybe the recent Eggertson-Krugman paper is actually in this vein?
Is there a single example in physics for where a micro model was successfully developed before a decent phenomenological macro description was available?
ReplyDeleteYes! Weather forecasting. Understanding the rules that governed particles and gases vastly improved our ability to use powerful computer models to predict the weather, above and beyond what had been possible with heuristic descriptions of weather phenomena.
Keynesianism has two sides, investment to get you out of the crisis, then paying off the debt when the crisis has passed.
ReplyDeleteEconomists and policy makers completely ignored the second part of that equation during a boom that lasted over a decade and generated unprecedented government revenues. That so many developed nations ran a deficit during the boom is a disgrace. You cannot reject Keynesian principles when times are good and run back to them when the inevitable downturn happens, the result is... well... this.
Is there a single example in physics for where a micro model was successfully developed before a decent phenomenological macro description was available?
ReplyDeleteI don't think Noah's example of weather prediction works but Einstein's explanation of Brownian motion probably qualifies.
As a non economist and not American (i am a physicist) who keenly read Krugman, Sumner, MMT theories and whatnot I would say that the problem is not whether Krugman has won or lost in the economics profession but that the economics battle is irrelevant.
ReplyDeleteThe more this crisis proceeds the more I think that the political science and sociology has a lot more interesting inputs than economics, because the issues, especially in Europe, proved not to be a technical miscalculation on the part of governments and financiers but a political struggle to extract as much rents as possible. I would accept the plausibility of a theory that NGDP targetting could solve all our problems last year, but today I see more clearly that it's simply an issue of political will. Even if NGDP targetting could do that, the interests of Merckel or the Republicans are not aligned with those that suffer due to the crisis.
The issue is much deeper than whether countercyclical policy can stop the suffering, the issue is to establish a political order that actually wants to stop the suffering and finds extensive unemployment as something inherently evil and not 'just deserts'.
I had seen the econ blogosphere as mostly about micro, and mostly about "everyday economics"
ReplyDeleteWell, if all you were reading were clever wankers like Cowen and Landsburg it might have seemed that way. Plenty of other bloggers were talking about the bubble and its implications.
It was kind of interesting, but not really. Then, suddenly, it became something very different. A war was on.
The econ blogosphere didn't interest you until a little internal theoretical spat flared up? And the gigantic bubble and the possible implications of it breaking didn't? WTF?
in principle, microfoundations are highly desirable for any macro model.
ReplyDeleteThe mathematical tools to go from micro foundations to macro predictions probably do not exist. The physicists are still struggling with Navier Stokes and have to resort to computer approximations for something as simple as the three body problem.
"The econ blogosphere didn't interest you until a little internal theoretical spat flared up? And the gigantic bubble and the possible implications of it breaking didn't?"
ReplyDeleteEconomists are trained to be uninterested in the economy.
Anyways, I wonder how widely laypeople perceive the argument between MIT and Chicago, or freshwater and saltwater, or whatever you want to call it, as an argument between Tweedledum and Tweedledee. The interesting work is elsewhere, and economists are trained to be ignorant of it.
In a different political context, Solow and Hahn argued against dominant trends in macroeconomics about 15 years ago. They had no impact on the club.
Your book idea is a very good. I think you could write an excellent lay-accessible explanation of the melee within the profession, which would shed a lot of light on the heat.
ReplyDeleteAnonymous:
ReplyDelete"The more this crisis proceeds the more I think that the political science and sociology has a lot more interesting inputs than economics, because the issues, especially in Europe, proved not to be a technical miscalculation on the part of governments and financiers but a political struggle to extract as much rents as possible. I would accept the plausibility of a theory that NGDP targetting could solve all our problems last year, but today I see more clearly that it's simply an issue of political will."
Yes. The whole 'financial crisis' start to finish has been a *political* struggle.
Economics is the ideological jargon used to misdirect attention away from the rude fact of political power. That's why it pays well, by academic standards.
The problem with economics, as I see it, is that a very simple idea - "Now what should we do?" - has morphed into a theology of nonsense.
ReplyDeleteA very simple and straigtforward picture of American economic history can be connected directly to the biologic patterns of life. See e.g. Albers & Albers, "The Golden Mean, The Arab Spring and a 10-Step Analysis of American Economic History,"- a peer-reviewed article in the Middle East Studies Online Journal, August 8, 2011. You did not need a PhD in economics to predict or understand the crisis.
Hi Noah. Australia reacted to the crisis with government stimulus. I think Asian governments have no problems with the idea that fiscal policy has a role in macroeconomic stability.
ReplyDeleteTo Noah
ReplyDeleteI suspect the real rason macro keeps on the same path is that most practitioners never took the models seriously. It is alarming how uninsulted new Keynesians are when DSGE theory is attacked. I think most feel that the profession requires that sensible arguments must be reconciled with rational intertempral maximization. Since this is always possible, no one is blocked by the norm. It also leads to nonsense whenever economists addicted to the approach discuss questions not related to their personal research (it is always possible to reconcile DSGE with reality but it takes time, so, when in a rush, DSGE addicted economists say things wildly inconsistent with reality).
I am more extreme than Krugman, but just this morning I was passing time working on a model in which people decide whether to move to a city by forecasting the effects of congestion in the future (maybe after centuries). It's fun (and sometimes publishable) to see what strange things one can make a micro founded macro model do by changing a simplifying assumption which is plainly false but which some once hoped might be innocent and harmless.
I predict we will keep playing the game until the next crisis finds us completely unprepared again.
To Ryan Cooper. I think chemists work with models based on early 20th century theory (Lewis dot) because they work. Macroeconomists don't have models which work. So we insist on elegant theory. It is all we have.
There is another war brewing too, about the relations between institutions, elite capture, inequality and democracy. Look forward for new insights into, and new interest in, political economy.
ReplyDeleteDevelopment economists have struggled with these issues for a decade of two, however, as the European Rim keeps collapsing, their ideas are destined to turn 'home'.
With the growing #Occupy movement in the US, they may have an audience there too.
If you are going to write a book and are leaning towards the heterodox then consider tackling Joe Schumpeter's creative destruction as implemented by the 1954 tax code elimination of dividends tax credits (yes it was Republicans that started the dividends double tax - you might ask why) and lowering of capital gains tax rates, supposedly to direct capital to IPOs, seed, venture, and growth but in practice we got more M&A with less internal investment and coy start-up funding (angel investors says it all). This is wrapped into the expansion of finance, both in the consolidation of commercial banks and the growth of the investment banking, PE, and hedge funds. I am just a lay person (software engineer) but from my 44 years of work in corporate America it looks to me like incentives that were intended to make US corporations more competitive just made them big and relatively less competent over time.
ReplyDeleteThe chart on page 8 and the following narrative agree with my thinking, but the authors do not go there but invite further research.
http://www.econ.ucdavis.edu/faculty/amtaylor/papers/w15512.pdf
This gated paper written by Patricia L. Bryan on "Leveraged Buyouts and Tax Policy" seems to support my point, but I was too cheap to read past the first page.
http://heinonline.org/HOL/LandingPage?collection=journals&handle=hein.journals/nclr65&div=57&id=&page
You cannot discuss this crisis and the response without a discussion of values. Krugman places high value on wealth equality and the policies that built the Great American Middle Class. Krugman is worried about the erosion of the middle class and the rise of wealth inequality.
ReplyDeleteKrugman's concerns and values are not shared by all economists. Many economists believe that individuals all benefit from maximizing their personal self-interest. They value the concerns of the elites and BigF because these are the people who can fund research and offer the greatest personal awards. They place high value on personal wealth accumulation and are not terribly concerned about wealth inequality.
There are other economists who are most interested in the economics of development and how to transition the poor into wealthier more productive contributing members of society.
The differences in VALUES are reflected in the differences in the policies they recommend. The models emphasize maximization of different sets of values.
Some models may be very good at maximizing certain types of values but perform very poorly in maximizing others.
Values are at the root of the disagreement. Values are difficult to discuss because they raise questions of fairness that are at the core of the psyche of the human social animal. We evolved as a social animal and our evolution includes an innate sense of fairness that is necessary to preserve social cohesion. Wealth inequality violates the innate sense of fairness and is a very unpleasant subject for people who have access to more wealth than is fair.
Values are at the core of the dispute and in part drive the models.
-jonny bakho
I dunno if the Keynsianism insurgency has failed, but the market monetarist insurgency is certainly taking off.
ReplyDeleteThis is the 1954 tax code Anonymous again.
ReplyDeleteI know that tax based investment incentives are not considered macro-economic. However, Charlie Munger pointed out in his 2003 speech to UCSB economics under grads:
http://www.tilsonfunds.com/MungerUCSBspeech.pdf
"Economics was always more multidisciplinary than the rest of soft science. It just reached out and grabbed things as it needed to. And that tendency to just grab whatever you need from the rest of knowledge if you’re an economist has reached a fairly high point in Mankiw’s [post.economics.harvard.edu/faculity/mankiw/mankiw.html] new textbook [Principles of Economics, www.amazon.com/exec/obidos/ASIN/0324168624/tilsoncapitalpar]. I checked out that textbook. I must have been one of the few businessmen in America that bought it immediately when it came out because it had gotten such a big advance. I wanted to figure out what the guy was doing where he could get an advance that great. So this is how I happened to riffle through Mankiw’s freshman textbook. And there I found laid out as principles of economics: opportunity cost is a superpower, to be used by all people who have any hope of getting the right answer. Also, incentives are superpowers."
My interpretation is that the long run outcomes of "investment" incentives that promote consolidation of corporate power and wealth are "steroidal" in promoting growth, but also "tumours" and other maladies of disproportion. The opportunity cost of misallocation of capital over decades is collosal. So, aside from bakho's somewhat moral interpretation of values, this is an economic value problem as well. Unsustainable distortions in capital allocated to productive investment and the attendant capture by financial intermediation has both reduced agregate demand and agregate productive capacity. Along with a strong dollar policy by virtue of foreign dollar reserves maintained indepent from trade balance FOREX and we are efficiently undermining our own national economy. I am all for wealth created by investment in the factors of production (which we still do offshore at discount rates), but not for casino finance.
In any case, good luck young man. Thankfully, "in the end we are all dead." (Keynes)
Noah, I think the Macro War moved to other field.
ReplyDeleteEurope, USA and Japan are trying to return to pre-2008 with austherity policies. Adn Krugman is right when says that will not work. Recession is coming too to USA. europe and Japan are currently at recession.
BRICS (South-Africa too is a member...) aparently are moving to State Capitalism. China is State Capitalism and I see Brazil moving at that direction (it will be less State Capitalism than China, but more State Capitalism than it is today... Brazil is at the first stages of preparation for a currency war, governement said want lower the Real value at more than 50%).
I think that around 2020 we will see who won: Europe, USA and Japan or BRICS...
Delenda Rome...
What has failed is an effective mutiny of the neoclassical paradigm given the navel gazing in academia and the irrelevance of neoclassical thinking particularly at the macro level; economics is a religion and the priesthood is the AEA with people like Mankiw, Lucas and Barro acting as guardians.
ReplyDeleteI was a chemical engineer in a previous life (before corporate globalization took hold and effectively gutted manufacturing)and as any process engineer will tell you scaling up what works in the lab will not work in the plant. Move this to the realm of economic "sciences" and one sees the micro foundations pablum existing unfettered along with a lot of other dubious ideas. When I saw the YouTube video of recent Sveriges Riksbank Prize winner --it isn't a real Nobel-- Christopher Sims defend DSGE at an INET conference I thought why could not the economics profession embrace real dynamic modelling that is stock-flow consistent given the computing power at our disposal today? Oh wait! That would admit that there is disequilibrium --something that scientist and engineers live with-- rather than equilibrium in markets. Divorcing politics from economics as Jevons, Marshall et al succeeded in doing in creating neoclassical economics has not succeeded in making it a science although the layperson is wont to believing this.
Conservatives understood long ago that they were fighting an unending war. They know that losing a battle -- even destroying careers of their own political champions -- today is of no importance, as long as they gain a better position for tomorrow.
ReplyDeleteThe important thing is to move the goal-posts.
Krugman is anathema to conservatives because he has moved the goalposts against them. A lot.
Liberals may not recognize Krugman's action over the last few years as a "victory"... but conservatives certainly know it has been a "defeat"!
This story is mostly myth and fantasy -- about the nature of the econoblogosphere, about the scientific problems of macro, about Krugman's role, etc., etc.
ReplyDeleteIt's inaccurate and shallow stuff.
Keynesianism of all types are the failures.
ReplyDeleteOf course, only academics don't recognize this because they live in their little bubble.
This story is mostly myth and fantasy -- about the nature of the econoblogosphere, about the scientific problems of macro, about Krugman's role, etc., etc.
ReplyDeleteIt's inaccurate and shallow stuff.
How so, in your opinion?
Ah anonymous trolls in the comment section. You must be onto something.
ReplyDeleteThe book is a great idea. During the crisis every major advanced nation enacted governmental stimulus. The U.S., Europe, Canada, Japan, China, etc. After the crisis past, they reverted to form.
The non-Keynesians are driven by one impulse. They are anti-government. Bush, Hank Paulson, Bernanke all were Keynesians in a fox hole as you write. When the chips are down is when it matters.
What Ryan Cooper said.
ReplyDeleteNoah, you're leaving out some obvious and important things.
ReplyDeleteWhat are the incentives in the economics professoriate?
It's clear now that for the Big Macro Boyz that being wrong in the service of the elites is no crime, or even shame, while being correct against the wishes of the elites is a crime.
The profession doesn't have the incentives to get it right.
Aelilea asked if there was a case in physics where a micro-scale model predated a macro-scale model. Noah put forward weather prediction. This is ridiculous. Aelilea did not ask if later, micro-scale models were better. People have been predicting the weather using atmospheric pressure for hundreds of years, and it has worked pretty well.
ReplyDeleteAbsalon disputed weather prediction and proposed Brownian Motion. This is a much better suggestion. However, decades before Einstein, Brownian Motion was described statistically with no understanding of the underlying physics.
None of this really matters. The point of the original question was a good one: are economists pretending their field is more of a hard science than it really is? Are they hiding behind complexity when they misunderstand the basic phenomenological aspects of their science?
Economics makes my brain hurt. It seems to me that some prominent micro-foundation economists seem to misunderstand as simple a concept as Ricardian Equivalence because they have their heads crammed way too far up their complex models.
Noah, in my view the debate was never about Keynesianism. Or to be more accurate, it makes much more sense to describe the war as a war between Popperian economists and anti-positivist economists, because if you describe it as a debate on keynesianism you fail to take into account people like Scott Sumner who are not Keynesians but supported similar policies as Krugman.
ReplyDeleteFor much of the last 30 years, you basically had economists who believed their theories are provisionally true only if they make correct predictions on the one hand and economists who thought their theories are true if most other economists agree they are true.
In this sense, Popperians won big time in the US and anti-positivists won big time in Germany. In the 80s if you pointed out that, for instance, even sticky-prices RBC models would have failed to predict the most significant economic events in the century, you would have been laughed at, NOT because you were Keynesian; you could be very Keynesian (like Stiglitz and Krugman) as long as you based your arguments on math rather then empirical evidence.
This is not the case any more. Pretty much everyone agrees that we rank models based on the accuracy of their predictions; in this sense we only keep the microfoundations that help our models make better predictions. In this sense the Keynesians won, but only to the extend that they won the right to claim that their models were more useful in predicting reality.
Contrast this to what happened in Germany; with a few (very few) notable exceptions, German economists don't think models need to have any predicting power. It's not that they dislike Keynesianism (to my experience they dislike Monetarism much more), they dislike the notion that economics should be held to the same standards as other sciences. This by the way, explains why German economists have very few publications in top journals, while German doctors rule very competitive fields such as oncology: If you don't get your philosophy of science right you are bound to fail in every science, for obvious reasons.
@Ryan Cooper
ReplyDelete"Why is that? And I mean that as a serious question. My training is in organic chemistry, and chemists are highly promiscuous about models, using ones like Lewis dot structures which are absolutely preposterous from a micro--i.e., quantum mechanics--level. Only when the model fails do they bring in some stripped-down version of the relevant physics (like the explanation for the Diels-Alder reaction, if you're curious)."
Because in economics your particles are human beings who think and then decide what to do, and any effect of what you do depends crucially on whether people anticipate it or not, and whether they expect it to be permanent or not. In a way in natural sciences all parameters are structural.
Your background is physics yet you decided to study Economics. I hope you do not get caught up in the false attempt to model human behavior as if it were knowable and predictable. Science attempts to be precise and predictable. Economics is anything but science regardless of what the academics desire. Wonderful blog though and I will try to come by and add my two cents from time to time.
ReplyDeleteI'm also a chemist (organic/biochemist). It seems to me, and is indeed completely obvious to any scientist, that a model with no predictive value is a model without value. However, reading economics blogs, it appears that many economists do not share this belief, as far as I can see. Before, I thought that economics wasn't very good at predicting because it is hard. After reading you guys for awhile, there seems to be a troubling lack of interest in even being able to predict.
ReplyDeleteI was watching a NOVA episode on behavioral economics and the interviews with Fama etc. really brought this to the fore. He seemed to think it preposterous that we should even expect his discipline to be able to predict even enormous, Earth-shattering events such as the present crisis.
Well, we do; and if you can't, you're pretty much worthless. I don't care if it's aggregate based or microfounded, it should just work. If it doesn't, your ONLY interest should be in making it work. Isn't anything else a complete waste of time?
I read Krugman because he does make predictions, and he has been right about alot of things. He said there was a housing bubble (mostly by referring to other people's work) and by God it popped. He predicted that austerity in Ireland and Britain would lead to stagnant economies and they look pretty stagnant to me. He did not predict the subsequent financial meltdown and he presented a paper describing that as a significant failure, basically because he didn't know very much about the financial system.
I haven't seen Fama, Cochrane, Mankiw and the rest of the Hee Haw gang predict anything that mattered. Now please tell me why we should look to microfoundations. I don't get it. It would be like statistical thermodynamics that didn't account for entropy. It would be worthless and no one would use it for anything. Somebody would come up with a better model and get themselves a well deserved Nobel Prize.
conesnail
Krugman is influential because his analysis comports very well with events. In addition to that his motivation has a basis in what most people regard as traditionally valid morality. That is, it's better to have some concern for other people than to not have that concern.
ReplyDeleteBut he's fighting a rather powerful constituency, the economic & political elite. I do think his contribution will tend to grow in as much as the internet allows information to flow in unprecedented ways. But it takes time.
As for microfoundations-- seems to me the accumulation of error is a killing problem.
conesnail, you are right in your observation that many economists still don't accept that models should be judged on the basis of predictive value, however:
ReplyDeleteFirst, you are referring to academic economists and in fact a very small minority of them, concentrated in a very specific demographic (professors older than 60 at a freshwater school). All other economists are either explicitly Popperian (like Krugman, or the marginal revolution guys). Keep in mind that all private sector economists (except those that produce research for the clients of their firm) care EXCLUSIVELY about the predictive power of economic models.
Second, yes there are indeed guys like Fama, but keep in mind that his research is not very related to macroeconomics, but rather finance and theory of the firm. Also Fama is one of the "good guys" considering that his research on market efficiency is based mostly on empirical evidence, not microfounded theory. He's not an anti-empiricist like Prescott or Lucas, he's just a really bad macroeconomist.
Krugman is influential and smart but frankly my two main issues are that he waffles (although often he carefully threads the needle) and too often comes down on the nanny-state side.
ReplyDeleteOn macro policy he advocates fiscal policy -but in fact, he has advocated the fed to do more many many times. He never said QE is ineffective. Fiscal policy in an NK model is effective "when not offset by monetary policy." The case for fiscal policy is when the Fed can't, or wont, commit to closing the output gap. He's advocated that the Fed should do much more, even endorsing an ngdp target. His graphs of "fiscal policy working" are just as much an indictment of the central bank... and i can show you his own words (and NK models) to prove it. Same with DeLong.
Emphasis on the "cant" because the feds own research says QE has lowered the "virtual" nominal fed funds rate to -2%, the real rate being even lower.
So why focus on fiscal policy and not continually emphasize both fiscal and monetary?? Why is fiscal better than monetary? How come he is not bashing the Fed like he did the BOJ back in the day? (to be fair he's done a fair bit of Trichet bashing).
And when it comes to health insurance accepting for a minute that we will have universal coverage, why shouldn't consumers be exposed to pricing and why shouldn't the market be decentralized (Sweden or Japan come to mind). Krugman advocates Medicare-for-all.
Seems to me on every issue, there are good arguments to be made on both sides yet Krugman comes down consistently on the ultra-liberal govt-can solve-all your problems side.
now, i have my theories about why he does not bash the Fed like he did the BOJ, but ultimately I don't really see a "failure" of New Keynsianism - in fact I somewhat see a sucess because we've proved there are really deep nominal rigidities in the economy.
the difference is that i just don't see a huge appetite for big govt spending (ironically, most people got sick of that under Bush) which, fortunately, still leaves foom for the FOMC to manage AD shocks (also an NK idea).
What Insurgency ?
ReplyDeletewho cares about arcane details of macro models: untill we have economists yelling, tax the rich, tax cap gains, it seems like how many complex equations can you fit on the head of a pin
ps: your captchas are way, way to hard
Those capcha's are *really* hard to read.
ReplyDeleteNoah: "Basically, Keynesian policy briefly regained its old throne when everyone was panicking in 2009 - everyone became a Keynesian in a foxhole, as Bob Lucas would say - but this brief consensus fell apart under an assault from austerity-minded European central bank economists in 2010-11."
An other way to say this is that when getting it right mattered, the Chicago Boyz used Keynesian models, because the work. When getting it right either didn't matter, or they wanted to get it wrong, they abandoned them.
Double, of course, for Stephen 'Inflation!!!!!!!!!!!!!!!!' Willamson.
The problem with physicists:
ReplyDeletehttp://www.smbc-comics.com/comics/20120321.gif
"Doctor, we have this pet physicist and... something's wrong. It keeps babbling about linguistics and neurology and climate science."
ReplyDeleteThe reason natural sciences have mainly moved from macro to micro is that the macro world is the accessible one, while getting at the micro world is invisible to the naked eye and has required increasingly powerful instruments to study. The LHC is still trying to fill in some holes in the microfoundations. In economics it's not so clear whether the same should be expected.
ReplyDeleteThat said, DSGE certainly looks nothing like a microfounded model from any real science. Rational expectations isn't really a microfoundation; it's a way to handwave away the issue that it's quite difficult to specify what human behavior should be when making decisions about the future based only on presently available information. Big problem that everybody on Wall Street would love to have the solution to, econ doesn't have the solution, so... just assume the problem is solved! Constrain your choices of models to only examine those that can be solved as a closed-form equation, and make sure the equation describes a world of people who know that equation. Apparently you can throw enough complications into DSGE to get whatever result you want, but that doesn't mean it's flexible enough. You can't assume a world of rational agents if you can't provide a solid theory of how to be a rational agent in the real world. An equilibrium model with a closed form solution can't describe a world with butterfly effects. That possibility shouldn't be assumed out of existence; deterministic chaos can be found in far simpler systems than the economy. (Incidentally, the logistic equation shows that deterministic chaos can be virtually guaranteed even in some systems that have an equilibrium solution.)
One reason often cited for believing in rational expectations in a world of humans of known psychological weaknesses is that even if most people aren't rational, the ones who get it right will control a larger and larger fraction of the market, so ultimately the market as a whole will behave rationally. This seems intuitively wrong to me; if you have a housing bubble going on, for as long as it continues, it is those who believe in the bubble who see their fortunes grow, while the ones who sit it out control less and less of the market, at least until it pops. It also occurs to me that if this is the reason to believe in rational expectations, then that is all the more reason to throw it out as an assumption. After all, if the argument is correct, you don't need a model of a world where everyone knows the model, you just need a model of a world where agents act on a variety of models assuming a variety of parameters, and rational-expectations-like results should be an emergent feature of the model. But if it doesn't naturally arise in a world where people disagree on what to expect, then it won't arise in the real world. So do away with it already. Then do away with the idea that you must get a closed form equation, and learn to code in python or something already; everyone in the real sciences needs to these days.
Economics is more analogous to biological sciences than the physical ones so all you physicists and chemists should stop whinging that economics isn't like your disciplines.
ReplyDeleteI'd add an example to my comment about incentives.
ReplyDeleteNoah, Stephen Williamson has been predicting inflation for a couple of years now, has been wrong, has been unconcerned about being wrong, and has no good model to justify being wrong.
Yet you still treat him with respect.
Various short replies...
ReplyDeleteEconomics is more analogous to biological sciences than the physical ones so all you physicists and chemists should stop whinging that economics isn't like your disciplines.
BUT, you can cut open a human body...not so a human economy.
The problem with physicists:
http://www.smbc-comics.com/comics/20120321.gif
Lulz.
ps: your captchas are way, way to hard
Sorry, Google doesn't give me a choice...one of these days I will go in the HTML and change it.
What are the incentives in the economics professoriate?
It's clear now that for the Big Macro Boyz that being wrong in the service of the elites is no crime, or even shame, while being correct against the wishes of the elites is a crime.
The profession doesn't have the incentives to get it right.
I agree with this. But that's a topic for another day (or several other days).
Noah,
ReplyDeleteI was thinking more along the lines of evolution and ecology.
The fact that evolutionary theorists also use game theory should give a clue why economics is more like the bio sciences than the physi ones.
ReplyDeleteOn a related note people interested in economic methodology should read 'The Bounds of Reason' by Hert Gintis.
ReplyDelete*Herbert Gintis
ReplyDeleteNoah have you never heard of MMT. MMT became a serious alternative viewpoint. Even Krugman noticed it. We don't need to worry about debt, because if you have your own currency, you can simply avoid debt.
ReplyDelete"in principle, microfoundations are highly desireable for any macro model"
ReplyDeleteRight. As an economist and an engineer working in complex systems, it seems to me that macro represents the emergent behavior of a system of many units, each of which have microfoundations all right. the aggregate system may or may not act as a big "representative agent" - when there is emergent behavior, it will not. This happens all the time in real life.
Timur Kuran wrote a book about preference falsification which might explain why so many US graduate students and professors pretend to believe in the validity of modern neoclassical macro. Few really believe in it, but they pretend to do so in order to go along with the perceived correct viewpoint. It takes a lot of courage to stand up for your true views, and not many have that courage. In keeping with the thesis, I remain anonymous.
ReplyDeleteYour blog and the impact it has with set-mainstream Economists; inspired me to start a parrallel economics blog as well! Many thanks!
ReplyDeletehttp://uncoveredinterests.blogspot.co.uk/
I think the comparison with Physics is an interesting one. There are massive areas of physics where we still don't have completely provable microfoundations - we can't solve the Schrodinger equation exactly for anything but the simplest atoms, for example. This hasn't held us back because we can still use non-microfounded models - the emphasis then goes into determining the extent of validity of those macro models and testing where they break down - no physicist would use Newton's laws for objects travelling at very high speeds, or classical mechanics for subatomic particles. It seems to me this is the big missing piece in economics - people get hold of an idea and then try to use it everywhere, regardless of whether the conditions are suitable. You see it in the arguments about crowding out in a liquidity trap or austerity during recessions. I guess this is partly due to the polarized nature of the politics, but it seems to me that in the quest for microfoundations economists are running before they can walk. Once you have your macrofoundations stable enough that you can explain when and where each model has validity, you'll be in a much better position to push downwards for an economic theory of everything
ReplyDeleteMore misc. short respones...
ReplyDeleteso many US graduate students and professors pretend to believe in the validity of modern neoclassical macro. Few really believe in it, but they pretend to do so in order to go along with the perceived correct viewpoint. It takes a lot of courage to stand up for your true views, and not many have that courage.
I think this is right.
Your blog and the impact it has with set-mainstream Economists; inspired me to start a parrallel economics blog as well! Many thanks!
Yay!!!
As an economist and an engineer working in complex systems, it seems to me that macro represents the emergent behavior of a system of many units, each of which have microfoundations all right. the aggregate system may or may not act as a big "representative agent" - when there is emergent behavior, it will not. This happens all the time in real life.
Yep!!
It seems to me this is the big missing piece in economics - people get hold of an idea and then try to use it everywhere, regardless of whether the conditions are suitable.
This is exactly right. Many economists totally ignore the concept of domains of validity. The fact that they say lines like "All theories are wrong" reveals how little attention the concept has received...
Noah: "This is exactly right. Many economists totally ignore the concept of domains of validity. The fact that they say lines like "All theories are wrong" reveals how little attention the concept has received... "
ReplyDeleteTo my mind, that statement is far too frequently used to ignore the fact that the speaker's model is more wrong than other models.
I know you must keep your eyes on the prize and can't deviate. But Krugman leading the argument? I see you are quite impressed by his Nobel.
ReplyDeleteWhat about MMT, you really should broaden your horizons now that your Phd is done and the advisors have no more power over you.
Not tooo far south of you is UMKC where lurk the hotbed in this country of MMT and their blog "Neweconomicperspectives" also see one of the grand daddies, Billy Mitchels blog, "Billyblog"
These guys have been looking at the mainstream macroeconomics and critiquing it for years. since the mid 80s I think.
They also have pertinant criticism of models such as your dmsg and krugman's is/lm
sorry DSGE
ReplyDeleteNoah when you say this: "Policymakers have emerged from their foxholes; they no longer feel a need to deviate from the comfortable pre-2008 consensus that monetary policy is the only necessary tool of demand management"
ReplyDeleteI can't help but think of Sumner. I mean no one has in a more high profile way tried to argue that monetary policy is all you need more than him. You dont mention him at all though he would seem to have been the headquarters of the counter insurgency against Krugman.
Do you not mention him as you don't want a war?
It seems to me obvious that we are still in the well (the liquidity trap) with the water at our necks. Only an idiot could not see it happening worldwide. The only way out is to poor more water in, till we reach the border and the water(liquidity) will be spreading everywhere nurturing the ground. At that moment, everything will start growing again and monetarists will be right to reduce deficit spending in order to control inflation.It is this reality so difficult to be understood?
ReplyDeleteWith your unscientific pseudoeconomic rambling Noah, I hope your dissertation committee slaps you in the face very hard. You do not deserve to be called a doctor of economics with your insistence on substituting whiny bullshit opinions for analytical rigour.
ReplyDeleteBegun the macro wars have.
ReplyDeleteWith your unscientific pseudoeconomic rambling Noah, I hope your dissertation committee slaps you in the face very hard. You do not deserve to be called a doctor of economics with your insistence on substituting whiny bullshit opinions for analytical rigour.
ReplyDeleteAt least now I know you're not an American. Whew!
Man, if you got a girlfriend, I'm guessing your trolling would get a lot funnier and more creative... ^_^
It must be annoying when fools write in with abuse instead of anything like a reasoned argument. I hope you can ignore them. My background is physics as well and I find this blog both entertaining and enlightening. I do not believe that macro economics is either easy or should be more like physics except to extent that it honestly recognize its limitations. Some of the assumptions remind to too much of the drunk looking under a lamppost for lost keys. The keys were dropped on the other side of the road but this is where the light is.
ReplyDelete.
If other individual who post to this site can offer some insight I would appreciate it. My question is this:
ReplyDeleteIn Noah's blog "What I learned in econ grad school" (April 29, 2011) he states:
"(C)oming as I did from a physics background, I found several things that annoyed me about the course ... One was that, in spite of all the mathematical precision of these theories, very few of them offered any way to calculate any economic quantity. In physics, theories are tools for turning quantitative observations into quantitative predictions. In macroeconomics, there was plenty of math, but it seemed to be used primarily as a descriptive tool for explicating ideas about how the world might work. At the end of the course, I realized that if someone asked me to tell them what unemployment would be next month, I would have no idea how to answer them. ...
(T)he second problem I had with the course (was): it didn't discuss how we knew if these theories were right or wrong. We did learn Bob Hall's test of the PIH. That was good. But when it came to all the other theories, empirics were only briefly mentioned, if at all, and never explained in detail. When we learned RBC, we were told that the measure of its success in explaining the data was - get this - that if you tweaked the parameters just right, you could get the theory to produce economic fluctuations of about the same size as the ones we see in real life. When I heard this, I thought "You have got to be kidding me!" Actually, what I thought was a bit more...um...colorful."
From everything I've read about macroeconomics, these criticisms are 100% valid.
So may questions, if anyone will answer them, are: (1) ARE these criticisms valid?, and (2) if they are valid, then why is so much professional prestige wrapped up in keeping them going?
Noah writes in the same post:
"But all the same, I absolutely don't blame the grad-level professor for teaching what he taught. Our curriculum was considered to be the state of the art by everyone who mattered. Without a thorough understanding of DSGE models and the like, a macroeconomist is severely disadvantaged in today's academic job market; if he had spent that semester teaching us Kindleberger and Bagehot and Minsky, our professor might have given us better ways to think about history, but he would have been effectively driving us out of the macroeconomics profession.
Thus, DeLong and Summers are right to point the finger at the economics field itself. Senior professors at economics departments around the country are the ones who give the nod to job candidates steeped in neoclassical models and DSGE math. The editors of Econometrica, the American Economic Review, the Quarterly Journal of Economics, and the other top journals are the ones who publish paper after paper on these subjects, who accept "moment matching" as a standard of empirical verification, who approve of pages upon pages of math that tells "stories" instead of making quantitative predictions, etc. And the Nobel Prize committee is responsible for giving a (pseudo-)Nobel Prize to Ed Prescott for the RBC model, another to Robert Lucas for the Rational Expectations Hypothesis, and another to Friedrich Hayek for being a cranky econ blogger before it was popular."
Is this not a devestating critique of an entire prestigious profession?
Will anyone answer these claims and tell me how and where Noah Smith's claims are wrong?
This comment has been removed by the author.
ReplyDeleteI'm 100% for the Krugman insurgency. In fact, 200%. The extra 100% is to pay for the stimulus by taxing the rich and the corporations.
ReplyDelete"Yes! Weather forecasting. Understanding the rules that governed particles and gases vastly improved our ability to use powerful computer models to predict the weather, above and beyond what had been possible with heuristic descriptions of weather phenomena."
ReplyDeleteYeah right. Weather prediction is extremely reliable. It's a fun but useless exercise as far as human beings are concerned. Maybe someday though. in the meantime it's nothing more than an toy.
So the same for mainstream economics. No matter how poorly the theory serves us in understanding economic systems and forecasting outcomes the blood-letters continue to rely on it. After all, it only hurts many billions of innocents, but the practitioners get status and the wealth that comes with it.
You seem to have completely forgotten the 2nd Law of Thermodynamics, which if I recall is a macro theory. Maybe they don't teach that in Physics. That's why economics should be presented as an engineering problem. Then it has a chance of becoming a real science.
Micro doesn't cut it for aggregate economic issues. We live in an aggregate economy - a closed system, but we ignore the closed system constraints.
Micro misses the forest for the trees.