The role of money in the monetary policy strategy of the ECB

Speech by Otmar Issing, Member of the Executive Board of the ECB
Workshop on “What Central Banks can learn from money and credit aggregates”
Eltville, 28 October 2005

I. Introduction

It is a great pleasure for me to attend this conference, jointly organised by the Deutsche Bundesbank and the University of Bonn. It will surprise no one that I view the title of the conference – “what central banks can learn from money and credit aggregates” – with particular pleasure. This is a topic which should be of interest to all central banks and one to which I have devoted a considerable effort over time, from both a policy and academic perspective. I congratulate the organisers for its selection and for the impressive programme that they have put together.

In my remarks this morning, I will address the role of money in the ECB’s monetary policy strategy. This is certainly not a new topic. Nevertheless, there are a number of reasons why a return to it today is both timely and necessary.

First, at the ECB we continue to face criticism that our monetary analysis has not and does not provide information relevant for the decision-making of the Governing Council. Some critics have even argued that the monetary analysis is superfluous. Others went so far as to suggest that the monetary analysis has been used to deliberately obscure the real underlying policy making process. I reject these suggestions out of hand.

Ironically, many of those making these arguments against monetary analysis advocate a “full information” approach to monetary policy. The inconsistency in their argumentation is immediately clear. As I will develop further in a moment, there are clear empirical and conceptual reasons – not least the robust relationship between monetary growth and inflation over longer time horizons – why money can and should play a role in the design of a monetary policy aimed at the maintenance of price stability.

It is thus time to move on from the question of whether monetary analysis should play a role in monetary policy-making. This has proved an unfortunate distraction from the real question: how should monetary analysis be undertaken in order to provide the best support for monetary policy decisions. This is where I will focus the bulk of my remarks.

This brings me to the second critique of the ECB’s policy, which I wish to address today. A number of critical observers have urged us to merge the monetary analysis with the economic analysis. As I will discuss in greater detail in a moment, in my view this would simply not constitute an efficient and effective framework for recovering the information in monetary developments required for policy decisions. It is a potential answer to the question of how to organise monetary analysis, but not a good one. In my view, the two-pillar framework designed by the ECB has proved an effective means to organise and present the analyses of the monetary and economic data and has served to support a well-designed and effective monetary policy.

To illustrate this point, I will conclude my remarks with some comments on the conduct of monetary analysis in practice. Analysing the design and conduct of monetary policy-making in the euro area is no longer an abstract exercise. Almost seven years after the start of Monetary Union, we have gained some valuable lessons regarding how to conduct monetary analysis and I want to report on these today. They provide the most compelling evidence in support of the approach we have taken. To use an English phrase: “the proof of the pudding is in the eating”.

II. Challenges at the start of Monetary Union

European Monetary Union is an important milestone on the long and winding road to European integration. The Maastricht Treaty established the European Central Bank, endowed it with a high degree of independence and assigned it the primary objective of preserving price stability. History – not least here in Germany – has demonstrated that this institutional framework underpins successful policy making.

Yet it was clear from the outset that good institutional design alone was not sufficient to ensure the success of Monetary Union. As a brand new central bank, the ECB did not have an established track record of success in maintaining price stability to inspire confidence and build credibility. In this context, it was imperative that the ECB established a policy framework – our strategy – to govern and guide its actions, so as to reassure the general public in the euro area as to its intentions and thereby stabilise inflation expectations.

For any central bank, the choice of a monetary policy strategy is complex. In its preparatory work for Monetary Union, the European Monetary Institute (EMI) – on the basis of enormous effort and successful cooperation among the national central banks – had analysed deeply the issues relevant for the design of a strategy for the still to be established ECB. Some options, such as exchange rate targeting, had been ruled out by the EMI as appropriate strategies for the euro area. Among the remaining candidate strategies still under consideration in 1998, the most prominent were monetary targeting and inflation targeting.

Advocates of monetary targeting at that time drew on a number of compelling arguments.[1] Adopting a monetary target clearly emphasises the central bank’s responsibility for monetary conditions and price dynamics over the longer term. Monetary targeting therefore informs the public about the central bank’s fundamental orientation and can thus help to stabilise inflation expectations. By the same token, monetary targeting helps to distinguish the central bank’s responsibility from that of other policy institutions, serving to promote meaningful accountability. By adopting a monetary target, the ECB would have exposed itself to strict discipline, as any deviation from the target path would have required a full public explanation and justification.

Furthermore, prior to the start of Monetary Union, the available empirical studies – which, of necessity, had to rely on somewhat artificial data constructed by aggregating national series – indicated that the technical prerequisites for a monetary targeting strategy were in place. In particular, the demand for broad money in the euro area appeared to be stable over the preceding two decades. Indeed, money demand was more stable at the area-wide level than in individual countries.

Against this background, it is not surprising that a large number of observers advocated the adoption of monetary targeting by the ECB. For example, in 1997 the IMF noted that “a monetary target would also provide the ECB with the advantage of continuity with the framework used by the dominant central bank of the ERM, the Bundesbank, and it might thereby facilitate the transfer of credibility”.[2] The Sachverständigenrat (German Council of Economic Advisers) also favoured this approach arguing that “Until the opposite is proved, the European Central Bank should assume that there are sufficiently stable underlying monetary relationships in the single currency area and that unexpected instabilities can be brought under control”.[3]

Yet, mindful of the overriding need for a robust framework at the outset of a new – and thus uncertain – monetary policy epoch, I maintained what, with the benefit of hindsight, proved to be a healthy degree of scepticism. Monetary union marked a significant policy regime change, which could be expected to have a substantial impact on the growth of the money stock and its indicator properties regarding the outlook for price developments. The Lucas critique was central to my thinking. Bearing this in mind, in 1998 I suggested that elements of a strategy of monetary targeting should be supplemented, in a pragmatic but not arbitrary fashion, by some elements of the inflation targeting strategy.[4]

Having arrived at the ECB in June 1998, what had, until then, constituted some kind of thought experiment became an immediate and real challenge. As you might expect, a number of somewhat theoretical concerns suddenly took on a very concrete form. For example, empirical studies based on the then latest available euro area data cast doubts on the short-term controllability of the broad euro area aggregate M3. While this result had to be interpreted with great caution, given the inevitable statistical and economic uncertainties, it warned against the introduction of a monetary target. More fundamentally, with the establishment of a new central bank coinciding with the introduction of a new currency and a new monetary regime, it was clearly going to be a very ambitious task to separate temporary from more longer-lasting deviations from the monetary target, and thus to interpret the implications of such deviations for the outlook for price developments and monetary policy decisions. In sum, these considerations argued against the announcement of a monetary target.

Nonetheless, notwithstanding the practical problems we faced at that time, a number of basic, quasi eternal principles of economics and central banking needed to be respected. In particular, on both theoretical and empirical grounds, the ultimately monetary nature of inflation could not be challenged. That prolonged periods of high inflation are associated with high monetary growth is essentially undisputed in academic and central banking circles.[5] While other factors (such as variations in aggregate demand or technological changes) can influence price developments at shorter horizons, this does not negate the underlying long-term relationship between prices and the money stock. This key relationship called for the assignment of a prominent role for money in the ECB’s monetary policy strategy.

Against this background – and given the need to balance pragmatism on the one hand with respect for key underlying economic principles on the other – the ECB had to adopt a new and distinct monetary policy strategy of its own.

III. The ECB’s monetary policy strategy

The main elements of the ECB’s strategy were announced in October 1998. As is well-known, they consist of a quantitative definition of price stability, the primary objective given to the ECB by the Treaty, and a two-pillar framework for the analysis underpinning monetary policy decisions. The first element of this framework, the monetary pillar, focuses on the analysis of monetary developments, while the second element consists of a broad assessment of the outlook for price developments and risks to price stability.

The important role of money in the ECB’s strategy was signalled by the announcement of a reference value for the annual growth rate of M3. The reference value was derived such that M3 growth at the reference rate would be consistent with the maintenance of price stability over the medium term. Yet already at this stage, the Governing Council made clear that the reference value was not a monetary target. When presenting the strategy, the ECB immediately emphasised that it would not mechanically try to correct deviations of M3 growth from the reference value.[6] Rather the reference value was an indicator, a guide-post for policy making, derived from the long-run relationship between money and inflation. Moreover, from the outset it was clarified that the analysis of monetary growth with respect to the reference value would be accompanied by a thorough analysis of the components and counterparts of M3, in particular credit.

In the context of its reflections on the ECB's monetary policy strategy in the spring of 2003, the Governing Council decided to specify the objective of price stability more precisely and to clarify the role of the two pillars in the formulation of policy decisions. In May 2003, the Governing Council therefore underlined that the economic analysis should uncover the short- to medium-term risks to price stability, while the monetary analysis should allow the ECB to identify the longer-term and more persistent trends in inflation. The Governing Council deemed it essential to continue to cross-check the information stemming from the analysis undertaken under both pillars in coming to its monetary policy decisions.[7]

The evaluation of the strategy thus recognised explicitly the medium- to long-term perspective on the outlook for price developments provided by the monetary analysis. It thus clarified that the main challenge facing the monetary analysis is to see through the inevitable short-term disturbances of the underlying relationship between money and prices, so as to extract the longer-term inflationary risks identified by monetary developments. In practice, the crucial issue is to be able to identify changes in underlying monetary trend in real-time, so as to obtain information about the changes in inflationary trends timely.

In addressing this challenge, a number of tools have been used. First, as already mentioned, other monetary and credit indicators (notably the components and counterparts of M3) are thoroughly monitored, and institutional factors and financial innovation are taken into account. Second, various measures of the liquidity situation are regularly constructed and analysed in order to provide an assessment of the level of liquidity relative to estimated equilibrium values. Third, developments in M3 are analysed in conjunction with other economic indicators so as to understand better the causes of monetary developments and thereby to extract the underlying monetary dynamics driving the evolution of price trends that should govern monetary policy decisions. Finally, money and credit dynamics are analysed in close connection with asset price developments. As demonstrated in several recent papers[8], extraordinary increases in asset prices in financial history have typically been accompanied by strong monetary and/or credit growth. This empirical relationship suggests that monetary and/or credit aggregates can be important indicators of the possible emergence of asset price “bubbles”, and thus are crucial to any central banks’ approach to maintaining macroeconomic and price stability over the medium term.[9]

These are the main elements of the framework we have implemented over the past six years.

IV. Monetary analysis in real time

As I have said, “the proof of the pudding is in the eating”. A crucial test for any kind of analysis is whether an assessment of the data against the principles of the strategy can be translated into practical and relevant policy advice. As regards monetary analysis, this boils down to whether the important signal in monetary developments about future inflation trends can be separated from the inevitable noise in the published monthly monetary data.

As several recent academic papers applying statistical filtering techniques to monetary and price data – including some presented at this conference – confirm, resolving this signal extraction problem seems feasible ex post, once long time series are available. However, in real time the challenges facing central bankers are formidable. Not only must new data be monitored continuously, but tools and methods used must be perpetually enhanced, refined and tailored to address the specific challenges at hand.[10]

Over the past four and a half years, the ECB has gained considerable experience with the challenges faced in this respect. The economic, financial and geopolitical shocks we have faced over this period – a stock market crash; wars in the Middle East; accounting scandals on both sides of the Atlantic, to name but a few – have influenced monetary developments in important ways.[11]

As early as 2001, sizeable portfolio shifts into money – seen as a safe haven at a time of heightened uncertainty – took place, leading to the emergence of a sequence of “positive shocks” in traditional money demand models. In other words, standard money demand models had difficulties in explaining monetary dynamics on the basis of developments in the traditional determinants of money demand, such as output, prices and interest rates.

Such developments would have presented considerable problems for a conventional monetary targeting strategy. However, the ECB’s richer approach to monetary analysis quickly allowed the development of a pragmatic framework – embodying econometric, statistical and judgemental elements – within which to carry out a real-time assessment of monetary developments. This framework permitted us to distinguish those developments in M3 which were distorted by identifiable special factors from those which represented changes in the underlying monetary trend relevant for policy decisions.

Let me illustrate these deliberations in more detail. Between early-2001 and mid-2003, the annual growth rate of M3 was well above the reference value. The strong growth of broad money could only partly be explained by developments in the traditional determinants of money demand. This raised a number of fundamental questions: had money demand become unstable? Was strong monetary growth the result of a temporary and transient shock or did it reflect a structural shift in money demand? Did strong monetary growth imply upward risks to price stability?

From the outset, it was clear that these questions could not be answered solely on the basis of an analysis of the M3 time series. A deep and broad analysis of the components and counterparts of M3, capital market and international capital flows and the global liquidity situation was needed to single out the relevant signals from the inevitable noise in the monetary data in real-time. Based on this encompassing monetary analysis, the ECB identified and quantified so-called “portfolio shifts” – mostly from holdings of foreign securities into money market fund shares/units – which were exerting a significant influence on monetary dynamics. The repatriation of funds previously invested abroad into domestic monetary assets reflected a flight to safety in the aftermath of the stock market crash and terrorist attacks of 2001 and the subsequent geopolitical instability.

Viewed in this light, strong monetary growth was seen as reflecting a temporary change in savings behaviour, rather than a shift in the underlying or fundamental trend in monetary dynamics. As such, although it was recognised that the associated accumulation of liquidity could pose inflationary risks in the future should it be converted into transactions balances at a time of strengthening activity and confidence, the baseline scenario implied rather modest risks to price stability over the medium term stemming from stronger monetary growth.[12]

This assessment was based on the results of a variety of tools – a combination of model-based and judgemental analyses – that are employed on a regular basis at the ECB. For instance, extended money demand functions have been developed which take stock market developments into account, thereby allowing the impact of the global fall in stock prices on M3 growth to be quantified. Using this and other techniques, an M3 series corrected for the estimated impact of portfolio shifts was constructed and has been published by the ECB since the autumn of 2004. When analysed in parallel with the official M3 series, two scenarios for assessing inflationary risks could be derived. Reflecting the assessment that the inflationary impact of portfolio shifts was relatively benign, the modal scenario for price developments was derived using the M3 series corrected for the estimated impact of portfolio shifts, whereas analysis based on the official M3 time series provided a quantification of the upside risks to this modal scenario.

It deserves to be mentioned that – with the benefit of hindsight – we can see that the outlook for price developments made on the basis of the encompassing monetary analysis[13], and thus correcting for the estimated impact of portfolio shifts, has performed as a better real time indicator of inflation developments over this period than most other frameworks, including macroeconomic forecasts and projections made at the ECB and elsewhere.

With the end of the war in Iraq in early 2003, economic and financial uncertainty fell to more normal levels and past portfolio shifts began to unwind. Annual M3 growth fell from mid-2003 until mid-2004. This appeared to confirm the assessment that the earlier surge in M3 growth had been transitory, and was not a harbinger of strong underlying monetary dynamics implying a significant upward trend in price developments.

However, the assessment of modest medium-term risks to price stability stemming from the monetary analysis changed in the second half of 2004, when broad money growth accelerated again. Although less pronounced in terms of headline growth, this acceleration has been judged to be of a different quality to that observed between 2001 and 2003. On the counterparts side, stronger monetary growth has been associated with increasing demand for loans to the private sector rather than capital flows from abroad. On the components side, higher M3 growth has been driven by its most liquid components, pointing to a significant impact of the low level of interest rates on monetary dynamics. Stronger monetary dynamics can no longer be interpreted as reflecting heightened financial and economic uncertainties, which have normalised since 2003. Since faster monetary growth appears to have been more fundamental in nature since mid-2004 than was the case between 2001 and 2003, the likelihood that strong monetary developments ultimately find their way through to higher prices must be seen as considerably higher.[14] Moreover, strong money and credit growth in a context of already ample liquidity in the euro area implies that asset price developments, particularly in housing markets, need to be monitored more closely, given the potential for misalignments to emerge.

Of course, as always, the signals coming from the monetary analysis have to be read with and cross-checked against the assessment derived from the economic analysis. This is the essence of the two-pillar approach. There is certainly no mechanical reaction of monetary policy to monetary developments. At the same time, it is apparent from my description of how the ECB has set about its monetary analysis over the past six and a half years that organising this analysis in a distinct and coherent pillar has facilitated its development and discussion among policy makers.

V. Conclusions

Looking back on almost seven years of successful monetary policy, we can draw a number of conclusions.

First, in its assessment and policy considerations, a central bank cannot ignore the information from monetary developments. This is a natural and immediate implication of the “full-information” approach to the design and conduct of monetary policy. The indications provided by developments in monetary and credit aggregates are simply too important to be left out. The relevant debate is not whether central banks should conduct monetary analysis, but rather how they should do so.

Second, when coming to a policy assessment a simple look at developments in headline M3 is obviously not enough. To analyse developments of money and credit, the ECB has continuously broadened and deepened its set of tools. The necessity of this approach was confirmed during the challenging period when the global economy entered troubled waters and financial and economic uncertainty rose.

Third, by cross-checking the available information, the two-pillar approach guarantees an important role for monetary analysis and thus supports the medium-term orientation of the ECB’s monetary policy, avoiding the trap of “short-termism”. The internal coherence and clarity of the monetary analysis is also served by maintaining a distinct pillar for it within the overall strategy.

Conducting monetary policy in the euro area – and let me recall that seven years ago, this currency area did not yet exist – has certainly proved to be an ongoing process of learning. This has been true in the field of monetary analysis, as elsewhere. It has been a challenging but necessary task to identify the information in monetary developments that is relevant for monetary policy decisions. I remain convinced that monetary analysis can, should and does play an important role in the ECB’s monetary policy framework, notably as a means to check, from a medium to longer-term perspective, the assessment of short to medium-term risks to price stability obtained from economic analysis. It thereby helps to ensure that the Governing Council, in forming its overall judgement of the risks to price stability, does not overlook potential implications of monetary developments for future price stability.

I started with reference to critiques of our strategy in general and the monetary pillar in particular. At the end of my talk, let me simply reverse the question: How can any central bank afford to ignore the information stemming from monetary analysis?


Borio, C. and Lowe, P. (2002), “Asset prices, financial and monetary stability: Exploring the nexus” BIS working paper no. 114.

Detken, C. and Smets, F. (2004), “Asset price booms and monetary policy” ECB working paper no. 364.

ECB (1998), “A Stability-Oriented Monetary Policy Strategy for the ESCB”, Press Release, 13 October 1998.

ECB (2001-2005), “Introductory Statement”, various issues.

ECB (2003), “The ECB’s Monetary Policy Strategy”, Press Release, 8 May 2003.

ECB (2004), “Monetary Analysis in Real Time”, Monthly Bulletin October, pp. 43-63.

ECB (2005), “Money Demand and Uncertainty”, Monthly Bulletin October, pp. 57-74.

IMF (1997), “World Economic Outlook”, October, Washington, D.C.

Issing, O. (1997), “Monetary Targeting in Germany: the Stability of Monetary Policy and of the Monetary System”, Journal of Monetary Economics 39, pp. 67-79.

Issing, O. (1998a), “The European Central Bank as a new institution and the problem of accountability”, Deutsche Bundesbank, Auszüge aus Presseartikeln, Nr. 19, pp. 1-7.

Issing, O. (1998b), “Welche geldpolitische Strategie für die EZB?“. In: Oesterreichische Nationalbank (ed.), “Wirtschaftspolitik 2000 – Die Rolle der Wirtschaftspolitik und nationaler Notenbanken in der WWU“, 26. Volkswirtschaftliche Tagung der Oesterreichischen Nationalbank,

Issing, O. (2004), “Financial Integration, Asset prices and Monetary Policy”, Dinner Speech held at the Symposium concluding two Years of the ECB-CFS Research Network on “Capital Markets and Financial Integration in Europe”,

Lucas, R.E. (1972), “Expectations and the Neutrality of Money”, Journal of Economic Theory, pp. 103-124.

Lucas, R.E. (1996), “Nobel Lecture: Monetary Neutrality”, Journal of Political Economy, Vol. 104, pp. 661-682.

McCandless, G.T. and Weber, W.E. (1995), “Some Monetary Facts”, Federal Reserve of Minneapolis Review, Vol. 19, No. 3, pp. 2-11.

Sachverständigenrat (1997), “Wachstum, Beschäftigung, Währungsunion – Orientierung für die Zukunft“, Annual Report, November, Council of Economic Advisers.

[1] It should be noted, however, that the practical experience with monetary targeting has not been dogmatic. The Bundesbank, for instance, has time and again pointed to the medium-term nature of its strategy. Furthermore, as part of ”concretizing its target”, the Bundesbank has always taken the underlying monetary situation at the time the decision was taken into account. See Issing (1997) for further details.

[2] See IMF (1997), p. 54.

[3] See Sachverständigenrat (1997), p. 356, translation by the author.

[4] See Issing (1998b), p. 112.

[5] See, for instance, Lucas (1972, 1996) or McCandless and Weber (1995).

[6] See the Press Release entitled “A stability-oriented monetary policy strategy for the ESCB”, 13 October 1998.

[7] See the Press Release entitled “The ECB’s monetary policy strategy”, 8 May 2003.

[8] See Borio and Lowe (2001) and Detken and Smets (2004).

[9] See Issing (2004) for a more detailed discussion of this issue.

[10] See also the Monthly Bulletin article “Monetary Analysis in Real Time” (October 2004, pp. 43-66) for a more detailed description.

[11] See the Monthly Bulletin article “Monetary Demand and Uncertainty” (October 2005, pp. 57-74) for a more detailed description

[12] See, for instance, the Introductory Statement of October 2001: “… a number of temporary factors play an important role in explaining recent monetary developments. Notably the uncertainty in stock markets and the relatively flat yield curve until August have led to portfolio shifts... For these reasons, we do not judge that monetary developments signal risks to price stability at this juncture”.

[13] See the Monthly Bulletin article “Monetary Analysis in Real Time” (October 2004, op. cit., pp. 56-57) on this issue.

[14] See, for instance, the Introductory Statement of 6 October 2005: “To sum up, the economic analysis indicates that oil and petrol increases, in particular, imply upward revisions to the outlook for short-term price developments. Some of the increases can be expect to be of temporary nature, while others are likely to be more lasting. Domestic inflationary pressures over the medium term still remain contained in the euro area, but significant upside risks have to be taken into account. Moreover, the monetary analysis identifies risks to price stability over the medium to longer term”.

Media contacts