Economic Placebo Effects
Matthew Martin
10/21/2012 03:32:00 PM
Tweetable
But not everyone is so optimistic on the value of placebo effects. John Taylor claims here that "policy uncertainty" is the main reason for the slow recovery, where otherwise expansionary short-term policies like fiscal and monetary stimulus are actually suppressing the economy by way of a negative placebo effect.
I'm a little disheartened to see these arguments, since they look very non-scientific arguments designed to protect each side's ideological bent. Nevertheless, I'm curious to see how the evidence actually works out. For all his huffing and puffing, John Taylor has yet to point us to a single study that actually asks the question of whether there is a causal link between business confidence and economic outcomes, even if we accept his premise that policy uncertainty is harming business confidence, which also remains unproven.
This seems to me to be a good time for an instrumental variables regression. We can easily regress economic outcomes on business confidence, but to a large extend causality goes the other direction--good economies foster high confidence--meaning that the estimates are meaningless. So we need to find an instrument that affects business confidence, but that we know is unrelated to economic outcomes in all other ways other than through business confidence. Any suggestions?
Unknown
10/21/2012 06:42:00 PM
matthew if we are going with the placebo affect(without actually doing anything)then obama is your man.he has not done a dam thing ,except that medical tax ,obamacare .oh and fast&ferocious you know killing 300Mexicans,and one american.oh and helping the black panthers at polling booths.
Matthew Martin
10/21/2012 09:26:00 PM
David, you are entitled to your opinions about Obama and his policies. However, I do not want to let that black panthers comment slide. It is insensitive and untrue--it is language that divides society, leaving black people feeling oppressed and white people feeling fear, all for no reason at all.