This paper analyses and explains the effects of central bank independence on a country's inflation rates and its economic performance thereafter. It deals with the benefits believed to come along with independence and the delegation of monetary policy to the central bank, the determinants and accuracy of the index of central bank independence (CBI), and the different impact that CBI has on developed and developing countries. The studies and test conducted have shown that CBI lowers inflation in developed countries but in developing countries it might have the reverse effects, mainly due to the degree of independence, and factors like traditions, the law, and the statue of the economy which vary across countries.

1. Introduction

This paper intends to study the relationship between central bank independence (CBI) and inflation levels among different countries; developing and industrialized. The main research problem that I intend to examine is whether central bank independence can lower the inflation rates of the countries that grant independence to their central banks, and whether this can lead to improved economic performance.

What is discussed in this paper is not only whether central bank independence (CBI) can lower inflation and hence inflation variability, but also whether this can be achieved at low cost. The economies presented in the study include both those of the developed countries, e.g. the U.K., as well as those in transition e.g. Russia that have recently gained entrance in the EU. In this paper I analyze the impact of CBI on inflation, the benefits that are believed to come along with CBI and the factors used in measuring CBI. You will see that CBI can have different forms of measurement that produce slightly different results, hence the effectiveness of the CBI index is also analyzed.

The reason for choosing this topic for further study is simply because during the past two decades there has been a considerable move towards central bank independence across several countries, with the belief that this will improve their inflation levels and thus contribute to economic growth. However, as you will see further on, this is not always the case, as some studies have revealed contradicting results, and economists and academics continue their studies to get a clerer picture of this issue.

2. The Spread of CBI and The Reasons for It

To begin with, it is amazing how fast CBI has spread among countries and governments since the late 1980s. One simple explanation of this spread is A.Alesina's (1988; 845) statement:

...independent central banks have been associated with a lower average inflation rate and may have been responsible for reducing politically induced volatility of monetary policy and inflation

2.1 Should a central bank become independent?

Folder (2005) explains that CBI was adopted to avoid possible disputes between political parties as a central bank is seen as a provider of information. Many economists have expressed their opinion on the spread of CBI; others have linked it to a way of avoiding the blame of political failure by some governmental parties (Miller, G. 1998, White 1994). Others have linked it to the infrequent changes of the government (de Haan and van't Hag, 1995).

That is, central bank independence in many cases was adopted after periods of high inflation in order to reduce it, due to the inflation targeting function a central bank is capable of pursuing. It can also be associated with the attraction of foreign investment and hence economic growth as a consequence of the targets set and the autonomy with which the bank can then operate (Maxfield, 1997). In countries within the European Union, CBI is a perquisite following the Maastricht Treaty (1992) for adopting the euro currency.

Overall and according to Folder (2005), independence has always been related to the adoption of anti-inflationary measures for pursuing monetary policy, but its explanation lies within ...the sociology of the financial elites and the politics legitimizing their policy preferences. The reasons behind achieving price stability through gaining central bank independence, Cukierman (1996) explains are several and include; the breakdown of other institutions like the European Monetary System (EMS) that had been responsible for maintaining price stability which is considered as the single and most significant objective of a central bank.

Ilieva and Gregoriou (2005) suggest that in transition economies central bank independence has increased mainly due to the desire of such countries like e.g. Czech Republic, Poland, Romania, etc., to join the European Union and the acquis communautaire that applicant countries should adopt. As they continue to reason the addiction to CBI, they add that another incentive for CBI is the international financial institutions such as the IMF (International Monetary Fund) that require certain criteria to be met before making unconditional loans, and these criteria are feasibly met with the help of CBI. Also, countries are attracted to CBI as this will attract potential investors by improving the nation's creditworthiness.

Cukierman, A. (1996) analyses developments since the late 1980s to the legal independence of central banks and to its meaning; the measurement of CBI, the interaction of central banks with the government, its effect on the economy, its determinants, etc

According to Cukierman, the trend towards CBI is due to a quest for price stability which is due to the following two reasons:

First, following the stagflation of the seventies and the adverse economic performance of some high inflation countries, in Latin America and elsewhere, conventional wisdom concerning inflation and real growth has changed. Whereas during the sixties the accepted view was in line with Keynesian dogma, that some inflation is good for growth, during the eighties and nineties became that inflation and the associate uncertainties retard growth. (1996; 3)

The good economic performance of Japan and Germany, countries with already low inflation added more value to the above concept.

Second, the rapid growth and internationalization of capital markets raised the importance of price stability as governments and private investors sought to enhance their access to broadening world financial markets. (1996; 3)

2.2 Types of Central Bank Independence

Independence with regards to central banking can be categorized into different groups, depending on the degree of freedom and the subject from which the central bank becomes independent. The major types of independence are;

Legal independence, where the bank is partly accountable to the government and legislation provides a framework within which the central bank and the government cooperate on certain issues. This form of independence varies significantly among countries as it depends on how strong in the law in each country and the degree to which it is followed. However, the degree of legal independence, namely LVAW, as it will be shown below, has been used by many as a major index of measuring the degree of CBI.

Goal independence refers to the case where the central bank is allowed to set its own goals, e.g. price stability, money supply, inflation targeting. However in most cases under this type of independence, the bank will decide on its goals with the confirmation of the relevant governmental departments. In this way, goal independence helps avoiding conflicts among fiscal and monetary policies, and increases the level of transparency and credibility of the central bank over its goals.

Operational independence is the most common form of independence and is followed by many central banks around the world, for instance, the Bank of England since 1997. It involves the government setting the bank's goals e.g. a 2% level of inflation, but the central bank being free to choose the instruments e.g. interest rates, to meet the targets set by the government.

Another form of independence is managerial independence, by which the central bank has the power of appointing its own stuff, set its budget, etc. This form is a necessity for the existence of the other abovementioned forms of central bank independence and is therefore granted to all central banks that can call themselves independent.

2.3 The case for central bank independence

There is a huge surge towards central bank independence by both the public and the governments, in the belief that independent central banks will not only achieve low inflation rates and price stability, but will subsequently lead to long-term economic growth and development. However CBI is an issue that needs further research before determining whether it should be adopted by all countries. This depends on the economic state of the country, whether it is a developed or a developing country or even on the demand of autonomy by the political parties within the country since by granting independence the government must pass to the bank the responsibilities of e.g. controlling the interest rates, etc. over which it used to have the power.

Another issue that needs to be examined before granting independence to a central bank is the political stability and the degree of uncertainty within the country. This is because in times of uncertainty and instabilities, e.g. prior to elections, the public favours CBI as an independent central bank is more objective in its role and always forward looking without ignoring the long-term effects of its decisions.

The majority of the parties affected by the actions of an independent central bank, i.e. the government as well as the general public are attracted by CBI because of the greater accountability and transparency the bank is equipped with when adopting a greater degree of autonomy. Moreover, it is expected to bring lower levels of inflation and this is the main reason why people welcome CBI and the number of central banks becoming independent has been increasing over the years. The main reason behind this expectation is because a central bank generally acts in favor of the public and in addition to the fact that it becomes free from the government and any political pressures, it is in a position to avoid short-term temptations regarding low interest rates which the government usually uses prior to electoral periods, for the sake of long-term low inflation and price stability, which in combination with other exogenous factors can result in economic growth.

Moreover, when a central bank gains its independence through institutional reform it becomes capable of appointing its own governor thus it moves away from political interference, and can also set an explicit inflation target. Additionally and as Carlstrom, T.C. and Fuerst, S.T. (2006) explain independence helps a central bank in constraining the behavior of fiscal authorities. That is, it can prevent people and especially the government following fiscal policy from acting in their short-term best interests, recognizing that any actions taken in the short-term e.g. lowering the interest rates to attract investments, may become undesirable in the long-term, e.g. rising inflation levels as with higher demand from low interest rates, the prices will likely increase. In this way, CBI also prevents the fiscal authorities from inflating the short-term for delivering e.g. favorable exchange rates. Hence, monetary policy can run in a more credible way and following the targets set, markets will know what to expect thus shocks will be limited.

However an independent central bank is also likely in extreme cases to bring so low levels of inflation that can be harmful to the economy. According to Epstein, G. (2007), the 3.5% drop in inflation levels by countries adopting an inflation-targeting monetary policy (IMF, 2006) is questionable as to whether this decline will improve economic growth. Explicitly, if the inflation level of a country is already low and the central bank adopts an inflation-targeting monetary policy then the resulting lower inflation level might prove dangerous to the economy by generating economic cycles.

Cukierman (1996) has developed two separate approaches for reasoning the urge towards central bank independence and explaining the benefits that can be enjoyed from independence.

These include; the theoretical approach according to which in the short-run monetary policy can be conducted in such a way that it allows for some inflation so that it can achieve employment, high economic activity and low interest rates. Hence, policy makers can expect some degree of inflation which they will present in the form of nominal wage and capital market contracts. In this way however, policy makers will have to keep inflation at a level that would balance the real equilibrium if they had been committed to zero-inflation.

As a result of this discretionary use of monetary policy, this is subject to inflationary bias, and this bias can only be minimised if monetary policy is delegated to an independent central bank because only this institution is free to choose how to operate monetary policy and takes interest mostly if not only to price stability.

And the empirical approach by which the case of CBI lies on empirical evidence showing that countries with an independent central bank have lower inflation rates and higher growth rates per capita output. An example of such a country is New Zealand:

2.3.1 The case of New Zealand

New Zealand is a country whose central bank managed to drop the inflation level after being granted with greater independence. The Reserve Bank of New Zealand was granted independence in 1989 following the Reserve Bank of New Zealand Act of 1989 and had therefore established an explicit inflation target.

The result was to reduce inflation levels from 7.6% during the years 1955-1988 from when the reserve bank was not independent, down to just 2.7%, after becoming independent, during the period 1989-2000. The latter rate is now considered one of the lowest among industrialized countries.

It is obvious that among all OPEC countries, the central bank of New Zealand managed to achieve the lowest inflation rate, especially during the 1990s.

What happened during the period of the inflation reduction was that the reserve bank of New Zealand went through a reform that resulted in it being granted with independence and a greater degree of autonomy, leading to low inflation.

Specifically, prior to 1989 it used to be an arm of the government. Monetary policy used to be subject to the ministry of finance and therefore the government. As a result, the level of independence was one of the lowest among industrialized countries, while the level of inflation was of the highest. Even then, the relationship between central bank independence was negative, even though the results were the reverse of what is considered optimum, i.e. greater independence, lower inflation.

In 1989, the Reserve Bank of New Zealand Act was passed by law. This act codifies inflation targeting and gives more autonomy to the country's central bank in order to meet its objectives. According to the Act the central bank's primary function is:

...to formulate and implement monetary policy directed to the economic objective of achieving and maintaining stability in the general level of prices. (Reserve Bank of New Zealand Act, 1989 as quoted in Carlstrom T.C. and Fuerst, S.T., 2006, p.3).

The impact of the Act on New Zealand's economy and specifically the Reserve Bank's autonomy can be seen in figure 2 below, which compares the degree of independence across different time periods and among different countries.

The findings of the New Zealand case show that if the country had adopted independence earlier then its average inflation rate would be 3.4% rather than 7.6% that it actually used to be, assuming all other things being equal. Following this assumption, CBI itself would be sufficient to reduce worldwide inflation levels from 5.6% down to 3.8%.

Despite the considerable drop in New Zealand's inflation rate it is still questionable whether this drop was caused solely by CBI, and it is difficult, if not impossible, to quantify by how much the inflation reduction was due to CBI. Firm conclusions cannot be made yet since the data used in this case is of limited sample size and comparisons would therefore be insufficient. What is true is that the relationship between CBI and inflation is similar across time. Any changes to the strength of this relationship are mostly due to macroeconomic and other factors such as the state of the economy, the state of the government, e.g. democracy, etc. and others that will be explained later in this paper.

3. Measuring Central Bank Independence

The degree of CBI for each central bank varies according to the state of each country and to compliance with the law. As Cukierman explains, in developing countries where compliance with the law is poor, a suitable proxy for CBI would be the turnover of central bank governors, whereas in industrialised countries such a proxy would be legal independence. Generally, when the appropriate index of independence is used, the results indicate an inverse relationship between CBI and inflation.

However, care should be taken not to mistake legal independence with actual independence, as legal independence is necessary but does not guarantee actual independence; ...legal independence is a necessary, but not a sufficient condition for a truly independent CB. (Cukierman, A., 2001; 7). Exceptions exist, like developed countries, where legal independence seems to be a good proxy because law is highly complied.

For a clearer picture of the effectiveness of CBI on the economy, it is preferred that some variables that make up the CBI index are used in combination, or that some indices are used only for a specific purpose. For instance, legal independence is a good proxy for actual independence in developed countries rather than in developing ones.

3.1 The Cukierman Index of CBI

The method that will be used the most in this paper to measure the degree of central bank independence and its relationship with inflation will be Cukierman Index (1992), the most widely accepted and used index for this purpose. Initially, the exact definition of the Cukierman Index according to Siklos, P. (1992; 65) is: An indicator of the degree of autonomy enjoyed by several central banks.

Cukierman Index to demonstrate graphically the measures of CBI and inflation during two different decades, namely the 1980s and the 1990s:

As can be seen in the above graphs during the 1980s even though CBI was not common across countries, there was a negative relationship between CBI and inflation level. This means that the greater the level of independence of a central bank, the lower the level of inflation within the particular country. It is thus obvious that the correlation between CBI and inflation is negative, whereas the errors overall are not fitted closely on the regression line. We should note however that the decade of 1980s was before central banks especially those within industrialized countries underwent major reforms in their statutes which then allowed them a greater degree of autonomy.

During the 1990s as Siklos, P. (2002) explains, most central banks went through a reform, as there was a trend towards CBI. As a result the overall degree of CBI increased and all index values were revised upwards, the government granting more autonomy to central banks, in the belief that greater independence would just be adequate for lowering the level of inflation.

However, the relationship between CBI and inflation during the 1990s turned out to be the reverse of that of the previous decade. That is, the correlation between CBI-inflation now became weaker but positive since the regression line on the scatter gram in figure 3b has an upward slope, meaning that inflation increases with the degree of independence. It is hard to explain what was wrong with the findings of the 1990s that caused the correlation to be positive, however one might argue that CBI increased for all countries during the 1990s and so it also reflects the inflation performance of the previous decade, although the more independent central banks have delivered lower inflation levels in the 1980s.

Furthermore, the Cukierman Index used is believed to contain some inaccuracies concerning the measurements of the degree on independence and thereafter the relationship of that with inflation because it was extended from the 1980s towards the 1990s in a different way than the one initially specified. For this reason more tests will be carried out to explain and compare the effectiveness of measuring CBI using the Cukierman Index in contrast to other indices developed for the same purpose, for instance Alesina and Summer's Index.

The Cukierman Index will also be used to test the effect of CBI on inflation in transition economies, based on Ilieva and Gregoriou (2005) paper regarding inflation performance, i.e. average inflation and inflation variance, and CBI in transition economies during the period 1991-2003.

3.2 The determinants of the CBI index

The degree of independence varies across countries. This is not only due to factors such as the type of independence of each bank, although the most common is operational independence, the degree of law compliance in each country, and tradition, but some other systematic factors as well. Such factors are described and categorized by Cukierman, who presents some hypotheses on these factors:

Hypotheses about the determinants of CBI

Initially, it is widely accepted that any form of inflationary bias raises the independence of central banks to the degree that politicians wish to grant to the CB. The main idea behind this concept is that the benefits of delegating monetary policy to an independent central bank will be higher when inflation bias is higher in instances of e.g. employment reaction to inflation shocks. This delegation according to Cukierman helps in preventing the competing political party from taking on activities not favoured by the government.

Secondly, Cukierman et al (1992, 2001) make the hypothesis that:

...the wider are the financial markets and the more elastic the supply of funds to government with respect to the interest rate, the more likely is the CB to be independent. (2001; 19).

Additionally, Maxfield (1995) supports that political authorities favour CBI where there is need for funds. When this need is high as he explains, the government delegates more authority to the central bank in order to signal the nation's creditworthiness.

Finally, the cases of countries that have experienced extremely high levels of inflation in the past, like Germany, Austria, and Brazil, show that such countries are more likely to delegate independence to a central bank so that politicians do not interfere with monetary policy.

3.3 The measurement of the CBI Index

Due to the widespread concept that the degree of independence of a nation's central bank plays a crucial role upon the policy actions and inflation, Cukierman (1992) presents an analysis of the effects of CBI on inflation and provides various indicators of CBI. However, as he explains, the degree of CBI is determined by several factors from legal to cultural some of which are difficult to measure and quantify, therefore the impact of CBI on inflation varies among countries and there is a certain degree of uncertainty about the level of CBI. As a result, the measurement and the creation of an index of CBI have been based on legal independence, as the degree of CBI also depends on the degree of independence granted to the bank by the law.

Despite the variations in the degree of CBI, it can be deduced that a low degree of CBI is linked with higher levels of inflation and inflation variability, while the level of credibility of a central bank with a low degree of CBI will be lower.

Cukierman presents three different sets of indicators of CBI; a proxy for legal independence and proxies for the deviations of actual from legal independence. Independence measured under these proxies is limited specifically to the Central bank's ability to meet a single objective; price stability. The reason for using several proxies in measuring CBI is because each proxy ... is a noisy indicator that captures a somewhat different aspect of CB independence' (Cukierman, 1992; 370), so using a combination of them reduces this noisiness of the overall measure

3.3.1 Measuring and Coding Legal Central Bank Independence

Using a proxy of legal independence is vital in making comparisons with previous studies on the impact of CBI on economic issues because all existing attempts on the features of an independent central bank rely on the bank's legal independence.

Cukierman presents the indices of legal aspects of CBI by separating into four groups the variables which make for a legally independent central bank. These groups are:

Chief executive officer: CEO Policy formulation: PF

Final Objectives: OBJ Limitations on lending: LLand codes them by the degree of independence of each group for the central bank of each of the countries included in the study.

The main assumptions made are; the central banks whose single objective is price stability are considered to be more independent, so are central banks with stricter limitations on lending from the CB. The coding involves sixteen different variables in a scale from 0 (least independence) to 1 (maximum independence), during the time period 1950-1989, separated into four different decades.

Due to the narrow definition of each of the variables used and the consequent lack of precision and multicollinearity problems that may arise, these variables are aggregated into eight legal variables by just calculating the unweighted mean of the codings used.

Furthermore, it is necessary to have an additional single index of legal independence for each country to assess the aggregate legal independence of the CB. This index can have two alternatives, the LVAU and the LVAW, that are computed by calculating the average of the codings of the first eight variables as described above. Table 1 in Appendix A shows the ranking of the countries according to the legal independence of their central banks as measured by the LVAU during the eighties decade. The LVAW would also give a similar picture.

Looking at the table of results one can see that among the seven most highly-ranked countries four are developed (Switzerland, West Germany, Austria and U.S.), while among the seven least-ranked countries four are less developed (Morocco, Panama, Yugoslavia and Poland). Generally, the top 10% of the rankings is comprised of developed countries, whereas the bottom 10% is concentrated with less developed countries. One should also note that there had been no hyperinflation experienced by developed countries during the 1980s, while some of the Latin America countries have, e.g. Brazil and Bolivia with a rate of 230%. This according to Cukierman may suggest that legal CBI may be neither necessary nor sufficient for low inflation. (1992; 382).

3.3.2 The turnover rate of Central Bank governors as a proxy for actual independence

As already explained, the legal status of the central bank is just one of the several determinants of actual CBI. There is no clear systematic indicator of actual CBI, but Cukierman (1992) presents two sets of such indicators. One is based on the actual turnover rate of the central bank's governor, and the other is based on the answers given to a questionnaire on CBI.

Table 2 in Appendix B shows the CB governors' turnover rates for the period 1980-1989. It is assumed that the lower the turnover rate the higher the degree of actual independence. Although the results are chronologically old, it is obvious that turnover rates in less developed countries occupy a range that has never been experienced by developed countries. It is indicative that more than half of the less developed countries have a turnover rate higher than the maximum of the rate of developed countries. It is clear that less developed countries experience higher inflation rates, on the grounds of lower actual CBI.

On the other hand, low turnover does not necessarily imply a high level of CB independence on the grounds that a relatively subservient governor will tend to stay in office longer than a governor who stands up to the executive branch. (Cukierman, 1992; 385)

Critically assessing the results, since the maximum turnover rate for developed countries is 0.2 (.e. five years) suggests that the turnover proxy may not be effective proxies for actual CBI for the sample of developed countries, whereas this proxy can be considered indicative for the sample of developing countries since these have turnover rates exceeding 0.2.

3.3.3 Central Bank Independence from answers to a questionnaire

Another aspect of characterizing CBI is the questionnaire. Under this method, answers were obtained from qualified central bankers from twenty-four countries during the period 1980-1989. The main questions asked covered the issues of; legal independence, final monetary policy objectives, monetary policy instruments, actual independence and its divergence from the law and intermediate targets and their indicators.

In coding the variables of the questionnaire, the bank is assumed to be more independent, all other things being equal, if the following hold; the term of office of the CB governor is longer than that of the government, limitations exist on lending from the CB which the government is in no position of altering, and in cases where stock targets exist because these mean that the CB is more free to meet its price stability target.

Table 3 in Appendix C shows the ranking of central banks by aggregate indices of independence according to questionnaire responses. The aggregate indices of QVAU and QVAW reflect the law and the way it is implemented in practice respectively, as well as important information about actual independence, and are very similar (?=0.99). The rankings agree to earlier studies that central banks of developed countries are more independent. However, the median of QVAU for developed countries, that is 0.6 for Britain and Lebanon, is greater than the median for less developed countries, that is 0.49 for Uruguay, and this contradicts the above findings for legal independence using the LVAU.

When measuring the degree of CBI it should be taken into account that the measures used above fail to quantify all the aspects of CBI as some are difficult to quantify. Such aspects are the quality of the bank's research department and its standing in comparison to other economic research institutions within the public sector (Cukierman, 1992). Independence is generally higher in countries with highly-developed financial markets according to Cukierman because the supervision of financial institutions is under the authority of the CB, so the larger the market the more wide the span of the CB's authority.

4. Central Bank Independence and Inflation Targeting

In this section the impact of central bank independence on inflation, inflation variability and the economy overall is analyzed using a model to test whether CBI can actually lower inflation, and comparing the effects of CBI by using both the Cukierman and the Alesina indices of CBI. Additionally, the costs of achieving lower inflation through central bank independence are also explained.

MacCallum, B. (1995) believes that it is strong will that is necessary for proper policy behaviour by central banks, not rules and regulations. A policy maker, i.e. a central banker in this case should act immediately to an inflation shock to restore the problem without letting any space for inflation bias to creep in. Rogoff, K. (1985) adds to this point by suggesting that nations should appoint central bankers who are more risk averse than the public, so that they act vigorously and minimise the problem of inflation bias that may arise. However, in an attempt to maintain low levels of inflation, central bankers may also push the economy into a phase of recession.

4.1 CBI against Inflation and Inflation variability

Fuhrer, C.J. (1997) critically assesses the theory of inherent inflation bias while maintaining CBI and low inflation on the basis that an independent central bank should have instrument and not goal independence.

A good monetary policy should be designed to be effective under all conditions thus a central bank shall not be tight to a single rule but instead to a policy that can be continuously adjusted to fit all needs and conditions according to the country's state of the economy and needs.

The graph below shows the history of the inflation rates of OECD countries since 1970. As can be seen from the graph, during the 1970s inflation was high for most countries, mostly due to oil price shocks, a condition also experienced nowadays. On the other hand, inflation dropped during the early 1980s. This disinflation is explained by Fuhrer (1997) as the favourable consequence from the will of the monetary policy makers, both the central bankers and governments, and exported disinflation, i.e. through trading from the leading countries towards the rest. Exceptions to this rule are Australia and New Zealand that managed to lower inflation later, during the mid-1980s after their corresponding central banks underwent some reforms, but have since then the best performance.

4.1.1 Comparing between two different indices of CBI

Despite the considerable drop in inflation rates as shown above, some questions still need further tests and study. For instance, are measures of CBI correlated with monetary policy success? Do countries with a higher degree of CBI achieve lower rates of inflation?

What the two scatter graphs above reveal is the expected answer, that central banks can control and hence lower inflation through their independence, however there are some important differences between the two graphs that call for further studies on the measurement of the CBI index. This is because the first graph (1a) uses the Alesina and Summers Index of CBI and reveals a strong negative correlation (-0.8561) between CBI and inflation. The second graph (1b) uses the more widely accepted index of CBI, the known Cukierman Index, and reveals a negative correlation too, but this time weaker than the previous one as its value is now -0.6576. Thenegative correlation indicates that CBI and inflation are negatively related, that is, the greater the degree of independence of a central bank, the lower will the inflation level of that country be.

As far as the robustness of the correlation between CBI and inflation is concerned, this is tested econometrically by Fuhrer (1997) and the results are listed on the table presented on the next page.

The table lists simple regressions for a variety of specifications and countries. Row 1 All countries lists simple regressions of average inflation rates during the period 1960-1989 on the average CBI index, corresponding to the scatter-graphs used earlier. Row 2 Countries with unemployment data lists controls for the cross-country variation in macroeconomic variables. Row 3 OECD countries lists the effects of annual variations in macroeconomic variables. The last two rows list a summary of statistics for fixed-effects regressions including the CBI index as a time-varying regressor.

Moreover, the CBI index is set to control whether CBI index and inflation are jointly determined at the annual frequency. If yes, then the regression coefficients are subject to bias of simultaneity, because; an inflation shock may lead to a legislative change in CBI index, which in turn it could be correlated with the annual inflation residual, causing the bias. The fixed-effect used involves the country-specific intercept term in each regression, and includes the cross-sectional variations in inflation, i.e. the t-statistic for regression on CBI index, and the R2.

and according to Fuhrer, there is no significant correlation between inflation and CBI in cross-country variations, neither in time variations within countries. The only significant negative correlation between CBI and inflation arises within the Alesina and Summers bivariate regression model in the cross-country sample. The OECD countries indicate a less significant coefficient of CBI.

4.2 A fully independent or a conservative central bank?

In this section a model of the actions of the government regarding CBI and economic performance is presented as conducted by Weymark (2006) in an attempt to distinguish the objectives of the government and the central bank and reach the optimum degree of independence for each central bank in order to produce a clearer view on the actual impact of CBI on inflation.

The inflationary bias problem mentioned earlier arises due to a potential inflation shock that can increase output. The reason behind this problem is that economic agents expect the government to use inflation shock to increase output, so they demand higher wages, neutralizing the impact of this bias on prices.

Rogoff (1985) suggested that the solution to inflationary bias would be an independent central bank that would put more weight on inflation control because of instrument and goal independence. However the degree of independence of central banks differs across countries, and even the most independent of all central banks are still under some governmental oversight.

There is a belief that partial central bank independence can also be optimal, and for this reason Weymark (2006) supports that the inverse relationship between CBI and inflation especially across industrialized countries is not casual, but there are some country-specific factors which determine inflation performance and the degree of CBI. There is a need to distinguish the concepts of central bank independence and central bank conservatism that used to mean the same in some studies on measuring the degree of CBI as such studies assumed CBI to be exogenous.

4.2.1 The economic model

In the model there are two groups of agents; private agents who are assumed to be rational, and the government, whose actions through inflation expectations determine economic performance.

The government is the fiscal authority in the economy and has two instruments; taxes and transfer payments. The central bank is the third agent in the economy and the monetary authority whose instrument is money supply. Assuming all agents share the same information, then the basic structure of the economy is described by the following equations:


et = expected inflation rate from rational agents

b = proportion of change in pre-tax income/(output) transferred to the wealthy

t = inflation rate in period t τrt = growth in transfer payments

yt = short-run output growth τt = overall growth in government tax revenues

mt = nominal money supply growth τxt = growth in discretionary tax revenues

τ = income-tax rate ut, εt = random disturbances independently distributed

s = savings rate α, β, b, s, τ = parameters assumed to be positive

According to the equations above, inflation is increasing by the rate expected by economic agents and output is positively related to growth in the real money supply. Equation (3) represents the government's budget constraint; the government must either sell bonds to the public or to the central bank, or levy taxes to finance its expenditure.

When formulating the economic structure, Weymark has assumed than government expenditure do not affect output, which therefore implies that the government can affect output and inflation performance only by its choice of central bank design (2006; 301). Also, she has assumed that the tax revenues for the government do not come from increases in the income-tax rate (τ).

4.2.2 Government and Central Bank objectives

The central bank and the government have different objectives; while the main objectives of the government are inflation stabilisation, output growth, making transfer payments and levying tax for revenue, the central bank's main objective is price stability that can be met through controlling inflation.

The government's loss function is represented as:

Where: π = the government's inflation target

λg1 = the weight assigned to output growth

λg2 = weight put on income redistribution

λg3 = weight put on tax penalty

θ = share of output to be allocated to the rich

What the loss function of the government tries to capture is that the government loses from inflation deviations from the target, and reflects the political reality. That is, that voters support the government when output growth is increasing. Lastly it represents the ideal transfer payments for the government. In order to make these, the government must sell bonds to the central bank or increase taxes.

By selling its bonds, there is an increase in the money supply in the economy which in turn increases the level if inflation. On the other hand, the freedom of the central bank to establish its own objectives depends on the degree of independence that it has been granted by the government. The latter however can impose several restrictions on CBI in an attempt to achieve governmental targets instead.

As Weymark explains, the government may have automatic credit facilities which force the central bank to finance government expenditures upon demand. There may be provisions that reduce the central bank's economic independence [...] without having to take into account the objectives of the government. (2006; 302).

The central bank's objective function which maintains economic independence and at the same time full political dependence is:


= the central bank's degree of economic independence, and λcb = the weight the CB assigns to output growth.

δ measures the degree of independence that a government grants on a CB. CBI is incomplete if the government has the power to impose any of its own objectives on the central bank.

Assuming that the government and the central bank have the same target regarding inflation, and that it has full control of λcb then the central bank is regarded as being politically dependent on the government, and the analysis that proceeds will indicate how this dependence can affect the relationship between CBI and inflation. A policy game has been developed to explain this, in which the government chooses the parameters δ and λcb , and both the government and the monetary authority, i.e. the central bank, choose their policy instruments (τxt, τrt , mt). The central bank is assumed to control the money supply by having full instrument independence, and must set mt at a level to minimize the losses from the degree of CBI, δ, and conservatism, λcb, granted by the government.

It is also assumed that the government and the central bank can observe white-noise disturbances (ut, εt) before they make expectations about inflation rates, thus their expectations are rational. Considering that private agents cannot observe those disturbances, and that the government's expected loss is constant for all combinations of λcb and 0< δ <1, then the government's expected loss is given by:

Whereas the relationship between δ and λcb is given by:

Where ... = 1+αβ-β(b-θ-σb), and σ = τ+s(1-τ).

4.2.3 Optimal conservatism and economic independence:

Considering the above equations, Weymark supports that partial central bank independence can also be effective if central bank conservatism is chosen appropriately. By differentiating the equation of δ above gives ∂δ/∂λcb < 0 which means that the degree of central bank independence and conservatism are substitutes. Hence when the best combination between δ and λcb is met, then λcb < λg1 requires:

Therefore the design of the central bank, i.e. the degree of its conservatism and independence, depend on the government's preference of output expansion (αβ), tax aversion, and the inflation cost of monetary expansion. Consequently, as Weymark explains, governments with a priority of output expansion are more likely to grant more independence to the central bank so as to prevent the economy from a high inflation rate that might result from the expansionary policy. Also, the problem of inflation bias can be eliminated completely if the following condition is satisfied: δ = λcb = 0, which is only optimal if λg3 = 0 as well.

The above model however can be criticised for not taking into account the political process involved during the design and structuring of the central bank while Cukierman (1992) has argued that political instability is fundamental in determining the optimum level of CBI. It can also be criticised of containing more instruments than objectives.

It is clear though that more CBI is preferred than less, and that in some cases complete economic independence (i.e. δ=0) is not the only solution to the high inflation problem, as a more liberal central banker can also be appointed where appropriate.

4.3 Inflation and Central Bank Independence

Earlier studies from Cukierman (1992), Alesina and Summers (1993) and others have shown that the inflation rate and central bank independence have a negative relation in high-income countries and support that inflation control is best when central banks are granted full independence, but Weymark argues that if this is true, then CBI has no impact on inflation if independence is accompanied by central bank conservatism.

An interpretation of this is the fact that this relationship between CBI and inflation rate arises from the cross-country variations in economic structure and government objectives which jointly determine both the choice of institutional arrangements and inflation performance (Weymark 2006; 309).

Weymark uses the coefficient used by Gini in Dolmas et al (2000) to explain the positive relationship between income inequality and average inflation and CBI thereon. As they explain, the government's willing to redress income inequality depends on the weight put on income redistribution and on tax-levying. This income redistribution is given by (2006; 310):

.According to this formula, the transfer of income to the poor is negatively related to the government's will of avoiding tax penalties and positively related to the government's will of income redistribution. And since the inflation target and expectation is given by: t = + λg3 , inflation increasing for all combinations of central bank independence and conservatism.

Considering the policy game described by Weymark, when the government designs the optimum monetary institution there is a negative relationship between central bank independence and inflation. On the other hand, when central bank conservatism is low, i.e. when λcb is high, CBI and inflation can be positively related.

Subsequently, governments of low income countries are more likely to appoint central bankers with low conservatism given their independence. This will later prove dangerous to the economy as it can be reflected in cross-sectional data as a positive relation between CBI and inflation. The reason is that given the most efficient degree of central bank conservatism:

The inflation rate when a central banker with weight on growth of λcb = λcb + ηis appointed, is given by:

For which CBI and inflation are positively correlated for η > 0.

The above agrees with Cukierman's (1992) conclusion that countries with high income are expected to have a negative relationship between CBI and inflation, whereas low (and middle) income countries experience a positive correlation between CBI and inflation. This different relationship between CBI and inflation experienced among developing countries will be explained in the subsequent chapter.

4.4 The accuracy of the CBI Index

Evidence suggests the assumption that the CBI index used so far is free from bias is not true. For instance, taking account of the different results from the scatter-graphs of the correlations between CBI and inflation obtained using two different CBI indices (i.e. Alesina and Summers, and Cukerman), urges us to use some econometric models to find out whether the index used for measuring CBI lacks integrity

Brumm, J.H. (2006) supports that the use of methods being inaccurate in their CBI proxy results in obtaining spurious results. As a result, Brumm uses alternative methodologies to test the accuracy of the CBI index used by other authors.

Beginning with Cukierman's Index, Cukierman, A. (1992) used the LVAW to measure the central banks' legal independence. However this is criticized as being improper as legal and actual independence of a central bank differ especially within countries where the monetary policy practice deviates from the law, so a different model must be used to account for this error in the forms of independence.

Campillo and Miron in their studies have used ordinary least squares (OLS) to estimate average inflation during the period 1973-1994:



For which the following apply:

regressand: INFL7394 = Average inflation during 1973-1994


INF4872 = average inflation 1948-1972

POLITICL = political instability

IMPSHARE = average of imports as a measure of GDP

EXCHRATE = the log of 1980 GDP per capita, a variable for exchange rate regime

DEBTSHAR = the 1975 national debt as a % figure of GDP

DATAQUAL = a data quality score

CBI = the LVAW index for central bank independence

The results are shown in the table below:

Prior studies have set 0<CBI = LVAW<1 where 0 denotes minimum and 1 maximum degree of independence. In this study however, before setting CBI = LVAW, one should firstly assume that CBI is measured without errors. That is:

LVAW= λ1CBI+δ1.......... (2)

Where: λ1=1 (coefficient) and δ1=0 (error measurement). We must also take account of measurement errors in order to have reliable findings, so Brumm suggests that the covariance structure analysis (CSA) to be used instead. Under this, equation (1) is constructed as above, and then measurement is made using equation (2) plus two additional equations:

TURNOVER=λ2CBI+δ2 .......... (3)

VULNERBL=λ3CBI+δ3 .......... (4)

Since LVAW is an indicator of CBI, then λ1>0. However, under the two above equations, TURNOVER is a measure of change in central bank government, thus λ2<0, λ3<0, and VULNERBL is an index of central bank political vulnerability. Therefore, the higher the values of these two variables the lower is the autonomy and degree of independence of a central bank.

After the tests explained above, the results from table 1 (p.40) show that these are not credible due to the measurement error since the LVAW is not statistically significant, as the coefficient of all three proxies are positive.

Table 2 below shows the estimates of equations 1-4 using the maximum likelihood method (ML) plus the coefficients of CBI (γ) and the CBI indicators (λ1, λ2, λ3). As can be seen from the table, the estimated coefficient of each indicator is negative, so is the estimated coefficient of the CBI which is also statistically significant at the 1% levels of significance. Additionally, LVAW is not significantly different from zero.

It can be deduced that central bank independence is measured with error and this can be inferred by the error bias present when substituting a CBI proxy into an inflation regressor as with equation (1). Instead, the CSA method has been used which revealed a strong negative relationship between CBI and inflation, the cause of it might be the omitted variables from equation (1), for instance political instability, exchange rates, data quality, etc.

4.5 The costs of achieving lower inflation

Although CBI is associated with low inflation and output variability, Demertzis, M. (2001) supports that this is not always the case as the consequences of monetary policy run by an independent central bank depend on the will of the government and political stability. In order to examine the impact of low inflation from CBI on the economy, Demertzis undertakes some tests based on Alesina's model by varying the degree of electoral uncertainty which as it will be shown, is a major determinant of CBI.

The gains of inflation achieved from CBI should be measured on the ability of each nation's central bank to maintain and promote growth at the same time as they maintain low and stable inflation. There is a conflict however on whether low inflation as a result of CBI comes at no cost; hence Demertzis is testing this numerically.

Alesina and Summers (1993) conclude that there is no relation between legal independence and output variability across industrialized countries. On the other hand, Rogoff (1985) finds a trade-off between inflation and stabilization policy, while Crosby (1998) argues that countries with lower output variability tend to choose to have an independent central bank. Alesina and Gatti (1995) support that both inflation and output variability under an independent central bank depend on electoral uncertainty and reach their conclusions assuming a two-party system with a probability of elections 50 per cent to each.

Since equal probabilities for each party are rare, Demertzis presents another version of the original Alesina and Gatti model to find the merits of CBI on the economy under different levels of political uncertainty.

4.5.1 The model

The model as initially presented by Alesina and Gatti (1995) and modified by Demertzis (2001) assumes a supply function in which output is positively affected by the inflation and output supply shocks, ε, with zero mean and finite variance.

yt = πt - πtε + εt, where ε ~ (0, σ2ε)..........(1)

As the authors explain, the timing is such that the shocks occur only after the private sector has negotiated wages to which the authorities react. Hence the government optimises the loss function as expressed below in terms of output and inflation:

minπL = [(πt)2 + β(yt - k)2..........(2)

For which (yt - k) represents the source of bias in the objective function and β the government's preference parameter with respect to the inflation and output parameters. The government aims an output level of k > 0, i.e. higher than the natural rate.

Substituting the above two equations and minimising the loss function with respect to π, gives:

π = βk - [β/(1+β)]ε the policy rule which embodies the inflation bias, πε = βk. Substituting these into equation (1) gives:

y = [1/(1 + β)]ε

And it follows that:

E(π) = βk E(y) = 0

Var (π) = [β/(1+β)]2σ2ε Var (y) = = [1/(1+β)]2σ2ε

Which show that; if inflation was the responsibility of a body aiming at reducing inflation -the independent central bank in this case- the results would indicate clearly that inflation is lower but at the expense of greater output variability. This inference agrees to Rogoff's (1985) studies.

Alesina and Gatti use the above described model to include an assumption under which two political parties compete for the government, wages are negotiated and one of the parties wins the elections. The shocks (ε) occur and the winning party has to react, which creates electoral uncertainty and bias over the private sector's inflation expectations.

Under this scenario, minimising inflation from the political decision process brings both lower inflation and lower output variability in the economy due to output instability. Output instability is a result of; exogenous shocks and uncertainty over which party will rule after the elections. Hence, by delegating monetary policy to an institution other than the government, output variability due to uncertainty is removed, thus output variability as a whole is reduced, whereas the variability of output is determined by the size of the exogenous shocks and political uncertainty.

Alesina and Gatti argue in Demertzis that the advantages of delegating monetary policy to an independent conservative body are; price stability and the elimination of output instability.

Central Bank Independence thus produces both lower inflation and lower output variability, a clear Pareto improvement in other words. Political parties consent to delegating monetary policy to an independent authority because they are both made better off as a result. (2001; 11)

With an independent central bank running monetary policy, what we get is:

E(Ι) = bk Var(Ι) = [1/(1+b)22ε

Whilst the expectations prior to elections and without the Independent Central Bank acting but with the two competing parties (D for Democrats and R for Republicans) instead, we get:

It is worth emphasizing Demertzis' concluding comment that for Central Bank Independence to come at no extra cost, it is important that both inflationary and output variability are reduced with an autonomous monetary institution. (2001; 12).

The analysis continues with examining the benefits of an independent central bank under a varying political uncertainty, i.e. P[0,1], and the way inflation and output variability are affected. For Demertzis this analysis is important in realising whether low inflation can be achieved at no cost because political uncertainty varies while the actions of an independent central bank with fixed preferences do not.

A limitation to the analysis as presented above is that it is limited in providing policy answers towards attaining inflation and output stability by using a single instrument. The analysis does not take into account other forms of policies and objectives, for instance, fiscal policy, thus the effectiveness of an independent central bank to actually deliver low inflation at no cost is not absolutely analysed.

5. CBI and Inflation In Economies In Transition

During the early 1990s, all formerly socialist economies (FSE) like Bulgaria, Czech Republic, Hungary, Slovakia and Slovenia, in an attempt to depart the Monobanks from their economies and restructure their economies into market economies, have established new and independent central banks whose features are alike those of the western style central banks and the Bundesbank. However, due to the lack of experience of being part of the market economy, the central bank still is participating in advising and decision-making on issues of converting the economy from centrally-planned to market economy, such as the appropriate use of instruments and the use and setting of the exchange rate.

Following Cukierman's (2001) point of view, even if there is a high degree of legal independence in the central banks of FSE, due to the problems of economic transition like the limited ability of the governments to finance debt, the independent central banks will need more than just their legal independence to lower inflation. That is, deficits will only be reduced by high levels of inflation as such countries are financed by inflation; hence the only way to lower inflation is for the government to balance the budget.

As a result, the relationship between CBI and inflation in developing countries is likely to differ from the relationship among developed countries. Below is an example that contradicts the effects of CBI on inflation as examined from the case of New Zealand (p.11): The Russian case.

5.1 The Central Bank of Russia and Russian Inflation

The case of the Russian economy comes to contradict the view that CBI and inflation are inversely related. This is because despite the Russian Central Bank (CBR) being legally independent, inflation is still high. CBR has announced that the factors that keep inflation persistently high are mostly non-monetary, and relate factors like monopolistic pricing, deregulation, etc.

According to Popov (1994) the CBR used its independence to boost inflation as its Finance Minister, Fyodorov, argued during the early 1990s that the source of inflation was indeed monetary, so he granted to the CBR the responsibility of monetary expansion as a solution to the economic problem.

the difference in inflation rates especially during the 1990s decade is huge between countries of the Former Soviet Union (FSU) and the rest of the world, excluding Latin America. Inflation in FSU is still higher than developed and developing countries the European inclusive, and the U.S, even during the past decade, a long time after the split of the FSU and the adoption of independent central banks by each of the countries that formed part of the FSU in the past.

The independence of the CBR however is doubted by several economists as Popov (1994) explains, the Russian government used to borrow money from the CBR at low nominal interest rates (10%) during the period 1992-1993. Also, the CBR acts not like an independent central bank should but instead as a proxy to the parliament which leads to higher inflation because:

... parliament represents many constituencies and because the CBR is highly dependent on parliament rather than being independent. (Popov(1994) in Cukierman 2001; 40).

This incidence of high inflation and inflationary bias cannot vanish from the Russian economy with just granting the central bank more independence, but actions are required from every party involved with the nation's financial matters; government and the banking system:

Once inflation has been allowed to develop, raising the legal independence of the CB without complementary measures is unlikely to stop inflation. (Cukierman, 1992; 449)

5.2 Institutional developments and CBI in transition economies:

The Bundesbank's constitution formed the basic structure of an independent central bank and has been followed by the ECB (European Central Bank) to deliver low inflation.

The acquis communautaire of Stage 2 of the European Monetary Union that must be followed by all new E.U. entrants states:

...Central Bank Independence, co-ordination of economic policies, and adherence to the relevant provisions of the 'Stability and Growth Pact'. New member states will have to forego any direct central bank financing of public sector deficits as well as privileged access of public authorities to financial institutions.

(European Commission, Agenda 2000, Supplement 13/97, p.44)

It is therefore clear that central bank independence is a legislative requirement for induction in the E.U.

Attaining CBI for autonomous monetary policy, as Ilieva and Gregoriou (2005) explain depends largely on the term of office of central bankers. Cukierman (1992) also uses the rate of central bank governors' turnover to determine the degree of central bank independence. Generally, the higher the turnover rate the lower the level of independence. Of course, some exceptions are always found. For instance the National Bank of the Czech Republic (NCB) has had its governor for over a decade now, following the political turbulences of the country before and after the split of Czechoslovakia in 1992.

In transition economies, central bank independence is focused on the central bank's economic independence, meaning its ability to choose on its own the instrument for pursuing monetary policy and achieving its target of maintaining price stability. The constraint in that according to Ilieva (2005) is the use of monetary instruments for financing public debt.

Ilieva and Gregoriou (2005) continue their research on CBI in transition economies to conclude that in their sample of countries most countries have adopted CBI as their

Central banks have gained legislative independence from their governments in order to fulfil the EU induction requirements as well as the requirements by international institutions concerning loans, debt rescheduling, etc. However, despite the legislative independence of most transition economies, Central Banks are often trapped to follow instructions from the government even though these contrast the existing legislation.

As far as the central banks' governors' turnover rate is concerned, Dvorsky (2000) in her study of five countries with economies in transition, namely Czech Republic Hungary, Poland, Slovakia and Slovenia, has come to a conclusion that conflicts with Cukierman's (1992) study of CBI and its relation to this turnover rate. Dvorsky's purpose was the measurement of actual independence according to the governors' turnover rate using an updated index. The study showed extremely low levels of turnover rates, even below the 0.25 benchmark, meaning that the turnover rate is not sufficient for reflecting the degree of actual independence in transition economies. However, the observation only lasted for eleven years so the results may not be adequate for firm conclusions.

On the other hand, Cukierman et al. (2000) conducted the same study but using a larger sample of twenty-six rather than just five transition economies concluded that:

...the levels of their legal independence are substantially higher, on average, than those of developed economies during the eighties. (p.10). One should note though that the indices used in the study were designed for OECD countries, therefore a new CBI index shall be designed for non-OECD countries, i.e. those in transition. Generally, in structuring a new index for measuring Central bank independence it is not just the turnover rate of the central banks' governors that should be taken into account, but also the objectives of each central bank e.g. price stability that affect its performance, autonomy of the implementation of monetary policy, and the legislation and tradition that can create trade-offs and interfere with the banks' independence and thereon the inflation levels. These are reflected in the questionnaire used for forming a new CBI index, in Appendix D along with a ranking of independence.

Maliszewski (2000) supports that the main cause for the high inflation in Eastern and Central Europe, i.e. in countries with economies in transition, are political factors which have led to monetary expansion. He finds that the more advanced transition economies also have more strict lending limitations. As a result the CBI coefficient used in the model for measuring the inflation variable with CBI might be biased if the CLI is not included in as well. Hence, the CLI is used as an explanatory variable in the following model for measuring the effect of CBI on average inflation during the period 1991-2003:

Iit = β1 + β2CBIit + β3CLTit + β4IMFit + β5DEFit + β6BZit + εit


Iit = transformed inflation for country i in time t

CBIit = central bank independence index for country i in period t

IMFit = a dummy equal to 1 at the signing of the IMF agreement, and equal to 0 thereon

DEFit = general budget government deficit for county i at period t

BZit = a dummy equal to 1 before the adoption of a country's own currency, and equal to 0 afterwards.

β1 = constant

εit = random error, εit ~ (0, σ2)

Table 1 below shows the results of the above model, where * denotes statistical significance at the 95% significance level, with the t-statistic in bold and the standard error in brackets.

Time Period











0.83 (0.13)

0.03 (0.014)

0.09 (0.015)

0.21 (0.07)

0.31 (0.07)

0.22 (0.08)








Table1: Central Bank Independence and Average Inflation

From the results above we can see that 74% of the inflation variation (i.e. the value of R bar squared) can be explained by the variations in the explanatory variables in the above model, which CBI is a part of it. Moreover, all explanatory variables used are statistically significant, and most importantly, inflation is reduced as central banks become independent. According to both Cukierman (1998) and Maliszewski (2000) this is true when the country adopts its own currency and IMF (β4) programmes are introduced into the model. The inverse occurs when government deficit (β5) is greater.

A similar study was conducted by Cukierman et al. (1998) with the only difference of using inflation variability measured by the standard deviation of the transformed inflation of a country at different time periods, in place of the inflation transformer used above. The results showed a lower value of R2 of 36%, but here again the fundamental finding was that as central banks become independent, inflation and its variability become lower.

6. Testing For Regression

In this chapter the focus is on a test I have conducted on whether there is a regression between the inflation rates of the U.K. during the periods 1986-2007, that is, the decade prior to and the decade after, the Bank of England (BoE) was granted with independence in 1997, in relation to the country's GDP and exchange rates as set by the Bank of England, during the same period too.

Using the simple regression method of Ordinary Least Squares (OLS) I formed the following equation:

INF9707 = α + β1INF8696 + β2GDP + β3XR + εt

INF9707 is the regressand of the equation and stands for the average inflation during the years 1997-2007. The regressors of the equation are factors which reflect the impact of inflation in their value; Gross Domestic Product (GDP) on the basis of total income and the exchange rate of the pound sterling to the U.S. dollar. Both are represented using quarterly rates for the period 1997-2007, post-CBI. Another regressor is the average inflation a decade prior to the Bank's independence (i.e. 1986-1996). α is the constant to the equation, and εt represents the errors to the equation.

On the basis of the above equation the following hypothesis is formed and tested:

H0: β = 0

H1: β 0

The null hypothesis (H0) represents the case where the variable INF8696 is insignificant and thus it has no effect on the dependent variable INF9707. The general rule under the t-test for testing the null hypothesis is that the null shall be rejected if the t-prob of the test is more than 0.05 (5% significance level).

This means that the null hypothesis: H0: β = 0 is rejected. In other words, the variables are significant and have an impact on the rate of inflation during the period 1997-2007.

Additionally, the Durbin-Watson test (DW) of 0.503 is not close to 2, meaning that indeed there might be a correlation between the variables used in the test, whereas the R2 figure of 0.4177 indicates a weak but positive correlation.

A problem of the above test is that it is of limited data as there have only passed eleven years since the BoE's independence, so the data used are not efficient for firm conclusions.

7. Recent Trends and Expectations

Inflation has been rising during the past years, and for this reason the governor of the FED, Ben Bernanchi has suggested to expand the role and responsibilities of the FED so that it has more power over the supervision of commercial banks and other financing institutions. Particularly, the governor argues that Congress shall grant more power and hence responsibilities to the FED in an attempt to reduce the impact of the current financial crisis.

According to an article from The Economist, the FED and the BoE must raise interest rates in order to preserve their credibility due to the rising inflation in conjunction with the crisis in loans from which the U.S. economy is moving into a recession.

Opinions on the issue seem to conflict between the ECB and the FED, as the ECB plans to raise interest rates by early-July whereas the latter has decided not to proceed with any changes in the rates. However, the general public seems to trust and support the independent financial institutions in their attempt to control the inflation shock.

The president of the FED, commenting on the inflationary shock and the financial crisis, admitted that the knowledge of central bankers is still rather vague on how the actions of a central bank can affect and minimize the losses of such inflationary shocks. More specifically, he argued that people are concerned about real wages, hence the expectations of higher inflation may lead central bankers to raise rates even more which in turn can increase costs thus inflation level may increase more than it is now. Unfortunately little is known on the future inflation expectations and how these expectations along with the actions of the central banks can have a direct impact on the price level.

After the inflation rise to 4.0% this June, France is pressing the ECB to think twice before increasing the interest rates to 4.25% to fight inflation. According to French president, Nicholás Sarkozy, increasing the rates is not the solution as the euro is still appreciated against the U.S. dollar and inflation is mostly due to rises in prices of goods and services.

Under the same opinion, the German minister of finance, Peer Steinbrueck, argues that the ECB must consider the impact on economic growth that will result from an increase in interest rates, which is not likely to reduce inflation in the short-term, since most countries within the euro-zone suffer or expect to suffer from a negative growth.

The president of the ECB, Jean-Claude Trichet warns of an inflation burst if central banks do not act immediately to fight inflation. Talking to the German newspaper Die Zeit, Trichet supported that central bankers can control this inflation spiral, and even though economic growth might be negatively impacted by the raised interest rates, the consequences will be far worst if no action is taken at all.

Joaquin Almuña argues that if the governments along with the central banks do not act immediately then there is a danger of simultaneous low growth and inflation in the economies.

8. Conclusions

Taking account of the several studies undertaken regarding the impact of central bank independence on inflation on developed countries and economies in transition one can conclude that central bank independence lowers average inflation indeed. In the study within transition economies earlier what is expected is for CBI to lower both transformed inflation and inflation variability. However, no relationship was found to exist between inflation and legal independence. This may be due to the fact that compliance with law varies among countries; in less developed countries this is lower, which creates bias over its relationship with inflation.

As central banks can create inflationary bias, this can also lead to a rise in inflation rates, for which there is a consensus that a solution to the bias can be achieved by a well-structured central bank. This is one that is independent to choose the instruments to pursue its targets given that it can run monetary policy effectively, hence CBI and inflation are linked after all.

One can expect the reduction in inflation levels as a result of CBI to be a normal consequence, however the problem is that we cannot yet make concrete conclusions as to whether CBI by itself is adequate for creating this inflation reduction as the ceteris paribus assumption of all other things being equal does not hold in the real world. Moreover, the measurements of the degree of CBI used are just proxies to the actual level of independence, as CBI is determined by a country's traditions, legislation and objectives that can cause interference by the government in the formulation of monetary policy. In addition, one should not forget that a central bank can never be absolutely independent as in most countries it is the government that decides on the targets that must be achieved whereas the central bank is only free to choose on the instrument it can use to achieve those targets. Therefore, the impact of CBI on inflation depends on the type of independence a central bank owns.

Moreover, other factors contribute to the inverse relationship between CBI and inflation which have not been examined in detail in this paper, such as the exchange rate, the supply and price of scarce resources like oil, etc. Furthermore, political uncertainty especially prior to electoral periods, and the degree of conservatism of a country's central bank are also crucial on the degree of impact on independence to inflation. A fully independent central bank is not the unique design as it has been proved that central bank independence and conservatism can be substitutes. Also, for any given degree of central bank conservatism, CBI is negatively related to inflation performance.

A pertinent conclusion derived from the tests made and described above is that there exist large discrepancies between legal and actual independence among developed countries. This means that the legal aspects of CBI used in the studies of the relationship between CBI and inflation in developed countries may not produce a clear picture of this relationship.



Casu, B., Girardone, C., Molyneux, P. (2006): Introduction to Banking. Harlow: Prentice Hall.

Cukierman, A., Miller, G.P., Neyapti, B. (2000): Central bank reform, liberalisation and inflation in transition economies. An International Perspective. Mimeo, July

Cukierman, A. (1992): Central Bank Strategy, Credibility, and Independence: Theory and Evidence. 1st ed. London: The MIT Press.

Maxfield, S. (1997): Gatekeepers of growth. Princeton University Press.

Siklos, L.P. (2002): The changing face of Central Banking: Evolutionary Trends since World War II. London: Cambridge University Press.

Conference Papers:

Buiter, H.W. (2006): Rethinking Inflation Targeting and Central bank independence. Old Theatre, London School of Economics.

Epstein, G. (2007): Central Banks, inflation targeting and employment creation. International Labour Organization: Employment Analysis and Research Unit, Economic and Labour Market Analysis Department: Switzerland.

Fraser, B.W. (1994): Central Bank Independence: What does it mean?. Reserve Bank of Australia Bulletin, 20th SEANZA Central Banking Course.

Popov, V. (1994): Central Bank Independence and Inflation in Russia. Paper presented at the conference on the Constitutional Status of Central Banks in Eastern Europe, University of Chicago Law School, April.

Journal Articles:

Aboagye, S.H., Tufte, R.D. (2006): Central Bank Independence, inflation variability and the revenue smoothing hypothesis. International advances in economic research, 12: 147-160.

Alesina, A., Summers, L. (1993): Central bank independence and macroeconomic performance: Some comparative evidence. Journal of Money credit and banking, 25(2): 151-161.

Alesina, A., Gatti, R. (1995): How Independent should the Central Bank be?. American Economic Review, 85: 196-200.

Barro, R.J., Gordon, D.B. (1983): Rules, discretion and reputation in a model of Monetary Policy. Journal of Monetary Economics, 12: 101-120.

Brumm, J.H. (2006): The effect of central bank independence on inflation in developing countries. Economic Letters, 90: 189-193.

Carlstrom T.C., Fuerst, S.T. (2006): Central bank independence: The key to price stability? Federal Reserve Bank of Cleveland, p.1-4.

Chadha, S.J., Macmillan, P., and Nolan, C. (2007): Independence day for the Old lady: A natural experiment of the implications of central bank independence. The Manchester School, 75(3): 311-327.

Crosby, M (1998): Central Bank Independence and Output variability. Economics Letters, 60: 67-75.

Cukierman, A., Lippi, F. (1999): Central Bank Independence, centralization of wage bargaining, inflation and unemployment - theory and some evidence. European Economic Review, 43: 1395-1434.

Cukierman, A. (1996): The economics of central banking. In: Holger, W. (2007): Macroeconomic Policy and financial systems. The Macmillan Press.

Diana, G., Sidiropoulos, M. (2006): Central Bank Independence and the cost of disinflation: Why the wage contract lengths matters? International Advances in economic research, 12: 287-297.

Folder, J. (2005): Why is Central Bank independence so widely spread? Journal of economic issues, 39(4): 843-865

Fuhrer, C.J. (1997): Central bank independence and inflation targeting: Monetary policy paradigms for the next millennium. New England Economic review, p.19-36.

Huang, D.X.K. (2007): Effective central bank communication under uncertainty. Journal of economic issues, 31(3): 661-680.

Levy, D.M. (2006): Sound monetary policy, credibility and economic performance. Cato journal, 26(2): 231-242.

Maliszewski, W.S. (2000): Central bank independence in transition economies. Economies of transition, 8: 749-789.

Preece, H. (2007): Central Banks' Independence under threat: Political pressure mounts against inflation-targeting. Economic Trends and Analysis: FINWEEK, P.48.

Rogoff, K. (1985): The optimal degree of commitment to an intermediate monetary target. Quarterly journal of economics, (100): 1169-1190.

Tager, M. (2007): Central Bank Independence: A research note of the case of the United Kingdom. The social science journal, 44(2): 359-366.

Weymark, N.D. (2006): Inflation, government transfers and optimal central bank independence. European economic review, 51: 297-315.

Wray, L.R. (2007): A post Keynesian view of Central bank independence, policy targets and the rules versus discretion debates. Journal of post Keynesian economics, 30(1): 119-141.

University Work Papers:

Ilieva, J., Gregoriou, A. (2005): Central Bank Independence and Inflation Performance in Transition economies: New evidence from a Primary Data Approach. Manchester Metropolitan Business School Working Paper Series.

Ismiha, M., Ozkan, F.G. (2003): Does central bank independence lower inflation?. York University Paper, Department of Economics.

Maxfield, S. (1995): The politics of Central Banking in Developing countries. Manuscript, Department of Political Science, Yale University, January.


European Commission (1997): Agenda 2000. Supplement 13.

Demertzis, M. (2001): Central bank independence: Low inflation at no cost? A numerical simulations exercise. Econometric research and special studies department: Research Memorandum, No.656.

Electronic Sources:


accessed on 9th April 2008.


accessed on 8th April 2008.

The Monetary Policy Committee. (2008) U.K. inflation report for August 2008, http://www.bankof england.co.uk/publications/inflationreport/ir08/avg.pdf ,

accessed on 5th August 2008.

U.K. National statistics online, Inflation figures. http://www.statistics.gov.uk/instantfigures.aspaccessed on 17th June, 20th July.

Cyprus National Statistics online, http://www.mof.gov.cy/mof/cysat/statistics.nsf/index_gr/index_gr?OpenDocument , accessed on 18th June 2008.

U.K. Inflation History:

http://www.inflationdata.com/inflation/inflation_rate/HistoricalInflation.aspx ,

http://www.moneyextra.com/dictionary/UK-Inflation-history-011485.php ,

accessed on 2-5th August 2008.

Internationl Inflation rates:

http://www.inflationdata.com/inflation/inflation_rate/InternationalSites.com, accessed on 1st August 2008.