i guess basically we're fairly good at making generally good directional predictions for large structural policy changes at the macro level, but we don't have the kind of precision that would be required to nail something like a 6% inflation rate vs 2% in a noisy environment
-
-
its an odd situation where typically we can predict roughly the direction of change that would occur in response to eg a policy or environmental change. but predicting what will happen at the level is very difficult
1 reply 0 retweets 1 like -
more concretely "if we do more of x, x will probably increase" is the sort of prediction we're fairly good at making and this is often though not always reliable but if you were to ask us in a void "what will inflation be next year" predictions are somewhat harder
1 reply 0 retweets 1 like -
all of this is built around "ceteris paribus", ie "all else equal", which is a way of saying "this prediction depends on stuff that matters but not what we're explicitly considering staying about the same" when the environment gets volatile the assumption gets weaker
1 reply 0 retweets 1 like -
in many cases the prediction about the difference--if we do x, y will increase-- stays valid despite environmental volatility, but it won't be obvious, because the environmental changes have an effect on y that overwhelms the effect that x had on y
1 reply 0 retweets 1 like -
so--and im not saying this is what happened, i don't know enough here--a person might have reasonably predicted that the effect of the spring stimulus on inflation was going to be negligible, but then a bunch of supply issues caused inflation so were they wrong? who knows, maybe
1 reply 0 retweets 1 like -
it gets even trickier because there are often interactions between the environment and the explicitly considered action. maybe the stimulus wouldn't have caused inflation without a supply crunch, but with the supply crunch happening it did accelerate inflation
2 replies 0 retweets 1 like -
in the latter case, was the prediction good or bad? at this point i think we are getting into talmudic territory and i defer to those who are better equipped for such considerations :) anyway its all very tricky and i mostly wish people would speak with less confidence.
1 reply 0 retweets 1 like -
Replying to @eigenrobot @Historycourses
If only someone who had studied a ton of Talmud was around... Thanks though!
1 reply 0 retweets 1 like -
I guess the obvious question is, after we've forgiven economists for doing their best, what predictions are good to make at all? Like "do stimulus, it shouldn't hurt inflation much" seems helpful, but what's the point of predicting what inflation will actually be if they can't
2 replies 0 retweets 1 like
yeah i don't think much of things like inflation forecasting from statistics some of this you can instead use markets to infer, using market rates on government debt of various durations i think i prefer that approach and it doesn't require any real statistical inference
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.