تورم، انتقال مطلوب دولت و استقلال بانک مرکزی

The problem of monetary policy delegation is formulated as a two-stage non-cooperative game between the government and the central bank. The solution to this policy game determines the optimal combination of central bank conservatism and independence. The results show that the optimal institutional design always requires some degree of central bank independence and that there is substitutability between central bank independence and conservatism. The results also show that partial central bank independence can be optimal and that there are circumstances under which it is optimal for the government to appoint a liberal central banker.

The problem of inflationary bias, which was first identified by Kydland and Prescott (1977), has at its root a time inconsistency problem that arises because there is a potential for surprise inflation to generate output increases. Inflationary bias occurs when rational agents, recognizing that the government may be tempted to use surprise inflation to expand output, demand higher wages, thereby neutralizing the impact of the price increase on output. Rogoff (1985) showed that inflationary bias can be reduced by delegating monetary policy to an independent central bank that assigns a higher relative weight to inflation control than society does.1 In his analysis, Rogoff implicitly assumed that the central bank was free to use its policy instrument as it wished (instrument independence) and also free of government influence over its policy objectives (goal independence). In practice, however, countries differ widely in the degree of independence they confer upon their central banks and even the most independent central banks, who typically do have complete instrument independence, are not entirely free of government oversight. There now exists an extensive empirical literature that provides convincing evidence of a negative correlation between central bank independence and inflation in industrial countries. However, this inverse relationship appears not to be causal. This suggests that there are underlying country-specific factors which jointly determine inflation performance and the preferred degree of central bank independence. In the empirical literature, central bank independence generally refers to the central bank's ability to make decisions independently of the government. It is this type of independence that the various indices of central bank independence attempt to measure. Relatively few theoretical studies of monetary policy delegation have focused on the determinants of the optimal degree of central bank independence in this sense; in most cases, central bank independence has been identified with central bank conservatism.2 This is somewhat unsatisfactory because the failure to distinguish between central bank conservatism and independence is formally equivalent to assuming that the central bank's ability to make independent decisions is complete, and exogenously determined. The need to distinguish formally between the concepts of central bank independence and conservatism has also been recognized by Eijffinger and Hoeberichts (1998). Using a game-theoretic framework, Eijffinger and Hoeberichts show that there exists a continuum of combinations of central bank independence and conservatism which may be optimal. They are also able to demonstrate that the optimal combinations of conservatism and independence are functions of the behavioral and structural parameters in their model. However, although Eijffinger and Hoeberichts model the game between the monetary authority and private agents (wage-setters), they do not allow for strategic interaction between the monetary authority and the government. This has two consequences. First, although the upper and lower bounds of the optimal range for combinations of independence and conservatism can be derived, none of the intermediate optimal combinations can be identified. Second, the analysis cannot overcome McCallum's (1995) criticism that delegating monetary policy to the central bank merely relocates the time inconsistency problem rather than solving it. The issue of monetary policy delegation has typically been studied using models that do not allow for strategic interaction between the central bank and the government. Alesina and Tabellini (1987), Lohmann (1992) and Debelle and Fischer (1994) are notable exceptions. Neither Alesina and Tabellini nor Lohmann are directly concerned with the determinants of central bank independence. In both studies, the degree of central bank independence is treated as exogenous, and neither study considers the potential benefits of less than full independence. Debelle and Fisher also treat the degree of independence as exogenous, but they recognize that different degrees of independence may be more or less desirable from society's point of view.3 In this article, I use a model of strategic interaction between the government and the central bank to explore the relationship between central bank independence and conservatism. In order to do so, I distinguish between central bank independence and conservatism and also between the objectives of a fully independent central bank and those of the government. Central bank independence is defined as the ability of the central bank to set its own policy goals. The objectives of the central bank center only around inflation stabilization and output performance, whereas the government has wider ranging concerns which include, in addition to inflation and output performance, the ability to make transfer payments and the desire to avoid using tax increases to finance these payments. The interaction between the government and the central bank takes the form of a two-stage non-cooperative game. In the first stage of the game, the government appoints a central banker and chooses how much independence to grant the central bank. In the second stage, the central bank and the government move simultaneously; the government sets taxes and the central bank sets the size of the money supply. The government is subject to a budget constraint and understands that the characteristics of the monetary institution established in period one will have a significant impact on monetary policy and therefore on the government's ability to make transfer payments in period two. Because the degree to which monetary policy is delegated to the central bank is a Nash equilibrium, neither policy authority has an incentive to deviate from the solution to the delegation game. Insofar as the game I describe mitigates inflationary bias, it overcomes McCallum's criticism and demonstrates that monetary delegation can be viewed as a solution to the inflationary bias problem. The rest of this article is organized as follows. The economic model is introduced in Section 2 and the objectives of the government and the central bank are specified and discussed in Section 3. The policy game between the government and the central bank is described and solved in Section 4. In Section 5, the solutions to the game are used to characterize the determinants of optimal central bank conservatism and independence. The relationship between central bank independence and inflation performance is analyzed in Section 6. Concluding comments may be found in Section 7

Central bank conservatism and the degree of independence that the government confers on the central bank are two key components of the design of monetary institutions. In earlier studies of monetary policy delegation, only the degree of central bank conservatism has been endogenized; central bank independence has been treated as exogenous. In this article, I have used a two-stage non-cooperative game between the monetary authority and the government to endogenize both of these design features. The framework I employ makes a clear distinction between the objectives of a fully independent central bank and the government. In particular, a fully independent central bank is concerned only with inflation and output performance while the government, in addition to these two objectives, is also concerned with achieving an expenditure target and minimizing tax increases. It turns out that the government's degree of tax aversion plays a critical role determining the nature of the optimal institutional design. The results show that the optimal institutional design is not unique; there is a continuum of optimal combinations of central bank independence and conservatism. Furthermore, there is substitutability between central bank conservatism and independence, so a fully independent central bank represents only one of many possible optimizing designs. As long as the government places some positive weight on the avoidance of tax increases, some of these optimal combinations may be characterized by a liberal central bank. The results of my analysis show that a conservative central bank is more likely to be optimal when the marginal contribution of monetary expansion to the achievement of the government's expenditure target is large relative its impact on inflation. In this case a conservative central bank is needed to curtail the use of monetary policy as a means of financing government expenditures. In addition to studying the relationship between central bank conservatism and independence, the game I have described can also be used to explore the relationship between independence and inflation performance. A large number of empirical studies have found a negative correlation between inflation and central bank independence in industrial countries. However, a causal relationship has not been convincingly established. The theoretical results obtained here show that inflation performance is invariant to changes in the optimal combination of central bank conservatism and independence. However, for any given degree of central bank conservatism, the optimal degree of central bank independence is negatively related to the government's inflation target. For optimizing countries, the negative correlation between central bank independence and inflation that has been found in cross-sectional empirical work may be a reflection of differences in inflation targets among the countries included in the sample. I also show that suboptimal combinations of central bank conservatism and independence result in a positive correlation between inflation and central bank independence. Specifically, when the central banker appointed by the government is too liberal, inflation and central bank independence are positively correlated.