I got more very useful comments on my section on the rise of the Efficient Markets Hypothesis, and I will get down to editing it before long. In the meantime, here’s my draft section on Implications of the EMH
At least in the draft, I’m following a standard structure: One chapter per dead/zombie idea, with sections on Beginnings, Implications, Failure and What Next? It seems to go OK for EMH, and we’ll see how it works for the others.
As before, comments of all kinds, and particularly pointers to (putative) errors, are most welcome.
The EMH: Implications
Although Paul Samuelson and Eugene Fama were jointly responsible for the formulation of EMH, they had very different views of how it should be interpreted.
Samuelson maintained the distinction, characteristic of the Keynesian-neoclassical synthesis between microeconomics, where a standard competitive market analysis was applicable, and macroeconomics, where a Keynesian analysis was needed. He argued that while tests of the EMH showed that financial markets were micro-efficient, the experience of bubbles and busts showed that they were ‘macro-inefficient’.That is, the efficient markets hypothesis works much better for individual stocks than it does for the aggregate stock market.
Samuelson’s position means, for example, that is difficult, if not impossible to outperform the stock market by examining the price history of individual stocks, or by poring over company reports. But it is possible to identify bubbles in the stock market as a whole, and to propose policies to stabilise asset markets. Writing in 1998, as the dotcom bubble was approaching its peak, Samuelson called for increased interest rates to deal with the ‘quasi-bubble’ on Wall Street. And, repeating Keynes response to the idea that rational speculators would always prevent such bubbles getting out of hand, Samuelson wrote
‘We have no theory of the putative duration of a bubble. It can always go as long again as it has already gone. You cannot make money on correcting macro inefficiencies in the price level of the stock market.’
But by the 1990s, Samuelson was in the minority, and his view that the EMH was consistent with macro-inefficiency commanded little support. The alternative interpretation, more logically consistent (if less consistent with reality), that the financial market price of an asset was not merely the best estimate of its value relative to other assets of the same kind, but was the best possible estimate, given available information.
This maximal interpretation of the EMH was espoused in academic works by Fama and his many students and followers and by the 1990s, accepted by most finance theorists, and nearly all policymakers. It was popularised by such writers as Thomas Friedman, whose The Lexus and the Olive Tree warned governments that the could not possibly hope to resist the collective financial wisdom embodied in the ‘Electronic Herd’ of global financial traders.1
The implications of the efficient markets hypothesis go well beyond financial markets. The EMH provides a case against public investment in infrastructure, and implies the macroeconomic imbalances, such as trade and current account deficits should not be regarded with concern and, provided they arise from private sector financial transactions, are actually both beneficial and desirable.
[anImage_0.tiff]
1 The term is an allusion to the ‘Thundering Herd’ a nickname for the iconic Wall Street investment bank, Merrill Lynch. In October 2008, Merrill Lynch was rescued from imminent collapse through a takeover by Bank of America, which was in turned bailed out to the tune of billions of dollars by the US government.
Asset bubbles and imbalances
The efficient markets hypothesis implies, in essence, that there can be no such thing as a bubble in the prices of assets such as stocks or houses. Such a claim, seeming as it does to fly in the face of centuries of experience, requires a fair bit of faith in the analysis underlying the EMH. But, in the euphoric atmosphere of the 1990s, such faith was in abundant supply.
The argument begins with the claim that, if a bubble in stock prices were indeed observable, speculators would sell the asset in question and go on to sell it short (that is, sell assets they did not hold in the expectation of being able to buy them later at a lower price). This would ensure that the price returned rapidly to the true market value.
But as Keynes had pointed out decades earlier, this argument only stands up if bubbles are short-lived, so that speculators are quickly vindicated. As Keynes said
“He who attempts it must surely lead much more laborious days and run greater risks than he who tries to guess better than the crowd how thc crowd will behave; and, given equal intelligence, he may make more disastrous mistakes. There is no clear evidence from experience that the investment policy which is socially advantageous coincides with that which is most profitable. It needs more intelligence to defeat the forces of time and our ignorance of the future than to beat the gun. “ Ch12 p140
More succinctly, in words widely (though apparently apocryphally) attributed to Keynes himself ‘the markets can stay irrational longer than you can stay solvent’. Lots of investors, from small-scale individual speculators to billion-dollar fund managers like Julian Robertson of Tiger Investments bet against the stockmarket bubble of the 1990s and lost.
A second argument is that, even if bubbles are real, there is little nothing policymakers can or should do to burst them. This was the conclusion reached by a number of central bankers who, unlike Greenspan, saw irrational exuberance in stockmarkets and property markets as a likely source of future trouble. A study by Borio and Lowe of the Reserve Bank of Australia pointed out the dangers of asset bubbles. However, in a policy environment where the only way of restraining speculation was to raise interest rates, they concluded that central banks could do little more than issue warnings.
It is worth noting, that although Australia experienced a serious land price bubble, lending standards were held to much stricter levels than in the US or other markets. It may be that the warnings of the Reserve Bank had some influence on the policy decisions of prudential regulators.
Finally, some supporters of the EMH argued, along lines first popularised by Austrian economist Joseph Schumpeter,, that even if bubbles lead to massively wasteful investment, they generate innovations that are beneficial in the long run (the phrase ‘creative destruction’ was widely used in this context).
Supporters of the EMH also engaged in a fair amount of historical revisionism to argue that famous historical examples of bubbles, from the Dutch tulip mania to the Roaring Twenties and beyond, were actually rational responses to new market conditions, often exaggerated in subsequent retellings. The Dutch tulip mania saw the price of a contract for a single tulip bulb exceed 10 times the annual income of a skilled craftsman Economist Peter Garber argued that these contracts were never fulfilled, and were little more than “bar bets”, and hence did not violate the EMH. Earl Thompson also defends the EMH, but argues that the prices were in fact rational in the light of the market rules prevailing at the time.
Implicit in all of these arguments was the conclusion that, in a well-developed modern market, with transparent dealing and with all parties subject to the scrutiny of auditors and ratings agencies, an irrational bubble could not possibly develop or be sustained. This conclusion formed the basis of financial policy in the decade leading up to the Global Financial Crisis of 2008.
The growth of the financial sector
In most of the simple models from which the Efficient Markets Hypothesis was derived, the financial sector did not exist as an industry. Financial markets were assumed to set the price of assets without any cost to the economy. As we have seen, the question of how those with information or forecasting skills could gain the returns they needed to justify their efforts was a significant theoretical problem for strong forms of the EMH.
Although no truly satisfactory analysis of the role of financial institutions in efficient markets was ever produced, advocates of the EMH came to accept that the cost of financial market transactions was equal to the value of the information they incorporated in asset prices. It followed that, as financial transactions were liberalised and the economy became ever more sophisticated, it was economically and socially desirable that the financial sector should grow.
And grow it did. In simple economic terms, the growth of the financial sector since the mid-1970s has been staggering. The financial services industry’s share of corporate profits in the United States rose from around 10% in the early 1980s to 40% in 2007 (Már Gudmundsson, http://www.bis.org/speeches/sp081119.htm), at a time when the profit share of national income was also growing strongly.
Volumes of financial activity grew at rates that defy any simple interpretation. The Bank for International Settlements has estimated the global volume of futures market transactions in 2007 at $1.6 quadrillion dollars, or about 30 times total world output. Notional volumes of outstanding derivative contracts are similarly massive. In the normal course of events, most of these transactions net out to zero, but even a small mismatch can produce losses (or gains) of many billions.
Along with all this, the income and wealth of those working in the financial sector grew massively. The salaries of financial sector executives reached four times the level prevailing in other industries, at a time when executive salaries in general massively outpaced growth in the incomes of ordinary workers. Such massive accumulations of wealth translate naturally into political power. Particularly in the United States, both major political parties were heavily influenced by Wall Street firms.
But the political power of the finance sector does not depend solely on command over economic resources. After the economic dislocation of the 1970s, the financial sector became, in perception and to some extent in reality, the most important guarantor of economic stability and prosperity. Governments sought desperately to gain and maintain the AAA ratings issued by agencies such as Moodys and Standard & Poors.
Private and public investment
Casual observation suggests that both the private and public sectors have difficulty in managing investments. Public sector investments, from the time of the Pharoahs onwards, have included plenty of boondoggles, white elephants and outright failures. But the private sector has not obviously done better. Waves of extreme optimism, leading to massive investment in particular sectors, have been followed by slumps in which the assets built at great expense in the boom lie unfinished or idle for years on end.
The EMH supports the first of these observations, but suggests that the second must be mistaken. Since public investments are not subject to the disciplines of financial markets, there is no reason to expect their allocation to be efficient.
By contrast, according to the EMH, private investment decisions are the product of an information system that is automatically self-correcting. The value given by the stock market to any given asset, such as a corporation, is the best possible estimate. If the managers of a given corporation make bad investment decisions, the value of shares will decline to the point where the corporation is subject to takeover by better managers.
The EMH, which enshrines the market price of assets as the summary of all relevant information, is inconsistent with any idea that managers should pursue the long-term interests of corporations, disregarding short term fluctuations in share prices. On the EMH view, the short-term share price is the best possible estimate of the long-term share price and therefore of the long-term interests of the corporation.
If the EMH is accepted, public investment decisions may be improved through the use of formal evaluation procedures like benefit-cost analysis, but the only really satisfactory solution is to turn the business over to the private sector. In the 1980s and 1990s this reasoning fitted neatly with the global push for privatisation, discussed in Chapter …
The EMH implies that governments can never outperform well-informed financial markets, except in cases where mistaken government policies, or a failure to adequately define property rights, leads to distorted market outcomes. If governments are better informed than private market participants they should make this information public rather than using superior government information as a substitute for public policy.
To sum up, the EMH implies that private enterprise will always outperform government, and that governments should confine their activities to the correction of market failures, and to whatever income redistribution is needed to offset the inequality of market outcomes.
Macroeconomic implications
There are also important implications for macroeconomic variables such as the balance of international trade. From a traditional Keynesian perspective, large imbalances in trade are a sign of trouble to come, since they will inevitably produce an unsustainable buildup of debt. Economists like Nouriel Roubini and Brad Setser were particularly vocal in warning of the trouble ahead for the US economy. I made the same point in an article published in the Economist’s Voice, with the self-explanatory title ‘The Unsustainability of the US Trade Deficit’.
By contrast, the EMH leads to the conclusion that economic analysis should be focused on asset values rather than on income flows. Observations of current income flows are informative only about the present, whereas asset values capture all relevant information about current and future income flows. An increase in asset values implies an increase in the present value of future income and therefore in the optimal level of consumption.
Once the EMH is accepted, there is no need to worry about imbalances in savings and consumption. International capital movements can be seen as the aggregate of a large number of transactions between ‘consenting adults’, buying and selling financial assets in markets which, according to the EMH, have already taken into account all available information about future risks. If a national government has better information, the appropriate response is not to act on it, but to release the information to the markets.
On the traditional, income-based view, by contrast, asset-based arguments are misleading and dangerous. By the time sentiment shifts in asset markets, the opportunity for an orderly adjustment will already have been lost. Advocates of the traditional view pointed to episodes of contagious panic in financial markets, most recently the Asian financial crisis of 1997.
The traditional response to macroeconomic imbalances such as trade deficits was the adoption of contractionary monetary and fiscal policies aimed at reducing demand for imports and at forcing domestic producers to seek export markets as a response to lower demand at home. The resulting ‘stop-go’ policies caused substantial suffering and economic dislocation.
Once it is realised that sustained macroeconomic imbalances ultimately reflect financial market failures, this response can be seen to be inappropriate, as can the benign neglect associated with the consenting adults view. The appropriate response is to intervene in financial markets to restrict the unsound lending practices that drive the growth of such imbalances.
{ 46 comments }
Michael Harris 07.25.09 at 8:39 am
Quick comment: this para needs a bit of tidying up.
“Along with all this, the income and wealth of those working in the financial sector grew massively. The tens of millions paid annually . Such massive accumulations of wealth translate naturally into political power. Particularly in the United States, both major political parties were heavily influenced by Wall Street firm.”
Kevin Donoghue 07.25.09 at 8:46 am
This looks as if it as written in a hurry, with sentences such as “If a national government”
As to the substance, it seems to me that you lump the EMH together with Panglossian attitudes to all manner of social ills. I’m sure it’s true that lots of EMH fans do think inequality is no big deal, but I’d be surprised if Fama for example actually tried to argue that the EMH shows we don’t need nationalised healthcare, or anything like that. It’s fair enough to treat the EMH and rational expectations generally as components of an ideology but it’s not fair to suggest that belief in the EMH implies belief in the whole package. This post reads as if you are suggesting just that. Fairness aside, I think you could make it more interesting by arguing along the lines that the EMH has a natural appeal for conservatives, but some of the ‘Austrian’ contingent detest it since it seems too fatalistic and leaves little scope for enterprise. The undead ideas you are writing about are all related but there’s a danger of making it seem like they are all they same idea.
Katherine 07.25.09 at 9:45 am
But by the 1990s, Samuelson was in the minority, and his view that the EMH was consistent with macro-inefficiency commanded little support. The alternative interpretation, more logically consistent (if less consistent with reality), that the financial market price of an asset was not merely the best estimate of its value relative to other assets of the same kind, but was the best possible estimate, given available information.
I think this paragraph is missing the conclusion/explanation: that this alternative interpretation is that the EMH is inconsistent wth macro-inefficiency? Or that if the EMH is correct, this means there can never be macro-ineffieciency? Both?
It may seem obvious to the writer or an economics expert, I suppose, but an extra line of clarification would help me.
Katherine 07.25.09 at 9:50 am
But as Keynes had pointed out decades earlier, this argument only stands up if bubbles are short-lived, so that speculators are quickly vindicated. As Keynes said
“He who attempts it must surely lead much more laborious days and run greater risks than he who tries to guess better than the crowd how thc crowd will behave; and, given equal intelligence, he may make more disastrous mistakes. There is no clear evidence from experience that the investment policy which is socially advantageous coincides with that which is most profitable. It needs more intelligence to defeat the forces of time and our ignorance of the future than to beat the gun. “ Ch12 p140
Again, I may be being a dunce, but I don’t see how the quote follows on from the sentence it is supposed to be illustrating. You have talked about bubbles and whether they are shortlived, then the quote refers to someone attempting “it” (what?), and socially advantageous investment policies vis a vis profitability. I don’t quite see the link.
Katherine 07.25.09 at 9:53 am
By the way, throughout this section it’s unclear which version of the EMH you are talking about. When you use the phrase, “Supporters of the EMH also engaged in a fair amount of historical revisionism to argue that famous historical examples of bubbles”, for example, I think you must mean the strong EMH, since in the previous section you indicated your own support for/acceptance of the weak form, yes?
John Quiggin 07.25.09 at 10:21 am
Michael and Kevin, I’ve made some edits to respond to your points, removing the snark to which Kevin objected.
Katherine, I mentioned in the previous section I was focusing on strong EMH from now on – this is a problem with doing it section by section I guess.
On #4, I’ll spell out that “it” is betting against bubbles. If this could be done profitably, then bubbles must soon burst.
#3 is a bit trickier. The Fama interpretation of EMH precludes macro efficiency. As I hint, this is closer to the “spirit” of EMH than the Samuelson view, but the Samuelson view preserves a lot of the theory and evidence supporting EMH.
Michael Harris 07.25.09 at 10:21 am
John, a thought that occurs based on the discussion of macro implications at the end.
Macro has been taught (esp. at undergrad level) with its focus squarely on the income flows. As you point out, debates about current account deficits have been discussed in “rational actor” terms (e.g. the Pitchford “consenting adults” Line on the CAD in Australia) which has market efficiency underpinning it. Hence the connection to the idea of assets being priced efficiently as per the EMH. But there’s not a lot of attention paid to asset values and balance sheets and their implications for macro performance in standard texts, at least as I’ve experienced them.
Finance tends to deal with asset pricing and so on, but it’s not well connected with macro models from where I sit. It may be in “monetary economics” texts and courses because that’s where the obvious overlap is, but I’m not familiar with that area very much.
But it seems to me that modern undergrad macro does have to move beyond the textbook-Keynesian focus on income flows, or the fancied-up dynamic macro stripped down into digestible form, and get its head around asset values, debts and balance sheets. If there’s a where-to section of the book (well, your “What Next?” I guess), that seems something worth pursuing.
Michael Harris 07.25.09 at 10:30 am
Katherine @ #4.
In case it still isn’t clear (although it may be by now), the EMH requires that bubbles be self-corrected, and that the self-correction comes via people betting rationally against the bubble because they know it is a bubble and hence will have to burst.
But if the bubble has strong enough momentum, betting “rationally” against it will send you broke before you can burst the bubble and claim your prize. You bet short, hoping to buy assets cheaply when it bursts and sell for your higher contracted price, but if the bubble doesn;t burst and the price has kept inflating, you’re screwed.
The EMH also implies that the markets have internalised everything efficiently, so that “socially advantageous” investments are also the privately profitable ones.
John Quiggin 07.25.09 at 10:55 am
Michael, I agree, though with a bit of a twist. In general, any situation where a focus on flows gives different conclusions to a focus on asset values is likely to be problematic. In particular, when flows appear unsustainable but asset markets don’t reflect this (eg when a country with a chronic trade deficit has a strong currency), I’d conclude that asset market failures may be driving unsustainable flows.
Akshay 07.25.09 at 11:39 am
Hi, Could you explain
But the political power of the finance sector does not depend solely on command over economic resources. After the economic dislocation of the 1970s, the financial sector became, in perception and to some extent in reality, the most important guarantor of economic stability and prosperity. Governments sought desperately to gain and maintain the AAA ratings issued by agencies such as Moodys and Standard & Poors.
I mean, if finance becomes a guarantor of economic stability, surely this is because of its command over economic resources? If you mean that its power comes from a) campaign financing/wealth and b) the threat of capital flight, I think it is clearer if made explicit.
Michael Harris 07.25.09 at 11:40 am
John, sounds first-approximation right to me.
The thing that drove the issue home to me was more domestic than international. It was that I’d leanred about central banking and monetary policy as being about controlling prices. When it became apparent that consumer prices were more or less under control but that asset prices were not, I realised that I couldn’t look to undergrad macro to tell me how to think my way through it.
The CAD/exchange rate issue is another, international, one where the asset price signals a problem with the flows.
P O'Neill 07.25.09 at 1:29 pm
You could specifically cite the “Lawson doctrine” for the idea among policymakers that current account imbalances could be ignored as long as they were on the private account.
Note also that Garber, cited re Tulipmania, worked extensively for Deutsche Bank in the latter glory years for the City (which roughly correspond to New Labour).
Ceri B. 07.25.09 at 2:30 pm
The EMH, which enshrines the market price of assets as the summary of all relevant information, is inconsistent with any idea that managers should pursue the long-term interests of corporations, disregarding short term fluctuations in share prices. On the EMH view, the short-term share price is the best possible estimate of the long-term share price and therefore of the long-term interests of the corporation. This is the deepest a-ha of the project so far for me, anchoring an observed behavior to its bad theoretical foundation.
King Rat 07.25.09 at 3:41 pm
One of the things that the news does that you should not is use unattributed arguments. One of the reasons I got frustrated with the news and more or less have stopped watching it are statements like: “In a move that some are calling disingenuous…” without ever naming who specifically are those “some”.
Stuff like “supporters of EMH argued …” should say who exactly argued that, whether it’s an individual or an organization or whoever. If you don’t, readers will be frustrated at not knowing who makes these arguments. ANd the people who made the arguments will say “I didn’t say that” and claim you are building straw man arguments.
Billikin 07.25.09 at 4:06 pm
“The EMH, which enshrines the market price of assets as the summary of all relevant information, is inconsistent with any idea that managers should pursue the long-term interests of corporations, disregarding short term fluctuations in share prices. On the EMH view, the short-term share price is the best possible estimate of the long-term share price and therefore of the long-term interests of the corporation.”
I do not think that that follows. In fact, if the managers of a corporation decided not to pursue its long-term interest, and published that fact — not in those terms, of course –, doesn’t the EMH imply that their share prices would immediately drop? Or does the EMH imply mind-reading?
Billikin 07.25.09 at 4:19 pm
Michael Harris: “The EMH also implies that the markets have internalised everything efficiently, so that “socially advantageous†investments are also the privately profitable ones.”
I don’t think so. In terms of the society as a whole we are talking about an n-person game, and not all of the stake-holders participate in the markets. If I am a poor person, and the success of corporation X harms me, I cannot just invest in corporation X to offset that harm. Does the EMH imply that everyone participates in the market?
Mike Bruce 07.25.09 at 5:59 pm
It might be helpful to have some discussion about why one asset price is “better” than another.
It seems intuitively obvious that prices in a bubble are “too high”, but I don’t have a good technical sense of what that means.
Neel Krishnaswami 07.25.09 at 7:23 pm
Argh! I know everyone uses this comparison, but please don’t join the bandwagon — it’s a terrible comparison, since output and trading volume are different units, like volume and pressure. More illuminating comparisons, which would convey a better sense of scale, would be to either a) the volume of derivatives trading compared to regular stock markets, or b) outstanding derivatives contracts to outstanding insurance contracts in things like car, fire, or health insurance.
If you want to be funny, you could also compare trading volumes to the total amounts bet in Las Vegas. E.g., Vegas casinos took in 10.6 billion dollars in gross gaming revenue in 2006, which assuming a 1% house edge implies that casino gambling volume was roughly a trillion dollars. So the futures markets were about 1600x the size of Vegas’s casinos.
EdSez 07.25.09 at 8:58 pm
I think that the description of the beliefs held by the EMH flock is too dryly descriptive, without any assessment of these faith-based beliefs. I think that a phrase such as ‘magical thinking’ should be sprinkled somewhere near the sentence
“By contrast, according to the EMH, private investment decisions are the product of an information system that is automatically self-correcting.”
I gather that you feel such commentary is putting your thumb on the scale. However, since the title of the book gives us the expectation of an argument, it would be helpful if you highlighted particularly absurd notions in the theories you intend to discredit.
bbartlog 07.25.09 at 11:27 pm
I always thought that the title (with subtitle) of your blog was a sufficient counterargument to the EMH. But I guess it always sounded stupid to me.
Hix 07.26.09 at 1:15 am
Fama himself certainly holds right wing extremist views and does apply some kind of emh view on every aspect of life. To lump this personal views, which are no doubt more likely for emh type economists entirely together with emh is a mistake. Emh can well be applied very narrow just to some parts of euqity market. That is how emh makes most sense, and in that context, emh can very well lead to rather leftish conclusions.
Hix 07.26.09 at 1:23 am
For those that doubt Fama and French have some rather extreme idears take a look at their blog for example that post:
http://www.dimensional.com/famafrench/2009/03/qa-capitalism-under-obama.html
Michael Harris 07.26.09 at 1:40 am
Billikin @16:
Katherine @ 4 was asking about the quote John used from Keynes, which included:
“There is no clear evidence from experience that the investment policy which is socially advantageous coincides with that which is most profitable.”
I was just wanting to make the point that a simple markets-work view of the world would imply via the EMH that asset markets are informationally efficient and (via some handwaving) invisible-hand-efficient too, in the sense that the “right” (social-wealth-maximising) investments are the ones undertaken by the private sector because they’re also the most privately profitable.
If as a poor person I was harmed by the success of Corporation X, that is presumably as a result of some kind of externality, which if unchecked does break the link between informational efficiency and invisible-hand efficiency. But that wasn’t the point I was trying to explain to Katherine in the context of Keynes’ quote. Admittedly, I was trying to keep it simple, which always presents traps.
Billikin 07.26.09 at 2:12 am
Michael Harris @ 16:
Thanks for the explanation. As a layman when somebody says “socially advantageous” I think of externalities. ;)
Abby 07.26.09 at 2:43 am
To Billikin’s point (#15) and your original:
“The EMH, which enshrines the market price of assets as the summary of all relevant information, is inconsistent with any idea that managers should pursue the long-term interests of corporations, disregarding short term fluctuations in share prices. On the EMH view, the short-term share price is the best possible estimate of the long-term share price and therefore of the long-term interests of the corporation.â€
I think this is a little bit backwards. The implications of EMH-strong are that if a manager were to attempt to pursue anything other than the long-term interests of the corporation in an attempt to boost the stock price in the short term, the market would see through the effort and the price would fall (since the long-term interests were no longer being maximized). Now, empirically, this is patently not true–it is perfectly possible to inflate the stock price through short-term efforts, which I think provides evidence that EMH-strong is not a theory that fits how the world actually works.
There is a huge amount of academic work that isn’t clear about this (I am mostly aware of swathes in the accounting literature, but I expect it occurs frequently in economics and finance too). They assume, when convenient, that insiders can manipulate stock prices or that stock prices over-react to trivial news (i.e. missing analyst expectations by a penny or something). But then some other part of their argument rests on EMH-strong. It drives me nuts.
Hix 07.26.09 at 7:33 am
I was just wanting to make the point that a simple markets-work view of the world would imply via the EMH that asset markets are informationally efficient and (via some handwaving) invisible-hand-efficient too, in the sense that the “right†(social-wealth-maximising) investments are the ones undertaken by the private sector because they’re also the most privately profitable.
No. The market would still overproduce products for rich people compared to the social optimum and underproduce the ones for the poor.
Michael Harris 07.26.09 at 9:29 am
Hix
“Efficiency” and “distribution” are nominally distinct, but every efficienct outcome is conditional on whatever the income distribution happens to be. If you change the income distribution (thought-experimentally), the equilibrium outcome will change. Each outcome can satisfy Pareto efficiency subject to the usual conditions.
If people have, for example, other-regarding preferences or preferences about income distributions per se then your comment has some relevance, but not really in the conetxt of the point I was making regarding the Keynes quote.
Kenny Easwaran 07.26.09 at 9:56 am
Maybe it’s just my background in logic, but these phrases again sound awkward: The alternative interpretation, more logically consistent (if less consistent with reality), Consistency is an all-or-nothing thing, and I think of coherence as the natural counterpart that comes in degrees (though it’s not quite the same). Coherence also has the nice feature that it suggests you’re not talking about pure logic here, but some more substantive notion. So I would say “more internally coherent (if less coherent with reality)”.
Max 07.26.09 at 10:16 am
Nobody knows what the sustainable growth rate is for an economy (or part of one). So obviously, the market will always be wrong, since it isn’t a magic box that predicts the future. That’s true even if the market is perfectly efficient and perfectly aligned with the most enlightened expert opinion.
John Quiggin 07.26.09 at 10:17 am
I’ve adopted quite a few comments above, citing the Lawson doctrine (which, I was a little surprised to discover, predates the corresponding debate in Australia by a year or so), comparing derivatives to the value of world stockmarkets, clarifying some points about the political power of finance and some minor points.
Still thinking about consistency and coherence. I think in ordinary language, consistency comes in degrees, and “coherent with reality” is off-key.
Also, I need to fix a position on attribution and citation. I may post on this because it raises a lot of questions.
Katherine 07.26.09 at 12:29 pm
Katherine, I mentioned in the previous section I was focusing on strong EMH from now on – this is a problem with doing it section by section I guess.Katherine, I mentioned in the previous section I was focusing on strong EMH from now on – this is a problem with doing it section by section I guess.
John, fair enough, and I remember you said that, now you’ve reminded me. This might be a recurring problem though, whether reading blog posts or reading a book. Would it perhaps be useful to adopt a clearer nomenclature? WEMH, SSEMH and SEMH, for example – just suggestions, and requiring minimal extra letters.
Katherine 07.26.09 at 12:38 pm
Oh and thank you Michael Harris and John for further explanations. I do understand the points made – I hope my questions will be useful for clarifying the text.
Michael Harris 07.26.09 at 12:42 pm
Attribution and citation: I agree it’s a central issue to get right. Too little and it’s just (the equivalent of) a collection of expandecd newspaper articles with assertions and broad descriptions intermingled, and prone to be being dismissed as straw-man positions.
It obviously takes more work to include lots of citations and quotes to back up the general descriptions in the main chapters, and it can block the flow of the argument. The compromise route might be to use the endnotes approach of popular history books, where a sentence like “But by the 1990s, Samuelson was in the minority, and his view that the EMH was consistent with macro-inefficiency commanded little support” has an endnote that gives the claims explicit support.
John Quiggin 07.26.09 at 8:38 pm
I really, really hate endnotes.
Michael Harris 07.26.09 at 9:33 pm
Judging by the link, John, they hate you!
John Quiggin 07.27.09 at 8:57 am
Indeed, but I still really, really hate endnotes
Tracy W 07.27.09 at 12:00 pm
Casual observation suggests that both the private and public sectors have difficulty in managing investments. Public sector investments, from the time of the Pharoahs onwards, have included plenty of boondoggles, white elephants and outright failures. But the private sector has not obviously done better
Hold on Dr Quiggin. Earlier you were selling this book as ideas that were dead. You in particular were stating that “The EMH says that financial markets are the best possible guide to the value of economic assets and therefore to decisions about investment and production.” (see https://crookedtimber.org/2009/07/13/bookblogging-keynes-and-the-efficient-markets-hypothesis/#more-12028), and implying (by your use of headers) that this is a dead idea. For the purposes of brevity, I am going to label the hypothesis you claim is a dead idea as “FMBFIP” (short for “Financial Markets are Best For Investment and Production decisions”, I know this is not a catchy name, I also had problems as a child naming my toys).
Now for an idea to be dead, I don’t think it’s enough to say that a number of people in the relevant profession disagree with it, I think this implies that it’s not an idea that the vast majority of members of the profession is going to defend (excluding of course the odd eccentric, in a world where breatharians can claim that we don’t need food or water to live, I think that setting a target of no people defending an idea is too high a target for “disproof”). For you to disprove FMBFIP you would thereby need to prove that some other way of making decisions is better. Fond as I am of mathematical proofs, I am willing to make due allowances for the use of the words prove and disprove here, but I think you are ridiculously overreaching here in claiming that FMBFIP has been disproved. You would need to show, in a way that no one else can rationally argue with, that some other system can do better and show it very convincingly. You haven’t. You’ve launched a round of criticism of financial markets, but you haven’t even stated what system you think can do better at guiding investment and production decisions. Implicitly I suppose you think of government, but clearly not all governments are better at investment and production decisions than markets (see the economic performance of the centrally planned states versus the economic performance of the capitalist states).
Of course, not actually identifying which policy-making system you think does better than FMBFIP is convenient for rhetorical purposes, it’s hard to point out flaws in something when you don’t even know what it is, and if most of your readers are already inclined to agree with you they won’t notice the massive missing hole in your argument (I know I am bad at identifying flaws in arguments I am emotionally inclined to agree with, this is part of why I hang out on Crooked Timber). But government policy makers need to either do something in particular or not do anything, and failing to even identify whatever system you think is better than FMBFIP means that you’re leading yourself wide open to a comparative insitutions analysis response.
Obviously, merely naming is not enough, you also have to show it’s better, and much more convincingly than saying “not obviously done better”, which would merely be a reason for openmindedness about FMBFIP, hardly a reason to conclude that it’s a dead idea.
You’ve just waged another counter-argument in the ongoing debate in economics about the proper roles of government and markets, you haven’t remotely managed to disprove FMBFIP or to prove that it’s a dead idea. My advice is to rename your book “Economic ideas that John Quiggin doesn’t think are right.” (Full disclosure, my advice is not motivated by a desire to maximise sales of said book).
Tim Wilkinson 07.27.09 at 12:41 pm
I hate endnotes if they are hard to find, which they usually are, because on the originating page you have the chapter title and a page number, while in the endnotes you have instead a chapter number. When endnotes have ‘Notes to pages xx-xx’ rather than ‘notes to chapter xx’ at the top of each page (or worse, interspersed in the text), or when the numbers are unique throughout the book rather than each chapter, I don’t mind them so much. Even if the chapter title as well as number were given in the notes, that would be some help, though still annoying. The way things are normally set up is backwards – as though readers start from the endnotes and want to find the text they relate to.
Long footnotes can be pretty annoying though, too. Perhaps footnotes for everything, which in turn contain endnote refs,with a page number, for (part of) those comments which are too long to be sensibly put in a footnote. Sounds Byzantine but could easily be made to work (long endnotes being referred to as appendices or something). It navigates between the Scylla of ‘Why is this endnote so inaccessible – are you not actually expecting anyone to read it?’ and the Charybdis ‘Why is this footnote cluttering up my page when it clearly isn’t important enough to go in the text?’.
BTW – ‘ibid‘ and ‘op. cit.’ used to refer back between footnotes are REALLY annoying – you endup leafing back through the previous footnotes trying to find the originating citation. Either repeat it or refer to a bibligraphy.
But then neither of these decisions are made by authors I don’t suppose, but instead decided on by people whose job it is to make books as clear and usable as possible. So a big ‘well done’ to them.
Though endnotes at least don’t disappear completely, like hyperlink targets…
Tim Wilkinson 07.27.09 at 1:10 pm
JQ – To my eye Tracy’s remarks @37 are off-target since she confuses pointing out that non-weak EMH is commonly cited in support of the FMxxxx, and is untrue, with asserting that the latter is disproved. Further, the remarks about whether an idea is ‘dead’ miss the point since (your claim is that) the ideas have been ‘killed’ – i.e. objectively discredited – but refuse to ‘die’ – i.e. be abandoned by theorists and especially practitioners.
But the remarks do I think indicate a certain lack of clarity [here and in the sequel, please insert ‘IMHO’ as required by local standards of politeness] which may well originate in the sources rather than just your exposition.
In addition to Tracy’s quote, you say that according to EMHs (or according to various people who advocate them?):
the short-term share price is the best possible estimate of the long-term share price
and
the stock price will be the best available estimate of the future value of the stock
and
the financial market price of an asset was not merely the best estimate of its value relative to other assets of the same kind, but was the best possible estimate, given available information.
The first, unless fairly vacuously true, i.e. all ways of estimating these things are all completely useless, would seem to be in tension with the EMH itself. Not sure if the other two are the same (as the first or each other), or thought to be by whomever, but it’s a bit unclear where the conclusions come from and what they amount to.
And to my inexpert eye, an increase in asset values implies an increase in the present value of future income and The EMH implies that governments can never outperform well-informed financial markets – the latter having nothing to do with stock valuations of any kind – seem unwarranted.
The (or a) problem you don’t explicitly identify, and which I think underlies the unclarity that I perceive, is that these EMHs don’t tell anyone anything useful at all. A bit like the assertion that there is no free will, they give no guide to action (except insofar as they incorporate more specific directions such as ‘don’t use charts’). And they are not really about ‘correct’ market valuation; they are about the actors being effectively identical so that no-one can get ahead (with the corollary, I suppose that no-one can be left behind).
On the standard nc approach as I understand its logical consequences, there isn’t any such thing as correct market valuation – market valuation reduces without remainder to market price. In a market with any speculative element, price determination (insofar as there is ‘elbow room’ for indistinguishable price-taking traders to do such a thing) is both a non-trivial function of the expected behaviour of the market, and (under the ‘revealed preference’ behaviourist method/metaphysic) can only be done as a piece of such behaviour. There is no Archimedean point from which to assess such a valuation.
Or anyway something like that – without doing (or possibly reading) some substantial, intricate and convoluted philosophical work (and porbably rather a lot of economic theorising), that’s the best I can do at present.
The assertion The efficient markets hypothesis implies, in essence, that there can be no such thing as a bubble in the prices of assets such as stocks or houses may be true (although NB I don’t think the ‘in essence’ bit sounds right, and ‘implies in essence’ sounds very odd), but if it’s taken as saying ‘there is indeed a possible situation which would count as a bubble, but it won’t happen because of EMH’ it seems to overstretch. I notice that when the EMH-based argument against bubbles is supplied, it deals only with bubbles that are observable – but what is there to observe?
If the structure of the market permits a speculative element in trading decisions, then the expected behaviour of traders is part of the input to trading decisions. When a bubble bursts, firms really do plummet in (not-merely-subjective) value. The firm whose shares drop has suffered a catastrophe every bit as real as an uninsured natural disaster. It has lost market confidence, or whatever you want to call it. To each (price-taking?) trader, market confidence is an objective external reality. To adapt a familiar formula, the market confronts the trader with the objectified essence of his own trading activity.
More prosaically, market valuations cannot be inaccurate because there is nothing for them to be accurate about (remember we are still in free-market economics world here.) The only thing that could be wrong with them is that they turn out later to have been involved in an insufficiently profitable deal. Of course, even to get that far the old bottom drawer will need to be opened up and a few of the assumptions relaxed – because there can be no deals if everyone has the same valuation and the same preferences, and there can be no (supernormal) profit on standard assumptions (nor on the EMH).
SUGGESTION:
That’s one of the reasons why I’d like to see the discussion pegged much more closely to the shameful secrets kept in that bottom drawer (or maybe the attic – some kind of Dorian Gray analogy surely possible here but can’t quite get it right). The EMH like much else is formulated on the back of tons of such assumptions, but no-one ever seems to get round to actually pointing out which ones, and which ones are unrealistic. So even just as an exercise I’d like to see a list of the standard idealising assumptions on which these EMHs are based, and a relaxing of each one to see if the EMH still holds. This would help to remove the baffling nebulousness of the concepts and assertions involved (I suspect that economists find these less baffling mainly because they’ve learned not to feel baffled.)
As a non-economist who has been subjected to some teaching in the subject, I would stress – you cannot explain any part of nc economics without explicitly describing these assumptions, and I would expect them to be implicated in many ways in all of the doctrines you discredit – which might add a unifying theme, too.
Personally I think there should be a checklist of standard assumptions/idealisations against which every economic assertion must routinely be checked. It should then be audited and certified as robust or not against the relaxation of each of the assumptions. All transactions and decisions are instantaneous – non-robust; no barriers to entry – non-robust; omniscience/unlimited reasoning power; non-robust, etc. A card with big green ticks and (lots of) red crosses would be good. More detail of how far the assumptions can be relaxed salva veritate could be added once this salutory practice makes its effects felt.
TITLE:
Being in the ‘Zombies are a bit too comic/pop’ camp, I’d like to recommend a different title based on superstition or magic or unreality, but can’t think of a good one off hand. Maybe a ‘new enlightenment’ type of idea, with allusion to the virtus dormitiva and theological elements or pre-enlightenment science. Possibly a play on the unscientificness of the failure to properly separate ideal from empirical theory: ‘Dismal non-science’ or something (though obviously something rather a lot better). Something about a twilit world between idea and reality (an Eliot reference might be nice!). Or, moving away from the sublime, something about elephants in the room?
kevincure 07.27.09 at 1:16 pm
I meant to note this in response to a comment in the last EMH post, but EMH, even in a strong form, does not imply that bubbles are impossible, or anything similar. There is an enormous literature on “rational bubbles” – roughly, they can exist but are difficult to generate. Statements that EMH says otherwise are simply incorrect.
Further, at least in the US (and at least while I was working at the Fed) the “don’t burst asset bubbles” argument was never made in terms of “asset bubbles are impossible” but rather in terms of “it’s too hard to know which asset price rises are bubbles and which aren’t, and it’s rather presumptuous to assume we know more than the market”. There is still interesting commentary on the latter, but the chapter above seems to imply the former. Most of Greenspan’s research from this era is online, so his exact views here are pretty easy to acquire.
Salient 07.27.09 at 1:54 pm
Indeed, but I still really, really hate endnotes
Surely you infinitely jest! :-)
Tracy W 07.27.09 at 5:29 pm
JQ – To my eye Tracy’s remarks @37 are off-target since she confuses pointing out that non-weak EMH is commonly cited in support of the FMxxxx, and is untrue, with asserting that the latter is disproved.
Can you please tell me what FMxxxx refers to? I may or may not be confused about Quiggin’s arguments, but I certainly am confused about your statement here.
Further, the remarks about whether an idea is ‘dead’ miss the point since (your claim is that) the ideas have been ‘killed’ – i.e. objectively discredited – but refuse to ‘die’ – i.e. be abandoned by theorists and especially practitioners.
I do not miss the point. I am arguing that Quiggin has not objectively disproved (not merely discredited which merely means to damage something, Quiggin’s claim is that the idea was dead, which I am translating as “disproved”) what I am calling FMBFIP.
The labour theory of value is dead, the Philip’s Curve is dead, but, at least based on Quiggin’s arguments here, FMBFIP is not dead.
The EMH like much else is formulated on the back of tons of such assumptions, but no-one ever seems to get round to actually pointing out which ones, and which ones are unrealistic.
The reason that no one ever seems to get around to pointing out such assumptions is that the EMH was never formulated on such assumptions. As Fama says in his 1975 article (http://www.ekonometria.wne.uw.edu.pl/uploads/Main/1970.Fama.EMH.pdf, see page 8 of the pdf), the EMH was formulated as an explanation of empirical findings that stock prices were a random walk. And Fama himself says that he never believed the strong form of his EMH and was unsurprised that it was disproved. I know that Quiggin has been claiming that EMH was derived from various assumptions, but he’s never produced this claimed derivation, and I’ve never seen it elsewhere.
In other words, you are never going to see a list of the standard idealising assumptions on which Fama’s EMH are based because such assumptions never existed.
John Quiggin 07.27.09 at 11:06 pm
I’m aware of the rational bubbles literature, and I regard it as incompatible with the EMH. As Tracy’s repeated comments make clear, there is ultimately no way of resolving disputes of this kind, since there is no official or generally accepted authority laying down a precise position. Most obviously, as pointed out above, Samuelson and Fama have very different views. And as others have pointed out, if you take Fama as sole authority you have the problem of determining which of his views are derived from his interpretation of the EMH, and which are just general rightwingery.
Also, in my view, the idea that you can never tell when there is a bubble is essentially the same as saying there is no such thing as a bubble, merely occasions when market participants make errors. As will be pointed out in the next section, the history of the past two decades suggests that bubbles can indeed be observed in the process of formation and constrained to some extent.
Neel Krishnaswami 07.28.09 at 1:37 am
This seems like too binary a dichotomy you’re proposing. Markets come in degrees of efficiency, and identifying a bubble is easier the less efficient it is, because the less like a random walk prices will be. For example, in a very inefficient market such as real estate, you can easily identify bubbles with very simple autoregressive models — the real estate bubble was blatantly visible in the Case-Shiller indices. But the argument for contemporaneous dot-com boom was much less compelling, because the statistical tests we needed to identify the bubble from the noise were much more complex and hence much more contestable. (Unless this is your point..?)
Bruce 07.29.09 at 3:57 am
I like the distinction between micro-efficiency and macro-efficiency. It frames the discussion in a way I haven’t heard before and helps clarify the argument. However I have to agree with Tracy W that in order to dispute the notion that financial markets are do not provide the best estimate of value, an alternate model must be proposed.
I will be interested to your the evidence that bubbles can be identified as they are developing, because that is the only alternate method of valuation you have yet made reference to, therefore your entire argument seems to rest on this. I’m wondering whether this will be fundamental analysis which seems to be highly reliant on the particular asset (and thus difficult to replicate for bubbles in different types of assets such as tulips, technology and property), or technical analysis that just looks at the level of increases.
For example, you mention the Australian land price bubble. Australia shared the massive increases in prices with the rest of the world, but has had a very soft landing, with maybe a 6-7% drop. Is Australian land still in a bubble?
The ABS price index for established homes only seems to go back to 2002, but it shows real prices have increased 64% since then, even with the recent drops. I am interested to know what methods you would recommend the Australian government use to decide whether to deflate this potential bubble.
John Quiggin 07.29.09 at 8:11 pm
Neel, I agree that the dotcom bubble was hard to detect by the kinds of statistical methods that work where weak-form efficiency is violated. But it was easy to tell that the NASDAQ as a whole was a bubble – the prices being paid could only be justified if every single dotcom was going to be a spectacular success, and (given the rise in other stocks), their success came at no cost to old-economy companies.
Bruce, I still think there is a bubble in Australian land, and I’m hoping that the gradual withdrawal of first homebuyers grants etc, combined with a very slow increase in interest rates, will produce a gradual deflation.
Comments on this entry are closed.