Brief Thoughts on Rules vs. Discretion for Monetary Policy

In typical economist fashion, I see the merits in a mixture of both. I think the switch to long-run average inflation targeting in the early days of the pandemic was certainly a large step in the right direction. The consistency of the rules-based policy is a desirable attribute – the central bank should provide stability so the market can operate smoothly. However, that isn’t the entirety of their responsibility. First of all, the commitment to a 2% inflation every year rule isn’t flexible enough. 2% isn’t some magic number that is unequivocally superior to 3% or 4% or even 1% – as long as inflation is low and stable. A certain maximum upper bound level in any given year is much too restrictive when it isn’t obvious the significance of that number. Furthermore, the consistent undershooting of inflation throughout the 2010s has meant there is pent-up slack in the economy that should be allowed to be unleashed by slightly overshooting the 2% target for some period. That slack has resulted in unemployment that is higher than necessary in the past in a quest for the arbitrary magic number of 2% inflation. Again, this is where a rule is harmful – if a central bank has two objectives, and there is a rule for one, that is the one that will gain superiority. This unemployment neglect is exacerbated by the fact that there is no longer an inherent tradeoff between inflation and unemployment (let alone one that is causal and stable), as we saw with the reverse-stagflation (low inflation, low unemployment) pre-covid. This is further complicated by the fact that monetary policy (i.e. interest rates) is a blunt tool – much of the time fiscal policy would be more direct and appropriate (as we saw with stimulus checks during covid), however that would require non-expert politicians to come to a quick consensus, which is seemingly impossible now (not even accounting for the likely inflation bias). Finally, the transparency attribute of the rules-based policy is underrated and important. The general public knows very little economics (i.e. are not the rational agents of the Barro-Gordon model), yet their irrational inflation expectations matter. Transparency of central bank decision-making (though it has risks) and economics education (like Powell has attempted) are important in educating the public so their expectations (which have real effects) are more predictable. [Aside: While I was initially strongly in favor of everyone getting basic economics education, I am now much more hesitant. It may be better for people to know that they know nothing about economics than for people to think they actually do know how the economy works based on taking basic economics courses and Neoclassical models without understanding the assumptions and how these basic models poorly approximate the real world.] Overall, I think I am in favor of a long-run average target level for inflation, but short-term discretion for central banks to react to unexpected shocks (like covid). I think it might also be valuable to introduce a similar target unemployment rate to bring it up to the same level of importance as inflation (which would inherently lead to more discretion as the rules would likely be conflicting), and to continue with central bank transparency and education.