The reduction of systemic risk in the United States financial system.

• Author: Scott, Hal S.

1. SYSTEMIC RISK REDUCTION: THE CENTRAL PROBLEM II. CAPITAL REQUIREMENTS A. CCMR Recommendations Aligned with the White Paper B. CCMR Recommendations That Differ from the White Paper and Pending Legislation: How Much and What Type of Capital 1. How Much Capital: Regulation and Markets 2. What Counts as Capital? III. CLEARINGHOUSES AND EXCHANGES FOR DERIVATIVES A. The Ability of the Clearinghouse to Reduce Counterparty Risk 1. Customized or Illiquid Contracts 2. Contracts Involving Nonparticipants in the Clearinghouse B. The Optimal Number and Scope of Clearinghouses C. Ownership of Clearinghouses D. Collection and Publication of Data E. Exchange Trading F. The International Dimension IV. RESOLUTION PROCEDURES A. The Importance of Resolution Procedures. B. What Institutions Should be Subject to Special Resolution Procedures? C. Imposition of Losses under Special Resolution Procedures D. Funding the Cost of New Procedures E. The International Dimension V. EMERGENCY FEDERAL RESERVE LENDING VI. REGULATORY REORGANIZATION A. Regulation of Systemic Risk B. Supervisory Authority C. International Developments CONCLUSION This Article concentrates on the central problem for financial regulation that has emerged from the 2007-2009 financial crisis--the prevention of systemic risk. The discussion largely focuses on the relevant recommendations of the Committee on Capital Markets Regulation (CCMR) in its May 2009 report. (1) Where appropriate, the Article compares the CCMR recommendations to those of the United States Treasury in its June 2009 report (2) and its suggested implementing legislation, and also to pending congressional legislation. (3)

   The CCMR is an independent, nonpartisan research organization founded in 2005 to improve the regulation of United States capital markets. (4) "Thirty leaders from the investor community, business, finance, law, accounting, and academia comprise the CCMR's membership." (5) Its "co-chairs are Glenn Hubbard, Dean of Columbia Business School, and John Thornton, Chairman of the Brookings Institution." (6) The Author of this Article is the Director.

1. SYSTEMIC RISK REDUCTION: THE CENTRAL PROBLEM

   Going forward, the central problem for financial regulation (defined as the prescription of rules, as distinct from supervision or risk assessment) is to reduce systemic risk. Systemic risk is the risk that the failure of one significant financial institution can cause or significantly contribute to the failure of other significant financial institutions as a result of their linkages to each other. Systemic risk can also be defined to include the possibility that one exogenous shock may simultaneously cause or contribute to the failure of multiple significant financial institutions. This Article focuses on the former definition because proper regulation could have the greatest potential to reduce systemic risk in this area. (7)

   There are four principal linkages that can result in a chain reaction of failures. First, there are interbank deposits, whether from loans or from correspondent accounts used to process payments. These accounts were the major concern when Continental Illinois Bank almost failed in the mid-1980s. (8) Continental held sizable deposits of other banks; in many cases, the amount of the deposits substantially exceeded the capital of the depositor banks. These banks generally held such sizable deposits because they cleared payments, such as checks or wire transfers, through Continental. If Continental had failed, those banks would have failed as well. Section 308 of the FDIC Improvement Act of 1991 gives the Federal Reserve Board powers to deal with this problem. (9) The Act permits the Board to limit the credit extended by an insured depository institution to another depository institution. (10) Limitation of interbank deposits may be feasible with respect to placements by one bank with another because the amount of credit extended is fixed for a given term. Indeed, it appears that the chain-reaction risk arising from bilateral credit exposures from overnight Federal Reserve funds transactions is quite low: Losses would not exceed one percent of total commercial banking assets as long as loss rates are kept to historically observed levels. (11)

   Exposures are more difficult to identify with respect to interbank clearing accounts where the amount of credit extended is a function of payment traffic. For example, Bank A may be credited by its correspondent Bank B for an incoming wire transfer of $10 million. Bank A is thus a creditor of Bank B for this amount. If Bank B were to fail, Bank A is seriously exposed. (12) Without material changes in the payment system, such as forcing banks to make and receive all payments through Federal Reserve rather than correspondent accounts, it would be quite difficult to limit these types of exposures.

   Second, a chain reaction of bank failures can occur through net settlement payment systems. If one bank fails to settle its position in a net settlement system for large value payments, such as the Clearing House Interbank Payments System (CHIPS) in the United States, other banks that do not get paid may, in turn, fail. (13) This risk was the major systemic risk concern of the Federal Reserve until CHIPS changed its settlement procedures in 2001 to essentially eliminate this risk. (14)

   Third, a chain reaction of bank failures can occur through imitative runs. When one bank fails, depositors in other banks, particularly those whose deposits are uninsured, may assume that their banks may also fail and so withdraw their funds, ex-posing these banks to a liquidity crisis and ultimately to failure. This result comes from a lack of information in the market about what specifically caused the first bank to fail. (15) The Federal Reserve plays the classic role of lender of last resort to stem irrational imitative runs in situations such as this one.

   Lastly, and especially prominent in the current crisis, a chain reaction of bank failures can occur as a result of counterparty risk on derivative transactions, such as credit default swaps (CDSs). (16) Here the concern is that if institution X fails to settle its derivative position with institution Y, both X and Y will fail. If Y in turn cannot settle its positions, other institutions will also fail. This risk proved potentially significant in the failure of the hedge fund Long-Term Capital Management in 1998. (17) Concerns of this type also underlay JPMorgan Chase's assisted acquisition of Bear Stearns and the injection of federal funds into AIG. (18) This is one area in which the failure of non-banks is a major concern, but the severity of this form of systemic risk and the degree of interconnectedness among financial institutions is currently unknown. (19) A report by the Special Inspector General for the Troubled Asset Relief Program (SIGTARP) on the government's investments in AIG indicated that Goldman Sachs, a major counterparty, would have been made whole in the event of an AIG default. (20) The report further indicated that the Treasury and Federal Reserve were primarily concerned with losses that would be incurred by investors in AIG in the event of a default, including $10 billion of state and local government money, $40 billion in 401(k) plans, and $38 billion in retirement plans. (21) The report's explanation of the government's action also mentioned concern over stemming runs on money market funds, which held $20 billion in AIG commercial paper. (22) Similarly, in their recent testimony on the "Federal Bailout of AIG," Treasury Secretary Timothy Geithner and New York Federal Reserve General Counsel Thomas Baxter also emphasized factors other than derivatives counterparty risk, including the impact that the failure of AIG would have on money market funds, personal savings and retirement plans, and insurance policyholders. (23) If prospective investor losses, rather than the fallout of interconnectedness, were the true basis for the government policy with respect to AIG, it may be that the concern with systemic risk is overstated. Further study and better disclosure from the Treasury and Federal Reserve is needed to support informed estimates of the magnitude of the problem. In any event, gauging the impact of systemic risk is difficult to determine and beyond the scope of this Article. (24)

   The threat of systemic risk (whether real or imagined) results in both the need for government bailouts at taxpayer expense and in an increase in moral hazard. These results occur because both equity and debt holders, as well as counterparties, may be protected against losses. Of course, the government could decide not to intervene, but this laissez-faire approach could put the entire global economy at risk, an even worse outcome. As the financial crisis has illustrated, banks cannot always count on the government to cut off systemic risk when it occurs. The politics of supplying money to banks are unpopular and unsustainable by the Federal Reserve over the long term without intense public scrutiny and loss of independence.

   At the outset, it is also worth noting that the "Volcker Rules" and related limitations on bank size announced by the Obama Administration on January 21, 2009 (25) do not have much if any potential to reduce systemic risk. The Volcker Rules would prohibit bank holding companies and all of their subsidiaries from engaging in proprietary trading, as well as from investing in or sponsoring hedge fund and private equity operations. Although President Obama has characterized proprietary trading as trading "unrelated to serving customers," (26) a precise legal standard has not been given. The related size limitations were initially described as straightforward caps on each bank's market share of non-deposit liabilities. As Deputy Treasury Secretary Neal Wolin describes, however, the size limits would not require banks to divest existing operations or restrict organic growth, but...