Central bank mistakes: more on count 2

Martin Sandbu in the FT picks up on my post on central bank mistakes. While he says that the first and third I identify are “on point”, he says the second is simply wrong. I think this is because he (and many others) misunderstand the point I am making, which in turn probably means I’ve failed to be clear about it. But it is really important.

My second criticism is that central banks did not make it clear what the impact of reaching the zero lower bound (ZLB) was, and as a result were too quiet about the adverse impact of fiscal austerity. That is not the same as saying there is nothing central banks can do at the ZLB, or that unconventional monetary policy is impotent. As I said in the post, what the ZLB meant is that central banks could no longer do their job effectively, and that unconventional policy “was untested, and it is just not responsible to pretend otherwise”.

Take three instruments: interest rate changes, fiscal policy changes, and unconventional monetary policy. The first two are tried and tested. There is still much uncertainty, but we can have a good guess at orders of magnitude when it comes to working out how much we need to do to achieve some end result (particularly when interest rate changes will not undo fiscal policy’s effects). Unconventional monetary policy has some impact, but we have little prior knowledge of how big that effect will be (or equivalently, how much we need to do to achieve some end result.) Given lags between instrument changes and results, this is a very serious disadvantage.

A simple analogy. The central heating is broken, and it is freezing outside. It can be fixed quickly with the right kit. You ring two plumbers to come and fix it. One says he can be there immediately, the other says they can come in two hours. You are getting very cold, so you naturally choose the plumber who can come straight away. However when they arrive, they tell you their equipment required to fix the problem quickly is broken, but they can nevertheless probably bodge something within the next day or two. You ring the other plumber, and they do have the right equipment. What would you do? Would you not get cross at the first plumber for not telling you their equipment was not working properly when you first contacted them? Now suppose the first plumber did not tell you anything, and you only found out about the kit that could have fixed the problem quickly later on. Would you employ that plumber again, particular when you discover that since his ‘repair’ your central heating is not working as well as it used to?

In a way this strikes at the core of the independence issue. Without independence, the government would be able to choose the best instrument available, which at the ZLB is fiscal policy. But central banks have been made independent and the task of stabilising the economy has been delegated to them. This institutional change should not mean that we no longer use the best instrument to do the job. [1] But if the central bank fails to be frank, perhaps because it feels bad about admitting that it no longer has the best tools to do the job, that is a clear mistake on its part. In this respect it is not important whether the central bank being honest and clear would have actually made a difference on this occasion. That it might have done is all that matters.

I think central banks can at this point get confused with political neutrality. But pointing out the facts as they see it about their own relative competence should never be seen as ‘political’. Here Tony Yates makes a good suggestion, which is that the central bank should be mandated to comment “on whether its ability to meet the inflation target [or whatever its objectives are] was being hampered by government fiscal policy.” 

Advocacy blogging is so ubiquitous that some presume that in pointing out this and other mistakes I must be arguing against central bank independence (CBI). To repeat, I am not. What I think is indisputable is that CBI done badly can be worse than no independence. It does not serve the cause of well designed and well implemented central bank independence to gloss over past mistakes.

[1] Suppose you erroneously think concerns about government debt were valid. Was that a justification for central bankers to argue against fiscal expansion? Absolutely not. With QE, any fiscal expansion could have been money financed. What central bankers should have said is that short term concerns about excessive government debt were unfounded, because they were acting as a lender of last resort. They did not say this.