Same money, opposite choices

Jack and Jill both have five million dollars today. Yesterday Jack had one million and Jill had nine. A theory that looks only at what people own calls them equally well off. Anyone who has met a human knows Jack is dancing and Jill is not picking up.

We never evaluate outcomes as final states. We feel them as changes from a line drawn by what we had, what we expected, what we were promised, what feels normal. Above the line is a gain. Below it is a loss, and across the experiments the book reports a loss hits one and a half to two and a half times as hard as a gain of the same size. Anthony owns one million and Betty owns four. Each is offered a sure two million or a coin toss between one and four. Anthony takes the sure thing, a gain. Betty gambles, because two million is a loss. Everything below follows from that.

The wine you will not sell for a hundred

Richard Thaler noticed a professor who loved wine and would not pay more than 35 dollars for a bottle. He also would not sell one from his cellar for 100. Selling had become a loss. The effect was then built in a lab with university mugs worth about six dollars. Sellers asked roughly twice what buyers offered, and fewer than half the trades economic theory predicted ever happened.

The exceptions reveal more. At baseball card conventions, experienced traders swapped a randomly assigned mug for a chocolate bar at nearly the rate a textbook expects, while beginners mostly kept whatever they were handed. Ownership stings when the thing is for enjoying. It stops stinging once you see it as inventory.

Par is a line you do not cross

A reference point can be a target rather than a possession. Across more than 2.5 million putts, professional golfers were 3.6 percent more likely to sink a putt for par than a putt for birdie at the same distance. Missing par is a bogey, a loss. Missing birdie is a gain forgone, and the loss simply summons more effort.

The same asymmetry shapes what people will accept from each other. After a snowstorm, 82 percent of respondents called it unfair for a hardware store to raise a snow shovel from 15 to 20 dollars. An old price becomes an entitlement. People let a firm pass on its own losses, and in experiments they paid to punish a firm that used fresh market power to impose one. Reform runs into this every time. Its losers are concentrated and fight. Its winners shrug.

A cockroach ruins a bowl of cherries. One cherry does nothing for a bowl of cockroaches.

Certainty has a price tag

Probability gets bent at both ends. Raising your chance of a million from zero to five percent creates hope out of nothing. Raising it from 95 to 100 removes the last doubt. Both feel far bigger than a five-point move in the middle. Parents in one study would pay an average of two dollars and thirty-eight cents to cut a child-poisoning risk from 15 to 5 cases per 10,000 bottles. They would pay eight dollars and nine cents to eliminate one of those risks entirely. Zero carries a premium arithmetic cannot justify.

Put that bent scale together with loss aversion and you get Kahneman's fourfold pattern. People take the sure thing when a gain is likely, gamble when a loss is likely, buy lottery tickets for unlikely gains and buy insurance against unlikely losses. A strong plaintiff settles cheap; a weak defendant rolls the dice.

Vividness bends it further. During the Israeli bus bombings, 23 attacks and 236 deaths among roughly 1.3 million daily riders, Kahneman knew his own risk was tiny and still hated stopping next to a bus. Presentation can do the same on purpose. Forensic professionals told that "10 of every 100" similar patients would commit violence refused discharge almost twice as often as those told there was a "10 percent probability." One version shows you the ten people.

The ticket you already paid for

Two fans hold the same basketball ticket. One paid for it, one got it free. A blizzard arrives, and the paying fan is the one more likely to make the dangerous drive, because staying home would close his mental account on a loss. The money is gone either way. Investors sell the stock that is up so the account closes as a win and keep the one that is down, even when the loser is the better sale after tax.

Regret works on the same bookkeeping. Mister Brown almost never picks up hitchhikers, does it once, and is robbed. Mister Smith does it all the time and is also robbed. Eighty-eight percent expected Brown to feel worse, because his unusual act is so easy to undo in imagination. Kahneman's cure is a wider frame. Paul Samuelson's friend refused one coin toss to lose 100 dollars or win 200, then said he would take a hundred of them. For someone who can carry the stakes, that is right. A hundred such bets have about a one in 2,300 chance of losing money overall. Scoring each alone made a good portfolio look like a series of threats.

Lives saved, lives lost

A disease is expected to kill 600 people. Program A saves 200 for certain. Program B has a one-in-three chance of saving all 600. In the original study 72 percent chose A. Now reword it. Under Program C, 400 people will die for certain; under D there is a one-in-three chance nobody dies. Seventy-eight percent chose D. The programs are identical pair for pair, and people shown both versions often could not say what principle chose.

Physicians choosing between surgery and radiation picked surgery 84 percent of the time when its outcome was framed as 90 percent survival, and 50 percent of the time when the same outcome was framed as 10 percent mortality. Culture does not rescue anyone from a form either. Organ donation ran near 100 percent in Austria and 12 percent in Germany, 86 percent in Sweden and 4 percent in Denmark. The high countries presumed consent with an opt-out box; the low ones asked people to opt in. The default predicted the decision better than culture did. "Lives saved" is not false and "lives lost" true. Each recruits a different intuition, and both feel settled until the other appears.

Presenting risk without exploiting it

Every product, contract and policy already has a frame. Kahneman's test is what the presentation makes easier. Show both wordings side by side, survival next to mortality. Give the numerator and the denominator, the percentage and the frequency. Name the reference point, because someone chose it. Make a default visible and easy to reverse, and tie it to a goal you could defend in public. Then reframe the choice and ask again.

If the answer survives the other frame, it is a preference. If it flips, it is a reaction. Jack and Jill had the same five million. What they needed was someone honest enough to say where the line had been drawn.

Three things to keep

Ask where the reference point is. Someone chose it, and it may not be yours.

Reframe an important choice both ways. If your answer flips, you have not decided yet.

Treat repeated small risks as a portfolio and stop scoring each one alone.