In discussing macroeconomics' Faustian bargain, Simon asks:
By putting all our macroeconomic model building eggs in one microfounded basket, have we significantly slowed down the pace at which macroeconomists can say something helpful about the rapidly changing real world?
Let me deepen this question by pointing to five newish facts about the "real world" which any good, useful macro theory should be compatible with.
1. The unemployed are significantly less happy than those in work. This doesn't merely provide the justification for an interest in macroeconomics. It also casts grave doubt upon RBC-style theories which unemployment is voluntary. Sure, at the margin, some people might choose to be unemployed. But the fact that most unemployed are unhappy suggests that most unemployment is involuntary.
2. Price and wage stickiness is over-rated. One Bank of England study has found that:
Nearly half of firms changed their prices within three months of an increase in costs or a fall in demand
This suggests that price stickiness isn't universal. And whilst pre-recession evidence pointed to some wage stickiness, more recent evidence has cast doubt upon this. Models in which stickiness plays a central feature might not, therefore, be well-grounded empirically.
3. The failure of a handful of organizations can have massive macroeconomic consequences. The great recession originated in the collapse of a few banks. This tells us that we need models in which micro failures generate macro ones, of the sort Xavier Gabaix has proposed.
4. Supply shocks do happen. It's improbable that all productivity fluctuations are due merely to labour hoarding in the face of demand shocks. One way in which they might happen has been described by Gabaix - the failure of a few big firms shows up in macro data as a fall in productivity.
5. Interactions between agents can magnify fluctuations. We know there are expenditure cascades, which occur because consumers copy other consumers. And it's likely that animal spirits can spread in much the same way as diseases do (pdf) - because optimism and pessimism are infectious. We can talk ourselves poorer.
These facts are a challenge to both RBC and New Keynesian models. But they have something in common. They stress the heterogeneity of agents: some unemployed respond to benefit changes but most don't; some prices are sticky but others aren't; some firms are big enough to have macro effects, others aren't. This, I fear, means that the problem with conventional macro isn't so much its microfoundations per se as the assumption that these microfoundations must consist in representative agents.
Now, you'll object that any model that tries to embed these five facts will very quickly become very complex. But that's my point - the economy is a very complex system and any theory that doesn't see this is very dubious. It could be that the best way to understand it - if at all - lies in agent-based models rather than conventional DSGE ones.
However, such models are - for now - beyond the understanding of most economics students (and me!). Which poses the question: what should students be taught?
"what should students be taught?"
Well, at a minimum, all econ. students should be forced to read the canonical works of the discipline. Read Smith (both Wealth of Nations and Theory of Moral Sentiments), Marx, Ricardo, Malthus, Bastiat, Fisher, Keynes,so forth and so on. Being forced to engage intellectually with the various distinct insights these individuals had might give them a greater ability to come up with more realistic models for agent behavior.
Posted by: SIS | February 19, 2014 at 03:36 PM
You're preaching to the choir SIS
Posted by: chris | February 19, 2014 at 05:48 PM
"what should students be taught?"
Whatever they're teaching at UMKC, and the opposite of most of what they're teaching at Harvard :)
Posted by: Tyler | February 19, 2014 at 06:10 PM
Totally agree with this.
(Except for the minor point that unhappiness implies that unemployment is voluntary - it doesn't, it might just be the result of the least worst option being crap.)
The main thing I agree with is that economics as it stands not as far from a hopeless waste of time as most economists would like to admit.
Have you thought about climatology and meteorology? They deal with very complex systems. There are disciplines which don't treat chaos theory as a fringe heterodoxy.
And how about a wider acknowledgement of the importance of credit?
A culture of intellectual humility, empiricism and scepticism would also be welcome.
Posted by: Andrew | February 19, 2014 at 09:30 PM
To clarify, I am talking about macroeconomics as per the post, not the range of good micro activity in the field.
Posted by: Andrew | February 19, 2014 at 09:34 PM
Chris - Daron Acemoglu is doing a lot of great work here, both in finance (http://18.7.29.232/bitstream/handle/1721.1/77606/Acemoglu13-03.pdf?sequence=1) and in the real economy (http://economics.stanford.edu/files/Acemoglu10_10.pdf). Thought you'd like this on your radar if it isn't already.
I think you move from your observations 1 through 5 to bashing DSGE too quickly, though. Roger Farmer uses the barest modifications to simple RBC-type representative agent modelling and comes up with models with at the very least satisfy 1, 2, 4 and 5. His is a far more successful research programme than agent based modelling, as far as I can see.
Similarly the Acemoglu work isn't really agent based modelling in the sense you describe it (being about the analytical properties of networks as much as simply computation) and models 3 very successfully without describing the economy as a complex system.
Posted by: JADHazell | February 19, 2014 at 10:36 PM
Yes, and also modern macro is the cause of global warming and steals chocolates from babies.
1. Already incorporated. What do you think 'diminishing marginal utility' (plus Inada conditions) implies for people with no income? Once you add in some sort of capital market imperfection (credit constraints, risk aversion, asymmetric information) then these standard neoclassical assumptions start to really matter. DSGE models have been analysed at length incorporating all of these issues. Also, you will be aware that DSGE as a general class of models do not rule out involuntary unemployment. What you refer to is a very special textbook case. Really at best this is a straw-man argument.
2. Do you mean the same price and wage rigidities that are ignored in the the textbook RBC models referred to in your point 1?
3. Agreed but only up to a point. You could perhaps instead refer to 'the financial sector' rather than 'a few organisations'. But this is exactly what mainstream macro has looked at more than anything else for the past 7 years - within the DSGE framework. No 'Faustian bargain' here. Furthermore the paper you cite was published in Econometrica - perhaps the flagship journal of the discipline. What does that tell you about mainstream economics' reaction to 2007/8?
4. Who, within modern macroeconomics, has ever denied the importance of supply shocks? Certainly not RBC models, and neither NK models for that matter.
5. Yes these phenomena are important, but also are known and can be (and have been) incorporated within DSGE. Can you explain how these phenomena and DSGE are incompatible? Please don't cite the textbook caricature again. Imagine that the utility function, agents' information sets, institutional rules and so on, may change within the context of DSGE.
What should we teach the students? Well, a mix of models which can do a reasonably good job of explaining reality - e.g. the IS-LM model, and models which are internally consistent and provide a good platform for extending upon in research projects, - e.g. New Keynesian DSGE models.
Oh wait, that's already what happens.
Sorry I don't mean this post to come across unnecessarily abrasive. Just peeved with the endless caricaturing.
Posted by: left&rational | February 19, 2014 at 11:20 PM
Some assorted points...
Very few people have ever taken the voluntary unemployment things literally, it's always been a place holder for something else.
Price stickiness in modern macro is there to get real effects of monetary policy, so what matters is that firms do not change prices in such a way as to neutralise MP. They can change prices every other minute for other reasons, that's beside the point. I think best reason why firms do not adjust prices to neutralise MP is Sims rational inattention. It's far to bothersome for far too little gain. They may change prices after they experience a change in nominal demand, but that's after MP has worked.
I guess some agent based models might have millions of agents, but otherwise you are going to use representative agents on some level. What's needed is sufficient heterogeneity. I agree with others that better treatment of debt would be desirable, although some mainstream work is already heading in that direction.
There are some pretty gentle agent based model text books around, you'd have no problem with them. Although of course they can be criticised along similar line so the mainstream, they are highly unrealistic along most dimensions
Posted by: Luis Enrique | February 20, 2014 at 08:56 AM
@ L&R - thanks for those comments. I appreciate your discontent with caricatures of economics; I've expressed it myself sometimes. I wasn't trying to write one of those "economics is wrong" pieces, but rather merely trying to emphasise some points that simpler models (of the sort students start with and often finish with) under-emphasize.
I'm not sure point 1 is wholly a straw man: I was thinking of Casey Mulligan when I wrote that (not to mention a big chunk of the general public!).
I used "some organizations" rather than "the financial sector" because it's possible for non-financial firms' failures to have macro effects - eg when the Fukushima nuclear plant closed: Acemoglu's done some work here:
http://www.iadb.org/intal/intalcdi/PE/2013/12532.pdf
As for your idea that ISLM should be taught, take it up with Simon Wren-Lewis:
http://mainlymacro.blogspot.co.uk/2012/02/why-introductory-macroeconomics-should.html
Posted by: chris | February 20, 2014 at 09:34 AM
No economist will ever answer my question when I ask about how or whether any economists address deterministic chaos and emergent properties. Am I to assume that they assumed away?
Posted by: AlanDownunder | February 20, 2014 at 09:49 AM
@ Alandownunder. Some have. See for example some of Alan Kirman's work
http://ideas.repec.org/e/pki52.html
His book Complex Economics is good.
Or try Brian Arthur
http://tuvalu.santafe.edu/~wbarthur/complexityeconomics.htm
Part II of Eric Beinhocker's The Origin of Wealth is also good.
Posted by: chris | February 20, 2014 at 12:33 PM
My received potted history is that economists got excited about chaos etc about the same time as everybody else (1970s?) but interest waned after little of interest emerged. People like Herb Gintis are keeping the flame alive, see recent Economic Journal on his website, but the general mainstream impression seems to be that whilst there's some interest in these models and much promise, they have thus far not proved to be much use answering the sorts of questions economists seek to answer ( monetary policy, tax policy, industrial organisation etc). Some of that is no doubt inertia and resistance from old fogeys invested in old ways, but my impression is that once you take the sort of critical attitude to agents based models as people are wont to do to the mainstream! there are plenty of cracks. Personally I think they are fascinating and probably the future of economics, but wiser heads than me are more skeptical
Posted by: Luis Enrique | February 20, 2014 at 12:48 PM
Thank you,
It's wonderful to see the starting point a question rather than an assertion about something that is obviously not true to anyone who is not an economist.
Please add Politics to your list. Politics matter. I think flawed, failed or incomplete the majority view of what economists do know did not get implemented.
It's pretty well understood that there needs to be counter-cyclical spending. Agents that reflect consensual co-operative action including the government need to be modeled.
Posted by: Dan | February 20, 2014 at 01:47 PM
Point (2) is quite misleading indeed. The report states that the main cause of changes in prices are raw material costs and wage costs -- which is consistent with fixed-price theory.
The point here is: although half of firms may change their prices in the face of changes in demand it is by no means clear that they change the prices to clear markets -- this is the key to vindicating flexible price arguments.
Also, so far as I can see (I didn't read the whole report), the report does not take into account the SIZE of the firms that respond to changes in demand quickly. If 50% of firms do respond quickly but these 50% are small firms -- which logic tells us they would be -- then they may make up maybe 10% of GDP.
Again, I think what you said in point (2) is quite misleading.
Posted by: Philip Pilkington | February 20, 2014 at 02:04 PM
On point (2), you cite Greenslade and Parker (2012), but that study says that when asked how prices for their main products were determined, the second most important explanation given by firms was mark-up pricing -- a form where prices are general;y not responsive to demand -- with variable mark-ups (58%) and constant mark-ups (44%) both being important (Greenslade and Parker 2012: F10).
And you only need to read S. Hall, M. Walsh, and A. Yates. 2000. “Are UK Companies’ Prices Sticky?,” Oxford Economic Papers 52.3: 425–446 as a further refutation of the idea that "Price and wage stickiness is over-rated."
E.g., Hall et al. asked firms: what happens when there is strong demand and this cannot be met from inventories or stocks?
The result was as follows:
Increase overtime | 62%
Hire more workers | 12%
Increase price | 12%
More capacity | 8% (Hall, Walsh, and Yates 2000: 442).
______
Only 12% would increase price: that's a pretty damning finding contradicting your point (2).
A full summary of the evidence here:
http://socialdemocracy21stcentury.blogspot.com/2014/02/mark-up-pricing-in-uk.html
Posted by: Lord Keynes | February 20, 2014 at 02:29 PM
@ Lord Keynes. I fear the HWT paper is a little dated. Greenslade and Parker find that 39% of firms chnage prices more often now than they did 10 years ago.
@ Philip - G&P find that big firms are more likely to change prices often, so the flex-price sector might represent a decent share of GDP. I agree that it's quite possible that they don't cut prices enough (and I can imagine cognitive biases that would cause this, such as Bayesian conservatism). But I'd welcome micro evidence on this.
I should emphasize that I'm NOT saying that prices are perfectly flexible - just that they're not fully sticky. The qn is: how much price stickiness do Keynesian models need? How would things change if we have a sticky price sector and a flex price sector?
Posted by: chris | February 20, 2014 at 05:11 PM
@chris
It's not just something like Bayesian conservativism. It's far more fundamental than that.
How are prices set? In a General Equilibrium model -- i.e. pure flex-price, market-clearing model -- they are set by the Walrasian Auctioneer. That is, prices are centrally planned.
Famously, flex-price theorists have never truly explained how prices ACTUALLY get there. Walras himself hinted at some sort of "groping". Okay, so let's take that at face value.
In order for firms to "grope" until they found market-clearing prices they would have to change prices, oh I don't know, maybe 5-10 times before they got the Goldilocks "just right" price. Not only does this not match up with experience but it seems highly unlikely.
To me, the idea of fix-price markets is a sort of pure theory encompassing all the many frictions of the real world. I'm not 100% convinced that it does need to be proved empirically.
That said, I'm a Post Keynesian and we have a long tradition of mixed economies with two sectors. Kalecki, for example, in his Theory of Economic Dynamics discusses fixed and flexi price markets in the first chapter (and, interestingly, focuses on the latter as primarily a cause of inflation**). Kaldor took a similar line. And JK Galbraith famously argued the same case in is New Industrial State.
But I'm still more agnostic still -- perhaps even atheistic. I just don't see how the actual logic of market clearing prices can ever take place in the world we live in. It seems like a silly thought experiment that got taken seriously before anyone (barring maybe Joan Robinson) ever considered how it could truly ever be applied.
___________
** I wrote up a piece on this here if you're interested:
http://www.nakedcapitalism.com/2012/10/philip-pilkington-why-free-markets-accommodate-speculation-and-lead-to-disequilibrium.html
Posted by: Philip Pilkington | February 20, 2014 at 05:45 PM
Mr Dillow adds another thoughtful chapter to the current conversation on the state of (macro)economics.
He comes close -- but does not quite arrive at -- a conclusion that should not be that hard to draw: economics is simply too complex to be "modeled" with a spreadsheet.
Most sciences long ago abandoned any pretense of "universal understanding". As knowledge of physical phenomena becomes ever greater, scientists have accepted ever-narrower windows onto the universe. Specialties are more and more detailed.
Why should economics escape this general rule?
Why should we be surprised that we need specific models for specific phenomena? The requirement for any "partial" model being, as in any science, that it fit the past and predict the future, for the particular phenomenon it addresses.
Posted by: Robert Nelson | February 20, 2014 at 07:02 PM