Central Bank Independence, Economic Freedom, and Inflation Rates

Article excerpt

WILLIAM A. LUKSETICH [*]

Measures of central bank independence combine many attributes that may or may not affect inflation. Central bank attributes are chosen as a result of political calculations over the distribution of resources between competing interest groups. Simultaneity bias results from regressions of central bank independence or of economic and political freedom on inflation or growth. Our estimates demonstrate the connections between economic and political freedom and central bank attributes that lead to inflation. Countries showing high degrees of economic freedom adopt structures that lead to lower inflation; those that show high degrees of political freedom do not adopt inflation-reducing institutional structures. (JEL E0, E5)

I. INTRODUCTION

The relationship between inflation rates and central bank independence from political control has been the subject of a number of recent studies. Generally these studies provide evidence that greater central bank (CB) independence is associated with lower inflation rates (Banaian et al. [1983, 1995, 1997]; Burdekin and Willett [1991]; Masciandaro and Spinelli [1993]; Alesina and Summers [1993]). This relationship appears to be weaker when developing countries are included in the analysis (Cukierman [1992]). All of these draw rather strong conclusions about the importance of CB independence as a force constraining the political system from using the monetary system to achieve political goals that are likely to be hostile to price stability. Supporting evidence is not strong, however, and the indices used to measure CB independence are far from perfect. Moreover, even if the CB independence measures were accurate, they may simply be proxies for political and economic freedoms that are the more important checks on political manipulations of the economic system.

Others question the evidence and draw different conclusions. Lohmann (1994, 1997) and Posen (1993, 1995, 1997) argue that the correlation between inflation and central bank independence is spurious. Lohmann argues that the federalist nature of the German government and its frequent use of coalitions in forming governments limit the number of threats to Bundesbank independence. Posen (1995) finds that political support for independence comes from the financial sector; if they effectively lobby for price stability, an independent central bank tends to accompany that lobbying. Miller (1998) provides a framework for understanding the behavior of various interest groups in controlling the rate at which governments inflate their currencies.

Cukierman's (1992) GB independence indices have been widely used in empirical studies to assess the effects of CB independence on inflation rates. [1] The term of office of the head of a CB (which we will call the "central banker") is one measure of bank independence. He assigns a ranking of 1 if the term of office (TOO) is greater than eight years, 0.75 if the TOO is great than or equal to six years and less than or equal to eight years, 0.50 if the TOO is equal to five years, 0.25 if it equals four years, and 0 if it is less than four years. He does this for numerous measure of bank independence. Since the publication of Cukierman's index nearly all the aforementioned researchers have either used these individual rankings or a weighted average of these rankings in their efforts to determine the importance of CB independence on inflation rates.

In this article, we extend the research in this area in three ways. First, the use of a multipoint ranking of central bank attributes imposes restrictions on the marginal impact of changing structures or attributes of the CB law that have no a priori basis. Moreover, the weights within the aggregate Cukierman index are quite arbitrary and do not hold up under any reasonable degree of scrutiny (see Banaian et al. [1998], hereafter referred to as BBW). Constructing dummy variables for each of the classifications would be more appropriate than using the classification scheme Cukierman devised as a continuous variable. …

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.