Endowment Effect and Market Returns as They Are Affected by Algorithmic Portfolio Management

In this post I hope to explain the basic premise behind a behavioral economic theory that I have composed. While I have not yet searched for quantitative evidence to demonstrate the extent to which it would affect markets, I believe that the principles behind my logic are upheld by fundamental concepts within behavioral economics.

A brief search of the endowment effect will explain the basic premise as described by Richard Thaler: ‘This pattern—the fact that people often demand much more to give up an object than they would be willing to pay to acquire it (Thaler, 1980).” As humans, we tend to assign more value to the items to which we assign ownership. This is seen in the form of pride.

It has been shown that stocks that outperform for a period of time continue to outperform. The fundamental basis of what drives price upward is that there are fewer sellers than buyers. The endowment effect contributes to the dearth of sellers. Those who own the outperforming stock feel pride in their ownership and are proud of the intelligent decision they had made to purchase the stock, thus, they are reluctant to sell.

The endowment effect can lead to strict loss aversion within money management. That is to say it is mentally difficult to sell a losing stock because we innately assign it a greater value than the market. This behavior can lead to an overly optimistic assumption about the reality of the outcomes available for a particular equity. The urge to hold on to a falling stock, even when the original investment rational has been proven false, can be attributed to the endowment effect. This leads to greater loses than would have otherwise been realized in a portfolio run by an Econ.

An Econ, as described by Daniel Kahneman in Thinking, Fast and Slow, is one who does not fall victim to the mental traps outlined within behavioral economics. If an Econ were to be running a portfolio in which a losing stock needed to be sold, there would be no occurrence of the endowment effect. Realistically speaking, Econs do not exist in the form of humans—however, they have recently been created in the form of algorithms.

My theory states that as algorithmically driven portfolio management continues to grow, we will see less price appreciation on outperformers and increased depreciation in price of underperformers compared to historical averages. This outcome is the result of two things: eliminating the reluctance to sell winners once they have become overvalued, and eliminating the reluctance to sell losers once an investment rational has been breached. Both outcomes result in an increase in sellers—which, in the long run, will dampen market return rates.