New article: “All Road Lead To r*”
I discuss the inevitability of r* in mainstream models. The argument is simple: central banks need to know what the effect of a policy rate is on the economy. bondeconomics.com/2020/10/all-ro…
I discuss the inevitability of r* in mainstream models. The argument is simple: central banks need to know what the effect of a policy rate is on the economy. bondeconomics.com/2020/10/all-ro…
At any given time, they need to know whether an interest rate setting is supposed to cause the economy (in some sense) to accelerate. For a linear model, there is always going to be a “no acceleration point” (which varies over time).
They need models to estimate that point.
They need models to estimate that point.
r* is not this non-acceleration point; it is a steady-state neutral point. But if we are a neoclassical, we usually impose assumptions to enforce the existence and uniqueness of that steady state.
(Deviations of other variables determines the single period neutral rate.)
(Deviations of other variables determines the single period neutral rate.)
I discussed linear models in the article, which could lead to complaining about nonlinear models. Although I will get around to discussing them, the nonlinear models are a red herring.
If they cannot be fit to data properly, they are just economic fables.
If they cannot be fit to data properly, they are just economic fables.
Once we recognize the inevitability of r*, we need to ask ourselves one question: does the concept work in practice?
If it doesn’t, that means we need to abandon the whole class of models. That is the answer that nobody wants to hear, so it will be ignored.
If it doesn’t, that means we need to abandon the whole class of models. That is the answer that nobody wants to hear, so it will be ignored.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
