Sunday, February 14, 2010

The Invincible Markets Hypothesis

There has been a lot of impassioned debate over the efficient markets hypothesis recently, but some of the disagreement has been semantic rather than substantive, based on a failure to distinguish clearly between informational efficiency and allocative efficiency. Roughly speaking, informational efficiency states that active management strategies that seek to identify mispriced securities cannot succeed systematically, and that individuals should therefore adopt passive strategies such as investments in index funds. Allocative efficiency requires more than this, and is satisfied when the price of an asset accurately reflects the (appropriately discounted) stream of earnings that it is expected to yield over the course of its existence. If markets fail to satisfy this latter condition, then resource allocation decisions (such as residential construction or even career choices) that are based on price signals can result in significant economic inefficiencies.
Some of the earliest and most influential work on market efficiency was based on the (often implicit) assumption that informational efficiency implied allocative efficiency. Consider, for instance, the following passage from Eugene Fama's 1965 paper on random walks in stock market prices (emphasis added):
The assumption of the fundamental analysis approach is that at any point in time an individual security has an intrinsic value... which depends on the earning potential of the security. The earning potential of the security depends in turn on such fundamental factors as quality of management, outlook for the industry and the economy, etc...

In an efficient market, competition among the many intelligent participants leads to a situation where, at any point in time, actual prices of individual securities already reflect the effects of information based both on events that have already occurred and on events which, as of now, the market expects to take place in the future. In other words, in an efficient market at any point in time the actual price of a security will be a good estimate of its intrinsic value.
Or consider the opening paragraph of his enormously influential 1970 review of the theory and evidence for market efficiency:
The primary role of the capital market is allocation of ownership of the economy's capital stock. In general terms, the ideal is a market in which prices provide accurate signals for resource allocation: that is, a market in which firms can make production-investment decisions, and investors can choose among the securities that represent ownership of firms' activities under the assumption that security prices at any time “fully reflect” all available information. A market in which prices always “fully reflect” available information is called “efficient.”
The above passage is quoted by Justin Fox, who argues that proponents of the hypothesis have recently been defining efficiency down:
That leaves us with an efficient market hypothesis that merely claims, as John Cochrane puts it, that "nobody can tell where markets are going." This is an okay theory, and one that has held up reasonably well—although there are well-documented exceptions such as the value and momentum effects.
The most effective recent criticisms of the efficient markets hypothesis have not focused on these exceptions or anomalies, which for the most part are quite minor and impermanent. The critics concede that informational efficiency is a reasonable approximation, at least with respect to short-term price forecasts, but deny that prices consistently provide "accurate signals for resource allocation." This is the position taken by Richard Thaler in his recent interview with John Cassidy (h/t Mark Thoma):
I always stress that there are two components to the theory. One, the market price is always right. Two, there is no free lunch: you can’t beat the market without taking on more risk. The no-free-lunch component is still sturdy, and it was in no way shaken by recent events: in fact, it may have been strengthened. Some people thought that they could make a lot of money without taking more risk, and actually they couldn’t. So either you can’t beat the market, or beating the market is very difficult—everybody agrees with that...
The question of whether asset prices get things right is where there is a lot of dispute. Gene [Fama] doesn’t like to talk about that much, but it’s crucial from a policy point of view. We had two enormous bubbles in the last decade, with massive consequences for the allocation of resources.
The same point is made somewhat more tersely by The Economist:
Markets are efficient in the sense that it's hard to make an easy buck off of them, particularly when they're rushing maniacally up the skin of an inflating bubble. But are they efficient in the sense that prices are right? Tens of thousands of empty homes say no.
And again, by Jason Zweig, building on the ideas of Benjamin Graham:
Mr. Graham proposed that the price of every stock consists of two elements. One, "investment value," measures the worth of all the cash a company will generate now and in the future. The other, the "speculative element," is driven by sentiment and emotion: hope and greed and thrill-seeking in bull markets, fear and regret and revulsion in bear markets.

The market is quite efficient at processing the information that determines investment value. But predicting the shifting emotions of tens of millions of people is no easy task. So the speculative element in pricing is prone to huge and rapid swings that can swamp investment value.

Thus, it's important not to draw the wrong conclusions from the market's inefficiency... even after the crazy swings of the past decade, index funds still make the most sense for most investors. The market may be inefficient, but it remains close to invincible.
This passage illustrates very clearly the limited value of informational efficiency when allocative efficiency fails to hold. Prices may indeed contain "all relevant information" but this includes not just beliefs about earnings and discount rates, but also beliefs about "sentiment and emotion." These latter beliefs can change capriciously, and are notoriously difficult to track and predict. Prices therefore send messages that can be terribly garbled, and resource allocation decisions based on these prices can give rise to enormous (and avoidable) waste. Provided that major departures of prices from intrinsic values can be reliably identified, a case could be made for government intervention in affecting either the prices themselves, or at least the responses to the signals that they are sending.
Under these conditions it makes little sense to say that markets are efficient, even if they are essentially unpredictable in the short run. Lorenzo at Thinking Out Aloud suggests a different name: other things in economics, such as rational expectations, EMH needs a better name. It is really something like the "all-information-is-incorporated hypothesis" just as rational expectations is really consistent expectations. If they had more descriptive names, people would not misconstrue them so easily and there would be less argument about them.
But a name that emphasizes informational efficiency is also misleading, because it does not adequately capture the range of non-fundamental information on market psychology that prices reflect. My own preference (following Jason Zweig) would be to simply call it the invincible markets hypothesis.


Update (2/16). Mark Thoma has more on the subject, as does Cyril H├ędoin. Brad DeLong and Robert Waldmann have also linked here, which gives me an excuse to mention two papers of theirs (both written with Shleifer and Summers, and both published in 1990) that were among the first to try and grapple with the question of how rational arbitrageurs would adjust their behavior in response to the presence of noise traders. In a related article that I have discussed previously on this blog (here and here, for instance), Abreu and Brunnermeier have shown how the difficulty of coordinated attack can result in prolonged departures of prices from fundamentals.


Update (2/17). My purpose here was to characterize a hypothesis and not to endorse it. In a comment on the post (and also here), Rob Bennett makes the claim that market timing based on aggregate P/E ratios can be a far more effective strategy than passive investing over long horizons (ten years or more.) I am not in a position to evaluate this claim empirically but it is consistent with Shiller's analysis and I can see how it could be true. Over short horizons, however, attempts at market timing can be utterly disastrous, as I have discussed previously. This is what makes bubbles possible. In fact, I believe that market timing over short horizons is much riskier than it would be if markets satisfied allocative efficiency and the only risk came from changes in fundamentals and one's own valuation errors.


Update (2/20). Scott Sumner jumps into the fray, but grossly mischaracterizes my position:
Then there is talk (here and here) of a new type of inefficient markets; Rajiv Sethi calls it the invincible market hypothesis.  I don’t buy it, nor do I think the more famous anti-EMH types would either. The claim is that markets are efficient, but they are also so irrational that there is no way for investors to take advantage of that fact.  This implies that the gap between actual price and fundamental value doesn’t tend to close over time, but rather follows a sort of random walk, drifting off toward infinity.
This is obviously not what I claimed. There is absolutely no chance of the gap between prices and fundamentals "drifting off toward infinity." All bubbles are followed by crashes or bear markets, and prices do track fundamentals pretty well over long horizons. The problem lies in the fact that attempts to time the market over short horizons can be utterly disastrous, as I have discussed at length in a previous post; this is what makes asset bubbles possible in the first place. I used the term "invincible markets hypothesis" not as a "new type of inefficient markets hypothesis" but rather as a description of the claim that markets satisfy informational (but not allocative) efficiency. And I did not endorse this claim, except as a "reasonable approximation... with respect to short-term price forecasts."

Sumner continues as follows:
Sethi argues Shiller might be right in the long run, but may be wrong in the short run. I don’t buy that distinction. If Shiller’s right then the anti-EMH position has useful investment implications, even for short term investors...
This too is false. I believe that Shiller is right in the short run (since he argues that prices can depart significantly from fundamental values) and also right in the long run (since he believes that in the long run prices track fundamentals quite well). This does not have useful investment implications for short term investors because short run price movements are so unpredictable and taking short positions during a bubble is so risky. Over long horizons, however, Shiller's analysis does suggest that risk-adjusted returns will be greater if the P/E ratio is lower at the time of the initial investment. One could rationalize this with suitable assumptions about time-varying discount rates, but as Thaler points out, such rational choice models are incredibly flexible and lacking in discipline. Everyone acknowledges, however, that for most investors passive investing is far superior to short-term attempts at market timing. The disagreement is about whether prices can deviate significantly from fundamentals from time to time, resulting in severe economic dislocations and inefficiencies.


Update (2/24). For a sober assessment of why passive investing remains the best strategy for most investors despite modest violations of informational efficiency, see this post at Pop Economics.

Robert Waldmann's comments at Angry Bear are excellent, but need to be read with some care. His main point is this: there exist certain (standard but restrictive) general equilibrium models in which informational efficiency does imply allocative efficiency. But minor deviations from informational efficiency do not imply that deviations from allocative efficiency will also be minor.
Anomalies in risk adjusted returns on the order of 1% per year can't be detected. We can't be sure of exactly how to adjust for risk. However, they can make the difference between allocative efficiency and gross inefficiency.

For policy makers there is a huge huge difference between "markets are approximately informational efficient" and "markets are informational efficient." The second claim (plus standard false assumptions) implies that markets are allocatively efficient. The [first] implies nothing about allocative efficiency.
In other words, the link between informational efficiency and allocative efficiency is not robust. This is why the market can be hard to beat, and yet generate significant departures of prices from fundamentals from time to time. 


  1. I appreciate your analysis, which seems quite a bit more thoughtful than most of the current EMH bashing, which I think often mis-states the claims of the EMH, and vastly overstates its influence on policy.

    Allocative efficiency requires more than this, and is satisfied when the price of an asset accurately reflects the (appropriately discounted) stream of earnings that it is expected to yield over the course of its existence.

    Does this mean that allocative efficiency requires that the price precisely equal the net present value of future cash flows? That can be disproved by any single data point, such as a company that goes bankrupt before the cash flows eventuate.

    I have never heard anyone claim that market prices only reflect fundamental information, only that market prices are relatively better than other methods of valuing. Even the Fama quote explicitly calls the market price "a good estimate".

    If the "best possible estimate" claim of the EMH were true, then would this not be sufficient to make the claim for allocative efficiency?

    The best argument I have heard for why market prices may not be the best estimate involve the difficulty in making bets on downturns. If issues such as liquidity and statutory issues limit betting on downturns, then this could certainly cause mis-pricing. However if this is the case, perhaps rather than attempting to pop bubbles, we should make it simpler to make bets on downturns.

  2. Bruce, thank your for your comment. Many firms go bankrupt before cash flows eventuate, but this does not necessarily violate allocative efficiency. The key here is the term "expected". As long as there is a positive probability of future cash flows, a positive price is consistent with allocative efficiency. But it could be that the price is much higher than the valuation supported by this probability, simply because the price is expected to rise further in the short run - in this case we have a bubble and allocative efficiency is violated. And the price will be a poor estimate of fundamental information.

    It is not just difficult to make bets on downturns - it is also incredibly risky because one needs to get the timing right and have lots of liquidity. Plenty of assets allow us to do this, put options and Credit Default Swaps for instance. But many big names have lost their shirts trying. See my earlier post for examples. However, Geanakoplos has argued that the emergence of credit default swaps did cause the housing bubble to burst by completing markets and allowing pessimists to leverage. This would support your position.

  3. Quoting from your quotation from Jason Zweig:
    "The other, the "speculative element," is driven by sentiment and emotion: hope and greed and thrill-seeking in bull markets, fear and regret and revulsion in bear markets."

    Quoting Keynes, from The General Thoery:
    "Most, probably, of our decisions to do something positive, the full consequences of which will be drawn out over many days to come, can only be taken as the result of animal spirits—a spontaneous urge to action rather than inaction, and not as the outcome of a weighted average of quantitative benefits multiplied by quantitative probabilities."
    Book 4, Chapter 12, Section 7.

  4. Pardon me if I cite myself, but I think this post on Econospeak is germane. Inability to predict market movements is fully consistent with a wide range of systemic properties, few of which normal people would call efficient.

  5. My view is that the mistake was in thinking that the market is immediately rather than gradually efficient. If the market were immediately efficient, overvaluation would be a nonsense concept. Yet there is much evidence showing that the P/E10 value that applies today does a reasonably good job of predicting the price that will apply in 10 years. So immediate efficiency just cannot be.

    The information that is not being incorporated into the current price is the extent of overvaluation that applies at the time. If investors considered overvaluation when setting their allocations, overvaluation would be impossible (overvaluation diminishes the long-term value proposition of stocks, which should cause sales, which would bring prices back to reasonable levels). It is because we have been taught not to lower our allocations in response to overvaluation during the Buy-and-Hold Era that overvaluation grew so great as to cause a crash.

    If we encouraged investors to adjust their stock allocations in response to overvaluation, market prices would be self-correcting and we would indeed see something close to immediate market efficiency. But we have been urging just the opposite for over 30 years. We have been encouraging investors NOT to take valuations into consideration when setting their stock allocations. Immediate efficiency cannot be attained when investors are encouraged to ignore a critical factor.

    I recently published a Google Knol entitled "Why Buy-and-Hold Investing Can Never Work" that explores this theme in more detail:

    Why Buy-and-Hold Investing Can Never Work

    Rob Bennett

  6. "Roughly speaking, informational efficiency states that active management strategies that seek to identify mispriced securities cannot succeed systematically, and that individuals should therefore adopt passive strategies such as investments in index funds."

    I think you are defining the informational efficiency down. A large part of what you call "sentiment and emotion" is just a pure noise. I don't agree that prices which send noisy garbled messages should be called informationaly efficient.

  7. 123 - I was just using the definition that the EMH folks use when they conduct tests of the hypothesis. I agree that the name doesn't fit, which was kind of the point I was making. As I said at the end of the post, the term informational efficiency is misleading, "because it does not adequately capture the range of non-fundamental information on market psychology that prices reflect." So I think we actually agree on this point.

  8. Rajiv,
    I wanted to stress that market sentiment has two different parts - genuine shifts in investor risk preferences and changes in investor expectations that are often not based on any research at all. The problem is that EMH folks mix these two up, whereas only the first part of sentiment deserves the name of fundamental information. The important divide is that Fama says that these two parts are observationally equivalent, while Shiller does surveys that try to measure market return expectations of small investors.

  9. The recent critiques of the EMH point to the housing bubble as "proof" that markets aren't efficient and that the EMH must be wrong.

    But were housing prices "wrong" in an environment where someone with no documented income, no down payment, and a sub 600 credit score could get a $500k mortgage? I'd wager no.

    Those who saw the crisis coming believed that housing prices were out of whack because the underlying fundamentals (lending standards, etc) were completely messed up. Ex-post, this is obvious, but at the time, the contrarians were in the minority.

    Zuckerman's account of how John Paulson cleaned up on the housing crisis ("The Greatest Trade Ever") is an excellent example of how the EMH works in practice. Paulson believed the underlying fundamentals to the housing market were wrong (and therefore prices were wrong) and made big bets accordingly. In doing so, he flew in the face of pretty much everyone else on Wall Street and stuck to his positions. It is very, very difficult to be a contrarian when seemingly the rest of the world is calling you nuts.

    Efficient does not mean rational and when "The market can stay irrational longer than you can stay solvent", going against market prices can be very tough to do successfully.

  10. Matty, ask yourself why "someone with no documented income, no down payment, and a sub 600 credit score could get a $500k mortgage". In large part it was because lenders believed that housing prices would continue to appreciate for some time, so that they would not lose money on their collateral even if the borrower ended up in foreclosure. In other words, what you are calling "fundamentals" depend on expectations about asset prices. If you want to claim that asset prices reflect fundamental values (as Fama does in the quotes above) then you have to define fundamentals independently of asset price expectations. Otherwise the argument is circular. Also, at the end of any bubble (including the one in tech stocks a decade ago) there will always be a major redistribution of wealth and a few big winners. Housing is no exception.

  11. Nice piece indeed.

    Thaler did a piece in the FT a while back that made the same claims about the theory, but it may miss a crucial distributional aspect to the story. The FT printed this response:

    Sir, Richard Thaler’s excellent article “The price is not always right and markets can be wrong”, (August 5), on the Efficient Market Hypothesis, may still be too kind to the “no free lunch” aspect of the theory. Applied to the latest boom and bust its simply another version of Keynes’ famous claim that in the long run we are all dead. That is, profits in the upswing of a bubble are either dissipated on the downside or result in insolvency when we try to be smarter than the market.

    As Prof Thaler puts it, “shorting internet stocks or Las Vegas real estate two years before the peak was a good recipe for bankruptcy”. But surely the point is that knowing this doesn't stop people trying to get the free lunch of “getting in and out just in time and making a bundle”, which in part generates the very bubbles that draw the theory into question. If investors sense a free lunch, and flipping condos is but the most recent example, they will go for it again and again, regardless of any theory that says that they should not.

    I think the point still stands.

  12. Thanks Mark... I agree completely. In fact, Abreu and Brunnermeier (cited in an update to this post) go one step further and show that rational traders who become aware that a bubble is in play will also buy into the rise as long as they are reasonably confident that they are among the first few to become aware of the overvaluation. It's an important point.

  13. I think this is a good way to explain EMH in a more systematic way. There is a lot of confusion out there, especially for those who blame the EMH for recent financial troubles.

    To assume market prices are NOT efficient however requires one to find the "right" price. So, the statement that markets are not allocative efficient implies that one has the "right" price that makes allocation better. This is an ex-post statement and probably has no value technically. One could always counteract by saying that the price is efficient at any point in time, given the information and policy choices at that point. The fact that investment decisions and capital allocation appear to be sub-optimal ex. post does not imply that markets were inefficient when those decisions were made.

    Interested in your comments on my bolg ""

  14. Rob Bennett makes the claim that market timing based on aggregate P/E ratios can be a far more effective strategy than passive investing over long horizons (ten years or more.) I am not in a position to evaluate this claim empirically but it is consistent with Shiller's analysis and I can see how it could be true.

    Thank for adding these words in your 2/17 update, Rajiv. I am going to post the text of your entire 2/17 update at my Google Knol so that no one reading there misses it.

    We are in complete agreement re short-term timing.

    If there are people reading these words who would like to take a stab at empirical testing of the Valuation-Informed Indexing strategy, I would of course be happy to assist in that effort in any way possible and would be grateful for their willingness to take on the initiative.

    There has been extensive discussion of the merits of this approach in the Retire Early and Indexing discussion-board communities over the past eight years. I think it would be fair to characterize these discussions (collectively referred to as "The Great Safe Withdrawal Rate Debate") as the most controversial and the most exciting investing-related discussions ever held on the internet.

    John Walter Russell created an entire web site to present his research on this approach:

    Early Retirement Planning Insights

    A helpful graphic that makes the point in a quick way is available at the Financial Web Ring forum:

    "The evidence in pretty incontrovertible. Valuation-Informed everywhere superior to Buy-and-Hold Over 10-Year Periods."

    Please note that the graphic at the Financial WebRing link was prepared prior to the crash. So the VII approach would now show as being superior in the 1990s time-period as well as in all others.


  15. If there are people reading these words who would like to take a stab at empirical testing of the Valuation-Informed Indexing strategy, I would of course be happy to assist in that effort in any way possible and would be grateful for their willingness to take on the initiative.

    I'll take a stab, Rob.

    Here is an analysis that compares Buy-and-Hold to Rob's Valuation Informed Investing (VII).

    In order to compare Rob's VII versus a static, never-changing stock allocation, we need decision rules that tells us how to implement VII. Rob provides us VII decision rules here:

    A Valuation-Informed Indexer might go with a stock allocation of 50 percent at times of moderate prices (a P/E10 level from 12 to 20), a stock allocation of 75 percent at times of low prices (a P/E10 level below 12) and a stock allocation of 25 percent at times of high prices (a P/E10 level above 20).

    How do we determine P/E10 levels? P/E10 data is contained in an Excel spreadsheet on Robert Shiller's website. Here is the link:

    So if you were a VII investor and followed Rob's guidelines, you would have maintained a normal stock allocation of 50% when the P/E10 level ranged between 12 and 20. This was the case up to 1992. However, you would have switched to 25% stocks in 1993 when P/E10 first went above 20. P/E10 stayed above 20 for the next 16 years before dropping below 20 in October 2008.

    I will use the S&P 500 index fund for stocks and the Total Bond Market fund for the non-stock allocation.

    The returns for the S&P 500 (SP500) and the Total Bond Market (TBM) can be found in a spreadsheet created by simba and linked here:

    The first full year recorded for the Total Bond Market fund is 1987. So that is the year I start my data . . .

    Year SP500 TBM
    1987 4.71% 1.14%
    1988 16.22% 7.35%
    1989 31.36% 13.64%
    1990 -3.32% 8.65%
    1991 30.22% 15.25%
    1992 7.42% 7.14%
    1993 9.89% 9.68%
    1994 1.18% -2.66%
    1995 37.45% 18.18%
    1996 22.88% 3.58%
    1997 33.19% 9.44%
    1998 28.62% 8.58%
    1999 21.07% -0.76%
    2000 -9.06% 11.39%
    2001 -12.02% 8.43%
    2002 -22.15% 8.26%
    2003 28.50% 3.97%
    2004 10.74% 4.24%
    2005 4.77% 2.40%
    2006 15.64% 4.27%
    2007 5.39% 6.92%
    2008 -37.02% 5.05%
    2009 26.49% 5.93%

    So with a little spreadsheet work, we can apply Rob's VII guidelines and produce valuation-adjusted returns. The left column represents Buy-and-Hold with a constant 50% in stocks and 50% in bonds (50/50) and the right column represents Rob's VII:

    Year 50/50 Rob (VII)
    1987 2.93% 2.93%
    1988 11.79% 11.79%
    1989 22.50% 22.50%
    1990 2.67% 2.67%
    1991 22.74% 22.74%
    1992 7.28% 7.28%
    1993 9.79% 9.73%
    1994 -0.74% -1.70%
    1995 27.82% 23.00%
    1996 13.23% 8.41%
    1997 21.32% 15.38%
    1998 18.60% 13.59%
    1999 10.16% 4.70%
    2000 1.17% 6.28%
    2001 -1.80% 3.32%
    2002 -6.95% 0.66%
    2003 16.24% 10.10%
    2004 7.49% 5.87%
    2005 3.59% 2.99%
    2006 9.96% 7.11%
    2007 6.16% 6.54%
    2008 -15.99% -5.47%
    2009 16.21% 16.21%

    50/50 23-year annualized return = 8.47%
    VII 23-year annualized return = 8.30%

    So it appears that adjusting the stock allocation in response to big price changes did not produce higher returns. The valuation-adjusted returns were 8.30% annualized over the 23 year period from 1987 through 2009. This is slightly lower than a 50/50 Buy-and-Hold portfolio which produced annualized returns of 8.47% over the same period.

    I will note that there are other ways to Buy-and-Hold. An investor has other stocks fund options besides the S&P 500 index fund. For example, Wellington fund is a fine balanced fund. And the Coffeehouse Portfolio is an example of using multiple stock asset classes.

    To see how both the Wellington fund and the Coffeehouse Portfolio compares with Rob's VII, see this link . . .


  16. So it appears that adjusting the stock allocation in response to big price changes did not produce higher returns.

    It's obviously always possible to find a small number of cases in which a given strategy "works" on a looking-backward basis. That's not the proper way to test the concept. No one knows in advance whether that one lucky scenario is going to turn up in the future or whether some other more likely returns sequence is going to turn up. An effective strategy is one that works in a high percentage of the possible scenarios.

    I have a calculator at my web site called "The Investor's Scenario Surfer" that permits unlimited testing. The calculator generates random, realistic 30-year returns sequences and the investor can change his allocation in response to price changes as he sees fit and compare the 30-year results with the results obtained from a rebalancing strategy. Valuation-Informed Indexing strategies have ended up ahead (in some cases very far ahead) in roughly 90 percent of the tests I have run.

    So there are cases in which Buy-and-Hold produces a better long-term result. But at what price in terms of risk? A strategy that puts you ahead in only one out of 10 cases in a high-risk strategy compared to one that puts you ahead in nine out of 10 cases. I think it is fair to say that, on a risk-adjusted basis, Valuation-Informed Indexing always leaves you ahead.

    Here is a link to the calculator:

    Scenario Surfer Calculator


  17. Rob, I don't think that randomly generated returns (regardless of the distribution you are using) can provide a convincing test of your claim. What you would need to do is to use historical data (as Schroeder has done) with multiple starting points and horizons. But even this is not enough: the P/E thresholds you choose for switching portfolio composition must be such as to generate on average over time the same asset allocation as the buy and hold strategy. In other words, you can't pick the critical P/E thresholds (12/20) and the asset allocations (25/50/75) independently: they have to be selected jointly to match the buy and hold asset allocation over long horizons.

  18. Rob's Response -- Part One

    Thanks for adding your input, Rajiv. We have been having extensive discussions of these issues at other places for eight years now. You obviously have more authority to address these questions than many of those (including me!) who have participated at discussion boards and blogs. So you are helping a lot of people out by engaging the debate a bit here.

    I do understand, however, that the merits (or lack thereof) of Valuation-Informed Indexing are not directly (but only tangentially) related to the point of your blog entry. If you want to close this aspect of the discussion at any time, I fully understand. I am of course always open to questions about the Valuation-Informed Indexing strategy at my blog and encourage anyone with questions to ask them there. That said, I will make an effort to respond to your points here for so long as you are okay with that.

    I disagree when you say that randomly generated returns should not be used to test the claims. The reality (as I understand it) is that stock returns always contain an element of randomness. There are two major factors that influence stock returns: (1) economic productivity, which in the U.S. has always been strong enough to finance an average long-term return of 6.5 real; and (2) investor emotion, which is evidenced in overvaluation and undervaluation and which either adds to or subtracts from the 6.5 return during certain time-periods.

    The 6.5 return proceeds like clockwork. The proper value of stocks is the discounted value of all future returns. For so long as U.S. productivity remains roughly what it has always been, that 6.5 number is going to continue to apply. But the adjustment that must be applied to the 6.5 number to cover the effect of investor emotion (evidenced in valuation shifts) is a wildly changing thing. It is totally unpredictable (so far as I can tell).

    There are two factors, a largely predictable factor (economic productivity) and a highly unpredictable (or random) factor (investor emotion). The ONLY way (in my view!) to assess the validity of an investment strategy is to use an analysis that considers BOTH types of factors.

    Stock returns are to some extent random (because unpredictable investor emotions play a role in setting them). Therefore it is only through the use of (partly but not entirely) randomly generated returns that one can assess the efficacy of any investing strategy.

    Schroeder's result is cherry-picked. There are always some return patterns in which Buy-and-Hold works (the calculator shows that this happens in about one in ten of the possible returns sequences). An analysis that permits cherry-picking of results is always going to be able to produce a "showing" that Buy-and-Hold works. If starting the analysis in 1987 does not produce the result you like, you could start in 1982 or 1991. If using bonds as the non-stock asset class doesn't do the trick, you could use some other asset class.

    An analytical approach that permits such easy doctoring of the data is not convincing, at least not to me. By using randomly generated returns sequences (that conform to the general patterns that have always applied in the historical record), you greatly increase the number of returns sequences that can be examined and thereby solve the problem of cherry-picking of results.

  19. Rob's Response -- Part Two

    I half agree with your point about the P/E thresholds and I half do not. Please understand that I reject out of hand the idea that there is one correct switching strategy. The idea of going from 75 percent stocks to 50 percent to 25 percent is a reasonable one (one that was once suggested by Benjamin Graham). But I believe strongly that the choice of a switching strategy must be personal to the investor. Investors with different life goals and different financial circumstances and different risk tolerances should be going with different switching strategies. There is no one right switching strategy that can be tested to see if Valuation-Informed Indexing works because there is no one right switching strategy for all investors.

    You make an important point in saying that "the P/E thresholds you choose for switching portfolio composition must be such as to generate on average over time the same asset allocation as the buy and hold strategy." I understand from an analytical perspective why you say this. But there's a problem with the idea. A P/E10 value of 20 that comes up during a time when the long-term direction of stock prices is downward does not signify the same thing as a P/E10 value of 20 that comes up during a time when the long-term direction of stock prices is upward.

    Overvaluation in the stock market causes a massive misallocation of resources. The primary reason why we are in an economic crisis today is that stocks were overvalued by $12 trillion at the top of the bubble and millions of people were believing that their wealth was much greater than it was in reality. Trillions of dollars were misspent on cars and vacations and houses. Millions of businesses were started that the owners thought were properly capitalized but which were in reality wildly undercapitalized. Those millions of businesses are now in the process of failing. The millions of investors who spent money that they did not possess are now afraid to spend because they have come to the realizations that their retirements are wildly underfunded.

    These realities did not apply when we went to a P/E10 level of 20 in the early 1990s. Stocks were overvalued then too. But had we then informed investors of the risk of investing in overvalued stocks, that would only have brought the P/E10 value to 14 or so. We are now in circumstances where it is likely that a realization on the part of investors of what they have done to themselves is going to bring the P/E10 level to 7 or 8 (we have fallen to 7 or 8 in every earlier case in which we permitted the P/E10 value to exceed 25). So there never can be one stock allocation that applies for one P/E10 level. We cannot say that a P/E10 level of 20 is a P/E10 level of 20 is a P/E10 level of 20.

  20. Rob's Response -- Part Three

    I am truly grateful for your feedback Rajiv. You are of course making intelligent points. My sense is that you would like to quickly prove things one way or the other. That is of course a wonderful ultimate goal. However, I do not believe that it can be done in one step. What we need to do today is to lift the Ban on Honest Discussion of these matters.

    There is today a Social Taboo that blocks many people who could be participating in helpful discussions from doing so. There is a thought that it is somehow "rude" to say that there are grounds for believing that Buy-and-Hold never works. When that changes (let us all pray that this happens soon!), there will be all sorts of analyses put forward and a huge Learning Together process will begin.

    I don't have all the answers, Rajiv. I did not go to Stock Investing School. I have never managed a huge mutual fund. I have not studied economics, as you have. I am some guy who posts stuff on the internet. I am a journalist by background. On the journalism question, I believe that I can speak with some authority. The unwillingness of many (including many in your profession) to speak in plain and frank and open terms about these questions has created an exceedingly unhealthy situation. I have never before seen anything remotely like it. As a society, we have put ourselves in a position of great danger and then refused ourselves the permission to figure out how to get out of it. That aspect of all this needs to change. I am sure of that much.

    When that changes, there are going to be all sorts of people examining these questions from all sorts of angles. I doubt that I will have much to say when there are many people better qualified than me participating. The reason I am leading this initiative today is that no one else is doing it. We all need to begin talking about the implications of the finding (30 years ago!) that valuations affect long-term returns. Those implications are huge. All that I am trying to do (as a mild-mannered reporter) is to get the ball rolling in the right direction.

    Again, thank for taking the time to engage here. You have done more than many others have done and I am sincerely grateful.


  21. Rob, I'm traveling at the moment so this will be a brief response. The problem with using randomly generated returns to test your claim is that you are either forced to assume stationarity (returns in successive periods are drawn from the same distribution) or to specify some ad hoc rule by which the distribution itself changes over time. So you can't allow for realistic structural change, that can arise through changes in laws, taxes, institutions, financial innovation, or even changes in the composition of trading strategies. Using historical data is a bit better, but not perfect either because it doesn't deal with future structural change. In my view the best approach is to build an agent-based simulation model (as discussed by Farmer and Foley; see my last post) with competing trading strategies and endogenously determined returns. You can allow earnings to be randomly generated, but prices should be the outcome of trading strategies that are heterogeneous and interpret earnings data in different ways. Then one can try to identify circumstances under which your system works and those in which it doesn't.

  22. Rob's Response -- Part One

    Thanks, Rajiv.

    I'm all in favor of testing as many things possible in as many ways as possible. That's how we learn. So I would like to see many smart people (smarter than me, I hope!) get involved in efforts of that sort.

    My message (which as I note above is more of a journalist's message than it is an economist's message) is that for this learning process to begin in earnest we first must bury this idea that has somehow caught on that Buy-and-Hold has been scientifically proven to be the best strategy. Until we do that, the widespread belief in demonstrably false (according to me -- but at least possibly false in the eyes of any reasonable person) ideas cuts off all initiatives to move forward.

    We already have a mountain of evidence that valuations affect long-term returns. Shiller published research on this in 1981 and there has been a lot of follow-up in the decades since. But there are retirement studies available on the internet today that do not make adjustments for the valuation level that applies on the day the retirement begins. These studies do not even note that there are many smart people who believe that such adjustments are needed to get the numbers right. It gets worse. I have gone to places where these "studies" were being discussed and noted the need for valuation adjustments and lots of people saw the merit in that and yet further discussions were banned. Is this not odd? What's going on?

    What's going on is that today there are two main schools of thought about how stock investing works. One school is the Buy-and-Hold School. That is rooted in the Efficient Market Theory, which posits that the market always prices stocks properly. The other model is the Behavioral School or the Rational School, which posits that investing is done by humans and therefore human emotions (which can cause stocks to be wildly mispriced for long periods of time) must be taken into consideration.

    One of these schools is in an ultimate sense right and one is in an ultimate sense wrong. They cannot both be right because the two models begin from opposite premises (either mispricing cannot exist or mispricing matters). It seems to me that all of us (economists, journalists, stock experts, investors) need to come to terms with this reality. We all have opinions as to which model is right but not one of us knows with absolute certainty. So we need to launch a national debate aimed at bringing us to a better understanding.

    What is going on today is that we are ducking these questions. There is a Social Taboo against saying what I say in the title of my Knol, that Buy-and-Hold Cannot Work. But that is one of the views that must be on the table if as a community of people we are to come to terms with all of the questions and come to satisfactory answers.

    You said something remarkable in your February 17 update to this blog entry, Rajiv. You said that my claim that long-term timing works "could be true." That's fair enough. Skepticism re bold claims is proper and healthy. But that statement alone should change the history of investing (and would if all the others who believe this would say so in clear and understandable terms). I can tell you from my experiences discussing these matters with tens of thousands of middle-class investors that the vast majority of them do not believe that any form of timing can work. They believe things that are not necessarily so. They are investing according to beliefs that do not necessarily stand up to scrutiny (my personal belief is that they in fact do not stand up).

  23. Rob's Response -- Part Two

    The economics profession has played a role in creating this extremely unfortunate state of affairs. There is obviously a lot of evidence that the market is not immediately efficient. Buy-and-Hold is rooted in a belief that the market is efficient; if the market is not efficient, there is no justification for ignoring price when buying stocks. The economics profession has done a poor job of getting the word out to the middle-class investor that the case for Buy-and-Hold has been seriously undermined by the academic research of the past 30 years.

    The sense that I get is that many in this field believe that the case that Buy-and-Hold cannot work has to be 100 percent proven before the arguments that this is so can even be put forward publicly. It of course has to happen the other way around. We cannot prove something if we cannot even discuss it as a possibility. What we should be doing is telling people that there are some smart people who believe that the market is immediately efficient and that Buy-and-Hold is a good strategy and there are other smart people who believe that the market is only gradually efficient and that Buy-and-Hold cannot work.

    If Buy-and-Hold is a mistake (and there is certainly strong evidence that this is so), the consequences could be truly frightening. Stocks were overvalued by roughly $12 trillion at the top of the bubble. If overvaluation is a real phenomenon (overvaluation can of course not exist if the market is immediately efficient), we are in the process of seeing millions of people lose $12 trillion from their portfolios. It's not hard to understand why we would be in an economic crisis if the efficient market theory really always was just a horrible bad guess.

    We don't need to rename the Efficient Market Theory, in my view. We need to figure out a way to have civil and reasoned discussions about both what is good in it and what is bad in it and then quickly share with everyone what we come to know so that the new knowledge that we have been acquiring for three decades now can do practical good. If there is even a small chance that the Efficient Market Theory is invalid, the millions of people who have put together retirement plans pursuant to studies assuming that the EMT is valid need to be put on notice that they need to return to the workforce.

    My bottom-line point here is that this isn't just an academic discussion. There are flesh-and-blood people affected by the extent to which we as a community permit discussions of investing strategies not rooted in a belief in the Buy-and-Hold Model. I believe strongly that once serious people published research throwing doubt on the intellectual framework for Buy-and-Hold, questioning of the wisdom of Buy-and-Hold strategies should have commenced everywhere. It's been 30 years now and this has not yet happened to any significant extent.

    The average investor does not today know how dubious a strategy Buy-and-Hold is. That needs to change. It may be that it is too soon to tell people that Valuation-Informed Indexing is the answer. But it is certainly not too soon to begin the discussions needed to develop new approaches. The economic crisis which many have long predicted was going to come as a result of the continued reckless promotion of Buy-and-Hold is now a current-day reality.

    The market is not "invincible" if long-term timing works. You acknowledge that my claim that long-term timing works "could be true." We should not be telling people that the market is either efficient or invincible if these things may not not so; that would be dangerous and irresponsible. We should be telling people that we do not know for sure today what works in investing and that we support efforts to launch a national debate aimed at figuring things out together.

  24. Rob's Response -- Part Three

    My "system" is just to accept the reality that valuations affect long-term returns. I am entirely non-dogmatic re the question of how investors take that reality into consideration. I of course have my own favorite strategies. But so long as investors know not to follow the Buy-and-Hold approach (which holds that there is no need whatsoever to take valuations into consideration when setting one's stock allocation), I am happy. My view is that Buy-and-Hold is the most extreme of all possible strategies and that millions of middle-class people are buying into it because it is promoted as being "scientific" and that this misperception is causing great human misery. There is nothing "scientific" in the year 2010 re claims that there is no need to take valuations into consideration when setting one's stock allocation. It's an opinion that some smart people hold. But it's not science, not in a world in which there is so much academic literature pointing the other way.

    Your fundamental point (as I hear it) is that the scientific case for the Rational Model is not yet perfect. That's fair. My point is that we are not going to get there without first acknowledging that the scientific case for Buy-and-Hold hasn't been perfect for a long, long time either. People should be able to hear both sides and that is today practically impossible. The economics profession could play an important role in changing this unfortunate state of affairs, in my assessment. We all should be working together on this in a spirit of respect and affection and cooperation.

    I again thank you for your willingness to engage in some back-and-forth re these matter, Rajiv. Your fair-minded and sincere and helpful challenges force me to sharpen my thinking re where I am coming from re all this.


  25. These discussions seem to imply that simple allocation strategies (based on "valuation" as proxied by crude accounting metrics such as P/E) provides significant alpha. If this is the case, why don't you just start a hedge fund, lever it up and attempt to corner all the wealth in the world? or at least, demonstrate consistent alpha?

    Academic prescription of "buy and hold" also is coupled with a diversified portfolio with an allocation that matches the individual's consumption needs. It is yet to be proven that a balanced portfolio of stocks, bonds, commodities and real assets is inferior to any other concoction - however fancy the mathematics is. This observation leads us back to the fact that markets are reasonably efficient.

  26. Why don't you just start a hedge fund, lever it up and attempt to corner all the wealth in the world? or at least, demonstrate consistent alpha?

    Using leverage to profit from overvaluation would likely prove to be a disaster, Gill. Stocks reached insanely dangerous valuation levels in early 1996. We didn't see a stock crash until late 2008. The data indicates both that long-term timing always works and that short-term timing never works.

    It is yet to be proven that a balanced portfolio of stocks, bonds, commodities and real assets is inferior to any other concoction

    How about cash or other super-safe investment classes?

    In January 2000, when stocks were priced to provide a most likely annualized 10-year return of a negative 1 percent real, Treasury Inflation-Protected Securities (TIPS) were paying a guaranteed 4 percent real. My view is that it makes sense to invest more heavily in the asset class paying 4 percent real than in the one paying a likely negative long-term return.

    This observation leads us back to the fact that markets are reasonably efficient.

    I certainly acknowledge that there are lots of good and smart people who believe this, Gill. There are also lots of good and smart people who do not. My personal view is that the market is not immediately efficient but becomes efficient gradually over the course of about 10 years.


  27. Rob: Just debating..

    "Using leverage to profit from overvaluation would likely prove to be a disaster."

    I use this as a test for people who say they have a superior investment strategy. If the strategy is superior - i.e, there is an intrinsic alpha - that alpha should be levered. If one is not willing to do that, it implies that there is some risk that is not accounted for. This risk, indeed is a jump and so any common place strategy such as low P/E or even more sophisticated ones such as the Fama-French model may show alpha in regimes that are smooth and but will wipe everything out in a discontinuity. This basic fact is not understood by those shouting on TV or elaborating stratagies in the press. The existence of alpha in common strategies over specific time windows, does not mean much. This is why a well diversified and static portfolio is still the best.

    "How about cash or other super-safe investment classes?"

    A diversified portfolio should include real assets - such as patents, R&D and other types of IP positions. Most of the value in the economy is generated from new stuff (and not from quarterly financial statements of inefficient companies showing low P/E). So, a balanced portfolio in this way will always over perform cash. Treasuries carry risk (regime risk) and no country is immune from it - US included.

    "This observation leads us back to the fact that markets are reasonably efficient."

    No sure this has anything to do with smartness. This is probably te easiest concept to understand.

  28. Rob: Just debating..

    Absolutely. That's precisely the right spirit to take to this in my view, Gil.

    I use this as a test for people who say they have a superior investment strategy. If the strategy is superior - i.e, there is an intrinsic alpha - that alpha should be levered.

    All that I feel that I can say is that I believe it would be a disaster to take on leverage because of a belief that the market is overvalued. I don't see how leverage could be a good idea when we are talking about something that only works in the long term.

    Could it be that your test was developed by a mindset that is focused on the short-term? I don't mean that as a dig. My sense is that Buy-and-Hold was a first-draft effort to move us from a short-term focus to a long-term focus and that the people who advocate it have not yet thought through all the implications of what it means to make the transition.

    This is why a well diversified and static portfolio is still the best.

    This statement begs the question, Gil. You believe that a "static" portfolio is one that always maintains the same stock allocation. I believe that the thing that you should want to be "static" about is your risk profile. If valuations affect long-term returns, then your risk is changing with changes in valuations. So, to be "static" in a meaningful sense, you would need to be willing to make changes in your stock allocation in response to valuation shifts.

    a balanced portfolio in this way will always over perform cash.

    That's the theory behind Buy-and-Hold. The Rational Investing theory is different. The Rational Investing theory posits that the price you pay affects the long-term return you obtain.

    What you are describing makes sense from one way of looking at things. You are describing what would be so if only investors were rational. But all investors are humans. How many humans do you know who never engage in irrational behavior? I have known people who have destroyed their marriages, their jobs, or their health for "reasons" that made no sense to objective observers. Is it so hard to believe that these same humans would invest in ways not in accord with what the books say should be done?

    For so long as investing is done by humans, it is possible for stocks to be so overpriced that the long-term return on super-safe asset classes can for a time be far higher. Or so the theory that I am putting forward maintains.

    Not sure this has anything to do with smartness.

    My strong sense is that it has a lot to do with what sortof smartness an individual possesses. Every advocate of Buy-and-Hold that I have met has been smart in an intellectual sense. But most have been "smart" in the ways that engineers and accountants and economists (!) are "smart." They like logic, they like numbers. They don't always feel comfortable talking about things like human motivation. They are not necessarily "smart" in the ways that novelists and detectives and journalists (!) are smart.

    I have known people who are not able even to describe these ideas accurately after years of being engaged in discussions of them. I have known others who "get it" the first time they are exposed to the ideas.

    I believe that one of our problems is that the people who we look to for help in understanding investing tend to come too much from one or two of the basic personality types. I believe that we need more diversity in this field re the types of intelligence being brought to bear on the various questions.

    Thanks for asking some good questions, Gil. I hope I have provided responses that appear sincere and at least reasoned in some foggy sort of sense, even if they are not entirely convincing to a good number.


  29. Rob:

    A balanced portfolio (not based on valuations such PE ratios) but based on the individual's consumption needs is dominant. "Balanced" means that it has to have all investable assets - including private assets. Portfolio will have to be balanced over time as the investor’s needs change (and not because IBM has low PE or Cisco has high PE)

    Regarding rationality, I think there is a lot of confusion around this. Market efficiency does not require all individuals to be rational. It just requires a large number of participants with access to information and capital.

    The fact that individuals do stupid things in life is not a sufficient reason to assume that market prices are not in equilibrium. The market price is determined by supply and demand. Assuming away liquidity constraints, fraud and regulation, price in the market is the equilibrium price.

    To generate an alpha, by investing in public markets, the investor has to have preferential access to information (illegal) or knowledge that is special (unlikely). If you believe you have a system or process that creates alpha consistently in all time windows, all the wealth in the world should belong to you.

    There are two important items here.

    (a) It has to be alpha (not returns)
    (b) It has to be positive in all time windows (not just in some)

    Most strategies are missing one or the other.

  30. To generate an alpha, by investing in public markets, the investor has to have preferential access to information (illegal) or knowledge that is special (unlikely).

    Anyone who takes valuations into consideration when setting his or her stock allocation today has the benefit of knowledge that is "special," Gil.

    The forum is the largest investing forum on the internet. They are today celebrating their third birthday. The forum was founded to escape me. I used to post with all the others there at the Vanguard Diehard board at and I would point out that valuations matter and the Buy-and-Hold advocates would get angry and ask that I be banned and Morningstar refused to ban me. So they set up a separate board and today they have all their discussions free of the input of Rob Bennett, who would annoy some there by pointing out the effect of valuations.

    Do the people there know about the effect of valuations? How could they? Discussion of it is prohibited. So do those who discover by some means or another that valuations matter come to possess "special knowledge"? Absolutely.

    You are working under the presumption that "special knowledge" must be hard to come by. It isn't! It's sitting in front of all of our noses. A majority of us do not want to see it! It is special not because it is hidden but because it offends the sensibilities of those with a strong belief in Buy-and-Hold and they elect not to take note of it.

    That's all it takes for the market price to get wildly out of whack, Gil. It's not rational to ignore information that would make you a better investor. But it sure is human! Humans do this sort of thing all the time when pursuing all sorts of human endeavors. But the Buy-and-Holders assume that this basic rule of how humans behave cannot apply in the investing realm. Why?

    Have you ever known someone who was dating someone who was no good? Every one of the person's friends knows it but the person involved does not. How could that be? It's not that the person involved is not smart enough to figure out the realities. It's that he or she does not want to figure out the realities. We all possess a great power to rationalize away inconvenient realities. A man hears what he wants to hear and disregards the rest.

    This has nothing to do with alpha. It is human emotion that is driving this.

    The P/E10 value tells you how much emotion is present in the market price at a given time. Overvaluation is always irrational. If the valuation were rational, the prefix "over" would not apply. Overvaluation by definition cannot be taken into account by the market in setting the price of stocks.

    What we need to do (in my view!) is to encourage investors to respond to overvaluation by lowering their allocations. If they did this, the market price would be self-regulating. Each case of overvaluation would be met by selling, which would bring prices back to fair value. The only thing that can keep market prices on track is a general lowering of stock allocations in response to overvaluation. This is the very thing done away with through the promotion of Buy-and-Hold strategies!

    The great irony is that the market would be efficient if it were not for the promotion of Buy-and-Hold. The success of Buy-and-Hold depends on the efficiency of the market and the promotion of Buy-and-Hold destroys any hope that the market could be efficient. Yikes!

    All the wealth in the world should belong to you.

    From your lips to God's ears, Gil. I have a giant basket here that I think should do a good job of catching all the twenties and fifties and larger-denomination bills that you and I agree "should" belong to me. I'm ready, Lord!


  31. This comment has been removed by the author.

  32. For a sober assessment of why passive investing remains the best strategy for most investors despite modest violations of informational efficiency, see this post at Pop Economics.

    I like the post at Pop Economics, Rajiv. I think it moves the ball forward and I am happy to see you linking to it.

    What I like about the Pop Economics post is that it offers a non-dogmatic defense of Buy-and-Hold strategies. There are obviously millions of smart and good people today who believe strongly in Buy-and-Hold. Those people of course need to post their sincere views and Pop has given voice to views held by many with that post. He is offering a civil and reasoned defense of the Buy-and-Hold approach. That's a plus because that opens the door to the presentation of civil and reasoned presentations of the other point of view.

    There are people who very much do not believe what you believe and what Pop believes, Rajiv. Those voices too need to be heard. If we all could agree that the differences here are differences of opinion, we all could enjoy a wonderful Learning Together experience. The reason why I wrote the Knol was to help launch such a discussion.

    I obviously believe that the end result of such a Learning Together experience is going to be that Buy-and-Hold is going to fall. But others obviously believe otherwise and those others have every bit as much a right to express their honest views as I do to express mine. What we all need to do is to encourage those trying to engage in discussions in which respect and warmth are shown by those on both sides of the discussion to those on the other side. When we do it that way, things just get better and better and better over time.

    I am going to send an e-mail this morning to Pop asking him if he would be willing to post a Guest Blog Entry by me in which I would respond to the points he makes in his defense of Buy-and-Hold. I can also write a Guest Blog Entry here on any points that you would like me to address in the event that you see value in me doing that.

    As noted in some of the comments above, these discussions have been going on for a long time. I believe that we have developed some wonderful insights. But a good bit of friction has evidenced itself in these discussions. The core problem is that the question being addressed -- whether there is a type of market timing that always works -- is so fundamental to the investing project.

    If I am right that long-term timing always works, that changes everything. In a positive way. But still it shakes people up to learn that pretty much all that we once believed about stock investing has turned out to be wrong.

    It's important that people on "the other side" try to appreciate that there has never been any evidence put forward that long-term timing doesn't work. That was always just a guess. There was lots of evidence put forward that short-term timing doesn't work and smart people just guessed that it's the same with long-term timing. If it's not (the difference is explained by a conclusion that the market is not immediately efficient but only gradually efficient), then there are lots of exciting things for us to discover together in the days ahead.

    I am grateful for what you have done here in opening up this discussion to some people who were until now not aware of it.


  33. Rajiv,

    Could you explain what the policy implications would be given that we know that information efficiency does not necessarily lead to allocative efficiency?