PsyFi Search

Thursday 10 December 2009

Mental Accounting: Not All Money Is Equal

Coherent Holes

Now listen carefully, because this is where the bizarre, byzantine and seemingly disparate behavioural biases that afflict our every monetary movement start to coalesce into a single, coherent whole. Or at least a set of coherent wholes. Or is that coherent holes?

Mental accounting isn’t about the madness of the dark denizens of accounting departments, although that’d be as good a subject as any. No, in fact it’s the strange way that we humans turn the perfectly designed medium of exchange we call “money” into an irrational, segregated and non-transferrable set of silos to which we then apply all the normal battery of biased behaviours, thus gearing up our irrationality to a marvellously unhinged degree.

Not All Money is Equal

The result of these odd and unnecessary accounting practices is that not all money is equal, some can’t be spent – ever – and some is frittered away wastefully without a second thought even when we need it for other purposes. Roughly speaking, mental accounting is at the heart of some of the most gratuitously stupid displays of monetary mismanagement we engage in. Which is quite some statement, given everything else we’re financially useless at.

If you’ve ever had a special pot of money for paying the household bills, or stashed it separately for a holiday or even just set some aside for a rainy day then you’re engaging in mental accounting. Nothing wrong in this, in its own right, but when applied to wider savings and investments it all starts to look familiarly illogical, only multiplied several times over.

Jack-in-the-Box Thaler

Mental accounting is the brainchild of Richard Thaler, who keeps on popping up like a psychological jack-in-the-box at the moment, largely because his ideas have caught the zeitgeist of the time: (traditional) economics is dead, long live (behavioural) economics. The world is more than ready to listen to someone whose basic thesis is that people do stupid things all the time and that expecting us to behave idiotically is the only sane way to proceed. Even better, mental accounting actually helps us make predictions about the way that people will behave. Some of them even come true.

So, roughly, mental accounting suggests that people segregate money between different “accounts” and that they then apply different sets of rules to these accounts. This can lead to all sorts of peculiar behaviour in investing.


One of the many properties of money is that it’s “fungible”. This means that it can be used interchangeably: money is money is money. Only mental accounting, at root, says that this isn’t true and, if that’s the case, it fundamentally calls into question a lot of economic theories that the world’s been using for the last four hundred years or so. So tough luck, Adam Smith.

The idea that humans use illusory but non-fungible boundaries to separate different pots of money – i.e. that an account assigned to long-term saving is not easily transferrable to weekly household spending due to psychological barriers – leads to a whole range of predictions. So, for instance, losses will hurt less if they can be offset by other gains in the same account and purchases are more likely if assigned to an account not in the red.

Anchoring, Anchoring, Anchoring

Now, of course, the issue of what actually constitutes a gain or a loss implies that people are operating against some reference point – as we saw in Anchoring, The Mother of Behavioural Biases attaching ourselves to arbitrary reference points seems to be a part of the human condition. What mental accounting adds to this is that anchoring applies, simultaneously, to all of the mental accounts that we run. People may take risks to reverse losses on one account – such as the oft-observed phenomena of racetrack betters taking big risks on longshots on the last race of the day in order to offset losses on their day’s betting account – that are irrelevant to other accounts. So, as discussed in The Lottery of Stockpicking, we simultaneously both gamble on the lottery and buy insurance.

All of which suggests that many of the behavioural biases we’ve previously seen are not to be looked at as single phenomena affecting individual people in the same way, but will be dependent on how each individual sets up their internal accounts. Loss aversion, for instance, may be a major driver of the way people create their accounts as they aim to ensure that multiple losses can be combined within the one account, minimising the feelings of regret. This also explains why people will buy more of a stock at price lower than their original purchase but not at a higher one.

Setting Boundaries

For investors, the way that individual holdings are assigned to mental accounts clearly matters. If each stock is held in a separate account then the standard traits of loss aversion and regret will apply to each and every stock. If combined in a single stock account then the aggregation of losses against other gains can reduce the pain and the likelihood of trading – noting, of course, that mental accounting suggests quite strongly that investors will prefer to sell winners and hold losers if separately accounted for, a prediction confirmed by, amongst others, Terrance Odean in Are Investors Reluctant To Recognise Their Losses? (Answer: Yes).

The fact that entirely fictitious and partly arbitrary boundaries can change the way we behave means that how these boundaries are set is critical. As Investors, You've Been Framed described, by modifying boundaries we can potentially reframe a profit to a loss or vice versa – many’s the stopped-clock expert who’s used this to their advantage by pointing out that the stockmarket boom/crash that they predicted many years ago has finally happened. However, the idea can also be used to explain one of the great mysteries of stockmarket investment – the so-called equity premium puzzle.

The Equity Premium Puzzle

For reasons that have long been debated equities seem to provide significantly greater returns than bonds over long timescales. This finding is a bit like discovering dinosaurs alive and well in the management suite of Goldman Sachs – both slightly surprising and somewhat undermining of mass extinction theories.

In particular the equity premium doesn’t make any sense because if we are to believe the fundamental theories of economics then the excess returns of equities should attract investors greedy for additional returns, driving prices up and eventually leading to lower returns equivalent to those of other asset classes. The persistence of these effects over very long periods suggests that there’s something awry in the economic machinery.

The standard explanation is that equities carry greater risks – so an individual stock is more open to unexpected and unpredictable stuff happening. This drives risk-adverse investors into hiding in safe government bonds, thus seeing their returns decimated over long periods. However, the research doesn’t support this theory, risk aversion isn’t sufficient to explain the finding.

Getting Jiggy With It, Occasionally

Riding to the rescue come Thaler and Benartzi who’ve proposed that the equity premium is, in fact, a consequence of mental accounting. They suggest that if you have two possible investments, one that pays a safe 1% and one that pays 7% over long periods but with massive variations over shorter periods (think bonds and stocks) then loss averse investors need to avoid accounting for losses too frequently, because the fear of losing money will drive irrational behaviour. This is known as myopic loss aversion.

The authors' analysis suggests that the equity premium is caused by investors' need to see that their choice of investment is outperforming over their chosen evaluation period. Working backwards from this they suggest that most investments are analysed about once a year and, presumably, adjusted based on the findings.

Essentially if investors frequently and constantly reset their anchoring points on their separate stock mental accounts so that they are continually assessing their status against current success or failure then they will be tricked, constantly, into getting jiggy with their stocks: selling their winners and holding their losers. This effect is enough to allow long-term investors the luxury of excess returns. Basically investors are the equivalent of serial fornicators, constantly getting their rocks off on new stocks.

Go Long-Term Young Man

If correct then mental accounting suggests that long-term investing in stocks based on fundamentals will, eventually, lead to significant outperformance over more active traders. The catch is that “long-term” can be excessively extended – Arnott and Bernstein suggest that the equity premium is really only about 2%: enough to make you very rich over a long time but not enough to allow you to avoid long periods of underperformance.

Of course, we’ve just had a very long period of stockmarket underperformance but whether this heralds a longer-term resurgence in stocks is impossible to predict. If nothing else markets can stay irrational for considerable periods so, as usual, all the sensible investor can do is invest in multiple asset classes and rebalance occasionally: remember, just don’t mentally account for each asset class separately.

Related Articles: Save More ... Tomorrow, So What's Behavioural Finance Ever Done For Us?, Investors, You've Been Framed


  1. My take on the equity risk premium is that there is no such thing. I don't believe that stocks pay a high average return because investors "demand" this. I believe that stocks pay a high average return because stocks ultimately must pay whatever return is justified by the productivity of the U.S. economy and the productivity of the U.S. economy has always been strong enough to finance a high average return.

    The mystery is why non-stock asset classes pay so much less. There really is a market process setting the returns of bonds and certificates of deposit and so on. If investors had confidence that they could obtain 6.5 percent real from stocks, market pressures would push the returns for the alternative asset classes higher. The correct way to say things (in my view!) is not that there is an equity premium but that there is a non-equity penalty.

    I believe that the non-equity penalty exists because of stock volatility. If we did away with volatility (I believe this can be done by educating middle-class investors about the realities of stock investing), the non-equity penalty would go away. Stocks would be less risky than they are today and all other asset classes would have the same risk but would pay higher returns! Neat, huh?


  2. Hi Rob

    I don't entirely agree with that, but mostly the arguments are academic in that I don't think we know enough to decide one way or another with enough certainty to say we're certain. So my opinion is no more than that, opinion.

    However, having spent a lot of time researching it I'm pretty much sure that educating anyone, no matter what their class, about this stuff isn't enough. For what it's worth that's a reversal of the position I originally adopted here. Unfortunately I don't (yet) have anything to replace it with, but I am looking ...