Why the Taylor Rule isn't a rules-based approach to monetary policy

From Gavyn Davies:

…these three different interpretations of the Rule were expected by Ms Yellen to differ by as much as 2 per cent in the appropriate level for the Fed Funds rate from 2012-15. Given these wide discrepancies, any of which could presumably be chosen by the Fed under the proposed legislation, it seems pointless to try to force a rule-based system on the FOMC just for the sake of it.

Furthermore, the Rule does not really help with several key problems faced by the FOMC today. The first is how and when to reduce the size of the Fed’s balance sheet, and how that decision should relate to the appropriate level of short rates. Next is how to determine the right relationship between economic objectives and financial stability when setting short rates. Based on their views on these two issues, the FOMC might decide that short rates should be either much higher, or much lower, than suggested by the Rule.

The Rule is also largely silent on another of the Fed’s main headaches right now, which is whether to treat the official unemployment rate as a good indicator of the amount of slack in the labour market. Many members of the FOMC, including the Chair, have argued that the amount of slack is greater than implied by the unemployment rate, because the labour participation rate has been temporarily depressed by the recession. The use of the Taylor Rule does not solve this debate, it simply treats it as if it does not exist.

From the FT there is more here. You don’t have to regard any of those points as arguments against a “Taylor Rule.” But it is disingenuous to think that peddling the Taylor Rule as a monetary option counts as a rule for those who have general reasons for believing in monetary rules over discretion. The Taylor Rule is lucky enough to be called a “rule,” and besides, any reaction function can be described as a rule of sorts. In those two senses it is indeed a rule. But the Republicans who are behind this are fooling themselves if they think this will yield the (supposed) traditional benefits of monetary rules, namely stability, predictability, non-ambiguity, transparency, and so on.

Addendum: Nick Rowe has some comments.