Real Estate is NOT on "the frontier"

The Finance Paradigm (Modern Portfolio Theory, the Capital Asset Pricing Model and the Black-Scholes Option Pricing Model) was conceived in the context of investment in securities. In a widely quoted article written in 1977 Richard Roll observed that an index based on "the market" portfolio should include "human capital and other non-traded assets", of which we presume real estate is a part.

It is not clear how much Harry Markowitz intended his ground breaking work of 1952 be applied to real estate, but in a 1991 article titled "Individual vs. Institutional Investing", he says "The 'investing institution' which I had most in mind when developing portfolio theory for my dissertation was the open-end investment company or 'mutual fund'."

In its various forms, the Finance Paradigm imposes strong conditions. Three of these are: perfect liquidity, perfect divisibility and perfect reversibility. Even if one could contrive ways to produce the first two of these via steep price discounting and securitization, the third requires short selling, an idea that is at best farfetched for real property. As a thought experiment imagine arriving on someone's doorstep to announce "I just sold a shopping center I don't own and need to borrow yours temporarily so I can deliver it".

All theories draw criticism. The ones that endure are those that survive scrutiny. Professor Markowitz did indeed advance the field of investment and risk analysis by introducing a two parameter model. It is not clear that two parameters can describe a complex world but it would seem that they do a better job than one. Of the two parameters in the Markowitz model, one (variance) is for risk. We now face a number of questions:

1. Does the stable four-parameter model improve our understanding of a complex world?
2. If so, should risk be further expanded into several (alpha, beta and gamma) parameters rather than one?
3. Do the generality of stable laws improve our understanding of risk over the special case of the normal?
4. Are real estate returns distributed non-normally?
5. If they are, are they distributed Stable non-normal or in some other non-normal way?

These questions all wait for answers.