Discussion on my last post on reanimated zombie ideas in economics touched on a lot of the themes I want to talk about in this one, about the efficient markets hypothesis and why this undead monster can never be laid to rest. (Warning: favorable references to Popper ahead!).
The ultimate zombie is one that is completely invulnerable. Neither special bullets nor hammer blows nor even decapitation can finally lay this undead being to rest. But dramatic logic requires that a zombie invulnerable to external threats must be subject to a subtle, but ultimately terminal, flaw that ends in its own destruction.
Ultimate zombies arise quite commonly in science and economics in the form of ideas that are immune from refutation. The classic examples arise from the popularised versions of Freudian psychology, centred on the Oedipus complex, named for the Greek tragic hero who unknowingly killed his father and married his mother. If a son hates his father, this is, obviously, evidence of the Oedipus complex. But, if he loves his father, this is explained as a repressed Oedipus complex. With rules like this, Freudian psychology can never be refuted.
But as a string of philosophers of science, being with the late Karl Popper, have shown, a theory that can’t be refuted by any conceivable evidence isn’t really a theory at all. All it says, in the end, is ‘anything can happen, and probably will’.
The global financial crisis, along with the earlier dotcom crisis has shown that, on any ordinary understanding of its terms, the efficient markets hypothesis can’t be right. Despite reaching a scale and sophistication unparalleled in history, global financial markets have shown themselves subject to the same manias, bubbles and busts that were seen in the Dutch tulip craze of the 17th century.
So supporters of the efficient markets hypothesis have sought a redefinition that would make it invulnerable to refutation. Their central argument is one that has already been discussed – if it is possible to diagnose the existence of a bubble, then it is possible to make arbitrarily large profits betting against it. And if someone like Warren Buffett has in fact done this, that can be put down to luck. Only if everybody can make money betting against the market can the EMH be wrong. But of course, it’s impossible for everyone to bet against the market – the market is just the aggregate of bets.
This argument in one form or another has been put forward by all the leading defenders of the EMH, notably including Eugene Fama and John Cochrane of Chicago and Scott Sumner of Bentley University
This set of observations from Scott Sumner in a blog post aptly titled ‘Defending the indefensible’ at least recognises the difficulties of the position
But why is Fama’s theory now in such disrepute? Because in the past ten years the world economy has seen two very important bubble-like patterns, indeed arguably the only two such market cycles in the US during my lifetime with macro significance. And they were both predicted by lots of experts because they violated popular theories of fundamentals. So start with the cognitive illusion that people have that makes them see bubbles even where there don’t exist.  People think they have made accurate predictions because an upswing is always EVENTUALLY followed by a downturn. Then add in the fact that The Economist really did make accurate predictions in two of the most important events in modern history. Do you think it will be possible to convince them that they just got lucky? About as likely as a husband convincing an already suspicious wife that he is innocent after twice being caught in bed with two separate women. So I feel sorry for Fama. He’s probably right, but I don’t see how he could ever convince anyone in this environment. It would be like trying to convince someone that neoliberalism was the right policy in 1933.
As a well known blogger would say, ‘Indeed’. Looking at the evidence of the two gigantic bubbles of the last decade, it’s hard to see how Sumner maintains his own faith, and he never really gives an explanation, except to say that it’s easy to misperceive bubbles. As far as macroeconomics is concerned, the experience of the Great Depression and of the current Global Financial Crisis (which as Sumner implies, really began with the 2001 recessions) is pretty strong evidence that neoliberalism is not the right policy, at least not for all occasions and not in the forms that prevailed in the 1920s or the 1990s.
But the ultimate response to this invulnerable zombie must be the same as Popper on Freudian psychology. If the Great Depression, the dotcom boom and bust and the current Global Financial Crisis are all consistent with the efficient markets hypothesis, the hypothesis can’t tell us much of interest about anything. At most, it says that even when markets are way out of line with economic reality, it is hard to exploit this fact to make a profit. Most of us (me and Krugman at any rate) already knew that, and confined ourselves to getting out of stocks when they seemed absurdly overvalued. Â
{ 126 comments }
David Moles 02.27.10 at 6:35 am
Is something seriously up with the formatting on this post, or is it me?
John Quiggin 02.27.10 at 6:44 am
Improved a bit now, not quite sure why there was so much white space.
Kenny Easwaran 02.27.10 at 9:03 am
That was a much more inoffensive Popper reference than I was expecting! I was worried about another Taleb-style “black swan” or Soros-style “open society” or something similar.
Hidari 02.27.10 at 10:20 am
I think Popper gets a hard time from ‘proper’ philosophers. Sure he’s not up there in the first rank. And sure, there’s all sorts of problems with the ‘falsifiable, unfalsifiable’ dichotomy. But as long as one views it as a common sense heuristic and not as some de facto ersatz scientific law, it’s an extremely useful tool to identify lunacy. It’s also quite a good tool to identify ‘ah but, you see, what you don’t understand is’ kinda reasoning.
JoB 02.27.10 at 10:29 am
4- yeah, Popper may not have been Wittgenstein but he’s done much more than almost all of those ridiculing him after Wittgenstein put together. Falsifiability, historicism, … good concepts to take stock of what is blatantly absurd.
Hidari 02.27.10 at 10:40 am
It’s also worth bearing in mind that Popper is not quite as shriekingly right wing as you might think from reading some of his ‘admirers’ (few of whom actually bother to read him, of course). To take just one example, Popper is generally credited with popularising the (silly, IMHO) assumption of ‘methodological individualism’. But this is something of a myth. It was really his student, J.W.N. Watkins, who was to blame. There are other, similar, examples.
JoB 02.27.10 at 11:31 am
To add insult to injury: something similar is the case with Hayek. His solutions weren’t the best thought out ones (although the worst ones he’s credited with were invented by others and associated to him), but his motivation and political concepts are very OK.
Neil 02.27.10 at 12:08 pm
I doubt very much we need Popper for this. Indeed, I suspect Popper just obscures what’s going on. The question is not “is there evidence that falsifies the theory” (since Quine many philosophers have accepted that there is no evidence that falsifies any theory whatsoever). Rather, the question is how what the probability of the theory’s being true, given the evidence. The GFC didn’t falsify neoclassical economics; rather, the probability that it is true fell as a result.
Hidari 02.27.10 at 12:11 pm
Hayek??!! Well I’m not sure the milk of human kindness is running that freely in my veins. It’s certainly true that the Austrians had some interesting ideas, although they are at their most interesting when they are at their most abstract (especially not just the links between Hayek and Popper, which are obvious and well known, but the links between Hayek (and the Austrians generally) and Wittgenstein. Trivial Pursuit fact: Hayek was Wittgenstein’s cousin).
Obvious the stress that Hayek put on ‘bottom up’ ‘self-organisation’ is important (although he course have acknowledged, but did not for ideological reasons, that this is also something one finds in anarchist thought). And his critique of certain forms of the planned economy is also important (although most people misunderstand what he is getting at here and where he is coming from).
But ultimately he was an idealogue who twisted facts to suit his theories (or prejudices) and there can be no greater intellectual sin. Look at his record on trade unions for example.
bert 02.27.10 at 12:30 pm
Kuhn? That’s falsification pretty much, with a sociological aspect to it. If the gaps between a conceptual framework and the observable facts become too glaring, competing frameworks can gather the allegiance of the herd.
bert 02.27.10 at 12:35 pm
Sorry, Kuhn is your initial premise. That these theories are subject to refutation.
Popper is your response to the superzombie tendency to sinuously evade refutation.
As you were.
JoB 02.27.10 at 12:39 pm
9- I’m not going to defend him and obviously he went overboard on the solutions (but I do admit that trade union legitimacy is problematic for me as well, conceptually). But – as a political thinker (as you point out yourself) – he was interesting. And I do think, to get into falsifiability, that whereas in ecomical science it is crucial to be falsifiable; it is not that straightforward in political theory. The ideology of Hayek was freedom – and, in principle, there’s nothing wrong with that. Where he was wrong is in the solutions he proposed (somewhat too black-to-white in light of his country’s recent history) to get at that ideological goal. At least he’s better than people that do not start with freedom as the ultimate goal.
roger 02.27.10 at 3:24 pm
Actually, instead of philosophers, I’d look to sociologists who have studied societies in which belief systems collapse. The idea that we are dealing here with science, and that all players are moved by truth, seems to me sociologically dubious. These belief systems seem to be about your standard things: social control, status, power, emotional work, etc. And the beliefs supplanting them will no doubt function in the same way. Although, admittedly, this collapse is a bit different – this is a small subculture. The larger collapse of course is of larger, and very much American, beliefs. Just as the collapse of the Soviet Union involved, on one level, the collapse of beliefs of the relatively small number of Soviet ideologues, and the larger collapse in the expectations of the population at large.
One of the reasons I would make this move is that economists are not obviously a positive science in Popper’s sense. They are a more myth oriented subculture – hence, the need for a “perfect market” model that is, of course, acknowledged as a myth – as well as the expansion of the concept market from its institutional meaning to meaning any situation that can possibly be designated by something called ‘supply’ and ‘demand’. These are clearly social and cultural symbols in the way that things in biology or physics are not. And they require different tools.
Paul Gottlieb 02.27.10 at 5:26 pm
I think Roger is exactly right: This is not a question of scientific theory, but one of religious belief. Our best guide to understanding the motivations of people like Fama and Casey Mulligan is not Popper, but Leon Festinger. His “When Prophecy Fails” is still the essential text for understanding the behavior of the (degenerate) Chicago school
bianca steele 02.27.10 at 5:45 pm
I don’t often comment on John Q’s posts because I really don’t understand economics. I feel somewhat familiar with the broad strokes but when I try to read an introductory econ textbook, even the math (and I think of myself as pretty comfortable with math) seems alien. But I was pretty surprised a few years ago to hear a strong version of the efficient markets hypothesis being treated as gospel: if so, where are the attempts to reduce this theory to the behavior of individuals and groups, and where are the examinations of the mechanism by which it operates? I had been under the impression that it was generally well accepted that economics needed to focus more on how it could be supported (and possibly refuted) by evidence.
Apparently I was wrong, and not only right-wing ideologues really believe in efficient markets.
bianca steele 02.27.10 at 6:16 pm
JoB @ 7: Unfortunately, the worst ideas Hayek is generally credited with are not as bad as the worst ideas he actually did invent.
bianca steele 02.27.10 at 7:09 pm
And it does seem to be a form of religious belief: for example, see how easily the idea that “I can trust, in the sense of not having to worry to much, that things have been arranged (by God) so that they’re pretty much for the best, for me” slides without effort into “(it is metaphysically true that) (God arranges things so that) everything a person does, assuming they’re a good person, is exactly perfect.” Lots of people find it easy to believe the former most of the time, but would be horrified to discover the latter was attributed to them (which I think is anyway an argument for pragmatism, that the former is just a useful belief and not metaphysically True).
But this is already too long for a comment box.
Jim Harrison 02.27.10 at 7:38 pm
The ideological function of the efficient market hypothesis is pretty obvious: economists aren’t paid to cause problems for their funders. It seems to me, however, that part of the appeal of the hypothesis is formal. Accepting it makes it easier to treat economics as an elegant deductive science instead of merely a branch of something empirical and non-U like sociology. Of course, just as the cleverness of jokes obscures or excuses their underlying crudeness and cruelty, the decorative mathematics of the model builders also has an indirect functional value. That doesn’t mean that the economists don’t enjoy the game for itself however.
Walt 02.27.10 at 7:57 pm
Popper’s not considered in the first rank of twentieth century philosophers? Really?
JoB 02.27.10 at 8:04 pm
16- example?
Probably the whole problem is that economists have been seduced to do politics, and are now wearing so much make-up that they have become the drag queens of science. Economists are the new metaphysicians and they think they can heal everything if we just let them find the Holy Grail.
Matt 02.27.10 at 8:09 pm
Walt at 19- I think that what’s meant here is two things- first, there are not that many people today who consider themselves “Popperians” (at least inside philosophy- some outside it do) or who spend a lot of time studying his writings. A lot more people now, for example, consider themselves “neo-Carnapians” and think it’s worth while to spend the time studying Carnap than they do Popper. Secondly, few people think that he was right on the issues that were most important to him, or that his path, while not quite right, opens up new and important pathways- I have in mind “falsification”, not as a useful tool for many cases (which it probably is) but as a full solution to the “demarcation problem”, his particular attack on induction, his rejection of the theory of evolution (and especially his reasoning there), his mind/body dualism (and his reasoning there) and the “third realm” idea, his historical scholarship on Plato, Hegel and Marx, and so on. I won’t say that no one thinks there’s something worthwhile in these areas, but they were central to Popper and now almost no one thinks they are right or even promising paths.
Walt 02.27.10 at 10:25 pm
What’s odd about that is that Popper’s victory among scientists is almost total. I don’t know how many people can trace the idea back to Popper, but almost everyone understands that saying that a theory is not falsifiable is an incredibly serious charge.
Matt 02.27.10 at 10:47 pm
I’m sure that’s right, Walt, but I strongly suspect that scientists have something somewhat different in mind than what Popper (so far as I understand him- I’m not an expert) had in mind- my impression is that scientists generally have a pretty loose idea in mind, while Popper had something pretty specific, something that doesn’t seem to fit very well with what scientists actually do- “critical experiments” that falsify a theory, etc. If you read Popper’s account of science it doesn’t seem very much like science, I think. Scientists are, rightfully, willing to mess with auxiliary hypothesis and the like if an experiment doesn’t go the way they think it should then they should be if they were actual Popperians. And, his rejection of induction doesn’t fit well with much science. Consider that he, for a long time, thought evolutionary theory wasn’t falsifiable, and so not science, and then changed his mind and held that it was falsifiable, but in fact falsified, and you get some idea of how Popper’s ideas of science don’t actually fit well with what scientists do. I don’t doubt that many consider themselves to be Popperians, but I expect that reading very much Popper would cure them of that.
Spectre 02.27.10 at 11:20 pm
Popper’s philosophy of science is a terrible description of what scientists actually do. That is one of Kuhn’s contribution, showing that unless Popper is proposing a normative theory of science, he is barking up the wrong tree. Falsifiability lives the life of the undead: Keynes’s dictum about practical men being in thrall to defunct economists applies equally aptly to scientists and philosophy of science. That’s not necessarily a bad thing, but using falsifiability against (supposedly) scientific economic theories is not the strongest attack one can mount.
tomslee 02.27.10 at 11:27 pm
@Matt (but not particularly Matt). You know what always puzzles me about philosophers? This apparent need to weigh up the good and bad contributions of a philosopher and then rate them by some arithmetic sum (OK I’m exaggerating). The practice seems different in the natural sciences, where the good bits of someone’s work are recognized and the bad bits just thrown aside. So no one says “sure Linus Pauling did great work on the theory of chemical bonding, but wasn’t that Vitamin C stuff weird?” or “Sure Newton discovered gravity and all that, but he got all that alchemy and biblical stuff (which was more important to him) wrong”. Instead, scientists tend to just not talk about the weird stuff, like it’s an impolite subject at a dinner table or something. It’s an approach with its own problems, but seems to me a more generous approach in that we recognize great contributions, not so much great people.
As just one other example, in a recent thread here on Zizek someone said that like most good thinkers, he hits about 400. My feeling then is, why is anyone reading the remaining 0.6 of his production? Why not just take the good bits and forget the rest?
tomslee 02.27.10 at 11:30 pm
“Consider that he, for a long time, thought evolutionary theory wasn’t falsifiable, and so not science, and then changed his mind and held that it was falsifiable, but in fact falsified, and you get some idea of how Popper’s ideas of science don’t actually fit well with what scientists do.”
I’d say that where Popper seems to have taken hold is in scientists’s attitudes to their daily practice of science, rather than in the grand sweep of the history of ideas. I think one can be Popperian (at least as an aspiration, and recognizing the many grey areas and limits of falsifiability) in the one area without being so in the other.
Matt 02.27.10 at 11:38 pm
I’d say that where Popper seems to have taken hold is in scientists’s attitudes to their daily practice of science, rather than in the grand sweep of the history of ideas.
This is what I’m not so sure of, at least depending on what you mean. I’m not a scientist and I’m not an expert of Popper (I’ve read a fair amount, and some criticism, but I’ve not read close to all of his works.) But my strong impression is that what scientists actually do in their normal practice is very much not like what Popper thought scientists should do, and that this is one of the more important criticisms of his view- that he tries to give us an account of what science is, what scientists do, and what separates science from non-science, but that his account turns out to have very little resemblance to actual practice. (Most scientists don’t spend that much time making bold conjectures and refuting them, rejecting whole theories on the basis of “critical experiments”, etc., at least not most of the time.) This is to say, one of the good reasons to reject Popper is that science is our best way of knowing, but that he got science, it seems to me, pretty clearly wrong.
Matt 02.27.10 at 11:40 pm
Why not just take the good bits and forget the rest?
If people could decide on what the good bits were they might be more inclined to do so!
bianca steele 02.28.10 at 12:29 am
JoB,
Speaking generally, most unforgivably: (1) He had an antitotalitarian argument that he preferred to direct only at Communism, allowing the reader to conclude that he had no issue with the far right. (2) Even after he moved to the United States, he implied that the US had a Communist economy, by asserting that all proper governments had the level of economic intervention characteristic of Fascism, which happened to be more intervention that the US had at the time. Either he didn’t know that or he was happy to implicitly accuse his hosts of Communism. In the latter case, he cannot be defended re (1) by saying he was only acting in accordance with his hosts’ beliefs (which is bad enough since it implies he believed the US was Fascist).
yx 02.28.10 at 12:29 am
If it is possible to come up with a strategy which can consistently beat the market, then EMH is a poor model of the market in question. For example, Vegas is clearly not operating under EMH: the house consistently wins, and optimal strategy is not to play at all. I don’t see how this is unfalsifiable. One can always argue that the success of a strategy is just luck, but by that line of reasoning, the hypothesis that smoking causes cancer is unfalsifiable–that’s what he have statistics for.
Herman 02.28.10 at 12:31 am
The defenders of the EMH make two claims that cannot both be true:
1. Nobody knows what the correct price based on fundamentals is.
2. Prices accurately reflect all relevant information.
The problem is if the first statement is true then the second becomes meaningless. If we do not know what the right price is, there is no way to tell whether or not all relevant information is accurately reflected in the existing price. Without a pricing model the EMH is meaningless. It becomes not a scientific hypothesis but a faith based initiative.
tomslee 02.28.10 at 12:37 am
Matt – well I’m no Popper expert either, and it’s a while since I’ve been a practising scientist. It is true that scientists don’t, as you put it “spend that much time making bold conjectures and refuting them” and so on. But I think they do spend time saying “if that’s true, then let’s see where it takes us” and that amoutns to much the same thing so long as they are prepared to dump the hypothesis if it takes them somewhere obviously wrong. But I’m prepared to admit this may be aspiration rather than execution; a goal more honoured in the breach than in the observance.
Herman 02.28.10 at 1:06 am
Continuing from my previous comment :
If the EMH is taken together with a pricing model ( say CAPM) then the combined hypothesis does become a scientific hypothesis that can be tested, has been tested, and shown to be false.
On the topic of hypothesis testing, as abused in macroeconomics, may I recommend Jeff Frankel’s paper on The Search For Perfect Nothingness
dsquared 02.28.10 at 1:25 am
A pet theory of mine, based on not really much more than casual empiricism of my family, is that while Popper’s philosophy of science doesn’t describe science particularly closely, it exactly describes the process that you have to go through if you are trying to fix a piece of lab equipment, and this recognition factor might account for the sociological truth that it is utterly hegemonic among practicing lab scientists.
Concerned Economist 02.28.10 at 2:05 am
John,
Perhaps it would be helpful for you to state the particular form of the EMH that you had in mind as you wrote your post.
You could then explain why the 2001 and 2008 crashes imply that by “any ordinary understanding of its terms, the efficient markets hypothesis can’t be right.†It seems that you feel that asset price bubbles by there very nature defy the EMH. I’m not sure this is actually correct.
You could also explain in somewhat greater detail why you think the revisions of the hypothesis is not falsifiable (by the way I’m not aware of any actual revisions so some greater detail here would also be helpful).
Phillip Hallam-Baker 02.28.10 at 2:46 am
“Popper may not have been Wittgenstein”, can the theory that he wasn’t be disproved?
hix 02.28.10 at 3:04 am
Probably not to DFA owhich seems to own Chicago University business school or anyone else that finances Chicago, but definitly for the finance industry as a whole.
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1105775
Things like that ares one big attack on wall streets most profitable activities.
Phillip Hallam-Baker 02.28.10 at 3:18 am
Popper was not really trying to describe science, he was trying to separate pseudo-science from science. Hence his popularity amongst scientists.
It is very hard to do proper science and even harder to do it according to the rules set out by Popper. The difference between real scientists and pseudo-scientists is not what they do but what the intent. Real scientific fields do attempt to test meaningful theories. Pseudo-science does not.
The efficient markets types are just as obnoxious and arrogant as the Marxists ever were. And like the Marxists, the conclusions that come out of their theories invariably support a particular ideological view.
I find it really hard to understand what an efficient market could be. Do these people seriously believe that the ‘value’ of IBM varies minute by minute in real time? That the amount of coke that some twenty something daytrader for a city firm has stuck up his nose has impact on the real value of IBM and not just the price?
If you define the value of a thing as the price it fetches then markets become efficient by tautology. But very clearly, Bernie Madoff’s clients did not find the markets very efficient. It took twenty years for the value of his accounts to reach the real value.
If the theory is that the prices will eventually come into balance then it becomes uncontrovertial and uninteresting. If the theory is that the prices are efficient second by second it is stupid. In between the theory is just meaningless.
yx 02.28.10 at 3:56 am
If the theory is that the prices will eventually come into balance then it becomes uncontrovertial and uninteresting. If the theory is that the prices are efficient second by second it is stupid. In between the theory is just meaningless.
In between is not meaningless. EMH may be able to explain, for example, why the price of health insurance stocks vary with the expected fate and composition of the healthcare bill, but not explain all minute-to-minute fluctuations. Would it be a useless theory in that case? Also, it seems a real empirical question, with the EMH answer neither tautological nor inherently silly, whether the equity premium is nonexistent because investors correctly factor equity risk, or does exist as result of some sort of psychological aversion to equity. Theories in the population sciences need not explain all details of all cases.
John Quiggin 02.28.10 at 4:03 am
@Concerned Economist – I have already written on all of this in earlier parts of the chapter, available at zombiecon.wikidot.com and posted in pieces here.
Phillip Hallam-Baker 02.28.10 at 4:44 am
@xy nobody seriously doubts that reality has some impact on market prices eventually. But the fact that that is not in doubt means that EMH must be something else if it has any meaning.
It is EFFICIENT, not ‘sorta, approximately bears some relationship to reality occasionally’ Market Hypothesis.
The point which is untestable superstition is the claim that the market participants are correctly pricing the effects. Clearly they are not even trying to as they are trying to predict the choices of other market participants over the next few minutes hours or days. So they are not calculating outcomes or even interested in outcomes. They are attempting to predict the prejudices and stupidity of their peers slightly earlier than they do.
Concerned Economist 02.28.10 at 5:08 am
John,
I know the earlier pieces but it is not clear what the precise claim you are making actually is.
In your chapter you actually mention in passing that the weak form of the EMH seems to work pretty well which would seem to disqualify it from having “zombie status.” For the strong or moderate forms, I am assuming that you are saying essentially the following: (i) the EMH says there can’t be bubbles in the market; (ii) we know that the dot-com episode and the recent market turbulence was due to speculative bubbles therefore (iii) the strong / moderate EMH must be wrong. Is that the line of argument?
Sorry about “boxing you in” but I would like some greater clarity if possible. (In particular, are you confident in (i) above?)
piglet 02.28.10 at 5:10 am
“EMH may be able to explain, for example, why the price of health insurance stocks vary with the expected fate and composition of the healthcare bill, but not explain all minute-to-minute fluctuations.”
Market prices are influenced by the expectations of market participants. Is that what you call EMH? If so, it is utterly banal and you are just confirming that EMH is useless as a predictive theory.
yx 02.28.10 at 5:23 am
I agree that EMH must mean something more than that reality has some effect on prices. But to take the position at the opposite extreme, and say that any deviation makes the theory useless, seems misguided also. I don’t think there is any theory in the population sciences which can fully explain the phenomena it is meant to account for–can you think of any? I think the best approach is to ask, for particular situations, how well the data fits EMH vs. opposing models (without the implication that either of the conflicting models is a perfect representation of reality even in that limited context). That’s why I brought up the equity premium puzzle–there is a real scientific question here as to the extent to which EMH is a sufficient explanation, and the extent to which we must bring in behaviorism.
It is not necessary for EMH that market participants attempt to calculate fundamental values for its own sake.
piglet 02.28.10 at 5:24 am
Concerned Economist: I’m not sure either what JQ defines as EMH but a hint would be to look at the Sumner post he links to: “But this is also what efficient markets look like. They are erratic. They have peaks and valleys. It looks like they have bubbles, but that is a cognitive illusion.”
The blog post is long and itself quite erratic but the only point it seems to make is this: markets are hard to predict. Therefore it is hard to know whether a pattern that looks like a bubble is really a bubble. Does this appear like a defensible theory to you? Does it make any falsifiable predictions? If you or anybody else has an answer to that question, let’s hear it. I agree with those who say that Popperian falsifiability is a useful heuristic.
yx 02.28.10 at 5:51 am
Market prices are influenced by the expectations of market participants. Is that what you call EMH? If so, it is utterly banal and you are just confirming that EMH is useless as a predictive theory.
What I call EMH is the hypothesis that all information is quickly and correctly reflected in the prices of assets. By “correctly” I mean there is no way to use information available to all participants to make a profit above the market by valuing assets at a non-market price. This hypothesis is false if we require it to fully cover all cases (which no one does). What I am arguing is that particular instances or degrees of failure should not cause us to abandon the theory, if it is useful as a partial explanation for some phenomena.
Does it make any falsifiable predictions?
I gave one above (30). EMH predicts that there will be no strategy which can consistently beat the market. The existence of such a strategy falsifies EMH.
John Quiggin 02.28.10 at 6:13 am
@Concerned Economist. I’m not too fussed about the weak form EMH, and I don’t think there was ever that much support for super-strong forms, relying either on private information or on the incredibly strained versions of RE found in the Casey Mulligan piece I criticised last post.
So, I am basically concerned with semi-strong efficiency, and your summary of my argument is pretty much right, with the addition that, if you thought the dotcom bubble was a fluke, you ought to have been convinced otherwise by what happened next.
What I’m saying now is that, maybe you can reformulate the semi-strong story in such a way as to avoid refutation by cases such as this, but if so, you have also lost the ability to derive policy implications.
Ted 02.28.10 at 6:52 am
by any ordinary understanding of its terms, the efficient markets hypothesis can’t be right.
What a curious, telling, argument. Given how much time has been devoted to the unquestionably UN-ordinary development and evolution of this EMH scientific hypothesis in arcane scholarly journals – and the subsequent debates in universities, Central Banks, and Treasuries around the globe, not to mention the research in management consulting firms (such as McKinsey) and other such professional organizations – why would you at the eleventh hour downgrade your critique to the mere standard of “ordinary understanding?” By what authority do you justify this sudden privileging of the “ordinary” over the academic and scholarly, and indeed very meaning explicitly used by Fama and EMHers themselves?
Given that we all know that you know this “ordinary meaning” standard is not in any way valid, what is going on?
piglet 02.28.10 at 7:02 am
“This hypothesis is false if we require it to fully cover all cases (which no one does).”
Well scientists sort of do require that a theory cover all cases. If it doesn’t, if violations are observed, then it is either immature (i. e. the underlying phenomenon is not well understood), or false, or needs to be refined. Which do you opt for?
And I have to take issue with your definition of “correct”. Terms like correct or efficient are normative – they suggest that the market is “right” in some meaningful sense. What you really seem to be saying is that there is no such a thing as a correct interpretation of the information available, therefore accepting the verdict of the market is the best one can do. This is something very different.
John Quiggin 02.28.10 at 7:02 am
Ted, you’ve delighted us long enough.
Your silly misreadings and general trollishness have exhausted my patience. Please go and find some other blog to infest.
John E 02.28.10 at 7:17 am
1. The core of popper work is not “falsification”, but creating a theory of non-justificationist epistemology. In his later work he almost entirely abandoned the concept of falsification in favor of “criticism”.
2. The main difference between Popper and Kuhn is that popper thinks the shift between theories can be rational, and it can be based on reason, while Kuhn describes this shift as sociological and beyond reason.
3. Popper is not a Methodological Individualist! He used this term, but actually he takes account of institutions. Joseph Agassi describe his attitude as “Methodological Institualism” (See his attack on individualism at chapter 14 of the Open Society).
Concerned Economist 02.28.10 at 7:21 am
John,
OK, I think I understand the basic outline of the claim. I’m not sure that EMH actually rules out speculative bubbles but for the purpose of this post let’s assume that it does so that the claim is somewhat meaningful. Here are some questions for your consideration:
1. What is a bubble? People talk about bubbles a lot but I’m not aware of a sharp definition of what a price bubble actually is. (Economists don’t have particularly well-formed ideas of what price bubbles are either so I’m not claiming that we know the true answer …)
2. Could you prove or argue that the sub-prime event (and subsequent events) was caused by an asset price bubble given the definition in (1)? I think most people would claim that it was obvious (of course it it’s really obvious then demonstrating so would be easy …).
3. Would it make any difference if the EMH were right and the price reduction was caused by a sharp revision in the market’s assesment of the value of sub-prime assets?
yx 02.28.10 at 7:33 am
Well scientists sort of do require that a theory cover all cases. If it doesn’t, if violations are observed, then it is either immature (i. e. the underlying phenomenon is not well understood), or false, or needs to be refined. Which do you opt for?
In practice, in the population sciences no theory fulfills this criteria, as far as I know. If you have a counter-example please share it. The answer from your list is “immature”, but by that standard, all sciences except physics and its close relatives are “immature,” and may remain that way for quite a while. From a policy perspective, being close enough is acceptable, given no superior alternative.
My own field is in ecology. In ecology, the way things work is someone comes up with a nice mathematical model, and then lots of people test particular systems for how well they fit the model (perhaps relative to an alternative.) No one says if 90% of systems fit the model better than any other that we should throw it away and look for a superior one on account of the 10%. Perhaps the model needs to be refined (although it’s possible that the other factors operating are too small and variable to reliably detect). And policy is based on the hypotheses that test pretty well, not on the (non-existent) hypotheses which succeed without exception. No one seems to attack any of the other population sciences for the faults noted for theories in economics, yet they seem very similar to me.
Robert Waldmann 02.28.10 at 9:54 am
When you write “But as a string of philosophers of science, being with the late Karl Popper,” I think you meant “beginning”. I also think you are wrong and that Pierce and Carnap said that before Popper did.
I admit I probably don’t know what I’m talking about, but my guess is that Popper’s contribution was to coin the word “falsification” which has the same meaning as the plain English “verification.”
Or forget Carnap and Pierce. What about Mach ? He asserted that meaningful statements were all statements about measurements. If so, unfalisfiable claims mean nothing. He wrote well before Popper.
To claim that the string of philosophers began with Popper, you have to disprove all claims that someone else said it first. I don’t think you can.
Well you can’t, unless you redefine ” a string of philosophers of science, begining with the late Karl Popper, have shown, a theory that can’t be refuted by any conceivable evidence isn’t really a theory at all. ” so that it is unfalsifiable.
JoB 02.28.10 at 10:48 am
John quotes Popper aptly. His core was a correspondence theory of truth where atoms (propositions) needed to correspond to atoms (facts). He held on to his core because it was the only way he thought one could avoid “I said so” reasoning. EMH – according to John – is “we say so” and fails according to Popper. If you want it to fail as well under a more complex holistic coherence account of truth, there is more work to do. That may be interesting work but I think that the Popperian heuristic together with our common sense understanding of the concepts works just fine. On the other hand, the most you’ll achieve in this way is saying EMH says nothing. With more work you may be able to be saying that markets cannot be efficient if there are markets that condemn the people in them to starve to death.
bianca@29- (1) is rather one-sided as an observation & (2) is about the person; I’m not going to deny he was a bit of a prick
Personally I tend to give anybody with an anti-totalitarian ambition the benefit of the doubt. There is quite enough shouting already.
JK 02.28.10 at 10:55 am
dsquared,
A pet theory of mine, based on not really much more than casual empiricism of my family, is that while Popper’s philosophy of science doesn’t describe science particularly closely, it exactly describes the process that you have to go through if you are trying to fix a piece of lab equipment, and this recognition factor might account for the sociological truth that it is utterly hegemonic among practicing lab scientists.
As a lab scientist I think there is something to that. Of course it’s helped by the Feynman / Steven Weinberg observation that philosophy of science is about as useful to scientists as ornithology is to birds. Hegel in the beginning of his Logic put it this way: “if we said that eating was impossible before we had acquired a knowledge of the chemical, botanical, and zoological characters of our food; and that we must delay digestion till we had finished the study of anatomy and physiology” then “these sciences in their field, like philosophy in its, would gain greatly in point of utility; in fact, their utility would rise to the height of absolute and universal indispensableness.”
alex 02.28.10 at 11:39 am
Since it is a well-established fact that company boards manipulate the information available about their activities in order to maximise the short-term price of their stocks – or ‘increase shareholder value’ as it’s known – how can a ‘market’ which is based solely on the aggregation of such manipulations be in any sense an ‘efficient’ representation of anything except the shifting sands of gullibility and suspicion amongst investors?
Blissex 02.28.10 at 11:54 am
«boards manipulate the information available about their activities in order to maximise the short-term price [ … ] how can a ‘market’ which is based solely on the aggregation of such manipulations»
This is not relevant to the discussion — because the EMH people would object that in that market you can always short those manipulations, making a killing later on and thus arbitraging away the manipulation. Therefore the market is not based “solely on the aggregation of such manipulations”, but also on the aggregation of arbitraging it away.
But perhaps you are arguing that the information provided is not just manipulated, but false. Well, in that case he EMH does not apply, because one of the core assumptions of the model on which the EMH is based is that everything is clean and clear.
The EMH in its weak formulation states that you cannot beat the market using public information (and implicitly but sneakilu that all relevant information is indeed public).
Some true propagandists might also argue that even if companies acted fraudulently in misreporting their numbers, the same board members that misreported them would short the shared price to make a killing, and thus arbitrage away the misreporting, or in the long term market participants would learn to discount the information provided.
Shmoe 02.28.10 at 2:30 pm
Now, if he we can just get him to acknowledge Carnap’s genius. Down with Metaphysics!
Barry 02.28.10 at 4:25 pm
hix:
“Probably not to DFA owhich seems to own Chicago University business school or anyone else that finances Chicago, but definitly for the finance industry as a whole.”
Could you fill me in on what DFA is? (I checked Wipipedia, but didn’t find any readily apparent items)
john c. halasz 02.28.10 at 6:47 pm
@58:
“Dimension Fund Advisors”- a hedge fund originally spawned out of U. of C. business school.
bianca steele 02.28.10 at 7:06 pm
JoB, I believe the issue is not about the person but about the content of his writings, which are handed out like lollipops to undergraduates curious about “Movement Conservatism.”
Barry 02.28.10 at 8:05 pm
Thank, John.
Clod Levi-Strauss 02.28.10 at 9:12 pm
“Steven Weinberg observation that philosophy of science is about as useful to scientists as ornithology is to birds.”
Philosophy, especially of the “Continental” sort might do Weinberg a world of good, since his rationalism runs to arrogance and then to the irrational: unmitigated racism. Welcome to the 19th century. Freud would have something to say about this.
And since the author of this post used Popper to attack Freud I should point out that the Oedipus complex true or false refers to excess of one emotion or another. We now refer to manic depression as “bi-polar disorder”. Would you rather we study one of them at a time?
There’s no scientific cure for intellectual laziness, and there never will be.
Clod Levi-Strauss 02.28.10 at 9:19 pm
Right book, wrong link. The relevant chapter is
here..
It’s straightforward, fundamentally historically inaccurate and intellectually obscene.
bert 02.28.10 at 9:53 pm
My memory of Plato as described in The Open Society and its Enemies is of a proto-fascist. Although I wouldn’t argue with a reading of volume 1 as non-specific anti-totalitarianism. In any event, it’s a real stretch to criticize Popper for operating some sort of double standard as regards fascism.
If you want to settle on a criticism, perhaps a better option would be that he identified shared features of totalitarian governments of right and left and that, by placing them in opposition to political freedom, he opened the door to characters like Jonah Goldberg eager to smear their domestic opponents. But really in this case the proper target of that criticism should be the current movement conservatives, and their unprincipled abuse of intellectual source material.
Blissex 02.28.10 at 10:52 pm
«The ideological function of the efficient market hypothesis
is pretty obvious: economists aren~t paid to cause problems for
their funders. It seems to me, however, that part of the appeal
of the hypothesis is formal. Accepting it makes it easier to
treat economics as an elegant deductive science instead of
merely a branch of something empirical and non-U like
sociology.»
Indeed the purpose of Economics in the USA is to support (by any
means necessary) the central truthiness of Economics, that the
distribution of income in the USA is justified by the enormously
higher productivity of the creative and productive elites
For those who are not totally committed to that purpose there is
no need for empirical refutations of the EMH, because it is
based on models that contain gross mathematical mistakes, as
well as absurd simplifications (a single agent, etc.).
Since the Cambridge Capital Controversy and the Second Best
Theorem (both of them about 50 years ago) both the validity and
relevance of the Arrow-Debreu-Lucas model and anything remotely
derived from it are entirely discredited, but this simply does
not matter to those who want to have a beautifully well rewarded
career in USA Economics.
What matters is to uphold (by any means necessary) the central
truthiness of Economics.
Those who study instead Political Economy (mostly outside the
USA) address entirely different questions.
Blissex 02.28.10 at 10:54 pm
«In practice, in the population sciences no theory fulfills
this criteria, as far as I know. If you have a counter-example
please share it. The answer from your list is ~immature~, but
by that standard, all sciences except physics and its close
relatives are ~immature,~ and may remain that way for quite a
while.»
That’s what Rutherford said: “All science is either physics or
stamp collecting”, just a reminder.
But even “in X% plus or minus Y% of cases” is a testable
hypothesis, just not one as precise as “in 100% of cases”. Then
usual example “almost all swans are white [as of 2010 on Earth]”
is well established and useful, because it tells you something
you can rely on, and implicitly warns you that some swans may
not be be white, and who knows about swans in the past, the
future , or other planets.
«From a policy perspective, being close enough is acceptable,
given no superior alternative.»
However the crux of the matter is “policy perspective”. As to
policy, having on’es arguments supported by “science” instead of
“close enough” sounds a lot better; those propagandists that
repeat the EMH as a talking point need absolutes and deep
mathematics (however flawed) because that gives a valuable
flavour of both respectability and obscurity to the central
truthiness of Economics about the income distribution being
justified always and only by productivity.
As to me, I think that all really useful knowledge is about
calls of judgement and getting odds right, but I do recognize
that absolute scientific truths sound a lot better for
propaganda purposes. Since there are very few of the latter
outside physics (and even most of those are
temporary/questionable), what has become crucial is to play word
games; and the favourite word game of astute propagandists is to
resort to the usual rhetoric tricks, mostly begging the question
and redefining words so that statements become tautologies.
As to begging the question the best trick that Serious
Economists use to to posit premises that imply the desired
results (Arrow-Debreu-Lucas is the major example), and even if
they have made some mistakes as to that, to fix them they have
either added more ad-hoc premises (e.g. rational expectations),
or just wrapped up the issues into obscure verbiage.
Complicated mathematics and obscure verbiage are just meant to
hide that the premises are meant to imply the desired conclusion
about the central truthiness of Economics.
Blissex 02.28.10 at 10:55 pm
« if the EMH were right and the price reduction was caused by
a sharp revision in the market~s assesment of the value of
sub-prime assets?»
Ah sure, but I note the tautology here: since those are market
prices, if there was any price reduction it must have happened
because the majority of market participants revised their
assesment of the value of sub-prime assets.
If the EMH is just the observation that market prices are a
function of assessments of value by market participants, it
reduces to a game of words.
The big deal is whether in the short term the assessments
of value by market participants can deviate significantly from
“true value”, whatever that means (except for “the assessment of
value by market participants”).
As to that, finite time horizons, finite speeds, and
agent-principal conflicts, can easily cause deviations for some
time. Perhaps Fama would be right over eternity and infinite
markets if market participants were immortal and had no
agent-principal conflicts (which are in effect the assumptions
behind A-D-L, and at the root of the Cambridge Capital
Controversy).
Also there is a point that is often overlooked: “all available
data” (the usual qualifier) includes both measurements and
interpretive models. The latter *can* be systematically wrong,
for good reasons (everybody is using the same flawed theory) or
bad reasons (a lot of people have conflicts of interest).
Ultimately there are two camps: the propagandists that believe
in the EMH and all the premises that they have wished to make it
“provable”, and those who know that “markets can remain
irrational for longer than you can remain solvent” (which is an
argument about finite time horizons and resources and boundary
conditions, and also about agent-principal conflicts).
politicalfootball 03.01.10 at 1:39 am
3. Would it make any difference if the EMH were right and the price reduction was caused by a sharp revision in the market’s assesment of the value of sub-prime assets?
Hmm, you might want to rephrase this – it’s got the tautological feel to it that JQ is complaining about. A price reduction can’t be anything but a revision in the market’s assessment.
I think what’s required is a definition of “efficient” that is 1. non-tautological and 2. useful for explaining events.
EMH adherents want to conflate efficiency with unpredictability, but even the Unpredictable Markets Hypothesis doesn’t seem to actually work in real life.
hix 03.01.10 at 1:46 am
Hedge fund is quite a stretch. They sell hm lets say, quantitative mutual funds.
http://www.dfaus.com/
Gene Fama works as salesman for them:
http://www.dimensional.com/famafrench/
Heres why i said they own Chicago Business School:
http://www.uchicago.edu/features/20081106_booth.shtml
yx 03.01.10 at 3:05 am
But even “in X% plus or minus Y% of cases†is a testable hypothesis, just not one as precise as “in 100% of casesâ€.
I agree. That’s the correct way to go about it.
that gives a valuable flavour of both respectability and obscurity to the central truthiness of Economics about the income distribution being justified always and only by productivity.
I don’t see how the EMH connects to “income distribution being justified always and only by productivity.” If anything, I think the broad utilitarian focus of economics suggests drastic wealth redistribution. There is no notion in economics of “rights”, just of utility.
Hidari 03.01.10 at 8:10 am
‘Philosophy, especially of the “Continental†sort might do Weinberg a world of good, since his rationalism runs to arrogance and then to the irrational: unmitigated racism’.
I understand what you are getting at by stating that Weinberg’s rationalism has turned to irrationalism (CF Eddie Izzard’s inspired sketch about how very, very, very, left wing people can suddenly go a bit further ‘left’ and find themselves as being very very very right wing, and so on), but you could also argue that what Weinberg needs a dose of is empiricism. After all, a quick look at the empirical data about the ‘founding’ of the State of Israel would quickly show that every single thing he states about it is not only false but blatantly and obviously false. However, interestingly, this chapter contains few if any footnotes: when one is arguing from personal prejudices, facts would just get in the way.
Another issue about this chapter, which struck me when I was browsing through his book in a book shop is that it is sandwiched between lots of other chapters discussing his brilliance at physics (which is not being questioned). This therefore seems to me to be a particularly malign example of the Argument from Authority: ‘I am a great physicist, therefore you should accept unquestioningly everything I have to say about Israel.’ There’s also a bit of ‘physics is true, everything else is stamp collecting’ here: after all, if history is mere ‘stamp collecting’ why bother reading any of the empirical data?
bianca steele 03.01.10 at 3:37 pm
tomslee,
In the physical sciences, if one doesn’t worry too much about scientists’ failures (like Newton’s dallying with religious mysticism), it is because personalities don’t matter much in science; one learns science from textbooks that treat the history of the discipline in a cursory manner. (At least at the beginners’ level–which is not to say that the science beginners learn is almost but not quite entirely unlike the science researchers have to learn before they can begin doing research–though even at the cutting edge, I don’t think researchers concern themselves much with old primary sources, unless they are doing history of the discipline for its own sake.)
And in any event, I wouldn’t be surprised if e.g. Pauling’s excursions don’t look funny in a few decades, much as rationalists’ and empiricists’ dabblings with spiritualism in the 19th century are now seen as a pathological overreaction to the limits of science.
mpowell 03.01.10 at 4:17 pm
The real stink of it is that while EMH can never be refuted on its own terms due to its meaningless malleability, it is still successfully wielded as a policy argument regardless of its applicability. That is why I think Fama and his fellow advocates are odious people- because I do not believe this bait-and-switch is accidental.
bianca steele 03.01.10 at 4:17 pm
As a contrast, look at Adam Gopnik’s piece on cookbooks in the New Yorker a month or two ago: He describes learning as necessarily being transmitted in face to face interactions between teacher and pupil, and by practice, never by means of words on the page. He is talking about learning to cook, but IIRC he does generalize.
someguy 03.01.10 at 4:22 pm
EHM isn’t a claim of best in the sense of good or excellent. It is a claim of best in the sense of better that any lame attempt by you or me.
EMH is a horrible name.
It should be called –
A defense of free market resource allocation that acknowledges that a free market is a really horrible and crappy way of allocating resources that guarantees gigantic waste and is prone to insane mania’s and yet sadly but correctly claims that it remains the least crappy way of allocating resources.
But that doesn’t role off the tongue quite the same.
I am not even sure asset prices reflect value over the long let alone at this point in time. But even if market prices only come close only in the long run that is a huge built in advantage.
Even given 20/20 hindsight how would you have allocated resources better? Is there even a conceivable mechanism for doing so? How could we measure to make sure you did a good job?
I wouldn’t have invested so much in housing or dotcoms doesn’t make the grade.
piglet 03.01.10 at 4:50 pm
yx 02.28.10 at 7:33 am
You are missing the point and maybe we are wasting our time here but I’ll try once more.
A theory that says “X will happen except if it doesn’t” is useless, can we agree on that? Then what exactly is that makes EMH, in your view, “acceptable, given no superior alternative”? You state that “EMH predicts that there will be no strategy which can consistently beat the market”. But we know that there are market participants who seem to “consistently beat the market”. There have to be because “the market” is just an aggregate of winners and losers. So how do we know whether Buffet is just lucky or whether he has a superior strategy? We don’t. EMH can’t be falsified that way.
Here’s a suggestion. I think what has been referred to as the “weak” form of EMH states that there can be no algorithm that would predict market prices. If such an algorithm existed, everybody would try to benefit from its predictions and this would render them useless. Now this is different from attributing normative characteristics (“efficient”, “correct”) to markets. The EMH would probably be a lot less controversial if it were called the “Unpredictable Market Heuristic (UMH) instead.
As an aisde, does it follow from the EMH (or UMH) that hedge funds only make sense if fraud is involved? I think it does, do others agree?
Walt 03.01.10 at 4:51 pm
That’s not actually what the EMH means, someguy. Really, there’s a whole academic literature and everything.
piglet 03.01.10 at 5:01 pm
Bianca: “I don’t think researchers concern themselves much with old primary sources”
I agree and I think this is a characteristic of any mature science. A mature science is based on contributions over centuries but each generation works through and reformulates the material. Newton’s laws are still taught but they are not taught in the way they were originally written down. Partly this is simply because we would have difficulty understanding even his notations, partly this is because new knowledge has changed our understanding of the meaning of Newton’s laws. Iow primary sources in the natural sciences are of only historical interest because the original ideas have been subject to so many revisions. When it is necessary to study the original centuries-old sources of a field of thought, it is probably religion not science.
someguy 03.01.10 at 5:09 pm
Walt,
In ordinary terms yes it is.
And after checking to make sure things haven’t change since my time
http://www-stat.wharton.upenn.edu/~steele/Courses/434/434Context/EfficientMarket/malkiel.pdf
[I just read the summary, to make sure that I still roughly understood the concept, I do not necessarily agree or disagree with any of the conclusions of the paper.]
Yes it is.
bianca steele 03.01.10 at 5:17 pm
piglet,
not taught in the way they were originally written down: okay
but each generation works through and reformulates the material: not sure
Newton didn’t write f = ma but this does not need to be reformulated in each generation (the problems we understand the formula to apply to do not change in each generation either). Haven’t calculus textbooks remained pretty much unchanged in the US and England at least for nearly 200 years? Science is science because there is no need to worry too much about formulation of the ideas, or what is the same thing, because the formulation is already abstract.
I’m not talking about metaphysics; Newton’s ideas of what is ultimately physically real are no longer considered true. But his ideas about the relationship among force, mass, and acceleration in circumstances involving medium-sized objects are still valid.
In fields where notation changes rapidly (perhaps statistics), this is because underlying theories are changing, not because underlying theories need to be reformulated to match users’ language.
probably religion not science: makes no sense, as we can’t verify your own paragraph without doing precisely this
Walt 03.01.10 at 5:23 pm
No, it’s not, and you don’t get to have the argument you’d prefer to have. The argument is not “Sure there are $100 lying on the sidewalk for a long time, but it’s better than socialism.” It’s that “There aren’t $100 lying on the sidewalk for very long.” If people forthrightly moved to the first argument, it would be an improvement, since we would then be working in the realm of facts, rather than in the realm of tautology.
bianca steele 03.01.10 at 5:39 pm
re my @82: of course when you get to building bridges, f=ma is modified for convenience’ sake. But this is not support for each generation rethinks and reformulates the ideas of its predecessors, nor for engineers don’t need to understand f=ma and aren’t in fact taught f=ma, or for that matter taught the fact that engineering formulas can be and have been derived from the principle in a way they can look up and verify.
someguy 03.01.10 at 5:59 pm
Walt,
No. In regards to EMH and free markets best has always – always meant better than any alternative. You cannot predict the walk.
[Even that isn’t 100% true. I am pretty sure Warren Buffet has more than once. I only briefly believed all the hand waving and tail distribution business and understand how annoying it can be.]
It is an argument you cannot avoid.
If you are formulating a meaningful discussion of policy implications you cannot just point to the dotcom and housing bubbles and claim that markets aren’t always rational and completely random.
You need the second bit. A better mechanism.
The first bit is important. It really is. And yes some folks are still waving hands and talking about stopped clocks.
But you still need the second part.
Even in some kind of weaker than weak form EMH is a pretty powerful concept with important implications and it has been pretty well tested.
Phillip Hallam-Baker 03.01.10 at 6:04 pm
Anyone who has actually run a business of any size knows that working out how well you are doing is a huge problem. Depending on your assumptions you may be making huge profits or running the company into the ground. You don’t need to resort to Enron tactics to produce accounts that bear no relationship to reality.
If the people inside a company don’t know how much they are making, how can the markets?
The other bizarre feature of this debate is the miss-statement of Rational Choice I swa earlier today. In common with many people, the author described rat choice as the assumption that people make rational choices. That is a totally idiotic assumption, but it is one that many economists give the impression of believing.
My understanding of rat choice is that according to the theory, you can build usefull models of how people react if you assume that they are rational. In other words, rat choice is the economist’s equivalent of assuming that there is no friction in certain mechanics problems. Physicists know that they have to account for friction in many cases, but often it can be ignored.
Almost twenty years ago now, when I was looking at different Web designs and the impact that certain low-level protocol choices would make on adoption, I used a series of rat-choice type models. I did the same thing ten years ago when trying to get a handle on Internet crime. You can really learn a lot from those models. But one of the things I discovered early on is that it does not matter much what the structure of your assumptions are. The curves look pretty much the same. I started off with five or six different models of the value of a network protocol, but the thing was that we could beat Gopher no matter what the assumptions were. And the graphs would look the same.
EHM is assuming that there is no friction. It is a reasonable approximation but a totally irrational and foolish dogma if presented as a fact.
Walt 03.01.10 at 6:17 pm
That’s silly, someguy. I’m happy to believe that financial markets are the best of real-world mechanisms to allocate capital. In fact, in many ways I _do_ believe it. I don’t want to replace the stock and bond markets with government agencies. I’m not happy with the current regulatory regime, but that’s not the same as a full-throated endorsement of central planning.
And I’m going to go ahead and point out the obvious. The EMH is _not_ well-tested. There’s basically one fact that supports it: the fact that short-term stock prices are unpredictable, and then lots of facts that don’t support it, accompanied by long-winded explanations of why those facts don’t count. The Malkiel paper is typical in that regard.
piglet 03.01.10 at 7:21 pm
Bianca: “Haven’t calculus textbooks remained pretty much unchanged in the US and England at least for nearly 200 years? Science is science because there is no need to worry too much about formulation of the ideas, or what is the same thing, because the formulation is already abstract.”
Caught me by surprise. I never read a 200 years old calculus textbook but I can assure you that many of the most fundamental ideas taught in modern calculus weren’t around by Newton’s or even Euler’s or Gauss’ time. The pioneers didn’t even have a sure-footed definition of the real numbers, one modern version of which is due to Dedekind. They used infinitesimals in a naive way, lacking a precise definition of convergence (Cauchy-Weierstrass). More generally, the whole of modern Mathematics is based on set theory that was only developed in the late 19th century (Cantor) and received its modern form in the early 20th century (Zermelo-Fraenkel). Algebraic structures were only rigorously explored in the 19th and 20 the centuries. Geometry was revolutionized in the 19th century from Lobachevsky to Riemann and Klein. The axiomatic approach which is the base of most modern Mathematics underwent significant questioning and revision during the foundational crisis (Russell, Hilbert).
Modern calculus builds – to some extent – on all these traditions, as well as on Newton and Leibniz. If your textbooks didn’t reflect that, you have been cheated. The practice of Mathematics has changed significantly over time, which makes it difficult to understand the original works even for an expert.
Walt 03.01.10 at 7:23 pm
You could safely use a 100-year old calculus textbook, I suspect.
piglet 03.01.10 at 7:26 pm
Walt: perhaps, if you could understand it. Why don’t you give it a try and share the results with us.
someguy 03.01.10 at 7:30 pm
Walt,
You know of studies indicating that people can predict relative long term prices? That in the long run the walk is non random?
I am glad you are happy to believe that free market capital allocation is the best real world approach.
I am very doubtful John Quiggin and others share your faith. Those who don’t think it is the best approach need to explain why and how their approach would be better.
bianca steele 03.01.10 at 7:41 pm
piglet,
I think that’s mostly BS. I’m aware of the claim that mathematics experienced a crisis around the turn of the 20th century because it didn’t have foundations, which is why Russell determined to give it some good ones. I don’t actually buy that and think Russell would have been better off with a defense against corrosive skepticism (which goes beyond even Rorty’s pragmatism, leading to the question why I even bother with a philosophers’ blog). I doubt calculus will be replaced with real analysis either.
Walt 03.01.10 at 7:46 pm
someguy: Not only do I know of examples, I know of an easy place you can find references — that Malkiel paper you linked to. Of course Malkiel is dismissive, so I don’t recommend taking his word for it.
piglet: You’re radically overestimating the amount calculus has changed over the last century. Thompson’s Calculus Made Easy dates from 1910 (very large pdf here), and is basically comprehensible.
michael e sullivan 03.01.10 at 7:59 pm
“So how do we know whether Buffet is just lucky or whether he has a superior strategy? We don’t. EMH can’t be falsified that way.”
It can’t be falsified because one person out of a universe of millions of investors has done very well. But that’s because you don’t have all the information. You *can* do statistical tests on groups of investors, and if too many outperform the market by too much over too long a time, that would be very strong evidence against the efficiency of that market.
The problem with picking out Buffet (or whomever) is that you don’t know enough about the results of everybody else. It is highly likely that out of many many millions of investors, if their profits were determined purely by random die-rolling, some investor somewhere would be on top and have made a ridiculous amount of money.
But if you study a closed system, you can get real answers to how many such outperformers there should be if results were spread purely by luck, and compare them to the actual results. Of course the problem with randomness and statistics is that you never have an absolute 100% result, but you might well be able to get pretty certain one way or the other.
And people have done such tests, and most of them seem to tilt in favor of no significant evidence for skill-based outperformance in the broad stock or bond markets. And other tests have been done of various trading algorithms based on various technical theories and about various models of fundamental value, and only very rarely do they come up with a positive result. That’s actually pretty good evidence in favor of weak-form EMH, although the existence of any counter-results at all is pretty damning to the strongest forms.
Unfortunately the *same* problem exists with looking at events like the 2000 internet crash, the 2006 housing bust and the 2008-2009 recession (and associated booms beforehand). For this to be damning evidence against the EMH, there needs to be a specific claim about what fundamentals were out of line and how you would have profited from identifying them, and then how that algorithm would have performed in other circumstances. The algorithms of “stocks suck just stay away” and “always listen when lots of pundits say there is a bubble” would have soundly beaten the S&P in the first half of 2000 or the fourth quarter of 2008, but as a long-term general strategy the first doesn’t seem to do well at all, and there have been collosal failures in the past of the second — I think the jury is out on whether such a strategy actually beats the very irrational market.
One of the keys here is that the claim of market efficiency is *not* a Leibnizian claim that this is the best of all possible worlds. It is a very specific claim that arbitrage opportunities greater than the bid/ask spread do not persist. When people use it to suggest that the market price is not merely a good or best available estimate of value, but that it is impossible even in principle to allocate capital more efficiently than the financial markets, they cannot be talking about the forms of EMH that stand up to the slightest scrutiny. That kind of policy suggestion would only be backed up by the super-strong forms which not only seem obviously false at first glance, but also have been pretty much demolished by empirical tests. I don’t think Sumner is guilty of this last. I’m less sure about Fama.
John Quiggin 03.01.10 at 8:07 pm
I’d be dubious about using a 100-year old calculus textbook if it was in English. I recall Littlechild writing that confusion over such basic questions as “what is a function?” persisted there for a decade or two after French and German mathematicians had sorted all this out. This was the era of the Cambridge tripos, after all.
Walt 03.01.10 at 8:15 pm
Well, I linked to a 100-year old calculus textbook that is still in print. (Though it has been revised.)
John Quiggin 03.01.10 at 8:26 pm
Walt, your book solves the problems of functions, limits, and sets so on pretty easily. None of these concepts is even mentioned. It just starts with a handwaving definition of infinitesimals and runs with it.
This book could have been written in 1810, and wouldn’t have been much different.
Walt 03.01.10 at 8:48 pm
Maybe I’ve lost track of whatever point was being made, but isn’t that basically how it’s taught at most places? We’re talking about calculus texts, not real analysis texts.
Walt 03.01.10 at 9:00 pm
I went back and skimmed the thread. Yes, calculus books are different now, in that they will include epsilon-delta proofs. I don’t know when they first appeared in college-level textbooks. They’re in Courant’s textbook, which is from 1934.
yx 03.01.10 at 9:25 pm
piglet, I never actually defended EMH as “correct.” My argument is merely that EMH is a useful and interesting scientific hypothesis by the standards of the population sciences; falsifiable, empirically testable, and neither tautological nor obviously false. Moreover, we should be open to the possibility that EMH is a partial explanation of some markets, or a good explanation of some markets and a poor explanation of others, rather than claiming that one person playing slots somewhere should cause us to throw it out entirely.
Yes, I agree that “X will happen except if it doesn’t” is useless. “X will happen around 80% of the time under conditions a,b,c” is not useless, even if we have no good explanation for the 20%. Or, “X explains 60% of the data” is not useless. To take an example from earlier in this thread, “X explains the particular phenomenon we are looking at, but doesn’t explain all minute-to-minute fluctuations” is not useless. These are the kinds of hypotheses that population scientists use all the time. Epidemiology, for example, has been massively productive using models which are “false” if taken as a perfect literal description, but which can account well enough for what we are interested in to be useful to policy. If you disagree with me, please do what I’ve asked for repeatedly, and give me an example of a hypothesis in the population sciences which does not fail the tests you apply to EMH.
But we know that there are market participants who seem to “consistently beat the marketâ€. There have to be because “the market†is just an aggregate of winners and losers. So how do we know whether Buffet is just lucky or whether he has a superior strategy? We don’t. EMH can’t be falsified that way.
I’ve already given two examples of EMH being (potentially) falsified. 1) In a casino, EMH is false because optimal strategy for most people is not to play at all, yet casinos manage to attract customers. 2) If the behaviorists are right about about equity premium, then you can beat the market simply by investing in equity. Now, perhaps it is difficult to distinguish between the behaviorist and rationalist position on equity premium, but I don’t think it’s impossible–certainly people do real empirical work on this topic.
To answer your question specifically, there are two good ways the problem you note can be solved. First, come up with an actual strategy and use standard statistics to demonstrate that it consistently beats the market. Second, come up with a class of market participants and show using standard statistics that this class consistently beats the market. Both methods have actually been attempted.
Now this is different from attributing normative characteristics (“efficientâ€, “correctâ€) to markets. The EMH would probably be a lot less controversial if it were called the “Unpredictable Market Heuristic (UMH) instead.
There is no normative content in EMH. Scientists use words different from philosophers (and both use words different from common language)–this is hardly a novel discovery.
As an aisde, does it follow from the EMH (or UMH) that hedge funds only make sense if fraud is involved? I think it does, do others agree?
Yes. This is a point I have made before. It is extremely weird that people think EMH exists as an academic theory for the benefit of Wall Street, when a major job of investment bankers is to convince people that they can beat the market. Real proponents of EMH say that the best thing to do is just find some passive mutual fund and leave it alone.
politicalfootball 03.01.10 at 9:27 pm
You know of studies indicating that people can predict relative long term prices? That in the long run the walk is non random?
someguy, as Walt points out, you need to read your link more carefully:
Either there is a conflict between the short-term voting mechanism and the long-term weighing mechanism, as he quotes Graham to suggest, or there isn’t. I think you are right, though, that in wanting to have it both ways, you are imitating Malkiel.
The joke goes something like “markets can stay irrational longer than you can stay solvent.” The point of that joke is to differentiate between efficiency and predictability. They are two different things.
Sure, there’s a considerably stronger case to be made for unpredictability of markets, rather than efficiency – but even there, the case is deficient, as Buffett demonstrates (and says, when asked.)
Blissex 03.01.10 at 9:28 pm
«I’m happy to believe that financial markets are the best of real-world mechanisms to allocate capital.»
«It is a very specific claim that arbitrage opportunities greater than the bid/ask spread do not persist.»
Even this is hopelessly naive — because “arbitrage”, “persist” and “spread” are not clear terms. For one thing, how do you define to which markets it applies? Just the NYSE? Does the Kabul SE qualify? And so on.
I think that the proper version of the EMH is more like “no trading strategy can make a specific individual trader outperform the market forever”, which may seem obvious (reversion to the mean), unprovable (except within models like Arrow-Debreu-Lucas) and profound at the same time
«When people use it to suggest that the market price is not merely a good or best available estimate of value, but that it is impossible even in principle to allocate capital more efficiently than the financial markets, they cannot be talking about the forms of EMH that stand up to the slightest scrutiny.»
But you both are right. The actual Arrow-Debreu-Lucas model does “prove” that “free” markets (suitably defined to ensure that is the result) are the best allocators of capital in every market and forever, and this means that the central truthiness of Economics (by which I mean Serious Economics as practiced the the likes of Mankiw) is also proven.
It is also true that the whole Arrow-Debreu-Lucas model itself can’t “stand up to the slightest scrutiny”, but that is insignificant to Economics — what matters is “proving” the central truthiness under *some* assumptions, then turn that into propaganda that “free” markets will realize “a Leibnizian claim that this is the best of all possible worlds”.
People like Mankiw do that all the time — they manufacture premises that imply a certain conclusion, “prove” that conclusion, and then turn the conclusion into a generally applicable propaganda claim. And the more mathematical and obscurely serious they look the better.
someguy 03.01.10 at 9:47 pm
politicalfootball,
Hmmmmm?
Walt said –
“There’s basically one fact that supports it: the fact that short-term stock prices are unpredictable, and then lots of facts that don’t support it, accompanied by long-winded explanations of why those facts don’t count.”
Malkiel somewhat concedes that in the short run stock prices might be predicted using various price models. Even if you could he doesn’t think that it would be worth it once you factored in transaction costs and he doesn’t think any such predictablility would last long.
In otherwords evidence exists that you just might be able to make predictions in the very short run. There is no evidence of that in the long run.
Hence –
“I am convinced that Benjamin Graham (1965) was correct in
suggesting that while the stock market in the short run may be a voting mechanism,
in the long run it is a weighing mechanism. True value will win out in the end.”
As for me I am not even sure if prices reflect return reflects value in the long term. But as long as they come fairly close the market is probably better than any other mechanism.
But I am open to change. How anyone can look at a graph of the business cycle and conclude that there is no theoretical way of improving resource allocation is beyond me.
The extremely tricky part is how do you do it?
John Quiggin 03.01.10 at 10:25 pm
Someguy, I don’t think it’s exceptionally hard to conclude that governments can improve on the allocation of resources by capital markets, both by investing in long-lived infrastructure assets and by countercyclical fiscal policy. After all, they did so on a large scale for several decades after 1945, and have continued to do so to some extent since then – whether the good outcomes were accidental can of course be debated, but the prima facie case looks pretty good.
The EMH in semi-strong and strong forms appeared to provide a proof that this could not be done, and that’s why it was interesting in policy terms. The question of whether you can profitably predict movements in the stock markets is both less interesting and harder to answer.
michael e sullivan 03.01.10 at 10:45 pm
“Even this is hopelessly naive—because “arbitrageâ€, “persist†and “spread†are not clear terms. For one thing, how do you define to which markets it applies? Just the NYSE? Does the Kabul SE qualify? And so on.”
Well, I wasn’t trying to be rigorous, I don’t have the subject familiarity to do that without a lot of re-reading.
My own general sense is that many big markets are relatively efficient in this sense (weak form efficient), but that many markets (and most smaller/thinner markets) are not. How many players there are and how much money moves relative to how much the largest potential participants control affects how efficient a market is. It has always seemed ludicrous to me that you could have every market be perfectly efficient outside of some hypothetical sandbox.
I doubt you could even say that the NYSE is completely weak form efficient, but I’m betting the prices of high-volume stocks mostly are.
Walt 03.01.10 at 11:00 pm
someguy, you have it backwards. The US stock market is unpredictable in the short run, but has some predictability in the long run.
someguy 03.02.10 at 2:57 am
John Quiggin,
Virtually no one, outside the Austrians, claim that resource allocation cannot be improved upon by countercyclical government policy.
Correctly or not a vast majority of folks believe that investment in things like education and basic science R & D results in improved resource allocation.
I guess the strong and semi strong versions of EHM weren’t really all that popular or didn’t exert all that much influence on policy?
For me the question – Are resources are being allocated efficiently between sectors and firms is very interesting and important question. EMH convincingly, if not conclusively, indicates yes.
Walt,
I will concede the probablity of an equity premium. [The evidence is good, I just have always had a personal gut feeling of no, and some doubts I never follow up on.] Not sure that means exactly x or y policy wise but again you could argue that markets misallocate on a macro scale. That they are too risk averse.
But see above about resource allocation between firms and sectors. Longer term your ability to predict relative stock prices is worse.
John Quiggin 03.02.10 at 3:44 am
“Virtually no one, outside the Austrians, claim that resource allocation cannot be improved upon by countercyclical government policy.”
That seemed to be the case until we actually needed fiscal stimulus. Then lots of Chicago economists suddenly found their inner Austrian.
But, as you suggest, the equity premium and its implications are crucial here. If you accept an anomalous equity premium, it follows that, for capital intensive infrastructure, governments are likely to do better than the market because their discount rate is closer to the social discount rate.
Ahistoricality 03.02.10 at 6:40 am
Then lots of Chicago economists suddenly found their inner Austrian.
Wait, there’s a difference? Hayek and Friedman represent different economic theories the same way that Coke and Pepsi represent different drink choices.
anarcho 03.02.10 at 10:47 am
“The ideology of Hayek was freedom”
Far from it, it was glorification of capitalism. Given a clash between freedom and capitalist/property, the “Austrians” always supported the latter. The 1920s praise of fascism by von Mises springs to mind, not to mention von Hayek’s position on Pinochet.
Not to mention, of course, their support for wage-labour (wage slavery, to use a popular anarchist term for that particular social relationship). Proudhon’s “Property is theft” and “Property is despotism” was lost on him — as was the obvious point that capitalist corporations suffer from much the same communications problems he suggested central planning in states have…
And as someone mentioned, anarchists had been making the case against central planning from the left long before von Mises and von Hayek did so from the right, and we did not exclude capitalist hierarchies from the critique…
piglet 03.02.10 at 3:17 pm
Bianca: “piglet, I think that’s mostly BS.”
And the basis for your claim is what? Quite frankly, your original claim about calculus not having changed in 200 years is so ignorant it didn’t merit the polite response I gave you and frankly there is no point to argue any more.
Walt 03.01.10 at 7:46 pm: I did take a look. Chapter 2 is named “on different degrees of smallness”, and chapter 3, “on relative growings” (never heard that term before). Page 19: “What does (dx)2 mean? Remember that dx meant a bit – a little bit – of x. Then (dx)2 will mean a little bit of a little bit of x.” etc. Is that how calculus is still taught in England or US???
As somebody with a mathematical background, I can read this stuff and translate it into modern concepts. The notation is close enough to modern notation. But if I were teaching calculus, I certainly wouldn’t want my students to see it.
Now this is a textbook. It is not an original source. It is an early 20th century representation of 17th century concepts. It is close enough in time for us to be able to follow. My original point however was that mathematicians don’t read original sources. Few mathematicians would ever want (or need) to read Newton, Leibniz, Gauss or Riemann or even early 20th century sources in the original, and they would be able to do so only with great effort.
THR 03.02.10 at 4:10 pm
Popper clearly didn’t understand Freud, and Quiggin’s rendering of the Oedipus complex above is not accurate, even as a snapshot.
someguy 03.02.10 at 6:24 pm
John Quiggin,
I think that as long as they in theory support monetary stimulus they are being intellectually consistent. Support for fiscal stimulus is not necessary.
Anyway I often enjoy your posts. Good luck with the book.
bianca steele 03.02.10 at 6:28 pm
@piglet: I assume you’ll expect the same generous treatment you dole out. I’m tempted to escalate hostilities, but maybe your English isn’t good enough to recognize one level of viciousness from another.
piglet 03.02.10 at 7:34 pm
“but maybe your English isn’t good enough” – no further comment…
John Quiggin 03.02.10 at 7:36 pm
@THR – this kind of comment is useless. If you think I’m wrong here, at least give a couple of sentences to explain way.
John Quiggin 03.02.10 at 7:37 pm
Piglet/Bianca I would prefer to avoid the kind of sniping seen above in my comments threads.
THR 03.02.10 at 8:05 pm
The classic examples arise from the popularised versions of Freudian psychology, centred on the Oedipus complex, named for the Greek tragic hero who unknowingly killed his father and married his mother. If a son hates his father, this is, obviously, evidence of the Oedipus complex. But, if he loves his father, this is explained as a repressed Oedipus complex. With rules like this, Freudian psychology can never be refuted.
You and Popper are confusing psychoanalytic theory with psychoanalytic interpretation.
Freud built a psychoanalytic theory that was expressly falsifiable at various points, and which he adapted in the face of new evidence. Various generations of psychoanalysts have done likewise. Furthermore, a good deal of Freud’s ‘metapsychology’ was couched in terms whereby it was clear that Freud considered his thoughts speculative, rather than definitive or ‘scientific’. Even Grunbaum, a staunch anti-Freudian, concedes these points:
http://www.escholarship.org/editions/view?docId=ft4w10062x&chunk.id=s1.3.37&toc.depth=1&toc.id=ch3&brand=eschol
Now, interpretation in psychoanalytic practice has an entirely different epistemological status. It’s true that, as far as interpretation goes, resistance on the part of the analysand could be regarded as confirmation of the analyst’s hypothesis. Different psychoanalytic theorists treat this problem in different ways, but generally speaking, interpretation is not reducible to truth claims (psychoanalysis is not, after all, CBT). Consequently, claims of Freud being unfalsifiable are either erroneous or facile.
Finally, the Oedipus complex has a little more to it than what you’ve suggested here. Admittedly, you were probably striving for concision rather than exegesis. However, whilst Freud thought that the Oedipus complex was a universal step toward psychosexual development, I don’t think he (or any other analyst) tried to reduce relations to one’s parents (for instance) solely in these terms. The Oedipus complex was never merely about loving/hating a mother/father, but about an extended dialectical dance of alienation, separation, triangulation, identification and various other things. Even those who are hostile to Freud (and there are many reasons to be) could at least give the guy a little credit.
politicalfootball 03.02.10 at 8:46 pm
I think that as long as they in theory support monetary stimulus they are being intellectually consistent. Support for fiscal stimulus is not necessary.
“Support” isn’t really the right measure and intellectual consistency isn’t the issue we’re discussing. The question is: Is fiscal stimulus possible? Some Chicagoans hold that it is not. This is a ludicrous belief.
bianca steele 03.02.10 at 10:09 pm
Sorry, John. (I should have ignored “piglet’s” first reply where the argument turned on a premise that “original ideas” can be “subject to . . . revisions,” whatever that means. I won’t do it anymore.)
piglet 03.02.10 at 11:38 pm
John: “Sniping”? Excuse me? If you want to name names, why not call it what it is – trolling.
John Quiggin 03.03.10 at 12:32 am
Piglet, call it what you want, but please stop.
THR, I’m not committed to Popper’s view of Freud, though I remain sympathetic to it. Note that I referred to popularized versions of Freud, rather than making claims about Freudianism as a whole. I’ll add a note linking to the source you gave and observing that debate continues as to whether Freudian theory is really as non-refutable as the popularised/interpretative version.
someguy 03.03.10 at 4:48 pm
politicalfootball,
A lot of intelligent and fair minded folks doubt the general effectiveness of fiscal stimulus policies.
Fom what I understand there is very little if any proof that any have worked. [Yea all macro is pretty much like that.]
You sell a bond and spend the money. Now if the person who bought the bond used the money they had in a mattress, because of fear and uncertainity or expectations of price declines, you win.
But if I cash in my stocks and buy bonds things aren’t as clear. If I cut back my spending and invest in bonds, that isn’t a clear cut win. If I cash in my savings and buy bonds that isn’t a clear cut win.
Say we employ an army to dig and than fill in holes. Some portion of otherwise idle capital was employed. That is great. But at the same time how much productive capital has been diverted into ditch digging?
I don’t think doubts about the effectiveness of this or that bit of fiscal stimulus or fiscal stimulus in general are all that ludicrious.
bianca steele 03.03.10 at 8:01 pm
@anarcho: same communications problems as corporate capitalism
Not only that. If Hayek is assuming direct communication of the kind necessary for this coordination is impossible, any project requiring more than one person is impossible, barring some kind of exploitation (which Hayek doesn’t seem worried about except in some situations), and that means it is immoral. This would be worrying if he had a good argument against the possibility of such communication, and even so, it seems to be an open question whether all communication is impossible or not, and if not where the line is to be drawn.
Bruce Webb 03.05.10 at 9:24 pm
Well all I can add on Hayek was that authorizing this literal cartoon version (18 panels) of The Road to Serfdom was close to criminal.
Freudians and Marxists at least have the defense that the Masters were not responsible for ‘Vulgar’ Freudianism or Marxism, Hayek put the most absurd and reductionist version of his belief in a form that was published in Look magazine and seemed to have formed the world view of many of those who know spend their time in Glenbeckistan.
http://mises.org/books/TRTS/
A professional economist friend of mine defended Hayek on the basis that his original work was much more nuanced and that he latter in life admitted that the actual record of the post-war Social Democracies in Europe showed that the cruder form of his argument was wrong, but by that time the damage was done by works like the one linked, the Bircher mentality was firmly entrenched in large parts of American society. Obviously Hayek doesn’t bear sole responsibility or even a preponderance of it. But still here it is.
(Note that the work is still hosted at Mises.org, and not in a why that suggests it is just some archival curiosity.)
bianca steele 03.06.10 at 1:46 am
From the comic: Let’s get a man who can make a plan work!
No, nobody would see an analogy to corporate management there.
Comments on this entry are closed.