Economics, good and bad

Attacks on mainstream economics such as this by Aditya Chakrabortty leave me hopelessly conflicted.

On the one hand, I sympathise:

 - Some mainstream economists certainly do try and use economic theory to defend elites - as in Greg Mankiw's defence (pdf) of the 1%.

 - It could be that a belief in rational markets is performative, and did help to contribute to the bubble in credit derivatives. This is one message of Shleifer and Mendel's chasing noise (pdf) paper.

 - Any economics course that doesn't teach its students some history of the discipline or some behavioural economics is both a lousy education and a poor preparation to be a practicing economist. One of my old bosses in investment banking used to say that the most depressing phrase in the English language was "I'm a trained economist" - the point being that economists should be educated, not trained.

But on the other hand:

 - As Andrew points out, the fact that most economists failed to predict the crash is actually a vindication of mainstream economics, which says that such things should be unpredictable. I'd add that forecasting isn't part of proper economics at all, so a forecasting error tells us nothing about the merits or not of economics.

 - Some of Aditya's claims, such as trying to blame Black-Scholes, are wrong: it might have been to blame for the 1987 crash, less so the 2008 one.

 - The fact that the recovery from recession was so weak for so long actually vindicates mainstream Keynesianism - which, as Simon says, is "at the heart of any undergraduate macro course."

 - The crisis was a failure not (just) of markets but of organizations. This can be usefully analysed with mainstream principal-agent theory.

Above all, though, I fear that Aditya is missing a bigger point. The division that matters is not so much between heterodox and mainstream economics, but between good economics and bad. I'll give just two examples of what I mean.

First, good economics tests itself against the facts. What makes Mankiw's defence of the 1% so risible is that it ducks out of the empirical question of whether neoclassical explanations for rising inequality are actually empirically valid. Just because something could be consistent with a theory does not mean that it is.

Secondly, good economics asks: which model (or better, which mechanism or which theory) fits the problem at hand? For example, if your question is "should I invest in this high-charging actively managed fund?" you must at least take the efficient market hypothesis as your starting point. But if you're asking "are markets prone to bubbles?" you might not. As Noah says, the EMH is a great guide for investors, but not so much for policy-makers.

It's in this sense that I don't like pieces like Aditya's. Ordinary everyday economics - of the sort that's useful for real people - isn't about bigthink and meta-theorizing, but about careful consideration of the facts.

Comments

[Luis Enrique]

one problem universities face is that if you offer a history of economics module, few student sign up for it. I suppose you could respond it should be made obligatory.

I would qualify your point "economics.. says that such things should be unpredictable" in that economics does not rule out some sage not shackled by group think predicting crises - Steve Keen channelling Minsky can predict stability will end in crisis - but it says that it must have been unpredictable to most people thinking in whatever fashion most people do - by definition few market participants could have predicted it, and the free thinking sage must have been ignored.

perhaps another way of putting your point that the real division is between good and bad economics is that people like Aditya overweight the merits of the heterodox and underweight the merits of the mainstream. To a ridiculous extent, in his case (zero weight on mainstream).

[Luis Enrique]

further ... so this means what economists should really be accused of isn't failing to predict the crisis, it is having failed to prevent it. If we'd been warning that left to their own devices even "sophisticated" Western banks are quite capable of such astonishing feats of self-destruction, they might not have been left to their own devices. Or perhaps if economists had be loudly pointing it out, even the banks themselves might have realized the risk they were taking and not taken it.

[Neil Wilson]

Ordinary everyday economics is about micro considerations. And that has no bearing on the macro economy or its policy.

For the fairly obvious reason that we have no idea what the aggregation function is.

Mainstream economics gets it completely wrong because it is obsessed with clockwork models and their underlying assumptions - layering on epicycles to deal with each anomaly. And then with suggesting policy that forces the real world to be more like their models.

Essentially they study the straitjacketed man, and always require that the straps are tightened to ensure that it conforms more to the perfect straitjacket.

Anybody suggesting that studying what would happen if the straitjacket were removed is treated with derision and scorn.

[Anonymous]

"I'd add that forecasting isn't part of proper economics at all, so a forecasting error tells us nothing about the merits or not of economics."

Perhaps so, but to argue that economics is not a predictive discipline must be wrong.

Even at the most basic levels, economists posit that price and quantity are inversely related, such that quantity demanded reduces in response to a price hike.

Economists get things wrong. Their saying everything was hunky-dory prior to the crash is a case in point.

[Luke]

Economics is like football management - everyone thinks they know better than the professionals. I'm not sure which has the most effect on outcomes.

[FromArseToElbow]

Building on Luke's analogy, there is the famous case of Len Shackleton's autobiography, wherein one chapter, entitled "The Average Director's Knowledge of Football", was a blank page.

The problem is not that we are all fools who ignore evidence and cleave to prejudices, but that many of the powerful are. The hold of EMH on politicians is akin to the theories of Charles Reep. They made perfect sense on paper, but were a terrible sight to behold on grass.

I think professional economists are beginning to overdo the self-pity, just as Aditya and others (e.g. the other Ferguson's 'Inside Job') are overdoing the "we name the guilty men" schtick. Economists are ideologically important, which is why they will always be corrupted, and why political economy went out of fashion.

[Deviation From the Mean]

Economists who claim that questions such as, "should I invest in this high-charging actively managed fund" are real economic questions need to go back to school I would say.

All economic questions are subject to countless variables and most economic questions are limited ones. I.e. they have assumptions. In the question above, you might want to ask how does investing in high-charging actively managed fund affect everyone else? You may then ask is an high-charging actively managed fund economically useful? And then you may want to ask, useful for who? And then you may want to ask......

The same issue can be applied to so called 'policy' decisions.

The economics profession is uncritical and only wants to celebrate the system. It is totally tainted.

[Magpie]

"I'd add that forecasting isn't part of proper economics at all, so a forecasting error tells us nothing about the merits or not of economics."

And I'd reply that you are wrong, very much so.

I would not be alone on that, and you don't need to take my word for it. At least 1976 Nobel Prize recipient Milton Friedman would agree with me:

"Viewed as a body of substantive hypotheses, theory is to be judged by its predictive power for the class of phenomena which it is intended to 'explain.' Only factual evidence can show whether it is 'right' or 'wrong' or, better, tentatively 'accepted' as valid or 'rejected.' As I shall argue at greater length below, the only relevant test of the validity of a hypothesis is comparison of its predictions with experience." (The Methodology of Positive Economics)

Friedman, it will be recalled, accepted that economics need not be descriptive: its predictive ability is what makes economics useful and its associated explanations plausible.

No prediction? No explanation! And no description...

So, what is economics good for, other than justify economists' salaries?

(Incidentally, is this assumed predictive ability that makes Ptolemy's geocentric model a good comparison with economics.)

[Pablo Mira]

"the fact that most economists failed to predict the crash is actually a vindication of mainstream economics, which says that such things should be unpredictable"
Actually, mainstream economists say those problems has been solved BECAUSE they are unpredictable. A nonsense, of course, but that is what you get when combining mainstream macrotheory and macroprescriptions: since those events are unpredictable, let's deregulate markets!

[Luis Enrique]

magpie

I think forecasting (trying to say what some number will be in some period) is a very different thing from coming up with models that make good predictions (on average, relative to an unobserved counterfactual, all else being equal and so forth).

[Magpie]

Luis Enrique,

I won't play these semantic games. Prediction is synonymous with forecasting. Ask Noah Smith:

What does it mean to have "predicted the crisis"?
http://noahpinionblog.blogspot.com.au/2013/05/what-does-it-mean-to-have-predicted.html

He damns Keen for lack of details.

Fair enough. But what's sauce for the goose, is sauce for the gander.

[Luke]

"Economists who claim that questions such as, "should I invest in this high-charging actively managed fund" are real economic questions need to go back to school I would say."

It is a real economic question to me, unless "will I make more/less money with more/less risk" is not an economic question for me. I'm not sure if "what if everyone else does?" is a real economic question or not. Not *for me*, I suspect.

[Luis Enrique]

magpie

it's not a game. I think Chris is using the word forecasting in a particular way and you have misunderstood him.

[Deviation From the Mean]

Luke, if we ignore for a moment that all economics revolves around you and consider economics as an objective discipline, aiming at some scientific kudos.

Then I would claim the question, "should I invest in this high-charging actively managed fund" is not really the question to gain such Kudos.

But I would advise blowing the lot on wine, women and song. And from that a whole lot of other economic considerations come into affect, though because we are now back with you at the centre of the universe, these questions are redundant.

[Philip Pilkington]

Weird post... Very internally inconsistent.

Chris Dillow writes: "The division that matters is not so much between heterodox and mainstream economics, but between good economics and bad. I'll give just two examples of what I mean.

First, good economics tests itself against the facts."

Okay great.

Mainstream economics didn't think there would be a crisis. There was a crisis. It failed the test. Therefore it is, by your own criteria, 'bad' economics.

That's the end of that debate. Unless you want to change the terms that you laid out. But that would be cheating, wouldn't it?

[Luis Enrique]

Philip

the bit of mainstream economics responsible for addressing the question "will there be a financial crisis in the Northern hemisphere" failed the test, big time. And everybody knows that. The profession - a few loonies notwithstanding - is not in denial about that.

To its discredit, mainstream macro was busy doing something else (assuming shocks arrive and looking at how to respond) and not trying to predict shocks. Those models *might* be useful for "how should we respond to shocks" questions ("good economics"), but they are not and never could be model to predict crises. Some mainstream macro economists are now working on endogenous crisis models, having realised they failed a test.

International macro has plenty of models that predict crises (sovereign debt etc.)

Financial economists had various models that showed banking crises and other financial market failures could occur, but didn't get those lessons listened to by those luminaries who thought there could not be a financial crisis. Perhaps they thought their models did not apply to sophisticated Western banks. Another fail, on that count.

Other bits of economics have models that could help us understand why the crisis occurred (principal agent models)

I think this means to label all of mainstream economics 'bad' is daft.

[Martin Connelly]

One of your less convincing posts Chris. Economists go around forecasting/predicting all the time. If they should not be doing so, then its up to other mainstream economists to correct those who do when they do so. And, if economists might no thave predicted the GFC, but others did.

[Philip Pilkington]

Luis,

Okay then. Let's be more specific: all mainstream economics that deals with macro shocks -- that is, ALL mainstream macroeconomics -- failed Chris' test. Therefore, ALL mainstream macro is 'bad economics' by his own criteria.

Again, he is free to change the criteria. But that will be seen as cheating.

[UnlearningEcon]

Chris I am surprised you endorsed the idea that since the EMH says things are unpredictable, it accounts for the 2008 crash. This just seems to me to be hand waving, and given the other problems with the EMH, I'm not sure why anyone would consider it an explanation at all.

Lilco also goes on to repeat some standard right-wing talking points...moral hazard means the gummut done it, regulation somehow caused the crisis, and it was an example of creative destruction. But none of these really make any sense, I really can't be bothered to repeat why.

In any case, the problem is that macroeconomic models simply didn't predict it. Regardless of whether economists have some key lessons correct, such as moral hazard, the discipline as a whole was blind to the systemic problems faced by financial institutions, and by the link this could have to the economy as a whole. You can claim the discipline has changed, but to claim it predicted and accounted for the crisis is beyond the pale.

[Charlieman]

@Luis Enrique: "one problem universities face is that if you offer a history of economics module, few student sign up for it."

Are you sure that it is not being taught? In engineering design, students are taught history indirectly; to get to where engineers are now, they tried ideas, which work sometimes but not always. Teaching should record failures.

Few engineers need to build a sprag clutch from scratch, but a half decent engineering education would alert those few that somebody else in the past did a good job.

The preselector gearbox was conceived in the 1930s, and functionally it worked well in British racing cars into the 1950s. Nobody developed preselector gearboxes for two human generations, but the concept is essential to flappy paddle gear change mechanisms in super cars.

One day, I will write up my critique on Henry Ford that 'History is bunk'.

[Left&rational]

Chris,

I can't see why you're conflicted. Mankiw is not representative of mainstream economics, at least not in his "defence of the 1%". In the same journal issue that published his paper you've got Piketty and Saez and co-authors - who have done more than any other academics to highlight the problem of growing inequality, and indeed do a much better job of explaining it than Mankiw does.

Quite why Mankiw gets the coverage he gets - for what amounts to a misinformed opinion piece, whilst Piketty and Saez - who are making a real mainstream research contribution, don't is an interesting question that you might like to consider in a future post.

I don't really understand your second point, but there's a lot we dn't understand about beahviour in financial markets. There's a lot of active research here as well.

On the history of economic thought, Luis is right, this drifted off the syllabus not through any neoliberal conspiracy, it was due to sheer lack of demand at the university I went to. I guess there is a case for making this mandatory, but this would hardly represent a revolution in the teaching of economics. Indeed the history of economic thought material that I was exposed to - taught by Mark Blaug at Exeter - seemed to underpin the mainstream apparatus rather than undermine it.

Chakrabortty's piece is misrepresentative of economics to the point of absurdity. I also think the precedent of students dictating the syllabus is wholly wrong, not to mention downright dangerous. Creationism? Astrology? Homeopathy in medical schools? Perhaps not the same thing, but where do you draw the line?

[Luis Enrique]

Why are the parts of mainstream macro that have nothing to do with predicted crises 'bad' because they failed to predict the financial crisis?

Saying that there was a major failure within mainstream macro - it ignored the big problem we now in hindsight know should,have been its top priority and had blithely assumed there was no danger from banks - is quite different from saying everything mainstream macro was doing is bad, for purposes other than predicting financial crisis.

[Philip Pilkington]

@Left&rational

I didn't see any demand by the students to "dictate the syllabus". They requested that there was more diversity in the department because there are other approaches that are recognised as valid by a significant and growing segment of the discipline.

To equate that in the first place with creationism is patronising and ridiculous. But regardless of that, no one is asking that students should choose the syllabus. Perhaps you were reading a different article to the rest of us?

[Philip Pilkington]

Luis,

What are you defining as macro? Is a specialist in trade a macroeconomist? I would say: no. He is a trade economist. What about in labour markets? Nope, he's a labour economist.

http://en.wikipedia.org/wiki/Macroeconomics

"Macroeconomics (from the Greek prefix makro- meaning "large" and economics) is a branch of economics dealing with the performance, structure, behavior, and decision-making of an economy as a whole, rather than individual markets."

That seems to me a perfectly good description. Macroeconomists deal with THE ECONOMY AS A WHOLE. The financial crisis had a major impact on the economy as a whole. That is a fact.

By Chris' criteria if a theory cannot deal with the facts it is 'bad' economics. Thus, if mainstream macro cannot deal with something that has an enormous impact on the macroeconomy then it is 'bad' economics.

That's really the end of the debate. Again, you can change the criteria. You can say something different to Chris. I'm basically just pointing out that his post is pretty incoherent. But then, we've been seeing a lot of incoherent and sloppy posts by mainstreamers on criticisms of their discipline in the past few months. I'm just trying to clean up the slop a little.

[AndrewLilico]

@UnlearningEcon | October 30, 2013 at 08:47 PM

My piece was disputing/debunking the claim that standard economic analysis offers no account of the 2008 crisis. You say the three main points I identified don't "really make sense" but that translates as "You don't agree". You don't dispute that these are elements of how standard economics accounts for the 2008 events - indeed, you describe them as "some standard right-wing talking points", reinforcing the point that those points are indeed part of what is normally offered in an account.

Say you don't agree, by all means, and tell us what you think instead - we'll all spend decades debating these events before we come to a consensus view and even then that might be wrong. But you (unlike my targets in my article) are not, I suspect, so foolish, naive and ignorant as to claim that standard economic analysis is incapable of offering any account of 2008.

[AndrewLilico]

PS, @UnlearningEcon | October 30, 2013 at 08:47 PM,

I suspect part of your own position is the complaint that standard macro models did not contain an explicit financial intermediation sector, and that that has been exposed as a key flaw. I might well have some sympathy for that, at least on a going-forwards basis, though I would say that any model has to abstract from many things, and standard macro models don't contain extractive industies, energy, or manufacturing sectors either. It's quite tough to know whether in the next crisis we face, having an explicit, and complicated to model, financial sector but not other sectors will add insight or obscure key issues.

That said, from the late 1990s I argued, in a series of papers on price-level targeting, that base slippage from inflation targets mattered partly because it failed to recognise volatility (indeed, under some conditions temporary collapse) in the money multiplier from base to broad money. I don't claim the standard macro models all performed perfectly (or were even that relevant). But the productive critiques and criticisms of standard macro models come from within orthodox economics - almost never from without.

[Magpie]

@Luis Enrique and Chris,

Sorry, fellows. I enjoy this blog, but I'm not convinced (RE: prediction).

[chris]

@Left&Rational - this raises the old qn of what mainstream economics is. There's a danger in committing the "no true Scotsman" fallacy here. I don't think it unreasonable to assume that a Harvard professor and author of standard textbooks writing in the JEP is "mainstream."
@ UnlearningEcon - I'm not saying the EMH explains the crisis - it doesn't. I'm merely saying it explains why mainstream economists failed to forecast it.
@ Philip & others - We must distinguish between forecasting, prediction and explanation. Yes, most economists didn't forecast the recession. But only some macro theories predict there can't be great depressions - these are the wrong ones. As for macro explanations, these are quite plentiful; one of my favourites is Ravi Jagannathan's:
http://apps.olin.wustl.edu/FIRS/PDF/2010/1692.pdf
I fear that those who are looking for a "science" that can forecast and prevent depressions are on a quest for the impossible.

[Left&rational]

Yes "mainstream economics" a more slippery concept than most recognise. But I think we can draw a separation between Mankiw's textbook and Harvard university (and indeed much of his earlier research on New Keynesian economics) - all in some sense mainstream, and his specific paper- "defending the 1%". You can call the latter "mainstream" by association if you insist, but like you I think that it is just plain bad research. But more to the point, as I tried to argue in my previous post, it is unrepresentative research.

Essentially what you are doing is saying that the Beatles are crap, because Paul McCartney's "frog song" is a crime against musicality.

[Luis Enrique]

Philip

here is the last macro paper I looked at.

http://elsa.berkeley.edu/~saez/michaillat-saezNBER13july.pdf

I'm pretty sure that counts as macro. It doesn't have anything to say about financial crises. One of the things it shows is the possibility of wage-led growth in presence of inequality. Is it "bad economics"?

hmm, hang on, some of the ideas in there are pretty left-field. Maybe it's a heterodox paper? It's good! Oh no, the authors are extremely high status mainstream economists. it's bad.

[Guano]

I am not an economist though I did study it at A-level more than 40 years ago. Back then it was a qualitative subject. In combination with history and geography you could learn a great deal about the world.

I was a bit surprised recently when my children said that university courses in economics required applicants to be good at mathematics. Economics has become a quantitative subject. It appears to be dominated by modellers looking for even greater precision. That is what its main customers (governments, financial institutions) are looking for: what is the optimal interest rate? But the risk of looking for precision is that you can be precisely wrong. You find a precise number for the output from your model that has a certain set of inputs, but it tells you nothing if you have omitted a key input.

Black swans. It's interesting to note how many economists (and politicians) use the term to mean things that could not be predicted. That is not the original meaning, which was things that should have been predicted but weren't because they were low frequency events. I get the impression that economists are still not thinking about these black swans.

Over-risky behaviour in the financial sector; climate change; antibiotic resistance - we know the risks are there though we cannot predict exactly what will happen and when. Can economists have a role in helping us to manage these risks? I think they can, in theory; they should be able to help us devise incentive structures to regulate the financial sector or reduce the harmful use of antibiotics. In practice, though, this appears to be difficult because it is so far removed from the assumptions of the models that economists are working on at present.

[Luis Enrique]

Guano,

you can obtain a "precise" answer (i.e. a number) whilst understanding there will be errors around it. In fact more advanced economic students would deal in probability distributions (around inputs and outputs).

Qualitative arguments also go astray if the omit a key input.

[Deviation From the Mean]

"you can obtain a "precise" answer (i.e. a number) whilst understanding there will be errors around it."

The problem with this is that the question being asked already contains bias and therefore the 'answer' is already corrupted. All economics is meta theory and all economics is normative in the final analysis.

This doesn't mean economics is a waste of time, that economic proposals cannot be analysed, evaluated. But any economists who claims that x will lead to y is firstly, ignoring a, b, c, d, e, f , g etc etc and secondly is failing to factor in variable 1, 2, 3 ,4, 5 ad infinitum.

This is why Marx's critique is more worthy than bourgeois economics because it doesn't try to ask questions such as "should I invest in this high-charging actively managed fund" but attempts to demonstrate that capitalism is a system of exploitation etc. or as Harvey would say, it is looking at the generality and not the particularity. Though, this can be problematical.

[zixiutangbeepollen.org Review]

Stumbling and Mumbling: Economics, good and bad

[UnlearningEcon]

@luis

Capitalism may 'deliver the goods', but the marxist argument is that they are unevenly distributed and often wasted, while production is often focused on superfluous things.

[UnlearningEcon]

OK, left the wrong comment on the wrong post. Let's try again:

@chris

OK, but there is circularity in the statement "economics predicts the crash couldn't be predicted, and economists didn't predict the crash, so this proves economics is right!" We first have to prove the first part eg defend the EMH. I think that this proposition as at least questionable.

@andrewlilco,

OK, OK, I hear you. I wrote that comment in a somewhat jaded state (does it show?) In any case, I'll leave a comment on your blog.

[Ramanan]

"Il Gattopardo (The Leopard) is a sweeping movie, based on the novel by Giuseppe Tomasi di Lampedusa, about social tumult and class conflict in Sicily in the 1860s. Directed by Luchino Visconti and starring Burt Lancaster, the film follows the Prince of Salina who is intent on preserving the existing aristocratic class order in the face of a rising bourgeoisie. As the crisis grows, Tancredi, the prince’s wily nephew, speculates that things must change if they are to remain the same. And they do. After the revolution, the old aristocracy remains in charge, allied via marriage with the new urban elite.

The concept of gattopardo is directly relevant for understanding the response of the economics profession since the financial crash of 2008. The response has been gattopardo economics, which is change that keeps things the same." - Tom Palley