(Bloomberg Opinion) -- Harry Markowitz, who won the Nobel in economic science in 1990 for his work on portfolio theory and the relationship between risk and reward, died last week at the age of 95. He was not only among the most famous and influential economists of the 20th century, but perhaps also the most misunderstood. Milton Friedman, another Nobel-prize-winning economist from the University of Chicago, is supposed to have said of Markowitz’s dissertation, “Nice work, but it’s not economics.”
Neither of the two men remembered Friedman saying that at the time, but both agreed it was not that Friedman denied the importance of securities pricing in economics, it was that Markowitz’s interest was all in the algorithms for constructing portfolios rather than any underlying economic theory.
Markowitz’s career consisted of many seminal advances in computer algorithms, and some early innovations in using computers to beat the market, not in anything to do with efficient markets or theoretical economics or finance. Students who took his finance classes at the University of California at San Diego report lots of enthusiastic lecturing about computers, not a lot of talk about prices. To Markowitz, financial prices were just data for his algorithms.
The hardest part about teaching Markowitz’s Modern Portfolio Theory, or MPT, today is explaining to students why there was ever any competing idea about securities prices. Markowitz focused on the statistical properties of portfolios of securities. In plainer language, he treated investing as gambling.
This was deeply offensive to traditional investment experts. The legal theory that dominated American law from the early 19th century to the 1970s was the “ prudent man theory.” It treated each investment decision on a stand-alone basis. If a trustee bought a stock that went down, the trustee could be liable to make up the loss. The trustee couldn’t argue that the overall portfolio of stocks bought had outperformed bonds, any more than a person who caused an automobile accident could argue that he or she had avoided lots of other accidents and had a better-than-average driving safety record.
As a result, institutional investors and many individuals avoided stocks, risky bonds, commodities and other securities that might provide useful diversification and higher expected returns than the safe, low-yield bonds favored by the prudent-man approach.
Professionals were adamant that investing was not gambling. Everyone knew it was hard, and the best investors often bought securities that went down, but each loss was a mistake, something to be analyzed for lessons to inform future choices. Dismissing losses as something a manager expected as part of the bell-curve of results — and directing clients to focus on the mean and standard deviation of the curve rather than the loss — was seen as irresponsible to the point of fraud.
The real distinction between investing and gambling is that investors deal in the real economic risk that comes from economic activity, while gamblers create risk with dice or cards or sporting events or other devices in order to bet. But that economic point did not occur to Markowitz, none of his work recognized the connection between security risk and the economy, he optimized portfolios exactly the same way he would approach a gambling game.
Only when real economists picked up on Markowitz’s work did it become economics. The crucial added element was the Efficient Market Hypothesis, or EMH, which is an economic assumption, not a mathematical one. When the economics of EMH is added to the mathematics of MPT, you get the financial revolution that created the Capital Asset Pricing Model and its many descendants. Markowitz is often thought of as part of this movement. In truth, he provided essential tools for the journey, but did not make it himself. He stayed home with computers, mathematics and algorithms without any serious foray into economics.
Despite the MPT revolution, most investment coverage today is about which securities will go up or down in price, not which ones are shrewd bets, and definitely not which ones have desirable correlations to be part of portfolios with attractive risk/return ratios. The mathematics used by cutting-edge investment quants has little relation to the sparse matrix theory and other areas Markowitz favored. While the law has changed to allow investment managers to introduce portfolio theory arguments in court, class action lawsuits still follow every big investment loss. While Markowitz forced a fundamental change in the economic theory of investing, his central insight remains a theoretical one for most investors.
More From Bloomberg Opinion:
- Nobel Laureate Robert Lucas Remade Economics: Tyler Cowen
- Nobel Prizes Need to Solve Their Bias Problem: Ndileka Mandela
- How to Fix Bernanke's Fixes? That's a Future Nobel: Stuart Trow
Want more Bloomberg Opinion? OPIN <GO>. Or subscribe to our daily newsletter.