Central bank policies in and out of the crisis: From non-conventional to new conventional

Speech by José Manuel González-Páramo, Member of the Executive Board of the ECB,
at the “Challenges to monetary policy implementation beyond the financial market turbulence Workshop”
Frankfurt am Main, 1 December 2009

I can hardly think of anything good to come out of the financial crisis, with the few exceptions of the high-quality papers presented at this workshop, the great interest that it has received from academic economists and the central banking community, and the intensity and thoughtfulness of the debates it has triggered over the last day and a half here in Frankfurt. I therefore congratulate you all for having contributed to the success of this venture and I am looking forward to following up on your research work as it may help to further improve the design of our institution’s operational framework and the quality of our policy decisions.

In my concluding remarks, I wish to touch upon two topics: 1) the impact of non-conventional monetary policy measures; and 2) the (near) future of monetary policy implementation in the euro area. I shall refer, where appropriate, to the papers presented at this workshop.

Aims of the non-conventional monetary policy measures

Let me first briefly recall what the non-conventional monetary policy measures have aimed to – and still aim to – achieve. The measures taken by the Eurosystem were:

  1. Fixed rate tenders with full allotment in all liquidity-providing operations

  2. Introduction of additional refinancing operations with one-month and three-month maturities, as well as the provision of funding at longer maturities (six months and one year)

  3. Broadened collateral framework (lowering of rating threshold to BBB-; acceptance of selected foreign currency assets and of securities issued in some non-regulated markets)

  4. Covered bond purchase programme (up to EUR 60 billion)

  5. Introduction of foreign currency-providing operations

Taking into account the balanced outlook for the risks to price stability, at present the measures aim to reduce specific distortions in the financial sector by narrowing exceptionally wide risk or term spreads and by increasing the credit flows in particular markets from depressed levels. Moreover, the measures aim not only to reduce the cost of credit, but also to increase the flow of credit to businesses and households.

Indeed, as documented in the interesting paper by Hempell and Sorensen [1] presented at this workshop, supply-side constraints on lending may have emerged during the crisis, related to banks’ (lack of) access to wholesale funding and to their stressed liquidity positions. Non-conventional monetary policy measures aim to mitigate, among other things, these distortions.

Another problem that central bank actions have tried to address from the beginning of the crisis was the hoarding of liquidity by counterparties, which severely disturbed the functioning of the money market, as shown by Acharya and Merrouche [2] for the UK and by Ashcraft et al. [3] for the US. The re-emergence of segmentation in the euro interbank market was also an issue calling for central bank action. This topic is discussed in several papers presented in this workshop, including those co-authored by C. Holthausen and N. Cassola [4]; F. Castiglionesi, F. Feriozzi and G. Lorenzoni [5]; M. Bruche and J. Suarez [6]; and N. Cassola and M. Huetl [7].

Thus, mitigating the malfunctioning of money and financial markets was, among other reasons, the motive behind the non-conventional monetary policy measures enacted by the ECB/Eurosystem. Research papers like those that I just referred to are now carefully documenting the emergence and impact of such distortions and drawing the relevant policy implications.

When considering whether or not to enact non-conventional monetary policy measures, attention had to be paid to their possible negative side effects in financial markets. While non-conventional measures were introduced to address market distortions, ultimately they themselves could distort credit allocation and market functioning. A key challenge for the central bank thus was – and still is – to assess what constitutes “normal” or “sustainable” market conditions. There is a challenging trade-off between doing too little and doing too much.

Have the non-conventional monetary policy measures worked?

Assessing non-conventional central bank measures is very difficult. It requires considering the context in which they were implemented, their impact on the problem they were supposed to address, and their possible side effects.

The ultimate yardstick for the effectiveness of non-conventional policies is their impact on aggregate spending and price stability. There are encouraging signs on both of these fronts: according to the most recent data, the euro area economy is now technically on its way out of recession and a global economic recovery seems to be taking off. At the same time, inflation expectations, in particular in the euro area, are well anchored and consistent with our price stability mandate. Nevertheless, given the time-lags of monetary policy, any firm conclusions about the macroeconomic effects of non-conventional monetary policy measures would seem to be premature at the current juncture. Moreover, as always, the counterfactual is unknown, and modelling the effects of non-conventional measures is even more difficult than for conventional policies.

Assessing the immediate effect of non-conventional measures undertaken in the interbank market (longer-term operations, broader collateral, etc.) is challenging. Non-conventional refinancing operations have often been introduced incrementally. These measures have become part of day-to-day central bank operations. Moreover, they have in many cases been implemented against the backdrop of important support measures taken by the governments (e.g. capital injections to individual institutions, guarantee schemes for financial firms’ debt) which also played a critical role in alleviating market tensions. Hence it is difficult to disentangle the contributions of the different types of non-conventional policy measures to the overall impact, including for the refinancing operations.

Notwithstanding these caveats, several papers presented at this workshop took up the challenge and tried to overcome the methodological difficulties by using a skilful combination of analytical tools, informed institutional knowledge, and careful event studies, focusing on specific operational targets and markets. In my view, these approaches are very promising.

For example, non-conventional monetary policy measures contributed to improving the functioning of the commercial paper market as shown by Hirose and Ohyama [8] in the case of Japan, and by Duca [9] in the case of the US.

For the euro area, Donati [10] and Nobili [11] explain and show the positive impact of ECB/Eurosystem measures in compressing money market spreads, namely by affecting liquidity risk components along the money market yield curve.

Moreover, as shown by Vaidyanathan [12], it is possible to identify a positive effect on conditions in foreign exchange swap markets after the announcement and implementation of swap lines between central banks.

Thus I am very glad to learn from you in this workshop that a positive assessment can be made at this juncture.

A topic that could not be discussed by the papers presented at this workshop is the impact of the covered bond purchase programme (CBPP) of the Eurosystem. The programme was announced back in May and was launched in July with a duration of up to one year. To encourage research in this field, let me give you some very preliminary, encouraging evidence of the positive impact of the programme. First, primary market issuance recovered significantly in the euro area following the announcement of the programme. Second, the upward trend in covered bond spreads was reversed and in some cases significant declines in spreads can be observed. Third, one of the main objectives of the programme is to support bank financing to the economy. It will be very important to analyse the effects and the mechanisms of the CBPP in these three respects.

Let me now turn to the last topic of my talk.

Exiting from non-conventional measures

My colleague Jürgen Stark recalled yesterday the main principles that guided the operational framework of the Eurosystem and that must be preserved in the future. I will not repeat them now. These principles have contributed to a clear signalling of the policy stance, low volatility of the overnight interest rate and the smooth provision of refinancing to the banking system. They have also contributed to lower term premia. There is reason to believe that they will continue to be useful in the future.

However, let me add a few further considerations.

The phasing-out process will be gradual and, towards its end, the average maturity of ECB/Eurosystem refinancing operations will be much shorter than it is today. With improving market functioning, lending at term maturities by market participants will resume. Thus, not all the term-lending operations carried out by the central bank will be needed to the same extent as in the past. A timely and gradual phasing-out should avoid distortions associated with maintaining non-standard mesures for too long.

Nevertheless, the exact allocation of the future share of refinancing among the MROs is still an open question; and the option of adjusting the future regular operations in light of the experience acquired with the non-conventional operations should not be excluded at this juncture.

This topic remains largely unexplored in the academic literature. A first attempt at tackling them has been made by Eisenschmidt and Holthausen [13] in their paper presented at this workshop. I would encourage some of you to further explore this highly relevant field.

In this respect, I can thus say that some elements of the liquidity policy measures of today may belong to the new conventional central bank toolkit of tomorrow.

An important aspect of the exit process relates to the return to variable rate tenders in the refinancing operations of the ECB/Eurosystem. A key advantage of the variable rate format – besides allowing the central bank to steer the liquidity volume in the desired direction – is the incentive that it provides for counterparties to resume pricing funding liquidity risk appropriately. Indeed, one important lesson from the crisis, which cannot be ignored, is that the longer-term refinancing operations of the central bank play at times the role of a liquidity insurance mechanism; therefore, the spread between the marginal rate of these tenders and the respective OIS rate can be viewed as a liquidity insurance premium and its dynamics over time and dispersion across banks may provide useful information to the central bank about emerging tensions in funding markets. The liquidity insurance mechanism associated with the longer-term refinancing operations has to be viewed in conjunction with the collateral framework of the central bank. Research on these mechanisms from an insurance theory viewpoint is therefore very welcome.

Let me conclude by making a few remarks on the role of the collateral framework.

From the very early stages of the crisis, counterparties were able to explore the funding possibilities offered by the collateral framework of the Eurosystem. These opportunities were furthered by the temporary broadening of this framework which was announced in October 2008.

The possibilities offered by the broad collateral framework, and their interaction with the generous conditions of term funding provided by the ECB/Eurosystem (fixed rate full allotment), allowed counterparties to fund assets whose market liquidity had evaporated.

Thus a broad collateral framework has merits as an immediate, quasi-automatic crisis mitigation tool. The potential welfare effects of the broad collateral framework are discussed in a formalised manner by Ewerhart and Tapking [14] in their paper presented at this workshop.

However, taking a more pragmatic view, the broad collateral framework poses some challenges going forward.

First, the acceptance as collateral of assets for which markets remain seriously dislocated increases liquidation risk and, in the case of “own use” of secured assets, also the concentration risk assumed by the central bank on its balance sheet. Hence, broad collateral frameworks require constant monitoring; furthermore, the risk control framework needs to be constantly adjusted in order to counteract unwarranted practices in the use of collateral by counterparties.

This requires a very sound information base and high level of human resources for monitoring financial market innovations, for developing pricing models and for refining risk control measures. The ECB/Eurosystem has been aware of an increase in residual financial risks in the context of its collateral framework since the outbreak of the financial market crisis, and has continuously redefined its risk control framework to manage this risk.

Second, as the Bank for International Settlements warned in its 2008 Annual Report, the large-scale intermediation of ailing capital markets by the central bank may create price distortions in the longer term. Making a wide range of liquid and illiquid assets eligible for central bank refinancing may – if not adjusted for by the central bank via risk control measures and adequate pricing policy – lead to a preferential treatment of illiquid assets relative to liquid ones, raise the relative price of illiquid assets and lead to oversupply and a consequent impact on credit allocation.

Third, a collateral framework that allows the own use of secured instruments in credit operations with the central bank as a permanent feature could reduce incentives for bank issuers to revive their third-party investor base and to reactivate markets.

Last, a broad collateral framework risks distorting the incentives for banks to manage liquidity risk properly, by allowing them to replace highly liquid assets such as government bonds with illiquid assets.

How to preserve the features of an effective, immediate crisis-mitigating tool, while at the same time containing unwarranted market distortions and not diluting incentives for prudent risk management, is an important area for future analysis and policy research. In this respect, I warmly welcome the papers presented at this workshop devoted to discussing some aspects of this difficult topic, such as those by Cheun et al. [15] and Fegatelli [16].

If a new conventional perspective on monetary policy implementation were to emerge where liquidity insurance is explicitly recognised, articulating collateral extensions with standing facilities and/or longer-term refinancing operations, the key challenge would be to design it in an operationally feasible way, and in such a way that market functioning would be preserved, while providing the right incentives for prudent liquidity provisioning and management by banks.

Let me once again thank you all for having contributed to the success of this workshop.

[1]H. S. Hempell and C. K. Sorensen, “The impact of supply constraints on bank lending in the euro area – crisis-induced crunching?”

[2]V. V. Acharya and O. Merrouche, “Precautionary hoarding of liquidity and interbank markets: evidence from the sub-prime crisis”.

[3]A. Ashcraft, J. McAndrews and D. Skeie, “Precautionary reserves and the interbank market”

[4]C. Holthausen and N. Cassola, “The 2007/2009 turmoil: a challenge for the integration of the euro area money market?”.

[5]F. Castiglionesi, F. Feriozzi and G. Lorenzoni, “Financial integration, liquidity and the depth of systemic crisis”.

[6]M. Bruche and J. Suarez “Deposit insurance and money market freezes”.

[7]N. Cassola and M. Huetl “The euro overnight interbank market and the ECB’s liquidity management policy during tranquil and turbulent times”.

[8]Y. Hirose and S. Ohyama, “Identifying the effect of the Bank of Japan’s liquidity facilities: the case of CP operations during financial turmoil”.

[9]J. Duca, “Preventing a repeat of the money market meltdown of the early 1930s”.

[10]P. Donati, “The effect of conventional and unconventional monetary policy on the money market spreads in the euro area”.

[11]S. Nobili, “Liquidity risk in money market spreads”.

[12]K. Vaidyanathan, “Quantifying financial market turbulence through deviations in covered interest parity”.

[13]J. Eisenschmidt and C. Holthausen, “The minimum liquidity deficit and the maturity structure of a central bank’s open market operations: lessons from the financial crisis”.

[14]C. Ewerhart and J. Tapking, “Repo markets, counterparty risk and the 2007-2009 liquidity crisis”.

[15]S. Cheun, I. von Koeppen-Mertes and B. Weller, “The collateral frameworks of the Eurosystem, US Federal Reserve and Bank of England and the financial market turmoil”.

[16]P. Fegatelli, “The role of collateral requirements in the financial crisis: one tool for two (diverging) objectives?”

Media contacts