“Mental Accounting and Consumer Choice” Thaler (1985)
Flexibility in real accounting standards allowed Enron to mislead financial analysts and investors by overstating profits for many years. Although a perfect accounting system should clearly communicate value to potential investors, we have to make do with a second-best solution which has the potential for manipulation. This paper by Thaler proposes that our own mental accounting system might come with a similar flexibility.
The paper opens with some observations of how, what are ultimately arbitrary, codings of gains and losses can significantly affect our behaviour. Someone down $1,000 in a poker game might continue to gamble recklessly in hope of getting even, while if their long-term investment portfolio had fallen in value by the same amount their play would be unaffected. A perfect accounting system of You Inc. wouldn’t care about this: all of your assets (including the present value of future labour income), and liabilities would be present on the balance sheet, with the difference being equal to your net worth. This paper tries to model the mental accounting manipulations that we go through, and how different representations of gains or losses can have a large impact on utility.
The paper uses prospect theory as a starting point: individuals are risk-averse for gains and risk-seeking for losses, with the marginal disutility of a small loss being greater than the marginal utility of a small gain (loss aversion). Mental accounting says that groups of gains and losses can either be evaluated together or separately. For example, because people are risk-averse for gains (diminishing marginal utility), splitting a large gain up into many smaller gains would lead to a net gain in utility. To minimise their impact, losses should be integrated. Each small loss hurts a lot, while the aggregate reduction in utility is less when they are grouped together. A lot of normally large expenses seem inconsequential when bundled together with the bigger loss of buying a house (OK so this isn’t really a “loss”).
The lessons for marketers are clear (see also my piece on dark nudges): offer many small discounts instead of an economically equivalent larger one, and when selling high cost items use the opportunity to bundle many smaller and overpriced extra options. I believe that mental accounting might help explain something already discussed on this blog: how investors take insufficient care to minimise costs.
When buying an investment product, such as a mutual fund, your total return will be composed of two parts: the gross return of the portfolio, and the impact of management costs. The two are always presented together as one figure: the net return. Management costs always represent a loss. No matter the gross return, mental accounting says that integrating the two will lead to an increase in utility. Suppose the gross return is negative. We therefore have two losses, and integrating those losses increases utility. Suppose the gross return is positive. Now, because of loss aversion, cancelling the loss against the gain will increase utility compared to separate evaluation. My point is that the financial industry has found a way of making their costs have an artificially low hedonic impact.
Costs are a huge driver of the sub-par returns many individual investors achieve on their portfolios. In order to make investors more sensitive to costs I suggest disaggreggating their impact from the portfolio’s gross return. This could be done by adding an extra line on investors’ statements, or perhaps by requiring investors to pay their costs out of their current accounts. The second option would be somewhat disadvantageous for investors with low liquidity, although I suspect it would have a larger impact on behaviour. Disaggregating returns would also decrease investors’ utility in the short term. This is something we would usually not want to do, but would be worthwhile if their utlility had been artificially boosted by a mental accounting trick. As Enron showed, all accounting tricks eventually get found out.