Behavioral finance

Every paper so far has assumed an investor who sits down, weighs the odds with absolute precision, and chooses. I have never met that person, and probably neither will you. This chapter is about the investor who actually exists, which means it is about us. Three papers: two psychologists first showed how people really judge what is likely, then how they really choose when money is on the line, and twenty years later two economists opened the accounts of sixty-six thousand ordinary households and added up what all of it had cost them. I find it the most humbling chapter in the whole history, because there is nothing in it I have not done myself.

Tversky and Kahneman, 1974

In 1974 Daniel Kahneman and Amos Tversky published seven pages in a science journal, and the claim they made is the following: when people judge how likely something is, they do not compute, they use shortcuts. The shortcuts are efficient, which is why we have them, and they fail in the same direction every time. Representativeness: we judge by resemblance, so a fund that has done well for three years looks like a fund that does well, and we buy it as if the past were a forecast. Availability: we judge by what comes easily to mind, so the risk that was on the news last week feels larger than the one that was not. Anchoring: we start from whatever number is in front of us and adjust too little, so the price we paid, or the analyst’s target from last year, keeps pulling on our judgment long after it means anything.

Why did seven pages matter so much? Because until then, economists assumed that when people made mistakes, the mistakes were random: one person guesses too high, another too low, and in a big enough crowd it all cancels out. Kahneman and Tversky showed that it does not cancel out. We all lean the same way. And if everyone leans the same way, the mistakes add up instead of cancelling, and they can push a price. That single observation is what turned psychology into something an investor has to take seriously.

Two warnings from the paper that people forget. The shortcuts are not stupid; they are what let a person cross a street without solving equations. And knowing about them does not cure them. Kahneman studied these biases for fifty years and said, repeatedly, that he still made them. So the defense is not to be smarter. It is process: a written base rate before every decision, rules committed in advance, and a decision journal. That last one deserves its own sentence, because it is the cheapest edge available to anyone who manages money. Write down, for every meaningful decision, what you expect, how confident you are, and why. Read it a year later. The counterparty you are trading against is your own former self, and that person is easy to beat.

Kahneman and Tversky, 1979

Five years later they published the most cited paper in economics, in the journal of the very theory they were attacking, so that it could not be ignored. The old theory said people evaluate outcomes by what they do to total wealth. Prospect theory says people evaluate outcomes as gains or losses relative to a reference point, usually wherever they are standing now. And the two sides are not symmetric. A loss hurts about twice as much as a gain of the same size feels good.

Once you accept that, a great deal of otherwise strange behavior becomes predictable. People sell what has gone up, to bank the good feeling, and hold what has gone down, to avoid making the bad feeling real. People who are already behind take wild risks to get back to even, which is the instinct that turns a setback into a ruin. Small chances get overweighted, so lottery-like stocks are chronically overpriced. And there is a stranger consequence. Because everything is judged against a starting point, the way a result is presented changes the decision people make about it. Show a family its whole fortune as a single number once a year, and they see a portfolio that went up a little. Show the same family the same money as a list of thirty positions every month, and they see red numbers, every month, on something, and someone at the table will want to sell it.

The practical consequence is one I have organized my own money around. A position sized so that I can hold it through a forty percent fall is worth more than a larger position I will abandon at the bottom, because the second one delivers the loss and none of the recovery. And the family version: loss aversion multiplied by family dynamics is the great destroyer of long-term plans. The defense is to write the response to a bad year before the bad year exists. Sizes, rebalancing triggers, what spending changes and what does not, who calls whom. Everyone signs it while markets are calm. Prospect theory guarantees that the same people would sign something different at the bottom.

Barber and Odean, 2000

Then the field test. Brad Barber and Terrance Odean obtained the records of sixty-six thousand households at a discount broker, from 1991 to 1996, and measured what the biases cost. Before costs, households earned roughly the market. After costs, the average household lagged by about one and a half percent a year. The most active fifth, turning their portfolios over more than two and a half times a year, lagged by about six and a half. Worst of all, the stocks people bought went on to do worse than the stocks they had just sold.

Why were these people trading so much? They traded because they were sure they knew something, and the more sure they were, the more they traded, and the more they traded, the worse they did. It would be comforting to think this only happens to amateurs with a brokerage account. It does not. The professionals who trade the most also tend to do the worst; the effect is smaller, but it is there. I keep the conclusion very simple: for most people, most of the time, every extra trade is money out the door. Doing nothing is usually the best move available.

What I take from it in practice: activity is a cost center with a negative expected payload, and every layer of an investment process should be audited for the turnover it generates without a stated, tested reason. Add up the true annual cost of trading in every account, spreads and commissions and taxes and impact, set a budget for it, and require a written justification to exceed it. If someone in the family wants a trading account for “the interesting part,” give them a small one, name it entertainment, and keep it away from the rest. And be honest about advisors who are paid by activity; this is the paper that describes what they are paid to do to you.

There is an exercise from my own reading that hurts and is worth doing once. Take your last five significant sales. For each, write down what you sold, what you bought instead, and how both did over the following twelve months. Then compare with what Barber and Odean found. Most people who do this stop needing to be persuaded.

What to keep

Defend against your own psychology with structure, not awareness: a journal, rules committed in advance, control over how results are framed, and a turnover budget. The mistake I see most often is reading all of this and thinking it describes other people. It describes you. If you do not build the rules, knowing the theory will not save you.

Gala: the person most likely to lose you money is not a stranger. It is you, on a bad day. Build things so that version of you cannot reach the money.