Harry Markowitz's Lasting Influence on Investing

Special Note: We’re offering a special holiday weekend sale on AAII Growth Investing, a new and better approach to growth investing that finds stocks likely to outperform the market in any economic condition.

The world of finance lost one of its groundbreakers last week. Economist and Nobel laureate Harry Markowitz died at the age 95 on June 22, 2023.

“It was Harry Markowitz who provided a theory and a process to the notion of diversification,” wrote Massachusetts Institute of Technology (MIT) professor Andrew W. Lo and Western University (Canada) professor Steven R. Foerster in their book “In Pursuit of the Perfect Portfolio” (Princeton University Press, 2021). “He helped to create the industry of portfolio management. Before his seminal ‘Portfolio Selection’ article in 1952, Markowitz recalled, there was no ‘notion that you should have a theory about what makes a well-diversified portfolio and what is the trade-off between risk and return.’”

Markowitz’s key insight was to consider the covariance of the investments held within the portfolio. Before adding an investment to a portfolio, look at how its returns are expected to vary relative to the other investments in the portfolio rather than consider simply the investment’s individual return characteristics. This is the underlying basis of modern portfolio theory (MPT).

For example, say an investor is gung ho about the prospects for artificial intelligence (AI). This investor believes semiconductors are best positioned to benefit from it and creates a two-stock portfolio of Nvidia Corp. (NVDA) and Advanced Micro Devices Inc. (AMD). Neither stock provides much diversification benefit to the other, as they will be affected by similar industry trends and investor attitudes toward chip stocks as well as AI.

A two-stock portfolio comprising companies from the same industry is an extreme example, but it exemplifies how Markowitz changed the industry. Prior to his 1952 paper in The Journal of Finance, many portfolio managers were focused on choosing assets based solely on their individual prospects (often by assessing valuations). Markowitz argued that these managers and their clients would be better served if the extent to which a new investment increased or reduced overall portfolio risk was considered.

Put another way, our investor could be better served by adding some small-cap value stocks, some international stocks and even bonds to their AI-oriented portfolio.

This basic concept of diversification is now familiar to most investors. But Markowitz’s concept of focusing on individual investments’ impact on overall portfolio risk and return was groundbreaking at the time he proposed it.

Similarly groundbreaking was his concept of an efficient frontier. The efficient frontier plots the optimal level of risk an investor should take for a given level of return. It also theorizes that there is an optimal level of return an investor should receive for a given level of risk.

This trade-off is not linear as the above chart shows. After a certain point, the amount of additional return diminishes as the level of risk is increased (upper right-hand end of the line of the chart shown above). Similarly, a portfolio that is too conservative does not even generate enough return to provide adequate compensation for the low level of risk taken.

Most investors—both institutional and individual—do not try to optimize their portfolios along the efficient frontier. Markowitz didn’t either. In fact, he split his portfolio into 50% stocks and 50% bonds. Markowitz explained his rationale to Lo and Foerster by saying he’d “look silly” if he was completely out the stock market when it went up and he’d look silly if he was 100% in stocks when they went down. (Notably, Vanguard founder John Bogle expressed a similar line of thought for explaining why he held a 50%/50% allocation.)

Much criticism has been levied at MPT over the years, including by AAII founder James Cloonan. I personally cannot recall the last time I saw anybody use the efficient frontier as the basis for their allocation—neither in an actual portfolio nor in papers about portfolio construction. Nonetheless, two underlying concepts of MPT remain sound: 1) The market does not compensate you for taking unnecessary risk (e.g., by holding an overly concentrated portfolio of similar investments) and 2) investors should seek to be adequately compensated in terms of expected return for the risks they take.

Markowitz is best known for MPT, but his influence on how people think about investing is much broader. Markowitz developed mathematical ways to measure downside risk. He was among the first to use computers in the realm of finance. Markowitz has even been called the grandfather of behavioral finance for laying the initial groundwork in this field.