If you’ve researched the theoretical foundations of index investing, you’ve no doubt come across Modern Portfolio Theory and the Efficient Markets Hypothesis. And if you read the commentaries of active money managers and the financial media, you’ve probably seen countless articles that dismiss both as obsolete. Modern Portfolio Theory is declared dead after every market crash, and all stock pickers, almost by definition, believe markets are not really efficient. Many of these critics think passive investing is folly—only the warm embrace of active management can protect you and your portfolio.
In his provocative book, Risk, Financial Markets & You, the Winnipeg-based financial advisor Alan Fustey adds his own criticisms of these two decades-old models. But his conclusion is surprising. When I interviewed him recently, I asked what investors should do if these models were broken. “Well, the first thing you do,” Fustey replied, “is you index.”
The background
Before going further, let’s review these two landmark financial theories, both of which revolutionized investing. Modern portfolio theory was devised in 1952 by Harry Markowitz, who later shared a Nobel Prize for his contribution. Markowitz showed that by combining risky assets that have less than perfect correlation, you can create a portfolio that has lower risk and a higher expected return than its individual components. Index investors understand this as the “free lunch” offered by diversification.
The efficient markets hypothesis was formulated by Eugene Fama in a 1970 paper. It suggests that security prices reflect all public information, and therefore analysis of individual companies does not allow investors to identify “mispriced” stocks. (Occasional mispricings probably exist, but they cannot be reliably exploited.) In an efficient market, buying an index fund is the best way to get market exposure at the lowest possible cost.
‘You’ve got to have a better alternative’
These two ideas are among the pillars of the Couch Potato strategy. But as Fustey explains in Chapter 4 of his book, both have some shortcomings. For example, he says, MPT presumes that asset class correlations remain constant, and that market returns follow a normal distribution (represented by a bell curve). We know neither of these assumptions is valid. He also points out that it is possible for some participants to have an informational advantage, so capital markets are not truly efficient.
But if you think Fustey is building a case against passive investing, think again: he argues that indexing still offers investors their best chance of success. “It comes down to this,” he told me. “If you’re going to throw out these ideas, then you’ve got to have a better alternative. If you reject the efficient markets hypothesis, you’re saying that you have the ability to get superior information, and that you can make money from that knowledge. But that flies in the face of reality. Someone is always going to have better information than someone else, but are you going to be able to profit from that on an after-fee basis? I don’t think so—and the history of active management proves that.”
Book giveaway
Alan Fustey is a CFA and a portfolio manager at Index Wealth Management, a Winnipeg firm that also has offices in Calgary and Vancouver. I recently added the firm’s three offices to my Find an Advisor directory.
Alan sent along two copies Risk, Financial Markets & You to offer to readers who are interested in learning more. Leave a comment below to be entered in the draw for the books. Contest closes at midnight EST on Wednesday, May 2. I will announce the winners in Thursday’s post.
Speaking of contests, a big thanks to everyone who requested a ticket to PWL Capital’s event featuring Carl Richards, author of The Behaviour Gap. The answer to the skill testing question was Buckingham Asset Management (BAM Advisor Services was also acceptable). Winners will be notified by email.
Dan,
Thanks for the summary of EMH and MPT; I’ve read about them before but every retelling adds a slightly different perspective.
I’ve added Alan Fustey’s book to my reading list on account of your article. In light of global financial uncertainty, it (and your model portfolios) couldn’t be more timely.
-CJ
I’ve been watching the indexes, and my impression is that indexing works well for the U.S. but not for Canada. The TSX is so heavily biased toward resources that indexing works well in a decade where commodity prices rise, but not well if you have a decade or two when commodity prices fall.
@Beardie: To follow up on Fustey’s comment, then, if indexing does not work in Canada, what strategy should you use in order to achieve index-beating returns? Remember, you have to identify this in advance, not look back 10 years from now and declare what would have worked.
So basically he’s saying: “Theoretically it works, in practice it doesn’t, but the alternative is even worse so you should do it anyway.” This may be raising eyebrows but I guess it makes sense…
Is Fustey’s book a good read for a novice investor? I completely devoured Dan’s book as well as Andrew Hallam’s Millonaire Teacher and I’m looking for something else to read while commuting :)
Hi Dan,
Great job as always. Keep up the good work. Please add me to the draw for the book.
Thanks
@Jason S: I would frame it more this way. Democracy is far from perfect, but that doesn’t mean we should toss out the whole idea in favour of monarchy or communism. It’s still the best model we have, even with its imperfections.
Alan’s book is pretty advanced, and I wouldn’t say it’s s great fit with a novice investor. Jason, you might like this book for your commute:
https://canadiancouchpotato.com/2010/07/12/review-the-gone-fishin-portfolio/
Shocking results! Index funds win again!
Please count me in for the draw.
To butcher a quote from Winston Churchill: “It has been said that indexing is the worst form of investing except all the others that have been tried.”
Excellent post, as always. Please count me in for the draw.
@CCP For Canada, you would have to overweight the sectors that have proven themselves to be good investments over 30-40 years and underweight the others. Try a mix of XIU, XDV, and XFN.
Please include me in the draw. Thanks Dan
Please add my name to the draw. Thanks.
compare me to the possible not perfection
Do you have any information regarding the fact that market returns doesn’t follow a normal distribution. A sum of random variables follow a normal distribution. The efficient markets hypothesis suggest that stock movement should follow random movement. Thus following a normal distribution.
Count me in the draw please.
Would like to be entered in the draw. Index Funds all the way!
Please add me to the draw. This site is a wonderful resource. Keep up the great posts!
Cam
@Sebastien: The sum of random variables approach the normal distribution only when their means are finite. In serial (yearly) returns, we are summing the logarithms of returns, and there is no guaratee that the means are finite. For example, a finite probability of going bust makes the log mean undefined. I’d suggest reading Mandelbrot’s book The (Mis)behavior of Markets for more on this. Much of what Taleb wrote about Black Swans, etc., began with Mandelbrot’s ideas.
Always interested in reading about investment theory!
Hi Dan,
Thank you so much for your work on this site. It’s a great source of guidance for young investors like myself.
In this article, you say: “It is possible for some participants to have an informational advantage, so capital markets are not truly efficient.” I’m curious about this. What sort of “informational advantage” could a person have (aside from inside information, which I understand is illegal to act upon.)
“…participants to have an informational advantage, so capital markets are not truly efficient.” I believe this statement is correct, just ask Nancy Pelosi and others in U.S. government. Not sure the “Stock Act” will curtail!
Very interesting. Please include me in the draw. Thank you!
I’m new to this whole investing business and I did a LOT of research before plunging in – and I chose to go the route of indexed funds. I’m still building my diversified portfolio but with the exception of a personal weakness for the precious metals so I split my emerging market portion 50/50 with precious metal indexes, I am can see the logic. The future concerns me – I tend to feel there is a LOT of money out there in the financial world and its just sloshing around…. But there are very few alternatives to grow money safely (with the low interest rates effective killing any form of ‘guaranteed’ savings and western government loath to raise rates in case it drives more home buyers into debt distress) and that big bubble of money is never going to be perfectly stable and working equally for stocks, bonds, savings and any other vehicle… So, a diversified index fund portfolio such as that recommended by Coach Potato makes sense to me – to reap the rewards in the direction the bubble of money leans as big investors lean one or the other and rebalance accordingly.
If there are ways to tweak that in the book, I’m in… I’ve got a lot of time to make up!
So indexing seems to be a reasonable tool for portfolio management…
One assumes that this is a reflection of the fact that that indexes and overall market performance are driven predominantly by institutional investors. One assumes that, for the most part, the portfolio management decisions of these leviathans are guided by MPT and the EMH. Their size entails attendant limitations on their flexibility (i.e., on their ability to respond quickly to new information in the system). Does this not leave that leave room for the retail investor to rationally exploit volatility on both a stock-by-stock and sector-by-sector basis without lapsing into the realm of mere speculation, and with reasonable prospects of outperforming both institutions and indexed ETFs?
@Sebastien: A sum of random variables reflects whatever the underlying distribution is. e.g. A sum of random variables according to a Poisson distribution will not follow a normal distribution.
There is not a lot of empirical evidence for the random walk hypothesis. If you’re interested in a rigorous treatment of the hypothesis, I recommend A Non-Random Walk Down Wall Street by Lo et al.
@Michael: Normal distribution are, indeed, not defined under infinite mean. But your point about the mean not been finite is complete mathematical non-sense. Think of every day variation as a random variable with a certain positive mean. Summing theses random variables (under the hypothesis that they are) will yield a normal distribution. This statement is mathematically true and non- arguable. Whether every day variation in the market is a random variable with some mean is arguable. In my understanding of the efficient markets hypothesis, every day market movement should follow random variable.
@Cody: In the book, Fustey gives the example of brokerage firms that “co-locate” their computers closer to those of stock exchanges so they get information and execute trades faster. Another example is micro-cap stocks that may have no analysts following them closely. These inefficiencies exist, but retail investors can’t benefit from them after accounting for costs.
@Chris: The sum of many Poisson distributions would appraoch the normal distribution. In the limit of an infinite number of such distributions, the sum is normal.
@Sebastien: Have a look at the Cauchy distribution (http://en.wikipedia.org/wiki/Cauchy_distribution). It’s mean is undefined and the sum of many of them does not approach the normal. This was one of the points that Mandelbrot made in his book.
I could not agree more with your post, please enter me for the draw.
The link to the article by William Sharpe below also explains the point of your post quite well. I believe you already refer to it in the past.
http://www.stanford.edu/~wfsharpe/art/active/active.htm
I’ve been a regular reader for a while now slowly building up my knowledge on the markets. Please include me in the draw, I’d love to check out these books as they sound quite interesting.
@Michale: I agree, but then the variables would be drawn according to a Cauchy distribution. Their would not be random variables in the sense of a maximum entropy distribution (i.e. http://en.wikipedia.org/wiki/Maximum_entropy_probability_distribution). Thus making my point that daily movement may or may not be random variables, but if they are, their sum must follow a normal distribution.
@Sebastien: I think we’re getting off-topic, but I can’t understand your logic. If you go to the bottom of the article you point to you’ll see that “not all classes of distributions contain a maximum entropy distribution.” Variables drawn from a Cauchy distribution are still random variables — it’s just that they have ‘wild’ properties such as not having a well-defined mean.
In any case, the main point I wanted to make is that it is definitely possible for markets to have a distribution that is not normal without violating the Central Limit Theorem that “under certain (fairly common) conditions, the sum of a large number of random variables will have an approximately normal distribution.”
Folks, let’s try to keep the language clean on this site. Next reader who mentions Poisson or Cauchy distributions will be banned. :)
What confuses me about indexing is the effect of asset valuations. Intuitively, it’s hard to buy assets when they appear overvalued like some bond funds right now. I understand that portfolio rebalancing mitigates that risk but the science of how often to rebalance is missing for me. Does this book shed some light on this?
@Jon Evan: I think once you accept that markets are mostly (though not perfectly) efficient, the whole idea of “overvalued” gets called into question. You have to ask yourself, why do I think that my view about the “correct” price for bonds is more accurate than the consensus view of tens of millions of market participants? And even if I am right and everyone else is wrong, how can I profit from my superior knowledge?
This would be a good question for all the people who shied away from “overpriced” bonds over the last three years and missed their higher-than-average returns over that period. Like Bill Gross, for example:
http://www.theglobeandmail.com/globe-investor/bill-gross-admits-mistake-in-dumping-us-debt/article2146053/
Please add my name to the draw. Thanks for the blog – I read every post.
In it to win it! Please add me to the draw.
Dan, to be a successful investor over the long term one must consider the emotional side of it when purchasing assets. You appear to be underplaying the reality of overvaluation and saying we should just ignore it because it’s impossible to get it perfectly correct. In daily purchasing decisions of anything (be it a car, house or dishwasher) surely everyone considers valuation and price is a normal consideration to buying anything: it’s human to consider that. The example is XBB. Over the period of 2005-2010 it’s price was relatively steady at ~29/unit but thereafter it rose to over 31 dollars! I can’t imagine there is much up side left and considerable downside with economies improving. Intuitively, I’m afraid to buy more XBB. This is the part of indexing which is difficult for me!
Hello Dan:
I do enjoy your blog and have learned a great deal here.
Please enter me for the book draw.
Phil
Thanks for yet another great post, this blog got me on the road to retirement savings about a year ago and I really appreciate the straight talk. Please enter me for the Draw.
@Michael, can you give an example of when the sum of a finite number of logs gives an infinite number? I assume you are using a finite number of data elements?!
@Jon: First let me acknowledge that bond returns in the foreseeable future cannot be as strong as they have been over the past 30 years—that’s just math, not forecasting. And if you want to keep your bond maturities shorter because that makes you more comfortable, that’s not a big deal. There’s no rule that says you have to hold a broad bond fund like XBB.
It’s just important to be aware that it is too easy to get caught up in this kind of thinking. Do you think stocks are overvalued, too? What about real estate? Gold? If you’re adjusting your bond portfolio what are you doing the other parts? If you end up being wrong (and you will be sometimes, guaranteed), will you overcompensate next time? You can see this gets to be a slippery slope, and very soon you’re an active manager.
Great article.
@CCP iShares really tempts you to be your own active manager with all the exotic ETFs they have.
i guess key is to identify what risk you are comfortable with and what return do you absolutely need. i am a conservative low beta value investor type and i do get into analysis paralysis quite a bit. that rules out individual stock, sector and country picking for me. passive indexing using e-series and RBC D series has been good for me. only thing i may add in the future is rebalancing based on 200 day moving averages. i think that would reduce the large losses (2008) which are hard to overcome. CC has been helpful in keeping me (kind of) passive!
@Adam: If one of the returns is -100%, which is a multiply by zero, the log of zero is undefined (or infinite) and the overall sum is infinite.
A “Random Walk Down Wall Street” by professor B. Malkiel tackles both the EMH and MPT with great thought and insight. Take a peak at his thinking, he is no slouch when it comes to taking arguments pro and con with these topics and laying out an understandable foundation.
Thanks CCP for running such an informative and practical blog.
Please enter me into the book draw.
Great post, and great blog.
Please enter me in the book draw.
Cheers.
Looks like an interesting read. Sign me up for the draw.
Please add me to the draw. I just took a look at the firm’s website and note that, in addition to index investing via ETFs they are prepared to engage in options trading, too. That suggests to me that, although this firm’s managers may be convinced index investors, they are by no means passive.