The world lost a legend this week
Harry Markowitz passed away at 95 years of age
One of the world’s most renowned economists and mathematicians, Harry Markowitz died June 25th. He is remembered for his 1990 Nobel Prize and pioneering work in portfolio theory, often called Modern Portfolio Theory (MPT). Markowitz, Eugene Fama, Franco Modigliani, Merton Miller, James Merton, William Sharpe, Fischer Black, Myron Scholes, Bob Shiller and Richard Thaler are a select group of Nobel laureates who in less than a century between themselves developed and expanded the study of finance and markets with extraordinary insights into how the price discovery mechanism of markets works and how investors behave.
Markowitz’s work followed the development of the “Efficient Market Hypothesis” (EMH) by Eugene Fama, leading to the widespread adoption of the Capital Asset Pricing Model (CAPM) and its later variant, the Fama-French “three-factor CAPM” which modified the original model by including as variables the relative market capitalization of a security being subjected to valuation and the ratio of book value to price of the particular security.
Unfortunately, most of their work was (despite widespread acceptance and enormous impact on portfolio managers) simply wrong. Markowitz developed the portfolio theory that a properly diversified portfolio would produce the highest return possible for a given level of risk and measured “risk” as the relative volatility of the price of a given security traded in secondary markets to the volatility of the market as a whole, a term labeled “Beta” and return as the relative price performance of a given security compared to the market as a whole, labeled “Alpha”. The popular investment site “Seeking Alpha” is named after that term. The theory presumed the absence of taxes and freedom from transaction costs and was grounded on a belief that higher share prices comprised higher returns.
William Sharpe advanced the Markowitz, Modigliani and Miller portfolio theory by conceiving that if one could both borrow and lend at the conceptual “risk free rate” one could increase one’s portfolio return or reduce its risk through judicious use of leverage, coining the now famous “Sharpe Capital Market Line” and “Sharpe Ratio”. James Merton, Fischer Black and Myron Scholes developed the elegant Black-Scholes model for valuation of stock options. Richard Thaler (building on research done by Amos Tversky and Daniel Kahnemann) advanced concepts of behavioural finance demonstrating that investors over-reacted to adverse events and under-appreciated long term favorable trends. Bob Shiller developed the Case-Shiller Cyclically Adjusted Price to Earnings Multiple modifying conventional price-to-earnings ratios to reflect the relative market multiple to prior periods.
Merton Miller theorized that dividends and capital structure were irrelevant to firm valuation (the “irrelevance theory”) explained with an anecdote where Miller told a pizza restaurant he wanted his pizza cut into eight slices instead of four slices since he was hungrier that day, a wonderful way to make the “irrelevance theory” comprehensible to everyone.
All of these extraordinary economists built on one another’s work either explicitly or implicitly adopting EMH and MPT as elements of their work. Both EMH and MPT assume investors benefit from higher share prices.
They don’t.
Nor do we live in a world where there are no transaction fees, no taxes, and with the ability to borrow or lend at the risk-free rate. Their conclusions, which formed an almost religious acceptance of their theories by money managers worldwide and by courts dealing with securities litigation, were both useful (since they provided a framework for discussion of key issues in markets) and with the exception of Black-Scholes, were by and large wrong, which I will elaborate on below.
Securities prices are affected by changes in the operating performance and prospects of securities issuers, but the operating performance and prospects of those issuers are largely unaffected by the trading prices of securities. Investors don’t benefit from higher share prices, they benefit from lower share prices. If an investor can buy an interest in the same earnings stream and ultimately the same dividend stream at a lower price, the investor will enjoy a higher return. This is tautalogical.
Trading in the secondary market is not free and generates no wealth. All the wealth represented by the shares and bonds traded in secondary markets comprises profits earned by the issues of those shares and bonds. Share prices that rise at rates greater than the rise in underlying earnings and cash flows implictly discounted to the present simply reduce returns to investors, with gains by one investor coming not only at the expense of another investor but also causing reductions in returns to shareholders owing to the transaction fees, commissions, managment fees tied to assets under management (AUM) and taxes. The underlying stream of profits remains unchanged but money is transferred from owners of the securities to wealth managers, financial intermediaries and to governments.
Fama claimed it was impossible for an investor to “outperform” the market averages consistently over time. In fact, it is relatively simple to do so. An investor who builds a portfolio comprising all traded securities in the same weights as they are represented in market averages and eschews trading will “outpeform” the market 100% of the time over any reasonably long period simply by avoided the fees, commissions and taxes that punish those who trade and bring down the returns captured by the market averages. The rise in income funds and “exchange traded funds” (ETF’s) that mimic market averages demonstrates this is the case, only modestly underperforming market averages since they are still exposed to management expenses, fees and commissions (often referred to as MER’s when described as a percentage of a given fund or ETF).
Investors who confine their portfolios to a broad selection of companies with small market capitalizations and those with low price to book value ratios also do better than the averages, a fact that caused Fama to develop the “three factor CAPM” in express recognition of this reality. Investors who limit their investments to well-managed companies that tend to grow more or less in line with economic growth and earn returns on capital employed greater than the average return on market indices of about 9 to 10% (the average for the past century) virtually always produce better returns to investors than market averages, particularly if the investors in question simply hold for long periods and look to a stable or growing dividend stream as the source of the “performance” ignoring the short term trading swings in secondary market prices driven by macro-economic factors, changes in policy driven interest rates, fluctuations in commodity prices and changes in investor sentiment. Those factors are “noise” and affect well-chosen portfolios no more and no less than the market as a whole. Pretending a savvy “advisor” can “time the market” or choose individual securities likely to diverge in a positive way from market trends will make the advisors rich and impoverish the investors who rely on them.
Peter Lynch, John Templeton, Benjamin Graham, Warren Buffet, Stanley Druckenmiller and a handful of others have outperformed the market indices for decades by buying well and holding for long periods. MPT and EMH are useful paradigms for analysis of securities despite their real world limitations, and their widespread adoption has reduced the volatility of markets as a whole as professional money managers have applied those theories to large funds making the theories self-fulfilling prophecies to the extent the trading by these major portfolios has followed the theories. But it is plain and obvious that large funds - Blackstone, Vanguard, the Canada Pension Plan, OMERS and Ontario Teachers Pension Plan have been incapable of producing returns that track market averages and the lag they suffer can be traced to excess trading and high management and transaction fees.
The development of Black-Scholes is perhaps the first theory that applied a stochastic approach to valuation of securities, developed by finding a price that comprised an equivalence point between the risk and option would expire worthless and the risk it would produce market average returns or better during its life and setting that equivalence point as the “value” of the option. It’s limits arise from presuming past price volatility is a proxy for future volatility and that funds paid or borrowed as option premia can be invested or borrowed at a risk free rate of return, but its assumption that stock and commodity prices are log-normally distributed is empirically robust.
Harry Markowitz was also one of the first economists to study investor behaviour and well ahead of his time in theorizing how and why investors make investment decisions. More recent work by Richard Thaler and Bob Shiller (among others) is now advancing our understanding of the psychology of investors in an emerging field called “behavioural economics”. I expect future Nobel prize winners will emerge with more “real world” theories to guide investment management decisions but it may take decades for behavioral economics to become mainstream given the widespread adoption of EMH and MPT in the financial industry and the vested interest professional investment managers have in persuading investors they have some special skill that warrants paying them large fees for stock selection, a game certain to perpetuate lower returns for investors as a class.
I look forward to the news that a future Nobel laureate has published a study that demonstrates that EMH and MPT are passe and that the road to riches lies in common sense investing along the traditional lines followed by Peter Lynch and Warren Buffett. That would comprise a real breakthrough in portfolio theory worth of worldwide acclaim.