Abstract
This Selected Issues paper analyzes euro area policies and discusses the implications of the 2007–08 financial sector turbulence for real economic activity. It examines the linkages between the financial and real sectors in the euro area. The paper discusses the European Central Bank’s (ECB) monetary analysis and the role of monetary aggregates in central banking, surveying the ongoing theoretical and empirical debate. The paper also describes the introduction of a “European Mandate” for financial sector authorities in the European Union (EU), a proposal that is under consideration by EU member states.
III. Liquidity Management in the Euro Area22
A. Introduction and Main Findings
53. The financial market turbulence has highlighted the relative robustness of the euro–area monetary policy implementation framework, while exposing potential weaknesses in the risk management framework of individual institutions. The turbulence has tested the mechanisms of liquidity creation and transmission at the global, country, and individual institution level, and it has arguably been the first major test of the euro–area liquidity management framework. The Eurosystem’s relatively high reserve requirements, combined with the averaging provisions, acted as a buffer during the turmoil. The wide range of collateral and banks eligible to participate in the Eurosystem operations has limited the risk of disruptions. The collateral framework has provided an important automatic stabilizer to the financial system. Maintaining the separation between monetary policy objectives and liquidity operations has been useful, but it has put an increased premium on clear communication with the market and the public.
54. The ECB’s actions have been widely praised, but the exceptional monetary operations have also highlighted some open issues. Those can be grouped into two broad categories: (i) linkages between liquidity management and monetary policy implementation; and (ii) linkages between liquidity management and financial stability. The existing literature tends to focus on either the supervisory side or the monetary policy implementation side while this chapter treats these two sets of issues as interlinked, addressing them in an integrated fashion.
55. There is a scope for some minor adjustments. For example, narrowing down the interest rate corridor for steering the money market rate could provide a clearer signal to the market. Also, should the financial turmoil worsen appreciably, consideration needs to be given to less conventional measures, such as lowering the rate on the marginal lending facility, further lengthening the maturity of auctions, lowering haircuts on collateral, and (in the extreme) outright purchases of assets; however, it should be noted that such steps have drawbacks that, outside extreme conditions, predominate the benefits.
56. The collateral framework should feature incentives for improvement in the credit quality of collateral after the financial stress is over. The leeway left to counterparts to accumulate good quality collateral and shift more risky collateral to the ECB allows the framework to operate counter–cyclically. However, the incentives should work both ways so that the quality of collateral reverts back to a neutral average in good times.
57. The flexible liquidity management framework at the systemic level should be complemented by stricter liquidity management practices in individual institutions. This chapter finds a case for reviewing liquidity regulations and supervisory practices, which still differ significantly across the euro–area countries, more than is the case for capital regulations and supervision. Regulators need to ensure that financial institutions rely less on recent correlations, and incorporate more severe liquidity gapping and correlation jumps in their market risk models and stress tests. There is also scope for reducing uncertainty through increased transparency about risk management practices, and about the products being used to manage liquidity. Furthermore, this chapter argues for clarifying the links between emergency liquidity assistance and supervisory intervention, based on an EU–wide approach.
B. Distinguishing Monetary Policy Stance and Liquidity Operations
58. The ECB has responded promptly and forcefully to the financial turmoil. Since the onset of the turmoil, the volatility of euro money market rates has increased sharply, and spreads between overnight and longer–term interest rates have risen. The ECB has responded by special liquidity operations (Appendix I), aiming to avoid large deviations of the overnight rate from the policy rate, and also to calm tensions and boost confidence in money markets. The following should be noted (Figure 1):
During all months but December 2007, the average overnight rate stayed within ±15 basis points of the policy rate set by the Governing Council at 4 percent since June 2007 (there was more volatility in day–to–day values of the overnight rate).
There have not yet been obvious signs of aggregate, economy–wide liquidity or money demand shocks (M1 growth decelerated further and M3 growth remained at an elevated level).
The total volume of the ECB’s refinancing operations did not exceed trend growth (except for a brief period at the end of 2007), but the maturity profile lengthened.
59. The ECB’s liquidity operations had some success in limiting the “noise” in themoney market. A key stated reason for the special operations was to provide confidence to the markets in a situation of a sudden increase in liquidity demands from European banks, which pushed the EONIA rate substantially above the policy target on several occasions. An econometric analysis using a GARCH model (Figure 2) suggests that the operations have been successful to the extent that after an initial increase in the deviations of the EONIA rate from the policy rate, both these deviations and their volatility has declined towards pre–crisis levels. The deviations of the EONIA from the policy rates (plotted here in the text chart) remained somewhat above the pre–crisis levels; the difference, while statistically significant, was quantitatively small. The same is correct for the conditional volatility estimated by the GARCH model (Figure 2). The stabilization of the EONIA has been achieved by an increased activity in the open market operations: one indicator is the ratio of the open market operations to the reserve requirements (Figure 3), which has been characterized by significantly increased volatility since August 2007.
Euro Area: Money Market Rates and Monetary Aggregates
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: European Central Bank.Euro Area: Money Market Rates and Monetary Aggregates
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: European Central Bank.Euro Area: Money Market Rates and Monetary Aggregates
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: European Central Bank.
Euro Area: Volatility in EONIA and Open Market Operations
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: IMF staff calculationsEuro Area: Volatility in EONIA and Open Market Operations
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: IMF staff calculationsEuro Area: Volatility in EONIA and Open Market Operations
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: IMF staff calculations
Open Market Operations and Reserve Requirements
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: European Central Bank, and IMF staff calculationsOpen Market Operations and Reserve Requirements
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: European Central Bank, and IMF staff calculationsOpen Market Operations and Reserve Requirements
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: European Central Bank, and IMF staff calculations
Deviations of the EONIA from the ECB policy rate
(histogram)
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: IMF staff calculationsDeviations of the EONIA from the ECB policy rate
(histogram)
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: IMF staff calculationsDeviations of the EONIA from the ECB policy rate
(histogram)
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: IMF staff calculations60. At the same time, in the unsecured 3–month interbank money market, spreads with overnight interest rates have remained at unusually high levels. The spreads between a 3–month deposit rates and overnight rate swaps of the same maturity have become lower in the euro area compared to the United States and the United Kingdom (Figure 4). However, a more detailed econometric examination of the effects of these measures finds that the ECB action (as well as the Fed and Bank of England actions), while helping to reduce the money market volatility, may have had only a small impact on these spreads (IMF, 2008 and Box 1).
Spreads Between 3–Month Libor Rates and Overnight Interest Rate Swaps
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: Bloomberg LP.Spreads Between 3–Month Libor Rates and Overnight Interest Rate Swaps
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: Bloomberg LP.Spreads Between 3–Month Libor Rates and Overnight Interest Rate Swaps
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: Bloomberg LP.61. Communicating the distinction between liquidity operations and monetary policy implementation has been a challenge. In this regard, the emphasis in communication on the role of monetary aggregates in guiding policy decisions may have complicated the ECB’s task relative to that of other central banks. It has been challenging for the ECB at times to communicate that in conditions of pronounced uncertainty, the supply of abundant liquid instruments was aimed not at altering the stance of monetary policy, but at ensuring the regular functioning of the markets (in particular, preventing large deviations of the EONIA rate from the policy target and providing confidence for somewhat longer term maturities). Indeed, no additional liquidity was provided on net terms, when considering average liquidity provided over the reserve maintenance period (also, the ECB did not soften eligibility criteria for the collateral it accepts). More importantly, even if additional liquidity had been provided on net terms, this should not have been interpreted as a change in the monetary policy stance as the policy stance is enforced by controlling the overnight interest rate in money markets while money supply, more or less, adjusts endogenously. The desirability of policies to offset temporary shocks to money demand that are unrelated to total output has been pointed out e.g., by Diamond and Rajan (2006).
High Spreads in Interbank Markets: Counterparty Risk or Liquidity Premium?
Increasingly large term spreads in interbank money markets, adjusted for interest rate expectation effects, are a defining feature of the financial turmoil that began in the summer of 2007. Do they reflect changes in counterparty risk, or are they signs of increased uncertainty, funding pressure for banks, and liquidity hoarding?
CDS spreads, as a measure of counterparty risk, and interbank market spreads tell different stories: both have gone up in the early part of the turbulence, but while the interbank market spreads have remained high, CDS spreads have narrowed markedly since March 2008. The decline in CDS spreads coincides with the Federal Reserve organized rescue of Bear Sterns and the Fed’s introduction of a new lending facility to improve the ability of primary dealers 0 to provide financing to participants in securitization markets. This has clearly alleviated markets’ concerns about possible bank failures.
Costs of insurance against banks’ default 1/
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
1/ 5‐year senior CDS spreads, average. U.S.: Citigroup, Bank of Americ Wachovia, Wells Fargo, Washington Mutual. Euro Area: Deutsche Ban BNP Paribas, Credit Agricole, Societe Generale, Unicredito, Banca MD U.S. banks a, JPMorgan Chase, k, HVB, Commerzbank, P di Siena.Source: Datastream.Costs of insurance against banks’ default 1/
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
1/ 5‐year senior CDS spreads, average. U.S.: Citigroup, Bank of Americ Wachovia, Wells Fargo, Washington Mutual. Euro Area: Deutsche Ban BNP Paribas, Credit Agricole, Societe Generale, Unicredito, Banca MD U.S. banks a, JPMorgan Chase, k, HVB, Commerzbank, P di Siena.Source: Datastream.Costs of insurance against banks’ default 1/
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
1/ 5‐year senior CDS spreads, average. U.S.: Citigroup, Bank of Americ Wachovia, Wells Fargo, Washington Mutual. Euro Area: Deutsche Ban BNP Paribas, Credit Agricole, Societe Generale, Unicredito, Banca MD U.S. banks a, JPMorgan Chase, k, HVB, Commerzbank, P di Siena.Source: Datastream.The recently diverging trends between CDS and interbank money market spreads could indicate further market segmentation, or liquidity factors (e.g., increased funding pressure on those banks that were already too reliant on wholesale funding). Michaud and Upper (2008) use the CDS as a proxy for counterparty risk and decompose the spread between 3–month Libor and the overnight (OIS) rate into credit risk and liquidity risk. They approximate the latter as the residual in their calculations.
A more direct way of measuring liquidity risk is using the spread between the rates of the ECB’s MROs and LTROs. Reflecting the design of these refinancing operations, the spread should include very little or no credit risk. This spread was particularly elevated at the end of 2007; it fell through early March but picked up since then (see text chart). These developments reflect the following sequence of events: at the end of 2007, an already severe liquidity pressure combined with some purely technical (end–year) issues related to liquidity management in financial institutions. The ECB’s shift to longer–term refinancing operations has contributed to the fall observed in January and February 2008, even though this effect has likely been relatively small, given that the repurchase agreements account for only 2 percent of total liabilities (consolidated) of euro area monetary financial institutions. In the wake of the Bear–Sterns failure and also pushed by regulators, banks have become increasingly unwilling to accept maturity mismatches at the short end for fears of running into liquidity problems, which accounts for the increase in the liquidity premium since early March 2008.
Spread between LRO and MRO adjusted for 3‐M OIS
(in basis points)
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Spread between LRO and MRO adjusted for 3‐M OIS
(in basis points)
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Spread between LRO and MRO adjusted for 3‐M OIS
(in basis points)
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
C. Scope for Modifications in the Liquidity Management Framework
62. There is some, but limited, scope for further modifications in the current Eurosystem liquidity management framework. As indicated above, the monetary operations/liquidity management framework has so far proven rather robust during the turbulence. Volatility in the EONIA has been contained successfully. Interbank–OIS spreads have remained at elevated levels, but most likely for reasons such as counterparty credit risk, which are largely outside of the control of the central bank. The following is a discussion of the measures that the are within the central bank’s control.
Lengthening further the maturity profile of refinancing operations
63. The ECB’s actions can do relatively little about the high spreads between longer–term and overnight rates. The empirical analysis of the recent episode suggests that one important factor of the high term spreads in money markets is the perceived counterparty (credit) risk. There is very little a monetary authority can do about this, short of bail outs of institutions at risk, including outright purchases of bank assets that could otherwise be sold on to markets only at a steep discount.
64. The lengthening of the maturity profile of ECB’s refinancing operations likely had some tempering effect on longer–term rates, but it has complicated short–term rate management. The amount provided through LTROs almost doubled from euro 150 billion to euro 290 billion while the amount provided through MROs fell from around euro 300 billion to euro 170 billion (Figure 1), resulting in a lengthening of the maturity profile (text figure). This had a tempering effect on term interest rates (Box 1), but at the same time may have complicated to some extent the management of the EONIA. This was evidenced for example in December 2007, when the average difference between the EONIA and the policy rate was –12 basis points, compared to +6 to +7 basis points in “normal times” (ECB, 2008). Under current conditions, the costs (e.g., in terms of lower control over the EONIA) of dropping further the amount available for MROs may well outweigh the benefits of lengthening of the maturity profile of refinancing operations.
Average maturity of open market operations
(weeks)
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: IMF staff calculations on ECB data.Average maturity of open market operations
(weeks)
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: IMF staff calculations on ECB data.Average maturity of open market operations
(weeks)
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: IMF staff calculations on ECB data.D. Narrowing the Interest Rate Corridor
65. One possible adjustment is to narrow the range that the standing facilities setaround the policy rate(currently ±100 basis points). This would lower the penalties charged for use of the standing facilities, effectively narrowing down the corridor for the EONIA rate, and likely making the money market rate less volatile in periods of increased uncertainty. The ECB’s interest rate corridor is relatively wide in international comparison, with some advanced economy central banks using corridors as narrow as ±25 basis points.23
66. An analysis of the EONIA rates suggests that the interest rate corridor would have to be very narrow to become binding. For example, narrowing down the corridor to ±75 basis points (the same as used, for example, in Sweden) would not have made much difference in the recent turbulence, as the actual observations since August 2007 have ranged from –46 to +58 basis points. Even a ±50 basis point corridor would cover all but one observation from the recent turbulence (i.e., more than 99 percent of the observations). Narrowing it down to ±25 basis points (as used, for example, in Canada or Australia) would still cover 92 percent of observations, i.e. EONIA would be inside the corridor about 92 percent of the time (and the ratio might be even higher if the ECB used more active money market operations to steer the EONIA within the corridor).24
67. Narrowing the interest rate corridor involves some trade–offs. It would likely limit EONIA volatility in stressful periods, and reduce the need for special liquidity operations. This would come at the cost of lower flexibility and increase the money market’s reliance on central bank funding (which has, however, been very low). Another risk is that if the interest rate corridor becomes too narrow, a large investor may benefit from building up a futures position first and trading subsequently in the spot market while using the central bank facilities. However, the incentives for such speculation are low unless the corridor is very narrow, and it can be shown (Ewerhart and others, 2004) that the probability of manipulation decreases when the central bank uses an active liquidity management.
68. On balance, narrowing the corridor could perhaps be useful, but it is not a panacea. If there were fundamental reasons for banks to be concerned about solvency of their counterparties, this measure alone would not be able to prevent banks from distrusting each other and the interbank market, especially at longer–term maturities, from breaking down. However, a moderate narrowing of the corridor is unlikely to do much harm, especially not in normal times, while helping to provide a clearer signal about the ECB’s desired interest rate in times of uncertainty. The balance of the arguments, and the international experience, point toward some scope for narrowing the interest rate corridor.
E. Targeting the 3–Month Interbank Money Market Rate
69. Targeting directly the 3–month money market rate is an option, but with unclear benefits. Some observers (e.g., Morgan Stanley 2008) have recently argued that the ECB should follow the example of the Swiss National Bank and target the 3–month interbank money market rate instead of the overnight rate. They point to the fact that the 3–month rate is more closely correlated with bank lending rates which are crucial in the transmission of monetary policy. One could take the argument even further and argue, for example, that one–year lending rates have a much stronger impact on consumption and investment demand than 3–month rates, and could be more relevant for the monetary transmission than 3–month rates. However, given the structure of the markets at the longer maturities, it would be very difficult for any central bank to control, with market means, rates at such maturity. There are also technical reasons why targeting a 3–month or longer rate may be difficult: Libor or Euribor type of references are declarative and not actually traded, which raises question as to the extent to which they are reflective of market conditions, particularly in stressful times. In the end, it is mostly a technical issue which maturity a central bank decides to target: longer maturities may be more economically relevant, but the longer the maturity, the more difficult it is to effectively steer a rate. The case for targeting rates at longer maturities is not clear at present. Moreover, the ECB can take into account changes in the spread between the overnight and three–month rate when setting its policy rate. 25
F. Adjusting the Collateral Management Framework
70. The collateral framework has proven to be one of the few automatic stabilizers in the financial system. The Eurosystem’s range of collateral acceptable for routine repurchase agreement operations is broader than that of the U.S. Federal Reserve or the Bank of England.26 The wide range of acceptable collateral diminishes the Eurosystem’s control over the structure of the collateral pool, but it has important advantages. Namely, it makes the collateral framework responsive to market innovations, and it acts as a countercyclical stabilizer in stressful times. This latter role has been at display in the recent period: even though the general framework has not been changed during the financial turmoil, the quality and liquidity of collateral posted (slowly deteriorating already before the turmoil) has substantially decreased. The share of government bonds has come down (currently about 15 percent, compared to their 50 percent share of the stock of eligible collateral, and compared to the 60 percent share in pledged collateral in 1999) in favor of primarily bank bonds and, to a lesser extent, asset–backed securities. Banks that post lower–quality collateral reportedly tend to bid higher rates at auctions, reflecting the lower opportunity cost of the collateral.
71. The decreased quality and liquidity of the posted collateral is not a problem per se. Combined with a deteriorating credit quality of the counterparties,27 these developments shift risks to the ECB, and increase the potential for moral hazard. However, the increasing risks are acceptable if they are aligned with the Eurosystem’s risk tolerance. Indeed, in a well– designed collateral framework, collateral quality can be expected to deteriorate in a stressful financial environment (and this can be dealt with via the risk mitigation measures present in the ECB framework).
Structure of the collateral pool
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: IMF staff calculations, based on ECB data.Structure of the collateral pool
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: IMF staff calculations, based on ECB data.Structure of the collateral pool
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: IMF staff calculations, based on ECB data.72. However, current collateral developments are appropriately subject to close monitoring. There is a need for the Eurosystem to refine its collateral policy, not only to ensure that the quality of new collateral does not fall below the Eurosystem’s risk tolerance level, but also to control the market impact of the collateral framework and to limit the risk of market distortions (this includes, for instance, limiting the “manufactured collateral,” i.e. the risk that banks could use the cross–placement of bank bonds to artificially create eligible collateral). Keeping track of market developments in European fixed–income markets is challenging, since markets are undergoing rapid structural changes and products are tending to become ever more complex. A recent example has been the rapid developments in markets for structured finance and asset–backed securities (some of which are currently illiquid).
73. It is particularly important to ensure that the quality of the collateral pool starts to improve again when the financial stress subsides. However, the current collateral framework does not seem to feature sufficient incentives for any substantial recovery in collateral quality (Chailloux and others, 2008), although only time will tell.
74. A key medium–term challenge may very well therefore be to make the framework truly countercyclical. Possible steps involve adjusting haircuts, or, if that is not sufficient, introducing higher interest rates when lending against lower quality collateral. In a situation of abundant collateral, the latter measure is likely to be more biting. 28
75. If current financial market conditions were to deteriorate appreciably further—which is not the central scenario—additional, normally very costly measures may have to be considered. These measures include softening the collateral framework by broadening even further the allowable collateral (this would mean extending it beyond the “A” rating), lowering the haircuts, or (in the extreme) even outright purchases. All these measures have substantial negative side–effects and associated costs that in most circumstances outweigh their benefits. For example, lower haircuts or a broader collateral list would increase the volume of collateralized operations, but they would mean bigger exposures for the Eurosystem, and lower incentives for banks to find financing from other banks.29
G. Liquidity Management and Financial Stability
76. The turbulence highlights the case for strengthening liquidity regulations for financial institutions. Central bank liquidity operations are only one of the key elements of systemic liquidity management; the first line of defense against excessive risk taking and financial distress should be sound risk management practices in financial institutions and market participants. Most of the existing liquidity regulations date back to more than a decade ago, a fact that, by itself, would warrant a review, given the dramatic changes that took place in the financial system in the last decade. The recent turmoil provides another reason for assessing whether the existing liquidity rules can be amended in a way that, if not eliminating such episodes, can at least limit their likelihood and impact.
77. Increasing the resilience to liquidity stress should involve making liquidity management in financial institutions more forward–looking and less reliant on recent correlations. An important part of the approach should be for regulators to ensure that financial institutions incorporate correlation jumps in their market risk models and stress tests. Also, to eliminate information asymmetries, it is important to reduce uncertainty through standardization of securitized products and improvements in the rating system (e.g., IMF, 2008).
78. There is a case for aligning more closely liquidity regulations in individual euro–area countries. In contrast to the internationally harmonized capital–adequacy requirements and accounting standards, the regimes for supervision of banks’ liquidity still vary considerably across countries, even in the euro area. This results, among other things, in a variety of year–end and quarter–end effects. The various liquidity regulation regimes in the euro area differ in the relative mix of approaches (quantitative and qualitative) that they use. Quantitatively–oriented liquidity regimes focus on compliance with required liquidity indicators, while qualitatively–oriented liquidity regimes are based on a dialogue between supervisors and banks’ managers about the banks’ liquidity management procedures and systems. In recent years, there has been a marked shift towards more qualitative approaches (Basel Committee, 2008), driven by the increased complexity and sophistication of liquidity management in individual banks. Also, within the quantitatively–oriented systems, there has been a move from stock–based liquidity indicators (e.g., ratio of liquid assets to short–term liabilities) to mismatch–based indicators (e.g., liquidity–at–risk). These developments puts much more premium on consistent supervisory implementation across countries.
79. In principle, there is a clear distinction between the collateralized provision of liquidity, emergency liquidity assistance (ELA), and solvency support. ELA is the support given by central banks in exceptional circumstances, on a case–by–case basis, to temporarily illiquid institutions and markets. The main guiding principle is that the competent NCB takes the decision about providing ELA to an institution operating in its jurisdiction. This takes place under the responsibility and at the cost of the NCB in question.30 Mechanisms are in place to ensure an adequate flow of information so that potential spillovers can be managed in a manner consistent with the maintenance of the appropriate single monetary policy stance (ECB, 2000). In particular, individual NCBs are required to report to the Eurosystem about ELA usage.
80. In practice, in a crisis situation, the distinctions between collateralized support, ELA, and solvency support become very blurred. In practice, there is often uncertainty about the solvency of the institutions involved, and it takes time to make a solvency assessment. Recent trends in the financial system, such as consolidation, emergence of global financial conglomerates, and growth of complex financial products make these tasks even harder. While requests for liquidity support may arise very rapidly, it is unlikely that central banks or supervisory authorities will be able to make a valid assessment of the solvency of troubled institutions quickly enough. As a result, it is practically very difficult to draw the dividing line between the central bank’s responsibilities and those of other authorities by distinguishing between institutions with solvency problems and those with pure liquidity problems, at least during a sudden crisis. For similar reasons, it may also be difficult to ascertain the quality of collateral in a crisis situation, which makes the distinction between collateralized liquidity support and ELA blurred. Moreover, there may be incentives for individual NCBs to use collateralized liquidity support rather than ELA, since the potential costs of the collateralized liquidity support are shared in the Eurosystem, while the costs of ELA are borne by the individual NCBs.
81. An important step towards improving the EU’s capacity to prevent, manage, and resolve financial crises would therefore be to put in place a functioning EU–wide framework for pre–crisis sanctions and tools. A key part of this framework would be a scheme prescribing mandatory supervisory actions at certain “trigger points,” which should be the same for all euro–area financial institutions, and cover both liquidity and solvency concerns.31 Designing such a system is nontrivial, particularly with respect to the trigger points, and needs to strike a delicate balance between rules and discretion, considering that each financial crisis is likely to be different. However, in a cross–country setting the case for a more rules–based approach to dealing with financial stress is even stronger than in a single–country setting, where it significantly rests on limiting politically–motivated regulatory forbearance and limiting costs of failures to the public purse: it is key to lessening banks’ scope for prudential arbitrage, establishing trust among supervisors, and distinguishing more clearly solvency support and the various forms of liquidity support.
82. Making the framework work would require a consensus on the ultimate policy objectives, including the degree to which governments are willing to be exposed to contingent claims for solvency support. This would involve a basic political agreement. It would also require at least some degree of harmonization of countries’ legal frameworks for bank resolution. These are thorny issues on which some work in on–going but much more remains to be done.
Appendix ECB’ Monetary Policy Implementation
In the ECB’s monetary policy framework, the main operating target is the euro overnight index average (EONIA), a measure of the effective interest rate in the euro area overnight market, calculated by the ECB on a daily basis as a weighted average of the interest rates on unsecured overnight lending transactions denominated in euro (e.g., ECB, 2008). The ECB’s monetary operations are geared towards limiting the deviations of the EONIA rate from the policy rate set by the ECB’s Governing Council. The volume of central bank liquidity is in turn determined endogenously by the market demand for money at given EONIA levels.
The key instrument of the ECB’s monetary operations framework are main refinancing operations (MROs), conducted on a weekly basis with weekly maturity. The ECB also conducts fine–tuning operations (FTOs), and longer–term reserve operations (LTROs). To stabilize money market interest rates and create (or enlarge) a structural liquidity shortage, the Eurosystem maintains a minimum reserve system, characterized by relatively high reserve rates and long reserve maintenance periods (one month). A large number of banks are eligible counterparties for Eurosystem monetary policy operations (potentially 1,700 banks compared to only 20 primary dealers in the U.S.; see, e.g., ECB 2006a and IMF, 2008). For small banks, participation in the weekly MROs could be relatively costly as they may not need to adjust their reserve holdings that frequently. For this reason, the ECB has also been conducting the LTROs, with a 3 month maturity. At this maturity, the ECB has offered relatively small fixed amounts until mid–2007 (euro 50 billion during February–August 2007, or about 15 percent of total reserve operations) and acted as a price (rate) taker.
Use of the marginal facilities
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: IMF staff calculations, based on data from the European Central Bank.Use of the marginal facilities
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: IMF staff calculations, based on data from the European Central Bank.Use of the marginal facilities
Citation: IMF Staff Country Reports 2008, 263; 10.5089/9781451879896.002.A003
Source: IMF staff calculations, based on data from the European Central Bank.The marginal deposit and lending facilities provide overnight liquidity by offering access on any regular weekday at the discretion of banks, and effectively set a floor and a ceiling for the EONIA. Rates are set at the policy rate minus or plus a margin of 100 basis points, respectively. Reflecting the wide margins, the average daily use of these facilities has been less than 1 percent of banks’ reserve requirements (see the text chart).
This framework has enabled the ECB to react flexibly to the recent turmoil in financial markets. The large number of banks eligible to participate in the Eurosystem operations has limited market segmentation across banks, which could become a serious problem during times of financial stress. The wide range of collateral, while not without potential problems, has also helped in avoiding possible forced asset sales in markets that experienced a sudden and sharp drop in turnover that may have triggered significant bank losses.
When faced with the financial turbulence, the ECB, together with the other major central banks, intervened repeatedly from the first half of August 2007 to satisfy the demand for liquidity and curb the divergence of very–short–term interest rates from the official rate, resorting most frequently to fine–tuning and 3–month operations.
In mid–December 2007, the ECB extended the duration of the main refinancing operation, supplying an exceptionally large volume of liquidity. This was a part of a coordinated set of measures undertaken together with the U.S. Federal Reserve, the Bank of Canada, the Bank of England, and the Swiss National Bank, all aimed at addressing the elevated pressures in short–term funding markets. Actions taken by the Federal Reserve included the establishment of a temporary Term Auction Facility (TAF) and the establishment of foreign exchange swap lines with the ECB (and the Swiss National Bank). Also in agreement with the Federal Reserve, the ECB offered loans in dollars to euro–area counterparties. In March 2008, the ECB went even further in extending the maturity of its refinancing operations by introducing a new “supplementary” LTROs with six month maturity. 32
An important feature of the ECB response to the turbulence has been the increased use of LTROs, which has resulted in a substantial lengthening of the average duration of its lending. The duration profile has been increasing already before the financial turbulence, but it was late 2007 and early 2008 when the structure switched substantially (Figure 1; bottom panel).
Another important part of ECB’s response to the financial turbulence has been to alter the time pattern of liquidity provision during the reserve maintenance period. The ECB has provided relatively more liquidity in the early parts of the reserve maintenance periods, and correspondingly less in the latter part of the period (to keep unchanged the total volume of liquidity provided over the maintenance period). It has been implemented mainly to accommodate banks’ desire to frontload reserve fulfillment. This was a helpful step that illustrated the benefits of the Eurosystem’s flexible reserve requirement system that combines relatively high remunerated reserve cushions with long reserve maintenance periods (Bindseil, Gonzalez, and Tabakis (2008).
References
Basel Committee on Banking Supervision, 2008, “Liquidity Risk: Management and Supervisory Challenges,” February (Basel: Basel Committee on Banking Supervision).
Bindseil, Ulrich, F. Gonzalez, E. Tabakis, 2008, Risk Management for Central Banks and Other Public Investors (Cambridge: Cambridge University Press), forthcoming.
Borio, Claudio, and William Nelson, 2008, “Monetary operations and the financial turmoil,” BIS Quarterly Review, March (Basel: Bank for International Settlements).
Chailloux, Alexandre, Simon Gray, Ulrich Klueh, Seiichi Shimizu, and Peter Stella, 2008, “Central Bank Response to the 2007-08 Financial Market Turmoil: Experiences and Lessons Drawn,” IMF Working Paper (Washington: International Monetary Fund).
Čihák, Martin, and Thomas Harjes, 2008, “Bridge Over Troubled Waters: Liquidity Management in the Euro Area,” Draft IMF Working Paper.
Diamond, Douglas W., and Raghuram G. Rajan, 2006, “Money in a Theory of Banking”, American Economic Review, Vol. 96 (1), pp. 30- 53.
Eisenbeis, Robert A., and George G. Kaufman, 2006, “Bank Crisis Resolution and Foreign–Owned Banks”, Federal Reserve Bank of Atlanta Economic Review, Fourth Quarter, pp. 1–18.
European Central Bank, 2000, Annual Report 1999 (Frankfurt: European Central Bank).
European Central Bank, 2008, “The Analysis of the Euro Money Market from a Monetary Policy Perspective,” ECB Monthly Bulletin February 2008 (Frankfurt: European Central Bank), http://www.ecb.eu/pub/pdf/mobu/mb200802en.pdf.
Ewerhart, Christian, Nuno Cassola, Steen Ejerskov, and Natacha Valla, 2004, “Sporadic Manipulation in Money Markets with Central Bank Standing Facilities,” ECB Working Paper No. 399 (Frankfurt: European Central Bank).
Fitch, 2008, “Outlook for Major European Banks in 2008 - More Stormy Waters Ahead, “ Special Report, available at fitchresearch.com.
International Monetary Fund, 2008, Global Financial Stability Report (Washington, DC: International Monetary Fund), April.
Mayes, David G., Liisa Halme and Aarno Liuksila, 2001, “Improving Banking Supervision” (Basingstoke: Palgrave).
Michaud, François–Louis, and Christian Upper, 2008, “What drives interbank rates? Evidence from the Libor panel,” BIS Quarterly Review, March.
Prepared by Martin Čihák and Thomas Harjes. For details, see Čihák and Harjes (2008).
For instance, central banks in Australia, Canada, and New Zealand have operated standing facilities with rates ±25 basis points relative to the policy rate, and Sveriges Riksbank has used a ±75 basis point corridor. The U.S. Fed’s lending facility had a +100 basis point spread before the recent turbulence; but since then the Fed has lowered the spread to +50 basis points (the Fed has no deposit facility).
In principle, it would be possible to move to an asymmetric corridor. However, symmetric corridors are not prevalent in advanced economy central banks, and the benefits of an asymmetric corridor are unclear.
For instance, before the onset of the turmoil, markets widely expected the ECB to raise its policy rate by 50 basis points to 4.5 percent during 2007, but the ECB decided to keep the policy rate unchanged.
The current collateral framework relies on a single list of eligible assets (currently numbering around 24,000), published daily on www.ecb.int.The list includes asset–backed securities and debt instruments issued by corporations and others. The collateral has to meet “high credit standards,” normally defined as an “A” rating.
The counterparty risk has deteriorated substantially, as indicated by the indicators of bank soundness, including credit default spreads, and by developments in banks’ ratings and rating outlook (e.g., Fitch, 2008).
The ECB’s current framework provides liquidity via OMO operations at the same price to all counterparts, irrespectively of the credit quality and liquidity of the underlying collateral. Haircuts and margin calls are used, but these measures are not biting in a situation of abundant collateral.
Additionally, extending the available collateral much further could run into inconsistency with the requirement that the Eurosystem provides liquidity to the banking system only “based on adequate collateral” (ECB Statute, Article 18.1), and it would also be incompatible with the principles of transparency, equal treatment, and accountability, since it would entail much more discretion in the management of counterparty risk.
See, e.g., ECB (2000). In the euro area, the ECB does not have direct ELA functions; those are in the hands of the national central banks (NCBs).
Such schemes are sometimes referred to as “structured early intervention and resolution” (SEIR). Based on Mayes, Halme and Liuksila (2001) and Eisenbeis and Kaufman (2005), one can identify the following characteristics of an efficient SEIR: (i) the prudential authorities need to act as soon as a solvency shortfall or other warning signals are detected; (ii) if there is no improvement after the grace period, a capital injection should be imposed; (iii) if no private sector solution has been found and solvency drops below a certain level or another trigger point is met, there should be a mandatory and prompt suspension of shareholder rights, the bank resolution agency should take custody or receivership of the bank, and new management should be put in place; (iv) in custody or receivership, the bank resolution agency needs to make a quick early assessment so as to allow continuity in the bank’s core operations and minimal or no disruption in the availability of most deposits; (v) systemic and core operations of the bank, including basic retail services, should continue uninterrupted or after a minimal interruption not exceeding one or two days; (vi) the reopened bank should be recapitalized, restructured and prepared for sale to private acquirers within a relatively short time period.
