How independent should a central bank be?

The case for an independent central bank is becoming increasingly accepted. This new orthodoxy is based on three foundations: the success of the Bundesbank and the German economy over the past forty years; the theoretical academic literature on the inflationary bias of discretionary policy-making; and the empirical academic literature on central bank independence (CBI). The purpose of this paper is to examine each of the three legs of the argument for increased CBI. ; First we examine the empirical evidence on the relationship between CBI and economic performance. In this context, we compare the German experience with that in the US and conclude that there is indeed a tradeoff between price and output stability. We also examine the sacrifice ratios in recent disinflations and show that sacrifice ratios are positively correlated with CBI. ; Second we present a theoretical model which allows us to consider the optimal degree of inflation aversion of the central bank. We show that society will be better off if the central bank precommits to an inflation rate, provided the fiscal authority is reasonably well behaved. We tie these conclusions to the literature on optimal incentive contracts for central banks. ; Finally we make the distinction between goal independence and instrument independence for the central bank. Given that a tradeoff exists between output and inflation variability, the tradeoff should not be left to the central bank, that is it should not have goal independence. Rather the goals for the central bank should be clearly specified, so that the central bank then can be accountable for achieving these goals. However, it should be free in its choice of means to achieve these goals.