Analysis of European Central Bank’s Monetary Policy

Abstract

This dissertation analyses monetary policy of the European Central Bank (ECB) and investigates the effect of monetary policy rule of the European Central Bank on the economic performance of European monetary union (EMU). The stated goal of the ECB price stability given by annual inflation of at or below 2% is an indicator that the bank is following an inflation targeting policy. Given this goal, this study proposes to evaluate the efficiency of the ECB in maintaining an unbiased monetary policy with a target of price stability, and to measure the reaction of the member states’ economies to financial shocks.

To do so, the dissertation will assess the efficiency of the monetary policy instrument used by the European Central Bank and evaluate its national transmission.The emphasis is placed on the analysis of the monetary policy in the Euro area applying Taylor rule. This work contributes to the literature in that it applies the results of other works to the particular and unique circumstances of the euro area. Literature exists on the Taylor Rule for some individual European countries.

However, no studies located that have used the Taylor Rule to empirically test these implications for the efficiency of the European Central Bank using all euro area countries.

Acknowledgements

Both the failures and successes of the European Central Bank will affect not only member countries of the euro area, but also the international economy in general. Although the creation of the European Central Bank and its currency, the euro, has brought larger wealth to the euro area economies, its continuing stability and efficiency will depend on the effectiveness of the European Central Bank in addressing some important obstacles to its success.This paper proposes to evaluate the efficiency of the ECB in maintaining an unbiased monetary policy with a target of price stability, and to measure the reaction of the member states’ economies to financial shocks. To do so, the dissertation will assess the efficiency of the monetary policy instrument used by the European Central Bank and evaluate its national transmission.

Specifically, I will pose and resolve several questions.First, has the implementation of a common monetary policy considerably changed the monetary policy formerly practiced by the member state economies? Next, are the member countries adhering to the policy strategy put forth by the European Central Bank, and are they better off under such conditions? Then, are the benefits and costs of the common monetary evenly distributed to economy of each euro area country? That is, is the European Central Bank biased?The organisation of this paper is as follows. The Chapter 2 provides review the literature on the costs and benefits of a common monetary policy in Europe. Specifically, it provides a brief description of the European Central Bank, its announced objectives and monetary policy strategy and rules.

I will incorporate a discussion of the challenges that the European Central Bank faces due to the diverse structures of the national economies of the euro area, and the potential risks associated with shocks affecting the national economies asymmetrically. Chapter 3 discusses the policy of the ECB and how it goes about achieving its goal of price stability. This will include a discussion of the Taylor Rule, which is a monetary policy tool used by the European Central Bank. It will also include a discussion of the data.

Chapter 4 determines the efficiency of the ECB in achieving its goal by estimating the Taylor Rule and then testing to make certain that the Taylor Rule used by each individual country before the implementation of the euro is structurally different. I will as well extract the target inflation rate for each country. Chapter 4 also provides an estimation of the Taylor Rule for the euro area aggregate. Finally, Chapter 5 offers a summary of all results and findings of the paper.

To perform the above analysis, 12 Euro area economies will be tested. These countries include: Austria, Belgium, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal, and Spain. This work contributes to the literature in that it applies the results of other works to the particular and unique circumstances of the euro area. Literature exists on the Taylor Rule for some individual European countries.

The Treaty of Rome (1957) established European Economic Community. One of its main objectives was the creation of a single market. The Treaty of European Union (Maastricht Treaty), hereafter, the Treaty, signed on 7 February 1992 (entered into force 1 November 1993) amended the Treaty of Rome and specified the design of EMU which meant a transfer of task of conducting monetary and exchange rate policies to the Community level, while leaving fiscal policies, labour market and employment policies, as well as structural and microeconomic policies in the hands of national authorities (ECB, 2003, 1999).The Treaty describes the creation of EMU in three stages.

Stage one (July 1990 – 31 December 1993) was mainly characterised by eliminating all internal barriers to the free movement of capital within European Union (EU). Stage two began on 1 January 1994. It established the European Monetary Institute (EMI), prohibited the financing of the public sector by the central banks, removed the privileged access to financial institutions by the public sector, and required it to avoid excessive government deficits. (ECB, 2003, 1999)Stage three started on 1 January 1999 with the transfer of the monetary policies to the ECB and the introduction of the euro.

Initially 11 countries adopted the euro – Belgium, Germany, Spain, France, Ireland, Italy, Luxembourg, Netherlands, Austria, Portugal, and Finland, which become 12 with the participation of Greece as of 1 January 2001. Currently there are three EU countries – Denmark, Sweden and the United Kingdom that have opted for staying out of the EMU, and 10 new accession countries joined EU at 1 of May, 2004 and two at 1 of January, 2007. (ECB, 2006)EMU is the second largest economic and monetary area in the world – in 2000 total GDP was 15.7% of the world GDP, compared to the U.S. with 21.1%. (ECB, 2004) It is worth mentioning that EMU was drawn up on the premise that eventually all EU member states would adopt the euro.

Thus, an evaluation of current EMU monetary policy is of particular importance for the perspective candidates of the single currency area.2.2 Costs and Benefits of the EuroThe inception of the euro and its complementary common monetary policy is providing the participating countries with tremendous economic benefits, while contributing to regional stability. These countries were forced to comply with convergence criteria such as low inflation and interest rates, and minimal budget deficits, thus transferring the stability culture of some European states to the entire euro area.

Salvatore (2002a) highlights the benefits of the euro and the common monetary policy in Europe. He points out that a common currency implies that prices are all posted in the euro in all euro area countries. This serves to facilitate price comparisons within the euro area countries. Any inflationary tendencies in a country will no longer be able to be masked by currency differentials.

Consumers will thus seek to purchase any bundle of goods where they could be acquired at the lowest cost. This will eventually increase cross-border competition and greater prosperity throughout the area as a whole.Price stability and a common currency also impose greater economic discipline. The Stability and Growth Pact, which accompanied the introduction of the euro in January 1999 created new rules for economic policy.

One of these rules prohibits excessive deficits by the member states. Specifically, they must be less than 3% of national GDP. Economic discipline is particularly relevant for countries such as Greece, Italy, and Portugal that experienced significant inflationary pressures in the past. (Salvatore, 2002a)For example, Portugal, which experienced inflation of over 10% in 1998, succeeded in reducing the rate dramatically to less than 4% by 1996 to be included in the euro area.

Greece, which had an inflation rate of 15% in 1988 managed to reduce it to 5% by 1996. It is doubtful that these countries would have been as successful in reducing inflation had the euro area incentive not existed. Once part of the euro area the strict adherence of the ECB to price stability forces the member states to maintain low inflation rates. (Salvatore, 2002a)Eichengreen (1993) confirms the notion put forth by Salvatore that a common currency facilitates trade and eliminates the transaction costs previously associated with currency conversion.

In 2001, the European Commission proposed a regulation based on the principle that cross-border transactions should not cost any more than domestic transactions. As a result, by July 1, 2003, anyone withdrawing euros from a cash machine anywhere in the euro area will not pay more than they would for a withdrawal in their country of origin. In addition, the cost of a cross-border transfer will be the same as the cost of domestic transfers of the euro by July 2003. (ECB, 2004)The euro and common monetary policy also offer significant benefits at the supranational level.

For example, as the euro is increasingly used as an international currency, the ECB will collect more seigniorage, and as a result be able to increase spending. As outlined above, monetary unification will render Europe more competitive, more prosperous, and consequently more stable. However, the economic stability of Europe is not free from challenges. (ECB, 2004)Salvatore (2002b) points out the main challenge to ECB efficiency.

A common monetary policy can only achieve prosperity and stability for the entire euro area if the policy is more appropriate and responsive than the previous national monetary policy. For the ECB to implement an effective policy that is appropriate for all of the national economies, the business cycles of the countries must be synchronised. If they are not, then any economic shocks to the area, such as the oil crises of the 1970s or even September 11th, will affect the economies asymmetrically as will the reaction of the ECB.Prior to the creation of the common monetary policy, the national central bank would have been able to stabilise the economy through monetary policy or through the exchange rate.

Now, the European Central Bank can only respond to the aggregate effects that the shock may impose on the euro area as a whole. This raises questions about the ECB’s ability to stabilise the national economies in the face of a shock. (Taylor, 1999b)Without control of monetary policy or the exchange rate, fiscal policy and labour mobility would be able to help stabilise a depressed national economy. However the potential for these two tools is limited in Europe.

While the Stability and Growth Pact helps impose beneficial economic discipline on the national economies, it also hinders the reliance on fiscal policy by the member states. For example, the Pact prohibits bailouts of national government deficits, as well as the financing of these deficits. (Salvatore, 2002b)Salvatore (2002b) and Eichengreen (1990) refer to the reduced scope for fiscal policy as a stabilisation tool for the national economies. National economies are typically free to increase public spending or reduce taxes.

However due to factor mobility, the fiscal options that could be employed unilaterally are limited. This is because government spending must be equal to present value of taxes plus present value of seigniorage.With EMU, they cannot control seigniorage. Government borrowing today is limited by the taxes that can be raised tomorrow.

With free movement of labour and capital and goods, if people expect that taxes will be raised, they can leave or capital will flow out of the country. Fiscal policy is usually more politically motivated and thus cannot be relied upon as much as monetary policy to stabilise the economy. Also, one country’s fiscal policy affects the next so fiscal policies are essentially regional and not national in scope. (Salvatore, 2002b)In other common monetary zones, such as the US, if one region is in recession while another is not, flexible wages and the movement of labour away from pockets of unemployment towards where it is needed helps to stabilise both areas.

However, in Europe, the labour market has historically been extremely rigid, further exacerbating the challenge of the ECB. Despite the EU provision for free movement of labour, as well as goods and services, in practice labour generally encounters obstacles when seeking employment in another member state. (Taylor, 1999b)For example, a Greek doctor will encounter language, cultural, and social obstacles to practicing medicine if he or she were to relocate to Germany. Not to mention that European real estate markets are sometimes controlled by the government, which typically seeks to provide housing for its nationals before expatriates, even if they are fellow EU citizens.

Under such circumstances, the Greek doctor will be forced to live under sub-optimal conditions and might be forced to find another, less productive career. (Salvatore, 2002b)Obstfeld (1997) shows how in the presence of labour rigidities, readjustment or stabilisation after an asymmetric shock is more difficult. Under such circumstances, currency alignments are useful tools. The implications are that it could be costly for euro area states to surrender control of their currencies, and that unless social policy is revised in Europe, the euro area states will be vulnerable to asymmetric shocks.

Thus, the efficiency of the ECB is of critical importance if the euro area, including all of its members is to remain a stable and prosperous player in the global economy. If business cycles are synchronised, then the inadequacies of the labour and real estate market will not be as highlighted.Bayoumi and Eichengreen (1993) demonstrated the different effects on prices and output of demand verses supply shocks. Unlike demand shocks, supply shocks result in permanent changes in output.

In addition positive demand shocks raise prices, while positive supply shocks reduce them. Their results lead them to believe that there is a core of countries whose shocks are correlated and a periphery that does not exhibit correlation.De Grauwe (2000) highlights the challenge faced by the ECB when the member countries have either asymmetric national shock transmission mechanisms or face different shocks. Using a two country model, he shows that while euro area policy-makers take decisions on behalf of the euro area as a whole, they often weigh the effects of the policies on their home country more heavily than the others.

The only way that contradiction between minimising the national loss functions and minimising the euro area loss function could be prevented is if the national economies are symmetric.Under such an ideal situation, advancing the good of the euro area is synonymous with advancing the good of each individual economy. To determine how likely it will be for the euro area to realise such an ideal situation, de Grauwe (2000) optimises the Phillips curves for each country and then tests to see if their residuals are correlated. This has implications for synchronisation of the national business cycles.

He then assesses the effectiveness of the ECB in responding to asymmetric shocks under the different degrees of business cycle correlation.He finds that as the degree of asymmetry increases, the effectiveness of stabilisation of output and unemployment is reduced. De Grauwe also articulates that the ECB has declared that in order to find its optimal policy rule, national macroeconomic data should be first aggregated into euro-wide averages. Then this average will be used to determine the overall loss function and hence ECB policy.

However, de Grauwe (2000) discovers that a more efficient method would be for each nation to minimise its respective loss function and then aggregate these loss functions.2.3 Monetary Policy Strategy of the ECBThe objective of European System of Central Banks (ESCB) with ECB in its centre is defined in the Treaty (Article 105) as follows:”The primary objective of the ESCB shall be to maintain price stability. Without prejudice of the objective of price stability, the ESCB shall support the general economic policies in the Community with a view to contributing to the achievement of the objectives of the Community as laid down in Article 2.

The ESCB shall act in accordance with the principle of an open market economy with free competition, favouring an efficient allocation of resources…”.

An excerpt from the treaty establishing the European Community, Article 2:”The Community shall have as its task …to promote throughout the Community a harmonious, balanced and sustainable development of economic activities, a high level of employment.

.. sustainable and non-inflationary growth ..

. convergence of economic performance…

“The Treaty also assigns the ESCB and the Eurosystem a substantial degree of institutional independence, but supplements it with extensive obligations concerning transparency and accountability. In October 1998, the ECB announced its stability-oriented monetary policy strategy for ESCB. It declared that the price stability is ECB’s primary objective. To reach the goal of price stability ECB implements “two-pillar” strategy described in Figure 1.

(ECB, 2001a)Figure 1. ECB Monetary strategy (ECB, 2001a)The first element of the ECB monetary strategy is its main goal – price stability. The Governing Council of ECB has adopted the following definition: “price stability shall be defined as a year-on-year increase in the Harmonised Index of Consumer Prices (HICP) for the euro area of below 2%.” (ECB, 1999) In 2003, the Governing Council further clarified that, within the definition, it aims to maintain inflation rates below but close to 2% over the medium term.

Monetary policy of the ECB is conducted through open market operations, which affect the short-term interest rates in the economy. The main refinancing operations are the most important open market operations and represent the key monetary policy instrument of the Eurosystem. The interest rate applied to these operations is the key policy rate in the Eurosystem. Other interest rates set by the ECB Governing Council are the interest rate on the marginal lending facility and the interest rate on the deposit facility.

(ECB, 2001a)First pillar of ECB monetary strategy – reference value of M3 – is set because it is widely believed that inflation is ultimately a monetary phenomenon. (Surico, 2003) Nevertheless, the ECB emphasise that monetary growth is only a leading indicator of developments in the price level and that the ECB will not target monetary aggregate explicitly . The reference value for M3 growth is based on the quantity equation of money (M*V=YP, where M is the monetary base, V is velocity of circulation of money, and YP is nominal gross domestic product (GDP)) and is set to be equal to M3 growth at 4.5% per annum.

This value based on the following medium-term assumptions: first, price stability should be maintained below 2%, second, the trend real GDP growth lies in the range of 2-2.5% per annum, and third, over medium term, the decline in the velocity of circulation of M3 is in the range of 0.5-1% each year (ECB, 1999, 2001a).Second pillar – analysis of economic and financial indicators – includes indicators other than monetary indicators, which according to the ECB are believed to provide a signal about future price developments.

These variables include wages, exchange rate, bond prices and the yield curve, various measures of economic activity, fiscal policy indicators, price and cost indices and business and consumer surveys, as well as inflation forecasts produced by the ECB and other institutions and market participants. (ECB, 1999)Given the wide range of models that permeate the literature, the two pillars are meant to capture the uncertainty related to the true model of the economy. The first pillar can be seen as representing a group of models which embody a view of price level determination that assigns an important role to money and the transmission channels of monetary policy related to money. The second pillar is meant to capture a range of alternative models of the inflation process that emphasise the interaction of supply and demand, cost pressures and the transmission channels that operate through these variables.

(ECB, 1999)It is clear that the ECB monetary policy strategy is not based on a single monetary policy rule – it is much more involved and complex. Nevertheless, a simple policy rule, like the Taylor rule, may provide insight into those variables that are important in the policy decision-making process. In the context of the Taylor’s rule one can think of the output gap or real marginal costs capturing the second pillar variables’ properties and inflation – first pillar’s properties. (Surico, 2003)

Monetary Policy RulesThe recent literature about monetary policy rules revived an issue that dates back to the 1960’s. Milton Friedman (1968) was one of the first to talk about the need of the monetary authority to focus on a monetary policy rule. Friedman outlined two requirements for the monetary authority, first “…the monetary authority should guide itself by magnitudes that it can control, not by ones that it cannot control,” and second “…monetary authority should avoid sharp swings in policy.” (1968, p. 14, 15). He also stated that: “By setting itself to a steady course and keeping to it, the monetary authority could make a major contribution to promoting economic stability,” (1968, p.17). Friedman’s proposed a “fixed” rule – a growth of monetary aggregate at a level of 3 to 5 percent.Since Friedman’s proposal great deal of research addressed the issue resulting in agreement and disagreement over the use of rules (fixed versus feedback) versus discrete economic policy – monetary and/or fiscal. Under the rule monetary policy is assumed to follow credible commitment to a specified path of the instrument variable obtained from a dynamic optimisation problem, whereas, under discretion, the monetary authority chooses the path of the instrument re-optimising every period.

In the late 1980’s and in the 1990’s the different thesis in the literature seems to converge to the conclusion that rules are preferred to discretionary behaviour, or, in other words, monetary policy with commitment will yield better results than monetary policy without commitment (Taylor, 1993; Clarida et al., 1998)It is important to note that a policy rule is not a mechanical formula with specified coefficients that needs to be followed – a statement usually used by opponents of simple policy rules (ECB, 2001a). Taylor defines a policy rule as a “systematic policy” or “policy system” meaning, it is a methodical, according to plan, and not casually or at random conducted policy; or “a policy rule is a contingency plan that lasts forever unless there is an explicit cancellation clause,” (1993, p. 199).

One has to keep in mind this broad definition of the policy rule, which provides a framework for the conduct of monetary policy and sets guidelines derived from the properties of the specific rule.McCallum (1999) uses the word “goal” to refer to the ultimate but usually non-operational objective of the monetary authority. As to the term “target” he refers to it as an operational variable that is actually used in the conduct of the policy.There is a fairly close agreement among economists that in the long run monetary policy can affect only nominal variables.

It is believed, however, that although not being able to affect levels of real variables (output or employment), monetary policy can affect the variability of the variables. A common approach in the recent literature is to characterise monetary authority goal as a maximisation of welfare function of the society, which is equivalent to minimising a loss function of the society. The loss function is a function of target variables of a central bank, defined as a deviation of the particular variable from its target value.The focus of the literature centres on three possible target variables – inflation, output gap and interest rate smoothing.

Monetary policies are defined in terms of targeting these variables – SIT, FIT, and SOT with interest rate smoothing capturing financial stability targeting (Mishkin, 1999). The consensus that has emerged from the research is that inflation targeting is desirable, however, opinions on whether output or interest rate smoothing or both should also be targets of monetary policy differ (Taylor, 1999a; Svensson, 2002b).Revival of the interest in rule rather than discretion is attributed to the work of Taylor. Taylor’s 1993 article caused a revival of the discussion of monetary policy targeting and instruments to be used as it introduced a new approach to monetary policy targeting – an interest rate as an instrument for policy targeting with the target variables – inflation rate and Output gap stability.

Taylor finds that it is preferable for the central banks to set interest rates based on economic conditions of their own economies paying little attention to the exchange rates.As a result, he argues, placing some positive weight on both price level and real output in the interest rate rule is preferable. He also admits that the relative weights for price and output are not clear, but some weights on output in the policy rule is likely to yield better outcome than a pure pricing rule. In his 1993 paper Taylor recommended an interest rate policy rule of the form of:rt = pt + 0.

5 yt+0.5 (pt – 2) + 2 (1)where rt is the federal funds rate, pt is the rate of inflation over the previous four quarters and yt is the percent deviation of real GDP from a target level, given by 100 (Y-Y*)/Y*, where Y is real GDP and Y* is trend real GDP, and 2 is long-run equilibrium real interest rate (1999, p.202). This rule according to Taylor describes Federal Reserve monetary policy for the U.S. quite well during the 1987-1992 period.Taylor proposes interest rate as a monetary policy instrument while other instruments are advocated by McCallum (1988) and McCallum and Nelson (1999). These authors propose monetary base as an instrument and nominal income as a target (NIT).

For open economies Ball (1999) advocates as an instrument monetary condition index (MCI) based on both the interest rate and the exchange rate. This view is not shared by others who argue that the interest rate is preferred to a monetary base instrument, since monetary aggregates are not good policy instruments due to the implied interest rate volatility (Clarida et al, 1999; Bemanke and Woodford, 1997). Nominal income targeting has received less attention in the literature for several reasons: first, if there is a shift in the trend growth of nominal GDP, the rule does not provide a precise nominal anchor.Second, if an optimal policy adjusts interest rate to some linear combination of disturbances to real output and inflation, by targeting nominal GDP it arbitrarily assigns equal weights to both inflation and output gap, whereas it might not be desirable.

And third, high nominal GDP growth might occur when the economy is still recovering from recession and is still below its optimal output. Adjustment of the instrument in this case might restrain the recovery (Clarida et al, 1999; Rotemberg and Woodford, 1999; Ball, 1997).Given that a substantial body of the literature prefers interest rate over other policy instruments and that the current practice of several central banks is to use interest rate as their monetary policy instrument, a general class of simple Taylor-type interest rate policy rules that appear in the literature can be summarised with the following equation:it = a?t+byt +cit-1 (2)where it is the nominal interest rate, ?t is the inflation rate, and yt is real GDP measured as a deviation from potential GDP (intercept term for the purpose of presentation is ignored here). Policy rules considered in the literature differ with respect to the values of a, b and c, where the original Taylor rule is with coefficient c equal to zero, a is 1.5 and b is 0.5. There is no consensus in the literature on the magnitude of the parameter on lagged interest rate and whether it should be included at all.With respect to optimality Levin, Wieland and Williams (1999) find that additional term of lagged interest rate leads to an improvement in terms of reduced variability of output, whereas Rudenbusch and Svensson (1999) obtain a dynamically unstable system if lagged interest rate is included, although they acknowledge that the result could be sensitive to the specification of expectation formation mechanism – non-rational (backward-looking) within their model.

Taylor (1999c) finds that lagged interest rate does not necessarily improve performance of the rule, it might even increase the interest rate volatility, as well as create some instability of the system in terms of infinite variances of the variables.Besides simple policy rules in the form of (2) a growing number of papers (McCallum and Nelson, 1999; Rotemberg and Woodford, 1999) derive policy rules based on representative agent framework and thus have explicit microeconomic foundations to address the question of expectation formation (forward-looking versus backward-looking or contemporaneous).A good monetary policy rule should be both operational and robust. Operationality refers to a policy rule expressed in terms of instrument variable that can be controlled by central banks and which requires information that could be actually possessed by central banks.

Robustness of a policy rule refers to its ability to produce at least moderately good performance in a variety of macroeconomic models and not just within a single model (McCallum and Nelson, 1999).Taylor-type rules can be easily adjusted to account for operationality aspect using lagged values of inflation and output gap (backward-looking Taylor rule). However, there is no clear evidence of the superiority of the backward looking rule versus the contemporaneous one in terms of optimality (McCallum and Nelson, 1999; Rotemberg and Woodford, 1999). One of the criticism levied against the formulation of the simple Taylor type rule involves the unobservable potential output and different measures used in estimating it, as well as the unobservable real long-run equilibrium interest rate (ECB, 2001a).

Gali and Gertler (1999) propose real marginal costs as an alternative measure of the output gap and the driving force of inflation. They derive a New Keynesian Phillips curve (NPC) within a monopolistic competition market structure and individual firms profit maximisation problem. The model has micro foundations and its structural characteristics are very appealing, it also incorporates rational expectations of the economic agents.These critiques now withstanding several articles in the literature (Rotemberg and Woodford, 1999; Levin, Wieland and Williams, 1999; Rudebusch and Svensson, 1999) show that Taylor rules in most cases adjusted for lagged interest rate perform well within a wide range of the economic models when tested for robustness.

In addition, the studies point out that Taylor rule appears to provide a good description of actual monetary policy and can provide important analytical insights when evaluating monetary policy (Taylor, 1993 and 1999b, Clarida et al, 1998).2.4.1 Monetary Policy TargetingRecent empirical evidence shows an adoption of the interest rate as a policy instrument by several central banks, for example, European Central Bank (ECB), Royal Bank of New Zealand, and Bank of Canada.

This development has somewhat pushed research towards further analysis of the goals of the monetary policy, the optimality of the policy rules, and, while seemingly in agreement, although not complete, on interest rate as a best instrument for achieving the goals of the monetary policy. There is also a common agreement in the literature that in the long run the primary goal of monetary policy should be low and stable inflation, and that monetary policy can affect real variables like output and employment (if it can) only in the short run.Within this context a common framework to describe the goal of a central bank is to assume that it aims to minimise a loss function which is increasing in the deviation of a target variable (inflation, output, interest rate) from the target level of this variable. In the literature the objective function of the central bank is described by an intertemporal quadratic loss function (Rudebusch and Svensson, 1999; Svensson, 2002a), with targeted variables expressed as deviation from the target levels.

For the optimisation problem assuming discount factor equal to one, the intertemporal loss function can be interpreted as the unconditional mean of the period’s loss function which equals the weighted sum of the unconditional variances of the targeted variables.Svensson (2002b) identifies three types of policy targeting: strict inflation targeting (SIT) – focus is entirely on controlling inflation at the shortest possible horizon, the second is flexible inflation targeting (FIT), where the focus is on price stability, but with some weight given to stabilising the output gap, and third is strict output gap targeting. The trade-off between inflation variability and output-gap variability for the three types of monetary policy mentioned above is represented with the so-called Taylor’s curve (Svensson, 2002a, p.270).

However, the addition of an interest rate smoothing as additional target of the central banks is advocated in the recent literature, one reason being an increased financial stability. Another reason widely quoted in the literature is the empirical evidence of the data about actual central banks behaviour in setting interest rates (Levin, Wieland and Williams, 1999). Thus far there is no common agreement in the literature on which target variable(s) central bank should focus on.To complete the optimisation problem one has to define the macroeconomic framework of the economy by specifying aggregate demand (IS-type) and aggregate supply (Phillips curve) relationships.

Since within the given framework central bank will not follow a monetary base targeting policy, it is not necessary to specify money market equilibrium condition.Optimal linear monetary policy feedback rule can be derived then by minimising loss function of the economy subject to the current state of the economy given by aggregate demand and supply equations. This rule, however, will be a function of the underlying structural parameters of the economy (Clarida et al, 1999). Its application requires a detailed knowledge of the structure of the economy and specific values of its parameters.

Recent work on monetary policy rules involves evaluation and analysis of different policy rules using the optimisation framework described earlier. A significant part of this research is devoted to the specification of the underlying model. Clarida et al. (1999), Rudebusch and Svensson (1999), Rotemberg and Woodford (1999), Woodford (2001) have shown that simple monetary policy rules such as the one proposed by Taylor (1993) (with some modifications) is compatible with several features of optimal monetary policy.

Inflation Targeting

In the literature the “price stability” concept usually is associated with “inflation targeting” not “price-level targeting” and there is consensus that price-level targeting is not desirable (Clarida et al, 1999). The announcement of the ECB also means that it follows inflation targeting – the target is price inflation, measured by annual changes in HICP, below 2% to maintain its goal of price stability.

The reason for treating inflation targeting as the primary goal of monetary policy is strongest when monetary authority cares about medium and long-term horizons, as there is agreement between economists that in the long-term money is neutral, i.e. the monetary policy can affect real variables like output and employment only in the short run. Nevertheless, the short run effects of the monetary policy involve stabilisation of the output variability, in particular, around its potential level.

Main advantages of inflation targeting are spelled out in studies by ECB (2001a), Svensson (2002b), Bemanke and Woodford (1997). These are summarised below:- Inflation targeting provides a nominal anchor in the economy for the nominal variables and private sector expectations;- High and variable inflation distorts market mechanisms to achieve efficient resource allocation via information of relative prices;- Price stability reduces unnecessary hedging activities in asset and investment decisions;- Price stability prevents an arbitrary and inequitable redistribution of incomes and assets;- Inflation targeting improves transparency of monetary policy – a better communication of policymakers’ objectives and intentions to the public and financial markets;- It is believed that the public better understands the inflation concept than when the targeted variable is nominal GDP or the monetary base.The main disadvantage of inflation targeting follows from the fact that the inflation responds to changes in monetary policy only with a substantial lag. As outlined by Bemanke and Woodford (1997) this lag implies two related problems: first, the information the central bank requires in order to implement inflation targeting may be much greater than that needed to target intermediate variable, whose response to policy changes can be observed with less delay, and second, it is difficult for the public and financial markets to judge if the central bank is on target – it creates adverse effects on the bank’s credibility and accountability.

The ECB monetary policy outlined earlier corresponds to the definition of a strict inflation targeting (SIT) rather than flexible targeting (FIT). Nevertheless, the ECB may experience pressure from the outside to stabilise output in the short run. As will be discussed in later in the paper it is possible that the output gap is also a target (indirectly) of the ECB monetary policy, an issue to be addressed in this study.

Monetary Policy Transmission Mechanism

Monetary policy transmission mechanism relates to the number of channels through which monetary policy affects real and nominal variables. As it was stated earlier, the complete effect of monetary policy on the economy takes time and this lag substantially complicates the monetary policy decision-making. When making a monetary policy decision it is very difficult to assess the extent previous monetary policy actions have already passed through the economy. Kuttner and Mosser (2002) provide overview of the various channels through which monetary transmission mechanism operates.

The main channels through which monetary policy is transmitted through the economy are described below. The first is so-called direct exchange rate channel to inflation. When a central bank lowers interest rate it leads to depreciation of the currency due to capital outflow. Depreciation of the currency makes imported goods more expensive and since imported goods are part of the consumer price basket that makes up price level index (HICP for the ECB defined inflation measure), it directly causes HICP to increase.

There is also an indirect exchange rate effect on aggregate demand via increased competitiveness in international markets through lower prices of inputs (depreciation of the currency) and that raises the demand for exports.However, the importance of exchange rate effects depend on countries openness to international trade and the ECB claims that the exchange rate channel of monetary policy transmission is less important for a large and relatively closed currency area like the EMU. The second is so-called indirect or real-interest rate channel to aggregate demand. Lower short and long-term real interest rates (long-term real interest rate is affected through the market expectations) stimulate investment and consumption, and thus there is an increase in aggregate demand and output in the economy.

The third and a very important channel is the effect on inflation expectations within the economy.Monetary policy affects the wage and price setting in the economy. If monetary policy aimed at price stability is credible, inflation expectations will remain anchored at the low level of announced inflation objective. If credibility is lost, inflation expectation will lead to price and wage setting that accommodates increased expectations and that will lead to higher inflation.

ECB Policy

The European Central Bank has clearly, comprehensively, and transparently outlined its mission as the monetary authority for the euro area. It has highlighted medium term price stability as its primary objective, and thus monetary policy will be used as a means to this end. Specifically, the ECB has committed the euro area to a less than 2% year-on-year increase.

By maintaining price stability in the medium term, the ECB acknowledges that non-monetary shocks such as tax changes or variations in international commodity prices could cause short-term price volatility that cannot be controlled by monetary policy.By not responding to such temporary fluctuations and continuing to focus on the medium-term, the ECB will avoid creating uncertainty in the economy, thereby maintaining credibility, which is a critical component for a successful monetary policy. Since inflation itself is very difficult to control, the Central Bank must have tools that could be used to address the causes of inflation. The ECB has thus created a role for money.

Thus, the ECB can respond to inflationary pressures by increasing or decreasing the rate of money growth, which will impact inflation. Alternatively, the interest rate could be a useful tool for monetary policy. Historically, many of the euro area countries, as well as others, have relied on the interest rate as the monetary policy tool (Clarida et al., 1998).

Thus, I will test to see if the interest rate rule known as the Taylor Rule is being applied by the ECB.

The Model

Taylor (1993) proposed a simple interest rate rule that is currently known as the Taylor Rule.

This rule implies that the interest rate is a function of its lag, inflation and output as such:it = ?1 it-1 + (1 – ?1) i*t + ?t (3)and, i*t = is +?? ?t+1 + ?? yt (4)Where it is the current nominal interest rate, is is the long-run equilibrium nominal rate of interest, ?t+1 is the deviation of next period’s inflation from its target rate, and yt is the deviation of output from its potential. ?? and ?? represent the weights placed on maintaining the equilibrium value of ?t+1 and yt respectively. The monetary authority manipulates these weights or coefficients in such a way so as to affect the interest rate according to the plan.For example, if the value of ?? is greater than 1, and if inflation rises by 1%, the nominal interest rate will rise by more than 1%, causing the real interest rate to also increase.

Alternatively, if the value of ?? is less than 1%, and if inflation rises by 1%, the nominal interest will rise by less than inflation, which will cause the real interest rate to decrease. Thus, the Central Bank could engineer either a contraction or an expansion by adjusting the weights on inflation. A similar argument can be made for the weight on output. The monetary authorities also seek to smooth interest rates.

Thus, the actual interest rate is a function of the desired target rate as well as lags of the actual interest rate. This smoothing behaviour is captured the parameter ?. For example, if ? is 0.5, then the monetary authorities will move the actual rate one-half towards the rate suggested by the Taylor Rule, while the other 0.5 will be on maintaining the rate from the previous period or periods.Taylor (1998) recommends raising the interest rate by 1.5 percentage points for each 1 percentage point increase in inflation. This will result in a rise in real interest rates that would help slowdown the economy, thereby reducing inflation.

He also recommends reducing the interest rate by 0.5 percentage points for each percentage point decline in real GDP below its natural rate. This would serve to ease the recession and maintain price stability.Taylor (1993) showed that during the 15 years prior, the Taylor Rule correctly captured the behaviour of the United States Federal Reserve Bank in setting interest rates.

Clarida et al. (1998) also confirm that the US follows a Taylor Rule, but they added that Japan, England, France, Germany, and Italy also follow some form of the Taylor Rule in setting interest rates.

The Data

To estimate the Taylor Rule for all euro area countries, the equivalent of the United States federal funds rate is used for each respective country.

However, in some cases, where data availability did not permit use of the federal funds rate, another measure of the interest rate is substituted, provided that it exhibits a high degree of correlation with the federal funds rate. All data was taken from the International Financial Statistics of the International Monetary Fund.The deviation of output from its potential is represented by the “output gap.” This is calculated by regressing industrial production on a quadratic time trend.

The residuals are then used as the gap. The source for this data is the International Financial Statistics of the International Monetary Fund (http://www.imfstatistics.org).

Inflation is the percent change in consumer prices per year, where the base year is 1995. The source for this data is the International Financial Statistics of the International Monetary Fund (http://www.imfstatistics.org).

Finally, the 3 Spot Price Index is included to capture any forward-looking expectations of the monetary authorities. For example, the central bank might foresee a rise in inflation and thus adjust the interest in response, prior to the actual rise.For all variables and for most countries data is collected from 1974Q1 to 2005 Q2, based on availability of the data. Since a concern of this paper is to comment on the effectiveness of the ECB’s monetary policy as compared to the previous national policies, it is necessary to impose a structural break in the data.

The definitive movement towards a common monetary policy in Europe was implemented in the Maastricht Treaty of 1991.After this point the countries began to prepare their economies for the union in three stages, culminating with complete unification of policy and currency. As a result, it can be assumed that this point represents a departure from the national policies previously practiced. However, since most European countries experienced a severe currency crisis shortly thereafter, the post-Maastricht period is considered to begin instead with 1993

However, in the absence of the dummy variable, the coefficients all remain negative. Only a few cases (those marked in bold and italics) exhibited insignificance, however the coefficients are still negative.

Lag Selection

The smoothing component of the Taylor Rule implies that past periods interest rates are considered by the monetary authorities when setting the actual rate. To determine the most efficient amount of lags of the interest to incorporate into the model, the Akaike Information Criteria is compared for the inclusion of lags one through three for both the pre-Maastricht and post-Maastricht periods for all twelve countries. The results are summarised in the chart below, where the smallest AIC values are marked in bold and italics. These values represent the quantity of lags that is optimal to include in the model.

A change in the estimates and their corresponding levels of significance from the pre-Maastricht period to the post-Maastricht period, indicates that the ECB policy is in fact different from the national policies practiced prior to the implementation of the common monetary policy. Formal tests for structural change confirm or disprove any changes in the parameter estimates.To estimate the parameters of the model, I employ iterated generalised method of moments (ITGMM). In doing so, I can account for the simultaneous equations bias inherent in the model.

In addition, I will be able to extract the target inflation. The instruments used vary slightly by country depending on how many lags are included in each country’s model. In general the instruments include, at least two periods of lagged inflation, industrial production, and the interest rate, and the first difference of the spot commodity price index. The model to be estimated is derived as follows.

This creates the moment conditions required for GMM.The weight matrix used is the inverse of the information matrix. This matrix is the most efficient in that it gives the greatest weight to the moments with the lowest variance.Before reporting the estimates, some proof that the model is well specified for each country is useful.

Source: (http://www.imfstatistics.org).The parameter estimates reveal that, in some cases, the inflation-targeting ECB is in fact meeting its primary objective of minimising inflation in the national economies. Some results also indicate that the countries that were targeting industrial production prior to the common monetary policy are no longer doing so.

This latter result is also in accordance with the ECB policy of not targeting output. Out of the twelve countries tested, nine demonstrated a significant inflation parameter estimate (with or without the optimal value of the estimate), the elimination of the significance of the output gap estimate in the post-period, or both.4.2 Structural Change in the Taylor RuleA dummy variable test is performed in order to determine if the Taylor Rule practiced by the ECB is statistically different from the Taylor Rule practiced by the national monetary authorities prior to Maastricht.

Lagged interest 10.73 0.0011The above results of the test for variable structural change illustrate that for Austria, France, and Finland, both inflation and output experienced a break.

For France, Germany, and Spain, only output structurally changed, while for Italy and Portugal only inflation changed. This test supports neither a structural change in inflation nor output for Belgium, Greece, Luxembourg, and Ireland.4.3 Overall AnalysisIn terms of significance of the estimates, the above tests reveal that in nine out of twelve cases, inflation is a significant contributor to interest rate adjustments in the post-Maastricht period.

Regarding the economic theory that motivates the Taylor Rule, not only is the significance of the estimate relevant, but also its value. As noted above, in The ECB and the Taylor Rule, Taylor recommends that the interest rate be raised by 1.5 percentage points for each 1 percentage point increase in inflation. This will result in a rise in real interest rates that would help slowdown the economy, thereby reducing inflation.

He also recommends reducing the interest rate by 0.5 percentage points for each percentage point decline in real GDP below its natural rate. This would serve to ease the recession and maintain price stability. However, this recipe must be refined for a European Central Bank that does not target output.

The values of the parameter estimates for six countries support the recommendation of the Taylor Rule to raise the interest rate by more than 1 in the face of a one unit inflation shock. Three countries, Finland, France, and Italy, are targeting inflation more aggressively than the Taylor Rule recommends. Ten out of the twelve countries tested positive for structural change in the Taylor Rule around the time that the common monetary policy regulations were being implemented. All countries for both the pre and post period seek to smooth interest rates.

This is evidenced by the significance of the lagged interest rate coefficient in all tests. However, for inflation and industrial production, the results vary.The estimates reveal that Austria had been targeting both inflation and industrial production in the pre period. However, in the post period it is only targeting inflation.

The model test for structural change reflects a break around the time of the common monetary policy. The variable tests for structural change indicate a change in inflation, as well as industrial production.For Austria, the estimate for inflation increased from 0.60401 to 1.

109556 (both periods significant). The estimate for industrial production was significant at 0.241709 in the pre period and then became insignificant.The estimates indicate that Belgium targeted neither inflation nor industrial production in the pre period, while in the post period, inflation became significant with an estimate of 1.

The model test for structural change reflects a break in the model, but this break is not confirmed by the variable tests for change.For pre Maastricht Finland, according to the parameter estimates, neither inflation nor industrial production played a role in interest rate changes. However, in the post period Finland is targeting inflation, with an estimate of 2.

The model test for structural change confirms the break. The variable tests for change reflect a break in inflation, as well as a break in industrial production at the 10% level.For pre Maastricht, France, industrial production was of relevance in setting the actual interest rate.

In the post period, industrial production becomes irrelevant while inflation becomes relevant with an estimate of 2.22. This change in inflation and industrial production is confirmed by the variable tests for structural change, as well as by the model test for structural change.German parameter estimates depict that Germany considered industrial production (with an estimate of 0.

814212) when determining the actual interest rate in the pre period, while in the post period this emphasis shifted to inflation but with an estimate of only 0.630451. According to these results Germany raises the interest rate by only 0.630451 in response to an inflation shock, which implies that the real rate is actually falling.

Although inflation has become significant, the German response is not aggressive enough.However, when the second and third lags of the interest rate are removed from the model, the estimate for inflation becomes 1.24, which is more in accordance with both the Taylor Rule and Germany’s typically aggressive stance against inflation. The test for model structural change is positive.

The variable tests reveal a break in industrial production but not inflation.For Greece, both inflation and industrial production became significant in the post period compared to the pre period. However, the value of the estimate for inflation (0.16) is not even close what Taylor recommends.

Both the model tests and variable tests for structural change do not support a break.Italy’s estimates indicate that in the pre period, neither inflation nor industrial production played any role in interest rate changes. However post Maastricht inflation, with an estimate of 2.56 not only became significant but it also surpasses the recommendation of the Taylor Rule theory.

The model test for a structural break reflects a change in the model, and according to the variable tests for structural change, this break is attributable to inflation.Luxembourg showed a shift in the emphasis on inflation. In the pre period, only the lagged interest rate mattered, while in the post period, inflation became significant. However, the value of the estimate for inflation (0.

427) is not at par with what Taylor recommends in order to prevent a falling real interest rate. The model test for structural change reflects a break in the model, but this break is not confirmed by the variable tests for change.For the Netherlands, industrial production was significant in the pre period and remained significant in the post period. Inflation became significant from the pre to the post period, although the estimate of 0.

426 is not aggressive enough. The estimate for industrial production is 0.559878, which corresponds with the theory of the Taylor Rule but not with the policy of the ECB. The model test for a break reveals a change, but the variable tests show a change only in industrial production.

Portugal, Spain, and Ireland did not produce any significant results in either period. However, for all three countries the model tests for structural change indicate a break. For the variable tests for change, Portugal shows a break in inflation, Spain shows a change in industrial production, and Ireland does not show a change in either.4.

Target InflationBy reorganising the model, the target inflation rate can be extracted for each country for both the pre and post period.The equation being estimated is,it = is + (1 – ?1) ?? ?t+1 + (1 – ?1) ?? yt + ?1 it-1 + et+1 (17)where is, the long-run equilibrium nominal interest rate is defined as (1 – ?1) (r – ) and r is the average real interest rate calculated from the data (the nominal interest rate minus inflation).And,(1 – ?1) ?? ?t+1 = (1 – ?1) ?? ?t+1 – (1 – ?1) ?? (18)(17) becomes,it = (1 – ?1)(r + ) +(1 – ?1) ?? + (1 – ?1)???t+1 +(1 – ?1)?? yt + ?1 it-1 + et+1Defining(1 – ?1)(r + ) + (1 – ?1) ?? C (19)Where C is the constant given by the SAS output.

Spain 0.161679 2.086274The above results reveal that only France has an inflation target near 2%, which is the goal of the ECB. The other countries’ targets are either far above or far below the 2% target.

These results could be tainted for two reasons. First, the central banks might have been acting to minimise a loss function that seeks to maintain output above its natural rate. If so, they would be operating on an alternative Taylor Rule in the form of:i*t = is +?? ?t+1 + ?? (yt –{yN + k}) (21)Where yN is the natural rate of output and k is a constant representing the amount by which output exceeds the natural rate. Such an augmented Taylor Rule would cause the constant to be a different value.

This would alter the value of the constant produced by the model. Second, the estimate of the average real rate might not be completely accurate due to the inconsistent behaviour of the real rate during the period of estimation (1973-present).In 1996, Rene Garcia and Pierre Peron showed that the United States real interest rate generally hovered just below 0 for most of the 1970s, jumped to a relatively higher rate in the 1980s before dropping again to a somewhat lower rate in the 1990s. A Markov Switching Model best captures this non-stationary behaviour.

Graphs of the real interest rate for each euro area country (http://www.imfstatistics.org) reveal that the real interest rate in Europe may have followed a similar pattern as that of the US. The general pattern, which exhibits three different trends may not be captured be the conventional method of calculating the real interest rate.

Using GMM to extract the model parameters for the ECB Taylor Rule, the results are the following.Table 10. Estimation of the Taylor Rule for the Euro AreaEstimate T-Stat P-ValueConstant 0.007599 1.

This implies that the ECB is not biased when the actual interest rate is used instead of the rate predicted by the Taylor Rule

Conclusion

The above analysis reveals that the European Central Bank follows a Taylor Rule when setting interest rates. The Taylor Rule allows for interest rates to be adjusted for both inflation and output, however the ECB explicitly states that it will adjust interest rates only in response to inflation. This implies that the coefficient on the output gap variable in the Taylor Rule for the ECB should not be significant, while the coefficient on inflation should be significant and valued over one.

The results outlined in Chapter 4 for the post Maastricht period overwhelmingly show that the Taylor Rule for the euro area economies emphasises inflation while practically eliminating the role for output in setting interest rates.A comparison of the post period results to the pre period results reveals that in the pre period, many of the countries were following a Taylor Rule that either emphasised output more than inflation, or maintained a weight on inflation that was less than one. The latter rule would produce the undesirable effect of a decline in the real interest rate when a rise in the rate is the appropriate response to an inflation shock. Together the pre and post estimation of the Taylor Rule reveal that the ECB has been efficient in curbing inflation in countries that typically experienced high inflation in the past.

The tests revealed that a Taylor Rule accurately captured the behaviour of the monetary authorities for both the pre and post period (despite the inefficiency of the pre period), however the coefficients on the variables varied drastically across the two periods. Most of the tests for structural change support the notion that the Taylor Rule followed in the pre period is not the same Taylor Rule followed in the post period.The estimation of the Taylor Rule revealed that the ECB is targeting inflation according to plan. When the target inflation was calculated from the model, the rate of 1.

79% demonstrates the ECB’s commitment to its policy objective of maintaining inflation at or below 2%. In addition, the estimation of the Taylor Rule for the euro area as a whole yielded coefficients of 1.23 and 0.02 on inflation and output respectively.

Thus, the goals of the national economies (as revealed by the national Taylor Rule estimations) and the goals of the aggregate coincide.The above conclusions paint a positive picture for the euro area and the success of the ECB. However, the impulse response functions of Chapter 4 imply that this success might not be sustainable. Particularly the output gap functions highlight that shocks are not necessarily transmitted from one euro area country to another, with the exception of France and Germany which demonstrated a reciprocal relationship.

As such, a significant degree of synchronisation of the euro area business cycles is not obvious.The implication is that in the absence of business cycle correlations, the national economies are susceptible to asymmetric shocks to which the ECB might not provide the most appropriate or adequately aggressive response. If one country experienced a positive shock while another was hit by a negative shock, the effects of the shocks on national inflation and output could neutralise each other, thereby maintaining the status quo for the aggregate of the euro area. Under such circumstances, the ECB would not respond, and thus not accommodate the needs of the individual economies.

In addition, since output is not targeted by the ECB, even if an output gap shock to one country is permanently transmitted to another, one cannot expect that the ECB would respond to bring the spiralling output back under control.Likewise, the inflation impulse response functions do not necessarily reveal that shocks are readily transmitted anymore in the post Maastricht period than in the pre period. Although no correlation conclusions can be drawn from these functions, they do overwhelmingly suggest that if inflation shocks are transmitted, they are immediately corrected for more so in the post period. This further highlights the ECB’s commitment to maintaining its policy goal of price stability.

Overall, this paper suggests that the ECB fulfils its obligations and commitments by adhering to a Taylor Rule, which has been proven to be more efficient than the Taylor Rules followed by the national economies prior to the implementation of the common monetary policy. Finally, it is worth noting that although the ECB claims to be very transparent in its policy implementation, the communication to the public could be improved if the ECB were to announce the relative significance it attaches to each of the two pillars of policy and which factors it considers to be most important in formulating its policy.Future research may evolve in two directions. One is to apply methodology of this study to current member of the EU but not EMU (the United Kingdom, Sweden and Denmark).

Also, in the light of expanded EU (12 new countries) as of January 1, 2007, analyse their current monetary policies and relate the analysis to the findings reported in this study.

References

1. Ball L. (1999) “Policy Rules for Open Economies,” Chapter No.3 in “Monetary Policy Rules,” edit.
2. Taylor J.B., NBER Conference Report, The University of Chicago Press, pp. 127-144.
3. Bayoumi T. and Eichengreen B. (1993) “Shocking Aspects of European Monetary Integration.” Adjustment and Growth in The European Monetary Union.
4. Torres and Giavazzi. Cambridge University Press.Bemanke B.S.
5. , Woodford M. (1997) “Inflation Forecasts and Monetary Policy,” Journal of Money, Credit and Banking, Vol.29 (4), pp.653-684.
6. Clarida R., Gali J., Gertler M. (1998) “Monetary Policy Rules in Practice: Some International Evidence,” European Economic Review, Vol.
7. 42, pp. 1033-1067.Clarida R., Gali J.
8. , Gertler M. (1999) “Science of Monetary Policy: A New Keynesian Perspective,” Journal of Economic Literature, Vol.37 (4), pp. 1661-1707.
9. Eichengreen B. (1993) “European Monetary Unification” Journal of Economic Literature, SeptEuropean Central Bank (2001a) “The Monetary Policy of the ECB”, European Central Bank.European Central Bank (2006) “Euro Area Statistics” Monthly Bulletin, October, pp.67-87.
10. European Central Bank (2003a) “Issues Related to Monetary Policy rules,” Monthly Bulletin, October, pp.37-50.European Central Bank (2003b) “Statistics Pocket Book,” ECB, June.European Central Bank (2004) “The Monetary Policy of the ECB”, European Central Bank.
11. Friedman M. (1968) “The Role of Monetary Policy,” The American Economic Review, Vol.58 (1), pp.1-17.
12. Gali J. Gertler M., Lopez-Salido J.D.
13. (2001) “European Inflation Dynamics,” European Economic Review, Vol.45, pp. 1237-1270.De Grauwe P.
14. (2000) “Monetary Policies in the Presence of Asymmetries.” Journal of Common Market Studies Vol 38, No.4, pp. 593-612.
15. International Monetary Fund (2007) International Financial Statistics, Retrieved from http://www.imfstatistics.orgKuttner K.N.
16. , Mosser P.C. (2002) “The Monetary Transmission Mechanism: Some Answers and Further Questions,” Federal Reserve Bank of New York, Economic Policy Review, May, pp. 15-26.
17. Levin A., Wieland V., Williams J.C.
18. (1999) “Robustness of Simple Monetary Policy Rules under Model Uncertainty,” Chapter No.6 in “Monetary Policy Rules,” edit. Taylor J.B.
19. , NBER Conference Report, The University of Chicago Press, pp.263-299.McCallum B. (1988) “Robustness Properties of a Rule for Monetary Policy,” Carnegie-Rochester Conference Series on Public Policy Vol.
20. 29; pp. 173-203.McCallum B.T.
21. , Nelson E. (1999) “Performance of Operational Policy Rules in an Estimated Semiclassical Structural Model,” Chapter No.1 in “Monetary Policy Rules,” edit. Taylor J.
22. B., NBER Conference Report, The University of Chicago Press, pp. 15-45.Mishkin F.
23. S. (1999) “Comment on Ch.5 “Policy Rules for Inflation Targeting'” by Rudebusch and Svensson,” in “Monetary Policy Rules,” edit. Taylor J.
24. B., NBER Conference Report, The University of Chicago Press, pp.247-253.Obstfeld M.
25. (1997) “Europe’s Gamble” Brookings Papers on Economic Activity, Volume 27, Issue 2, 241-300.Rotemberg J.J., Woodford M.
26. (1999) “Interest Rate Rules in an Estimated Sticky Price Model,” Chapter No.2 in “Monetary Policy Rules,” edit. Taylor J.B.
27. , NBER Conference Report, The University of Chicago Press, pp.57-119.Rudebusch G.D.
28. , Svensson L. (1999) “Policy Rules for Inflation Targeting,” Chapter No.5 in “Monetary Policy Rules,” edit. Taylor J.
29. B., NBER Conference Report, The University of Chicago Press, pp.203-246.Salvatore D.
30. (2002a) “The Euro, The European Central Bank, and the International Monetary System”, Annals, AAPSS, January.Salvatore D. (2002b).” The Euro: Expectations and Performance,” Eastern Economic Journal, Vol. 2