The Nobel Prize

Harry M. Markowitz (b. 1927)

Harry Max Markowitz was born in Chicago, Illinois, USA in 1927. He remembers reading David Hume and Charles Darwin while still in high school and being especially taken with the problem of induction and by Darwin’s carefully drawn arguments. Markowitz studied at the University of Chicago on a bachelor’s programme for two years before electing to specialise in economics. His choice turned particularly on his interest in the economics of uncertainty; he was awarded a BPh in 1947, followed by an MA in 1950. Markowitz’s teachers at Chicago included subsequent Nobel Laureates Milton Friedman and Tjalling Koopmans. Markowitz clearly impressed as a student as he was invited to take up a junior research position at the Cowles Commission at Chicago, then directed by Koopmans (Nobel Foundation, 2004). He was awarded a PhD by the University of Chicago in 1954.

In 1952 Markowitz took a job as a research associate with the RAND Corporation. Following an invitation from James Tobin (see entry in this volume) he spent the academic year 1955–56 at the Cowles Foundation at Yale University writing a book Portfolio Selection: Efficient Diversification of Investment, published in 1959. He briefly left the RAND Corporation for a consultant’s position with General Electric in 1960 before returning in 1961. After further work in the business sector, Markowitz began to combine business and academic life. He has held professorial posts at the University of California, Los Angeles (1968–69), the Wharton School, University of Pennsylvania (1972–74), and Rutgers University (1980–82), and worked for IBM and Daiwa Securities Trust Company, before becoming president of his own firm. He has also been visiting professor at Hebrew University in Israel, the University of Tokyo and the London Business School. From 1982 to 1993, Markowitz was the Marvin Speiser Distinguished Professor of Finance and Economics at Baruch College, City University of New York; his institutional affiliation at the time of his Nobel Award. He is presently research professor at the University of California, San Diego.

Markowitz’s offices and honours include presidency of the American Finance Association in 1982, election to membership of the American Academy of Arts and Sciences in 1987, and the award of the John von Neumann Theory Prize in 1989 by the Operations Research Society of America and the Institute of Management Sciences. He is also a fellow of the Econometric Society. In 1990 he was awarded the Nobel Memorial Prize in Economics, together with Merton Miller and William Sharpe, ‘for their pioneering work in the theory of financial economics’, and more specifically in Markowitz’s case ‘for having developed the theory of portfolio choice’ (Nobel Foundation, 2004).

In his Nobel lecture, Markowitz (1991, p. 476) recalls a slightly ticklish episode during the defence of his PhD at Chicago: ‘Milton Friedman argued that portfolio theory [the subject of Markowitz’s dissertation] was not Economics, and that they could not award me a PhD degree in Economics for a dissertation which was not in Economics’. We shall come to the result of this apparent impasse later, but the debate itself is instructive – and this is Markowitz’s point in relating it – because it illustrates the novelty of his work on portfolio theory in the early 1950s. As Professor Assar Lindbeck noted in his Nobel Presentation Speech on behalf of the Royal Swedish Academy, before Markowitz’s pioneering contribution ‘there was hardly any theory whatsoever in financial markets’.

Markowitz’s initial paper on portfolio theory – ‘Portfolio Selection’ – was published in 1952 in the Journal of Finance. The paper began by rejecting the hypothesis that investment behaviour should seek to maximise discounted anticipated returns on a portfolio. Markowitz pointed out that such a strategy might easily underplay the role of diversification in portfolio choice when, in truth, ‘diversification is both observed and sensible’ (Markowitz, 1952, p. 77). The point of diversification, of course, is that it reduces uncertainty. An investment decision that turned solely on expected return could lead to either a portfolio composed of just one security, or one of several where securities have equal expected returns. The problem here is that investor-return myopia abstracts from any considerations about portfolio risk. Following this line of argument, for Markowitz (1991, p. 470), ‘It seemed obvious that investors are concerned with risk and return and that these should be measured for the portfolio as a whole’. However, he also pointed out that diversification per se might not be enough to reduce the risk associated with a portfolio. Previously, it had been ‘hinted’ by, for example, John Hicks, that the law of large numbers meant that risk could be eliminated by sufficient diversification (Brealey, 1991). Markowitz argued that because the returns from securities are too intercorrelated, the law of large numbers was not applicable. This meant that diversification had to be of an appropriate form – in other words it had to take account of the variance of portfolio return. Here then was Markowitz’s approach to portfolio choice: it had to balance ‘two dimensions … the expected return on the portfolio and its variance’ (Royal Swedish Academy of Sciences, 1991, p. 2). This gave rise to his ‘expected returns–variance of returns’ or E–V rule which states that investors should choose from a set of efficient portfolios that have a ‘minimum variance for a given expected return or more and maximum expected return for a given variance or less’ (Markowitz, 1952, p. 82). The paper identified the efficient portfolio set using graphical analysis. A central characteristic of the E–V rule is that it requires fruitful diversification. Markowitz highlights the futility of assembling a portfolio of securities based on firms from the same industry: their performances are likely to be highly correlated, and no matter how many securities are held, the problem will persist unless covariances are recognised as a critical issue in portfolio choice. For Brealey (1991, p. 15) this highlights Markowitz’s most important contribution to financial economics – his demonstration ‘that there is a role for a portfolio manager that is distinct from that of the security analyst and that this role has a formal economic logic’. In Markowitz’s (1952, p. 87) own words ‘The rule serves better…as an explanation of, and guide to, “investment” as distinguished from “speculative” behaviour’. In subsequent work (Markowitz, 1956; 1959; 1987; Markowitz et al., 1999) he provided an algorithm for tracing the efficient frontier of portfolio choice and offered elaborations on his earlier analyses.

In addition to his Nobel Prize-winning work on portfolio theory, Markowitz has also made important contributions to mathematical programming and the development of a computer programming language SIMSCRIPT, both of which were recognised in the award of the von Neumann Theory Prize.

Finally, we return to the debate initiated by Milton Friedman over the economic credentials of Markowitz’s PhD. Markowitz supposes Friedman to have been only ‘half serious’ as his degree was awarded with no undue delay. However, at a distance of almost 40 years, Markowitz is content to acknowledge that, then, Friedman may well have been right, yet he adds a twist: ‘at the time I defended my dissertation, portfolio theory was not part of economics. But now it is’ (Markowitz, 1991, p. 476). He might well have continued that the difference was initiated by his own pioneering work (Brealey, 1991; Royal Swedish Academy of Sciences, 1991; Varian, 1993).

Main Published Works
(1952), ‘Portfolio Selection’, Journal of Finance, 7, March, pp. 77–91.

(1956), ‘The Optimization of a Quadratic Function Subject to Linear Constraints’, Naval
Research Logistics Quarterly, 3, March–June, pp. 111–33.

(1959), Portfolio Selection: Efficient Diversification of Investment, New York: Wiley & Sons. (1987), Mean-Variance Analysis in Portfolio Choice and Capital Markets, New York: Basil

(1991), ‘Foundations of Portfolio Theory’, Journal of Finance, 46, June, pp. 469–77.

(1999), ‘A More Efficient Frontier’ (with F. Schirripa and N.D. Tecotzky), Journal of Portfolio
Management, 25, May, pp. 99–108.

No comments: