L = (y-y*)2-aV-7r*)2,

y*>y, a>0.

(9.23)

a may differ from a, the weight that society as a whole places on inflation. Solving the policymakers maximization problem along the lines of (9.10) implies that his or her choice of , given , is given by (9.12) with a in place of a. Thus,

( * - ) +

+ F

( - *).

(9.24)

Figure 9.5 shows the effects of delegating policy to an individual with a value of a greater than a. Because the pohcymaker puts more weight on inflation

FIGURE 9.5 The effect of delegation to a conservative policymaker on equilibrium inflation

view about the relative importance of output and inflation. The idea, due to Rogoff (1985), is simple: inflation-and hence expected inflation-is lower when monetary policy is controlled by someone who is known to be especially averse to inflation.

To see how delegation can address the dynamic-inconsistency problem, suppose that the output-inflation relationship and social welfare are given by (9.8) and (9.9); thus = + - and L = [( - y*)/2] + [ { - 7 *)/2]. Suppose, however, that monetary policy is determined by an individual whose objective function is

= * + ~(y* ~y). (9.25)

The equilibrium is for both actual and expected inflation to be given by (9.25), and for output to equal its natural rate.

Now consider social welfare, which is higher the lower is (y - y*)/2 - (7 - 7 *)2/2. Output is equal to regardless of a. But the higher a is, the closer 77 is to 77*. Thus the higher a is, the higher social welfare is. Intuitively, when monetary policy is controlled by someone who cares strongly about inflation, the public realizes that the policymaker has little desire to pursue expansionary policy; the result is that expected inflation is low.

Rogoff extends this analysis to the case where the economy is affected by shocks. Under plausible assumptions, a policymaker whose preferences between output and inflation differ from societys does not respond optimally to shocks. Thus in choosing whom to delegate monetary policy to, there is a tradeoff: choosing someone with a stronger dislike of inflation produces a better performance in terms of average inflation, but a worse one in terms of responses to disturbances. As a result, there is some optimal level of "conservatism" for central bankers.

Again, the idea that societies can address the dynamic-inconsistency problem by letting individuals who particularly dislike inflation control monetary pohcy appears realistic. In many countries, monetary policy is determined by independent central banks rather than by the central govemment. And the central govemment often seeks out individuals who are known to be particularly averse to inflation to run those banks. The result is that those who control monetary policy are often known for being more concerned about inflation than society as a whole, and only rarely for being less concerned.

Empirical Application: Central-Bank Independence and Inflation

Theories that attribute inflation to the dynamic inconsistency of low-inflation monetary pohcy are difficult to test. The theories suggest that inflation is related to such variables as the costs of inflation, policymakers ability to commit, their ability to establish reputations, and the extent to

This idea is developed in Problem 9.12.

than before, he or she chooses a lower value of inflation for a given level of expected inflation (at least over the range where ir > *); in addition, his or her response function is flatter.

As before, the public knows how inflation is determined. Thus equihbrium again requires that expected and actual inflation are equal. As a result, when we solve for expected inflation we find that it is given by (9.13) with a in place of a:

which pohcy is delegated to individuals who particularly dislike inflation. All of these are hard to measure.

One variable that has received considerable attention is the independence of the central bank. Alesina (1988) argues that central-bank independence provides a measure of the delegation of pohcymaking to conservative policymakers. Intuitively, the greater the independence of the central bank, the greater the govemments abihty to delegate pohcy to individuals who especially dislike inflation. Empirically, central-bank independence is generally measured by qualitative indexes based on such factors as how its governor and board are appointed and dismissed, whether there are government representatives on the board, and the governments ability to veto or directly control the banks decisions.

Investigations of the relation between these measures of independence and inflation produce a consistent result: independence and inflation are strongly negatively related (Alesina, 1988; Grilli, Masciandaro, and Tabellini, 1991; Cukierman, Webb, and Neyapti, 1992). Figure 9.6 is representative of the results. Thus it appears that delegation is an important determinant of inflation.

There are two limitations to this finding, however. First, the fact that there is a negative relation between central-bank independence and inflation does not mean that the independence is the source of the low inflation. As Posen (1993) observes, countries whose citizens are particularly averse to inflation are likely to try to insulate their central banks from political pressure. For example, it is widely believed that Germans especiahy dislike inflation, perhaps because of the hyperinflation that Germany experienced after World War 1. And the institutions governing Germanys central bank appear to have been created largely because of this desire to avoid inflation. Thus some of Germanys low inflation is almost surely the result of the general aversion to inflation, rather than of the independence of its central bank.

Second, it is not clear that theories of dynamic inconsistency and delegation predict that greater central-bank independence will produce lower inflation. The argument that they do predict this imphcitly assumes that both central bankers and govemment policymakers preferences do not vary systematically with central-bank independence. But the delegation hypothesis implies that they will. Suppose, for example, that monetary pohcy depends on the central banks and the govemments preferences, with the weight on the banks preferences increasing in its independence. Then when the bank is less independent, government officials should compensate by appointing more inflation-averse individuals to the bank. Similarly, when the government is less able to delegate policy to the bank, voters should elect more inflation-averse governments. These effects wiU mitigate, and might even offset, the effects of reduced central-bank independence.! In short,

In addition, the usual measures of central-bank independence appeat to be biased in favor of finding a link between independence and low inflation. For example, the measures often put some weight on whether the banks charter gives low inflation as its principal goal (Pollard, 1993).