Imagine real GDP were to grow by 2% a year - roughly its average for the last 100 years - for the next 1000 years. Average incomes in 3007 would then be 398 million times higher than they are now. But can you imagine the average person being 398 million times happier than they are now?
And if the long-run connection between GDP and well-being is weak, GDP must be a bad measure of well-being in the short-run; the Easterlin paradox is real.
In this new paper (pdf), Jeroen van den Bergh summarizes this and other faults with GDP as a measure of well-being - and he argues that we should abolish the measure entirely.
Measures of GDP are worse than useless, he says, as they cause people to panic about recessions or falling behind other countries, without offering any offsetting useful information.
This doesn't mean we should aim for lower economic growth. But it does suggest Bullingdon Boy was right: "it's time we focused not just on GDP, but on GWB - general well-being."
Of course, there are some other indicators of aggregate well-being. But we don't need to replace GDP with these.
Perhaps we don't need any aggregate measure of well-being at all.
Put it this way. Say there's just the two of us. We'd agree to do things that make us both better off, and we'd agree to not do things that'll make us worse off. And if something benefits you but harms me, we can negotiate a side-payment that either compensates you for not doing it, of compensates me for the harm (depending on who has the relevant rights).
In this Coasean world, we just don't need any measure of aggregate well-being of the two of us.
So, rather than fret about aggregate measures, governments should focus on moving us nearer to this Coasean world. This means: clearly assigned property rights, individual power to walk away from repressive bargains; cuts in transactions costs that prevent mutally beneficial trades; and mechanisms to ensure that externalities get internalized.