If money could buy happiness, how much would it take to bring it back after the death of a partner, child or spouse? Most of us would be loathe to assign such a value, if not offended by the question, but two economists have attached such dollar values to deaths by comparing the way that lost loved ones lower scores on happiness surveys with the way that greater incomes boost scores. More than just a gruesome exercise, they say they hope it will provide courts with a way to more fairly award damages.

"It's a very black thing to talk about," says economist Andrew Oswald of the University of Warwick in England, but courts regularly award damages to bereaved survivors after the death of a loved one. Such awards, however, are not necessarily based on well considered rules. In the U.K., the 1976 Fatal Accidents Act provides for a lump sum of $20,000 to a surviving spouse or the parents of a minor. Recent U.S. court cases have valued life at as much as $18 million or as little as $10,000, according to a 2005 study.

Looking for a more equitable way to assign damages, Oswald and Nattavudh Powdthavee of the University of London reviewed data collected from 10,000 Britons tracked by the British Household Panel Survey, begun in 1991, which records major life events and includes questions designed to gauge overall mental health. They identified the amount of money, on average, that raised a person's mental health score by the same amount that a loved one's death lowered it.

They calculated that it would take $220,000 annually to raise someone's happiness to pre-death levels after a spouse dies, $118,000 for a child, $28,000 for a parent, $16,000 for a friend and only $2,000 for a sibling. Taking into account that some people might be harder hit than others could as much as double those amounts, Oswald and Powdthavee wrote in paper reported at a conference held last week on happiness research, law and policy at the University of Chicago (UC).

Eric Posner, a UC law professor and co-organizer of the conference, says it is too early for courts to adopt Oswald and Powdthavee's method but notes that it may be less arbitrary than the existing way of assigning damages. "The courts ask juries to pick a number and they don't give juries a specific rule or principle for guiding their deliberations," he says. As a result, "it's either nothing or a very high number."

Social psychologist Jonathan Haidt of the University of Virginia says there is "so much more at stake when people suffer loss than simply the hit to their happiness." Actual suffering should factor into damage awards, he says, but so should other things such as feelings of outrage or injustice.

Oswald agrees that more research is needed before the findings should influence policy, but he stands by the concept. "Just because it's hard to value this subtle thing is no reason not to try to do the best in being fair to victim," he says. "We're trying to make it logical instead of random."