The ECB’s exit strategy

Speech by Jean-Claude Trichet, President of the ECBat the ECB Watchers ConferenceFrankfurt, 4 September 2009

Ladies and gentlemen,

It is a pleasure to be with you today to participate in this 11th edition of the ECB Watchers’ Conference. I am delighted to see that this conference has established itself so successfully as a venue for interaction between the European Central Bank and its observers.

My remarks today come in the context of the ongoing financial crisis, the most severe the developed world has witnessed for more than sixty years. As the Governing Council recognised yesterday, “there are increasing signs of stabilisation in economic activity in the euro area and elsewhere. This is consistent with the expectation that the significant contraction in economic activity has come to an end.” Nevertheless, it would be too soon to call the crisis over. The Governing Council also underscored that uncertainty remains high and that the recovery is expected to be rather uneven.

Exceptional times have called for exceptional actions. Faced with the financial crisis, we have introduced a set of non-standard measures, which we call “enhanced credit support”. These measures are intended to prevent the threat of a drastic loss of liquidity in the financial system as a whole and support the flow of credit to firms and households above and beyond what could be achieved through policy interest rate reductions alone.

Given their exceptional nature, these measures will have to be unwound once conditions return to normal. Of course, we designed technically the measures with their exit in mind, and today’s conference provides an opportunity for me to outline the overall exit strategy from the set of measures. To be clear, in this context, I would like to stress that, firstly, the term “exit strategy” should be understood as the framework and set of principles guiding our approach to unwinding the various non-standard measures. It does not include considerations about interest policy. Secondly, the discussion of the exit strategy should not be confused with its implementation. As I said, it would be premature to declare the crisis over. Now is not the time to exit. But I would like to make it clear that the ECB has an exit strategy, and we stand ready to put it into action when the appropriate time comes. My colleague Jürgen Stark, who will be on the first panel on the programme today, will go into more detail as regards issues related to timing.

Let me start by taking up things from where I left them a year ago.

When I addressed this conference at the beginning of September 2008, I reflected on the direction in which the economy was headed. I looked back over the preceding 12 months of financial turbulence, noting that this had marked the end of “a long phase of heightened risk tolerance”. At the end of my speech I expressed concerns that the correction of financial imbalances, and the associated re-pricing of risk, might not have run its full course.

As we now know, this was indeed the case. Only ten days later, the re-pricing in markets took a quantum leap. In mid-September the turbulence turned into a full-blown crisis. The chart of money market spreads tells this tale clearly.

[ Chart 1: Money market spreads]

As you can see, credit spreads soared. Financial market activity fell dramatically, volatility spiked and the global financial system came close to seizing up. Central banks stepped into the breach, addressing this unprecedented market failure. The ECB was in the vanguard of actions to contain these intensified tensions, taking a number of standard and non-standard measures. The measures by the leading central banks eventually paid off. As you see in the chart, money market spreads – a symptom of the market failures that I mentioned – retreated from their peaks. They are now back at levels comparable to those before the intensification of the crisis last year. But they still remain elevated in a longer time perspective.

The crisis triggered a freefall in economic activity. The chart shows that within a few months only, from January to July 2009, the projections for 2009 shifted massively to the left, in highly negative territory. After they centred around -1.5% in January 2009 and -3.2% in April 2009, they dropped to -4.5% in July, entirely outside the distribution range still a few months earlier.

[ Chart 2: The economic freefall]

One important element of the ECB’s actions has been the non-standard measures I have described as “enhanced credit support” ^[1]. This support collectively defines the special and primarily bank-based measures to enhance the flow of credit over and above what could be achieved through policy interest rate reductions alone, and avoid a systemic liquidity threat.

As you know, the ECB’s enhanced credit support comprises five elements:

Let me briefly recall the rationale underlying these five specific measures.

The first motivation for introducing enhanced credit support was to address the seizing up of the money market. If left unaddressed, this seizing up would have rendered the refinancing of many bank assets impossible, risking a massive and disorderly deleveraging by credit institutions. This, in turn, would have seriously hampered the monetary policy transmission mechanism and ultimately curtailed the expansion of credit to firms and households in the euro area. The core of the measures that we introduced in October was the “fixed-rate full-allotment” tender procedure. Banks were granted access to essentially unlimited liquidity at our policy interest rate in our regular open market operations. This support enabled them to maintain their crucial role in financing the real economy.

As a second measure, we expanded the collateral eligible in our liquidity providing operations. This offered an opportunity for banks to refinance assets that had become less liquid in the wake of the market turmoil. While the quantitative impact of this measure was not large in aggregate terms, it was important as it gave reassurance to banks that a larger share of their balance sheet could be refinanced at any time in case of need.

A third motivation was to give banks a more medium-term perspective in their liquidity planning, making them less vulnerable to short-term shocks and attenuating the maturity mismatch on their balance sheet. By giving banks a longer planning horizon we could further support their credit provision to the economy.

A fourth issue we needed to address was the currency mismatch in banks’ balance sheets and the malfunctioning of the international swap market. This was particularly relevant for US dollars, given the need of European banks to fund their dollar assets. We therefore began to supply liquidity to European banks in US dollars, backed by a swap facility established with the Fed.

Finally, the Eurosystem acted in financial markets directly. The Governing Council chose covered bonds due to their particular importance in Europe and the fact that, by carrying a signature from the issuing bank and the underlying asset, they are akin to our repo operations. Our aim was to have a catalytic role: we decided on a purchase programme with a volume that was significant enough to improve market functioning but not so large as to dominate the market as such. We also chose the covered bond market because we knew we could trigger new issuance activities, and thereby have a positive spillover effect on the overall financing of banks as well as on other market segments.

All in all, there are encouraging signs that these measures – in combination with the substantial interest rate reductions – are gradually bearing fruit. At the same time, there is no room for complacency and we are, as always, alert and prepared for any unexpected developments.

Having described the set of non-standard measures, let me now turn to the question of how to exit from them. The rationale for exit from these measures is the inverse of the rationale for their implementation. We have introduced exceptional measures under exceptional circumstances and will unwind these measures when the situation returns to normal and the rationale for the measures fades away.

Note that an exit strategy is not identical to a particular course of action. Rather, it lays out a framework and set of principles to govern actions in the face of circumstances in whatever form they take.

[ Chart 3: Definition of a strategy]

The distinction between a strategy and a particular course of action is well known in monetary policy. A monetary policy strategy provides the framework for assessing economic circumstances and responding to ever changing contingencies. But a strategy does not spell out interest rate decisions in the future.

In my view, the exit strategy from non-standard measures is comparable. As of today, we do not have all the information necessary to decide which exit path will be desirable or optimal from the perspective of tomorrow. But we have developed a set of criteria that will guide our future decisions. In this sense, exit considerations are subject to the same qualification that I always make when I speak on behalf of the Governing Council on monetary policy matters: we are not pre-committed.

Accordingly, as of today, we are not committed to a particular timing or sequence of actions for the exit phase. But our future decisions will not be arbitrary. They will be guided by a framework. In my view, four cornerstones have to be considered in our approach to exit from non-standard measures: the link to our monetary policy strategy; the forward-looking initial design of the measures; our technical and institutional ability to act; and our reputation for swift and decisive action when it is required.

[ Chart 4: Cornerstones of the exit strategy]

1. Link to the monetary policy strategy

The first cornerstone of the our exit strategy is the link to our primary objective and thus to our monetary policy strategy. Evidently, the crisis has not changed the ECB’s primary objective. Our actions have resolutely been taken with this primary objective in mind: to maintain price stability in the euro area over the medium and longer term.

By implication, any non-standard measure whose continuation would pose a threat to the achievement of price stability must be undone promptly and unequivocally.

At the same time, if no such risk existed, a non-standard measure could be maintained in case we continue to see significantly abnormal tensions, credit dislocations and impairment of the monetary policy transmission mechanism.

But financial institutions ultimately need to stand on their own two feet. Central banks can and have provided support to market functioning. But ultimately market activity relies on the confidence and trust of market participants themselves.

2. The design of the ECB’s measures

The ECB’s non-standard measures were designed with exit considerations in mind.

There was a strong consensus to embark on such policies that would prove unproblematic to unwind at a later stage. In this sense, the design of the non-standard measures was chosen to facilitate a smooth exit, representing the second cornerstone to be kept in mind.

This is reflected in two aspects of the design: the fact that a number of measures will phase out naturally; and the fact that the size and scope of the outright purchases for which exit would require either an unwinding or offsetting operations has been carefully calibrated.

a. Phasing out

A degree of phasing out has been built into the exit process through the design of our measures. In the absence of new policy decisions, many of the measures will naturally unwind over time. An overwhelming proportion of the additional liquidity provided during the financial crisis has been offered through repurchase agreements. Once these operations mature, their impact naturally phases out. Another example concerns the agreements with other monetary authorities to provide liquidity in foreign currencies. These agreements are priced in a way that makes it likely that they will face less demand once tensions in the market subside. Hence, here too a “natural phasing out” is envisaged, essentially reflecting lower demand.

b. Size and scope of the intervention

We have also carefully guarded against undertaking outright operations of a scope and magnitude that would make them difficult to offset or unwind. Notably, the covered bonds purchase programme is limited in size compared with the overall scale of our refinancing operations and compared with our balance sheet. Its effect on total liquidity can easily, if desired, be neutralised via adjustments to our monetary policy instruments.

3. Ability to act

Of course, notwithstanding these design features, exit will require a number of policy decisions. Therefore, an important question and the third cornerstone of our exit strategy is whether a central bank has the technical and institution capability to take such decisions.

a. Technical capability and the operational framework

On the technical side, the question is whether we need to set up new rules and procedures to exit. This is not the case. The ECB’s operational framework is well equipped to facilitate the unwinding of non-standard measures when the need arises. The framework embodies a rich and flexible set of instruments, including fine-tuning operations, for the absorption of abundant liquidity – promptly if necessary.

Moreover the framework permits short-term interest rates to be changed while keeping some non-standard measures in place, should continued credit support be needed. This feature is of great consequence: it means that the Governing Council can choose the way in which interest rate action could be combined with the unwinding of the non-standard measures. There is no pre-determined sequence between a change in interest rates and unwinding of non-standard measures. Hence, the Eurosystem retains appropriate flexibility as to the way in which interest rate action will be combined with the unwinding of the additional credit support measures, notably its extended framework for longer-term refinancing operations.

In short, the ECB has in place all the technical ability to absorb liquidity as required.

b. Institutional capability and independence

On the institutional side, the question is whether a central bank has the institutional status and necessary independence to carry out any action it sees necessary to fulfil its task. The Governing Council has the unfettered capability to take and implement appropriate policy decisions whenever circumstances warrant, given the ECB’s strong institutional independence. This reflects the maintenance of a clear dividing line in the euro area between the central bank and the fiscal sphere of responsibilities.

The fact that the ECB has refrained from purchasing government bonds is in line with this institutional framework. This approach avoids any possibly intricate interaction with other policy actors on decisions relevant for exit, and maintains a clear separation of responsibilities between the central bank and fiscal authorities.

4. Reputation to act when appropriate

Any strategy will only be credible if there is no doubt about the ex ante determination of the institution to implement the required actions. This is the fourth cornerstone. From this perspective, the ECB’s determination to act in a consistent way is beyond any doubt. Over the past decade, our institution has established a clear track record of vigilance and reputation for acting in a timely fashion.

Let me give only two examples: in December 2005 the Governing Council, on the basis of its monetary policy strategy, acted promptly to address mounting risks to price stability. Although many ECB watchers at the time favoured holding or reducing rates, on the basis of subsequent events it is widely recognised that this decision was appropriate and served to anchor longer-term private inflation expectations securely. A second example that underscores the ability to take decisions in difficult circumstances is the rise in interest rates in July 2008. It is now often put aside because subsequent developments were overtaken by the eruption of the crisis, but there is no doubt that inflation expectations were rising noticeably at the time. The decision of the Governing Council at the time demonstrated its determination to continue solidly anchoring inflation expectations including in difficult circumstances. At the height of the financial crisis, this reputation has helped to stabilise inflation expectations, including contributing to guard the euro area against the materialisation of the risk of deflation.

[ Chart 5: Inflation expectations in the euro area]

Allow me to conclude.

This morning, I have outlined four main cornerstones of the exit strategy from our non-standard monetary policy measures. They reflect: our reputation for swift and decisive action when it is required; our technical and institutional ability to act; the forward-looking initial design of the measures; and, above all, the link to our monetary policy strategy and the primary objective of securing price stability in the euro area over the medium and longer term.

Stressing the importance of the exit strategy should not be confused with its implementation. Notwithstanding some recent signs of improvement in the economic outlook, it is premature to declare the financial crisis over. From today’s perspective, the need for enhanced credit support remains.

Therefore, for the time being central bank liquidity is likely to remain abundant. This does not mean, however, that the broader concept of liquidity in terms of monetary aggregates grows strongly. And it is this broader concept, not central bank liquidity itself, which is related to inflationary pressures over the medium term. Of course, central bank credit and broader monetary aggregates are linked over time. In a modern banking system, and in normal circumstances, central bank credit is largely used by banks – in the form of required reserves – to create bank credit, and expand deposits and broad money aggregates. So, an increased provision of central bank credit can lead – more or less automatically – to an increase in broad money.

In current conditions, that link has been weakened, however. Today, banks – in the aggregate – tend to hold part of the central bank credit that they obtain in our refinancing operations not as a reserve against an expansion of loans and deposits, but in the form of excess reserves deposited with the Eurosystem. This attenuates the automatic link between the provision of central bank liquidity and broad money. And, as a consequence, it attenuates the inflationary potential of the Eurosystem’s outstanding credit to banks.

The exit strategy, in the end, will need to be invoked at the precise time in which the traditional link between broad money and our provision of liquidity to the banking system will re-establish itself. This calls for a constant and careful monitoring of the conditions at which the Eurosystem supplies central bank credit to banks in view of the evolution of the economy and of markets expectations.

Thank you for your attention.