4 ms·
Wouldn't an economic indicator for the "right level" be nice? Maintained by some economic research institute (without some obvious agenda)? Or does that alread
by karparov 2y ago
Wouldn't an economic indicator for the "right level" be nice? Maintained by some economic research institute (without some obvious agenda)?
Or does that already exist, I just don't know it?
Maybe it's not just one-dimensional though, i.e., it could matter where exactly spending occurs.
- simiones 2y agoThere's no such thing as an economic institute without an obvious agenda, and generally no way that we know to compute the optimal spending. And even if there was some way to decide how much to spend, there would still be the insurmountable problem of deciding what to spend this money on. It's pretty obvious that some uses of money are more useful than others - if the state spent large amounts of money on buying snow and storing it in the sun, that's not going to have the same kind of effect as spending the exact same amount of money on building new roads. But beyond obviously silly examples, no one can really agree on even the basics of the effects of government spending in other areas. For example, a question like "if the government pays very large pensions and social aids, does that increase the economy as much as spending the same amount of money on subsidizing private business?" is extremely divisive in economic circles - with opposing schools of economics claiming opposite results as the "obvious scientific answer".