Prince Henri Auditoire 02 BW

On the European monetary policy strategy and practice

by Mr. Yves Mersch, Member of the European Central Bank Governing Council

Tokyo, 23 October 2001.

Seule la parole prononcée fait foi

Ladies and Gentlemen,

Speaking on strategy reminds me of the following quotation from the "Recorded Sayings" of the Founder of Kencho-Ji Temple in Kamakura, Rankei Doryu: "If you have lost your true self, all phenomena bring you nothing but annoyance. If you discover your essence of mind, you can follow nothing but the true path."

Before starting with the main content of my speech, I would like to make some brief remarks concerning the common mistake to compare the US and European area economies with regard to both their cyclical position and their medium-term growth outlook.

The euro area economy has been affected by the international economic climate and, in particular, by the sharp slowdown of the US economy. This is a natural effect of the growing interdependence of regional economic areas, brought about by increase in international trade and financial flows. Nonetheless, it would be a mistake to draw excessive parallels between economies, which exhibit non-synchronised economic cycles, different structural characteristics, and hence unequal long-term growth prospects.

In fact, many of the macroeconomic and financial disequilibria, which have built up in the US, cannot be found in the euro area. Although investment has remained buoyant in recent years in the Euro area, there is no sign of surplus accumulation in capital stock. Moreover, although the European savings ratio has declined in recent years, it is still appreciably higher than in the US. Lastly, the private sector, with the likely exception of the telecom sector, is in a less vulnerable position, being characterised by lower relative debt and less exposure to fluctuations in financial asset prices.

The policy-mix in the United States is the result of a one-to-one relation between one monetary policy and one fiscal policy. The European monetary area displays an asymmetric relation in this respect. The policy mix is composed of a single monetary policy for the twelve participating member states, with fiscal policies remaining to a large extent in the hands of national authorities.

Against this background, it should come as no surprise that European monetary policy has been, is and will be different from the monetary policy of the Federal Reserve System. This discrepancy is, I believe, not so much due to differences in the statutory remit of the respective central banks but rather to economic situations departing from different bases and requiring different measures.

Having clarified - once more - the differences between the US and the Euro area economies, and hence the appropriate monetary policy, I shall concentrate on the strategy adopted by the Eurosystem in the remainder of my speech, without making too many references to the US.

The Eurosystem Monetary Policy Strategy

1.   The Need for a Monetary Policy Strategy

Quite naturally, monetary policy decision-makers operate in a world of uncertainty. Information or knowledge of economic activity and structures are necessarily imperfect. Complete descriptions of economic activity are only available after a significant time lag. Usually, they become available after the monetary policy decisions have been taken.

This simple fact has several implications being of particular importance for the conduct of monetary policy. A large academic literature has also addressed the question of optimal monetary policy in an uncertain environment. Together with the long-standing practical experience of Euro area national central banks, this literature has shaped the design of the Eurosystem monetary policy strategy.

It is of fundamental importance to ensure, as far as possible, that monetary policy itself does not become an additional source of uncertainty. Hence the ECB's monetary policy strategy announcement attempts to address the issue of uncertainty in providing a very clear and concise systematic framework for the analysis of information and the taking of policy decisions, without specific policy conclusions being predetermined in a mechanical manner. The design of the Eurosystem monetary policy strategy also recognises that any specific economic framework is incomplete and cannot capture the full complexity of the monetary transmission process in the 12 Euro area economies. This is done in employing different analytical frameworks, commonly referred to as the two-pillar strategy, and in relying on long-standing professional experience, rather than to solely rely on one single monetary or economic indicator. This considerably reduces the risk of making substantial policy errors, disciplines policy choices and keeps decisions consistently anchored to the mandated objectives.

ECB's Definition of Price Stability

In contrast to the FED's mandate, the Maastricht Treaty establishing the European Union states that "the primary objective of the European System of Central Banks shall be to maintain price stability". The Treaty thus assigns a clear mandate to the newly created central banking system, while leaving open a more precise and operational definition of the price stability objective, as well as the design of the monetary strategy to be adopted in order to fulfil this objective.

Making use of its independence to decide how to fulfil the mandate given by the Maastricht Treaty, the Governing Council announced its "stability-oriented monetary policy" in October 1998 and laid out how it intended to interpret and make operational the price-stability mandate. Several elements of that announcement are worth reminding and discussing here.

Firstly, the Governing Council stated that the "monetary policy strategy will focus strictly on this objective of price stability". In addition, a quantitative definition of price-stability was announced. "Price stability shall be defined as a year-on-year increase in Harmonised Index of Consumer Price (HICP) for the Euro area of below 2%", with price stability having to be maintained "over the medium term."

Hence the existence of short-term price volatility, which cannot be controlled by monetary policy, was explicitly recognised. The announcement contained an explicit and quantitative definition of price stability – which the Fed, among other central banks, lacks. This feature improves accountability and helps to anchor economic agents' expectations.

Do We Need a Quantitative Definition?

Monetary policy could target a constant price level, a zero inflation rate, an upward drifting price level or a low but nonzero rate of inflation (Fisher, 1996; Goodhart and Viñals, 1994). An advantage of a price level target is that it keeps uncertainty about price levels in the distant future smaller than it would with zero inflation target, which entails a base drift. The main disadvantage is that it puts a greater strain on monetary policy, because variations in the inflation rate are required to reverse the effects of previous shocks on the price level. There are in addition several reasons for aiming for a small positive inflation rate instead of a zero rate:

According to Debelle (1997) the true rate of inflation is below the measured rate, which is due to measurement problems,

As Akerlof, Dickens and Perry (1996) point out, if nominal wages are rigid downward, then reductions in real wages will involve less employment loss when brought about through inflation. Very low inflation therefore effectively reduces real-wage flexibility and may worsen the allocative efficiency of the labour market,

With a zero targeted inflation rate, there is a greater danger of deflation. As pointed out in the literature on financial crises, persistent deflation can create serious problems for the financial system.

The lower bound of zero on the nominal interest rate arises because cash carries a zero nominal interest rate, which imposes problems for monetary policy with a zero targeted inflation rate (Summers, 1991). If the expected inflation rate is zero, it is very difficult to engineer a negative short-run real interest rate (which may be needed during recessions).

These arguments led the Governing Council of the Eurosystem to conclude that price stability is the best way to insure an endurable growth in the euro area and to adopt a quantified definition of price stability which accepts a rate of inflation of up to 2%.

Important to note is also that, in addition to the quantitative specification of the price stability objective, the monetary policy strategy announcement emphasised very clearly from the very beginning that strategy and decisions will aim at stabilising inflation in the Euro area as a whole, irrespective of inflation developments in any of its constituent members.

2.   The Eurosystem's Monetary Policy Strategy

The Two-Pillar Strategy

Let me go back to the "stability-oriented monetary policy strategy". In addition to the quantitative definition of the price stability objective, a clear commitment to the goal variable which I have just discussed, the announcement contained a description of the strategy to be implemented. The monetary policy strategy consists of two key elements, the famous two pillars. To recapitulate briefly, under the first pillar, "money will be assigned a prominent role [Â…] signalled by the announcement of a quantitative reference value for the growth of a broad monetary aggregate". In addition, and in parallel with that analysis of monetary growth, a major role is assigned to "a broadly-based assessment of the outlook for price developments and the risks to price stability in the Euro area" - the second pillar.

We thus recognise different existing models of the structure of the economy and the nature of the monetary transmission mechanism. The first pillar represents a group of models and analytical frameworks which embody a view of price level determination that accords an important role to money. The second pillar encompasses a range of alternative models of the inflation process, predominantly those which emphasise the interplay between supply and demand in the goods and labour markets, as well as information useful for monetary policy stemming from financial variables, such as bond yields, asset prices (including exchange rates), credit developments and balance sheet positions.

First Pillar:

The prominent role for money reflects the monetary origins of inflation over the medium to longer term. One of the most remarkable empirical regularities in macroeconomics is the ubiquitous long-run relationship between the price level and money stock. A positive and often almost one-to-one relationship between monetary growth and inflation at longer horizons has been illustrated for a wide variety of economies, using a number of different analytical and empirical tools and employing various definitions of money data sets.

There is a broad consensus in the literature that any well-specified model of a monetary economy should exhibit this feature. On both empirical and theoretical grounds, the literature provides ample justification for assigning money an important role in monetary policy making.

This conclusion is particularly appropriate for the Euro area. In contrast to other countries where monetary developments during recent decades have been rather erratic, in the Euro area the available evidence continues to point towards the existence of a stable relationship between broad monetary aggregates – in particular M3 – and price developments in the long term. Moreover, M3 and other monetary and credit aggregates appear to possess good leading indicator properties for future price developments, especially at medium-term horizons. The reference value of M3 growth is presently set at 4 ½% and will be revisited in December. This value is considered consistent with the long-run economic growth rate in the Eurozone, the inflation objective of below 2% in the medium term and the trend reduction in money velocity.

However, deviations of M3 growth from the Eurosystems reference value of 4 ½% do not mechanically trigger interest rate changes. Rather, such deviations are to be seen as catalyst for further analysis, first searching for the causes of such deviations and thereafter assessing the implications for future price developments and the appropriate monetary policy.

Therefore, the reference value for M3 is not an immediate or direct monetary target in itself. The Eurosystem does not attempt to control monetary growth in the short term so as to hit the reference value at a specific horizon. Rather the reference value acts as an analytical and presentational device, which constitutes an important benchmark for assessing risks to price stability in the medium term.

The analysis under the first pillar is particularly important in making an evaluation of the magnitude of monetary impulses to the economy and may help in the assessment of possible bubbles in financial markets. Historically, booms and busts in asset markets have been strongly associated with large movements in monetary and credit aggregates in particular, and their implications on the economy may depend on the strength of the balance sheet position of the financial sector. This constitutes a further reason for the Eurosystem strategy to give special consideration to the analysis of monetary and credit aggregates and of financial intermediaries' balance sheets from which these aggregates are derived.

Second Pillar:

The second pillar of the monetary policy strategy is a broadly-based assessment of the outlook for price developments. Forecasts of crucial economic indicators, including inflation developments, play a significant role in this assessment. While inflation is ultimately a monetary phenomenon, monetary developments may, at least in the short term, not always be the only or even the most useful guide for assessing the economic disturbances threatening price stability.

A monetary policy aimed at the maintenance of price stability must therefore rely on the analysis of a broad range of variables. Therefore, in parallel with the analysis of monetary developments, the Eurosystem also evaluates a wide range of other economic and financial indicators in broadly assessing the risks to price stability in the medium term.

It is vital that these two pillars are not misinterpreted as targets or any form of monetary rule. On the contrary, the Governing Council has made it clear on numerous occasions that it will take its decisions on the basis of a detailed and comprehensive assessment of a broad range of economic indicators. Isolated developments of individual indicators, such as deviations of monetary growth from the reference value, will not prompt automatic and mechanistic monetary policy responses.

The structure of the monetary policy strategy clarifies the much-discussed role of the exchange rate. Along with other indicators, the exchange rate of the Euro is a component of the second pillar. It is not regarded as a target of the Eurosystem's monetary policy, but it is given due consideration in actual monetary policy decisions. Nonetheless, an increasing weight will be attached to the exchange rate whenever its development threatens the medium term internal price stability in the Euro zone. In any other case, monetary policy decisions will be largely unaffected. This is also true of all the other indicators of the second pillar, such as wage developments.

Ultimately, the Governing Council has to come to a monetary policy decision on the basis of its collective overall assessment of the economic situation and the risks to price stability. This decision will take into consideration all the available relevant information, and not just the information reflected in forecasts and/or deviations of M3 growth from its reference value.

3.   The Monetary Policy Strategy in Practice

A Closer Look at the Interest Rate Decisions by the Governing Council

The first column in table 1 shows the dates at which the Governing Council decided to change the key policy rate while the second column displays the level of that rate. The third column reports the cumulative change in that interest rate since the onset of EMU. The fourth column shows the annual HICP inflation rate for the Euro area corresponding to the month preceding the decision (and hence, one that is likely to be available at the date of the meeting). The fifth column displays its cumulative change, the sixth column reports the level of M3 and the seventh the cumulative change in M3 growth at the same time.

Interest Rate Changes, HIPC Inflation and M3 Growth
ECB Interest rate Cumulative Change HICP Cumulative Change M3 Cumulative Change
22 December 98 3.00 0.80 5.70
09 April 99 2.50 -0.50 1.00 0.20 4.90 -0.80
05 November 99 3.00 0.00 1.40 0.60 6.20 +0.50
4 February 99 3.25 +0.25 1.90 1.10 6.20 +0.50
17 March 00 3.50 +0.50 2.00 1.20 6.50 +0.80
28 April 00 3.75 +0.75 2.10 1.30 6.50 +0.80
09 June 00 4.25 +1.25 1.90 1.10 5.40 -0.30
01 September 00 4.50 +1.50 2.40 1.60 5.60 -0.10
06 October 00 4.75 +1.75 2.70 1.90 5.30 -0.40
December 00 4.75 +1.75 2.60 1.80 4.80 -0.90
11 May 01 4.50 +1.50 2.90 2.10 5.30 -0.40
31 August 01 4.25 +1.25 2.8 2.00 6.70 +1.00
18 September 01 3.75 +0.75 2.70 1.90 n.a. +1.00

Until May 2001 all decisions on interest rates, with the exception of the 50 basis point rate cut in April 1999, have taken the form of an increase of 25 or 50 basis points. By the end of EMU's second year, the interest rate has risen by 175 basis points, matching almost exactly the corresponding cumulative change in inflation. Hence the real interest rate was basically unchanged.

Despite the negative cumulative change until May 2001, the annual rate of M3 growth continues to be above the fixed reference growth of the ECB. The recent three-month average annual growth rate during the period June-August 2001 increased to 6.4%. The corresponding growth rate during the period May-July 2001 was 5.9%. These developments partly reflect the relatively flat yield curve up to August and the weakness in stock markets, which made the holdings of short-term deposits and marketable instruments included in M3 more attractive.

As pointed out previously, the eurosystem's surveys indicate that the data for M3 are distorted upwards by non-resident holdings of market paper and debt securities with a maturity up to two years issued by MFIs. These holdings may have contributed with three-quarters of percentage point to annual M3 growth over recent months. In fact, the subtraction of this amount reduces the excess with regard to the reference value of 4 ½%.

The Governing Council kept interest rates unchanged until the 11 th of May 2001. This can be explained because the deterioration of the price outlook was due to temporary factors, such as the oil price and the exchange rate of the euro against the dollar. In addition, the slowdown in US economy and the subsequent deterioration in the world economic outlook lowered inflationary risks as from the third quarter of 2000, paving the way in October 2000 for a halt in the succession of official interest rates rises.

On the basis of the evaluation of inflationary risks in the area and the development of the external environment, the Governing Council has always adopted a prudent stance, gradually modifying his assessment of macroeconomic conditions in the euro area. As a result, the Governing Council set the benchmark interest rate in May 2001 to 4.5%, estimating that the lower economic growth envisaged for the euro area should mitigate inflationary pressures in the medium term followed by a further cut of 25 basis points on August 31 st.

On 17 September 2001, the Governing Council took the decision to lower the key interest rates by 50 basis points. This was done in concert with the Federal Reserve System, and a direct response to the increased uncertainty regarding the global economic environment following the terrorist attacks on the US. With the benefit of hindsight, this interest rate decision can be viewed as a "front loading", both in terms of timing and of magnitude, of a reduction which would have occurred under more normal circumstances in subsequent Governing Council meetings.

The Governing Council monitors inflation developments in the euro area attentively. While the less expansionary economic climate tends to moderate inflation, risk factors remain. These are associated with uncertainty concerning wage developments as a consequence of price increases of energy and other imported inputs. While these factors could delay the reduction in inflation rates in the near future, the most recent developments allow one to expect further gradual reductions in recorded inflation.

I think it has become apparent to you that decisions on the money policy stance do not mechanically follow from deviations of M3 growth from the reference value. More generally, the medium term focus recognises the presence of different transmission channels affecting price developments with long, variable and uncertain lags. Money growth, in particular, is associated with inflation at a medium to long term horizon. In fact, extracting information from monetary developments is not a mechanical process. Over policy-relevant horizons, money and credit may be influenced by many factors, including cyclical shifts in the demand for money and credit and regulation. The same applies in particular for the publication of most economic projections. So, a main task of monetary analysis is to understand and quantify these various possible effects, drawing out the key implications for inflation outlook. In addition, acknowledging that the inflation is the central goal of the Eurosystem, does not imply that its monetary policy should ignore significant output movements.

Governing Council Decision on 31 August 2001

With regard to the first pillar, the increase in the growth rate of M3 of 5.9% must be interpreted carefully. In fact, it reflects both the weakness in stock markets and the relatively flat yield curve. In addition, there have been indications that monetary growth figures are distorted upwards by non-Euro area residents' purchases of negotiable paper included in M3.

Regarding the second pillar, there are clear signals of lower inflationary pressures coming from the demand side. Recent economic indicators show that the Euro area's real GDP growth in this year will most likely be lower than was expected a few months ago. However, the growth outlook in the Euro area is better than the USA's growth perspective and the present level of the ECB interest rate conforms to its strategy.

Finally, I think that financial and economic agents recognise that the monetary policy strategy of the Eurosystem has been implemented with a high degree of flexibility since the beginning of 1999. In the first few months of European Monetary Union, economists worried about threats of deflation and negative real effects following the international financial crisis. Then they worried about the strong US Dollar and the increases in oil prices and the generated inflation risk. Later they were concerned about the slowdown of economic growth in the USA and the much heralded new economy. Did it really exist, one may wonder? They worry about financial difficulties in Turkey and Argentina, as well as a prolonged recession in the USA, following the tragic events of 11 September, which is feared to spread around the globe.

All these issues are also taken into consideration when deliberating about the appropriate monetary policy decision in order to achieve price stability in the medium term.

4.   Future Challenges for the Eurosystem

Despite this success, many challenges for monetary policy and the Eurosystem persist. Among these are:

  • The introduction of notes and coins in January 2002 (to which I very much look forward),
  • The mastering of inflation phenomena and persuasion of other policy making institutions to share the Eurosystem's policy and objectives, particularly in this economic period of stress. While short term deviations from budgetary targets appear acceptable in the present environment, they must not lead to delaying or even endangering the set medium term fixed policy objectives.
  • The future EU enlargement and its implications on ECB future reform.

Please allow me to elaborate on this point before closing my speech.

Let me emphasise that the Eurosystem's attitude to EU membership of the accession countries is positive. However, it is important to realise that any political entry into the EU & into the EMU are fundamentally distinct. At the moment, accession countries are foremost interested in joining the EU.

Once, these countries have joined the EU, there is a very specific procedure for joining EMU. In fact, accession to EU does not guarantee automatic acceptance into the monetary union. After joining the European Union, the newcomers are expected to join the European Exchange Rate Mechanism (ERMII). This is necessary to start the reporting period clock. The Maastricht treaty specifically requires that compliance with the debt, deficit, interest rates and inflation targets be evaluated over the year before the examination. However, the reporting period is 2 years for the exchange rate criterion (ERM membership without devaluation), so the evaluation can (theoretically) not be passed until a 2-year track record exists.

After the 2 nd accession anniversary to the EU, the Commission and ECB need about 2 months to receive the final economic data and to write their convergence reports. These reports assess the suitability of all non-EMU Member States for EMU membership based on compliance with the EMU-related acquis and the Maastricht economic convergence criteria. The Maastricht criteria require inflation rates, government deficits and debts, and long-term interest rates to have been low enough and exchange rates to have been stable enough during the reporting period. While there are other criteria, experience shows that if these are met, one can expect that the members will be judged to be fit to enter Eurosystem. The Ecofin Council will then deliberate on the reports for one month, suggesting that the Ecofin vote can come no sooner than 27 months after EU accession. If the answer is "yes" the EMU-members-elect may need some time to adopt the euro. It took the founding members 8 months, while it took Greece 6 months. Hence, voting rights in the Governing Council should come about 6 months after the Ecofin decision. If the rules are followed to the letter, the process should take at least 33 months. This represents enough time to the Governing Council to resolve the problem of necessary reform.

This is of course the most rigid and swift entry scenario, which appears to be rather unlikely given the strong need for real convergence to set in. By real convergence we think mostly of convergence terms of GDP per capita and industrial structures, in order to minimise the risks and the effects of asymmetric shocks. Also, it is in the interest of 'in' and 'out' countries to enter EMU only when they are able to sustain the new monetary policy system. Let me explain why the nominal convergence is not sufficient to join the EMU.

In the year 2000, the degree of real convergence remained limited, as Euro area growth was nearly as strong as growth in the accession countries. Furthermore, it is unlikely that the medium-term growth differential will be significantly higher than 1% in favour of the accession countries. Together with the huge current GDP per capita differential of about 56% in purchasing power parity terms or about 78% in terms of current exchange rates, it seems that the GDP per capita gap between Euro area countries and accession countries will remain large for the next few years.

Therefore, it cannot be emphasised enough that substantial structural reform and prudent macroeconomic policies are crucial to enhance accession countries' growth potential, thereby fostering cohesion within the enlarged EU and hence minimising the risks and effects of asymmetric shocks.

5.   Concluding Remarks

My speech provided an overview of Eurosystem's monetary policy strategy in theory and practice. I have illustrated the monetary policy framework of EMU and its operation during its first two and a half years, and gave a brief outlook regarding the challenges that lie ahead of us.

The Eurosystem has shown the capability to ensure price stability in the medium term, given the means at its disposal. Other instruments affecting price stability lie in the hands of national governments, and hence are outside the control of the Eurosystem. The task for national governments remains to adhere to the stability and growth pact and to continue driving forward the necessary structural reforms. Economic growth and the maintenance of price stability also and especially depend on appropriate wage agreements. I am heartened by developments in this area, which bear testimony to the emergency and consolidation of a stability culture in Europe.

These developments are desirable from an economic angle and, given present times, also from a wider political standpoint.

Thank you very much for your interest.