Published by sunandoroy on Wed, 2008-11-05 12:43 ( first published in http://www.prmia.org)
The recent financial turmoil has highlighted the deep rootedproblems in the globally integrated interbank market. The recent loss of trust in the interbank market has contributed to bank collapses, huge erosion in market capitalisation and panic among investors. As we slowly and steadily restore a semblence of normalcy, it may be appropriate to take a close look at the crisis of confidence in the interbank sector and the possibilities of containing contagion triggered by this crucial segment of the global financial market.While the research below traverses the period prior to the global financial turmoil, it provides key insights to the contagion routes and possibility of restricting the cascade effect in an integrated financial landscape.
Risk Management in the Interbank Market:
The interbank market is part of the overall money market. Banks take on interbank exposures for two main benefits. First, banks need to pay out cash to customers on demand and to clear transfers of their customers’ deposits to other banks. Deviations of actual liquidity needs from banks’ expectations imply that banks may, ex-post, hold excess liquidity or need to obtain liquidity. Interbank markets are then used for risk sharing purposes, i.e. to manage bank-idiosyncratic liquidity shocks. Second, interbank markets may be used to hedge and transform other kinds of risks such as foreign exchange risk and interest rate risk. This allows efficient use of banks funds (Bhattacharya and Gale, 1987).
On the other hand, the interbank market is also a source of risk to the financial system. Interbank exposures may create problems if aggregate liquidity provision is insufficient. In this case, banks would try to avoid liquidation of their long-term assets, and would therefore liquidate their claims on other banks, often in other regions. A financial crisis in one region could then spread by contagion to other regions and thereby introduce liquidity problems in the latter (Alien and Gale, 2000). This risk is significantly controlled when central banks play a key role in preventing aggregate liquidity shortages.
A second source of contagion is the domino effect. The failure of one individual bank may initiate a domino effect if the non-repayment of interbank obligations by the failing bank jeopardisesthe ability of its creditor banks to meet their obligations to their (interbank) creditors. Contagion occurs then “mechanically” through the direct interlinkages between banks.
“Spill-overs” through market expectations represent a third potential channel for contagion. There may be a broad based withdrawal, as expectations become self-fulfilling (Diamond and Dybvig, 1983). Regulatory intervention such as suspension of convertibility or deposit insurance may alleviate the problem of bank runs and banking panics (Freixas and Rochet, 1997).
While the banking systems all over the world have introduced improved risk management practices to control risks in their balance sheets, risks from complex interbank transactions in the course ofliquidity management and derivatives trading have often been neglected aspects of the risk assessment process of the individual institution. The international experience with financial crises of the 1990s is a clear pointer to the fact that inter-bank exposures may generate cascading effects, where failure of one bank results in a failure of other banks which may not necessarily directly affected by initial shocks. This risk of contagion depends on the extent and pattern of exposures. In this context, it is important to understand and manage the embedded risks in inter-bank exposures. Ever since the bailout of the LTCM by the US Fed, the importance of assessing concentration risk arising from inter-bank exposures has received considerable attention. There have both been theoretical explorations and empirical analysis of risks associated with inter-bank exposures in different markets. In the cross country analysis presented below, a summary of findings of these studies are presented. We divide the section into the following parts. Part II presents the theoretical developments in containing risks in interbank exposures. Part III presents the findings of country specific studies. Part IV contains a summary of the lessons learnt from country experiences in managing risks in interbank exposures.
II Theoretical Contours of Risk Management of Interbank Exposures
Several studies have been conducted in the recent past to make formal assessments of the inter-bank exposures. The central concern of risks in the interbank market arises from the systemic implications of individual bank defaults through contagion and self fulfilling expectations. Hellwig (1997) pointed out that a complex network of interbank debtor/creditor relations may result in sophis¬ticated maturity transformation, which, in turn, at the level of the individual institution, may mask interest rate exposures of the banking system. Since it is hardly possible to assess the risk of a banking system based on the evaluation of individual banks, a “system approach” is called for. While risk management methods may certainly be suitable for individual credit institutions, regulators, concerning themselves primarily with the stability of the whole banking system, have to get a clear idea of the risk borne by the banking system. This is impor¬tant since a systemic banking crisis, i.e. a situation in which financial intermediation collapses at a large scale, translates into substantial costs to the real economy.
The attempts towards modeling risks in interbank market received a major boost with the development of the network model. The network model was intro¬duced to the literature by Eisenberg and Noe (2001), who present an abstract, static analysis of a clearing problem. In the model, the structure of claims and liabilities of the banking system is shown as a matrix. Eisenberg and Noe (2001) presented consistent generalized clearing payment vectors thereby enabling scenario analysis. This further enables estimation of default frequencies, the loss given default and the contagious insolvencies. Using the relative frequencies of the individual events across the various scenarios, we may then conduct probability estimations.
Elsinger, Lehar and Summer (2002) developed the method of entropy optimization. This method attempts to distribute the mass of the row and column totals in such a way across the cells that the sum conditions are fulfilled and that as much consistency as possible is preserved with the a priori information about the unknown cell entries.
Market and Liquidity Risk
While interbank exposures have generally been associated with credit risk, there is a strand of literature that stresses that interbank exposures are not just a source of credit risk but also of market (price) risk and liquidity risk. One important conclusion of this strand of thought is that prudential regulation (in the form of minimum capital requirement ratios or other solvency constraints) when combined with mark-to-market rules can sometimes generate undesir¬able spillover effects. Marking to market enhances transparency but it may introduce a potential channel of contagion and may become an important source of systemic risk. Liquidity requirements can mitigate contagion, and can play a similar role to capital buffers in curtailing systemic failure. It seems intuitive to conjecture that when players are faced with illiquid markets, they would try to insure against liquidity holes by holding more liquid assets.
Huizinga and Nicodeme (2001) find that non-bank international deposits are positively related to wealth taxes. This suggests that non-bank international deposits are in part determined by tax concerns. With regard to international interbank deposits, the tax treatment of deposits also undoubtedly plays a role, although it is not the sole driver. Moshirian and Bishop (1997) show that international interbank movement of funds were determined, among other things, by the relative cost of capital (which is affected by differences in tax treatments) between countries
III Interbank exposures: Cross Country Experience
Research on interbank markets have mostly been done on advanced financial markets, who function in a manner similar to interbank markets in emerging economies, but differ on the extent of internationalization of financial markets. Select country experiences with interbank market are discussed below.
Austria
A research project conducted by the Economic Studies Division of the Oester-reichische Nationalbank (OeNB) and the Center for Business Studies of the University of Vienna was aimed at finding (a) the risk of interbank loans at a system level, accounting explicitly for complex credit chains/interdependencies, (2) making optimal use of the data sources as they normally exist in central banks. The framework consists in a network model of the interbank market. Based on specific assumptions about the resolution of insolvencies, the model endogenously explains the possible payment flows among banks in different future states of the world (scenarios) for a given structure of interbank liabilities and for a given structure of other bank assets and liabilities. The states of the world are described by the impact interest rate changes, exchange rate and stock price fluctuations as well as credit defaults have on diebanking business. The network model explicitly determines the possible interbank payments for each state of the world.
Using a cross-section of Austrian banks as at September 2001, the study found the Austrian banking system as broadly stable and the like¬lihood of systemic banking crises extremely low. The scenarios were created by exposing various balance sheet items to risk factors. In each scenario banks face gains and losses derived from market and credit risks. The median default probability of Austrian banks was below 1 per cent. (Elsinger, Lehar and Summer (2002)).
The methodology for the assessment of credit risk is as follows: (1) First, the balance sheet item “claims on nonbanks” is decomposed in line with the Major Loans Register data into several exposures to industries.
(2) Second, on the basis of the credit rating data, the estimated default frequency and its standard deviation is computed.
(3) In this way the necessary parameters for the individual credit portfolios and calculate a distribution of default losses for each bank is arrived at.
(4) The distribution then yields default loss scenarios generated by combining historical simula¬tion and the credit risk model (Elsinger, Lehar and Summer (2002)).
Luxemburg
Banks in Luxembourg have been historically active in interbank markets. Interbank markets are those through which Luxembourg banks’ linkages with foreign financial intermediaries are the strongest. In particular, Luxembourg’s large banks are very active as providers of funds to the interbank market as well as to the international groups they belong to. An important channel through which an international shock is likely to be transmitted to Luxembourg banks is through their credit relationships with correspondent banks, many of which are the parent banks of a group to which a Luxembourg subsidiary belongs. There are therefore, potential vulnerabilities in interbank relationships.
Empirical studies on the Luxemburg interbank market essentially concentrated on two alternate scenarios : (a) where defaults in interbank market create contagion affecting other Banks; and (b) where the interbank market shocks are free from contagion. The results of the tests, in terms of number of banks resulting under-capitalized as a result to sets of shocks to financial groups, reveal that the interbank market in Luxemburg exhibited stability. Even under the contagion scenarios, the number of undercapitalized banks was restricted to only two banks. In sum, the exercise revealed some concentration of interbank exposures for only two banks whose capital base might result in being more severely eroded than their peers. Overall, Luxembourg banks appear at present fairly resilientto shocks originating in major financial groups to which they belong and/or are most connected through the interbank market.
Belgium
The Belgian interbank market is international in character and highly concentrated. Degryse and Nguyen (2003) make an empirical examination of the systemic risk from inter-bank exposures in the Belgian banking system. They note that inter-bank exposure amongst Belgian bank was small at 15 per cent, suggesting that potential contagion risk arises mainly from foreign interbank exposures. The study is based on scenario analysis for contagion on historical data (using the maximum entropy distribution). The study showed that, over the last decade, the worst-case scenarios in the case of contagion triggered by a Belgian bank have been subject to three major changes. Between 1992 and 1997, the worst-case scenario consistently wors¬ened. During this period, the share of interbank assets in total assets tended to increase. This amplified the exposure of Belgian banks to other Belgian banks and increased the potential con¬sequences of contagion in the worst-case scenario. Between 1997 and 1999, the worst-case scenario improved. Since 1997, mergers may have had an impact on the worst-case scenario. The decrease over time in medium-sized players, which were large enough to cause other banks to “fail” in the contagion exercise, also dampened the contagion effect observed over time in the simulations. Thus, since 1999 the risks seem to have stabilized.
Moreover, following consolidation, large banks have further increased their cross-border interbank exposures. About 85 per cent of Belgian interbank loans are granted to foreign banks. Foreign interbank positions thus represent a potential source of contagion that may be more important than domestic contagion risk. The results of this section suggest that, in the Belgian con¬text, the international risk of contagion may deserve more attention than domestic contagion risk.
US
Furfine (2001) test the performance of interbank market during Autumn of 1988 in the backdrop of the Russian sovereign bonds default and the near collapse of the LTCM. The paper finds that the Federal Funds market was largely unaffected during the period and channeled liquidity to institutions in need at rates consistent with the target rates. Further, the risk premiums also remained stable even as lending volumes increased. Thus, he stressed the appropriate role of central banks in the management of risks in interbank markets.
Interestingly,Chevallier-Farat (1988) observed a linkage of developing offshore markets and heightened interbank exposures in the US. He noted that the creation of off-shore areas in the United States in 1981 (International Banking Facilities) triggered massive movements of international interbank funds.
UK
Wells (2002) brings to fore the systemic risk implications of the inter-bank exposures in the UK. Loans, CDs and reverse repos provided for the bulk of such exposures. They argue that even though large UK banks enjoyed good credit ratings, the implied pattern of large exposures suggest that insolvency of a large foreign bank could cause multiple bank failures in UK.
Chile
Cifuentes (2003) observe that 1990s has seen increasing bank concentration in Chile, which has led to increased systemic risk in inter-bank markets. Conducting simulations the paper reports measures on bank failures and damaged assets. Regulatory measures are also proposed to contain such risks.
Netherlands
Lelyveld and Liedorp (2004) study interbank contagion in Dutch banking sector. They conduct scenario analysis to measure contagion risks and find that bankruptcy of one large bank will put a considerable burden on the other banks, but would not lead to a collapse of the interbank market.
Germany
Upper and Worms (2004) estimated bilateral exposures in the German inter-bank market. They found that in absence of safety nets there was considerable scope of contagion, while institutional guarantees for saving banks and cooperative banks substantially reduces but does not eliminate the danger of contagion. They show that a failure of single bank could lead to the breakdown of up to 15 per cent of the banking system in terms of assets.
IV
Lessons from Cross Country Experience
The cross country experience throws up a number of issues. The central message that comes forth is that the Central Bank should be alert and observant to mitigate the risks that may arise from the interbank segment. From the cross country experience, the following risk mitigation techniques seem to be useful in reducing risks in interbank market .
First, the interbank market is not an isolated market. It is linked to other segments of the financial sector and macroeconomy. Macroeconomic shocks through their influences on liquidity and asset prices influence the interbank market and change its risk profile.
Second, opening up of the external sector may lead to growing cross-border interbank exposures and can increase risks in interbank markets. The level of internationalisation of the interbank market heightens the risks of interbank exposure. Exceptionally high proportions of cross-border interbank loans and deposits highlight a feature of the interbank market in developed countries which potentially transforms the risk of contagion, as well as the way it should be handled. Given that the lion’s share of the inter¬bankexposures are situated abroad in advanced countries, banks are more sensitive to international crises than to domestic ones, and any attempt to assess the impact of interbank markets on financial stability must be viewed in that perspective. India is fast integrating with the global economy, and the lessons from the global economy may be instructive.
Third, international experience shows that mergers reduce risks in Interbank markets. The cross country literature reveals that the interbank market risk reduces when mergers and consolidation takesplace. The world over, mergers are taking place in the banking sector at a rapid pace. This is likely to reduce the risk profile of interbank markets.
Fourth, government tax policies in domestic markets often lead to flow of funds towards international interbank markets, with heightened credit and liquidity risks.
Fifth, maturity structure of inter-bank liabilities bears a close relation to risk. If interbank loans and deposits show a relatively short maturity, banks use interbank markets mainly to manage their short-term liquidity needs, so the risks are limited.
Institutional Mechanisms towards Risk Mitigation
Internationally, several institutional mechanisms were found to be useful in reducing the risks in interbank market.
First, the existence of collateralized interbank market (a well developed and liquid repo market) reduces the risks of contagion. However, the existence of a repo market may lead to the disappearance of the uninsured international interbank market (Freixas and Holthausen, 2001). This can occur as a result ofasymmetric information; a bank that attempts to obtain an unsecured cross-border loan may be suspected of having had the loan denied by other domestic banks which have more information about the borrower.
Second, the use of netting contracts among banks is a mechanism for reducing interbank exposures. A problem at one bank is then less likely to initiate a “domino effect” on the interbank market. Emmons (1995), however, shows that netting of interbank claims shifts the bank default risk away from interbank claimants towards non-bank creditors, i.e. the risk is transferred to the banks’ creditors who are not included in the netting agreement.
The Central bank has a major role to play in reducing the risks in interbank market. Potential central bank intervention, as well as the presence of safety nets, lowers contagion risk in the interbank markets . There are several forms of central bank intervention to contain interbank exposure risks. They are :
(i) Liquidity Management
Central banks may decide to provide liquidity to the market as a whole when aggregate liquidity is insufficient, or directly to individual banks when the market fails to provide liquidity to sound financial institutions. Moreover, although interbank exposures are not explicitly covered by deposit insurance, issues such as “too-big-to-fail” may introduce implicit deposit insurance for these exposures.
(ii) Containing Market Expectations
Central Banks also play a crucial role in arresting “Spill-overs” of market expectations. Regulatory intervention such as suspension of convertibility or deposit insurance or even active communications policy may alleviate the problem of bank runs and banking panics (Freixas and Rochet, 1997).
(iii) Limits to large counterparty exposures:
Limits imposed by authorities on banks’ large exposures and proper monitoring and control of large exposures of credit institutions contribute to reducing contagion risk. Limits are usually formulated in terms of banks’ own funds.
(iv) Challenges to Banking Supervision
A regulator entrusted with the mandate of financial stability needs to differentiate between fundamental instability and systemic instability. Banks may default as a direct consequence of shocks (fundamental insolvencies), but also due to a chain reaction, i.e. because other banks defaulted in the first place. The latter cases may be regarded as insolvencies caused by chain reactions (contagious insolvencies). It may be noted that Banking supervision, which focuses on the individual institutions, has no in-built mechanism to arrest systemic insolvencies and thus inadequate to detect such risks. In this context, the need for a separate mechanism to monitor the systemic risks emanating from interbank markets can hardly be underscored.
(v) Effective use of data on Interbank Exposures
Also relevant is to devise a mechanism for optimal use of interbank data, including creation of database, regular monitoring, relevant reporting and assessment of risks. Periodic stress testing by the central banks may be necessary in this regard. In this context, it may be useful to further explore application of a network model in the Indian context by using advanced matrix analysis and entropy optimization.
References
Alien, F. and D. Gale (2000). ‘Financial Contagion’, Journal of Political Economy, Vol. 108, No 1, pages 1-33.
Banque Centrale de Luxembourg (2000), “Evolution du Marche Interbancaire a Luxembourg”, Bulletin, n°2.
Bernard, H. and J. Bisignano (2000), “Information, Liquidity and Risk in the International Interbank Market: Implicit Guarantees and Private Credit Market Failure”, 6/S Working Paper, n°86.
Bhattacharya, S. and D. Gale (1987), “Preference Shocks, Liquidity and Central Bank Policy”, in Barnett and Singleton: “New Approaches to Monetary Economics”.
Chevallier-Farat, T. (1988), “Le Role du Marche Bancaire International”, Revue d’Economie Politique, n°5, pp. 673-699.
Cifuentes, R. (2002), ‘Banking Concentration and Systemic Risk’, pre¬sented at the Annual conference of the Central Bank of Chile, December, forthcoming in the conference volume.
Degryse, H. and G. Nguyen (2003), “Interbank exposures: an empirical investigation of systemic risk in the Belgian bank¬ingsystem”, NBB Working Paper, forthcoming.
Deutsche Bundesbank (2000), “Longer-term trend in German Credit Institutions’ Interbank Operations”, Monthly Report, January 2000.
Diamond, D. and P. Dybvig (1983), “Bank Runs, Deposit Insurance, and Liquidity”, Journal of Political Economy 91 (3), pp. 401-419.
Eisenberg, L. and T. Noe (2001), ‘Systemic Risk in Financial Systems’, Management Science, Vol. 47, No. 2, pages 236-49.
Elsinger, H., A. Lehar and M. Summer (2002), “The Risk of Interbank Credits: a New Approach to the Assessment of Systemic Risk”, presented at the 17th Annual Congress of the European Economic Association.
Emmons, W. (1995), “Interbank Netting Agreements and the Distribution of Bank Default Risk”, The Federal Reserve Bank of St. Louis, Working Paper Series, Working Paper 016-A.
Freixas, X. and C. Holthausen (2001), “Interbank Market Integration under Asymmetric Information”, European Central Bank, Working Paper Series, Working Paper n°74.
Freixas, X. and J.-C. Rochet (1997), “Microeconomics of Banking”, Cambridge, The MIT Press.
Freixas, X., B. Parigi and J.C. Rochet (2000), “Systemic Risk, Interbank Relations and Liquidity Provision by the Central Bank” Journal of Money, Credit and Banking 32 (3), pp. 611-638.
Furfine, C. (1999), “Interbank Exposures: Quantifying the Risk of Contagion”, BIS Working Paper, N°70.
Furfine, C. (2001), “The Interbank Market during a Crisis”, BIS Working Paper, No. 99.
Hellwig, Martin, 1997. “Banks, Markets, and the Allocation of Risks in an Economy,” Sonderforschungsbereich 504 Publications 97-35, Sonderforschungsbereich 504, University of MannheimSonderforschungsbereich 504, UniversitätMannheim. (available at Hellwig page in IDEAS)
Huizinga, H. and G. Nicodeme(2001), “Are International Deposits Tax-driven?”, Economic Paper, European Communities, Number 152.
Moshirian, Fariborz and Robert Bishop. 1997, “International Business: Determinants of Interbank Activities.” Journal of International Financial Markets, Institutions and Money, v. 7, no. 4, December: 329-49.
Upper, C. and A. Worms (2002), “Estimating Bilateral Exposures in the German Interbank Market: Is there a Danger of Contagion”, Deutsche Bundesbank Discussion Paper 9.
Wells, S (2002), “UK Interbank Exposures: Systemic Risk Implications”, Bank of England Financial Stability Review, December 2002, pp. 175-182.
Leave a Reply