اصلاحات بانک مرکزی، آزادسازی و تورم در اقتصادهای انتقالی ، چشم انداز بین المللی

This paper develops extensive new indices of legal independence (central bank independence (CBI)) for new central banks (CBs) in 26 former socialist economies. The indices reveal that CB reform in the FSE during the 1990s has been quite ambitious. In spite of large price shocks, reformers in those countries chose to create CBs with levels of legal independence that are substantially higher, on average, than those of developed economies during the 1980s. The evidence in the paper shows that CBI is unrelated to inflation during the early stages of liberalization. But for sufficiently high and sustained levels of liberalization, and controlling for other variables, legal CBI and inflation are significantly and negatively related. These findings are consistent with the view that even high CBI cannot contain the initial powerful inflationary impact of price decontrols. But once the process of liberalization has gathered sufficient momentum legal independence becomes effective in reducing inflation. The paper also presents evidence on factors that affect the choice of CBI and examines the relation between inflation and CBI in a broader sample.

The, still ongoing, process of transition from plan to market in the former socialist economies (FSE) involves a fundamental process of change in the structure of those economies. In their attempt to create the institutional infrastructure needed for a market economy governments of the FSE scrapped old institutions and replaced them with new ones that were often patterned after similar institutions in the Western democracies. One element of this process was the creation of a Western type central bank (CB). Practically, all FSE either created a totally new CB by breaking the, typically socialist, Monobank into a CB and a private banking system or, in the case of several Central and East European (CEE) countries, substantially upgraded the legal independence of their preexisting CBs. Within a span of 8 years (1991–1998) all FSE created completely new CB laws, or reformed existing laws, at least once and sometimes even twice. Although there are substantial cross country variations among these new CB laws practically all of them embody substantially higher levels of independence than was the case in the pre-reform period. Prior to, and in some cases after, the enactment of the CB law most transition economies experienced high and variable inflation. In some cases those inflationary episodes even assumed hyperinflationary dimensions.1 The conjunction of high inflation and of CB reform provides a unique opportunity to examine the relationship between inflation and central bank independence (CBI) in environments with major structural changes and high inflation. This paper has two major purposes. The first is to document and quantify the cross sectional and over time variation in the level of legal CBI in transition economies in a manner that allows systematic comparisons with the independence of CBs in more mature market economies. The second is to examine whether higher legal CBI is associated with lower inflation as is the case in developed economies. 2 This negative association between inflation and legal independence is not obvious a priori for several reasons. First, as is the case in non-FSE developing countries, legal independence may be a poor proxy for actual independence because of substantial deviations between actual practice and the law. 3 Second, although CBI may be negatively associated with inflation in relatively stable Western democracies it may not be sufficient to contain the inflationary impact of large price shocks such as those that are induced by price decontrols and armed conflicts. A recent study by de Melo et al. (1996) reports that inflation is lower in transition economies with a higher level of sustained liberalization. A third purpose of the paper is to examine the relative contributions, if any, of liberalization and of legal CBI to the abatement of inflation. To this point there has been little systematic work on measuring legal CBI and its relation to inflation in the FSE. Two exceptions are the recent work of Loungani and Sheets (1997) who construct an index of legal independence for 12 FSE and relate it to the rate of inflation in those countries in 1993 and Neyapti (2002) who develops similar data for a sample of eight Central and East European countries between 1989 and 1996.4 This paper extends both samples along several dimensions. First, it includes 26 FSE and considers the association between inflation and CBI over the entire period between 1989 and 1998.5 Second, it provides indices for the detailed features of the new laws that are based on the codification system in Cukierman et al. (1992) and in Cukierman (1992, Chapter 19). This makes it possible to compare the level of independence of the newly created CBs with that of more established CBs in the world and to experiment with several alternative measures of independence. The indices of independence developed in the paper reveal that the legal independence of newly created CB in the FSE is higher than that of CBs in developed economies during the 1980s. In particular, at least eight of the newly created CBs possess levels of aggregate legal independence that exceed that of the highly independent Bundesbank during the 1980s. The evidence in the paper also shows that inflation and legal independence are negatively related but only after the process of liberalization has been sufficiently strong and sustained. The paper is organized as follows. Section 2 describes the methods used to construct detailed and aggregate indices of legal independence. It also presents the indices and puts them within an international context. Controlling for wars and the extent of sustained liberalization, Section 3 presents preliminary evidence on the relation between inflation and several aggregate indices of legal independence. The sample consists of a pooled cross section time series comprising three broad time periods; The period prior to the enactment of the first CB law, the period after the enactment of the last CB law and, for nine countries, a period between the enactment of a first and a second CB law. The main finding is that legal independence and inflation are unrelated. Section 4 takes a deeper look at this lack of association by also controlling for the temporary, but powerful, inflationary shocks induced by price decontrols and by allowing for the potential existence of an interaction between legal independence and sustained liberalization. The main finding is that the familiar, from developed economies, negative relation between inflation and legal independence appears also in the transition economies, but only after the process of sustained liberalization crosses a certain threshold. Section 5 examines the relationship between inflation and legal independence in a wider sample that includes the transition countries and the developed economies. Section 6 tests empirically several hypotheses about possible determinants of legal CB independence. This is followed by concluding remarks.

The creation, from scratch, of new CBs in the FSE provides a natural experiment for examining the extent of CB reform, its sociology and the relation between CBI and inflation in an environment characterized by fundamental structural reform. This paper develops extensive new data on the legal independence of CBs in the post CB reform period in the FSE. The new data is constructed in a manner that makes it comparable to earlier data on CBI in the industrial democracies and in, non-FSE developing countries. The data indicates that CB reform in the FSE during the 1990s has been quite ambitious. In spite of the large price shocks induced by the transformation from plan to market, or perhaps because of them, reformers in those countries chose to create CBs with levels of legal independence that are substantially higher, on average, than those of developed economies during the 1980s. But since it is likely that the average level of compliance with the law in the FSE is lower than compliance with it in Western democracies the discrepancy in actual independence may not be as large as appears to be the case from this comparison. Taken to the extreme, and in view of the large price shocks caused by liberalization, the preceding observation could lead to the hypothesis that differences in legal independence among the different FSE should not matter much for inflation. The evidence in this paper suggests that such a conclusion is too extreme. Controlling for other variables CBI is indeed unrelated to inflation during the early phases of liberalization. But for sufficiently high and sustained levels of liberalization, and controlling for wars and price liberalization, legal CBI and inflation are usually significantly and negatively related. These findings are consistent with the view that legal CBI, no matter how high, cannot contain the powerful inflationary impacts of price decontrols and of the liberalization of foreign trade and of the exchange rate. But once the process of liberalization has gone far enough legal independence turns out to be effective in slowing inflation down. The evidence also suggests that the negative association between CBI and inflation is stronger for narrow indices of independence that focus on the allocation of authority over monetary policy, the procedures for the resolution of conflicts, the focus on price stability and the seriousness of limitations on lending to government. The cumulative index of liberalization developed by de Melo et al. (1996) exerts a significant negative influence on inflation, as is the case in their paper, mainly at low levels of cumulative liberalization. Interestingly, at high levels of cumulative liberalization, after a sufficiently sustained experience with the new liberalized institutions has accumulated, CBI becomes relatively more important in keeping inflation down and cumulative liberalization becomes less significant. One possible explanation for the increase in the effectiveness of CBI at high levels of cumulative liberalization is that law abidance in transition economies rises with the level of cumulative liberalization. As a consequence the discrepancy between the relatively high legal independence of the CBs of economies in transition and their actual independence shrinks—and legal independence becomes more effective in keeping inflation at bay. A non negligible number of FSE have recently reached a range of values of cummulative liberalization for which CBI does have a significant moderating effect on inflation (see Table 7 of the appendix). After controlling for CBI, cumulative liberalization, wars and price decontrols, the evidence does not support the view that countries with currency boards enjoy lower rates of inflation. Examination of the relation between inflation and legal independence in an international sample composed of the transition and of the developed economies strengthens the conclusion that legal independence and inflation are negatively related internationally. Nine out of the 26 FSE in our sample had two CB reforms. The average level of independence of the first law in those countries was usually lower than the average level of independence in countries with only one CB reform. This was often followed by very high inflation in the countries with two reforms and then by enactment of a second CB law. The average level of independence embodied in the second law is substantially higher than that of the first law in those countries. It is also higher than the average level of independence in the 17 countries which had only one CB reform. A preliminary examination of the factors that determine the level of legal independence chosen by political authorities suggests that being on the first fast track to join the EMU has a positive effect on legal independence, and that the later the year of CB reform the higher the level of independence embodied in the new charter. A possible reason is that the importance of CBI as a signal of financial respectability, to gain access to international credit markets, rose through the 1990s with the further abolition of restrictions on capital flows and further widening of international capital markets. But the existing evidence does not support the view that countries with higher inflation subsequently grant higher independence to the CB. The fact that the average level of legal independence of the new CBs of economies in transition is substantially higher than that of developed economies during the 1980s at least partially reflects the shift in professional consensus among economists and policymakers in favor of CBI between those two decades. Our feeling is that, had CB reform in the transition economies taken place during the 1980s rather than during the 1990s, the level of CBI embodied in the new laws would have been significantly lower. This is one concrete illustration of the broader principle that existing professional consensus at the time of reform affects the pattern of reform.