Monday, May 12, 2014

Simon Wren-Lewis is much too kind to Tony Yates

Simon Wren-Lewis is extremely polite as usual. He wrote
Tony Yates today criticises Paul Krugman’s argument that economics had the answer to how to respond to the crisis, but policy failed to follow the prescription. As I completely agree with Paul Krugman on this, let me say why I think Tony’s criticism is completely wrong.
"completely wrong" isn't the half of it. my comment: Wonderful post as usual. I very much like the firmness of your opening declaration. However, I also think you are too kind to Yates. "Social Welfare" is an ethical concept. One doesn't have to be a utilitarian. It must be ad hoc, because there is no way to derive a social welfare function from theory or evidence. Yates is correct that we can't know that the policy response was suboptimal for the simple reason that we can't know what is right and what is wrong [Godwin's law violated deleted]. That said, it is impossible to reconcile public hatred of inflation with utilitarianism. Model based estimates of the cost of inflation imply that it is negligible compared to the cost of output below potential. The case that the social costs of output gaps and of inflation are comparable rests on one of two arguments. First they may be comparable, because we can't know what is good. Even if no model gives high costs of say 10% inflation, there may be high social costs because say such inflation causes intense displeasure to God or something. This is just a special case of the fact that we can't prove something is good and something is bad. Second, maybe output gaps must always be on average zero. If so, the social cost is second order in the standard deviation of output. That means it is reasonable to assume it is small (since squared numers are generally small for example 1000 squared is ... shutup). This is a reasonable argument based on NK models given the assumption that there is no zero lower bound. In other words it is total nonsense.

added here [In fact, of course, people hate inflation, because they think it implies higher prices for the same nominal income. When asked about inflation, they say they would prefer much higher real income. Economists understand that this is nonsense and have decided to just accept that inflation is very very bad. This is a betrayal of utilitarianism and of the otherwise dominant methodology of normative economics. The refusal to admit that they are doing this is proof of gross dishonesty. No serious person can question the plain fact that policy is fundamentally based on an elementary error of logic.]

Finally, Yates must argue that it is reasonable to guess that fiscal stimulus would have caused high inflation. The logic of this argument is we can't know this isn't true, because fiscal policy was about the same everywhere so the fact that inflation was about the same everywhere proves nothing. This argument is insane. In fact fiscal policy differed vastly across countries. Output responded (with a revised estimated multiplier of 1.5). Inflation was about the same everywhere. The argument that the huge range of policy from Greece to say the USA tells us nothing about what would have happened if there was more fiscal stimulus is the argument that data can't tell us anything about what would have happened if we did something else. The claim that there isn't proof that policy was suboptimal is the claim that there can't ever be proof that anything is suboptimal.

No comments: