Efficient Market Hypothesis - EMH

[martingale]

given an appropriate utility function for participants

That gets close to changing the parameters to hide the evidence of failure.

Sure. If you believe that progress in science is "failure" then you're right. Most people think progress is a good thing, but you can call it failure if you want. I am just pointing out two things that I think it's important people know:

1. As in any science, the work of behavioral economists will lead to an improved version of the EMH, not an outright rejection of it. The EMH is wrong the same way Newton's theory of gravity is wrong--it agrees with the better theory most of the time, but in some cases, other theories agree with the data even better.

2. The theory of efficiency in competitive markets derives not from the work of EMH advocates like French/Fama/etc., but from John M. Keynes and Adam Smith. Market efficiency (in general, not just the stock market) is a fairly well established core and fundamental theory of economics.

It seems to me some people hate the idea that the market may be efficient, and every time someone finds a flaw in the EMH (of which there will be many) they throw a party and go on as if the EMH can be ignored forevermore. It's something like the way creationists jump for joy every time someone discovers that the theory of evolution requires some modification.

Science progresses by discovering errors in older theories and producing new, better ones. It's not going to happen that the EMH is "totally wrong" it's going to happen that something else is better. But yes, if it makes you happy, the EMH is wrong. Relativity, quantum mechanics, and fluid dynamics, the theory of evoloution are all wrong too.

[NormR]

The EMH is a nice theory that got mauled when it ran into the real world. In most cases when this happens, the theorists suck it up and go back to the drawing board.

The Cost Matters Hypothesis is more fundamental and has been around longer.

I suggest that the most effective investment plan a taxable investor with a time horizon of more than ten years can make is to buy rock-solid blue chips, 100 shares at a time whenever enough cash is accumulated, and sit on them. This does not apply to the entire portfolio, of course, but to the taxable equity core.

Amen

[yielder]

The EMH is a nice theory that got mauled when it ran into the real world. In most cases when this happens, the theorists suck it up and go back to the drawing board.

I was thinking about the weak responses of Malkiel and Fama to their critics and the fact that neither seems to put their money where their mouth is. For either to come right out and say "Whoops" would be a career killing admission. Eventually, the anomalies will overpower EMH and Random Walk and behavioural finance research will become more widespread and robust. Investors and pros alike will eat it up because it will legitimize their efforts to beat the market. Think of the opportunities that'll present.

The Cost Matters Hypothesis is more fundamental and has been around longer.

Absolutely.

Amen

For ever and ever, amen. Can I hear a hallelujah? Whoops, excuse the irrational exuberance.

[martingale]

The EMH has done better with real world data than any other theory of the market ever presented, so I'm not sure where this "mauled by the real world" comment comes from. So far the EMH is still holding up better than any of the alternatives when it meets with real world data.

Everyone expects the EMH to be superceded by a better theory, the way every theory is. I am not sure how that gives anyone cause to ignore it. The ONLY credible threat to the EMH in the literature comes from the behavioralists, and they are expected to update the EMH rather than outright refute it. Behavioral finance economists agree with EMH advocates on far more than they disagree.

[jiHymas]

When economists look at the market and judge its efficiency liquidity is not only taken into account, but used to prove efficiency by demonstrating that the liquidity costs prevents participants from earning economic profit.

This, actually, is the crux of my argument. I assert that one man's liquidity cost is another man's liquidity profit. I further assert that it is through a single-minded approach to accumulating liquidity profits that brokers make their money. I further assert that in order to reduce market risk and capital requirements, brokers do not always seek to realize the full 'liquidity profit' of a given trade, but will attempt to lay off their positions in an expeditious manner (that is to say, if they buy a large position at a discount to the 'real' market of X, they will not insist on selling this position at the 'real' market price to realize a profit of X, but will be perfectly happy to sell their position at a discount of aX, a<1. Finally, I will assert that many successful rich/cheap analytical systems, including my own, are successful simply because they supply long-term liquidity to the brokers, who in turn supply short-term liquidity to the market in general.

Incidentally, according to the IDA Securities Industry Performance: January - March 2005. (June 2005), trading profits in the 1Q05 were $280-million. Not bad! There are fairly high barriers to entry into the business since (i) it's hard to raise sufficient capital to go after trades in size on a consistent basis (ii) if you do raise sufficient capital, you won't necessarily get a lot of client calls, since many buy-side firms semi-arbitrarily restrict their call list and want to deal with firms with a long history (iii) you need a lot of volume to justify the expense, since you have to be able to count on being able to lay off your market exposure in an expeditious fashion.

When I was trading governments, the call-list for a trade of, say 50-million a side was a lot shorter than the trade for 10-million a side. Less active dealers can't offer prices competitive to their more active competition on sizable trades, because they can't count on being able to lay them off as quickly. This is not just a profitability concern: there are also concerns related to regulatory capital and market risk in maintaining a large position for a lengthy period.

A trader with one of the bigger firms once told me that, all in all, he liked doing business with us. Well, he would hardly say otherwise! But in explanation, he said that despite his aggravation with our relatively complex trades and annoyance with a client who would get excited over pennies he found it very useful to have a client that would generally move counter to the street and help him square his books.

[jiHymas]

I guess one more problem I have with the strong-form EMH is that conclusions that may be drawn from it all turn out to be false (I have no such difficulties with weak-form EMH)

Strong-EMH says that you can't make any extra money by investing with a genius. Therefore, since investors are so completely rational, they will not pay for the prospect of superior returns - they will simply pick a spot on the efficient frontier to suit their comfort level and get there by indexing. All they really need by way of advice is a check that their portfolios are, in fact efficient.

But this is manifestly not the case! Regardless of whether excess returns are or are not achievable, investors have attempted, are attempting and will attempt to secure them!

All mutual fund advertising (of active funds, anyway) has the implicit message that the genius managers at XYZ Co. will get better returns than their competition. Newsletters sell subscriptions this way. The predicted result of the strong-EMH simply isn't happening.

All active managers attempt to sell their services on the basis of superior returns. Some of these are out-and-out frauds (the SEC allegations in S.E.C. v. Terry's Tips, Inc. and Terry F. Allen make interesting reading and may be true). But of all the active managers I have ever met, each one has known, in his heart of hearts, that he is the greatest portfolio manager in the history of the universe and it is only through the jealousy of his superiors, the incompetence of his staff and the soon-to-be-corrected perversity of the market that this fact is not universally recognized.

[Bylo Selhi]

When I was trading governments... A trader with one of the bigger firms once told me that, all in all, he liked doing business with us. Well, he would hardly say otherwise! But in explanation, he said that despite his aggravation with our relatively complex trades and annoyance with a client who would get excited over pennies he found it very useful to have a client that would generally move counter to the street and help him square his books.

That would be GBC Bond? As I recall (albeit dimly) throughout the '90s, when I held both, GBC's MER was ~90bp, or ~30bp more than PH&N Bond and the difference in returns was also ~30bp in favour of the wet coast. When all was said and done, what value was being added (for the investor)?

But of all the active managers I have ever met, each one has known, in his heart of hearts, that he is the greatest portfolio manager in the history of the universe and it is only through the jealousy of his superiors, the incompetence of his staff and the soon-to-be-corrected perversity of the market that this fact is not universally recognized.

IANAP but could those inflated egos be the result of 6 and 7 digits worth of annual [s]remuneration[/s]positive reinforcement? (Or maybe it's just something in the tanning process for the leather seats in those 911s.)

[Shakespeare]

Strong EMH basically says you can not make extra returns by paying attention to valuations. IMO, it's dead wrong. I do not have such problems with weak EMH, which allows that possibility.

[yielder]

Strong EMH basically says you can not make extra returns by paying attention to valuations. IMO, it's dead wrong. I do not have such problems with weak EMH, which allows that possibility.

And it's this statement that you cannot beat the market that is mainly responsible for the indexing mantra. Cost plays an understated role in the mantra. As I said upthread, the fact that the market can be beaten doesn't mean you shouldn't index. Cost, built in fund hurdles, manager behavioural problems are all strong arguments for indexing unless no index product exists or you don't know how to how to identify funds likely to outperform. Identifying these funds can be done but the possible added returns probably don't justify the added relative risk.

Norm has a great Buffet quote in his current Weekly Stingy news - "The stock market is a no-called-strike game. You don't have to swing at everything--you can wait for your pitch. The problem when you're a money manager is that your fans keep yelling, 'Swing, you bum!'" At overvaluation peaks, funds are awash in new money which gets invested in expensive stocks. At undervaluation troughs, there is a steady outflow of investor money. Buying high and selling low is a tough hurdle to overcome.

[Bylo Selhi]

Identifying these funds can be done but the possible added returns probably don't justify the added relative risk.

"Successful investing, based on indexing, depends on trading off the low possibility of doing better than the index, for the high probability of doing better than most other funds." ...Ted Cadsby, The Power of Index Funds

[NormR]

I find it interesting that even Bogle has seemed to move away from the EMH and has gone to the Cost Matters Hypothesis.

Also, it seems clear that indexing does not rely on the EMH to be true. Nor does the lack of success, of the average, fund manager rely on the EMH.

The EMH has done better with real world data than any other theory of the market ever presented, so I'm not sure where this "mauled by the real world" comment comes from. So far the EMH is still holding up better than any of the alternatives when it meets with real world data.

EMH has only survived so long because it is very hard to falsify or disprove. This is actually a significant disadvantage for a theory because it also means that it doesn't make significant predictions.

Is it a shock to anyone that earning unusually large profits in the market is both hard and, err, unusual?

It strikes me that the EMH presents a nice story but I don't see what it has actually given us. What are the solid predictions that it makes and which ones haven't run into a nasty 'anomaly'?

[NormR]

Who Gains from Trade? Evidence from Taiwan

In the presence of information and trading costs, informed investors should profit from
uninformed investors. We test this proposition by analyzing the performance of
institutional and individual investors using trades data for all market participants in the
Taiwan stock market during the five years ending in 1999. Before considering trading
costs (commissions and transaction taxes), on the average day, institutions realize trading
profits of $NT 178 million, while individual investors lose the same amount. In general,
the gains to institutions are not offset by their trading costs, while trading costs
exacerbate the losses of individuals. After costs, we estimate that the trading of
institutional investors adds one percentage point annually to their portfolio performance,
while the trading of individuals subtracts over three percentage points annually from their
performance. All major institutional categories (corporations, dealers, foreigners, and
mutual funds) earn profits before costs. Only corporations fail to do so after costs. We
also map trades to orders, which we classify as aggressive (demanding liquidity) or
passive (providing liquidity) based on order prices. Virtually all trading losses incurred
by individuals can be traced to their aggressive orders. In contrast, institutions profit from
both their passive and aggressive trades. Most of the institutional profits from passive
trades, which provide liquidity to market participants, are accrued within a few days of
the trade. In contrast, most of the institutional profits from their aggressive trades accrue
at horizons up to six months. All of the gains from trade are exhausted within six
months.

[NormR]

Norm has a great Buffet quote in his current Weekly Stingy news - "The stock market is a no-called-strike game. You don't have to swing at everything--you can wait for your pitch. The problem when you're a money manager is that your fans keep yelling, 'Swing, you bum!'" At overvaluation peaks, funds are awash in new money which gets invested in expensive stocks. At undervaluation troughs, there is a steady outflow of investor money. Buying high and selling low is a tough hurdle to overcome.

I guess that I'll keep my new 'quote of the week' going
It's always hard to know what is liked and what is not in an email newsletter.

[jiHymas]

When I was trading governments... A trader with one of the bigger firms once told me that, all in all, he liked doing business with us. Well, he would hardly say otherwise! But in explanation, he said that despite his aggravation with our relatively complex trades and annoyance with a client who would get excited over pennies he found it very useful to have a client that would generally move counter to the street and help him square his books.

That would be GBC Bond? As I recall (albeit dimly) throughout the '90s, when I held both, GBC's MER was ~90bp, or ~30bp more than PH&N Bond and the difference in returns was also ~30bp in favour of the wet coast. When all was said and done, what value was being added (for the investor)?

This was in the '90's, when I was with Greydanus, Boeckh (GBA) and yes, GBC Bond Fund was among our clients. We had a lot of respect for PH&N and we were on a lot of short-lists with them for institutional business. We traded governments exclusively, PH&N were more closely matched to the index. Note that GBA did not own GBC Bond in this period - we were the investment managers only.

What value did we add for clients? There may have been some exceptions, but basically our composite results (for institutional accounts with full discretion) were first quartile for all periods in excess of a year (first quartile for single years is generally achieved by cowboys who bet big and bet right. We didn't want to be there.). Again, my memory is hazy, but I think our composite returns for longer periods exceeded the index by about 25bp pa - an extremely good number for any bond portfolio, particularly one that eschewed corporates.

The problem with discussing active bond management returns in a retail context is that the MER is by far the dominant issue with respect to investor returns.

[Norbert Schlenker]

Let me throw something new into the discussion. This article by Rappaport is in the most recent issue of FAJ and well worth reading.

[Norbert Schlenker]

I'm responding to a few select items.

The point is that the market pays up on a regular basis for liquidity. Selling that liquidity is a very profitable business; the existence of that premium is not compatible with any strong-form EMH I have yet seen.

It seems to me hardly inconceivable that what is now the standard three factor model (up from beta alone as originally formulated) couldn't be adjusted to add "liquidity provision" as a fourth factor.

I guess one more problem I have with the strong-form EMH is that conclusions that may be drawn from it all turn out to be false

Remember this for just one minute. Here's a hypothesis from which false conclusions can be drawn, so it's flawed. Then, in the very next paragraphs ...

Strong-EMH says that you can't make any extra money by investing with a genius. Therefore, since investors are so completely rational, they will not pay for the prospect of superior returns - they will simply pick a spot on the efficient frontier to suit their comfort level and get there by indexing. All they really need by way of advice is a check that their portfolios are, in fact efficient.

But this is manifestly not the case! Regardless of whether excess returns are or are not achievable, investors have attempted, are attempting and will attempt to secure them!

Here's the contrary hypothesis - that strong EMH is false. The result is "have attempted, are attempting, and will attempt". Be honest for just a moment now. What is the long term record of the community of investors who do exactly this? I believe that it is scandalous underperformance versus the indices. By the original logic, this contrary hypothesis must also be false.

I was thinking about the weak responses of Malkiel and Fama to their critics and the fact that neither seems to put their money where their mouth is. For either to come right out and say "Whoops" would be a career killing admission.

At this point, I don't think either of them care that much about their careers. Malkiel is in his 70s, Fama is 66. They're made men even if EMH / 3FM are conclusively debunked tomorrow. I also wonder what you mean about putting their money where their mouth is. Is there conclusive evidence that they're not by and large indexers? Even if there is, do you have conclusive evidence that they're not just playing around?

Strong EMH basically says you can not make extra returns by paying attention to valuations. IMO, it's dead wrong.

Valuation is a subjective thing. The market price is in almost all cases the result of someone else's valuation, not yours! What makes you right and everyone else wrong?

hard to falsify ... What are the solid predictions that it makes and which ones haven't run into a nasty 'anomaly'?

The solid prediction is that the current price is about right and that, if you accept this and ignore the price, you are likely to do better than most people who don't. That prediction is borne out by history.

[Norbert Schlenker]

While I'm at it, what about this?

A little girl and her toy dog are knocking the stuffing out of a pack of pros at the half-way mark of Report on Business's ninth annual stock-picking contest.

Emilia Loewen, 5, daughter of Report on Business deputy editor Cathryn Motherwell, and her stuffed dog Sam are the wild-card entry in a group of professional market experts in this year's My One and Only stock-picking contest. The entrants, who are vying for a Globe and Mail coffee mug, must pick a stock trading at $1 or more on the Toronto Stock Exchange, and hold it for the year. ...

At the halfway mark of Report on Business's ninth annual stock-picking contest, Shell Canada Ltd., chosen by Sam, the stuffed dog owned by Emilia Loewen, has given the five-year-old a substantial lead over her more learned rivals.
June 30 6-month
Player Stock (symbol) Price Total return
Emilia Loewen Shell Canda (SHC-TSX) $32.89 +24.12%
Al Budal Xceed Mortgage (XMC-TSX) 5.20 +14.04
S&P/TSX Composite index 9,902.77 +8.11%
Contra Guys Cygnal Tech. (CYN-TSX) 1.44 -4.00
Marco den Ouden Sherritt Int'l (S-TSX) 9.31 -5.83
Robert Callander CGI Group (GIB.SV.-TSX) 7.36 -8.00
Michael Smedley Centurion Energy (CUX-TSX) 13.41 -8.78
Veronika Hirsch West Energy (WTL-TSX) 4.75 -16.96
Robert McWhirter Calian Technologies (CTY-TSX) 12.25 -17.23
Michael Vaughan Ondine Biopharma (OBP-TSX) 1.70 -47.69

Pyette, G&M

[Shakespeare]

What makes you right and everyone else wrong?

Still making money off your preferred spreadsheet, Norbert?

[Norbert Schlenker]

Adjusted for risk and time spent and the compensation due to liquidity providers, there is probably no economic profit whatsoever.

[jiHymas]

Who Gains from Trade? Evidence from Taiwan

Superb article! Thanks for the link - it's always nice to know that there's at least one academic who agrees with me! I've added links to this paper on my websites.

The point is that the market pays up on a regular basis for liquidity. Selling that liquidity is a very profitable business; the existence of that premium is not compatible with any strong-form EMH I have yet seen.

It seems to me hardly inconceivable that what is now the standard three factor model (up from beta alone as originally formulated) couldn't be adjusted to add "liquidity provision" as a fourth factor.

Here again, I think, we're approaching the point where the argument simply becomes playing with words. Is the strong-EMH with a liquidity provision still a strong-EMH? If a pig had wings, could it fly?

I guess one more problem I have with the strong-form EMH is that conclusions that may be drawn from it all turn out to be false

Remember this for just one minute. Here's a hypothesis from which false conclusions can be drawn, so it's flawed. Then, in the very next paragraphs ...

Strong-EMH says that you can't make any extra money by investing with a genius. Therefore, since investors are so completely rational, they will not pay for the prospect of superior returns - they will simply pick a spot on the efficient frontier to suit their comfort level and get there by indexing. All they really need by way of advice is a check that their portfolios are, in fact efficient.

But this is manifestly not the case! Regardless of whether excess returns are or are not achievable, investors have attempted, are attempting and will attempt to secure them!

Here's the contrary hypothesis - that strong EMH is false. The result is "have attempted, are attempting, and will attempt". Be honest for just a moment now. What is the long term record of the community of investors who do exactly this? I believe that it is scandalous underperformance versus the indices. By the original logic, this contrary hypothesis must also be false.

I disagree with your logic. The strong-EMH makes a prediction - in formal terms:

A -> B

I state that, in fact, we observe !B (not-B) and may therefore conclude !A, since

!B -> !A

is the contrapositive of the original statement, which is allowed.

You then claim that

!A -> !B

which is not allowed by formal logic. I can never remember whether that logical manipulation is called the converse or the the inverse, but one way or another it is not logically equivalent to my statement.

In terms relative to the original line of thought, I suggest that there are confounding factors which result in scandalous underperformance vs. the indices: lack of information, bad information, excessive transaction costs and, of course, plain bone-headedness.

Added later

According to Review of Concepts of Formal Logic (Purdue) if the original statement is P -> Q then

Converse : Q -> P
Inverse : !P -> !Q
Contrapositive : !Q -> !P

[martingale]

The strong form of the EMH predicts that some people will have excessive returns for extended periods of time. Similarly, if 1000 people flip a coin 10 times in a row, someone will flip heads ten times out of ten.

[Norbert Schlenker]

Here again, I think, we're approaching the point where the argument simply becomes playing with words. Is the strong-EMH with a liquidity provision still a strong-EMH? If a pig had wings, could it fly?

Strong form EMH says that the current market price is fair to within transaction costs; i.e. there's a willing buyer just under fair and a willing seller just over. I would argue that illiquidity is a transaction cost. That liqudity providers get paid is not much of a refutation of EMH.

I disagree with your logic. The strong-EMH makes a prediction - in formal terms:

A -> B

I state that, in fact, we observe !B (not-B) and may therefore conclude !A, since

!B -> !A

is the contrapositive of the original statement, which is allowed.

Agreed.

You then claim that

!A -> !B

which is not allowed by formal logic.

Agreed, but that wasn't the point I was making.

Strong EMH claims that there is no economic value in acquiring information re securities (unless you are the absolutely first one to exploit it, which is a costly exercise). The negative of that is that there is economic value in acquiring such information, i.e. going to the effort will result in financial reward.

I was just pointing out that the historical evidence for such financially rewarding results is so scant that this (contra)hypothesis is just as gravely in doubt.

The strong form of the EMH predicts that some people will have excessive returns for extended periods of time. Similarly, if 1000 people flip a coin 10 times in a row, someone will flip heads ten times out of ten.

To be fair, that's not what strong EMH says. Strong EMH says that there are no economic profits from the work involved. If that's so, then we fall back on regular statistics, which predicts a good chance that one in a thousand will flip ten heads in a row.

[jiHymas]

Here again, I think, we're approaching the point where the argument simply becomes playing with words. Is the strong-EMH with a liquidity provision still a strong-EMH? If a pig had wings, could it fly?

Strong form EMH says that the current market price is fair to within transaction costs; i.e. there's a willing buyer just under fair and a willing seller just over. I would argue that illiquidity is a transaction cost. That liqudity providers get paid is not much of a refutation of EMH.

Whatever you like. As long as I get paid, I don't care what you call it.

[martingale]

To be fair, that's not what strong EMH says. Strong EMH says that there are no economic profits from the work involved. If that's so, then we fall back on regular statistics, which predicts a good chance that one in a thousand will flip ten heads in a row.

Pretty close. If you substitute the word "expected" between "profits" and "from" then I'll agree with you completely. The EMH does rely on regular statistics around this statistical term "expected" and therefore predicts that some people will earn higher than average return--requires it in fact. Were there not enough people earning the higher than average return that in and of itself would invalidate the EMH.

However, as we seem to agree, those people earning the higher than average return are just lucky, and the unlucky cancel them out, so that any individuals expectation is normal profit, if the EMH is true.

In case I haven't beat this horse enough, normal profit is expected by the EMH. That means you earned something, but you could have done as well or better doing something else with your time and money. No economic profit is expected in an efficient market.

[Norbert Schlenker]

As long as I get paid, I don't care what you call it.

That's a very weak argument against EMH, James. Liquidity is a valuable service and the people providing it deserve to be paid. That would be so even if the market were perfectly efficient. That such things happen now is thus irrelevant to the issue of whether EMH is true or not.