A decision that turned out badly was not necessarily a bad decision

Two things that get scored as one

Any decision made under uncertainty has two separable components: the quality of the reasoning that produced it, and the outcome that followed. In a world with randomness in it, good reasoning sometimes produces poor results and poor reasoning sometimes produces excellent ones. Judging the first by the second is generally called outcome bias, and it is remarkably hard to resist.

The difficulty is that outcomes are visible and reasoning is not, particularly in retrospect and particularly other people’s. So the outcome becomes the available evidence, and it gets used as though it were evidence about the decision.

Money is a domain thick with randomness, which makes it a domain where this confusion does an unusual amount of damage. Almost nothing a household decides — about a purchase, a career move, a mortgage term, an investment — has a result determined solely by the quality of the thinking behind it. Circumstances intervene, and they intervene in both directions.

The four combinations, and which two mislead

Set it out plainly. A well-reasoned decision with a good outcome is straightforward and teaches the right lesson. A poorly reasoned decision with a bad outcome is also straightforward. The trouble is the other two.

A well-reasoned decision with a bad outcome invites the conclusion that the approach was wrong, when it may simply have been unlucky. A household that diversified sensibly and then experienced a poor period may abandon the approach at exactly the wrong moment, having learned something untrue.

A badly reasoned decision with a good outcome is worse, because it is reinforced. Someone who put a large share of their savings into one holding and did well concludes they were perceptive rather than fortunate, and the conclusion encourages a repeat with a larger amount. Success is a poor teacher when the process was unsound.

Hindsight tidies the story afterwards

The problem is compounded by the way memory works once an outcome is known. Events that were genuinely uncertain come to seem as though they were foreseeable, and people recall having expected what actually happened. That reconstruction is not dishonesty; it is how the mind integrates new information, and it operates without any sense of revision taking place.

The consequence is that the uncertainty present at the moment of decision becomes invisible in retrospect. A decision that was reasonable given what was knowable then can look negligent given what is known now, and the person judging it is often the person who made it.

This is why financial commentary explaining why a market movement was inevitable is so plentiful after the fact and so scarce before it.

Judging the process instead

The alternative is to assess decisions on the information available at the time, the range of outcomes considered, whether the risk taken was proportionate to what the household could absorb, and whether the reasoning would still look sound if a different outcome had occurred.

That last test is the useful one. If a decision would be described as reckless had it gone the other way, it was reckless; the outcome merely concealed it. A large concentrated bet that pays off was still a large concentrated bet.

Recording expectations in advance makes this possible, since it preserves what was actually thought at the time rather than what is remembered afterwards. Without a record, the reconstruction is unavoidable, and it will be flattering in the cases where things went well and harsh in the cases where they did not.

It also helps to ask how the same reasoning would have fared across a range of plausible futures rather than the one that happened. A choice that survives most of them was sound; a choice that depended on one particular path was a wager, whatever the outcome eventually showed.

What this changes in practice

It changes how a household reads its own history, which matters because that history is what future decisions are calibrated against. It suggests holding onto a sound approach through a poor stretch, and it suggests treating a windfall from an unsound decision with more suspicion than celebration.

It also suggests being gentler about the past. Households that made reasonable decisions and were overtaken by events sometimes carry a great deal of self-blame that the facts do not support, and that self-assessment influences how confidently they act next time.

None of this indicates what any household should do. Circumstances differ enormously, outcomes remain uncertain regardless of how carefully anything is reasoned, and a decision of real size is worth taking to a regulated adviser — who can, among other things, tell you whether the process was sound before the result is known.