A tribute to Harry Markowitz

On June 22, 2023, Nobel Prize winner Harry Markowitz, PhD father of modern portfolio theory died at age 95, by Jeffrey D Fisher

 
In 1952, Markowitz published a paper titled “Portfolio Selection” in the Journal of Finance, which laid the foundation for what is still referred to as “modern portfolio theory” or MPT. His research emphasized the importance of diversification and the relationship between risk and return in investment portfolios.

Markowitz introduced the concept of the “efficient frontier,” which represents a set of optimal portfolios that offer the highest expected return for a given level of risk or the lowest risk for a given expected return. Each portfolio on the efficient frontier represents a different optimal proportion of each investment alternative. His research emphasized the benefits of diversification, showing how combining assets with low or negative correlations can reduce overall portfolio risk.

The development of MPT was part of Harry’s dissertation at the University of Chicago. The story goes that when he was defending his dissertation in 1954, Milton Friedman who was chair of his dissertation committee told Harry that he wasn’t sure his dissertation was acceptable as an economics dissertation since it was “just math.”

But fortunately, Milton was likely joking, and Markowitz did pass his dissertation and his work on MPT ultimately resulted in a Nobel Prize in Economic Sciences in 1990. Dr Friedman attended the Nobel Prize ceremony, which he had received in 1976, and at the end of Harry’s acceptance speech he asked Milton if his dissertation was now acceptable economics.

Extensions of the concept of an efficient frontier resulted in the Capital Asset Pricing Model (CAPM) that assumes everyone should invest in the same portfolio on the frontier that results in the highest combination of that portfolio and a risk-free rate.

That portfolio then becomes the “market portfolio.” The CAPM is of course important in the concept that beta is the appropriate measure of systematic risk for a portfolio which is the only risk that should matter if investors can diversify away the rest of the risk.

The concept of an efficient frontier was important to the evolution of real estate in a multi-asset portfolio, especially by pension plans and other institutional investors. Research showed that adding real estate to a portfolio of stocks and bonds made it even more efficient; that is, it had an even higher return at every level of risk.

Or conversely, less risk for any target return.  Similarly, adding international real estate investments can also result in a more efficient frontier since the returns across different countries tend to not be highly correlated.

MPT has also been applied within the real estate asset class. Whereas MPT was originally applied to selecting individual stocks to form an optimal portfolio, this was not practical for real estate. It is not possible to invest in every single real estate investment.

But we can strategically invest in most property types and geographic areas. So MPT was applied to determining the optimal allocation of a real estate portfolio to different combinations of property type and geographic areas. This helped create the demand for all sorts of real estate investment consultants that continues to this day.

As the capitalization of REITs grew significantly in the 1990s, research also examined whether both publicly traded real estate and private equity real estate were needed in a portfolio to create the most efficient frontiers. Most studies show that there was a role for both with private equity tending to have the most allocation at lower target returns and lower risk investments (CORE), and REITs having a greater allocation towards higher risk and higher target return investments.

There remain challenges in the application of MPT to real estate such as the use of appraisal-based indices that were the only method available during the early years of applying MPT to portfolios that included real estate. Appraisal based indices were shown to be smoother (suggesting lower risks) and lag publicly traded asset indices.

This resulted in MPT output suggesting higher allocations to real estate in a multi asset portfolio. Several “unsmoothing techniques” were developed to deal with this issue and increase the risk metrics assigned to real estate. Questions were also raised as to whether the returns for private equity real estate were normally distributed as assumed by the MPT.

While MPT might not be formally applied by most portfolio managers today, most still apply the concepts behind MPT by considering how to diversify the portfolio, especially for CORE funds. Including property sectors and geographic areas that are not highly correlated is still important in the top-down approach to creating portfolios where allocations to property sectors and geographic locations are determined before selecting individual properties.

This is also one of the motivations for CORE funds pursuing investment in many alternative property sectors in recent years such as senior housing, self-storage, data centres, student housing, and others that tend to be less correlated with the main property sectors of apartment, office, industrial and retail.

MPT is covered in most if not all investment textbooks whether their focus is stocks, real estate, or any other asset class.  We teach MPT in the real estate academy offered through the National Council of Real Estate Investment Fiduciaries (NCREIF). While the math can be a bit intimidating to many, there are many computer models that are available to use in practice to create efficient frontiers and they can even be created in Excel using the Solver add-in if limited to a reasonable number of investments to be included in the portfolio. Clearly, we owe a lot to Harry Markowitz for his contributions to investment strategies and portfolio theory.
 

Published by our friends at: