4. The psychological drawdown: quitting at the very worst moment

Risk 3: Drawdown and ruin

There are two drawdowns in any investment and only one of them turns up in the reports. The first is the one the strategy endures: a curve that falls, stalls and rises again. The second is the one endured by the person running it, and that one ends sooner, almost always at the worst possible point. The distance between the two is where most of the return investors thought they had goes up in smoke.

A strategy feels no fear; you do

A system can stay 40% below its peak for three years with no internal consequence whatsoever: it does not sleep badly, it does not read the news, it does not have to justify itself. You do. Kahneman and Tversky documented that a loss weighs roughly twice as much as a gain of the same size, so a month of minus 8% is not emotionally offset by a month of plus 8% but by two of them.

Then add how often you look. A portfolio of shares observed daily is in loss on close to 45% of days; observed year by year, far less often. Nothing has changed in the portfolio: what has changed is the number of times you put your tolerance to the test. Looking more makes you perceive an investment as more painful than it is, and act accordingly.

The behavior gap

There is a way of measuring this with real money. The return a fund publishes assumes you bought at the start and touched nothing. The return the average investor gets weights by the money that was inside at each moment, and it comes out systematically worse.

Morningstar's annual reports on that difference have put it, decade after decade, at around one to one and a half percentage points a year in equity funds over ten years. Other firms publish much larger gaps, of several points, using fairly disputed methodologies; it is worth sticking with the conservative version, because even that version is devastating. One point a year over twenty years takes away around a fifth of the final wealth.

The mechanism is no mystery and it does not require anybody to be stupid:

• The money comes in after the good years, when the fund shows up in the rankings.
• The money goes out after the bad ones, once the fall has already happened.
• The result: little capital exposed to the good stretches and a lot of capital exposed to the bad ones.

It is a tax paid on when you act, not on what you chose.

The arithmetic of giving up halfway

Concrete numbers make the scale of it clear. Your portfolio is worth 100 at its peak. It falls 30% and you sell at 70. The market bottoms out shortly afterwards and you decide to go back in once it has recovered to 85, because you would rather wait for confirmation. Your 70 now buys 82% of what you had. The market goes right back to 100 and you are left on 82.4.

The market has ended flat. You have permanently lost 18% on a journey that, had you done nothing at all, would have been a draw.

The halfway version is no better. You fall 35%, you have 65, and you decide to halve your exposure so you can sleep. The market needs 54% to get back to its peak, you capture half of that and you end up close to 82. You have locked in the loss at the large size and left the recovery to the small size, which is the worst possible combination and the most common of the lot.

Decide the limit beforehand, not during

The only defense that works is a boring one: set the limit in the cold light of day, in writing, before the fall exists. Discovering it in the middle of a fall means deciding with the worse half of your head.

• Write it down before you start. One page: what maximum fall you consider plausible, how much time under water you are willing to tolerate and exactly what you will do if it happens, including the option of doing nothing.
• Size backwards. If the strategy has 40% falls and you can stand 15%, the conclusion is not to change strategy, it is that only part of your capital belongs in it.
• Separate stopping on evidence from stopping on pain. The first is an objective rule, defined beforehand, about what observation would demonstrate that the strategy has broken. The second is that you cannot bear it. Both are legitimate; mixing them is what ruins people.
• Cut down how often you look. Checking daily on a strategy designed for years adds no information and does add opportunities to give up.
• Rehearse the scenario. Write out, with your real figure in euros, what the statement will look like in the worst plausible month. The question is not whether you accept it in the abstract, it is whether you accept seeing that specific number.

And a warning: this is education, not advice. Nobody can tell you what your tolerable limit is; the purpose of this module is for you to decide it before the market decides it for you.

Real case: the best fund of the decade and its investors

Through the 2000s, an American equity fund, CGM Focus, was the best in its category with a return of around 18% a year. The average investor who passed through it lost around 11% a year over the same period, according to the flow-weighted calculations Morningstar circulated and the financial press of the day picked up.

Almost thirty points of annual difference between the fund and its own owners. There was no fraud, no hidden fees, no asterisk in the prospectus: the money arrived en masse after the spectacular year of 2007 and left en masse after the disaster of 2008. The return was there, available to anybody who stayed put. Almost nobody stayed put.