Last up in this series of deep dive articles on behavioral finance is mental accounting. Very generically, the bias describes some of the errors in judgment that occur when investors contextualize information. In this way it is similar to some of the other biases I have covered where improper context is the root cause of distorted thinking (e.g. Anchoring, Availability, and Representativeness).

Next week I conclude the deep dive by landing with A Theory of Behavioral Finance. To my knowledge the discipline of behavioral finance has avoided creating a theory. Odd. In any case that is where I am going to land. In the meantime…

A Helpful Mnemonic Device: LOCHAARM

The major behavioral biases are:

• Loss aversion
• Overconfidence
• Confirmation
• Herding
• Anchoring
• Availability
• Representativeness
• Mental accounting

The above mnemonic device may prove helpful for remembering the major behavioral biases. Specifically, the above is LOC HAARM, brain lock that harms investment performance.

Mental Accounting Bias: Origins

Mental accounting may be described as the artificial compartmentalization of money into different categories and applying different decision rules to those categories. Richard Thaler, yet another Nobel Prize winner for work in behavioral economics was first to formally describe mental accounting bias in 1985.^[1] Thaler, though, defined mental accounting mathematically and with respect to Prospect Theory and did not provide an explicit verbal definition.^[2] However, he did provide several salient examples demonstrating mental accounting, including:

“Mr. and Mrs. L and Mr. and Mrs. H went on a fishing trip in the northwest and caught some salmon. They packed the fish and sent it home on an airline, but the fish were lost in transit. They received $300 from the airline. The couples take the money, go out to dinner and spend $225. They had never spent that much at a restaurant before.”

In this example the two couples clearly had compartmentalized the monies received from the airline as “house” money, and not as their own. Consequently, they spent more at the restaurant than they ever had before. Both couples overlooked the fact that because money is fungible, that they really were spending their own money, not that of the airline. After all, both couples could have invested the money instead, or at long last fixed a broken house lamp.

Thaler’s research seeks to explain oddities in decision making that deviate from the predictions of traditional economic theory where the (bs) assumption is that decisions are made mathematically and rationally. His work demonstrates that lay people and CEOs alike are guilty of mental accounting.

Mental Accounting Bias: Nuances

There are several key nuances with mental accounting to be aware of and some fluency with them improves investment decision making.

Contextualization

Contextualization lies at the heart of mental accounting bias. That is, how you sort information in your mind determines how you think about it. Duh! But here is the thing, categorizing information is entirely arbitrary, but the sorting ends up affecting the emotions we feel and the decisions we make.

See if this example hits home. Most of us have worked in a corporate setting where we are asked to create an annual budget. Within the overarching plan there are categories of anticipated spending that management asks us to delineate (i.e. mental accounts). Thus, there are monies set aside for the data budget (e.g. Bloomberg, Factset), travel (e.g. conferences, visits to corporate HQ), continuing professional development (e.g. publications, courses), and so on.

Inevitably expenses exceed the budget in one category and frequently this leads to criticism from up the management food chain. But equally inevitably the expenses for another category are under budget and we do not feel the pain. But the monies are all fungible, all are a part of the same total budget, but for each pile of cash there are different decision rules, different utility functions, and consequently different judgments driven by arbitrary distinctions.

There are Two Different Values in Every Transaction

Said differently, the above example highlights a nuance of mental accounting. Namely, that there are really two values attached to any transaction:

1. Acquisition value, and
2. Transaction value.

Acquisition value is the expected price we have in mind that we are willing to pay for something. Whereas transaction value is the value one attaches to receiving a better price than expected. If there is a gap between these two values you have the obscure but important economic concepts of either producer surplus or consumer surplus (see, for example, Uber’s Surge Pricing a Study in Important Investing Topic: Consumer Surplus). Producer surplus is where acquisition value > transaction value; and consumer surplus is where transaction value > acquisition value.

Decisions to buy or not buy are always made under uncertainty. That is, I do not know my utility outcome (i.e. transaction value) with any purchase ahead of time. Consequently, ex ante, I never willingly purchase a good or asset where transaction value is negative. My total utility for any purchase = acquisition value + transaction value. I might buy something where transaction value = 0, but never below that level.

The whole thing is more obvious after I have made a purchase of a good or asset. If I believe I have overpaid for something (transaction value < 0) I feel “ripped off” and that the seller has made money illicitly and I am likely angry. Here there is consumer surplus.

But if the reverse is true (transaction value > 0) I believe that I have “gotten a good deal” and I am happy with this outcome. Yet, the acquisition value is arbitrary and determined by me. Another (ir)rational person confronted with the same situation may have a different acquisition value in mind, and thus a totally different emotional experience in the exact same circumstance.

Framing

A final detail on mental accounting bias are the frames suggested by Shefrin and Thaler’s behavioral life cycle hypothesis.[3] Here they postulate that people mentally frame assets as belonging to either current income, current wealth, or future income. Depending on how folks contextualize their money relative to these categories has implications for their behavior. Specifically, their marginal propensity to consume under each circumstance is likely different in each scenario.

Mental Accounting Bias: Manifestations

In investment management our ability to contextualize at a high level is one of the secrets to delivering great long-term results. Thus, we need to minimize the ill effects of mental accounting bias on our results by knowing the contexts in which it occurs.

Most obviously it is typical at the portfolio level to categorize assets. For a value investor here is an obvious example: overvalued, slightly overvalued, fairly valued, slightly undervalued, and undervalued. Yet, as I illustrated above with acquisition value and transaction value, our emotions attached to these different categories are arbitrary because of how we choose to think about acquisition value.

There is NO difference between acquisition value as described by mental accounting bias and estimated/fair value as described in a valuation process. Margin of safety is also exactly equivalent to transaction value.

Yes, a research analyst’s valuation model uses logic and is a rational process. But because the output of a valuation model – fair value – embeds dozens of assumptions about the future outcomes faced by a business, and because the future is unknowable, its price is arbitrary. As I describe in Most Misunderstood Investing Concepts: Fair Value, our estimate of the price of a security as derived in a valuation model is interesting only as a means of understanding the business and its operating environment.

The above example is for a value manager, but a growth manager might have in mind categories for the stocks in her portfolio of: high growth, some growth, and no growth. Each category likely has different decision rules in place and thus may be guilty of mental accounting bias, too

Another manifestation of mental accounting bias in investment management happens when we think of portions of the portfolio as being “long-term capital,” or “short-term capital.” This might especially be true in target date funds, for example. If we do not sell an underperforming corporate bond out of a portfolio because it is part of the diversified fund’s “long-term capital,” then mental accounting bias may be present. Maybe instead we sell a better-performing bond out of the “short-term capital” portion of the portfolio, thus hurting returns. Remember, the money is fungible, and the distinctions between long-term and short-term may be entirely arbitrary. Thus, we may experience detrimental effects on our investment returns.

Conclusion

I hope as illustrated above it is obvious that mental accounting bias has two things at its heart: improper contextualization of assets that can lead to different decision rules; and different embedded assumptions about time horizon. Of these two, improper contextualization is the problem that is more likely to lead to poor results. By contrast, many of the classic examples of mental accounting bias I have read about over the years disappears when there is an explicit agreement about the correct time horizon to consider for an investment.

[1] Thaler, Richard H. “Mental Accounting and Consumer Choice.” Marketing Science, Vol. 27, No. 1 (1985): pp. 199-214

[2] As an aside, Thaler’s explication of mental accounting and his extension of Prospect Theory is well worth reading. His framework provides a mathematical way of modeling investor’s behavior under different identifiable circumstances. In other words, he helps to make behavioral investment analysis a possibility from a quantitative point of view.

[3] Shefrin, Hersh M.; Thaler, Richard H. (1988). “The behavioral life-cycle hypothesis”. Economic Inquiry. 26 (4): 609–643