Why do we play it safe with gains but gamble to avoid losses?

We play it safe with gains and gamble with losses because we don't judge outcomes by how much we end up with, but by whether they feel like a gain or a loss from where we stand now. That is the heart of prospect theory, developed by Daniel Kahneman and Amos Tversky in 1979. Losses hurt more than equal gains please, so we grab a sure win, and we take wild chances to dodge a sure loss.

Their classic test makes it vivid. Offered a sure $450 or a 50 percent chance at $1,000, most people take the sure money, even though the gamble is worth more on average. Offered a sure loss of $500 or a 50 percent chance of losing $1,100, the same people now gamble, hoping to lose nothing. A quick calculation from those numbers shows the gamble is worth $500 on average in the first case and costs $550 on average in the second: both times, people pick the option a cool calculator would reject.

The theory had a second twist: we give too much weight to rare events and too little to near certainties, treating a 1 percent chance almost like 5 percent. That helps explain why people pay extra for insurance against unlikely mishaps.

The paper became the most cited in economics, and Kahneman won the 2002 Nobel Prize in economics, though he said he never took a single economics course. Tversky, who died in 1996, would have shared it, Kahneman wrote.