Thread by @JWMason1

People often say that an inverted yield curve (short rates above long ones) predicts recessions because 1) long rates reflect investor belefs about future short rates and 2) central banks lower short rates in recessions.

The logic is fine as far as it goes but there's an additional assumption which often goes unstated: that central banks are **ineffective** at stabilizing demand.

As any student of Milton Friedman could tell you, if changes in the policy interest rate are in general sufficient to offset shfits in demand, with recessions (or inflationary booms) due only to central bank error, then interest rates and the output gap should be uncorrelated.

In other words, low interest rates will only be assocaited with recessions if central banks respond to weaker demand by lowering rates, and this nonetheless fails to prevent weak demand from developing into a recession.

A familiar analogy is a driver with good control of a car - there's no correlation between the car's speed and pressure on the brake or gas pedal. Driver only uses pedals when car would otherwise go too fast or too slow, and actual speed changes are just due to random mistakes.

If we in fact we most often see the driver slamming on the brakes when the car is going fast, and pushing the accelerator to the floor when the car slows down, that proves that the driver in geenral is unable to control the speed of the car.

In the same way, if we see the policy rate lower in recessions and higher in booms, that proves that the policy rate (or at least the normal variation of it) cannot fully offset other sources of variation in demand.

Today, there is no reason a fall in long rates driven by an expectation of looser monetary policy should be a negatuve signal for the real economy, unless we also assume that looser policy will be unable to prevent a deep downturn.

So I'm fine with the conventional story of why the inverted yield curve predicts recessions. (There might be other valid stories as well.) But people making it should acknowledge that they're saying that the impotence of monetary policy is not special ZLB thing, but general rule.

Prompted by this from Krugman, which makes good case for conventional story. He notes that current inversion is due to falling long rates. So story where inverted yield curves predict recessions because they result from rapid tightening, won't work here.

https://twitter.com/paulkrugman/status/1166713277012987911?s=20

To be clear, I am not saying that orthodoxy says that central banks can completely offset demand fluctuations. The orthodox claim is that monetary policy can fully offset **predictable** fluctuations in demand, and that central bank errors are random.

For orthodoxy, conventional inverted yield story doesn't work bc if markets have enough information to know that future demand shortfall will likely require Fed to cut rates, then Fed does too, and can make cut large enough to fully offset predicted shortfall.

Of course no one claims that Fed will do this perfectly. But if they are trying their best - using all available info - then by defintion they're as likely to err one way as other. So in actuality low rates are as likely to be assocaited with positive output gap as negative one.

If demand shortfalls are predictable and Fed lowers rates some but not enough to offset them (as standard inverted yield story requires) then they must see adjustments large enough to fully offset as infeasible or too costly - it's a fuzzy line between can't and don't want to.