The macro foundations of microeconomics

by John Quiggin on October 25, 2013

Twitter alerted me to an amusing exchange between Chris Auld, posting a list of “18 signs you’re reading bad criticism of economics and Unlearning Economics, responding with 18 Signs Economists Haven’t the Foggiest. UL suggests that Stephen Williamson manages an impressive 9 out of 18 in his review of Zombie Economics (my response here with more from Noah Smith.

Scoring myself against Chris Auld’s list, I’d say I’m in the clear. But quite a few commenters on Zombie Economics have made complaints along the lines of his point 1, that I focus too much on macroeconomics (and finance). The implication is that, even if macro is totally wrong, only a minority of economists do it, and microeconomists are in the clear.

This defense doesn’t work, at least not in general.

The basic problem is that standard neoclassical microeconomics is itself a macroeconomic theory in the sense that it’s derived from a general equilibrium model as a whole. The standard GE model takes full employment (in an appropriate technical sense) as given, and derives a whole series of fundamental results from this. Conversely, if the economy can exhibit sustained high unemployment, there must be something badly wrong with standard neoclassical microeconomics.

Most notably, in a competitive GE with full information, no externalities and so on, prices of goods reflect the social opportunity cost of producing them. This means, that, other than redistributing the initial endowments of property rights, governments can’t do anything to improve on the competitive market allocation of resources.

Once you have involuntary unemployment, all of this fails. Keynes’ famous thought experiment of burying pound notes in coal mines made the point that an intervention that would be totally absurd in terms of standard microeconomic reasoning might nonetheless help to alleviate a recession and therefore make society better off.

The point can be made in more detail with respect to labour economics, finance theory, public economics and industrial organization. None of the standard conclusions of these fields of microeconomics can be assumed to be valid under conditions of sustained high unemployment.

Keynes specifically presented his macroeconomic ideas as making the world safe for neoclassical micro. If governments could stabilise the aggregate economy with fiscal policy, there was no need for comprehensive economic planning of the kind being practised, with apparent success, in the Soviet Union, or for ad hoc interventions like the price-fixing elements of the New Deal.

In summary, the failure of macroeconomists and finance economists to provide either a warning of the Global Financial Crisis or any consistent advice on how to deal with the ensuing Great Recession it isn’t just a problem for them. It undermines the whole economics profession. The sooner we realise that the entire discipline is in a state of scientific crisis the sooner we might start to do something about it.

{ 218 comments }

1

Kevin Donoghue 10.25.13 at 11:37 am

If economics was mine to dispose of I think I’d carve it up into pieces, entrusting some bits to mathematicians, some to management science types etc. I’m not sure this would solve anything much, but it could put an end to some pointless arguments.

2

Robert 10.25.13 at 11:37 am

After a group meeting yesterday, I was talking to my supervisor in front of the receptionist.

Me: I only asked that question, because I knew nobody wanted to go first. I thought I would act like a fool and encourage discussion.
Receptionist: And you do it so well.

3

Bruce Wilder 10.25.13 at 11:37 am

Lots of markets do not clear. One would think that someone would draw an implication from that about aggregate outcome. But, alas . . .

4

Robert 10.25.13 at 11:37 am

And my timing seems to be good too.

5

ed 10.25.13 at 11:48 am

John, given the criticism you and others make to mainstream economics (macro and micro), what should someone who wants to learn a more correct version of economics start with? Books that explain what is wrong with currently dominant economics is not a pedagogical starting point to recommend to someone wanting to get into the subject. So what currently available intro to micro or intro to macro textbook would you recommend as both pedagogical enough and correct enough to be a good starting point to read on a freshman level?

6

reason 10.25.13 at 1:57 pm

I actually love this sentence:
“Most notably, in a competitive GE with full information, no externalities and so on, prices of goods reflect the social opportunity cost of producing them”

Lets parse it out – in a competitive (meaning perfect competition which hardly ever exists) GE (a fleetingly unlikely event) with full information (given the limitations of the human brain not even theoretically possible), no externalities (we all live on separate planets) and so on …..

i.e. before you even come to the bit you want to reject, you have made four absurd assumptions.

7

Chatham 10.25.13 at 2:08 pm

That comment section is a pretty good example of some of the oddness coming from those critiquing mainstream econ:

Krugman had an awful theory of the euro. Back in the 90s he was pushing the Optimal Currency Area argument which utilises Robert Mundell’s work (Mundell used the argument to SUPPORT the single currency! flexible framework right!). Then when the crisis broke Krugman dropped that and basically robbed the MMT argument with accrediting it.

This from the unlearningecon guy:

It does make sense in that there are obvious things that could falsify Darwinism: bunny rabbits in the Cambrian, an organ of irreducible complexity. These things have never been observed, but as Darwin himself noted, even one of the latter would tear his theory to shreds.

Of course, this is the internet, so most criticism is going to be bizarre/ill-informed. I’m not sure it’s much more relevant than people getting death threats online.

8

In the sky 10.25.13 at 2:15 pm

The standard GE model takes full employment (in an appropriate technical sense) as given, and derives a whole series of fundamental results from this. Conversely, if the economy can exhibit sustained high unemployment, there must be something badly wrong with standard neoclassical microeconomics.

Lest readers mistakenly forms the impression from this post that economists have completely overlooked unemployment in GE models. They have not. There are plenty of GE models (and, heavens!, dreaded dynamic and stochastic GE models) where unemployment is a major component.

The standard GE model doesn’t have the sort of imperfect decision-maker where “nudges” can substantially affect the savings rate, either…

9

Straightwood 10.25.13 at 3:09 pm

Let’s not devote all of our scorn to the Internet rabble. Please note that Robert Rubin remains one of the ten members of the Harvard Corporation. Mr. Rubin and his proteges, including Lawrence Summers, made many of the disastrous decisions leading to the financial crisis, yet only a faint odor of opprobrium clings to them. Their lofty postions are unassailable because of the astonishing durability of reputational franchises in modern society. Once you are a “made man” in institutional circles, it seems that only a felony conviction can impair your standing.

10

TheF79 10.25.13 at 3:20 pm

Isn’t this just the Theory of the Second-Best? i.e. if we place a Pigouvian tax on an externality but firms have market power, we may correct the externality but by reducing output we exacerbate market power in a way that it is actually welfare decreasing? But in this case, we replace “firms have market power” with “agents are involuntarily employed,” and now the Pigouvian tax potentially changes the amount of involuntarily employment? At least that seems like the corollary to the burying notes in coal mines example. Microeconomists have thought about second-best issues for like half a century, so maybe there’s a more fundamental point here that I’m missing.

11

Jerry Vinokurov 10.25.13 at 3:22 pm

Hey, I have a list too. It’s called “One sign you’re dealing with someone uninterested in any critique of their discipline.” It goes like this:

1) Behaves like a smug asshole on the internet without offering any conclusive support for any assertions.

Chris Auld seems to fit the bill, by my lights.

12

nnyhav 10.25.13 at 3:49 pm

13

William Timberman 10.25.13 at 4:10 pm

To a lay observer, wherefore by their fruits ye shall know them seems to apply here, and to apply not only to the work of particular economists, but to the discipline of economics as a whole. Stephen Williamson may be wrong, and wrong in a particularly annoying way into the bargain, but he isn’t the only one who hasn’t any genuinely useful tips to offer to Greek pensioners, Chinese wage slaves, or American Walmart greeters, who, after all, are the ones forced to cope in a very personal way with the failure of his and every other economist’s recent expectations. Professor Williamson can’t be bothered, true, but even those who can be bothered don’t seem to have much left on their shelves for those in actual need. Or so it seems to me, peering into their shop from the outside. It’s not their fault, or at least not exclusively their fault, so I get no pleasure from seeing them humbled. I do wonder, though, where else we’re all going to go for that much-touted change we can believe in.

14

Donald Johnson 10.25.13 at 4:20 pm

Chatham–I know nothing about MMT beyond the fact that it is a heterodox theory in economics, so maybe your first cited comment was bizarre, but what was wrong with the following__

“It does make sense in that there are obvious things that could falsify Darwinism: bunny rabbits in the Cambrian, an organ of irreducible complexity. These things have never been observed, but as Darwin himself noted, even one of the latter would tear his theory to shreds. “

That’s a pretty standard (and correct) response made by defenders of evolution against creationists who take the line that Darwinism is unfalsifiable. His point, made in the next sentence which you omitted, is that neoclassical economists don’t point to similar observations that would falsify their theories. Is that the part that is silly? Because you forgot to include it.

15

Jerry Vinokurov 10.25.13 at 4:36 pm

Economics is still suffering philosophically from this particular piece of bad reasoning promulgated in Friedman’s Essays in Positive Economics:

The difficulty in the social sciences of getting new evidence for this class of phenomena and of judging its conformity with the implications of the hypothesis makes it tempting to suppose that other, more readily available, evidence is equally relevant to the validity of the hypothesis – to suppose that hypotheses have not only “implications” but also “assumptions” and that the conformity of these “assumptions” to “reality” is a test of the validity of the hypothesis different from or additional to the test by implications.

This kind of approach fundamentally confuses the issue and turns the entire discipline into a parameter-tuning exercise. If you’re not required to demonstrate that the axioms of your theory bear any relationship to reality, then you can conveniently decouple your model from any real-world phenomena and claim that it’s good for something as long as it validates on some out-of-sample data. Friedman then goes on mangle physics, with his entirely spurious comparison to expert pool players. There’s a lot of “seems” and “plausible” sprinkled along the way, with very little evidence adduced to demonstrate that any of this is coupled to any actual behavior on the part of the agents (firms, individuals, governments) under consideration. The whole black-box approach to human behavior (“humans might not literally be solving optimization problems in their heads but we can pretend they do because… reasons!”) is thoroughly misguided; it cannot possibly serve as a substitute for actually understanding what people actually do, what real cognitive processes drive decision-making under real-life constraints.

16

Daniel Velez 10.25.13 at 4:43 pm

I would question the statement that most micro economists believe that the world is anything like the model you just described. If that was the case no one would study how to use policies to correct externalities (environmental economics), asymmetric information (health/labor economics), irrational actors (behavioral economics). This “scientific crisis” you describe took place a long time ago and economists have been trying to understand these types of issues for decades. Simplifying microeconomists as people who take the Arrow-Debreu model as fact is intellectually dishonest.

17

ezra abrams 10.25.13 at 4:45 pm

one thing I know for sure: Both G Mankiw and P Krugman are very wealthy individuals
Both are authors of college textbooks, which sell for an exorbitant price.

Given these two facts, one might ask why they (in essence) think it is ok for a dictator (college prof) to charge ludicrous prices for textbooks that the dictator makes mandatory.

18

Chatham 10.25.13 at 5:40 pm

Chatham–I know nothing about MMT beyond the fact that it is a heterodox theory in economics, so maybe your first cited comment was bizarre, but what was wrong with the following__

The first comment is bizarre because almost every time I’ve seen Krugman talk about the Euro lately he’s mentioned the idea of an optimal currency area. To say that it’s something he talked about in the 90’s before he was left disarmed by the fact and forced to plagiarize from the MMT (“we might seem to be marginalized, but the most famous economists steal from us because they secretly know we’re right!”) is rather strange.

That’s a pretty standard (and correct) response made by defenders of evolution against creationists who take the line that Darwinism is unfalsifiable.

Eh? Irreducible complexity is an ill-defined creationist idea that assumes a drop of ignorance outweighs an ocean of knowledge. If scientists found a complex organ they couldn’t immediately explain the evolutionary origin of, they’d spend time trying to understand it rather than saying, “that’s it! Evolution doesn’t exist!” Likewise prehistoric bunnies.

I also don’t see the use of holding Darwin up as an expert, but his “18 problems with economics” post suggests that he thinks individuals who are viewed as the originators of an idea have some sort of magical connection to the idea.

His point, made in the next sentence which you omitted… you forgot to include it.

I just didn’t want people to know The Truth.

…is that neoclassical economists don’t point to similar observations that would falsify their theories. Is that the part that is silly?

It’s silly when he’s responding to a comment that compared economics to biology, implying that unlike theories in biology, economic theories are non-falsifiable because neoclassical economists provide theories that are non-falsifiable. Neoclassical economics is not the totality of economics; none of the economists I read call themselves neoclassical (and they do provide examples of how their theories could be proven false).

19

UnlearningEcon 10.25.13 at 5:52 pm

@Chatham

You are over-interpreting my statement about Darwin. Obviously, as Kuhn taught us, theories cannot be abandoned as soon as an observation superficially seems at odds with them. However, my point was that there are logical possibilities which would invalidate Darwinism.

“I also don’t see the use of holding Darwin up as an expert, but his “18 problems with economics” post suggests that he thinks individuals who are viewed as the originators of an idea have some sort of magical connection to the idea.”

Well, the originators of an idea are surely well equipped to understand it and IMO should be listened to. But I link to far more people than those who originated a particular idea, and also do a survey at the end…

“Neoclassical economics is not the totality of economics; none of the economists I read call themselves neoclassical”

Most economists don’t use the term “neoclassical” to refer to themselves, no, but that doesn’t mean the school doesn’t exist. Loosely speaking, it’s the ‘optimisation and equilibrium’ approach, which characterises things like consumer/producer theory DSGE, Walras, etc.

“(and they do provide examples of how their theories could be proven false).”

As I said to that commenter, I’d like to see one example of a neoclassical theory which has been truly “falsified” and never used again. Every time I ask this there is a deafening silence.

20

Robert 10.25.13 at 6:00 pm

“Washington has spent the past three-plus years in terror of a debt crisis that keeps not happening, and, in fact, can’t happen to a country like the United States, which has its own currency and borrows in that currency.” — Paul Krugman, in his column today in the NY Times.

Sounds like MMT theory to me. I am willing to accept that both some mainstream and some non-mainstream economists did not expect the Euro to survive without severe problems.

The idea that the use of the terms “neoliberalism”, “neoclassical economics”, or “neo-classical economics” is suspect is just stupid. Furthermore, Auld, in the comments section, falsely accuses another commentator of calling him a Nazi. Nobody should take this intellectually impoverished clown seriously or care if numbered items in his list apply to you. (I admit to a bias here.) Those who would like to see (too broad a group of) academic economists disrespected by others would want Auld to keep up his decades-long behavior.

21

Chatham 10.25.13 at 6:25 pm

However, my point was that there are logical possibilities which would invalidate Darwinism.

As well as economic theories (though perhaps not whatever theories you would define as neoclassical).

Well, the originators of an idea are surely well equipped to understand it…

Not if the idea is reliant on empirical research.

As I said to that commenter, I’d like to see one example of a neoclassical theory which has been truly “falsified” and never used again. Every time I ask this there is a deafening silence.

Are any of the commentators defending neoclassical theory? That could be your problem. You quoted the end of my sentence, but seemed to miss the beginning:

Neoclassical economics is not the totality of economics; none of the economists I read call themselves neoclassical (and they do provide examples of how their theories could be proven false).

And respond to that by asking for neoclassical economic examples. Eh?

If you want examples from economists regarding how their theories would be disproven, I’d be happy to provide them – though I should point out they’re not hard to find. As I said though, none of them consider themselves neoclassical, and since you admit that few economists actually do consider themselves that, I don’t see why we should pretend that they’re important when talking about economics as a whole.

22

Walt 10.25.13 at 6:49 pm

UnlearningEcon: If you asked Chicago-school macroeconomists that question, 100% would give (old) Keynesian economics as an example of a falsified theory. It’s the central myth in their origin story.

23

UnlearningEcon 10.25.13 at 7:09 pm

@Walt,

Probably, but even if we accept their argument, the Phillips Curve and many ‘Old Keynesian’ concepts/ideas are still taught and sometimes endorsed by economists.

@Chatham

You’re missing the point. I’m saying that neoclassical economics exists and has a definition, whether economists want to call themselves neoclassical, marxists or the rainbow brigade. I can provide this definition (loosely: “optimisation and equilibrium” and have yet to see a theory that fits it entirely abandoned.

24

adam.smith 10.25.13 at 7:23 pm

@23 UE – I’m sympathetic to a lot you’re saying, but stuff getting taught isn’t an argument. Newtonian physics and even the Bohr model of the atom are still taught, and that’s in disciplines with much clearer delineations of established knowledge and falsification.

25

Chatham 10.25.13 at 7:36 pm

@UnlearningEcon

I must be missing the point, because frankly I don’t see any point. If we’re interested in economics compared to biology, and interested in whether or not economists think their theories are falsifiable, I don’t see the relevance of trying to ferret out who we should be calling neoclassical.

There are plenty of economists out there who provide ways to falsify their theories. Krugman has said numerous times that soaring interests rates would show his model to be false. I don’t know if economists who you would consider neoclassical would say something similar, but I’m also not terribly interested in who, despite their protestations, we should be calling a neoclassicist. Nomenclature is just not the most important thing in my mind when I’m trying to evaluate an economist or economics as a whole.

26

Jerry Vinokurov 10.25.13 at 7:50 pm

Newtonian physics and the Bohr atom are taught as useful approximations to much more accurate theories. Furthermore, the domain of applicability for Newtonian physics and Bohr atoms is very well-defined, as compared to the domain of applicability for economic models.

27

Bruce Wilder 10.25.13 at 8:21 pm

The basic problem is that standard neoclassical microeconomics is itself a macroeconomic theory in the sense that it’s derived from a general equilibrium model as a whole.

While appreciating the point you are making, I’m not sure that’s the “basic problem”. It’s not about false axioms, per se; it is about the failure to embrace critical method, in forming an overall view and informed faculty of judgment. Macroeconomics and finance only seem to have greater weight, because it is in attempting to answer political questions — policy economics questions — that the failure of judgment comes to the fore.

Studying economics, as a college student or a college professor, seems to fry the brains of people. Even great scholars, who are capable of sustained and difficult feats of careful reasoning, nevertheless, embrace ideologies of maddening stupidity and simplification. And, as the several recent posts on the absurdities of the economic Nobels illustrate, this is a pattern, which applies at the highest levels of the profession.

If economics, as a scholarly discipline, were doing its job, its most reputable practitioners would be wise teachers, of nuanced judgement, able to raise the level of political debate and dispute, with useful facts and insights.

I have some acquaintance with economics, and antiquated experience practicing economics as a profession, and can respect the achievements of, say, Eugene Fama or Ronald Coase (just to choose people, who have been discussed recently on CT) or even, the arch-devil, Milton Friedman, but I cannot fathom their embrace of an overarching view of how-the-economy-works, which is so stupid, and so at odds, frankly, with what I would think a person trained and informed by economics, would arrive at, regardless of their personal temperament or idiosyncratic preferences. How Fama gets from a remarkable essay about how to formulate a null hypothesis of informational efficiency for research on prices in financial markets to a zombie idea that markets simply are efficient, and no one should suggest otherwise, I have no idea; how an expert on financial markets could ask, in relation to Obama’s 2009 stimulus proposals, “where’s the money going to come from?” is beyond my comprehension. It’s not really about axioms, is it?

Greg Mankiw’s textbook has its famous “10 lessons” on the flyleaf, and every one is ridiculously and, one would think, embarrassingly oversimplified. Mankiw is a genuinely smart guy, but he seems to prefer to be a smart-aleck 14 year old, who just finished reading Atlas Shrugged last Saturday. Someone, who learns economics from that textbook ends knowing less about economics than she would learn just walking around a supermarket with her eyes open.

It is not like there are many economist heroes out there. Brad DeLong’s series of self-smackdowns is charming on one level, but shouldn’t he have better judgement at his age, a deeper understanding? He seems to think he’s a terrible economist, and, as far as I can tell (and I can tell), he’s basically just a much nicer person than his conservative interlocuters, not a better (or even minimally competent) economist. I could be accused of worshipping Paul Krugman’s footsteps I am so enthusiastic about his blogposts sometimes, but if I really think about it, I realize that his view of the economy as a whole is remarkably stunted on a fundamental level, and he really doesn’t like to engage critically with those shortcomings. Josh Mason was making a cottage industry for a while of writing posts that took the form of “I agree with Paul Krugman, but . . . “, where he would take up one or another of the howlers, that seemingly make up the deep foundation of Krugman’s economics — the long-run neutrality of money, being one that comes to my mind, or the irrelevance of income distribution to the possibility of full-employment another, or the dictum that currency devaluation rebalances trade. These kinds of ideas — taken from analytic “proofs” of “theorems” — are accepted almost as articles of faith by many economists, Krugman included, though they cannot withstand an hour’s serious thought about the operation of the actual economy. And though, Krugman seems to relish excoriating his “bone-headed” (applied to Barro) colleagues or mourning a “Dark Age” at Freshwater schools, he seems to have been unaware, say, of Ben Bernanke’s views, though Bernanke hired him at Princeton and was his colleague for many years.

Maybe, economics feels too much like a big investment in Socratic-admitting-you-don’t-know-anything, that at the end, honest analysis is paralyzed analysis, of the “on the one hand, . . . but on the other hand . . . ” kind, and retreating to the verities of “Free to Choose” ideology is experienced as relief.

Maybe, the whole profession suffers from Dunning-Krueger syndrome; as a group, despite erudition or technical skills, most economists know so little about the economy outside of an area of narrow specialization, that they are incapable of discerning judgment (“good taste”) about broad and sweeping questions. That tends to be my experience of individual, unfamous economists — they know some little corner of the economy, where they have mastered subtle distinctions and have, sometimes, deep factual knowledge, but on big questions of policy and/or macro economics, where they know little, they fall back on some half-assed applications of the market failures toolkit and Friedmanite rhetoric about “markets” and “free trade”. They just “know” some things — rent control is bad, or the minimum wage reduces employment or the New Deal was a failure, which prolonged the Great Depression by raising prices, Keynes was discredited twenty years ago and we all know better now, . . . and on and on like that, each opinion no better supported than the political ideas of my barber.

28

Bruce Wilder 10.25.13 at 8:30 pm

You fellows want falsified?

“Market economy” is false.

Thank you for playing.

29

Trader Joe 10.25.13 at 8:50 pm

In most sciences, the current scientist stands on the shoulders of the great scientists who came before them so that they can see farther.

In economics, the current economist stands on the shoulders of the economists who came before them so he can more firmly push them into the quicksand and to provide at least some temporary support for their own theories before they too know the same fate.

30

Donald Johnson 10.25.13 at 9:49 pm

“Irreducible complexity is an ill-defined creationist idea that assumes a drop of ignorance outweighs an ocean of knowledge. If scientists found a complex organ they couldn’t immediately explain the evolutionary origin of, they’d spend time trying to understand it rather than saying, “that’s it! Evolution doesn’t exist!” Likewise prehistoric bunnies.”

You should blame Darwin for that–he’s the one who said that his theory could be disproven by the existence of an organ that couldn’t have evolved in a gradual step-by-step process. Perhaps he was being dishonest–it’s very hard to show an organ couldn’t have evolved. Behe just coined the phrase IC and provided some bogus examples. As for Cambrian bunnies, of course that would falsify evolution, or else reveal the existence of time machines or godlike beings with a sense of humor. That’s why creationists are so eager to find human footprints in Jurassic rocks.

From reading Krugman, rightwing economists don’t seem to acknowledge any test that would falsify their theories. Shown fossils of Cambrian bunnies, they would pretend they were trilobites.

31

Donald Johnson 10.25.13 at 9:51 pm

“From reading Krugman, rightwing economists don’t seem to acknowledge any test that would falsify their theories. “

I mean, of course, that according to Krugman rightwing economists don’t acknowledge evidence that falsifies their theories. If he is right, then why are those economists treated as serious people, rather than as cranks?

32

Metatone 10.25.13 at 9:56 pm

Bruce Wilder covers most of what I was going to say.

I’ll reiterate because maybe I’m not as nuanced as he is:

1) Like it or not, macroeconomists who get involved in policy matter quite a bit, because they have the power (e.g. Reinhart & Rogoff) to justify policies that cause enormous misery to many many people.

2) Microeconomists who don’t want to be tarred with that brush need to either start policing the dubious standards in macroeconomics, or they need to start splitting their discipline off. You can’t belong to the same professional bodies, publish in the same journals, feed off the same tenure tracks and then disavow any responsibility.

3) Further, the average microeconomist (and we get a lot of them in UK “health economics” so I’m speaking from a history of dealing with them) seems very prone to apply knee-jerk market theories to anything they don’t know much about. Again, if they don’t want to be tarred, they need to stop dipping in the tar barrel…

33

chris auld 10.25.13 at 10:22 pm

The idea that the use of the terms “neoliberalism”, “neoclassical economics”, or “neo-classical economics” is suspect is just stupid. Furthermore, Auld, in the comments section, falsely accuses another commentator of calling him a Nazi. Nobody should take this intellectually impoverished clown seriously or care if numbered items in his list apply to you. (I admit to a bias here.) Those who would like to see (too broad a group of) academic economists disrespected by others would want Auld to keep up his decades-long behavior.

My apologies, Bob, for failing to rise to the level of grace you invariably exhibit.

Chris Auld
Associate Smug Asshole and Adjunct Intellectually Impoverished Clown
Department of Economics
University of Victoria

34

Walt 10.26.13 at 7:23 am

Some of the points on that Chris Auld list are insane. “Neoclassical” is just what people outside the mainstream call the mainstream. Since “neoclassical” originally meant “marginalist” and since the mainstream analysis is part of a direct intellectual tradition that began with the marginalists, it even makes sense.

“Neoliberal” is a standard name for a certain political philosophy about markets and regulation.

35

Walt 10.26.13 at 7:26 am

Using Stephen Williamson for examples is not really legitimate. He is a living, breathing straw man.

36

ZM 10.26.13 at 9:48 am

“The sooner we realise that the entire discipline is in a state of scientific crisis the sooner we might start to do something about it.”
I don’t really understand why economics would be categorised as a scientific discipline rather than as one of the humanities in the first place. Given economic structures are largely culturally determined rather than natural, and that these structures change over time, and that an observer would have to take into account things like practice and individual agency, the object of study itself doesn’t seem appropriate to the scientific mode at all.

37

Barry 10.26.13 at 11:27 am

Donald Johnson: “You should blame Darwin for that–he’s the one who said that his theory could be disproven by the existence of an organ that couldn’t have evolved in a gradual step-by-step process. “

Incorrect; he said what he said; creationists look at anything that they don’t understand (or that their deliberately-kept-ignorant audience doesn’t) and claim that evolution is disproven.

Big difference.

38

UnlearningEcon 10.26.13 at 11:44 am

@chatham

If you want to get over term,s, jargon and wordplay, I’m with you. However, my point is that mainstream economists all too often deny there is a mainstream – there is “science” and there is heterodoxy – and definitions like this are meant to show them that, yes, their approach is just one way of doing things.

@walt

Yes, but he is so amusing. That he managed (16) on my list in a review of Quiggin’s book is quite remarkable.

39

Passing By 10.26.13 at 12:39 pm

Prof. Quiggin — “Standard neoclassical microeconomics is itself a macroeconomic theory in the sense that it’s derived from a general equilibrium model as a whole.”

The neo-classical micro-economics of individual markets existed, in well worked-out form, long before anybody embedded it in a GE framework.

The GE theorists came along with an extremely-restrictive model of the macro-economy that they could show was in equilibrium under certain rather-unrealistic conditions. And announced that their new toy underpinned those already extant theories … which had gotten on quite well without them.

Now you ‘re pointing out that the GE theory doesn’t account for the macro-economy. True enough, but talk to the GE guys, not the micro guys.

In my experience, nobody who applies micro theory to individual markets [e.g., to understanding a particular industry’s structure] worries about whether their application depends on the macro-economy being in general equilibrium.

40

Ronan(rf) 10.26.13 at 12:44 pm

Bruce Wilder
I’ve long wondered (genuinely) what school of economics would you subscribe to? I find your critiques of ‘neoclassicism’ always interesting, but curious what you’d replace it with….

41

William Timberman 10.26.13 at 1:08 pm

Ronan(rf) @ 40

Thanks for this. I’ve also been curious, but have refrained from asking myself because I assumed that the answer would be wind up being longer than would fit in a CT comment. (Certain that there had to be some, I even went so far as to look for Bruce Wilder books,)

Of course one can infer some things from the accumulated comments, but the circle of inference one gets as a result is far too large in diameter to be entirely satisfying. I hate to jump the gun on BW’s own reply, but if I had to guess, I’d guess that his take on the political part of Political Economy would be of the most value, at least to me, largely because a) that’s the part that interests me most, and b) the part that I would expect to be the most unorthodox — when measured against the current standards of orthodoxy anyway.

42

John Quiggin 10.26.13 at 1:23 pm

In my experience, nobody who applies micro theory to individual markets [e.g., to understanding a particular industry’s structure] worries about whether their application depends on the macro-economy being in general equilibrium.

That’s the point of the OP. The validity of the applications depends on implicit macro assumptions, but, as you say, this dependence is ignored. If the implicit macro theory is wrong, so is the derived micro.

43

Straightwood 10.26.13 at 4:11 pm

The ability of brilliant minds to justify almost anything is an embarrassing fact of history. Jurists in the German National Socialist regime neatly modified their legal theories to accommodate the regime’s murderous actions, retroactively legalizing summary executions and other “national security” measures. Heidegger, one of the most prominent philosophers of the last century, assented to the expulsion of the Jewish members of the faculty when he was rector of the University of Freiburg.

Thus it is not at all surprising that many prominent economists have become politically aligned with the political right wing and that this orientation has distorted their academic work. They view Keynsian economic theory the way some 20th century German scientists viewed quantum theory as “Jewish physics.” Ideology is easily masked as scholarship by a clever mind.

44

Peter K. 10.26.13 at 5:08 pm

As an amateur, self-taught marcoeconomist who has been following these debates since the 1990s, it strikes me that effectively the debate isn’t about micro vs. macro but on different macros which have different goals at the policy level.

In my mind Keynesian macro has held up, as Krugman points out. Prof. Quiggin’s zombie ideas which reappearerd after the recent crisis are an old-timey macro which echoed the ideas of the Austrians, the British Treasury View, and Mellon’s advice to Hoover: “liquidate, liquidate, liquidate.” Lowering unemployment and closing the output gap were not the goals unlike modern macro. Perhaps the micronauts have these unacknowledged, zombie ideas floating around in the background.

Or rather, they aren’t forced to choose between different versions of macro – one of which is wrong in many of its claims – and can simply focus on the micro.

45

Passing By 10.26.13 at 5:21 pm

Prof. Quiggin –“The validity of the applications depends on implicit macro assumptions, but, as you say, this dependence is ignored. If the implicit macro theory is wrong, so is the derived micro.”

The micro models I’m thinking of, and their applications, aren’t derived from GE theory. Of course, they can be so derived, with enough cobbled-together assumptions. But nothing about their logic requires a GE theory; which is why many of them antedate GE theory.

In particular, your statement in the original post that “[n]one of the standard conclusions of these fields of microeconomics can be assumed to be valid under conditions of sustained high unemployment” seems misplaced.

As an example, suppose we have a non-collusive duopoly market and want to know how its prices and market shares would shift with a specified third entrant. Not sure why the answer would change substantially under conditions of sustained high unemployment.

46

ZM 10.26.13 at 5:39 pm

JQ@42 “That’s the point of the OP. The validity of the applications depends on implicit macro assumptions, but, as you say, this dependence is ignored. If the implicit macro theory is wrong, so is the derived micro.”
I’m not really knowledgable about these things to a great degree, but you seem to me here to be ignoring the interdependence of these two aspects as the subject (author) arrives at one or the other. If micro corresponds in some way to ideographic information (maybe I’m wrong here?), an adult will always make observations and record them with a certain set of explicit or implicit assumption, conscious or otherwise. Macro would seem to correspond to nomothetic (?), but while nomothetic assumptions conscious or otherwise will guide the selection and depiction of ideographic information, an attempt at giving a nomothetic account of things would be arrived at through the subject’s study of the relevant ideography. I’m not sure if I’ve conveyed that well, and I guess it’s an obvious kind of point, but you seem to be leaving the interdependencies between the two out…

47

mattski 10.26.13 at 5:49 pm

However, my point is that mainstream economists all too often deny there is a mainstream

Do you have an example of this?

and definitions [jargon?] like this are meant to show them that, yes, their approach is just one way of doing things.

Do you have a ‘way of doing things’ that is distinct from, and news to, folks like Quiggin & Krugman, for example?

48

Lee A. Arnold 10.26.13 at 6:11 pm

John Quiggin: “The point can be made in more detail with respect to labour economics, finance theory, public economics and industrial organization. None of the standard conclusions of these fields of microeconomics can be assumed to be valid under conditions of sustained high unemployment.”

Aren’t the standard conclusions also invalid under conditions of any other divergent social cost as well? E.g. negative environmental externalities, etc? Wouldn’t the common-sense assumption (i.e. the non-economist’s assumption) be that an individual consumer’s preference to ACCEPT a high social cost is due to cognitive bias, ignorance, misinformation, and/or a really bad attitude?

49

mattski 10.26.13 at 6:16 pm

but have refrained from asking myself because I assumed that the answer would be wind up being longer than would fit in a CT comment.

Some people drink, some people go on…

50

john c. halasz 10.26.13 at 6:23 pm

@43:

“some 20th century German scientists viewed quantum theory as “Jewish physics.”

Werner Heisenburg.

@46:

Idiographic?

51

William Timberman 10.26.13 at 6:37 pm

mattski @ 49

Some people drink, some people go on…

Clearly you’ve never partied with Russians, who are masters of both, and have never been shy about combining the two with truly astonishing élan. I’ve learned a lot from them, but unfortunately have never been able to remember any of it after I finally regained consciousness, which was usually just as the sun was disappearing below the horizon yet again….

52

Herman 10.26.13 at 7:02 pm

John Quiggin:

Most notably, in a competitive GE with full information, no externalities and so on, prices of goods reflect the social opportunity cost of producing them. This means, that, other than redistributing the initial endowments of property rights, governments can’t do anything to improve on the competitive market allocation of resources.

Once you have involuntary unemployment, all of this fails.

Yes, but it fails at full-employment too since the required assumptions are impossible to attain in the real world, and the model is not robust to even slightly relaxing many of it critical assumptions. And, the theory of the second-best lurks.

The claim the the Arrow-Debreu GE has absolutely no real world domain of applicability isn’t some new or heterodox claim, it has been made by GE pioneers themselves.

Fank Hahn:

One would have to be far gone in philistine turpitude not to appreciate the towering intellectual achievement which the existence result and the whole Walrasian edifice represents. Nevertheless, this should not blind us to the possibility that what is constructed there has little if nothing to do with the world which we inhabit. Nevertheless, it cannot be denied that the prospect of so many refining the analysis of states which they give no reason to believe have ever or can ever come about is scandalous. It is probably also dangerous.

53

Tim Wilkinson 10.26.13 at 7:43 pm

just a tiny, digressive point: non-collusive duopoly should be ‘uncooperative duopoly’. If both adopt a tit-for-tat strategy in the anticompetitive PD game and they can once manage to reach the co-operative state, the result is cooperation functionally equivalent to, but not involving, some simple forms of collusion, e.g. most obviously in pricing. And they can reach that ccooperative state by other slightly more complex strategies which involve only slightly more exposure.

54

Passing By 10.26.13 at 7:59 pm

Tim Wilkinson —

Digressions are half the fun …

Don’t disagree with the point you’re making. However, I wrote “non-collusive” for a reason: it summarizes the relevant legal constraint in most jurisdictions. Applied analysis is interested in the full range of legal outcomes … including the sort of tit-for-tat outcome you describe.

55

John Quiggin 10.26.13 at 8:43 pm

@Herman and others: I’m aware that the standard assumptions of the GE optimality result don’t hold. But these are the kinds of problems microeconomists are happy to deal with – eg proposing Pigovian taxes as a response to externality, antitrust for monopoly and so on. The standard analyses of these things assume that the economy is at a full employment equilibrium.

As regards your quote from Hahn, I’m pretty confident he was making the same point as me – he was, after all, a leading Keynesian macroeconomist, and he certainly said things like this on other occasions.

56

John Quiggin 10.26.13 at 8:55 pm

@Passing By In a recession, and seeking protection from entry, incumbents commonly argue along the following lines.

We are continuing to pay underused workers so that we will be able to expand output when the economy recovers. We can do this because we are not competing aggressively on price. But, if a new entrant starts a price war, we’ll have to sack lots of people. In the end, at least one firm (the entrant or one of the incumbents) will go bust, and the productive capacity of the industry will be lower than without entry. Meanwhile, our job cuts will raise the general level of unemployment

Can your model show whether or not this reasoning is justified?

57

ZM 10.26.13 at 9:07 pm

@50 Sorry, I have awful spelling sometimes. I intended ideographic, obviously. Is that sort of the equivalent of microeconomics in the economics field?

58

John Quiggin 10.26.13 at 9:24 pm

“The neo-classical micro-economics of individual markets existed, in well worked-out form, long before anybody embedded it in a GE framework. “

This is just wrong. Walras created GE theory at the same time as he (and, independently, Jevons and Menger) developed neoclassical micro. The marginal conditions for GE are precisely those used to resolve the problems of classical economics.

59

Herman 10.26.13 at 9:46 pm

@John Quiggin

I wasn’t disagreeing with anything you wrote, merely highlighting a point that wasn’t explicit in the post.

Do we really need to shout from the rooftops about the complete irrelevance of GET even at full employment? Incidentally, my emphasis was on the lack of robustness, rather than merely lack of realism. After all, doesn’t everybody know this?

Not quite. In the policy debates of the last few years, it has become quite clear that a lot of freshwater macro people either really do not know basic micro or pretend not to. eg the frequent insistence that the default assumption must be that the FWT holds, unless we can explicitly model why it fails. eg Robert Barro or this student of Steve Williamson.

Things aren’t helped by the fact that both, the Neoclassical synthesis and the “New Neoclassical synthesis” (ie NK macro, Goodfriend and King 1997) assume that A-D is a useful approximation at full employment. Yes, I know that Krugman has recently broken away from this consensus — much to the chagrin of Roger Farmer — and that you agree with him.

To reiterate, I agree with what you say.

60

mattski 10.26.13 at 10:26 pm

William @ 51:

Touche!

61

John Quiggin 10.26.13 at 11:11 pm

BTW, I use “neoclassical” to refer to something specific, not as a pejorative. As a first cut at a definition, it’s the body of microeconomic analysis arising from Jevons and Walras, formalised into the supply-demand model by Marshall and then mathematized by Samuelson, Arr0w-Debreu and so on. It includes standard analyses of market failure, externality, monopoly and so on

Mainstream, but not neoclassical, economics includes

* Keynesian macro (as I mentioned, synthesized with neoclassical micro by Hicks in away that no longer commands general acceptance)
* Behavioral econ
* Game theory (this is a controversial call, but I’m willing to defend it)

I find that people who use “neoclassical” as a pejorative often switch between using it as a synonym for those who would describe themselves as “mainstream” and some specific meaning, often narrower than the one I’ve given here (for example, excluding market failure).

62

Richard H. Serlin 10.27.13 at 6:34 am

“that only microeconomics, derived rigorously from rational behavior, is real science.” – Paul Krugman today.

This is a key root of the problem, a very snobby and harmful definition of “science”. It’s not at all just, as they say, that all economics must be derived directly from microfoundations – as bad as that is, where we throw induction out the window, which has been incredibly powerful in scientific advancement. See: http://mainlymacro.blogspot.com/2012/03/microfoundations-is-there-alternative.html.

It’s that it has to be only a certain kind of deduction, where you must assume every microunit, i.e. person, is 100% rational, has 100% of all public knowledge in their heads, and has 100% perfect education and expertise in every area and sub-area, so they can always figure out what’s perfectly optimal, and with no time or energy cost. Not just only microfoundations is scientific, but only microfoundations that are very or incredibly unrealistic.

So, a key root problem is this extremely limiting and harmful snooty definition of “science” (that not coincidentially makes libertarianism and plutocracy look a lot better). What would be a much more useful definition, one that would create far more utility for society? How about just broadly, logical and truthful? The truth, itself, would have to be based on at least assumptions like what we see actually exists, the mathematical axioms, etc., but just basic commonly agreed assumptions like that, and the truths would often be probability based, and our best efforts, but at least that should be the effort.

This would include the models that dominate macroeconomics currently, because they are logical and true given their assumptions, but it would not include the false overly literal inferences to reality that these economists so often make.

And a nice definition of science like that would not just exclude often comically false overly literal interpretations to reality of freshwater and libertarian biased economists, it would also include microfoundations models that were based on more realistic microfoundations. What would such microfoundations look like? Please see my next comment.

63

Richard H. Serlin 10.27.13 at 6:44 am

“I suspect nearly all economists are naturally reluctant to embrace cases where agents appear to miss opportunities for Pareto improvement – I give another example related to wage setting here. However in most other areas of the discipline overwhelming evidence is now able to trump these suspicions. But not, it seems, in macro.”

– Simon Wren-Lewis, at: http://mainlymacro.blogspot.com/2013/10/nominal-wage-rigidity-in-macro-example.html

Consider these microfoundations from Chicago economist Richard Thayler:

• Suppose you had $100 in a savings account and the interest rate was 2 percent a year. After five years, how much do you think you would have if you left the money to grow? More than $102, exactly $102 or less than $102?

• Imagine that the interest rate on your savings account was 1 percent a year and that inflation was 2 percent. After one year, would you be able to buy more than, the same as or less than you could today with the money?

• Do you think this statement is true or false: “Buying a single company stock usually provides a safer return than a stock mutual fund”?

Anyone with even a basic understanding of compound interest, inflation and diversification should know that the answers to these questions are “more than,” “less than” and “false.” Yet in a survey of Americans over age 50 conducted by the economists Annamaria Lusardi of George Washington University and Olivia S. Mitchell of the Wharton School of the University of Pennsylvania, only a third could answer all three questions correctly.

At: http://www.nytimes.com/2013/10/06/business/financial-literacy-beyond-the-classroom.html?pagewanted=all

And these:

From a Center for Economic and Financial Literacy survey:

“65% of people answered incorrectly when asked how many reindeer would remain if Santa had to lay off 25% of his eight reindeer.”

“1 in 3 people didn’t know how much money a person would be spending on gifts if they spent 1% of their $50,000/year salary.”

– Personal Finance for Dummies, 7th edition, 2012, page 9

It’s not just the issue that they require that all macro models be microfounded, with the resulting limitations and problems (see: http://richardhserlin.blogspot.com/2012/03/haugens-critique-of-microfoundations-in.html). They also require that the microfoundations always be highly unrealistic in the way that they like, the way that fits their preferred paradigm, and/or makes their preferred libertarian ideology look more desirable.

One thing I find really interesting: I often hear economics and finance professors talk about how ignorant and incapable their students are – and then in their research they assume everyone is a genius! with not only perfect public information in their minds, but perfect expertise to analyze it with! And they see no contradiction (or don’t care).

Part of this is many will claim that you don’t need everyone to be smart and expert and informed to get the result, you only need a savvy minority. But for many things it’s easy to show that won’t be enough. Complete markets and perfect arbitrage won’t often exist like they do in models, and with regard to the smart getting more and more control of money, or becoming more and more common, remember people don’t live forever and they spend up lots of what they gain; the smart die off, and paraphrasing P.T. Barnum, there’s a new sucker born every minute, and anyway from Keynes, in the long run,… As well, the savvy are limited in how much they can push an inefficient price to its fundamentals by how undiversified their portfolios will become as they buy more and more, so each additional unit becomes riskier and riskier, more and more eggs in the same basket, and so worth less and less to them. For more on this, see this post of mine (http://richardhserlin.blogspot.com/2013/09/the-intuition-for-wallace-neutrality.html) which was just quoted heavily by Miles Kimball (http://blog.supplysideliberal.com/post/63442789541/wallace-neutrality-roundup-qe-may-work-in-practice)

64

ZM 10.27.13 at 8:36 am

@62 “So, a key root problem is this extremely limiting and harmful snooty definition of “science” (that not coincidentially makes libertarianism and plutocracy look a lot better). What would be a much more useful definition, one that would create far more utility for society? How about just broadly, logical and truthful? “

Although I suppose anyone may change the meaning of any word like Humpty Dumpty in Alice in Wonderland, if we simply take the concept of science (give it whatever name you wish to), I still fail to see why one would classify economics as a science unless one chooses to classify every discipline and body of thought as science.

Logic and truth are not typically only the properties of the pure sciences. Nor, in terms of a critique of libertarianism and plutocracy, are they necessarily involved in value judgements, in determining the good from the bad, the wise from the unwise etc.

Scientific seems to have come into use in early modern English through compounding the Latin scientia (knowledge – from Greek and Indo European roots to split) and ficus (making) to translate Aristotle’s epistemonikos which he contrasts with doxastikos. Epistemonikos “deals with the universal and the necessary, the body of facts resulting from demonstrations are universal statements that are always true.” (Hanley, Being and God in Aristotle and Heidegger). Doxastikos on the other hand was held to relate to an opinionative logos.

65

notsneaky 10.27.13 at 8:42 am

This is just wrong. Walras created GE theory at the same time as he (and, independently, Jevons and Menger) developed neoclassical micro.

Cournot had them all beat though. I guess the question is, what does “fully worked out” mean.

66

notsneaky 10.27.13 at 8:53 am

Btw, the fundamental difference between Chris’ post and UE’s is that Chris says “these are bad arguments” while UE says “these are bad people“. At best UE should put in a lot of instances of “some” in his post. That’s mostly why Chris makes valid points while UE is just trolling.

A secondary, but telling ,difference is that in his post when Chris provides links for support he links to actual examples of bad arguments. UE mostly just links to his own past posts, which makes it look like there’s some kind of support or reference, but it really is just circular self important posturing.

67

Robert 10.27.13 at 10:26 am

Poor Chris Auld knows my name is not Bob.

“UE says ‘these are bad people“. As far as I can see, this is simply false.

For what it is worth, I wrote most of the Wikipedia article on Neoclassical Economics. Of course, if you question any particular phrase, it would take me quite a bit of work to figure out whether I wrote that bit or put it exactly like that. I am comfortable with the idea that, say, game theory is mainstream, but not neoclassical.

I also wrote much of the Wikipedia article on General Equilibrium Theory. I did not write the section at the end on Other Schools. Nor did I write such untrue and unsupported lines as:

“As with all models, this is an abstraction from a real economy…”

“Therefore, when equilibria arise that are not efficient, the market system itself is not to blame, but rather some sort of market failure.”

“Supporters have pointed out that [non-uniqueness] is in fact a reflection of the complexity of the real world and hence an attractive realistic feature of the model.”

68

Martin 10.27.13 at 10:55 am

“The basic problem is that standard neoclassical microeconomics is itself a macroeconomic theory in the sense that it’s derived from a general equilibrium model as a whole. “

I don’t think that there would be many economists questioning that. If you believe that the central bank can stabilize the economy, then you also must believe that the central bank can destabilize the economy. If there is any disagreement it is about what counts as stabilization policy. Is there a role for fiscal policy? Is there a role for financial regulation?

69

UnlearningEcon 10.27.13 at 11:41 am

I link to some past posts because, unlike Chris, I wanted to do readers the service of explaining why these were bad arguments, rather than smugly assuring myself that they ‘just were’. It was natural that I’d made some of the points I made in that article before, so I linked to myself in places, and more often than not these posts contained clear examples of economists saying what I was criticising.

Anyway, isn’t most of the blogosphere just “circular self important posturing” ?

70

mattski 10.27.13 at 12:03 pm

Anyway, isn’t most of the blogosphere just “circular self important posturing” ?

Well, alright then. There you have it.

71

ed 10.27.13 at 1:03 pm

I’ll repost my question from earlier, in slimmer form:

what currently available intro to micro or intro to macro textbook would you recommend as both pedagogical enough and correct enough to be a good starting point to read on a freshman level?

Answering that question, and spreading that answer, is probably among the most important things to do for anyone critical of the current dominant themes within economics.

72

Jerry Vinokurov 10.27.13 at 1:53 pm

It’s that it has to be only a certain kind of deduction, where you must assume every microunit, i.e. person, is 100% rational, has 100% of all public knowledge in their heads, and has 100% perfect education and expertise in every area and sub-area, so they can always figure out what’s perfectly optimal, and with no time or energy cost. Not just only microfoundations is scientific, but only microfoundations that are very or incredibly unrealistic.

There go those nasty axioms again, always rearing their head. Turns out, when you start from demonstrably false assumptions, your theory might not end up reflecting what actually happens! Who coulda thunk it, except for, you know, most philosophers ever.

73

Rakesh Bhandari 10.27.13 at 3:12 pm

I have yet to see in a standard economics textbook any working out of what seems to be fundamental problem in equilibrium analysis.

It is usually presumed that the movement to equilibrium is rapid enough that the supply curve would not have jumped in the meantime.Neva Goodwin, et. al do realize certain to their great credit that “the forces that can work against quick movement towards equilibrium include such human characteristics as habit and ignorance. Slow production processes and long-term contracts may also slow down or stop adjustment.” p. 93

But if the adjustment process is slow enough, then it seems to me that due to on-going techno-organization change the supply curve will have shifted before the adjustment process is complete, so that the economy is best understood as one of permanent disequilibrium.

74

roger gathman 10.27.13 at 4:46 pm

I don’t buy into the idea that falsification is the defining principle of science. Have we learned nothing from Kuhn or Lakatos? But it does play a role, at least in the way sciences go at constructing and solving their problems.
In economics, there is a strangely broad area in which falsification is countered by assumptions that are granted to be fictitious. However, there are certain notions that are flourished about as if they surely must be true. For instance, preferences in the neo-classic paradigm, codified by Arrow and Debreu, are invariant and logically sorted by a simple transitivity rule, so that if preference a is preferred to b and b to c, a is preferred to c. The reason for this, it turns out, is that the transitivity rule is pretty. There is no test ‘in the wild’ that has ever reproduced this theorem.
Is it then thrown out? No, it is not thrown out.
What does happen, though, is that such fake insights – and here I’d include the maximizing behavior supposed to be the only factor necessary for an economy among individuals – are incorporated in policy that does nudge people towards more, say, maximizing behavior. Economists are always experimenting on populations, but they like to pretend that they aren’t, that that is only done by those nasty central planner types. What nonsense. In fact every economic policy suggestion is an experiment. Neo-classical economics just codifies certain experiments that have been run on the population since I’d say capitalism got upon its hind legs. In themselves, it is true, economists aren’t that important – but just as theologians in the Middle Ages created the theoretical structures needed for subduing peasant populations and changing their customs, economics is important for the way it works on the self-understanding of people under capitalism. Its greatest successes as a hermeneutic have been in Anglophone countries, which probably has a very complex sociological explanation – perhaps having to do with family dynamics. I think the great success of economics as a discipline has not been the light it throws on capitalism, but the way it makes people think that the typical subject is not someone so embedded in a family and community that far from thinking about his or her self-interest, most of his or her time and energy is spent thinking about and doing things for others – which is how most people I know simply live – but a robust profit maximizer. The true egoist is a rarity in nature and society. It isn’t just that we satisfice – I would say that it is a rare person who sits down and thinks of what she actually is and deduces her self interests from that. In the eighteenth century, perhaps, there were people who were doing that in their chateaux, like De Sade, but otherwise it just isn’t happening. Individualism is a nice credo, but it simply doesn’t crop up much in reality.

75

Rakesh Bhandari 10.27.13 at 4:53 pm

Another way in which a certain kind of individualism is suspect…
“….social conformity or emulation–such as wanting a consumer good because
other people already have it–can create positive feedback in the market,
with the potential for destabilization. In the simplest terms, erratic and
fragile market bubbles or ‘cascades’ can occur if individuals consider the
behavior of others to be a better source of information than their own
knowledge or preferences. ” Frank Ackerman, “Still dead after all these
years: interpreting the failure of general equilibrium theory” Journal of
Economic Methodology, 9:2, 119-139, 2002, p. 134

76

Alex K. 10.27.13 at 5:18 pm

I was going to write that John’s premise that microeconomics assumes macro claims is complete nonsense for obvious reasons: microeconomics makes some assumptions which then can result in various macroeconomic properties — it’s completely the reverse of the causality that John implies.

“In the sky” tried to point this out gently, by reminding John that you can make microeconomic assumptions (e.g. price rigidity) that result in unemployment even in microeconomic models. But John didn’t get the hint, and continued with the nonsense that microeconomics makes macro assumptions.

Now Krugman makes the same mistake, so if Krugman can be an idiot about this point, then maybe it needs to be made explicit.

Krugman’s other point, that historically, people used to take math-heavy microeconomics seriously only because of the need for macro economics is similarly without any weight: macroeconomics still does not have any prospect for being a science (There are strong negative aggregation results and “anything-goes” results that make the prospect of microfoundations for macroeconomics hopeless. Without microfoundations, the stochastic assumptions needed for doing empirical work are just pulled out of one’ s hat: i.e. they are unscientific. Furthermore, relaxing the hyper-rationality assumptions only increases the degrees of freedom of the system, which makes the negative aggregation results and the “anything goes” results only stronger, so that is not an available alternative for microfoundations.)

On the other hand, there is no reason why a microeconomic theory, limited to a well defined domain, cannot be scientific: you just model enough of the details of the situation until you get something more or less scientific. Of course, General Equilibrium will not get you there and neither will any psychological theory (even if you have a scientific psychological theory that describes correctly the mean and variation of some psychological index for the the entire population, the people making important economic decisions are often selected from the tails of that distribution; so you can not just assume that all important economic decisions are made according to the psychological theory) But it should still be perfectly possible to made scientific models of very concrete microeconomic situations.

77

Rakesh Bhandari 10.27.13 at 5:29 pm

Paul Samuelson: “The correct (equilibrium) set of prices of consumption goods and of productive services, the market quantities of outputs and inputs–all these are ‘unknowns’ whose numerical values are determined by a vast set of ‘simultaneous’ equations’: the condition that all prices be equal to producers’ ‘marginal costs,’ and to consumers’ relative ‘extra utilities’: that wages equal ‘marginal revenue productivities’; that profits be at a maximum, etc. Given such noneconomic facts as (1) production, (2) technology, (3) tastes, and (4) distribution of ownership of property, there are just enough individual and market relations or equations to determine a unique economic solution. Everything depends on everything else, but all together determine each other in just the way that stationary balls in a bowl mutually determine each other’s position. Who solves the complex equations of economic life? Certainly not mathematicians. Certainly not government bureaucrats or congressmen. Every person–whether a businessman, housewife, farmer or wage earner–is helping to solve them every time he decides to use one kind of labor and not another, or decides to buy butter and not oleomargarine, or decides to plant more or less intensively so much corn and so much wheat, or decides to quit glass blowing for type setting. We don’t have to use slide rules, and we don’t have to understand the pricing system to contribute to the solution of ‘general equilibrium….Prices keep moving as a result of our readjustments of behavior. If outside factors such as inventions, wars, or tastes were to remain constant long enough [IF!!!rb], then we might finally approach the ‘general equilibrium set of prices,’, at which all the forces of supply and demand , value and costs, might just be in balance, without any tendency to further change. Of course, in the real world, outside factors never stand still, so that as fast as equilibrium tends to be attained it is disturbed. Still, there is always a tendency–at least in a ‘perfect’ competitive system–for equilibrium to reestablish itself, or at least to chase after its true position. (Economics, 1981, ed.:p.593-94)

Isn”t the tendency for the economy itself to generate forces that disrupt any equilibrium before it is likely attained at least equally strong as this tendency towards equilibrium?

Or is the point that in the actually non-existent perfectly competitive system there is no endogenous tendency for disequilibrium shifts in the supply curve because firms don’t have the profits with which to make techno-organizational changes? And so therefore it’s ok in the mathematics of the actually non-existent perfectly competitive system to ignore endogenous causes of disequilibrium?

Doesn’t Schumpeter’s utter failure to synthesize a static Walrasian model with his own realistic dynamic vision not point to the simple impossibility of equilibrium economics?

78

William Timberman 10.27.13 at 5:48 pm

An interesting juxtaposition of comments, these last two (75 and 76, at the time I’m writing this). To paraphrase my understanding of Alex K’s contribution, he seems to be insisting that as long as we bite off small enough pieces, we can chew on as much reality as we like — and avoid messy things like psychology. Rakesh Bhandari, on the other hand, does a nice job of hoisting Samuelson on his own petard, but I wonder if he notices that Samuelson’s dump the ball into the box and pray, followed by his but aren’t we all screwed anyway? bears an eery resemblance to some of the old Soviet academicians’ debates about central planning.

In reading discussions of economics, macro or micro, I sometimes get the feeling that some people know too much, while others — like myself — know too little. Politically speaking, that might be a corrosive rather than explosive formulation, but it could easily be just as damaging to our hopes of doing better bye-and-bye.

79

roger gathman 10.27.13 at 6:10 pm

75 seems to me a classic instance of the rosary argument – that is, in a classic sense, there is no argument made there. Rather, much as you simply repeat the rosary in order to do the rosary, you repeat an argument that was made once and then countered as though the repetition was a refutation. And by so doing you reveal that it isn’t an argument that is being made at all, but rather, it is a magical charm.
Economics is full of magical charms. For instance, by adding scientific to model, you don’t really add anything to what you are saying – it is simply a model, in the same way that there is a model for the personality of those born under the sign of sagittarius – but you have assured yourself that it is scientific. How? By calling it scientific.

80

Alex K. 10.27.13 at 6:12 pm

“Alex K’s contribution, he seems to be insisting that as long as we bite off small enough pieces, we can chew on as much reality as we like — and avoid messy things like psychology.”

Whether we need psychology or not depends on the situation. If we want to model the behavior of people buying in a supermarket, then psychology might be useful, as the shopping population is representative enough for the general population. Although even then, you might need to make assumptions about the culture of the population. The silly tricks marketers play on the US population might not work as well for the price conscious population of China.

But if you want to model the some high-impact decision making, then psychology is not all that useful, as the people making those decisions are not representative of the psychological profile of the general population.

81

Alex K. 10.27.13 at 6:14 pm

” And by so doing you reveal that it isn’t an argument that is being made at all, but rather, it is a magical charm.”

This seems to be a perfect description of your post. Do you have any idea to what I am referring to when I talk about negative aggregation results and “anything goes” theorems?
If not, then maybe you have no business commenting on those issues.

82

William Timberman 10.27.13 at 6:20 pm

Alex K. @ 79

…the people making those decisions are not representative of the psychological profile of the general population.

Let me tell you about the very rich. They are different from you and me. They possess and enjoy early, and it does something to them, makes them soft where we are hard, and cynical where we are trustful, in a way that, unless you were born rich, it is very difficult to understand. They think, deep in their hearts, that they are better than we are because we had to discover the compensations and refuges of life for ourselves. Even when they enter deep into our world or sink below us, they still think that they are better than we are. They are different.

To all of which I say: bollocks. That doesn’t make me Hemingway, but then I’ve never needed to be.

83

Alex K. 10.27.13 at 6:26 pm

It’s not about the very rich William. The silly tests psychologists use for testing the answers of the general population regarding computing some expected value will work very differently in the department of mathematics. If the decisions makers are selected from departments of mathematics, then psychological results about how people make certain decisions will not apply to them.

This does not mean that therefore we should assume that everyone is rational — it just means that psychology is not a satisfying answer to the problems of unrealistic assumptions in economics.

84

mattski 10.27.13 at 6:51 pm

Alex,

There are strong negative aggregation results and “anything-goes” results that make the prospect of microfoundations for macroeconomics hopeless.

Well then, I’ll ask on Roger’s behalf (and mine.) Would you explain what you are talking about here?

85

john c. halasz 10.27.13 at 6:59 pm

@80:

I’m pretty sure Roger, as well as most people here, are aware of Mantel-Sonnenschein-Debreu.

86

mattski 10.27.13 at 7:06 pm

(I see a parallel between conservative arguments against, for example, regulation of financial transactions and the dismissive swipes at macroeconomics here. “Regulation? Impossible to fully and fairly implement, so why bother?” “Macro? Impossible to fully model reality, so why bother?” … Krugman, to take the best example I know of, understands that economics is an approximate science. Of course it has limitations, but if you dismiss the whole thing what are you left with? Cult-of-the-day?)

87

Nine 10.27.13 at 7:24 pm

Alex K.@82 – “If the decisions makers are selected from departments of mathematics, then psychological results about how people make certain decisions will not apply to them.”

Is the assumption here that because mathematicians are knowledgeable on decision and Game theory they are therefore immune to cognitive error ?

88

William Timberman 10.27.13 at 7:28 pm

Alex @ 83

It’s not about the very rich William.

No, it’s not. It’s about unjustified assumptions, chief of which is that people with the power to make decisions are a) uniquely qualified to make those decisions, and therefore, b) unlike the rest of us in significant ways, and c) as, such, prima facie proof of the futility of involving anyone without power in the decision-making process.

Psychologists make all sorts of wacky interventions in our understanding of how human beings work, as often as not because they’re chasing the same acknowledgement as scientists that economists are. Nevertheless, to decide that the psychology of a trained mathematician, even acting in his official capacity, is not only different from, but superior to that of a randomly-selected member of the great unwashed is quite simply a bridge too far. Do you really think that advertising executives are unaffected by the billboards that they themselves have had a hand in creating? If so, I doubt you understand as much about how economics works as you claim. It’s not your reasoning I doubt, you understand, but your premises.

89

William Timberman 10.27.13 at 7:30 pm

Nine beat me to it, and was much more concise in doing so as well. Credit where credit is due.

90

roger gathman 10.27.13 at 7:31 pm

Alex K is aggressive without being meaningful. However, I am a little encouraged by the fact that I’m cast outside with ignoramuses like Krugman.
Even so, he is a pretty good example of what is wrong with economics. The outrage that the vulgar would dare to intrude on occult secrets such as are possessed by the Great Alex is a rhetorical bit that you see all the time in the writings of the humoral physicians in the 17th century. How dare you interfere if you can’t distinguish the ileaster from the archaeus, and you have completely messed up the categories of the semina and the vesicles through which the animal spirits travel, in conjunction with the house of Jupiter, which can only be got by the most minute mathematical and scientific modeling! .
It is typical of the primitiveness of economists, from the perspective of sociology or psychology, that they think they can make claims like Alex K’s claim about mathematicians, without actually referencing a single empirical study, or even understanding how they are done. “The silly tests psychologists use for testing the answers of the general population regarding computing some expected value will work very differently in the department of mathematics. If the decisions makers are selected from departments of mathematics, then psychological results about how people make certain decisions will not apply to them.” This is delightful: true early modern thinking, in which experiment is disdained in favor of deduction. Many an alchemist or astrologer could agree entirely with the sentiments expressed in this form of thinking.
I don’t think all economics still resides in the pre-scientific age, but Alex K. is not so untypical as all that. One would have to do a larger study to see whether he’s an outlier or a more typical specimen of what passes for economic science nowadays, of course. But it would be interesting to see.

91

bob mcmanus 10.27.13 at 7:35 pm

74: I think the great success of economics as a discipline has not been the light it throws on capitalism, but the way it makes people think that the typical subject is not someone so embedded in a family and community that far from thinking about his or her self-interest

Well, I glanced at Gathman’s blog, and withheld comment because I presumed he understood that a focus on a humanistic individualism is not at all unique to neoclassical economics, but is an important part of the Western Enlightenment project. “Rights” inhere to individuals not groups, states, classes, families, so we must presume that rationalities, agency, etc also are a capability of the individual. And so on.

There are alternative initial presumptions, that agency and rationality is embedded in institutions (Keynes) or classes (Marx). And older systems. These will usually entail costs to individual autonomy we might find unacceptable.

92

Nine 10.27.13 at 7:42 pm

William Timberman@89 – If that’s Alex’s implicit assumption then it’s easy to see why he’s down on “psychology” by which i think he means behavioural economics. Anecdotally, I’m not sure if Monty Hall qualifies as a cognitive puzzle but no less than Paul Erdos himself got it wrong.

93

Rakesh Bhandari 10.27.13 at 8:02 pm

While aggregating a social demand curve may prove difficult with, say, the complications of income effects from price declines and the resort to a representative agent may be somewhat pathetic-I disagree with the importance that neoclassical critics such as Steve Keen and Duncan Foley give to this– consider the line of criticism that begins with taking the endogeneity of preference seriously.

Kaushik Basu notes (and this too shows how micro depends on macro in a certain way): “When we do have preferences, these often change depending on what others do and on what kind of equilibrium comes into existence…This immediately means that the methodological, individualistic assumption that one can begin by first describing the individuals and then going to to describe the society that they will create is untenable, since the society they create may in turn alter the way that the individuals are characterized–for example their preferences. For one it can easily lead to the non-existence of any equilibrium, which would leave much of our analysis in a vacuum. It could also lead to multiple equilibria, opening up interesting new questions concerning government intervention.” p. 33 of Beyond the Invisible Hand.

94

ZM 10.27.13 at 8:19 pm

@86 “Krugman, to take the best example I know of, understands that economics is an approximate science. Of course it has limitations, but if you dismiss the whole thing what are you left with? Cult-of-the-day?)”
What do ou mean by an approximate science exactly? Approximate means close but not the same as far as I am aware, so you seem to be saying that economics is close to a science but not a science?? And therefore, if it is not a science, what is it? You seem to be saying that it’s not a science, but if one were to allow it were not a science it could therefore only be a cult-of-the-day, presumably phrased as such to show a disdain for all that is not science (better to masquerade as science than the alternative)…

95

Alex K. 10.27.13 at 8:49 pm

As expected, Roger Gathman comes up with just hot air. And no, Roger, I don’t put you with Krugman at all — you would have to make some sort of argument in order for me to do something like that.

Nine: “Is the assumption here that because mathematicians are knowledgeable on decision and Game theory they are therefore immune to cognitive error ?”

No, just that they will be much better at avoiding some of the silly mistakes the general population makes on simple probability tests, therefore an empirical result, even a scientific one, about how the population does on that test does not allow you to model all important economic actors as making that mistake.

I’ll repeat: I’m not claiming that assuming that we have only rational actors is the correct way forward — only that naive psychologism is not the right way either.

Mattski, I am indeed talking about the Sonnenschein–Mantel–Debreu theorem and also about Franklin Fisher’s results about the impossibility of having an aggregate production function represent all the micro production functions, even if you want the representation to be just an approximation. These results are not formal proofs that scientific macroeconomics is impossible, but they are pretty strong hints.

96

Andre Cieplinski 10.27.13 at 9:08 pm

Ed @5 and 71: I think you could build up a decent knowledge in non mainstream economic theory out of few books. They aren’t exactly introductory textbooks, but none should be too hard. I’d recommend

For macroeconomics: Lavoie, M (Introduction to post-keynesian economics); Snowdon, B. & Vane, H. (Modern macroeconomics: its origins, development…), Lopez, J. & Assous, M. (Kalecki) or just go for Kaleckis originals papers or Theory of Economic Dynamics, they are not hard and very straightforward.

For microeconomics: Sylos-Labini, P. (Oligopoly and Technical Progress); Steindl, J. (Maturity and Stagnation in American Capitalism); and for a critique of neoclasical micro SRAFFA, Piero. The laws of returns under competitive conditions. The Economic Journal, v. 36, n. 144, p. 535-550, 1926.

Hope this is of any help, good luck

97

John Quiggin 10.27.13 at 9:43 pm

” These results are not formal proofs that scientific macroeconomics is impossible, but they are pretty strong hints.”

SMD shows the impossibility of macro in much the same way as Godel proved the impossibility of math, and with much the same effect (that is, almost none).

98

Collin Street 10.27.13 at 10:23 pm

As expected, Roger Gathman comes up with just hot air.

The same person will make the same mistakes: you can’t see your own mistakes until you’ve become someone different to the person you were when you made them.

Until that time, the logic of people pointing out your errors is going to look erroneous to you: you’ll see the errors you made, but from the inside, reversed.

I mean, yes, it’s possible that you’re right and the other person is wrong… but logically you’d expect errors to be evenly distributed, that in roughly matched company you’ll be making about the same number of errors as the next bloke, no? If you [believe that you] are getting strike rates significantly above fifty percent in disputes, it’s pretty solid evidence that you’re missing your errors.

99

ZM 10.27.13 at 10:38 pm

@97 Genuine question, isn’t Godel more about the necessary incompleteness of maths? I wouldn’t really agree that impossibility and incompleteness are exactly the same thing – incompleteness simply implies to me that something has its limits, whereas impossible implies something is not able to be done at all – so then maths can be done (it does seem to be done) within limits….

100

Bruce Wilder 10.27.13 at 10:44 pm

Ronan(rf) @ 40: . . . what school of economics would you subscribe to?

I’m an institutionalist. I think about political economy — the system of production and trade — as being structured and governed by institutions, the traditional primary topic of sociology (not that I know anything about sociology). Paul Krugman’s blogpost in response to the OP by Quiggin is worth looking at, for a brief precis on what happened to the institutionalists, who once dominated economics. It is a very useful complement to the OP. Krugman basically says that Keynes and Samuelson killed institutionalism, and that was, in his opinion, mostly a good thing, because the institutionalists talked too much and couldn’t come to a conclusion. In relation to Quiggin’s point, Krugman says — correctly, imho — that the apparent success of Keynes’ prescription, as carried out in WWII, combined with Samuelson’s codification of neoclassical economics into a Euclidean geometry, eliminated the institutionalists from the academy. John Kenneth Galbraith was the last prominent public intellectual with an institutionalist viewpoint; the hapless Arthur F. Burns the last Fed chairman trained with an institutionalist approach to the business cycle.

The institutionalists were the field workers of economics, the people, who went out and looked at what was going on and tried to understand it descriptively and operationally. In the days before government statistics departments created vast databases of survey research, they were the ones, who painstakingly assembled tables of iron production or unemployment rates, or traced out the historical ups and downs of the business cycle; in fact, the institutionalists were largely responsible for creating those government statistics departments. The first generation after the war — people like Samuelson himself — had been taught by institutionalists, and had a fund of factual knowledge and appreciation that reality did not conform to theory. Even the pre-Friedman Chicago school of George Stigler et alia knew that the real world was necessarily messier than their theoretical idealized analysis. But, people of Krugman’s generation did not have that appreciation or acquaintance, and economics, as a discipline lost its balance, becoming top-heavy with theoretical analysis, especially after the failures of the first efforts at macroeconomic forecasting with computer models.

Krugman’s caricature of institutionalists, confronting the Great Depression with helpless wordiness, is deeply ignorant as well as unfair. In fact, they were very busy with the institutional agenda of the New Deal — they, or their ideas, were why the New Deal could acquire an institutional agenda. They were structuring Social Security and Unemployment Compensation and reforming the Stock Market with the Securities and Exchange Commission and banking and finance with the FDIC and Glass-Steagall; they were reforming electricity with public utility regulation and public power projects; they were restructuring agriculture with the Agricultural Adjustment Act — the second version of which is surely one of the most successful instances of industrial policy ever implemented; commercial aviation was restructured with the FAA and CAB and commercial broadcasting with the FCC.

Even Krugman’s smug assertion that the American experience — of WWII spending ending the Great Depression — vindicated Keynes, leaves many aspects of that experience unexamined. The spending was not simply running a government deficit — it involved comprehensive price controls, forced savings and a sweeping, radical change in the distribution of income (sometimes referred to by economists as the Great Compression), and though much of the spending was on explosives, which made burying pound notes in an abandoned mine look sensible by comparison, the U.S. directed enormous private and public investment into plant and infrastructure — this is in contrast to the U.K., which consumed its capital base during the war, making the economic experience of the late 1940s and early 1950s in Britain compare much less favorably to the “depression” of the late 1930s. If that experience had been the lesson, Keynes would be forgotten. So, it is “complicated” after all.

Friedman’s ideology was tailor-made for an agenda of dismantling the institutions of the New Deal — he simply denied that those institutions had anything to do with how well the economy functioned in the 1960s. The heavily disguised authoritarianism of conservative libertarianism became the favorite interlocutor of the know-nothing neoliberalism of a college-educated elite, who were tired of class warfare and hanging out with trade unionists, racists and machine politicians. But, I digress.

Theory has gone out of fashion and lots of young economists are trying to make a career out of “empirical” work, including field or case studies, and this will tend to contribute, I suspect, to an unintended revival of institutionalist perspectives, as will general dissatisfaction among the public with the outcomes of the ill-designed institutional reforms of the neoliberals. The revival of institutionalism has been going on for a long-time. They gave Oliver Williamson’s long and lonely quest to keep the English language in use in economics and to make something of the 1962 seminar he took with Herbert A Simon and Kenneth Arrow the honor of a faux Nobel, so there’s some prestige potential. The Public Choice and law and economics folks — the other half of the Chicago school — have dedicated themselves to reviving institutionalism, so there’s support on the intellectual right. The maturing of game theory and behavioral approaches means there’s a rich “toolkit”, as the academics would say, for assembling an institutional interpretation of the economy (dare I say, hermeneutics? no, I can’t do it — it’s too horrifying). I live in hope that some bright young MMTer will stumble onto Martin Shubik’s three volume magnum opus on a dusty library shelf, and the revolution will be on.

101

Consumatopia 10.27.13 at 10:51 pm

Godel proved the incompleteness any system with a finite, predetermined set of axioms.

His work doesn’t stop mathematicians from investigating the consequences of new axioms, just like a scientist would investigate new hypotheses. It’s not that math is impossible, it’s that it’s always unfinished, always incomplete.

102

roger gathman 10.27.13 at 11:06 pm

Bruce, you make good points. I think that there are further points to be made regarding the institutionalists and the progressives in the first decade of the 20th century. The Great Depression was by no means the first great depression in American history. It was the institutionalists, in alliance with progressives (when that term was not a watered down label for liberals) who put into place the fed, the department of commerce, the first laws regulating interstate commerce, labor law reform, etc. The Keynesians have a bad tendency to think that everything began in the 1930s, because, well, that is when Keynes wrote his General Theory. But depressions happened well before that time. The architecture of the New Deal – everything that you mention – was backgrounded in the experience of 20 years before. Unfortunately, that architecture is in bad need of an overhaul, and instead of having a strong institutionalist base to call upon, we have well, mainstream economics, which is manifestly terrible at understanding regulation, institutions and enterprises. No surprise, then, that we keep coming up with gallimaufry solutions like ACA.

103

roger gathman 10.27.13 at 11:11 pm

ps – I’d cite as one of those failures the way in which economists, like Greenspan, literally did not understand connectivity when devising the regulatory regime for financial instruments like CMOs. To say things like, we don’t need to regulate derivatives because the trading is done by people who can afford the losses – millionaires and such – shows such amazing ignorance about embedded effects in the economy that you have to just gape.

104

Bruce Wilder 10.27.13 at 11:14 pm

Axioms — however much they may look to our prejudiced intuition like nearly self-evident propositions — function as a way to bound the problem. That may contribute to enabling our limited powers of reason to solve the problem, but it leaves solution as well as the problem bounded.

The classic example is Euclid’s parallel lines. Lots of problems are easier to solve, if we assume parallel lines can extend indefinitely across a plane; it makes the plane, flat, in a simple way, which is very convenient, but also limiting. If you are Einstein imagining yourself riding on the cowcatcher of a train travelling at the speed of light, the absurdities imposed by axiomatic parallel lines become obvious, as you try to imagine shining a flashlight in the direction the train is travelling: is the light of the flashlight travelling faster than the speed of light (the speed the train is travelling)?

Will there ever be one DSGE model to rule them all? I feel comfortable saying, “no”. That there are economists, who apparently can not work out the severe limits of this modeling approach, does give pause.

105

William Timberman 10.27.13 at 11:20 pm

bob mcmanus @ 91

Well, I glanced at Gathman’s blog, and withheld comment because I presumed he understood that a focus on a humanistic individualism is not at all unique to neoclassical economics, but is an important part of the Western Enlightenment project. “Rights” inhere to individuals not groups, states, classes, families, so we must presume that rationalities, agency, etc also are a capability of the individual.

It seems to me — and not just to me, I think — that both Marx and Keynes, as well as non-economists like Weber, Nietzsche or Freud, are inseparable components of your Western Enlightenment project. Specifically, their work seem to represent successive attempts to refine the relationship of the rational to the irrational, the self-reflexive individual to his determining collective, and, if you like, to define justice as a feedback loop between them.

Viewed from that perspective, the Western Enlightenment project can be understood not just as a parochial artifact of single historical period, or the arc of a single culture’s development, but the project of modernity as a whole: to escape the dead hand of custom, institution, and class while at the same time acknowledging both our debt to all of them, and their legitimate claims on our allegiance. Libertarians and atavists of all sorts would vehemently reject any such appreciation, but what they have to offer in its place seems, at this juncture, to be pretty thin gruel.

106

John Quiggin 10.27.13 at 11:44 pm

Godel is indeed about incompleteness – he showed that, if you want a complete system of math, in which every true theorem can be deduced from the axioms, you’re out of luck. Of course, this doesn’t stop you deriving theorems or doing all the stuff that mathematicians have always done.

Similarly, SMD show that if you want to derive a model of the entire economy, including aggregate demand and supply functions , from standard micro foundations, you’re out of luck unless you’re willing to make some unappealing simplifications, such as a representative agent. But no one really wants to do this, except maybe the most ambitious exponents of DSGE.

(New) classical economists don’t have any need for aggregate demand and supply – they rely on the welfare theorems I mentioned in the OP. Old Keynesians (like me) think that “General Equilibrium” is a special case, and that the real interest is in unemployment equilibria, hence Keynes’ title.

107

ZM 10.28.13 at 12:00 am

@105 I’m reading an interesting older essay by Jameson that actually, so far, locates Marx re: communism in the romance genre. I haven’t finished it yet though, so am unsure where it will go.

108

Alex K. 10.28.13 at 12:23 am

“Godel is indeed about incompleteness – he showed that, if you want a complete system of math, in which every true theorem can be deduced from the axioms, you’re out of luck. Of course, this doesn’t stop you deriving theorems or doing all the stuff that mathematicians have always done.

Similarly, SMD show that if you want to derive a model of the entire economy, including aggregate demand and supply functions , from standard micro foundations, you’re out of luck unless you’re willing to make some unappealing simplifications, such as a representative agent.”

This is all perfectly fine, except for the word “similarly.”

If, as some here have proposed, you want to get rid of Friedman’s “assumptions do not matter” then your macroeconomic models need to have some microfoundations — not necessarily microfoundations in General Equilibrium, but microfoundations nevertheless. If SMD (and a plausible equivalent to SMD, for economic actors exhibiting a variety of modes of irrationality; with some negative aggregation results added) precludes any form of non ad-hoc microfoundations, then you’re back to “assumptions do not matter.”

“Assumptions do not matter” might be a plausible approach if macroeconomists could exhibit some spotless predictive empirical record, in which case we would throw up our hands and claim that some macroeconomic model “just works” so it must represent some emergent property of economies.

Alas, macroeconomic prediction does not have a spotless record.

109

Rakesh Bhandari 10.28.13 at 12:26 am

What are unemployment *equilibria*? I understand the political point of such a phrase–it brings out how many shoves may be needed to push the economy to a higher level of employment when it is really stuck and small confidence boosters won’t do the trick.

But a situation of unemployment equilibrium would also probably be characterized by excess capacity and the fratricidal competition to which that would give rise, possible debt deflation cycles and likely neo-mercantalist politics and international tensions. In what ways would it accurate to describe such a situation as an equilibrium?

110

Peter T 10.28.13 at 1:04 am

Sometimes it seems to me that a large part of the problem is not with being “scientific” (whatever that means), but with trying to combine consideration of urgent practical problems with an approach grounded in high theory. There is no fundamental problem with seeking a unifying theoretical understanding of things. Perhaps if persisted in long enough it will yield results (it did for alchemy, and also for medicine, although in both cases it took a few centuries). It will certainly give rise to the odd clever insight. As an academic pursuit it can do no real harm.

BUT, if the subject is messy, complex, dynamic, uncertain and so on in all the ways actual economies or societies are, it is more practical to recognise that our current theories are hopelessly over-simple. In this case, there is no substitute for close examination of the actual case – what I take to be Bruce Wilder’s institutional approach, grounded in the details of the actual economy. In other words, to approach the economy as an artisan does his or her materials. With real knots, lumps, and impurities, unreliable suppliers and limits to workability.

The 30s and World War II are a case in point. Paul Krugman’s comment shows, as Bruce says, an astonishing ability to wave over the details. It was not just that the government boosted aggregate demand, it was that the government identified and invested in very specific sectors in response to very specific identified problems. Sure it was noisy and sometimes missed the mark – it was very untheoretical in ways that are anathema to the rigorous mind. It was heavily political, and so subject to all sorts of compromises. It worked.

What we need is a clear understanding that, while theoretical economics is fascinating, it has very limited applicability to our actual problems.

111

Ronan(rf) 10.28.13 at 2:15 am

Thanks Bruce.
I had that feeling, I seem to remember you implying as much before. Where would you start (aside from an old dusty three volume magnum opus !) to understand the basic’s of institutional economics? Galbraith?(bear in mind I know nought about economics)

Relatedly, have you followed and what do you think of complexity economics? It appears faddish, (like all theories), though interesting to me, but whenever Ive asked anyone who would know about it they say it has no practical use (now, or never, Im not sure)

112

ZM 10.28.13 at 2:40 am

@100 “Friedman’s ideology was tailor-made for an agenda of dismantling the institutions of the New Deal — he simply denied that those institutions had anything to do with how well the economy functioned in the 1960s. The heavily disguised authoritarianism of conservative libertarianism became the favorite interlocutor of the know-nothing neoliberalism of a college-educated elite, who were tired of class warfare and hanging out with trade unionists, racists and machine politicians. But, I digress.”

I’m not old enough to have memories of the 1960s myself, and I dare say they were different in Australia in many ways from the US, but the 1960s socio-economy, particularly as it went on seems to have been characterised by a lot of extreme social divisions and violence – Indochina overt and covert warfare, (using the bodies of young men), various protests against Vietnam, civil rights struggles, assassinations etc etc

I think American Pastoral (the only Roth I’ve read) posited that the off-shoring of labour at the time by business owners was directly responsive to the civil rights movement. The two bodily (rather than rhetorical) options taken by the daughter figure – bombing, total non-violence – could possibly be taken to point in some way to the subsequent appeal of pop po-mo neo-liberalism….

113

Ronan(rf) 10.28.13 at 2:57 am

..as an addendum to 111, I hear some people complain when ‘complexity’ becomes reduced to ‘shit happens’, and so want to create an empirically grounded, mathematically sophisticated, multidisciplinary school of social science.. I’ll adopt my mothers fatalism (and my own laziness) and argue this is obviously doomed to failure, although credentially and intellectually I cant argue it (who would listen?)

114

Lee A. Arnold 10.28.13 at 2:59 am

I think maybe each science has a different relation to the general definition of science, and they are located at different points in a continuum from exact historical description to precise prediction.

Economics might never be a precisely predictive science. There would be at least two different real reasons for that: 1) chaotic outcomes from n-body computation, and 2) new emergence of different (and often higher) logical types of ideas, innovations, and institutions.

1) N-body computation results are understood from early in science.

2) The second reason poses problems in mathematical prediction in another way, because the innovation or institution is a rule set that is logically incommensurate to the transformations or transactions it rules. They are different logical types: market transactions are about quantities of money; rules are about transparency and responsibility to common expectations, willingness and outcomes. They cannot act on each other, i.e. arithmetically transform each other, in the same equation.

So then, what are the options for formulating a scientific fundamental about institutions, to try to match this fundamental to reality? One idea is to think of it as a mechanical algorithm, and perform both the transactions and the rules by doing so in tandem reiteration. But then, we see something like a game theory tree, (after all, a game is an institution too,) but it quickly branches out to an infinite continuum of choices, and so we have to limit it for the real finite world, by inventing something like the concept of “bounded rationality”.

Instead, I think that a much simpler solution is to go at the problem from both sides at once. Start with the fact that there are two, co-equal, forms of economic growth. One form maps over the transactions, and the other form maps over the rule set: A) growth by specialization and trade, and B) growth by the reduction of the cost of making those trades, by the institution that the trades are under (or, the reduction of the transformation costs by the institution that the transformations are under, e.g., the work flow at a business firm).

Reducing the trade costs serves to increase the volume of trades (or transformations) between the specializations. This looks like an enormous source of growth. (Coase cited a paper by other economists which found that transaction costs, despite our efforts to ease them, still equal around half the GDP.) Why is reduction of trade costs co-equal to specialization and trade? Because this institutional cost-saving, at one level of transactions/ transformations, becomes the comparative advantage of a specialization that may be traded at the next level up. It is a combined effect. There is a hierarchy, with a tandem alternation of equalizing and maximizing at each level, and it goes in both directions along this following nesting: [ individual ]-[ firm ]-[ agglomeration ]-[ government ].

It can easily be understood when pictured in one model, though that model is not mathematics, it is para-mathematics or exomathematics:
http://www.youtube.com/watch?v=BXr5VjQ0OXE&list=PLT-vY3f9uw3AcZVEOpeL89YNb9kYdhz3p

Surely this cannot be predictive, but the visual is a more efficient description, which is a hallmark of science. Ernst Mach comes in around here somewhere… Also it might help to cure the rhetoric of market primacy, which is to think of markets as the main thing, and institutions as impingements upon it. This is not only wrong, it has finally become intellectually inhibitory as it is poisoning everyone.

115

mattski 10.28.13 at 3:12 am

ZM @ 94

What do you mean by an approximate science exactly?

Well first and foremost it is a study of aggregates, it looks at the activity of large numbers of people and tries to discern patterns as well as causal relationships. But it is difficult indeed to make statements about huge numbers of actors without making approximations. To me that seems uncontroversial.

But step back a minute. ALL science is approximate. It is just that some is more so than others. Newtonian physics is an approximation of quantum mechanics. What is quantum mechanics an approximation of? That remains to be seen. If you believe–as I do–that ultimate reality is unknowable except through direct experience unmediated by symbolic representation, then clearly all knowledge is approximate. (Isn’t it fairly obvious that language is metaphorical? Doesn’t that imply that any proposition is an approximation of some other phenomena, whether direct experience or some aggregation of observations?)

Economics suffers from the fact that it straddles the world of facts and the world of human desire. As such it will never be a hard science. But personally, I am persuaded by people like Krugman that it is possible to come to practical conclusions like, “in a depressed economy the government should pick up the slack through deficit spending.” And I distinguish the utility of Krugmans advice from a good bit of the utopian histrionics that one finds in these comment sections.

116

mattski 10.28.13 at 3:37 am

all knowledge is approximate

Should have read, all communicable knowledge is approximate.

117

ZM 10.28.13 at 3:42 am

@115 “clearly all knowledge is approximate” I agree with this part of your statement, more or less. I simply disagree with your implicit epistemological claim that all knowledge is scientific knowledge, and all objects of knowledge are best understood by a scientific subjectivity.
Krugman’s conclusion that you give seems to me to assume that national economies are not interrelated to a very significant extent, are not related to geo-political matters, and that the material resources of the planet are not finite.
Re: Utopianism, the Jameson essay I mentioned earlier says this, which I find interesting (as I’ve said I haven’t finished the essay)
“Romance now again seems to offer the possibility of sensing other historical rhythms, and of demonic or utopian transformations of a real now unshakeably set in place. … The association of Marxism and romance does not so much discredit the former so much as it explains the persistence and vitality of the latter….” (The Political Unconscious “Magical Narratives” pp. 104-105)

118

Bruce Wilder 10.28.13 at 3:48 am

Rakesh Bhandari @ 73, 77 . . . 109

Equilibrium analysis is a really powerful way to think, because it helps sort out, proportionately, all the forces and counterforces involved. To derive an equilibrium analysis requires a constraint as a starting point, some fixed limit that must be satisfied. Everything is balanced to satisfy the constraint. Consider, as an example, the analysis of how the movement of a stream of gas around an airfoil could generate lift. Bernoulli pioneered the analysis of fluid dynamics, using conservation of energy as a constraining principle. Euler is credited with the amazing feat of deriving an analysis of a no-viscosity fluid being turned by a plane, which satisfied at once, conservation of mass, momentum and energy. Euler is credited with a lot of things, really, because he was really, really good at derivations of this sort.

The Classical economists knew that they needed some kind of constraint, some kind of fixed standard, to derive an analysis. That’s what Ricardo was trying to do with the Labor Theory of Value, find that fixed standard, that constraint, which would allow him to derive an equilibrium analysis.

The Neoclassical economists of the marginal revolution gave up on finding a constraint or conservation law, and simply assumed equilibrium obtained, and proceeded with their analyses from there. Comparative statics was born. It’s a “natural” or homeostatic equilibrium, the product, not of a conservation law, per se, but of a regulated process approaching equilibrium as an attractor or set point. For Walras, that process was tatonnement, a double-auction with an all-seeing and infinitely patient auctioneer calling out prices and quantities until equilibrium equating, and exhausting, supply and demand obtained. For the American, Irving Fisher, it was something closer to the fluid dynamics of an hydraulic system — he studied under an engineer expert in hydraulics, and focused on the constraints of a gold standard. But, the failure to find a constraint with which to do a formal derivation meant that all depended on the equilibrium obtaining, without really having a good account of exactly how or why it would obtain. The details of the all-important equilibrium had to be both obscured and fiercely defended.

Keynes, bless his black liberal heart, upended this happy vision of the economy as a system in, or tending toward, equilibrium, by stumbling onto a limit, a constraint: full employment of resources. Full employment of resources is a qualitative standard for efficiency (idle resources imply the possibility of a costless Pareto improvement) and a quantitative standard for equilibrium: unemployment of resources implies markets have failed to clear. Something about the system of money, prices and markets (or “markets”) has to be seriously out of whack, if high unemployment persists.

The whole enterprise of neoclassical economics, though, depends on equilibrium obtaining of itself, and that equilibrium must be “natural”, aka magical and obscure.

I really don’t know what to say beyond that. There’s a history of how the neoclassical synthesis, building upon J.R. Hicks’ kludgey interpretation of Keynes, became the doctrine of the American Keynesians. Ultimately, though, I think I gave the correct answer @ 28: “Market economy” is false. Never mind asking what the alleged “equilibrium” looks like, where’s any kind of definition of the “market” metaphor, applied so loosely and universally in the assertion of a market economy? I look around and see very few “markets” within spitting distance of the kind of exhaustive auction Walras imagined. That’s not a criticism of Walras, per se; I imagine he thought he was working out some concepts, not describing in detail, reality. What rational people do with models like Walrasian general equilibrium or Arrow-Debreu is use them to notice all the many, many ways the actual, institutionalized world doesn’t look at all like them; that sort of learning from contrast can be really useful and productive of insight.

We do live in a political economy of distributed decision-making, but money prices, formed in anything like a market, are not our sole, or even dominant coordinating mechanism, nor, at any particular point in time, are all the “markets” working demonstrably well, clearing or tending toward equilibrium. As Oliver Williamson’s book title put it: markets and hierarchies. And, the hierarchies predominate, in case you haven’t noticed. Most prices are not arrived at by tatonnement; they are administered by bureaucracies. And, the prices administered, are administered, precisely because increasing returns to scale and sunk cost investments mean that, over the relevant range of output, marginal cost is less than average cost and decreasing — conditions that preclude a market-clearing equilibrium in price. By a strict neoclassical standard, there can be no market equilibrium; its existence is excluded. I guess those can be termed, “sticky prices”, but the welfare implications are a good deal more nuanced that you’ll find any appreciation of, in Woodford.

In relation to the OP’s insight, it seems quite plausible to me that the Great Depression was prolonged, precisely because the high rate of unemployment put general downward pressure on wages, at a time, when technological and productivity trends would seem to imply that the only feasible full-employment equilibria would be at higher wage rates. Wages had to rise, despite the deflationary pressure, for the economy as a whole to be able to reach a full-employment equilibrium.

In addition, some important markets were malfunctioning very badly, and had to be institutionally restructured, before their price formation, so to speak, could be relied upon. Electric utilities were one such case, and the case that contributed significantly to the stock market crash. Agriculture was a huge problem area, where a large fraction of the population was sinking hopelessly into poverty; technological and productivity trends were very positive in agriculture, but required the sector to shed resources, while making substantial sunk cost investments — commodity prices were highly unstable — there was no discernible equilibrium — and under great pressure from the elaborating food processing industries. And, the rapid advance of industrial areas of the U.S., relative to Europe, in the aftermath of WWI and under the new financial dominance of the U.S. and the constraints of the reparations regime, created further barriers to stable, general equilibrium in the global markets, financial and trade alike.

In the GFC of 2008, the administration of the systems for making mortgage loans, and securitizing those loans, broke down. I’m not sure what the point is, of endless blathering about these problems, as if there was a housing market, except to obscure the truth. Honoring such efforts to obscure, I suppose, is why a central bank would give a faux Nobel to Fama and Shiller, but we’ve already discussed that.

119

ZM 10.28.13 at 4:05 am

Mattski @116 “Should have read, all communicable knowledge is approximate.” Sorry, but this clarification makes me want to ask in Where/What/Whom is the locus of non-communicable knowledge which is not approximate, but, presumably, true?

120

ZM 10.28.13 at 4:07 am

But perhaps you meant being rather than knowing?

121

Robert 10.28.13 at 4:08 am

It has been a while since I’ve read this:

Geoffrey M. Hodgson, Economics and Institutions: A Manifesto for a Modern Institutional Economics (University of Pennsylvania Press, 1988).

122

Bruce Wilder 10.28.13 at 4:23 am

Peter T @ 110

Speaking for myself, I would not consider institutional economics to be atheoretical, just because it focuses on concrete and particular situations and solutions. An architect with no understanding of structural engineering (or aided by a structural engineer with no physics), would be a very dangerous decorator.

The actual economy has to be structured to deal with the limits to our knowledge and understanding, with what we don’t know, including what we don’t know that we don’t know. Axiomatic, deductive reasoning is not capable of comprehensively modeling genuine uncertainty, even though it may demonstrate the logical necessity of dealing with uncertainty and risk. That’s the leap that experience and skill must make: the structural engineer analyzes the loads and stresses and their distribution in line with the dictates of his theory, then he buys some insurance, by making things thicker here and there, introducing some redundancy, etc.

Arrow-Debreu demonstrates general equilibrium under uncertainty requires perfect insurance, such that no actor errs due to risk aversion: every actor bets the mathematical expectation and realizes the mathematical expectation, regardless of what actually happens. That’s far from practically possible, but it would seem to indicate a high value to pervasive provision of social insurance, to contain desperate or risk-averse behavior. Why the economists touting the theoretical goodness of general equilibrium reasoning are not all socialists I could not tell you.

123

john c. halasz 10.28.13 at 4:42 am

@122:

Actually Walras was some sort of socialist and K. Arrow was certainly in his personal “normative” views a very well-meaning social democrat.

124

William Timberman 10.28.13 at 5:03 am

It seems to me that the economist-touts of the theoretical goodness of GE reasoning are not all socialists because they’re agnostic about outcomes. For economists, what we’re aiming at as a civilization, what concept of the good we should endorse, has always been, and perhaps will always be, someone else’s business. (With the usual few notable exceptions, of course.)

Then again, look what happens when everybody decides to get into the act. Economists — and not only economists — are suddenly without identifiable patrons. Panic city. It would be nice, I suppose, if we all recognized the limits on human behavior, understood where those limits placed us and everyone else in the scheme of things, and had the wherewithal to abide by that understanding. In the real world, though, creating the one DSGE model to rule them all might actually prove to be an easier task.

125

John Quiggin 10.28.13 at 5:08 am

Not just in his personal view. Arrow’s 1963 paper “”Uncertainty and the Welfare Economics of Medical Care” is really the beginning of the modern economic analysis of social insurance and comes straight out of his GE work.

For Arrow at least, the whole point of GE was a modus tollens exercise. For the welfare theorems to hold we would need lots and lots of state-contingent markets. We don’t have them, so we need intervention.

126

John Quiggin 10.28.13 at 5:09 am

The actual economy has to be structured to deal with the limits to our knowledge and understanding, with what we don’t know, including what we don’t know that we don’t know. Axiomatic, deductive reasoning is not capable of comprehensively modeling genuine uncertainty, even though it may demonstrate the logical necessity of dealing with uncertainty and risk. That’s the leap that experience and skill must make: the structural engineer analyzes the loads and stresses and their distribution in line with the dictates of his theory, then he buys some insurance, by making things thicker here and there, introducing some redundancy, etc.

This, pretty much, describes my current research program. I’ll try to write a post on it soon.

127

ZM 10.28.13 at 5:19 am

@122 “That’s the leap that experience and skill must make: the structural engineer analyzes the loads and stresses and their distribution in line with the dictates of his theory, then he buys some insurance, by making things thicker here and there, introducing some redundancy, etc.”

The fundamental difference here is obviously in the inertia of the materials the structural engineer uses to make his or her structure. In terms of cultural structures, they both make and are made in the practices of living persons. Imagine Pygmalion’s bridge…

128

Chris Mealy 10.28.13 at 5:33 am

Bruce, I got curious about Martin Shubik and found this gem, A Curmudgeon’s Guide to Microeconomics. From the conclusion:

Unfortunately, microeconomics is probably not an elegant subject when really well done. In order to keep us going, every now and then it is a good idea to set almost all parameters equal to one or zero, keep everything convex, and make a few further simplifications here and there, and for this price we can be rewarded with some nice convergence theorem. The Invisible Hand appears, pats us on the head, and we feel elegant. Beyond that it is necessary to go back to the miasmal swamp of reality.

129

Bruce Wilder 10.28.13 at 8:07 am

Ronan(rf): Where would you start?

For an economics textbook, Samuel Bowles’ Microeconomics: Behavior, Institutions, and Evolution is good — definitely a textbook, though. In his last major efforts, John Kenneth Galbraith cribbed a lot, from the then young Bowles and Gintis, radicals hired at UMass-Amherst in the early 1970s.

Mary Douglas’ How Institutions Think, which grew out of lectures in reply to Mancur Olson’s Logic of Collective Action is stimulating.

The past master of the old institutionalism was Thorstein Veblen. The Theory of the Leisure Class is a classic.

The best institutionalists are anthropologists — read an anthropologist’s analysis of any economic policy, process or situation, if you come across such an essay.

130

Ed 10.28.13 at 8:25 am

Andre Cieplinski @96: Thanks. Just to be clear, I didn’t mean primarily for me, though I admit of not having read everything on your list. Rather, I have the impression that the biggest problem with economics is what is included and what is included in economics 101, or the courses taken by almost everyone who then continues with a career in business, politics or the right wing commentariat. How much practical impact does it have if there are this or that heterodox research on higher up levels that only reaches a tiny number of those taking economics? I think the best remedy for this would need to be systematically constructive: construction of textbooks that practically can replace Mankiw etc at the introductory level. Shouldn’t a big chunk of the discussion of the problem in economics be aimed at forwarding that goal? But that doesn’t look to be what is happening.

131

John Quiggin 10.28.13 at 9:21 am

“The best institutionalists are anthropologists” OTOH, there’s optimal foraging theory

https://en.wikipedia.org/wiki/Optimal_foraging_theory

132

Luis Enrique 10.28.13 at 9:39 am

John Q,

would you be kind enough to school me on SMD?

My current understanding is that those results say you cannot start from general micro foundations and be sure you’ll end up with nice downward sloping market demand curves you can use to do standard macro. On the other hand, you may well end up with downward sloping demand curves (it’s not ruled out). So if you think – for some reason other than having started from individuals and aggregating – it makes sense that market demand curves slope down, there’s nothing to stop you using that in macro models, and it’s not necessarily inconsistent with microfoundations.

So SMD leaves some wiggle room. Saying you cannot build up in general, or under weak assumptions, and derive the result of downward sloping demand curves, is not the same thing as saying nobody should be doing macro with downward sloping market demand curves and imagine what they are doing is consistent with microfoundations.

But as fear my current understanding is wrong – if so, I’d like to find out sooner rather than later.

Also – and now I am thankful for internet anonymity because I might really be about to make a fool of myself – the way I currently understand SMD is to think that if a price of a commodity rises, in GE people who aren’t endowed with it might buy less but those who are will be richer and being richer might demand more the commodity themselves, so you cannot say market demand will fall at higher prices.

If that’s right, I’d be interested in your thoughts on what market demands for what commodities might not be downward sloping in reality, and how that would matter. My guess is: labour.

oh, I don’t know whether you saw the recent Econ Journal Watch, but I thought this quote from Arrow was interesting. He thought of “general equilibrium as an ideal to be achieved by a mixture of private enterprise and public planning and regulation;”

133

In the sky 10.28.13 at 1:09 pm

In this thread there are at least two types of comments.

The first type are the ramblings of people who really have very little idea what they’re talking about, but really want to be read anyway. The sort of comment that dislikes economics for not talking about how multinational corporations act in malevolent ways in their unending pursuit of profit, but also for assuming that agents act rationally/firms maximize profits. Or dislikes economists for being neo-liberal/conservative, even though the voting behavior of professional economists are noticeably to the left of the average American.

The second type are well-informed, but miss the mark for one reason or another. For example, people roll their eyes at Arrow and Debreu and consider their famous theorem to be some ridiculous neo-liberal ploy. That’s nonsense. I fully “believe” in Arrow-Debreu. The result, very basically, is that a set of prices *exist* that will result in markets clearing. There is nothing about equilibrium necessarily holding. And Arrow and Debreu are not the type of Bush-era conservatives (that commentators here like to portray economists as) who think that equilibrium will pertain; rather, Arrow is well-known for his views on intervention and Debreu is the same Debreu of the aforementioned SMD theorem.

Allow me to take an example of what I consider to be well-informed, but somewhat misleading commentary:

@Bruce Wilder

And, the prices administered, are administered, precisely because increasing returns to scale and sunk cost investments mean that, over the relevant range of output, marginal cost is less than average cost and decreasing — conditions that preclude a market-clearing equilibrium in price. By a strict neoclassical standard, there can be no market equilibrium; its existence is excluded.

In the strictest, bare-bones neoclassical model Bruce has a point. And similarly, as I noted above, the strictest bare-bone model does not have scope for “nudges” to influence policy. This does not imply that economists do not believe in “nudges”; it implies that bare-bones models have to be adapted to what you’re investigating. Restricting the availability of credit to firms will provide equilibrium existence in Bruce’s example.

It’s all leading to a scenario where people like Ed, who have never taken Econ 101 or read Mankiw, are looking for alternatives that are “more correct”. Where the academic consensus that “opportunity cost” is a useful thing to teach kids in 101 is shunned. Because John’s views are way off in the tail of most the distribution (but he’s a little famous), and Bruce is a non-economist with a bit of an institutionalist axe to grind. Madness.

134

mattski 10.28.13 at 1:34 pm

ZM @ 117

Krugman’s conclusion that you give seems to me to assume that national economies are not interrelated to a very significant extent, are not related to geo-political matters, and that the material resources of the planet are not finite.

I don’t see this at all. Of course national economies are related. Of course resources are limited. (!) But Krugman’s analysis is disciplined by political realities. He is, in effect, trying to answer the question, where should we place our next footfall? Minds of less discipline have a tendency to squabble about where we should be walking 10,000 paces hence.

in Where/What/Whom is the locus of non-communicable knowledge which is not approximate, but, presumably, true?

In the mind, of course! (“Where is the mind” is not necessarily a meaningful question.)

135

mattski 10.28.13 at 1:52 pm

@ 133

The sort of comment that dislikes economics for not talking about how multinational corporations act in malevolent ways in their unending pursuit of profit…

It’s interesting the way “we” (left-of-center types?) like science when it debunks primitive, hierarchical ways of thinking such as Judeo-Christian explanations of our world and social structure. But if peoples commitment to scientific neutrality gets in the way of our preferred story of ‘the way things ought to be’ then all of a sudden we don’t have such a boner for science after all.

Compare and contrast:

The manner in which a creationist picks at the weak spots in Darwin’s theory, declares Evolution untenable, and then produces a Bible from his vest pocket and says, “and by the way, I just happen to have a book here which explains what Darwin cannot.”

The manner in which some lefty sorts pick at imperfections in economic modeling and then implicitly reach into their vest pocket for a lovely Marx-scented story about the way the world should be.

136

William Timberman 10.28.13 at 2:19 pm

133, 135

We’re stuck with the division of labor, no doubt about it, but I would argue that as a result economists, for all their pretensions to scientific rigor, are subject to the same limited horizons as the rest of us are, and in general, aren’t a whit more humble about it. What’s more, they’re equally adept at crossing over into the normative, no matter how heavily they tread on other areas of expertise in the process. So what us ignorant leftists with the lovely Marx-scented stories are up to should be considered self-defense, not vandalism — if, that is, if economists treasure objectivity as much as you guys say that they do.

It’s not about the value of science, or of objectivity, still less is it about if you’re so smart, how come you ain’t rich. It’s about economists who appear to be asking for a deference that recent world events suggest they haven’t earned. Messy interventions by outsiders like those you so dislike in this thread may be annoying, but they appear to be necessary, even among economists themselves, exactly as the last sentence of the OP says. Look at it this way — we’re from the rest of the world, and we’re here to help you.

137

Ronan(rf) 10.28.13 at 2:23 pm

The pushback on claims about Stalin’s expansionist tendencies have come from historians working in the Soviet archives (Geoffrey Roberts Stalins Wars, for example) Afaik a lot of the most recent work in US archives has also come to the conclusion that US intent atwere complicated
consensus (afaict) from those working in the US archives is much the same
What do you mean it leads you to ‘the fabled primary sources’

“In this thread there are at least two types of comments.”

And lets not forget the third kind, defensive junior academics who..

Look, the only point here is that if economics as a discipline was confined to its rightful place in ‘the hierarchy’ (a dusty corner where anoraks could model all sorts of phenomena to their hearts content) then no one would care one way or the other
If the council of economic advisors had a policy position akin to that of the council of dairy policy advisors, who would care?
Of course economics is a technical and complicated subject, which takes time and effort to learn properly, but its also a much more limited, much less useful subject than it’s practitioners would admit. Which is where the annoyance comes from..

138

Ronan(rf) 10.28.13 at 2:24 pm

Eh, sorry..first 6 lines are from a completely different comment..

139

Cranky Observer 10.28.13 at 2:26 pm

How does one distinguish between people who are “well-informed, but miss the mark for one reason or another” and people “who, after thorough study and/or substantial worldly experience, disagree with In the sky”?

Cranky

140

MPAVictoria 10.28.13 at 2:59 pm

“How does one distinguish between people who are “well-informed, but miss the mark for one reason or another” and people “who, after thorough study and/or substantial worldly experience, disagree with In the sky”?”

Why they are one and the same Cranky.

141

In the sky 10.28.13 at 4:44 pm

Snark, without substantive content: the liberal commentariat of CT never lets you down.

142

Robert 10.28.13 at 5:13 pm

I do not accept that “a set of prices *exist* that will result in markets clearing” is a result in the Arrow-Debreu model. It is not clear that markets are or can be defined in the model. I am aware that Debreu (1959) provides, in the style of Bourbaki, a definition of so-called “private ownership economics” and a proof of the existence of a so-called “market equilibrium”, given certain conditions. But what that has to do with actually existing capitalism or markets is unclear to me.

Quite a bit of literature on administered prices discusses the relationship between the markup over costs, retained earnings, and investment plans. I suppose the bit about retained earnings is a recognition of the lack of unlimited amounts of external finance being available and of differences between finance obtained by borrowing and diluting existing shares.

As I understand it, the famous paper by Miller and Modigliani shows that, in their model, whether firms borrow or sell shares is immaterial to the firm. And a quite a bit of literature introduces one friction or another into the model to try to explain why firms care. Can you cite a theorem showing that restricting the availability of credit will imply that an equilibrium exists under the conditions Bruce is describing, including increasing returns to scale?

I have seen many defenses of mainstream economics that rely on the existence of pluralism in advanced research. Many of these defenses acknowledge most undergraduate teaching in economics is problematic. So, if one wants to understand existing economies, rather than what economists teach students, it is not clear that one should pay any attention to mainstream introductory textbooks.

143

mattski 10.28.13 at 5:50 pm

William @ 136,

You write as though I am defending the economics profession as a whole from criticism. I thought it was clear that I’m fully on board with the sorts of criticisms of the profession coming from JQ and PK. Sorry if it wasn’t.

But I am defending the use of imperfect modeling techniques from the sort of “if it isn’t air-tight then you shouldn’t even bother” sentiments present in this thread. I am also asking that we aim for what is noblest in the scientific tradition, that is, to be patient observers open to new information and always trying to keep our personal prejudices from blinkering our eyes. And I really don’t think that cultivating that sort of approach prevents us from vigorously defending what we believe in. But it really helps if we know the difference between our knowledge and our values.

144

William Timberman 10.28.13 at 6:22 pm

Mattski, of course your 143 is easier for people of good will to endorse than your 135, but the contrast between them leaves some doubt as to whether it’s you or I who’s confusing our knowledge with our values.

I’m reminded of BW’s observation about what walking through a supermarket tells us. In mine there’s ten feet of shelf space, mid-shin to collarbone, devoted to different brands and formulations of toothpaste, all of which are within a penny or two of the same price, except for the store brand. Is price what’s clearing the market here, and, metaphorically speaking, at what price is it doing so? Or how about Detroit’s decades-long response to inferior quality control in its products? (examples provided upon request) Did they lower the price, or increase their investment in product development? No, they increased the advertising budget and complained about union wages. Should we be consoled by the fact that GM, Ford and Chrysler nearly went out of business as a result? What do U.S. foreign-aid purchasing stipulations legitimately have to do with clearing the market for F-16s? One might even be justified in asking what we’re talking about when we talk about a market for F-16s — unless it’s talking through our collective hats, of course.

I’m all for science, mattski, but when economists appeal to us on the basis of their scientific credentials, I think we’ve earned the right to a certain skepticism. God knows that the central planners of the grand Soviet experiment got plenty of perfectly justified attention on that score. What is it that makes the Friedmans of our own capitalist dreamworld immune, especially when they climb into bed with the Pinochets, and claim that they’re doing so on the basis of scientifically-derived best practices?

Irrelevant questions? Maybe so, at least insofar as they get entangled in a debate on the significance of the SMB theorem. But one person’s irrelevance is another’s pressing concern, and on the Internet, it isn’t only the masters who get to say who’s the dog.

145

roger gathman 10.28.13 at 6:29 pm

142 brings up a question that has also been pursued by Jamie Galbraith – that, in certain circumstances, price controls are a good thing.
Myself, I think the US healthcare market is long past the point at which price controls should have been imposed on, for instance, pharmaceuticals – especially as pharmaceuticals, far from being products produced and traded in competitive markets, are harnessed to very restrictive prescription laws, on the one hand, and monopoly on the other – a monopoly granted by the vast expansion of ip law.
The result is that american consumers pay more for everything in the medical market place, and we all wring our hands about this situation and seek nudges to make the private sector correct itself, ignoring of course that the private sector is a fiction.

146

William Timberman 10.28.13 at 6:34 pm

Make that SMD theorem. To many entanglements with server message block errors in my past, perhaps, or maybe its just aging neurons that are to blame.

147

Ronan(rf) 10.28.13 at 7:00 pm

In the sky @141

There’s plenty of substantive disagreement for you to argue with, (and ignore the snark), but that would mean moving beyond strawmen (The sort of comment that dislikes economics for not talking about how multinational corporations.. ) and dismissiveness (and Bruce is a non-economist with a bit of an institutionalist axe to grind. .)

148

In the sky 10.28.13 at 7:23 pm

Au contraire, Ronan, a chara, I was not being dismissive of Bruce. I placed him in the “well informed but somewhat off the mark” category, and intentionally listed him as a contrasting example to the “mostly bullshit” category. I’m not sure how you see that as being anything but disagreement (i.e., not dismissiveness).

As for the snark, suffice it to say that it causes neither tears on my pillow nor pain in my heart. Alas I do take it as an indication of the type of fellow internet traveler I’m talking to. Much like YouTube comments and debating with members of Stormfront, when unnecessary snark pops up, I find it’s a good sign that I’m essentially wasting my time. So on that note: I’m out, and apologies to Bruce if he thought my comments were dismissive.

149

mattski 10.28.13 at 8:03 pm

William, I have $5 that says my neurons are in worse shape than yours. :^)

I enjoy a challenge, but I don’t think I saw where you showed me confused about the difference between my knowledge and my desires. (I am sure I am guilty nonetheless, just not sure where.) But here’s something: Yes, I have a pet peeve, a pretty strong one, with references to Marx. And I’m no expert on Marx so bad on me. But I think bringing him up is not generally useful both because (from what I understand of him) he was thinking extremely broad-brush in addition to making wildly unjustified assumptions about human nature, AND it is a hot-button (I’m proof!) that derails discussions.

Same for lefty complaints about “capitalism.” I don’t think the word means a whole lot today. To me, the word simply refers to freedom to act in the marketplace as one wishes. Well, Aaaaaab-veeeee-ously, we can’t have complete freedom to do whatever we want economically or otherwise. We have to have laws regulating our behavior to some extent. And the really important questions are, “to what extent?” That gets into the details, and in my view renders kludgy abstractions like “capitalism” and “socialism” superfluous. Let’s talk about the specifics and leave the ideological blunderbusses to the side.

150

ZM 10.28.13 at 8:27 pm

Mattski@134
“I don’t see this at all. Of course national economies are related. Of course resources are limited. (!) But Krugman’s analysis is disciplined by political realities. He is, in effect, trying to answer the question, where should we place our next footfall? Minds of less discipline have a tendency to squabble about where we should be walking 10,000 paces hence.”
Yes, of course! I’m not clear on this, but by saying Krugman’s analysis is disciplined by (US?) political realities are you intending to imply he is a knave?
Where we should place our next footfall would depend on our understanding of where we have been, where we are, and where we determine we will be going (10,000 paces hence). And yes, I am certainly not politically disciplined as Krugman is, and yes I do not think the US should be going in the general direction it seems to be taking.

“”in Where/What/Whom is the locus of non-communicable knowledge which is not approximate, but, presumably, true?””
“In the mind, of course! (“Where is the mind” is not necessarily a meaningful question.)””
First. Evidently, there is not the mind but many many minds (unless you happen to be particular sort of Cartesian who not only equates being with the cogito, but also the trueness of one’s cognition with having cognition – but then again, assuming you don’t think you’re writing to yourself you would accede to the trueness of the existence of other minds in this case anyhow).
I think this is a meaningful if (Descartes aside) completely obvious point.
Second. Presumably if you are speaking of non-communicable knowledge you would mean knowledge with comes to the mind before being codified in language or expressed in the timbre of one’s voice, one’s facial/bodily expressions etc?
Third. What are your reasons for thinking this non-communicable knowledge is always and necessarily true? And does this hold for only your mind, or for the minds of others as well?

151

ZM 10.28.13 at 8:28 pm

Knowledge *which* comes to the mind

Sorry

152

ZM 10.28.13 at 8:35 pm

Bruce Wilder @129
“Marx said, “the country that is more developed industrially only shows, to the less developed, the image of its own future.” In large part, however, this prediction would still fall under Durkheim’s dictum that “a science of the future has no subject matter.””

153

john c. halasz 10.28.13 at 8:47 pm

@149:

The conception of agency as (cognitive) beliefs + desires is, broadly speaking, humean. The alternative conception of agency is that (cognitive) beliefs and desires interpenetrate, which would be broadly aristotelian. In the latter case, there would be qualitatively different levels of beliefs and desires and their interactions, which admit of rank ordering of lower, i.e. more deluded or narrowly egotistical, and higher, i.e. generalized and disinterested forms. (Most economists are instinctively, if unreflectively, humeans; Marx is an aristotelian, both broadly and narrowly).

In general, the fact/value (or synonymously, positive/normative and descriptive/prescriptive) distinction is equally necessary and impossible to make out. Make of that paradox what you will…

154

William Timberman 10.28.13 at 9:21 pm

mattski @ 149

…in addition to making wildly unjustified assumptions about human nature

Is the idea that human beings are rational agents — or can be reduced to rational agents for certain very important theoretical purposes — any less wildly unjustified than the idea that history is the ultimate driver of human development? In any event, I appreciate your extending a hand in my direction. Consider it taken. Peace.

155

bob mcmanus 10.28.13 at 10:10 pm

154: any less wildly unjustified than the idea that history is the ultimate driver of human development?

Trying to keep track of the negatives in there and avoiding quoting Lukacs…

Yes, when history is understood as a subjective process rather than as an external burden.

156

ZM 10.28.13 at 10:23 pm

@154, @155
“154: any less wildly unjustified than the idea that history is the ultimate driver of human development?

Trying to keep track of the negatives in there and avoiding quoting Lukacs…

Yes, when history is understood as a subjective process rather than as an external burden.”

I haven’t read Lukacs so I don’t understand that reference, and the negatives have confused me. Are you saying you agree that history (as praxis and as interpretation) is subjective rather than an (only? who could argue this?) external burden?
Also, I don’t entirely agree that Marx was strictly an historian – it is a very specifically teleological thing that he undertook….

157

William Timberman 10.28.13 at 10:45 pm

Oy. In response to public demand, let’s try it again without the negatives. 1) mattski asserts that Marx makes wildly unjustified assumptions about human nature. 2) Presumably he also believes that the assumptions underlying certain microeconomic theories are more congruent with human nature because they have scientific foundations. 3) I doubt that he’s made a better case for 2) than Marx makes for historical determinism. 4) I was making no claims whatsoever about the truth of Marx’s views about historical determinism, or about his views in general. 5) I was, however, observing that there’s no more reason to distrust Marx’s supposed views about human nature than there is to distrust those implied by current economic theories.

158

William Timberman 10.28.13 at 11:04 pm

Or, to put it another way, the world scribbles in our mind, and our mind returns the favor by giving us the wherewithal us to scribble on the world. Given the millions of us past, present and (one hopes) future, this is a helluva Gestalt to untangle. As far as I can see, no one has successfully accomplished that formidable task to date, and I doubt anyone is likely to succeed at it in the foreseeable future, except perhaps in part. It’s a chicken and egg thing….

And no disrespect intended, by the way, to those have given, are giving, or will in the future give it their best shot. If we really need heroes, they will certainly do.

159

mattski 10.28.13 at 11:14 pm

William @ 154,

Is the idea that human beings are rational agents — or can be reduced to rational agents for certain very important theoretical purposes

I wasn’t defending that assumption! I don’t think Quiggin or Krugman would either. But I feel your warm heart and I’m grateful for our discussion. Peace to you too.

ZM @ 150,

but by saying Krugman’s analysis is disciplined by (US?) political realities are you intending to imply he is a knave?

Goodness, no! I am praising him as exemplary. How do you navigate if your feet are not on the ground?

and yes I do not think the US should be going in the general direction it seems to be taking.

We agree. (So do hundreds of millions Americans, but that’s another story and not as simple.)

there is not the mind but many many minds

Hard to say. Isn’t it possible that the consciousness beings experience is a manifestation of a single awareness? So that one mind and many minds are not mutually exclusive? I am not making a claim here, only suggesting another way of looking at it.

knowledge with comes to the mind before being codified in language

Yes, in other words, ‘experience.’ The experience of eating an apple is not the same as a written description of eating an apple.

What are your reasons for thinking this non-communicable knowledge is always and necessarily true?

True and false are values ascribed to propositions. Experience is direct knowing. We only have problems with truth and falsity when we attempt to render experience into some symbolic format. And at that point we are removed from the experience we’re trying to communicate.

Still, there are levels of awareness. It is common for the quality of our attention to fluctuate from minute to minute, day to day, year to year.

160

ZM 10.29.13 at 12:36 am

@157,158 Thank you William Timberman, that makes it clearer.

@159 Thank you for your reply Mattski.
“How do you navigate if your feet are not on the ground?”
If you found yourself entangled in the woods, you might try to climb the tallest seeming tree to get your bearings before setting out perhaps?

” I am not making a claim here, only suggesting another way of looking at it.”
Keeping with the tree theme:
“II
I was of three minds,
Like a tree
In which there are three blackbirds.” Wallace Stevens

“The experience of eating an apple is not the same as a written description of eating an apple”
Disregarding the apple from the proverbial tree, eating an apple consciously combines being and knowing in the one deed (here I intend to imply that experience is not the same as [but also not entirely opposed to] knowledge – one at once eats an apple and observes the sensations of eating an apple and may wander into the meanings of eating an apple, or wander into wondering about horse racing or some such instead) – writing about eating an apple is indeed not the same as eating an apple.

“True and false are values ascribed to propositions.”
True and false are not only used in this way. I might describe a person as true or false for example.

“Experience is direct knowing.”
Surely this depends on the person (subject) who is experiencing something (plus I think you probably mean to include observing here as well, which is different again), the person’s experience of the action (if someone tells to me a thing which is false [in mistake or lie] I might directly experience it as either true or false – my experience is not necessarily a direct contact with the thing in itself), and the person’s knowledge of the experience which is temporal and generally extends from the subjective experience to reflections far from the experience itself. In sum, I’d argue that to experience a thing is not equal to knowing it.

“We only have problems with truth and falsity when we attempt to render experience into some symbolic format” Unless we are raised by wolves we are always acculturated beings. As we experience something we almost always experience it symbolically at the same time – it is much more difficult and far less common to attempt not to render something into a symbolic format.

“And at that point we are removed from the experience we’re trying to communicate.” With words almost always – spontaneous laughter, a hand clap, an involuntary cry – not so removed. But often what we are trying to communicate is not the thing itself but our thoughts on the thing – so I might ask then are we ever (whilst conscious) really removed from the experience of our thoughts…. Of course, one may not try to communicate one’s thought in writing, but may try to dissemble – and we are back to the start of your reply again – is some one an exemplary being, a fool, or a knave?

161

notsneaky 10.29.13 at 1:54 am

Anyone who does Macroeconomics, of whatever variety, uses some kind of representative agent, either implicitly or explicitly. As soon as you’ve written down a “Y” or a “C” or an “I” or a “P”, and done anything more than just describe the magnitude of these variables, you’ve assumed a representative agent of some sort. Hell, often times when these aggregate variables are actually measured and constructed from underlying micro data, there’s an implicit assumption of a representative agent. It might be a strange representative agent (despite what some folks say, a “representative” agent of some sort always exists) but she’s there. Just saying.

(One exception might be agent based models, but I haven’t seen any that are then mapped into actually measured aggregates. They may be out there)

(Oh yeah and the Austrians. Depending on who you listen to, they just might have a theory of business cycles, of macroeconomic aggregates, while simultaneously rejecting the validity of macroeconomic aggregates. Or they just don’t do macro.)

162

john c. halasz 10.29.13 at 2:53 am

@161:

Mr. Market as Gosplan super-planner, eh?

In my untutored opinion, macro-economics doesn’t have micro-foundations, and if there is a need for “foundations”, it’s at the meso- level: i.e. inter-sectoral dynamics.

163

Bruce Wilder 10.29.13 at 2:57 am

I write, “by a strict neoclassical standard . . . ” and In the Sky agrees, “In the strictest, bare-bones neoclassical model Bruce has a point . . .” OK, so far so good.

Then, boom: “Restricting the availability of credit to firms will provide equilibrium existence in Bruce’s example.” WTF?

I don’t think it’s a dismissive response; it’s more like it’s deflective, a nonsense phrase thrown out there to prevent any further engagement or dialogue; probably qualifies for one of Unlearning’s Signs. Robert @ 142 tried harder than I’m willing to, to construe that curious sentence as a sensible proposition (and as far as I can tell, still failed).

In the Sky: “Bruce is a non-economist with a bit of an institutionalist axe to grind. Madness.”

Now, that might have been dismissive. Forgive me, In the Sky, for having a point of view.

164

John Quiggin 10.29.13 at 4:54 am

@notsneaky That’s mostly right. But the central assumption in DSGE is that the representative agent is (very close to) a standard microeconomic optimizer. That can fail because actual behavior is non-optimizing, but also because of problems in the step from micro to macro, not just aggregation problems but co-ordination failures. So, the fact that the aggregate relationship isn’t what you would expect from standard micro need not be a problem

165

Bruce Wilder 10.29.13 at 8:17 am

A representative agent eliminates conflict from the narrative.

166

UnlearningEcon 10.29.13 at 12:00 pm

I don’t understand notsneaky’s characterisation of an representative agent here – it seems almost tautological. Surely, C, I etc. refer to aggregate phenomena and are not thought of as a single agent behaving in some way?

167

notsneaky 10.29.13 at 12:34 pm

@john h in 162

I don’t think that (Gosplan super planner) is directly related. Depending on which side of the bed I wake up on a given morning I might agree with the second sentence.

@john q in 164

I agree. But skipping right to a relationship between aggregates in the style of caricatured naive old-school Keynesians (who weren’t that naive) also obscures potential problems with these relationships. At the end of the day it doesn’t matter that much whether you derive a consumption function by maximizing a utility function which may or may not be meaningful, or just write down consumption function based on a intuitive understanding of how people in the economy as a whole behave.

I think it’s instructive how Krugman does it. He obviously writes down his ideas first in the form of an old Keyenesian IS/LM style model. Then sits down and translates them into DSGE form with maximizing representative agents and all. At the end of the day he gets the same damn answer/result. And if you go back to the 90’s when DSGE was still consolidating its foothold in the journals, and track some papers from their working paper version to published version you see a number of people doing exactly the same thing.

For example compare the 1997 NBER WP version
http://www.nber.org/papers/w6495.pdf?new_window=1

with the 2007 published version
http://www.ssc.wisc.edu/~mchinn/LeeChinn_JIMF.pdf

Probably some referee told them “you have to put a representative agent in there and maximize stuff!”. The empirics are identical.

There might be some value added in such an exercise in that you make sure you’re not missing anything that maximizing a utility function would highlight. But that’s generally not enough to justify the dominance of DSGE in the literature. Rather, the more plausible explanation is that it’s a simple signaling/screening mechanism,.

@UE in 166

As soon as you posit some kind of a relationship between aggregate phenomena we’re talking (implicit or explicit) representative agent. Obviously the people who do DSGE don’t think of their representative agent actually being a single individual either.

It might be equivalent in the sense that “relationships between aggregate phenomena” is just another phrase for “(some) representative agent”. What else would it be?

168

notsneaky 10.29.13 at 12:36 pm

(that should be “2006 published version”)

169

Patrick S 10.29.13 at 12:39 pm

Great discussions …..

In answer to Ed @ 5, 71, 130 I think one project to produce an alternative to ‘Economics 101’ textbook from a heterodox point of view could be ‘Economics: a new introduction’ by Hugh Stretton (Aus’ wide-ranging intellectual who originally trained in history but then made significant contributions in urban planning, before getting into economics …)
http://www.amazon.com/Economics-New-Introduction-Hugh-Stretton/dp/0745315313

I am getting into modelling of transport systems which includes a behavioural economics of its own, that of how people value and trade-off time etc in making travel and mode decisions. The field seems to be evolving in an agent-based direction after a long ‘equilibrium’ phase of its own …

170

Luis Enrique 10.29.13 at 12:55 pm

notsneaky, you couldn’t adjudicate on my questions/ramblings #132 could you? I’d like to improve on my current position.

171

notsneaky 10.29.13 at 2:34 pm

Luis,

To be perfectly honest I’m not clear on what the relationship between SMD and the aggregate demand of macro actually is. I thought I sort of knew once, then stopped thinking about it and moved on, then coming back to it I realize that I (and probably many others who invoke SMD in this context) don’t really know.

The SMD is a theorem about properties of theexcess aggregate demand function, derived from a “real” general equilibrium framework. The word “excess” in there means that it’s actually about the interaction of demand and supply, not just demand. The AD of macro is just a relationship between some index of (demand for) goods (“aggregate output”, GDP) and some index of their prices, independent of the supply side of the economy. It’s hard to say how they are related except that they both have the words “aggregate” and “demand” in them, though used in a different sense. It seems like it’s this superficial semantic similarity which causes confusion.

In SMD the “P” is a vector of relative prices of different goods, relative to some numeraire. In Macro the “P” is the price of a numeraire, determined by the units we measure money in. The SMD says there might be more than one vector of these relative p’s which clears all markets. The AD just says that if the index of prices increases, desired demand for the index of output goes down.

There might – probably is – some implication of one for the other but from where I’m sitting it’s pretty loose and unclear. I’m unable to see the mapping of the results of SMD into macro-AD.

I guess one could re-interpret the goods of the “real” GE model as commodities at different points in time with the relative prices being the real interest rates. But then we need an infinite number of goods in there. And I don’t think that’s what people are talking about when they link SMD to macro-AD.

I *think* the Gorman form representation theorems are actually the relevant ones here. I’m not sure though. Those are about being able to write total demand as a function of total wealth (as opposed to a vector of individual wealths), but prices are still a vector (rather than an index). For macro-AD we want to write total demand as a function of a price index, which enters through the interest rate. I guess that if we can write sum of individual demands as a function of total wealth, then we can separate that out and what’s left is the appropriate price index. But again, I’m not sure how exactly that fits in.

I wouldn’t mind someone enlightening me on this subject as well.

172

JW Mason 10.29.13 at 2:58 pm

Notaneaky, In the sky:

Would you agree with the following?

The operational meaning of microfoundations is a formalization in terms of constrained maximization. If your model begins with an equation of the form “X maximizes quantity Y subject to conditions Z” then your model is microfounded. If not, not.

It seems to me a lot of useless arguments if we could agree that the term microfounded, as used by economists, simply describes models with this particular structure.

173

Luis Enrique 10.29.13 at 3:03 pm

thanks notsneaky,

if you haven’t seen it, you might like:
New Developments in Aggregation Economics, Pierre André Chiapporiy, Ivar Ekelandz

I am trying to digest it. slowly.

174

JW Mason 10.29.13 at 3:05 pm

I meant “arguments could be avoided if if…”

175

Robert 10.29.13 at 3:19 pm

This is off slightly in another direction, but about the SMD results.

Consider a pure exchange economy in which the market supply and demand functions for individual goods are modeled as aggregated across individual consumers. Suppose each consumer decides on how much of their endowments they want to sell and how much of other commodities they want to buy by maximizing utility, given the prices of all commodities. Would you expect market supply functions to be such that more or a commodity is supplied when its price is higher? Would more of a commodity be demanded by consumers when its price is lower? The SMD results show that the neoclassical theory of the consumer gives no warrant for such a conclusion. Market supply and demand functions cannot be expected to be well-behaved, if you argue based on methodological atomism and individualism.

A number of economists (e.g., Alan Kirman) tried to find assumptions that would lead to better behaved market supply and demand curves. For example, suppose all consumers have identical tastes. And assume that endowments are all in the same proportions among all consumers, but the quantity of this composite commodity varies among the endowments for the consumers. Then market supply and demand curves can still be of completely arbitrary shape. The assumption that all consumers have identical tastes (in other words, effectively one agent exists in the economy) and that the proportions of commodities which they consume does not vary with income (effectively one commodity exists in the economy) may not be strictly necessary to ensure well behaved market supply and demand curves. But it does not seem possible to generalize those sufficient assumptions (of Gorman form) to get the introductory textbook story. (And, as I understand it, the situation is not improved by adding production to the model. )

This is all technical general equilibrium microeconomics. As I understand it, these results imply any dynamic behavior, as complex and chaotic as you like, is possible for macroeconomics. Basically, the Lucas program of seeking microfoundations is a failure, and no reason exists to expect macroeconomic equilibrium to obtain for long without government management.

176

Watson Ladd 10.29.13 at 3:29 pm

Robert, would you work more if you were paid less, even after the time given to change jobs or retrain? I think supply and demand curves make enough sense intuitively to accept monotonicity per variable.

177

Bruce Wilder 10.29.13 at 3:35 pm

Thanks, notsneaky, for 167 & 171

178

Jerry Vinokurov 10.29.13 at 3:41 pm

Experience is direct knowing.

This is tangential, but I would argue there’s no such thing as “direct knowing.” First of all, perception is highly mediated by experience, various priors from already learned things, and so on. As such, you never perceive anything “directly,” because there’s no such thing; everything is filtered through a complex processing stream, and by the time it arrives at the point of conscious experience, it’s no more “direct” than when I tell you the directions to my house.

179

Luis Enrique 10.29.13 at 3:42 pm

“Market supply and demand functions cannot be expected to be well-behaved …”

this is what I was trying to get at #132. What does “cannot be expected” mean? It could mean “cannot in general be shown to be well-behaved” whilst still allowing that well behaved market demand functions are perfectly possible, with how sensible it is to expect to observe them an open question. Or it could mean a much stronger “well behaved market demand functions are ruled out (or very special cases)” and should not be assumed.

one way, if I understand correctly, which I doubt, of stating the SMD result is that you can take any “badly behaved” market demand function and find individual preferences, satisfying some basic properties, that would support it. But that raises the question of how plausible those individual preferences would be. As somebody else put it: Should we worry that an increase in the price of milk will lead to an increase in demand for milk, because dairy farmers consider it a luxury?

180

ZM 10.29.13 at 3:51 pm

@176 People do this – sometimes not out of choice but need ( involuntary unemployment etc) and sometimes by free choice ( people that want to work in a particular job, perhaps in welfare, education or creative arts for example). Monotonicity is not a very good thing to accept, and it’s odd to think its acceptable intuitively unless you’ve never entered into dialogue in your life.

181

Bruce Wilder 10.29.13 at 4:20 pm

Robert @ 142: the famous paper by Miller and Modigliani shows that, in their model, whether firms borrow or sell shares is immaterial to the firm.

What it shows is that the cash flow of the firm, no matter how you might slice and dice it into various kinds of financial securities — equity claims or debt claims of various kinds — still has the same total market value. The market value of the cash flow, sliced and diced = the cash flow.

They didn’t show that it is “immaterial to the firm”, though. In a world of uncertainty and asymmetric information, etc. debt is a commitment mechanism. If management promises one group of creditors a rock-steady stream of payments no matter what the business results and cash flow turn out to be, and another group, the residual, which will vary with the success of the management, that will reveal information and could become a basis for motivating management to “avoid shirking” as the economists say. If you like feel-good stories, a company enjoying the good luck of a huge, cushy economic rent, making management fat, happy, lazy, can make financial commitments, which effectively remove the cushy, and force management to milk the situation for all it’s worth.

Returning to Quiggin’s point in the OP, macroeconomic circumstances, loosely defined, determine the appropriate leverage ratio to motivate the management to invest wisely and work hard. Not just the rate of aggregate resource utilization — the unemployment rate — or business activity, but also monetary and financial circumstances. A la Minsky, a long period of stability see companies uping their leverage ratios, which may be appropriate in the immediate circumstances to optimally motivate management to invest prudently (or to disgorge cash, which is usually the issue with established, listed companies sitting on big economic rents). A shock to the system can make the leverage ratios inappropriate, which means not only do the leverage ratios adjust, but the “wrong” leverage ratios motivate some pathologically risky (desperate gambles; embezzlement ahead of bankruptcy, perhaps) as well as risk-averse behavior (canceling positive net present value investments to pay down debt and achieve more appropriate leverage ratios for the new economic environment).

To loop back to the point I was making earlier about the predominance of hierarchy in price formation, if one were serious about general equilibrium analysis as a microfoundation (if I can misuse that term even after Josh has nudged us about how Lucas uses it), the atom of that analysis wouldn’t be a market, where price clears the market; it would be an administrative process, where price satisfies the creditors. (And, fatuous welfare and uniqueness claims would be absent.)

182

Alex K. 10.29.13 at 4:30 pm

Here is a link to a paper by Abu Turab Rizvi which explains the link between SMD and microfoundations.

183

Luis Enrique 10.29.13 at 4:38 pm

Alex K, check out Chiappori and Ivar Ekeland linked to above, contains quite a lot that’s new since Rizvi wrote that.

184

Alex K. 10.29.13 at 4:43 pm

JW Mason : “The operational meaning of microfoundations is a formalization in terms of constrained maximization. If your model begins with an equation of the form “X maximizes quantity Y subject to conditions Z” then your model is microfounded. If not, not.”

That may the actual usage of many macroeconomists, but it’s a pretty poor one. I don’t think that DSGE models with representative agents are microfounded just because they use constrained maximization.

I also don’t buy the attempts to restrict the negative results regarding the possibility for microfoundations just to results about microfoundations in General Equilibrium.

Macroeconomics can not have microfoundations in GE because imposing rationality as the only restriction on economic actors is not a sufficient condition for restricting the behavior of various aggregates. If you want to eliminate the rationality restriction, then you have even more degrees of freedom for your actors, hence even more freedom for specifying the behavior of your aggregate quantities — which is precisely what you do not need if you want to provide microfoundations for some macroeconomic model dealing in aggregated quantities.

185

ZM 10.29.13 at 4:46 pm

@181 I’m not sure if you mean you think welfare and uniqueness claims are foolish or the apparent administrator would think they were foolish?
And if uniqueness claims are foolish, what about claims to difference?
Leaving aside subjectivity and welfare, a problem therefore seems to me to be that you start with something hierarchical – hierarchy implies at the very least certain levels of difference (not exactly uniqueness, but not sameness by any measure) depending on the complexity of the hierarchy as a whole (to an Earl everyone lower in status is a churl – however, there are levels etc amongst the churlish too).

186

Trader Joe 10.29.13 at 4:54 pm

@176
For salaried employees – working more and getting paid less is a daily fact of life. On a per hour basis every incremental hour worked results in that outcome.

The two obvious reasons millions of salaried employees choose this are 1) failure to work the incremental hours results in dismissal with a massively negative utility function or 2) working the incremental hours increases the chance of a promotion with the potenital to recoup in future earnings some of the hours given now.

187

notsneaky 10.29.13 at 5:06 pm

JW Mason, I would not agree.

188

notsneaky 10.29.13 at 5:24 pm

Robert,

I’m not clear on what you mean by “market” supply and demand. If you’re talking about a market for a particular good then I think you’re wrong. First, since you’re talking an endowment economy, there’s no “supply” function (well, it’s a vertical line). You can have all kinds of funky individual demand functions (with the usual properties) and there’s absolutely nothing that stops you from adding them up and for that market the market demand function will be “well behaved” – downward sloping, monotonic all that. m(p)=x1(p)+x2(p). Then m'(p)=x1′(p)+x2′(p). Of course holding wealth constant.

The difficulties arise when we don’t hold wealth constant. If wealth effects are present you just cannot consider a simple market in isolation. Then a change in price not only moves people along their demand curves but redistributes income, shifting the individual demand curves. That’s where the result comes in that the excess aggregate demand curve can do funky things, if the shifts and the movements along interact in weird ways. But it doesn’t make sense to discuss the SMD theorem in a context of a single particular market. It’s general equilibrium.

If you’re just using the word “market” for “aggregate” (across goods and individuals) then sure, that’s what we’re talking about.

189

Bruce Wilder 10.29.13 at 5:48 pm

ZM @ 185 I was scorning the fatuous claims of latter-day Panglosses, who prefer to imagine a impersonal market god leading us smoothly by pareto-improving increments to the best of all possible worlds. I’m fine with welfare claims, which are realistic about a world of conflicting interests, in which deeply unfair bargains might be sought by powerful people.

190

Luis Enrique 10.29.13 at 8:00 pm

Not sneaky,

This may, or may not, help with “market demand” from above paper:

Technically, Sonnenschein stated two versions of the problem… In first version each individual receives some nominal income , whereas in the second they each receive a fixed endowment….
The first case corresponds to the market demand problem; what is considered in the second is the agents excess demand, demand as the difference between the agents desired consumption bundle and her initial endowment.
In both cases, aggregate demand depend on prices and initial endowments. The main question, however, is the characterization of aggregate demand as a function of prices.

191

John Quiggin 10.29.13 at 8:51 pm

@notsneaky I read Krugman somewhere saying that’s exactly what he does.

It’s reminiscent, BTW, of Dylan Thomas’ writing method as described by AJP Taylor (not an objective witness!)

192

notsneaky 10.29.13 at 10:10 pm

Alex K.,

The Rizvi paper doesn’t really address the issue at hand. That paper is really about how the research program (Dreze etc.) to turn real general equilibrium theory into a new kind of macroeconomics failed. It failed because it turned out that it was really difficult to incorporate sticky (fixed) prices and imperfect competition into the framework in a consistent way. Among other things. Rizvi focuses on why those attempts failed because of the SMD result. Yes. But as far as I understand it, those attempts failed for a whole number of reasons.

More or less, it’s like the whole “rationality” assumption. There’s one way to be rational (your preferences are complete and transitive). And a whole lotta ways to be irrational. If prices are flexible then they’re flexible, Walrasian auctioneer and all. But if they’re not flexible then there’s dozens of ways for them to be sticky. You got to come up with rationing rules etc and those are more or less ad hoc, market specific or institution dependent. Same with how you characterize imperfect competition (the tension between Bertrand and Cournot being an obvious example). So even if there was no such thing as the SMD theorem it would be very very hard to make real GE into a macroeconomic theory. But there is such a thing as the SMD theorem and it’s the *snort* that the econ GE/macro Gods have after they’ve humiliated you in other ways.

I still don’t see how the SMD result is related to the aggregate demand of standard macro, whether it be of the old school Keynesian variety or the NK DSGE variety. I think that John Q is right in that as far as macro people are concerned it’s more or less something we should note, acknowledge, then ignore and move on to other things. Like asking whether we really need something being maximized to do macro or can we built the models from aggregates on down.

The Chiappori paper linked by Luis seems more relevant but yeah, I got to work through that one slowly.

Put it another way – any kind of self respecting modern macro (not to be confused with MMT) wants to have sticky prices and imperfect competition in it. You’re not going to get that out of real general equilibrium, SMD or not. Which isn’t to say that real GE is no good. It’s just a different research question.

193

mattski 10.29.13 at 10:21 pm

Jerry @ 178

As such, you never perceive anything “directly,” because there’s no such thing;

There is no such thing, or there are impediments in the way? It seems to me helpful to distinguish between sense perceptions, the intellect and the imagination. Yes, we are deeply conditioned by past experience, nevertheless our eyes see only light, dark & color. Our ears only sound, etc. I would say that the quality of our perception, the degree to which it is obscured by intellect and imagination, is both subject to change and something we can actively develop if we choose to. Indeed, meditation is a technology fitted for this purpose.

But without getting too fussy about it, there is good reason why we consider experience to be at the root of knowledge. When you want to hire the best sea captain for your boat, you want (in Socrates phrase) “the one who knows.” Years on the ocean is going to be a prerequisite for having this kind of knowledge, as opposed to–say–learning seamanship in a classroom.

ZM, I will respond, time constraints right now.

194

ZM 10.29.13 at 10:21 pm

Bruce wilder @189, Thank you, that clarifies the statement for me.

195

Collin Street 10.29.13 at 11:47 pm

But as far as I understand it, those attempts failed for a whole number of reasons.

It failed because the economy is a system of interacting dynamic equations, because interacting dynamic equations generate chaotic behaviour, and because simplified models of chaotic systems have extremely limited analytical applicability.

Which, to be fair, is only something that’s been widely known to us since, what, the seventies? Early eighties? By “us” of course I’m explicitly excluding Chris Auld, who got cranky with people on his blog pointing it out and who declared it a known problem he’d never, ever, ever respond to.

[also, of course, chaotic modelling requires extremely detailed knowledge of starting conditions, and econometrics simply isn’t in a state to deliver the information we need.]

196

Jerry Vinokurov 10.30.13 at 1:22 pm

There is no such thing, or there are impediments in the way? It seems to me helpful to distinguish between sense perceptions, the intellect and the imagination. Yes, we are deeply conditioned by past experience, nevertheless our eyes see only light, dark & color. Our ears only sound, etc. I would say that the quality of our perception, the degree to which it is obscured by intellect and imagination, is both subject to change and something we can actively develop if we choose to. Indeed, meditation is a technology fitted for this purpose.

Sure, you can condition your perceptions. People learn to hear music, for example. But think about what it means to say that there are “impediments” in the way. In the way of what, exactly? The entire act of perception is simply (or, not so simply) a complicated information filtering process that your brain performs. In a very literal sense there is no perception until that process happens. That’s why speaking of “direct” or “immediate” or “unimpeded” perception makes no sense; perception is just process, and its outputs are not accessible outside of the process itself. Whatever meditation does, it’s not actually overcoming this problem.

But without getting too fussy about it, there is good reason why we consider experience to be at the root of knowledge. When you want to hire the best sea captain for your boat, you want (in Socrates phrase) “the one who knows.” Years on the ocean is going to be a prerequisite for having this kind of knowledge, as opposed to–say–learning seamanship in a classroom.

I don’t see how that’s relevant. Yes, of course experience is important; who says otherwise? I’m just pointing out that there’s no such thing as “direct experience” of anything. That’s a kind of naive 18th century empiricism that made a lot of sense before we had any idea about what’s going on in the brain, but it makes no sense today and should be discarded.

197

Alex K. 10.30.13 at 2:49 pm

“Rizvi focuses on why those attempts failed because of the SMD result. Yes. But as far as I understand it, those attempts failed for a whole number of reasons.”

Rizvi illustrates how SMD can be used to point out ad-hoc assumptions made by attempts at microfoundations. Since rationality does not impose significant restrictions on the aggregate excess demand functions, sooner or later the macroeconomic modeler will have to impose such ad-hoc assumptions.

Additional problems with macroeconomic models do not imply that criticisms based on SMD are not valid.

” I think that John Q is right in that as far as macro people are concerned it’s more or less something we should note, acknowledge, then ignore and move on to other things.”

Well, yes, macroeconomists are very good at ignoring theoretical problems with their models. It’s a special talent. This would not be so bothersome if macroeconomists would have some stellar predictive record. But with poor theoretical justification and with a poor predictive record, sometimes we have to ask, what exactly is it that macroeconomists are doing?

198

mattski 10.30.13 at 2:56 pm

ZM @ 160

I was of three minds,
Like a tree
In which there are three blackbirds.”

:^)

If you found yourself entangled in the woods, you might try to climb the tallest seeming tree to get your bearings before setting out perhaps?

Yes. But the problem with this metaphor, I think, is that the future is out of view no matter how “high” we think we’re climbing! So instead, let’s hone our walking skills, keep our eyes on the terrain in front of us and use our values and beliefs to set our general course. That’s my metaphor about why I like Krugman’s approach.

my experience is not necessarily a direct contact with the thing in itself

Yes. We spend a lot of our lives in our intellect. We are social creatures, we love to talk to each other, debate, disagree etc. When we’re using language most of our attention is in our intellect, not in our senses. So, our minds are using symbols to represent the world, and we are essentially creating a shadow reality to represent and help us negotiate the real world of things. (I believe this is where Plato’s parable of the cave comes from… the double-edged sword of language.)

it is much more difficult and far less common to attempt not to render something into a symbolic format.

Yes, this is our conditioning. But our “bright moments,” the moments when we feel most alive, are moments when our minds are especially clear and focused on their object. Maybe it is a sunset, a kiss, a sound (a song?) or even a thought (Einstein had many of these it seems!) But we feel most alive when our minds are focused on an object and this is a skill we can develop in ourselves, and in the process gain insight into the nature of our thoughts, which are typically repetitive and not terribly insightful.

Jerry,

That’s why speaking of “direct” or “immediate” or “unimpeded” perception makes no sense; perception is just process, and its outputs are not accessible outside of the process itself.

So, your level of alertness is even and unchanging? You never feel duller than usual or more lucid than usual? When you feel particularly alert and alive, is this the upper limit of what is possible? Or is it possible that you could improve on that? I admit I can’t really make sense of the latter part of the above statement.

I’m just pointing out that there’s no such thing as “direct experience” of anything.

I think you’re being argumentative without a clear purpose. Have you ever tried to hit a baseball with a bat? Play the violin? Anyone who has experienced moments of clarity and mastery knows full well about direct experience.

199

ZM 10.30.13 at 4:27 pm

Mattski @198
“I think, is that the future is out of view no matter how “high” we think we’re climbing! So instead, let’s hone our walking skills, keep our eyes on the terrain in front of us and use our values and beliefs to set our general course.”
The future, our values, and our beliefs are all not visible. I suppose I mean here something like we simply have to leave the ground to set our course, whether we like it or not we are always climbing those trees, sometimes even at the same time as we step towards our destination of the moment.

“When we’re using language most of our attention is in our intellect, not in our senses. So, our minds are using symbols to represent the world, and we are essentially creating a shadow reality to represent and help us negotiate the real world of things. (I believe this is where Plato’s parable of the cave comes from… the double-edged sword of language.)”
Perhaps someone more familiar with Plato could enlighten us, but in terms of the double edged sword I’ve the idea that Plato thinks the ideal form – which language and physical things both strive towards (in what order I’m unsure) – is most real and is the object to be sought. I could be way off the mark here.

“But we feel most alive when our minds are focused on an object and this is a skill we can develop in ourselves, and in the process gain insight into the nature of our thoughts, which are typically repetitive and not terribly insightful.”
My focus tends to wax and wane – and may alight on a physical object – or may delve into more intangible concerns. As you say, the object of focus may be in the physical surrounds or, may be intangible (Einstein’s thoughts as you say). May be more or less insighful. But I am hesitant to equate my focus on a thing/idea with direct access to the thing/idea because *I* am always and unavoidably the medium of my experience of something – I may forget myself for moments – but forgetting something does not mean it is not there.

200

mattski 10.30.13 at 5:12 pm

Allegory of the Cave

But I am hesitant to equate my focus on a thing/idea with direct access to the thing/idea because *I* am always and unavoidably the medium of my experience of something

Not direct access to the thing (I apologize if I said this) but direct access to experience. “The thing” is an idea. Our experience comes first, and we add layers of ideas on top of our experience.

As to whether our “self” is a “thing” mediating experience, I am persuaded that the only way to investigate the validity of this idea is to sit quietly and observe the mind. I can’t claim any personal insight here.

201

notsneaky 10.30.13 at 5:45 pm

Alex

Your statement “sooner or later the macroeconomic modeler will have to impose such ad-hoc assumptions” implies that just such ad hoc assumptions have to be imposed in order for macroeconomics to even get off the ground. In other words, whatever implications SMD may or may not have for microfoundations, these results have to be ignored … for macroeconomics to even get off the ground.

They can be ignored by positing a representative agent who’s maximizing something, or they can be ignored just by writing down ad hoc functional relationships between aggregates. But ignored be they must.

So your later statement that “Well, yes, macroeconomists are very good at ignoring theoretical problems with their models. It’s a special talent. ” is just a gratuitous swipe inconsistent with your previous claim.

Predictive power/track record is another matter entirely, independent of the SMD theorem.

202

Jerry Vinokurov 10.30.13 at 5:59 pm

So, your level of alertness is even and unchanging? You never feel duller than usual or more lucid than usual? When you feel particularly alert and alive, is this the upper limit of what is possible? Or is it possible that you could improve on that? I admit I can’t really make sense of the latter part of the above statement.

I’m not really following this. Here’s what I mean when I say “perception is just process, and its outputs are not accessible outside of the process itself.” The brain is some physical information-processing mechanism. That’s a gross simplification, but it’s not entirely inaccurate. Perception is the output of that process in the very direct sense that “to perceive” something just is to send it through the appropriate pathways that deliver the “percept” to your conscious experience. Vision is a good example of this because it’s well known that the path traveled by the information delivered to your eyes (which are just lenses plus sensors) is quite convoluted. There are all sorts of streams of processing going on before you “see” anything at all.

The notion of “direct access to experience” is basically a clever illusion played by the brain. That is, if you feel sad, you don’t feel that you feel sad, you just… feel sad. And so on. There doesn’t appear to be any intermediate stage between the feeling and your recognition of it. This isn’t true, of course; the feeling isn’t some raw thing that exists separate from you and your brain, it’s as much a product of your brain’s operation as anything else. What you’re feeling is not something distinct from that operation, it just is the operation of your brain that is what feeling is.

I think you’re being argumentative without a clear purpose. Have you ever tried to hit a baseball with a bat? Play the violin? Anyone who has experienced moments of clarity and mastery knows full well about direct experience.

I am not a baseball or violin player, but sure, I’ve experienced moments of clarity and mastery. I’m not trying to be argumentative for nothing, I’m reacting to this particular bit from you:

If you believe–as I do–that ultimate reality is unknowable except through direct experience unmediated by symbolic representation, then clearly all knowledge is approximate.

I argue that this is false. That there’s no such thing as “ultimate reality” and in any case there’s no “direct experience unmediated by symbolic representation” that would get you to it. In fact, “direct experience” in this way is a kind of nonsensical term, because there isn’t any way to “directly” experience anything at all, since experience itself is a symbolic construct built up by your brain doing what the brain does.

203

Alex K. 10.30.13 at 6:10 pm

“Your statement […] implies that just such ad hoc assumptions have to be imposed in order for macroeconomics to even get off the ground. In other words, whatever implications SMD may or may not have for microfoundations, these results have to be ignored … for macroeconomics to even get off the ground.”

Right. I’m saying that there is no law of the universe which says that macroeconomics has to get off the ground.

You can probably still think about the economy as a whole, as you would think of a complex ecological system, and figure out ways to make the system more robust ( for instance, by thinking about ways of making the financial system less coupled)

But I’m skeptical about the possibilities for precise prediction and manipulation of he economy as a whole.

204

Bruce Wilder 10.30.13 at 6:12 pm

“Adding layers” seems to me like the wrong metaphor.

Our eyes — the cones and rods on the retina — respond to light; they are sensors. Our “experience”, though, is mediated by our brains interpreting that sensory data into something like a computer’s virtual representation in a game. The color, green, is not a property of light of a certain wavelength; its a virtualization assignment made by our brains. Our brains make lots of calculations and guesses, to turn the clues afforded by a combination of binary vision, shadows, parallax, etc. into an experience of a 3d world.

When we do some science, including whatever you want to call economics, the differences include the conscious and social construction of both the “sensory” experience and the theoretical apparatus. We name things so we can discuss them, and we measure things, to make sense experience “objective”. As Hume noted, “cause and effect” in the mechanics of anything — even the physics of billiard balls — cannot be directly observed; mechanical operations and functional relationships are not in the sense data; they exist only “in our heads”, so to speak, because they are insights from theoretical analysis. We experience “cause and effect” as some things hidden inside a black box, which we can only infer by means of a theory that identifies the functional relationships among the things we can sense or measure.

I’d offer the possibility that the self is an epiphenomenon of being jarred by the failure of our theories to process the data stream adequately: the panic when trying to ride a bicycle for the first time, the astonishment at the tricks of a magician, or the frustration of talking to an economist about the economy, something he knows nothing about while claiming to know everything.

205

notsneaky 10.30.13 at 8:14 pm

@203, Alex,

With that I actually sort of agree. In my darker moments.

206

Collin Street 10.30.13 at 8:36 pm

They can be ignored by positing a representative agent who’s maximizing something, or they can be ignored just by writing down ad hoc functional relationships between aggregates. But ignored be they must.

In a non-chaotic system modelling works. An approximately-correct description [but simplified] of the system gives approximately-correct results [but with missing details]. In a chaotic system this does not work: simplified descriptions quickly give uselessly-erroneous results. You can make your models, but if they don’t do the same thing the real object does what’s the point?

And this is something that people have been pointing out since we got a proper understanding of chaos and dynamic instability in… what, the mid-eighties? And we get answers like yours.

207

Collin Street 10.30.13 at 8:37 pm

Err, dodgy blockquoting. You’re all clever, you can work it out.

208

ZM 10.30.13 at 9:05 pm

Mattski @ 199
Thanks, that makes more sense if I understand you’re not equating direct experience of a thing/idea with complete access to the thing/idea.

Alex K @203
I don’t know a lot about it, but looking at the human economy like a complex ecology (incorporating actions and materials/environment) sounds a little like the interdisciplinary discipline of Human Ecology.

From the Hunter School at CUNY:

“…introduction to the interdisciplinary study of human- environmental interactions as generally framed in 1972. The Editors’ Introductory Statement described their intention to publish research on “… the role of social, cultural, and psy- chological factors in the maintenance of ecosystems; effects of population density on health, social organization, environ- mental quality; new adaptive problems in urban environ- ments; interactions of technological and environmental change; and the genesis of maladaptations in man’s biological and cultural evolution …” (1971:1).”

209

mattski 10.30.13 at 9:20 pm

@ 202

I’m not really following this.

Nor I. Look, you say the brain is a physical mechanism. What do you mean by “physical?” Are perceptions physical? If not, then how does a physical mechanism produce them? Bruce makes the point–very true–that color is not a property of light. Color is produced in the mind. But what is the implication of this? To me this implies that it makes no sense to speak of a “physical world.” Nevertheless, we do because it is practical to do so.

And, Bruce, sure, my metaphor is far from perfect. And you’re right, cause and effect are ideas. As such they are produced by our minds *in a way* that light, sound, taste etc are not. The self, too, is an idea. The greater our alertness the further the self recedes into the background.

210

ZM 10.30.13 at 9:32 pm

Bruce Wilder @204

“The first mathematical analysis of bicycles suggested that this is also what keeps a moving bike on its wheels. But although the equations were written down in 1910, physicists always had nagging doubts about whether this was the whole story.

The most definitive analysis came exactly a century later. It involved an experimental bicycle that had all its gyroscopic effects cancelled out by a system of counter-rotating wheels. The effort of building such a strange contraption was worth it: the resulting paper was published the prestigious journal Science.

The publication plunged bicycle dynamics back into chaos. It turns out that taking into account the angles of the headset and the forks, the distribution of weight and the handlebar turn, the gyroscopic effects are not enough to keep a bike upright after all. What does? We simply don’t know. Forget mysterious dark matter and the inexplicable accelerating expansion of the universe; the bicycle represents a far more embarrassing hole in the accomplishments of physics.”

http://www.newstatesman.com/2013/07/mysteries-bicycle

211

mattski 10.30.13 at 9:37 pm

@ 204

the frustration of talking to an economist about the economy, something he knows nothing about while claiming to know everything.

I am smiling as I say this, but, SLOPPY, BRUCE, SLOPPY!!

212

notsneaky 10.30.13 at 10:10 pm

It failed because the economy is a system of interacting dynamic equations, because interacting dynamic equations generate chaotic behaviour, and because simplified models of chaotic systems have extremely limited analytical applicability.

Which, to be fair, is only something that’s been widely known to us since, what, the seventies? Early eighties? By “us” of course I’m explicitly excluding Chris Auld, who got cranky with people on his blog pointing it out and who declared it a known problem he’d never, ever, ever respond to.

Colin, in all honesty, I expect the reason why Chris told you, or whoever, that he wasn’t going to respond to “chaos theory” talk, is because most of the time it comes up the people bringing it up have no idea of what they’re talking about. Either about “chaos” in general, or how it works in economics. They’re just trying to sound “smart”.

And who says the economy is “chaotic”? It might be, but I see no reason to make that assumption or evidence to have it as a conclusion. Just because there’s a whole bunch of interacting equations doesn’t mean that you automatically get chaos.

213

William Timberman 10.30.13 at 10:31 pm

My precious self engendered even in part by an encounter with Stephen Williamson? Heaven forfend!

214

Peter T 10.30.13 at 11:16 pm

Fascinating thread. Can I come back to the point I made earlier @110. The issue is not theory per se, but the huge gap between all our formal theories and the messy reality. JQ kind of illustrated the point @ 131 – the link to optimal foraging theory is nice, and I am obscurely pleased to have acquired this snippet of knowledge, but optimal foraging theory will not tell you much about how to manage a nature reserve or a farm. Sure you can point to the way a mob of kangaroos move around and say “see – optimal foraging theory exemplified”. Then you can point again and say “see – optimal reproducing theory” and again and say “optimal predator-prey adjustment theory” and so on. And you will be right. It just won’t help you in any practical sense. And while you are arguing over which of these theories is better the roos will have moved on, or some sod will have shot them.

In the same way, arguments about microfoundations or which model are a distraction from the real problems and issues. Sometimes I have the feeling that they are a deliberate distraction, a way to avoid having political arguments, or to preempt their outcome. The strength of the institutional approach is that it directs attention to the details, which is where, in our present state of knowledge, the action has to lie.

Why is it, for instance, that the very notion of actually helping some specific sector because of its identified importance to the whole (political) economy is routinely met with cries of scorn? Whole classes of remedy are ruled out while we have the debate about which theory applies.

215

Bruce Wilder 10.30.13 at 11:17 pm

Chaos theory has become a metaphor for the poets, the way “relativity” became one, after Einstein — pretty soon people were talking about ethical relativism, which is going pretty far afield.

I expect the actual economy has some potential for chaos, but the economy is something we do, and not really something, like the weather, which simply happens to us. We can screw up, trip into social arrangements that spin out of all control — there are certainly enough examples in history of things going terribly wrong, that the metaphor, at least, can find applications, though I cannot judge the application for the maths.

The OP used this phrase, “in a competitive GE [general equilibrium] with full information, no externalities and so on” and in the countradiction of that phrase, of those assumptions, there’s plenty in conventional economics to turn our expectations for the actual economy, topsy-turvy. not sneaky said, “any kind of self respecting modern macro . . . wants to have sticky prices and imperfect competition in it” and suggested that made GE “no good” for that purpose. I’m pretty sure some practitioners think they’ve got imperfect competition and sticky prices, and their models track the actual economy reasonably well; I don’t agree, but, hey, I’m not an economist, just “well-informed.” (Oh, and sloppy.) I tend to believe becoming better informed would be a boon to economics; and maybe being a little sloppy — a little less ocd about the intertemporal optimization thing — wouldn’t be a bad thing, either.

216

Substance McGravitas 10.30.13 at 11:26 pm

Chaos theory has become a metaphor for the poets, the way “relativity” became one, after Einstein — pretty soon people were talking about ethical relativism, which is going pretty far afield.

Entertainment around that. Even further afield.

217

notsneaky 10.31.13 at 12:23 am

For the record, I actually like Optimal Foraging Theory. Also the Economic Defensibly Model. I wouldn’t go crazy with them but I consider both to be nice pieces of insight.

218

Chris Mealy 10.31.13 at 11:01 pm

Bowles’s “Microeconomics” textbook is really great. It’s not a critique like Debunking Economics or other “what’s wrong with economics” books. It’s sneaky. The problem with mainstream econ is that you start from a premise that individual self-interest leads to optimal outcomes. Sure, advanced econ is often about the exceptions to that, but the root of “the economic way of thinking” is still what you’d get if you read Milton Friedman’s “Free to Choose.” Not for a minute does Bowles let you think this way. Chapter One starts with a story:

Like the overnight train that left me in an empty field some distance from the settlement, the process of economic development has for the most part bypassed the two hundred or so families that make up the village of Palanpur. They have remained poor, even by Indian standards: less than a third of the adults are literate, and most have endured the loss of a child to malnutrition or to illnesses that are long forgotten in other parts of the world. But for the occasional wristwatch, bicycle, or irrigation pump, Palanpur appears to be a timeless backwater, untouched by India’s cutting edge software industry and booming agricultural regions. Seeking to understand why, I approached a sharecropper and his three daughters weeding a small plot. The conversation eventually turned to the fact that Palanpur farmers sow their winter crops several weeks after the date at which yields would be maximized. The farmers do not doubt that earlier planting would give them larger harvests, but no one the farmer explained, is willing to be the first to plant, as the seeds on any lone plot would be quickly eaten by birds. I asked if a large group of farmers, perhaps relatives, had ever agreed to sow earlier, all planting on the same day to minimize losses. “If we knew how to do that,” he said, looking up from his hoe at me, “we would not be poor.”

Only after he’s discussed the prisoner’s dilemma and the assurance game does Bowles introduce mutually beneficial gains from trade (he calls it the invisible hand game). By broadening the range of strategic interactions he’s able to dig into subjects that mainstream econ mostly ignores, like institutions and power. Bowles also dares to ask where preferences come from and how they change, rather than just take them as given.

Comments on this entry are closed.