Karl Smith presents a good case for not making too much of the Romer-Bernstein forecast. I like the method of turning it around on another policy and thinking about how we would react.
But I think something he tip-toes around should also be stated much more clearly: the methods by which economists forecast the future and the methods by which economists estimate stimulus impact are completely different, and the latter are much more accurate. People often talk as if the two empirical tasks are one and the same, without realizing that they are entirely different. They're so different, in fact, that when I learned forecasting and when I learned the sorts of methods they use to produce multiplier estimates, they were taught in entirely different econometrics classes. So if somebody says something like "their model predicting the impact of stimulus obviously doesn't work because their forecast was way off", they don't know what they're talking about. They're conflating two completely different things.
You can be (like I am) confident in their multipliers but not all that confident in their forecasts. This shouldn't be that surprising of a position. We trust that climatologists have a better sense of what's happened with climate in the past than they do of what will happen in the future. We trust that evolutionary biologists have an easier time describing the evolutionary tree from 50 million years ago, but we don't think they're particularly good at understanding what the tree will look like 50 million years from now. Only in very mechanical systems, like planetary orbits, do we trust long-term projections. Even in cosmology more generally nobody knows exactly what's going to happen in the far future.
"We trust that evolutionary biologists have an easier time describing the evolutionary tree from 50 million years ago, but we don't think they're particularly good at understanding what the tree will look like 50 million years from now."
ReplyDeleteWell, evolutionary biology is more than just about stuff that happened 50 mya. The predictive power of evolutionary biology is stronger in some areas than others; e.g. Alexander's prediction of a eusocial vertebrate - which was promptly confirmed by one my favorite of all animals - the naked mole rate.
My favorite animal? That's hard, but it is probably but Attenborough's long-beaked echidna - monotremes are just the coolest.
Anyway, can you go into more detail about how these two empirical tasks differ?
Forecasting is mostly just variations on "what happened today is the most likely thing to happen tomorrow". No joke. You can throw in lags and in addition to an error term you can have lagged error terms. That cleans things up. But it's still a version of "what happened recently is likely to happen in the near future". Some sense of the determinants of output help you specify a better model.
ReplyDeleteFor identifying the impact of stimulus, you are in a somewhat better position. You have variation in stimulus, and you have variation in GDP. You just want to be able to make meaningful use of the correlation. What you want is a change in stimulus that is unrelated to anything else associated with GDP.
As for the Alexander - I think it's more evidence for what I'm saying. Keynes predicted the impact of a liquidity trap before knowing that such a thing could actually even happen. That's much like Alexander's "prediction". Alexander didn't and couldn't say "in the future such and such a creature will evolve". He anticipated how a particular animal could thrive, but he couldn't predict a given evolutionary trajectory. Keynes outlined how the economy would react to a liquidity trap - but Keynes could never have predicted that in 2009 we'd fall into such a trap. "Prediction" seems like a strong word for what Alexander accomplished.
Well, it is what he accomplished. My point is that evolutionary biology is more than about deep time - mya or bya.
ReplyDelete