Norm proposes an experiment to find out whether money buys happiness:
Find two people of equivalent wealth, take all the money one of them has above that minimum threshold [ above which money is thought not to buy happiness], and give it to the other. Then ask [who is happier].
This won't do. Someone who has their money taken off them will feel unhappy because they'll feel a sense of injustice. Also, they'll feel powerless because someone can do such a thing to them; we know that autonomy makes people happy, so the loss of it will sadden them.
What's more, people adapt to their circumstances - this is thought to be one reason why money doesn't greatly increase happiness. So if those circumstances worsen, people feel uncomfortable. They are loss averse.
A better experiment would be to take two otherwise equivalent folk, give one of them a windfall, and see who's happier.
Andrew Oswald and Jonathan Gardner have done just this. They compared the well-being of people who won decent sums (£1,000-£120,000) on the lottery to that of people who didn't. And they found that lottery winners were significantly happier. What's more, this is true even two years after the win, so it's not just short-lived euphoria.
This isn't conclusive. Perhaps there's a selection effect here: the sort of people who choose to play the lottery are precisely those who are more likely to be happier if they had more money. But it's a better experiment than Norm suggests.