The point is . . . are those instructor who always leave the impression that any monopoly on the legal use of threat-of-punishment for obedience to rule of law... is morally evil . . . doing the student a favor for his interview for a job.
Caplan’s confusion is caused by conflating science and technology. Economics is particularly prone to this because it is so normal to seek technological advice that is based on this (social) science.
The basic distinction is easily explained. Science is, roughly, about trying to describe the world. Technology is, roughly, about trying to do things in the world. Scientific theories are often translatable into technological theories. For instance, suppose a science says, “Every X turns into a Y when N is added”. Then we can derive the technological theory (among other possible ones), “If you want a Y, then add N to an X”. But this technological “advice” is only hypothetical. It is not moral (and thereby categorical) advice.
However, not all science has (known useful) technological applications. And not all technology has a scientific explanation (“It just works in practice; we don’t have a theory of how”).
"Anytime an economist notices a discrepancy between (a) the world as it is, and (b) the world as people believe it to be, economics implies advice."
Peter wants to murder Paul by poisoning him. But Peter has underestimated the necessary dose. Therefore, "economics implies advice" to Peter to increase the dose to the correct fatal level. Is there something odd here or have I somehow misunderstood?
Well, even very rough estimations of costs and benefits can improve decision-making over not considering relative costs and benefits at all.
For example, if there had been any good faith attempt at weighing the costs and benefits of long school shutdowns during COVID, our policy would have been different.
You got shouted down as an ignoramus or a reckless killer in some states if you even attempted that discussion.
Perhaps (although I honestly disagree on the school shutdown example: online is the future, might as well start it with seven-year-olds already), but that's not what the post is saying. It says that "if X over-/underestimates the costs, the right advice will always be to do more/less of the thing A whose costs are over-/underestimated". But that's not true if the relative weight of A to other things is so under-/overestimated that this dwarfs the cost miscalculation!
("Weights" naively implies linear combination, but it doesn't have to, the argument still holds for non-linear and even - perhaps especially - non-monotonic effects of costs.)
There is a lot of work for economists to do, because we as a society seem to have forgotten how to do cost-benefit analysis.
Our catastrophic response to COVID reflected that. Probably also our response to climate change and many other things.
One of the worst aspects of cancel culture is the muzzling of people trying to bring the discussion to weighing of costs and benefits.
If it is advisory, then who hires your students?
Would you say 99% government?
Or a mere 90%?
The point is . . . are those instructor who always leave the impression that any monopoly on the legal use of threat-of-punishment for obedience to rule of law... is morally evil . . . doing the student a favor for his interview for a job.
A second, more explanatory, bite of the cherry:
Caplan’s confusion is caused by conflating science and technology. Economics is particularly prone to this because it is so normal to seek technological advice that is based on this (social) science.
The basic distinction is easily explained. Science is, roughly, about trying to describe the world. Technology is, roughly, about trying to do things in the world. Scientific theories are often translatable into technological theories. For instance, suppose a science says, “Every X turns into a Y when N is added”. Then we can derive the technological theory (among other possible ones), “If you want a Y, then add N to an X”. But this technological “advice” is only hypothetical. It is not moral (and thereby categorical) advice.
However, not all science has (known useful) technological applications. And not all technology has a scientific explanation (“It just works in practice; we don’t have a theory of how”).
"Anytime an economist notices a discrepancy between (a) the world as it is, and (b) the world as people believe it to be, economics implies advice."
Peter wants to murder Paul by poisoning him. But Peter has underestimated the necessary dose. Therefore, "economics implies advice" to Peter to increase the dose to the correct fatal level. Is there something odd here or have I somehow misunderstood?
This implies that once costs and benefits are known, the calculation is known, i.e. that weights that are assigned to them are necessarily correct.
Well, even very rough estimations of costs and benefits can improve decision-making over not considering relative costs and benefits at all.
For example, if there had been any good faith attempt at weighing the costs and benefits of long school shutdowns during COVID, our policy would have been different.
You got shouted down as an ignoramus or a reckless killer in some states if you even attempted that discussion.
Perhaps (although I honestly disagree on the school shutdown example: online is the future, might as well start it with seven-year-olds already), but that's not what the post is saying. It says that "if X over-/underestimates the costs, the right advice will always be to do more/less of the thing A whose costs are over-/underestimated". But that's not true if the relative weight of A to other things is so under-/overestimated that this dwarfs the cost miscalculation!
("Weights" naively implies linear combination, but it doesn't have to, the argument still holds for non-linear and even - perhaps especially - non-monotonic effects of costs.)
Careful, Bry, or you'll be quoting William James soon!