Back to Portfolio for the Future™

Jacobs, Levy, and Markowitz on Portfolios

Bruce Jacobs and Kenneth Levy, the founders and Chief Investment Officers of Jacobs Levy Equity Management, have brought out a new and considerably thickened edition of their classic collection of articles on equity investment.

This second edition of Equity Management contains all 15 articles from the original, and 24 of more recent vintage.

It also contains a new foreword, by Harry Markowitz, along with a reprinting of Markowitz’ original foreword.

As readers of AllAboutAlpha likely know, Markowitz is more than “a founding father,” he is arguably the founding father of modern portfolio theory.  In 1952 he expressed the key insight that, while diversification can reduce risk, diversification will never eliminate risk. Any portfolio will retain risk. This opened the way to other insights – that return variance can be identified with risk for  purposes of quantification, for example, and that there must exist an ”efficiency frontier” representing the best possible trade-offs of risk and return. By 1959 Markowitz had developed a “general mean-variance portfolio selection model.”

He received the Nobel Prize in Economics for this work in 1990.

But in the 1950s this was an academic theory, not something portfolio managers yet considered helpful, and in his foreword to the first edition of Equity Management, Markowitz credited Jacobs and Levy with “bridging the gap between theory and practice in the world of money management.”

In his foreword to this expanded edition, Markowitz congratulates them more specifically on their “efforts to extend the general portfolio selection theory to accommodate recent innovations in portfolio management.”

Some of their work, he says, laid the foundation for strategies such as the 130-30 long –short portfolios.

Two of the Classics

One of the original 15 papers reprinted here is Jacobs and Levy’s 1988 piece for the Financial Analysts Journal, “Disentangling Equity Return Regularities,” which contained an extensive bibliography of the literature on the subject until that time.  The efficient markets hypothesis, Jacobs and Levy wrote, “is strongly contradicted … prices do not fully and instantaneously reflect all publicly available information.”   

As a consequence of this inefficiency, stock price charts do not manifest a random walk. There are return regularities. This 1988 article contained a brief but valuable discussion of the macroeconomic linkages of these regularities.

A follow-up paper they wrote for the same journal the following year, “On the value of ‘value’” appears here as well. In it, Jacobs and Levy make the point that although from a certain abstract point of view the value of a stock should be measured by the dividend discount model, in ugly noisy fact the DDM “is but a small part of the security pricing story.”  At the same time, these authors acknowledge that DDM “appears to be useful in valuing equities relative to alternative asset classes.”

Two of the Additions  

Among the many new pieces in the expanded edition I will briefly and arbitrarily describe only two here. Only time will tell which become classics in their own right.

One of these is “Portfolio Optimization with Factors, Scenarios and Realistic Short Positions,” published in Operations Research in 2005. Jacobs and Levy co-authored with Markowitz on this one, and there is more of software engineering than of economics in it. The authors discuss efficient set algorithms that use factor models, that is, models that presume that return on a security depends in a linear fashion on one or more factors (such as the general market, the relevant industry, or a flight to quality) , plus a security’s specific or “idiosyncratic” term.

The three scholars also discuss loosening the til-then common assumption of modelers that “an investor can sell a security short without limit.” This of course isn’t true, but no very generally applicable alternative assumption presents itself, since “short side requirements vary from time to time, broker to broker, and investor to investor.”   So this assumption most be loosened only for specific clients “as she or he finds them.”

We also find in this compendious volume the article, “Enhanced Active Equity Portfolios are Trim Equitized Long-Short Portfolios,” which appeared in the Journal of Portfolio Management ten years ago.  As the title rather concisely indicates, Jacobs and Levy argue that 130-30 portfolios “can be seen as equivalent to an equitized long-short portfolio,” although they are more compact (hence the adjective “trim”) and less leveraged.