It's always been weird to me how many smart economists (much smarter than me usually) become members of the Cult of the GDP. I think it has to do in part with a desire to be as scientific as possible and therefore avoiding messy things like value judgments. However too often people fall into the trap of making a value judgment without realizing it.
There's no objective way to judge a policy based on its impact on the income distribution, and even if you're willing to apply some subjective criteria, coming up with consistent rankable metrics of outcomes based on those outcomes is problematic for a variety of reasons.
But that's no excuse for turning off our brains and deciding that the policy which leads to the highest level/increase in per capita GDP is strictly preferred to all other policies no matter what the outcome.
Is a policy which makes 1% of the population better off but 99% worse off a better one strictly because it raises the average? What about 20/80? What about 50/50, when it's the less well off people being made worse off and the more well off people being made better off? People can certainly have different opinions about these things, but what people shouldn't do is think that by focusing solely on real per capita GDP they're not making a subjective judgment. A policy change which impacts GDP also is likely to have an impact on the income distribution and just because you manage to avoid the latter issue doesn't mean that the issue isn't there. All you're doing is saying "GDP trumps all other considerations." If that's what you believe, fine, but it's a rather odd thing to believe.
