The impact of systemic risk depends very much on the collective behavior of financial institutions and their interconnectedness as well as on the interaction between financial markets and the macro-economy. Elements of interconnectedness can generally be measured by consideration of counterparty risks related to a financial institution's activities. Knowing the interconnectedness of a financial institution could enable a systemic risk regulator to determine how many additional failures could be caused by the failure of an individual firm.1
The theme of interconnectedness is still in its nascent stages, having garnered broad attention by the financial industry, regulators, and academia only since the Credit Crisis. The reason for the latter is attributable largely to lessons learned from the failure and/or government bailout of financial firms such as Lehman Brothers Inc. and American International Group in 2008.
As interested parties analyzed the risks posed by these firms during the Credit Crisis, it became very clear that the real risks posed to the global financial system did not only originate from the stand-alone size or direct counterparty exposures posed by such firms. Instead, it was only after the massive size and complexity of the direct and indirect financial, operational, and legal connections of Lehman and AIG became more clear that the true systemic impact of these firms became evident. As discussed in detail in Chapter 2, Lehman's bankruptcy impacted 8,000 subsidiaries and affiliates worldwide, approximately 100,000 creditors, and its 26,000 employees worldwide.2
Time and time again, from the Spring of 2007 on, policymakers and regulators were caught off guard as the contagion spread, responding on an ad-hoc basis with specific programs to put fingers in the dike. There was no comprehensive and strategic plan for containment, because they lacked a full understanding of the risks and interconnections of the financial markets. Some regulators have conceded this error. We had allowed the system to race ahead of our ability to protect it.3
Interconnections among financial firms can also lead to systemic risk under crisis conditions. Financial institutions are interconnected in a variety of networks in bilateral and multilateral relationships and contracts, as well as through markets.4
After reading this chapter you will be able to:
As mentioned earlier, the Credit Crisis illustrated that future risk analysis and monitoring by financial market participants and supervisors needs to go well beyond assessing the impact of a firm on a stand-alone basis. Analysis and monitoring must become more sophisticated and consider the interconnections that firms have around the globe. When a large, highly interconnected firm fails, that could lead to a significant risk of global contagion by spreading losses to hundreds, if not thousands, of counterparties.
For example, when two Bear Stearns hedge funds fell into distress in 2007, due largely to leveraged exposure to mortgage-backed securities, this propagated risk across the financial system. Due to continued losses at the two hedge funds in 2006 and 2007, investors began to steadily withdraw money. During 2007, the funds were forced to sell assets at distressed prices given that the prices and demand for mortgage-related securities both fell dramatically during this time. As this occurred, firms that had lent the funds money against this mortgage collateral began to demand higher margin terms while others refused to renew maturing secured financing trades, known as repos. For example, one of the fund's largest repo lenders, Merrill Lynch, seized $850 million of collateral that the funds had posted against their loans.5
Although ultimate parent Bear Stearns was not legally obligated to support the funds, the firm decided to take out repo lenders to the Bear Stearns High-Grade Structured Credit Enhanced Leveraged Fund by assuming about $1.8 billion of subprime assets onto its books, contributing to a write-down of nearly $2 billion of mortgage-related collateral by Bear in November 2007. This led investors, creditors, and rating agencies to more closely scrutinize Bear's leverage, asset quality, and liquidity. At that time, Bear was the second largest prime broker6 in the industry and was a top-three underwriter of private-label mortgage-backed securities. Over the subsequent months and into early 2008, lenders began demanding more collateral from Bear and rumors swirled of their diminishing capital and liquidity.
On Thursday, March 13, Bear's CEO informed the SEC that the firm would be “unable to operate normally” the next day. On Sunday, March 16, J.P. Morgan informed the Fed and Treasury that it was interested in a deal to buy Bear contingent on financial support from the Fed. The Fed, under section 13(3) of the Federal Reserve Act, agreed to purchase $29.97 billion of Bear's assets to get them off the firm's books through a newly created entity called Maiden Lane LLC.7 J.P Morgan financed the first $1.15 billion through a subordinated loan while the Fed bore the risk of the remaining $28.8 billion of assets. On the evening of March 16, J.P. Morgan announced a deal to buy Bear for $2 share, which was subsequently increased to $10 share on March 24.
Fed Chairman Ben Bernanke justified the Fed's saving of Bear by citing the beginnings of a “breakdown” in the $2.8 trillion tri-party repo market and their view that Bear “… was so essentially involved in this critical repo financing market, that its failure would have brought down that market, which would have had implications for other firms.”8
The Financial Crisis Inquiry Commission (FCIC) concluded that Bear experienced runs by repo lenders, hedge fund customers, and derivatives counterparties and was rescued by a government-assisted purchase by J.P. Morgan because the government considered it too interconnected to fail.
It's important to note that there is no single, universally accepted definition of interconnectedness within the financial industry, particularly since it's a relatively new concept. Moreover, the term can cover an extremely broad array of financial, operational, and legal relationships, contracts, or dependencies. As one real-life example, as previously referenced with respect to some of the failed financial firms during the Credit Crisis, such firms were party to hundreds of thousands of derivatives and collateral contracts across the globe. One can think of Lehman Brothers as the central node or hub in a complex web of connections with its losses spreading across this web and affecting, to different degrees, any entity that maintained a relationship with it.
Given this inherent complexity in defining interconnectedness, we begin by providing readers a broad taxonomy of categories according to research by The Depository Trust & Clearing Corporation (“DTCC”) that may assist firms in applying interconnectedness analysis to their respective organizations or other analytical endeavors.
Direct and Indirect Financial Connections:9
Operational Connections:
Global regulators such as the Basel Committee on Banking Supervision have agreed on five categories (listed in the following Table 15.1) for measuring the systemic importance of Global Systemically Important Banks (G-SIBs). Subsequently, the Fed adopted this methodology to determine which U.S. banks are G-SIBs. The methodology gives an equal weight of 20% to each of the five categories of systemic importance, which are: size, cross-jurisdictional activity, interconnectedness, substitutability/financial institution infrastructure, and complexity. For purposes of this chapter, we focus primarily on two of these five categories: (i) interconnectedness, which attempts to measure a G-SIB's financial interconnectedness; and (ii) substitutability, which focuses on critical services provided by the G-SIBs to the industry. See Chapter 9 for details and a listing of current G-SIBs.
Table 15.1 The G-SIB Assessment Methodology
Category (and Weighting) | Individual Indicator | Indicator Weighting |
Interconnectedness (20%) | Intra-financial system liabilities Intra-financial system assets Wholesale funding ratio |
6.67% 6.67% 6.67% |
Substitutability/financial institution infrastructure (20%) | Assets under custody Payments cleared and settled through payment systems Values of underwritten transactions in debt and equity markets |
6.67% 6.67% 6.67% |
Cross-jurisdictional activity (20%) | Cross-jurisdictional claims Cross-jurisdictional liabilities |
10% 10% |
Size (20%) | Total exposures as defined for use in the Basel III leverage ratio | 20% |
Complexity (20%) | OTC derivatives notional value Level 3 assets Held for trading and available for sale value |
6.67% 6.67% 6.67% |
According to a detailed systemic risk survey conducted jointly by Deloitte and SIFMA (June 2010), the factors identified by Basel are largely in line with the views of the broader market. The survey found that factors typically included in a definition of systemic risk include: size (of an individual financial institution or a combination of smaller firms); interconnectedness; and the potential for underlying issues, such as complexity and leverage, exposure concentrations, erosion of market practices, marketplace bubbles, and the potential of a failure (of a systemically important firm or group of firms) to serve as a trigger event that may impact the broader real economy. When considering the complexity of the global financial system, it is essential that key drivers of systemic risk are identified. Knowing these drivers will enable a systemic risk regulator to identify, measure, and monitor systemic risk events across the entire financial services system.
The OFR describes interconnectedness as the failure of a bank to meet payment obligations to other banks, which can accelerate the spread of a financial system shock if the bank is highly interconnected. The OFR has adopted Basel's five-part methodology for assessing a bank's degree of systemic importance.
One of the OFR's many different measures of the systemic risks posed by large banks includes a bank's total claims on the financial system, its total liabilities to the financial system, and the total value of debt and equity securities issued by a bank. For the first two of these indicators, the financial system includes banks, securities dealers, insurance companies, mutual funds, hedge funds, pension funds, investment banks, and central counterparties.
Table 15.2 illustrates the OFR's annual ranking of G-SIBs' systemic importance based upon the five equally weighted measures described earlier.11
Table 15.2 Global Systemically Important Banks
Bank Holding Company | G-SIB Bucket10 |
JP Morgan Chase & Co. | 4 |
HSBC | 4 |
Citigroup | 3 |
BNP Paribas | 3 |
Deutsche Bank | 3 |
Barclays | 3 |
Bank America | 2 |
Credit Suisse | 2 |
Goldman Sachs | 2 |
Mitsubishi | 2 |
Morgan Stanley | 2 |
Industrial & Commercial Bank of China | 1 |
Royal Bank of Canada | 1 |
Société Générale | 1 |
Bank of China | 1 |
Banco Santander | 1 |
Wells Fargo | 1 |
UBS | 1 |
Credit Agricole | 1 |
China Construction Bank | 1 |
Unicredit | 1 |
Agricultural Bank of China | 1 |
Mizuho | 1 |
Groupe BCPE | 1 |
Bank of New York Mellon | 1 |
State Street | 1 |
Sumitomo Mitsui | 1 |
Standard Chartered | 1 |
ING Group | 1 |
Nordea Bank | 1 |
Another measure of a bank's systemic risk performed by the OFR includes analysis conducted of the public sections of the eight U.S. G-SIBs' living wills. To minimize the risk of costly bank bailouts in the future, the Dodd-Frank Wall Street Reform and Consumer Protection Act (Dodd-Frank Act) requires every bank holding company with $50 billion or more in assets to prepare a resolution plan, or living will. Living wills describe how a failing bank would wind down. Each of the individual living wills for the eight U.S. G-SIBs differs somewhat in terms of format and content. However, most contain enough standardized information to allow the OFR to conduct analysis across banks and across time. One such category is called “complexity” and measures the number of core business lines, material legal entities, and critical operations of each bank, which can serve as measures of complexity and provide an indication of the ease by which each bank could be resolved.
The next category of complexity measured by the OFR is a study of the layers of corporate structure within the U.S. G-SIBs. According to the OFR, multiple layers of corporate structure can make orderly resolution of a bank more difficult. In this study, the average subsidiary is 4 to 6 layers below the parent and in some cases 20 layers below.
Another measure employed by the OFR is called intra-firm interconnectedness, which quantifies the total number of internal connections with other material legal entities within each organization. It also measures the total number of membership each G-SIB maintains in financial market utilities (FMUs), which provide payment, clearance, and settlement services to banks. This is deemed important as FMUs require G-SIBs to post collateral to them on a regular basis, and they also maintain certain early termination rights, both of which can complicate the resolution process.
As covered in Chapter 8 of this book, CPMI IOSCO publishes a set of 24 principles for financial market infrastructures which recognizes that interconnectedness among FMIs may foster knock-on effects throughout the financial system in a stress event, thereby impacting participants and markets across the FMIs. Specific examples include:
In this section, we will explore an approach proposed by DTCC that financial firms may employ to initiate their own risk analysis, consisting of the following elements:12
Let's explore each of these elements.
Make a comprehensive inventory of external entities on which your firm relies: Most financial institutions rely on adequate funding and liquidity, credit, access to markets and market infrastructures, as well as the provision of reliable and timely data—among many other processes. External entities that provide or support these services represent external interconnections to your firm. Given that insolvencies occur at a legal entity level, intragroup dependencies between distinct legal entities should also be represented as external interconnections.
Determine which interconnections are critical to your business: Use the following criteria to assess the level of criticality of your interconnections:
Quantify your critical interconnections if practical: Quantifying interconnections can be useful as a straightforward and objective way to aggregate, rank, and assess the related risks. It may also help prioritize risks and monitor their evolution over time. That said, operational interconnections with providers of data and other financial services may be harder to quantify than those with borrowers/lenders, trade counterparties, and funding providers. Therefore, interconnections should be quantified as appropriate depending on the circumstances and the effort involved in doing so.
Assess in detail how an impaired interconnection could affect specific areas: Depending on the circumstances, the failure of an interconnected entity may cause a credit or trading loss, but it may also cause a loss of revenue, affect funding, or have a different type of impact altogether. In assessing the effect of an impaired or failing interconnection, it may be more appropriate to consider peak volumes and associated risks, rather than average values.
Identify highly interconnected entities and assess the potential impact of their failure on your business entity: While the analysis described earlier is valuable, its real power lies in the aggregation of risks across areas that may be simultaneously affected by a single failure. The failure of a highly interconnected entity may have a combined effect—for instance, by simultaneously causing credit losses while affecting your firm's funding as well as your access to other financial services. While each of these impacts may be manageable individually, their combined effect may not be.
Manage exposures to interconnected entities holistically: Given the potential combined effect of the failure of a highly interconnected entity, it is important to manage the associated risks holistically. Among other things, that means that concentration risk should be managed not only by assessing the relative exposures to funding, trading, credit, and other counterparties in isolation, but also in its entirety across these various areas. Stress tests and scenario analyses can be very valuable in this respect, provided they explicitly incorporate these forms of interconnectedness.
Cooperate across departments: Organize cross-functional risk reviews and discussions to make interconnectedness awareness an integral part of your organization's risk management culture. Interconnectedness analysis should complement other disciplines, not replace them.
Take a gradual approach: As is the case for other risk management disciplines, interconnectedness analysis is an iterative process—start small and expand gradually. Periodically assess in which areas you may need to become more sophisticated.
Q15.1: Which event in the financial industry brought the concept of interconnectedness into the forefront for the first time?
Q15.2: Provide some examples of how interconnectedness risk manifested itself with respect to failed financial firms such as Lehman Brothers, AIG, and Bear Stearns during 2008?
Q15.3: What is the primary difference between interconnectedness risk and the way in which G-SIBs were analyzed prior to the Credit Crisis by risk managers and financial regulators?
Q15.4: Which regulatory institution was the first to create a formal methodology for quantifying the relative systemic risk introduced by G-SIBs?
Q15.5: What are the five primary categories that comprise the methodology used today by several regulatory bodies to assess the degree of systemic risk for G-SIBs?
Q15.6: Of the five categories referenced earlier, which two are most relevant for interconnectedness risk analysis and why?
Q15.7: What are the two broad categories of interconnectedness as detailed in this chapter? Please name three of the subcategories of each.
Q15.8: What three financial measures does the Office of Financial Research apply to G-SIBs to calculate each bank's total interconnectedness?
Q15.9: What three criteria does the Office of Financial Research apply to G-SIBs to calculate each bank's total substitutability risk?
Q15.10: What is the first step that firms should take when attempting to initiate a new analysis of the level of interconnectedness risk their firm is exposed to?