In prior posts I have taken swipes at traditional asset allocation, buy-and-hold investing, the Efficient Frontier, the Efficient Market Hypothesis and Modern Portfolio Theory (MPT).
Sure, I am trying to be provocative, poke a little at advisor complacency and provoke polemic on the comment boards…but I am also sincere. MPT relies entirely on investment history for investment analysis and conclusions. These tired and discredited methods are rubbish…and have cost investors trillions.
It is encouraging to see evidence of advisor post-mortems in progress as some advisors are seeking not to repeat the mistakes of the past. I was also entertained by John C. who cracked on the comment board, “What’s over 50 years old and still considered modern? MPT”
But pretty pie charts and Powerpoints are not so easily disposed of. As an anonymous critic incites, “The appeal of Modern Portfolio Theory in the investment advising community is its simplicity, graphic presentation value, and most of all, little or no investing judgment or skill is required; just pick, print, present, and hope; chasing efficient frontiers, hoping that investment history will somehow repeat itself, and just waiting for historical updates to generate new efficient frontiers to justify investment change.”
Nevertheless, some advisors are stubbornly standing by their man(tra).
Modern Lovers
Consider these edited comments that I received from Matthew K. in response to the Crisis = Opportunity post;
“MPT works. With the right allocation and systematic rebalancing to maintain percentages as well as in line with client’s goals, there is no lost decade. Markowitz knew what he was doing, and as an academic, he did not stand to profit…When MPT is juxtaposed with Daniel Kahneman’s Nobel Prize winning ideas on heuristics, you see how MPT does add value when used in line with client’s goals…Any classic definition of “Venture” includes the idea of risk taking. Where does that fit in CAPM or the efficient frontier?”
I cannot rebut a hopeless romantic, so let’s engage Matthew K. in a virtual volley with interlaced quotes excerpted from a FT article and a McKinsey interview with the especial epistemologist, Nassim Nicholas Taleb. Taleb is the author of two true investor instant classics and must-reads, Fooled By Randomnes and The Black Swan.
Taleb has a strong opinion on the matter of MPT and modern finance…and he is no modern lover:
MK- MPT works. With the right allocation and systematic rebalancing to maintain percentages as well as in line with client’s goals, there is no lost decade.
NNT-We learn from crisis to crisis that MPT has the empirical and scientific validity of astrology, without the aesthetics…In 1990 William Sharpe and Harry Markowitz won the prize three years after the stock market crash of1987, an event that, if anything, completely demolished the laureates’ ideas on portfolio construction….I would ban portfolio theory immediately. It’s what caused the problems…Portfolio theory simply doesn’t work. It uses metrics like variance to describe risk, while most real risk comes from a single observation, so variance is a volatility that doesn’t really describe the risk. It’s very foolish to use variance.
MK-Markowitz knew what he was doing, and as an academic, he did not stand to profit…When MPT is juxtaposed with Daniel Kahneman’s Nobel Prize winning ideas on heuristics, you see how MPT does add value when used in line with client’s goals.
NNT-Academic economists are no more self-serving than other professions. You should blame those in the real world who give them the means to be taken seriously: those awarding that “Nobel” prize… Every time I have questioned these methods I have been abruptly countered with: “they have the Nobel”, which I have found impossible to argue with. There are even practitioner associations such as the International Association of Financial Engineers partaking of the cover-up and promoting this pseudoscience among financial institutions. The knowledge and risk awareness we are accumulating from the current subprime crisis and its aftermath will most certainly not make it to business schools.
Thanks, (virtual) Nassim. I will take the next one.
MK-Any classic definition of “Venture” includes the idea of risk taking. Where does that fit in CAPM or the efficient frontier?
VP-Of course, venture implies risk-taking… they are nearly synonymous. A venture investor is knowingly acknowledging and accepting an implicit and quantifiable serving of risk that is decidedly less than a range of positive (asymmetric return) outcomes. Perhaps investors would have been better served if their notion of the risk that they were assuming in their efficient frontiers was not muted (and implied to be mitigated) by the marketing machinations of MPT. CAPM is a future-oriented model yet it essentially relies on historic data to predict future returns. The Efficient Frontier? I have seen the inputs, I have seen the outputs…and I have seen the results…the efficient devastation of unsuspecting portfolios.
Album: The Modern Lovers, The Modern Lovers, 1976
Popularity: 57% [?]
I agree with Taleb, but if one completely dismisses mean variance optimization, what should one use to manage risk? Taleb disses MPT, which is fine, but what does he offer as a potential solution?
“It’s very foolish to use variance.” Does Taleb actually believe what he says or does he better understand the power of hyberbole for marketing than he understands risk. If only had he written just about any other time period in the 20th century, he be unknown and poor. But no doubt he’s this crisis’s Elaine Garzarelli (yes, her ego was almost as hard to stomach in 1988, but I do applaud her for having her bullish forecast for 2008 still front and center on her website. Fine, probably a google result she doesn’t know how to fix).
Yet I agree on the importance of (Taleb’s) general thesis: models are incomplete and humans are fooled by randomness. I just wish I had the talent to turn a few elementary concepts into two best selling books. Maybe his next book will be Garbage in, Garbage out. As you point out, VP, that reality makes CAPM much less useful and an overall reliance on getting to the efficient frontier a fool’s game. But as my friend Fish implies on his much more pithy reply, are we supposed to ignore the positive impact of diversifying your non-systematic risk– the core concept of MPT? My friends at Lehman would say no. And I suspect you would, too.
Hey VP,
If you want to base your strategy on speculation, emotion or intuition – that is fine with me. To each his own.
If you can suggest a better solution – I’m all ears?
regards,
Mike Bayer, CFP
Very fair comment, Mike…the ball is in my court. Posts thus far have been focused on problems, not solutions. Stay tuned, and thanks again for engaging.
There are a couple of alternatives. Some of the more recent work has pointed at diversification of time dependencies and trading styles as a better predictor of minimum covariance. Taleb’s dismissal of MPT seems to me to be based upon the correct notion that the distributional assumption is wrong, and on this point I would wholeheartedly agree. The issue of disclosing the better solution is sticky, as their is no incentive from the informed investors perspective, in increasing market efficiency.
Here’s hoping the better solution eludes you.
MPT is not “wrong” at all, and its practitioners are not villains. MPT is simply a portfolio optimization statement and has a variety of solution methods. It has a set of assumptions which after 50 years folks are starting to question. They can change assumptions, and there have been some recent published results showing how. (e.g., a JPMorgan report.) But the basic optimization problem remains valid, where the matrix of risk measures can be anything you can practically model.
I particularly want to say there is nothing in MPT dependent on the underlying distribution of returns, only on the assumption the various means and variances are finite. Nowhere is a Gaussian distribution assumed. In this regard, note that the theory and practice of wireless system engineering relies on the mathematics of white noise. Of course, white noise cannot exist as modeled, since white noise has infinite power. But engineers know how to approximate.
Rather than read repeated rants of criticism, I’d prefer to see some useful alternates from those who whine that MPT is wrong.
MPT may not be “wrong”, but it certainly isn’t right…and it may not have “failed”, but it certainly didn’t work.
Are the MPT practitioners “villians”? It depends on whether one sees them as accomplices or victims of financial fraud (I bias to victims). However, if after experiencing five financial meltdowns in the past 22 years (Black Monday, S&L Crisis, LTCM/Russian default, Dot-com bubble and the subprime/credit default crisis) those same practitioners don’t change their behavior…IMO, they are violating a public trust.
MPT outputs endorse allocating to a diversified mix of asset classes that are not perfectly correlated. When an MPT portfolio is constructed, correlation coefficients of asset classes are used in designing the portfolio. However, the reality is that correlations change – if you calculate the correlations for the prior 5-year period they will be different than the past 3-year period, for example. In the year 2008, all asset classes moved toward a correlation of 1.0, as all asset classes declined, except for treasuries. I know its hard for advisor to toss out those pretty pie charts–but they need to let it go.
The notion of diversification has value–but MPT places that value higher than more important core portfolio managment values such as safety of principal, liquidity and achieving growth through allocations to investments that are charactarized by their favorably-skewed risk-reward ratios.
We do agree about the need for alternatives. I have presented mine–I welcome your comments on Hybrid Portfolio Theory.
Thank you Lloyd, for taking the time to comment.
MPT is a house of cards built on faulty assumptions. MPT is based on A gaussian distribution as long as you use standard deviations. It is by nature a rearward looking process.
Diversification is a crutch for the poor performer. The whole notion of investing in stocks misses the main substance of investing. Stock are a medium of transfer, not an investment. The investment is in the underlying business. In order to perform well an investor needs to understand businesses. In this day and age this is a skill to few investors maintain.
MPT focusing on past performance, assuming rational actors and equating risk with volatility is bound to disappoint. It is a proven failed formulaic investment model that could be carried out by a 15 year old with a computer and the correct software.
VP statement: “investments that are charactarized (sic) by their favorably-skewed risk-reward ratios” says it all. Tilt the odds in your favor investing in businesses that you know and understand and are obviously mis-priced. With so many companies selling at less than their cash and investments you don’t even have to look too hard. (So much for efficient markets) Guess what the odds of success are when you buy a few of these companies. The ability to understand a balance sheet is much more valuable than anything MPT can tell me about these companies.
There are plenty of other ways to identify successful investments, however, most will find them too much work. MPT is easy, unfortunately there is little reward in life for easy, most often the good stuff comes with hard work.
All discussion always comes back to risk. And obviously “diversification” and all the measuring ratios and other analytics (mean variance optimization) are woefully inadequate for measuring, let alone mitigating, risk.
The basic problem is that all the practitioners out there who people entrust with trillions of dollars – their futures – have never focused on the rules of the game, only the results of the game.
Speaking of “risk” what if you sat down to play the board game “Risk” and did not know the rules of the game. What would be the expected outcome. So obviously before you play you would want to know the rules. Risk can be a challenging game, like Chess, but the rules are fixed.
Think of the stock market as a game. What makes it so difficult is that mathematically the odds are stacked against you, as losses count more than gains (you know the drill lose 50%….) Therefore risk management is rule #1, #2, #3…. But the market does not come with a rule book, it is not fixed, it is a multi-dimensional, it morphs with ever changing realities of our constantly evolving world.
If one’s best effort is to try to compare today to the past, in this game, that basis is all wrong, and always has been. However, the so-called Black Swan of 2008 should have forever proven that in spades.
Those clinging to the same pre-2008 notions on investing are truly lost. But for the masses who have lost and continue to lose their money, all they hear from their private advisers, or the public ones like “talk to Chuck,” is the same old refrain.
The conclusion to me is obvious. Either learn the game or get out.
Good trading,
Russ Mascieri
Isn’t part of MPT just common sense? If you get 499 friends (no, I don’t have that many) and you EACH invest $10K in ONE of the stocks of the S&P500, you’ll each bear more risk, with NO MORE TOTAL RETURN FOR THE GROUP than if you each bought $10K in an S&P500 index fund.
Isn’t that part of MPT and isn’t that irrefutable?
TS–
Your illustration does not speak to the core of MPT.
Rather, I would recall some of George P. Jackson’s sentiments (from “Death to the Pie”)…
• Past correlations are largely a product of coincidence, and probably won’t hold in the future. In particular, at the very time that you want your carefully designed portfolio to hold together, at the time of a great crisis, most things will suddenly become highly correlated as various sectors all move down in unison. You should not put too much faith in an “efficient” portfolio performing at all well if world markets go mad for a little while.
• Because volatility, risk and correlations do not behave in the same way from one period to another, the concept of the “efficient frontier”, while apparently true, is not useful in practice. If you get a computer to chart risk and return for a variety of portfolios, you do get an upward limit on return for a given risk, that does follow the “efficient frontier” left to right convex curve. Although empirically we have found that efficient frontiers do exist, there seems to be no way of predicting in advance what asset allocations would give you a portfolio that lies on this curve. Therefore people that set a computer to carefully plot out a thousand portfolios to find which asset allocations have worked, will get one that only worked up until yesterday, and this portfolio will behave differently in the future.
• The basic premise is erroneous, that stocks always go up (look at Japan’s Nikkei 225, it was 38,957 in 1989, it is now 20-years later at 7,645, 80% below its all time high),
• As it is applied today, an asset class of large cap growth stocks is a different asset class, than say foreign stocks – to me it is one asset class, stocks – MPT says you get diversification from holding small value stocks and large cap growth stocks – but to me they are all stocks – and that is not diversification,
• MPT says that you should hold a diversified mix of asset classes that are not perfectly correlated. But, just as performance changes for asset classes, correlations change as well. When an MPT portfolio is constructed, correlation coefficients of asset classes are used in designing the portfolio. However, the reality is that correlations change – if you calculate the correlations for the prior 5-year period they will be different than the past 3-year period, for example. In the year 2008, all asset classes moved toward a correlation of 1.0, as all asset classes declined, except for treasuries.
• MPT measures risk using historical standard deviation of an asset class. Again, standard deviation in reality changes, it is not static.
• MPT uses complex mathematics in constructing portfolios. This complexity adds to its sex appeal, but the inputs are historical, so to me, the outputs are suspect,
• SAA sells winners and buys losers, the exact opposite of successful investing. You want to let your winners run and sell your losers. The application of SAA during 2008 would have you selling government bonds (winners) and buying stocks (losers) as your stocks went down and government bonds went up,
• MPT says that an investor who wants higher returns must accept more risk. As if there is a “magic dial” that you turn up and you get more return. My experience is that more risk usually ends up with lower, not higher, returns.
• There is no thought given to current economic conditions or trends. Real estate may be crashing, but under SAA you must continue to hold real estate as an asset class, for “diversification.” I’m not making this up!
• MPT says that investment returns are normally distributed, when in fact investment asset returns are not normally distributed, they are skewed with “fat tails.” The beautiful bell curves exist only in academia.
• MPT’s basic premise is grounded in the efficient market hypothesis (EMH), which says that risky assets reflect all known information – it’s a classic argument for a “random walk” and why you should buy and hold index funds. If EMH were true, you would not have asset bubbles. To me, markets are just that – a collection of people with all the behavioral deficiencies we all possess: overconfidence, bias, etc., etc.