This approach is enormously valuable.
It disciplines thinking, produces useful information and makes it easier to build professional consensus about what is known and which questions remain unanswered.
Though cost-benefit analysis is not perfect, is often the best route to getting informed experts to agree.
Used in isolation, however, it can lead to trouble.
In a paper presented at the annual conference of the American Economic Association (AEA) in January, Matthew Weinzierl, of Harvard University,
notes that the world is too complicated to be modelled with anything like perfect accuracy.
Many knock-on effects from policy shifts are unknowable beforehand.
He suggests that in the absence of perfect foresight, policymakers could turn to social principles or rules that have evolved over time.
These may reflect accumulated knowledge about some choices’ unintended consequences.
He gives an example.
Governments might choose to increase redistribution based on evidence that high inequality creates feelings of envy, and envy reduces the welfare of the non rich by making them feel worse.
Yet survey evidence suggests that people are largely opposed to redistribution that is motivated by envy.
Validating envy through tax policy could prove socially corrosive, in a way that economists’ models fail to capture.
Put differently, Mr Weinzierl contends that economists should take moral concerns more seriously.
That is something close to professional heresy.