Ladies and Gentlemen, dear colleagues, 
It is a great pleasure to speak to you today.
I would like to use this opportunity to reflect on a fundamental tension lying at the heart of monetary policy:
A central bank’s success in ensuring macroeconomic stability depends on it having a credible commitment to price stability. And to build and nurture such credibility, it should have a track record of responding in a consistent, transparent and predictable way to shocks hitting the economy.
At the same time, financial and macroeconomic disturbances are sometimes of a scale and complexity that they alter the underlying structural relationships between key economic variables. This, in turn, challenges monetary policy-makers to temporarily adapt their established strategies to a new environment without undermining their inflation-fighting credentials.
The current crisis provides a salient case study of such a tension: starting with a loss of confidence in the solvency of certain financial institutions, the crisis quickly morphed into a broad-based impairment of private financial intermediation. As a consequence, the link between monetary policy impulses and economic aggregates – which is the structural relationship most relevant for central banks – was severely affected.
Confronted with this new environment, central banks across the globe have expanded their toolkits to compensate for their reduced influence on economic outcomes. The measures they took have in many cases helped to stabilise the economy. But they have also triggered a lively debate on where to draw the line between crisis-fighting efforts on the one hand – and a general departure from established principles of monetary policy on the other.
I would like to consider how central banks can resolve the tension between anchoring long-term expectations and flexibly confronting short-term challenges. I will argue that the key to success lies in a clear-cut hierarchy between the different elements of a monetary policy strategy. In doing so, I will highlight two distinct elements: the objective of policy and the conduct of policy. The monetary policy objective of a central bank should always be the overriding concern and guiding principle of any policy action. Building on the credibility that derives from having a firm and stable commitment to its objective, the central bank should then flexibly tailor its policy conduct to the specific macroeconomic disturbances that it has to face.
As I will demonstrate, the ECB’s monetary policy strategy during the crisis has consistently taken this approach.
Let me first briefly expand on the distinction between the different elements of a monetary policy strategy.
What I refer to as monetary policy objectives are the overarching economic goals of a central bank. For most central banks around the globe, these goals include – as a primary component – the pursuit of price stability. This objective is often complemented by ancillary policy preoccupations, such as minimising deviations from sustainable growth and financial stability. In the modern central banking universe that has emerged since the macroeconomic failures of the 1970s, no one would doubt that while price stability is a desirable medium-term goal for monetary policy. Avoiding excessive economic fluctuations around the trend and minimising financial instability, however, often has a different status. These objectives are ancillary because they have more to do with the path to the medium-term objective – a path that should minimise unnecessary fluctuations under way – than with the medium-term objective itself, which remains price stability.
That being said, central banks attach different priorities to these ancillary policy goals as they seek to steer the economy to its steady state. In the US, for instance, the Federal Reserve System has a dual mandate: maximum employment and stable prices. By contrast, the Treaty on the Functioning of the European Union clearly specifies price stability as the primary objective for the Eurosystem. I will come back to this distinction between the Eurosystem and the Federal Reserve at a later stage.
The conduct of monetary policy refers to the way a central bank goes about achieving its objective. This usually includes systematic monitoring of specific indicators or macroeconomic aggregates that the central bank uses as guideposts towards its final objective. The aggregates are chosen because of their close empirical association with the variables underlying the ultimate monetary policy objectives. They may include, for example, broad monetary aggregates or market interest rates. In addition, the policy conduct involves certain instruments which a central bank uses to steer the financial conditions under its direct control. These include reserve requirements, the monetary base, as well as the key interest rates at which the central bank provides liquidity to market participants.
Changes in monetary policy strategy have very different implications, depending on which of these two constituent elements – policy objectives or policy conduct – they affect. From a practical perspective, objectives are more difficult to change, inter alia, because they are usually enshrined in law. By contrast, the conduct of policy is, to a large extent, at the discretion of the respective central bank.
In the euro area, for example, the price stability objective has “constitutional” status: it is explicitly laid down in Article 127(1) of the Treaty. At the same time, the institutional framework leaves considerable scope for the Eurosystem to calibrate its policy measures so that it achieves its price stability objective. There is one exception to this general fact, which I will turn to soon.
The success of a monetary policy strategy depends on the interaction between these elements: the more credible a central bank’s commitment to its objective, the more flexibly it can adjust its monetary policy conduct to specific challenges.
There are two reasons for this. First, if expectations are firmly anchored by the central bank’s objective, short-term “innovations” in policy conduct will not induce market participants to fundamentally re-assess their anticipations of where the economy will stabilise in the medium term, when all the current shocks have dissipated. This is particularly relevant in crises when the central bank is prompted to expand its toolbox so as to confront exceptional circumstances. By contrast, without a firm commitment to its objective, the adoption of new, temporary instruments may be misinterpreted as a profound shift in the central bank’s overall strategy. And this can change the way society perceives the basic parameters of the monetary policy framework. Inflation expectations – one prominent parameter of how the regime is perceived by the public – can become unanchored.
Second, a credible commitment to the objective allows a central bank to tailor the strength and timing of its monetary policy response more specifically to the shocks that hit the economy from time to time. Macroeconomic disturbances can be broadly categorised according to whether they affect mainly the demand side or the supply side of the economy. In the case of a negative demand shock, both economic activity and inflation will tend to fall. Typically, a financial crisis is one example of a negative demand shock: credit conditions are restricted, aggregate demand is restrained and firms are encouraged to undercut their competitors, thereby exerting downward pressure on inflation.
To the extent warranted by the magnitude of the shock, the central bank thus faces an unambiguous policy prescription, namely to provide additional monetary accommodation.
In the case of a negative supply shock, the central bank has a policy dilemma as economic activity and inflation move in opposite directions. A drop in productivity is a case in point: an increase in companies’ cost per unit of output makes production less rewarding, and thus lowers the amount of goods and services supplied relative to demand. Here, the central bank either temporarily accepts higher inflation, or – in an attempt to keep inflation right on target – it reinforces the disturbance driving down output.
But this trade-off is weakened when the central bank can credibly commit to its price stability objective. Knowing that the steady state remains the ultimate goal of policy, private agents will not factor in the temporary increase in inflation into their price- and wage-setting behaviour. As a consequence, the central bank can afford a more measured conduct of policy in the face of a negative supply side shock.
Of course, this is not to say that central banks have carte blanche to adopt any tool or tactic that may appear suitable for dealing with any short-run challenge. The way policy is conducted can, in itself, impinge on the central bank’s credibility to meet its objective. One example is the monetary financing of fiscal deficits, which may severely undermine the capacity of a central bank to freely decide on an appropriate course of action. The flexibility of policy conduct is clearly limited to those measures that preserve the central bank’s room for manoeuvre to independently determine its monetary policy stance – now and in the future.
To summarise, for a central bank to be able to confront short-run challenges while anchoring long-run expectations, it must stick to three principles. First, it has to consistently and unambiguously align any policy action with a clearly specified objective. Second, it has to adjust its policy conduct flexibly to the specific nature of the macroeconomic shocks hitting the economy. And third, in doing so it has to design its policy conduct in a way that does not bind its hands and thus possibly jeopardise its medium-term policy objective.
Let me relate this analysis to the monetary policy strategy of the ECB.
As I mentioned earlier, the Treaty unambiguously establishes price stability as the primary objective of the ECB. On this basis, the Governing Council has communicated an explicit definition of price stability, namely a range of positive inflation rates below 2% over the medium term. This is a range at which inflation does not distort the allocation of resources in the economy. Since the inception of the euro the Governing Council has broadly fulfilled this mandate since the growth rate of the euro area harmonised index of consumer prices has been on average 2.07%.
This clear objective, together with the ECB’s track record, has firmly anchored inflation expectations in the euro area. Since 1999, the average inflation expectation by professional forecasters at a five year horizon has been on average 1.91%.
In our framework, inflation is obviously an indicator of pre-eminent policy relevance. But it is not the only one. In conducting its policy, the ECB has looked at a broader set of indicators which from time to time have released signals which conflicted with those which we extracted from our inflation forecast. Under our monetary pillar, in particular, monetary and financial trends have acquired such a prominent analytical status that, we think, they allowed a better identification of the origin of macroeconomic disturbances. This in turn has provided for a more robust approach to policy overall.
This monetary pillar “spare tyre” has been relevant in conditions of persistent supply shocks. Let me give an example. Since the 1990s and for much of the last decade, inflation in a large number of mature industrial economies remained quiescent. Many underlying determinants can explain these worldwide conditions. A regained control of inflation expectations by the major monetary authorities was certainly a prime cause of price stability. But other factors contributed to a sustained trend of disinflation. They include the delayed impact on productivity of the digital and internet revolutions that burst upon the scene in the early 1990s as well as the supply chain implications of China’s emergence as the world’s manufacturing powerhouse. Positive supply side shocks were either generated domestically or imported through an expanded trade intensity of production.
In the euro area, despite the upward pressures that the ongoing adjustment in relative price levels across Member States was exerting on the euro area price index, “price pressures”- defined as the difference between the year-on-year harmonised inflation rate observed in each quarter and the upper ceiling of our price stability range of 2% - were absent or even negative in the years preceding the Lehman debacle.  Had the ECB relied on inflation as a summary statistic of the state of the economy, the conclusion would have been that the economy had reached an ideal state of balanced, non-inflationary growth – the “Great Moderation”.
Our monetary pillar was sending a quite different signal. The growth rate of the broad monetary aggregate M3 doubled during the recovery that followed the bursting of the dot-com bubble. At the same time, premia paid by borrowers on various categories of debt contracts had plunged to levels unseen in the recent history of finance. Credit growth closely followed the rampant growth rate of M3. These indicators – which are an integral part of our monetary analysis – were sounding the alarm in an otherwise complacent policy environment. The resulting cacophony of indicators gave some headaches to the ECB staff who had to interpret them and also to policy-makers, who were forced to challenge the rosy outlook offered by the mainstream narratives of the time.
Now, in retrospect, we know that the undertone played by the monetary analysis, which appeared so much out of tune at the time, was in fact the counterpoint, the hidden score of financial instability. Lehman’s demise and what followed revealed the score of the hidden voice, and its drama.
The moral I draw from this episode is the following. The price stability objective needs to be reinforced, in the day-to-day conduct of policy, by an intense strategic focus on monetary and financial indicators. When financial markets create excessive leverage – too much credit – and engage in imprudent maturity transformation – too much bank money – reacting to financial booms under the monetary pillar will not only ensure price stability but also counteract the build-up of risk. Vice versa, very subdued credit dynamics, such as those we are currently observing, provide crucial signals of how long the accommodative monetary stance is warranted. This probably requires a better understanding of the ebb and flow of liquidity and the drivers of endogenous money creation inside the financial system.
But the possibility of activating this fail-safe policy mechanism largely depends on how the price stability objective is defined. The fact that inflation can move within a certain range is an important pre-condition. Indeed, the Governing Council’s specification of the ECB’s primary objective has two characteristic features, which complement its monetary pillar. First, it is a range of values, not a point target. Second, it is based on a medium-term orientation.
This has two implications. First, there is some flexibility as to where inflation could be within the range at any point in time. In 2003 the Governing Council announced that within the range of inflation rates consistent with price stability, a rate “close to 2%” should be the aim of policy in the medium term. But there is scope for inflation to vary within the range along the way to convergence with the medium-term policy aim, and many path configurations are possible. Second, the medium term itself is not a pre-set time span, but it is dependent on the underlying shocks that may explain inflation dynamics from time to time.
The former type of discretion – related to the path of convergence of inflation to its medium-term level – is critical when supply shocks are positive, that is, expansionary and disinflationary. This dimension of the ECB’s monetary policy strategy is not always appreciated, but is important. When inflation pressures are absent or negative and positive shocks to aggregate supply feed financial buoyancy – the conditions which prevailed in the late 1990s and for a good part of the 2000s – having a range definition of price stability, rather than, say, a point target, is essential. The possibility of letting inflation temporarily drift down within its price stability range when faced with clear signs of financial exuberance increases the chances of hitting the inflation objective over the long run.
The latter type of discretion – related to the time horizon – is crucial in the presence of transitory negative supply shocks and demand shocks. For example, a cost-push shock deriving from a spike in commodity prices pushes measured inflation upwards and is a drain on demand and economic activity. In the face of such a disturbance, the medium-term horizon affords some time for the shock to work its way through the economy and allows the ECB to “look through” it to a time when the shock is forecast to have subsided. This in turn ensures that the central bank is not forced to introduce further macroeconomic volatility.
As for a demand shock, the policy-relevant time horizon shrinks. A negative demand shock which is recessionary and potentially deflationary will have to be counteracted quickly and forcefully. This is in fact what the ECB did during the most acute phases of the financial crisis. This is what we have to do if confronted with severe downside risks to price stability.
Against this background, let me briefly come back to the comparison between the ECB’s monetary policy framework and that of the Federal Reserve System in the US. As mentioned earlier, the mandates of these two central banks differ: the ECB’s framework entails a clear ordering of objectives in which price stability ranks first, whereas the Fed’s dual mandate focuses on both price stability and maximum employment.
Some observers have invoked this difference in mandates to argue that the ECB is, by comparison, not well placed to support employment and growth in downturns. But, in fact, this argument overstates the practical difference in policy-making arising from these different mandates.
To see this, we should bear in mind that monetary policy can only support real economic aggregates in the short run. By contrast, long-term economic growth is determined by the structural characteristics of the economy, including the supply of labour and capital, as well as overall productivity. Any attempt by monetary policy to permanently lift the economy above the growth potential determined by these factors is bound to fail.
The Fed’s dual mandate is not at odds with this principle, I believe. As I read it, the employment dimension of the Fed’s dual mandate is not meant to influence the structural path of unemployment. In fact, as I said before, it serves as an explicit acknowledgement that the central bank will be wary of employment dynamics when steering output back to its structural path after economic shocks have perturbed labour market conditions. But, as I just explained, the ECB’s price stability definition and policy horizon also allow for substantial flexibility in calibrating the policy response to shocks. Due to this flexibility, the ECB can pursue its long-term price stability objective without triggering socially costly short-run fluctuations in output and employment. In this regard, it is in line with the key purpose of the Fed’s dual mandate.
In the euro area crisis, the ECB’s capacity to react flexibly within its stable framework has been tested exhaustively. One challenge that has rendered this crisis particularly complex is financial fragmentation between Member States. Two equally sound and productive businesses incur distinctly different costs of capital depending on whether they are located in an economically vulnerable country or in a more stable country.
However, funding conditions for firms and households constitute a key channel through which monetary policy impulses translate into consumption and investment decisions. As a consequence, changes in our main policy rates have had uneven effects across countries, thus undermining the singleness of monetary policy.
The ECB has responded to this with a range of non-standard monetary policy measures. In doing so, it has adopted solutions that have addressed the specific short-run challenges, while reconfirming its medium-term price stability commitment.
Let me give three examples. The two 3-year very long term refinancing operations (VLTROs) that we launched in late 2011 and early 2012 aimed to alleviate adverse funding conditions for banks which were unable to satisfy their additional liquidity needs in the market. By removing the liquidity constraints on credit supply, it has avoided a major credit crunch. Without the launch of the two refinancing operations, banks would have defaulted on their obligations coming due in the market, or would have discontinued and withdrawn existing credit lines to companies. The liquidity injection was on a large scale, amounting to around €520 billion – taking into account the shifting of liquidity out of other operations. It helped to counteract the fragmentation, in that a large share of the overall liquidity uptake under the VLTROs came from banks in vulnerable countries.
At the same time, the interest rate on these operations is indexed to the ECB’s main policy rate. Thus, if the ECB were to change this rate for monetary policy reasons, the costs for the remaining period of the three-year VLTROs would also change accordingly. Hence, the three-year liquidity provision which took place at times of extreme stress does not hamper the ECB’s capability to transmit changes in short-term interest rates.
In a similar vein, the programme for Outright Monetary Transactions has been designed with a view to combatting fragmentation while preserving the credibility of the ECB’s monetary policy objective. By being tied to an effective macroeconomic adjustment programme, the initiative prevents the intervention on government debt markets to eliminate “redenomination risk” – the unwarranted perception of a risk of euro breakup - from becoming a subsidy for unsustainable national policies. The European Stability Mechanism will impose on sovereign issuers the same solvency conditions as those imposed by the ECB on banks when it provides them with liquidity. And sovereign solvency is a necessary condition to avoid fiscal dominance on monetary policy, and thereby for the ECB to fulfil its medium term objective.
A third example is the ECB’s action to improve the transmission to the real economy of the impulsions provided by the single monetary policy. The broadening of the range of assets accepted as collateral to monetary policy operations makes it possible for euro area banks to better mobilise claims on the real economy, and therefore to lend more. The ECB is today exploring options to further strengthen lending to the real economy and, in particular, to small and medium-sized enterprises, which form the backbone of the euro area economy. But the ECB can only act within its mandate. Assuming non-profitable assets today present in the balance sheets of banks, or seeking to artificially eliminate differences in financing costs arising from different risk environment or from insufficient capital would amount to substituting actions not taken by other actors. This would be to the detriment of the central bank’s independence and freedom of action.
To sum up, the ECB has stringently followed the guidelines necessary to confront short-run challenges while anchoring long-run expectations, and it will continue to do so.
Let me conclude.
The crisis has exposed the balancing act lying at the heart of monetary policy. On the one hand, central banks are key participants in addressing short-term challenges that the economy faces. On the other hand, their crisis-fighting efforts have to be designed and communicated in full consistence with their long-run price stability mandates.
To successfully manage this balancing act, monetary policy-makers must clearly distinguish between their objective, which serves to firmly anchor market expectations, and their conduct of policy, which should flexibly adjust to the specific challenges at hand.
The ECB’s strategic framework contains the necessary elements to achieve this balance. Financial conditions have generally improved in the euro area over recent months. The ECB will continue to effectively accompany the recovery that should materialise provided that reform efforts continue both at the level of each euro area economy and to strengthen the governance and efficiency of our economic and monetary union.
I thank you for your attention.
I wish to thank Fédéric Holm-Hadulla and Massimo Rostagno for their contributions to this speech. I remain solely responsible for the opinions contained herein.
For an illustration, see figure 2.13 in Fahr, Motto, Rostagno, Tristani and Smets (2011), “A monetary policy strategy in good and bad times: lessons from the recent past”, ECB Working Paper n° 1336.
Reproduction is permitted provided that the source is acknowledged.