ICI Comments on FSOC's Proposed Recommendations for Money Market Fund Reform (pdf)

Comment Letter of the Investment Company Institute on Financial Stability Oversight Council Proposed Recommendations Regarding Money Market Mutual Fund Reform Docket No. FSOC-2012-0003 January 24, 2013 I. Introduction and Executive Summary .................................................................................................................. 1   Executive Summary of Comments ..................................................................................................................... 5   II. Section 120 Requirements .................................................................................................................................. 12   A. FSOC Does Not Have the Authority to Make Recommendations Under Section 120 Regarding Money Market Funds ........................................................................................ 12   B. FSOC’s Determination of Systemically Risky Activity Under Section 120 Is Premised on Misleading and Incorrect Statements ............................................................................... 14   1. Conduct and Nature of Money Market Funds’ Practices and Activities ................................ 15   2. Size, Scale, and Concentration .......................................................................................................... 28   3. Interconnectedness .............................................................................................................................. 29   4. “Evidence” from the 2007–2008 Financial Crisis ........................................................................ 30   5. The 2010 Reforms ............................................................................................................................... 37   C. FSOC’s Proposed Determination Is Overly Broad ............................................................................... 40   1. Treasury and Government Money Market Funds ........................................................................ 40   2. Tax-Exempt Money Market Funds .................................................................................................. 47   III. Temporary Gates and Liquidity Fees .............................................................................................................. 49   A. Objective Trigger for Gates and Fees ........................................................................................................ 50   B. Enhanced Disclosure ..................................................................................................................................... 51   C. Temporary Gates and Liquidity Fees Specifically Address Systemic Concerns .............................. 51   D. U.S. and European Experiences with Suspension of Redemptions and Gating .............................. 52   1. U.S. Fund Experiences ........................................................................................................................ 52   2. European Fund Experiences .............................................................................................................. 53   E. U.S. Tax Implications .................................................................................................................................... 54   F. Operational Implications ............................................................................................................................. 55   1. Difficulties of Intraday Gating .......................................................................................................... 55   2. Systems Modifications for End-of-Day Gating ............................................................................. 56   G. Impact on Certain Transaction Types ..................................................................................................... 57   IV. FSOC Alternative One: Floating NAV ......................................................................................................... 57   A. A $100 Price Is Arbitrary and Without Precedent ................................................................................ 57   Financial Stability Oversight Council January 24, 2013 Page ii B. A Floating NAV Will Not Prevent Investor Runs ................................................................................. 58   C. Investors Would Seek Stable NAV Alternatives .................................................................................... 61   1. Tax Implications ................................................................................................................................... 62   2. Accounting Implications .................................................................................................................... 65   3. Specialized Business Applications and Automated Systems ....................................................... 66   4. Statutory Prohibitions and Investment Restrictions ................................................................... 67   D. Floating the NAV Would Harm the Market ......................................................................................... 67   E. Transition and Grandfather Issues ............................................................................................................. 69   F. Removing Exemptions Under the Investment Company Act Is Not Necessary ............................. 69   V. FSOC Alternative Two: NAV Buffer and Minimum Balance at Risk ..................................................... 70   A. Effects on Investor Liquidity and Market Size ........................................................................................ 71   B. Operational Implications ............................................................................................................................. 72   VI. FSOC Alternative Three: NAV Buffer and Other Measures ................................................................... 74   A. NAV Buffer ..................................................................................................................................................... 75   1. Requiring Fund Advisers to Commit Capital ............................................................................... 75   2. Requiring Funds to Raise Capital in the Market .......................................................................... 81   3. Requiring a Within-Fund Capital Buffer ....................................................................................... 82   B. Other Measures .............................................................................................................................................. 82   1. More Stringent Investment Diversification Requirements ........................................................ 83   2. Increased Minimum Liquidity Requirements ............................................................................... 83   3. Shareholder Transparency .................................................................................................................. 83   VII. Economic Impact of Proposed Recommendations .................................................................................... 84   A. FSOC’s Estimated Benefits of Proposed Recommendations Are Illusory ....................................... 85   B. FSOC’s Cost Analysis Is Narrow and Highly Speculative .................................................................... 89   VIII. Conclusion ........................................................................................................................................................ 92   January 24, 2013 Financial Stability Oversight Council Mr. Amias Gerety, Deputy Assistant Secretary 1500 Pennsylvania Avenue, NW Washington, DC 20220 Re: Proposed Recommendations Regarding Money Market Mutual Fund Reform (FSOC-2012-0003) Members of the Financial Stability Oversight Council: The Investment Company Institute (“ICI”)1 submits this letter in response to the Financial Stability Oversight Council’s (“FSOC” or “Council”) request for comment regarding its Proposed Recommendations Regarding Money Market Mutual Fund Reform (“Report”).2 FSOC issued its request under Section 120 of the Dodd-Frank Wall Street Reform and Consumer Protection Act (“Dodd- Frank Act”), which permits FSOC to make recommendations to a “primary financial regulatory agency”—in this case, the Securities and Exchange Commission (“SEC”)—to make changes to its own regulations governing nonbank financial institutions under its jurisdiction. I. Introduction and Executive Summary Money market funds, which date back to the early 1970s, are one of the most significant and successful financial product innovations of the past half century. Today, 61 million retail investors, as well as corporations, municipalities, nonprofits, retirement plans, and other institutional investors, rely on the $2.7 trillion money market fund industry to provide a low-cost, efficient cash management tool that offers a high degree of liquidity, stability of principal value, and a market-based yield. Money market funds also serve as an important source of direct financing for state and local governments, businesses, and financial institutions, and of indirect financing for households. Without these funds, financing for all of these institutions and individuals could be more expensive and less efficient, and competition among financial intermediaries would be weaker. Money market funds owe their success, in large part, to the stringent regulatory requirements to which they are subject under the federal securities laws, including most notably Rule 2a-7 under the 1 The Investment Company Institute is the national association of U.S. investment companies, including mutual funds, closed-end funds, exchange-traded funds (ETFs), and unit investment trusts (UITs). ICI seeks to encourage adherence to high ethical standards, promote public understanding, and otherwise advance the interests of funds, their shareholders, directors, and advisers. Members of ICI manage total assets of $13.9 trillion and serve more than 90 million shareholders. 2 The Report is available on the Treasury Department’s website at http://www.treasury.gov/initiatives/fsoc/Documents/Proposed%20Recommendations%20Regarding%20Money%20 Market%20Mutual%20Fund%20Reform%20-%20November%2013,%202012.pdf. Financial Stability Oversight Council January 24, 2013 Page 2 Investment Company Act of 1940 (“Investment Company Act”). The regulatory regime established by Rule 2a-7 has proven to be effective in protecting investors’ interests and maintaining their confidence in money market funds. The SEC deserves tremendous credit for crafting these requirements and administering them in a manner that has allowed money market funds to thrive and to serve so many investors. In recognition of the importance of money market funds, ICI and its members have devoted significant time and effort to considering how to make these funds more robust under even the most adverse market conditions—such as those caused by the widespread failures of banks and other financial institutions in 2007 and 2008. Over the past few years, the SEC, ICI, and ICI’s members have made a great deal of progress toward their shared goal of strengthening the resiliency of money market funds.3 Taking the initiative to respond quickly and aggressively to the events of fall 2008, ICI formed the Money Market Working Group (“MMWG”) to study the money market, money market funds and other participants in the money market, and recent market circumstances. The March 2009 Report of the Money Market Working Group (“MMWG Report”) addressed these topics and advanced wide-ranging recommendations for the SEC to strengthen money market fund regulation.4 In 2010, with ICI’s strong support, the SEC approved far-reaching rule amendments that incorporated many of the MMWG Report’s recommendations and enhanced an already strict regime of money market fund regulation.5 The amended rules make money market funds more resilient by, among other things, imposing new credit quality, maturity, and liquidity standards and increasing the transparency of these funds. In the event a money market fund proves unable to maintain a stable $1.00 net asset value (“NAV”) per share, the fund’s board of directors is empowered to take prompt action to assure an orderly liquidation of the fund and equitable treatment for all shareholders. These risk- limiting reforms proved their value during the summer of 2011 when money market funds—without incident—met large volumes of shareholder redemptions during periods of significant market turmoil, including the historic downgrade of U.S. government debt and a widespread financial crisis in Europe. Indeed, so far-reaching were these reforms that today’s money market funds are dramatically different from those of 2008. Yet, the calls for further changes continue. 3 For a detailed overview of these efforts, see Perspectives on Money Market Mutual Fund Reforms, written testimony of Paul Schott Stevens, President and CEO, Investment Company Institute, before the U.S. Senate Committee on Banking, Housing and Urban Affairs (June 21, 2012), available at http://www.ici.org/pdf/12_senate_pss_mmf_written.pdf, at 11-14. 4 See Investment Company Institute, Report of the Money Market Working Group (March 17, 2009), available at http://www.ici.org/pdf/ppr_09_mmwg.pdf. 5 See Money Market Fund Reform, SEC Release No. IC-29132 (February 23, 2010), 75 FR 10060 (March 4, 2010) (“2010 MMF Reform Release”). Financial Stability Oversight Council January 24, 2013 Page 3 Over the past year, former SEC Chairman Mary L. Schapiro outlined in various public statements what she believed a formal money market fund rule proposal should include.6 In response to these and other public statements, the SEC amassed an extensive record of comment letters, surveys, research, reports, articles, and other materials filed by a range of market participants and other interested commenters. The overwhelming majority of those commenters focused on the adverse consequences of the contemplated proposals.7 Commenters raised substantial concerns that these measures, if adopted, would drive funds out of business, reduce competition and choice, and alter the fundamental characteristics of money market funds, thereby destroying their value to investors, issuers, and the economy. Many also argued that, rather than making our economy and financial system stronger, such reforms have the very real potential to increase systemic risk by driving investors into less-regulated, less- transparent products. Based upon their review of the record and their analyses of the SEC staff’s draft release setting forth Chairman Schapiro’s proposals, a bipartisan majority of the SEC—Commissioners Luis A. Aguilar, Daniel M. Gallagher, and Troy A. Paredes—expressed deep concern about proceeding down the regulatory path suggested by the Chairman and opposed issuing the proposals without additional study and analysis.8 6 According to these statements, the Chairman’s agenda included capital buffers combined with redemption holdback restrictions and/or requiring money market funds to “float” their NAVs. See, e.g., Statement of SEC Chairman Mary L. Schapiro on Money Market Fund Reform (August 22, 2012), available at http://www.sec.gov/news/press/2012/2012- 166.htm; Remarks at the Society of American Business Editors and Writers (SABEW) Annual Convention (March 15, 2012), available at http://www.sec.gov/news/speech/2012/spch031512mls.htm; Remarks at SIFMA’s 2011 Annual Meeting (November 7, 2011), available at http://www.sec.gov/news/speech/2011/spch110711mls.htm. 7 Interested persons submitted comments on money market fund reform over a two-year period—first in response to the SEC’s request for comment on the 2010 Report of the President’s Working Group on Financial Markets: Money Market Fund Reform Options (“PWG Report”), see SEC Release No. IC-29497 (November 3, 2010), available at http://www.sec.gov/rules/other/2010/ic-29497.pdf, and later in response to Chairman Schapiro’s proposals. As a follow-up to its request for comments on the PWG Report, on May 10, 2011, the SEC hosted a roundtable (“SEC Roundtable”) on money market funds and systemic risk that consisted of SEC officials, representatives of FSOC, and participants from ICI, the fund industry, academia, the business community, and state and local governments. The unofficial transcript for the SEC Roundtable is available on the SEC’s website at http://sec.gov/spotlight/mmf-risk/mmf-risk-transcript-051011.htm. In response to the PWG Report, as well as Chairman Schapiro’s agenda, ICI filed eight submissions. See August 20, 2012, submission at http://www.sec.gov/comments/4-619/4619-262.pdf; June 20, 2012, submission at http://www.sec.gov/comments/4-619/4619-200.pdf; May 24, 2012, submission at http://www.sec.gov/comments/4- 619/4619-182.pdf; May 16, 2012, submission at http://www.sec.gov/comments/4-619/4619-177.pdf; April 19, 2012, submission at http://www.sec.gov/comments/4-619/4619-166.pdf; April 13, 2012, submission at http://www.sec.gov/comments/4-619/4619-164.pdf; February 16, 2012, submission at http://www.sec.gov/comments/4- 619/4619-119.pdf; and January 10, 2011 submission at http://www.sec.gov/comments/4-619/4619-49.pdf. For a detailed summary of additional comments, surveys, reports and other data filed with the SEC, see Letter from John D. Hawke, Jr., Arnold & Porter LLP, on behalf of Federated Investors, Inc. to the Honorable Mary L. Schapiro, Chairman, Securities and Exchange Commission (July 17, 2012), available at http://www.sec.gov/comments/4-619/4619-213.pdf. 8 See Statement Regarding Money Market Funds by Commissioner Luis A. Aguilar, Securities and Exchange Commission (August 23, 2012) (“Aguilar Statement”), available at http://www.sec.gov/news/speech/2012/spch082312laa.htm; Statement on the Regulation of Money Market Funds by Commissioner Daniel M. Gallagher and Commissioner Troy A. Paredes, Securities and Exchange Commission (August 28, 2012), available at http://www.sec.gov/news/speech/2012/spch082812dmgtap.htm. Financial Stability Oversight Council January 24, 2013 Page 4 Indeed, by all accounts the SEC actively continues to consider what actions, if any, are appropriate.9 For example, at the request of the three Commissioners named above, the SEC staff recently issued a study addressing the causes of investor redemptions in 2008, the efficacy of the SEC’s 2010 amendments to strengthen Rule 2a-7, and the potential impacts of future money market fund reform on issuers and investors.10 Unable to convince her own agency to put forth her ideas, Chairman Schapiro urged FSOC to intervene. In response, the Council invoked Section 120 of the Dodd-Frank Act, apparently with the goal of compelling the SEC to pursue the same proposals that a majority of the SEC had refused to consider just three months earlier.11 In fact, the Report appears to have ignored hundreds of comment letters opposed to these very proposals. FSOC took this action despite clear indications that the SEC remains engaged on this issue, and despite the fact that, as FSOC itself acknowledges, “[t]he SEC by virtue of its institutional expertise and statutory authority, is best positioned to implement reforms to address the risks that [money market funds] present to the economy.”12 For our part, ICI consistently has supported exploring reasonable options to make money market funds even more resilient while preserving the fundamental characteristics of these funds.13 We remain firmly committed to working with regulators on this important issue. We submit, however, that this process should be guided by two principles. First, we should preserve those key features of money market funds that have made them so valuable and attractive to investors. Second, we should preserve choice for investors by ensuring a continued robust and competitive global money market fund industry. As discussed in detail in Section III, we believe that, should regulators continue to believe further actions are necessary, the use of gates and liquidity fees by “prime” money market funds14 can provide further stability to money market funds consistent with these two objectives. 9 See, e.g., Remarks before the U.S. Chamber Center for Capital Markets Competitiveness by Commissioner Daniel M. Gallagher, Securities and Exchange Commission (January 16, 2013), available at http://www.sec.gov/news/speech/2013/spch011613dmg.htm (noting that a bipartisan majority of Commissioners are currently working with staff of the Division of Risk, Strategy and Financial Innovation and the Division of Investment Management to shape a money market fund reform proposal). 10 See SEC Division of Risk, Strategy and Financial Innovation, Response to Questions Posed by Commissioners Aguilar, Paredes, and Gallagher (November 30, 2012) (“SEC Staff Study”), available at http://www.sec.gov/news/studies/2012/money-market-funds-memo-2012.pdf. 11 The Report also proposes a third alternative (Alternative Three, discussed in Section VI.), which suffers from many of the same flaws that beset Chairman Schapiro’s reform concepts. 12 See Report, supra note 2, at 15. 13 For example, in a comment letter responding to the PWG Report, ICI detailed a concept for a private liquidity facility to further strengthen “prime” money market funds. See Appendix to Letter from Paul Schott Stevens, President and CEO, Investment Company Institute, to Elizabeth M. Murphy, Secretary, Securities and Exchange Commission (January 10, 2011), available at http://www.ici.org/pdf/11_sec_pwg_com.pdf and http://www.ici.org/pdf/11_sec_pwg_deck.pdf. 14 “Prime” money market funds are funds that may invest in high-quality, short-term money market instruments, including Treasury and government obligations, certificates of deposit, repurchase agreements, commercial paper, and other money market securities. They do not include tax-exempt, government, or Treasury money market funds. Financial Stability Oversight Council January 24, 2013 Page 5 Unfortunately, as discussed below in Sections IV, V, and VI, FSOC’s proposals are altogether at odds with these objectives. Rather, the proposals are of a nature that would eliminate or minimize the utility of these funds to investors,15 and would impose burdens on fund sponsors and intermediaries likely to impel many, if not most, to exit the business. Moreover, FSOC’s proposals would not achieve the very goals that FSOC has articulated for itself. To reiterate, ICI and its members are committed to working with policymakers to further strengthen money market funds’ resilience in the face of severe market stress. In light of the untested nature of the proposed recommendations and the highly speculative nature of any benefits and costs, however, the Council’s proposed experiment with a core sector of our financial system is simply unwarranted. We strongly oppose FSOC’s proposed recommendations. EXECUTIVE SUMMARY OF COMMENTS The four years since the peak of the financial crisis in September 2008 have been marked by a vigorous public dialogue about the characteristics, value, and regulation of money market funds. Regulators, fund sponsors, investors, issuers in the money markets, financial analysts, academics, and other commenters have contributed to a voluminous public record before the SEC, a record replete with economic research, legal analysis, and the perspectives of a broad range of participants in the short-term financial markets.16 Even as this dialogue was underway, events in the market—particularly the U.S. and European debt crises of 2011—have provided empirical tests that challenge regulators’ assumptions and assertions. This record and recent market experience have demonstrated the significant value of money market funds to investors, issuers, and the economy at large; the added resilience that the SEC reforms of 2010 have brought to these funds; and the widespread public opposition to the most prominent proposals for further changes in light of the harmful effects they would have. Our comments on the Report build upon this extensive public dialogue. We begin with a discussion of the predicates required under the Dodd-Frank Act for a proper Section 120 recommendation and conclude, based on the Report, that FSOC lacks the authority to make any recommendations regarding money market funds. We also analyze the misleading and incorrect statements the Council has used as a foundation for its case for fundamental changes to money market funds. We note that the Council’s proposed determination fails to reflect a nuanced and thoughtful analysis of the various types of money market funds and their characteristics, and demonstrate that no 15 FSOC itself even acknowledges that its proposals would have this impact. “[R]eforms that would provide meaningful mitigation of the risks posed by [money market funds] would likely reduce their appeal to investors.” See Report, supra note 2, at 29. 16 ICI has been an active participant in this dialogue; see Appendix A. Since 2009, ICI has produced a significant body of empirical research that bears directly on the question of money market fund reforms. Little of this research appears to have been taken into account, much less contested, in the drafting of the Report and its recommendations. To ensure that this research is before the Council and available for its consideration, we will be submitting a supplement to this letter with various key studies, reports, and surveys. Financial Stability Oversight Council January 24, 2013 Page 6 case can be made for further changes to the regulation of Treasury, government, and tax-exempt money market funds. Next, we describe the potential for temporary gates and liquidity fees to serve as effective tools to address redemption pressures in prime money market funds. We then outline our concerns with FSOC’s policy options for further money market fund reform: requiring money market funds to let their share prices float; implementing permanent restrictions in the form of a “minimum balance at risk” (“MBR”) requirement that would bar shareholders from redeeming all of their shares on demand; and requiring funds to maintain an explicit NAV buffer. We also discuss our views on possible additional measures, such as more stringent investment diversification requirements, increased minimum liquidity requirements, and shareholder transparency. Finally, we discuss how FSOC failed to properly evaluate the economic impact of its proposals under the statutory requirements of either the Dodd-Frank Act or the SEC’s rulemaking process. A summary of our comments follows. FSOC’s Determination Is Not Firmly Grounded in Law. The Council’s authority under Section 120 of the Dodd-Frank Act is expressly limited: it can only make recommendations regarding enhanced regulatory standards for a financial activity or practice conducted by “nonbank financial companies” or bank holding companies. Congress defined “nonbank financial company” generally to mean a company that is “predominantly engaged in financial activities.” Congress then expressly charged the Board of Governors of the Federal Reserve System (“Board”) with the responsibility to establish criteria necessary for applying this definition to specific companies—a task that the Board has not yet completed. This is more than a technical deficiency. It is indicative of the undue haste and lack of analytical rigor with which FSOC has pursued this matter. FSOC has not provided an adequate basis to support a determination that any money market funds would qualify as nonbank financial companies, and thus lacks the authority to issue these recommendations. Accordingly, we respectfully request that FSOC withdraw the proposed recommendations (Section II.A.). FSOC’s Determination Is Not Grounded in Fact. The Council’s basis for determining that “the conduct, scope, nature, size, scale, concentration, or interconnectedness” of money market funds pose systemic risks is also materially flawed in substance. The Report’s assertions about money market funds distort these funds’ nearly 40-year record of resilience, exaggerate the impact of money market funds on the financial crisis of 2007–2008, and ignore the substantial benefits of the 2010 SEC reforms. The Council’s determination also is based on the myopic premise that the features it ascribes to money market funds (e.g., risk-averse investors, lack of explicit loss-absorption capacity, and use of amortized cost accounting) are unique, and that any related risks are not attributable to the functioning of cash- management products or the short-term markets generally. Our comments demonstrate, with substantial empirical data, that this premise is incorrect, and point out that focusing attention on one product will not address broader systemic concerns in the short-term markets (Sections II.B.1.–3.). We also analyze the events of the 2007–2008 financial crisis, finding—consistent with the results of the SEC Staff Study—that many factors (including repeated shocks from failures by banks and other financial institutions and the lack of coherent, consistent government response to those failures) spurred redemptions from money market funds (Section II.B.4.). Last, we discuss the SEC’s 2010 reforms and Financial Stability Oversight Council January 24, 2013 Page 7 demonstrate that the Council’s concerns about the ability of money market funds to meet large-scale redemptions unquestionably reflect an out-of-date view of the industry that wholly ignores the 2010 amendments (Section II.B.5.). These misstatements and omissions are not merely incidental mistakes—they are the foundation of FSOC’s case for fundamental changes to money market funds. We strongly object to FSOC taking the drastic step of using its Section 120 authority based on faulty assumptions or data that do not reflect the current regulatory regime or actual market experiences of money market funds. Further Fundamental Changes Are Not Necessary for Treasury, Government, or Tax-Exempt Money Market Funds. FSOC’s proposed determination also fails to reflect a nuanced and thoughtful analysis of the various types of money market funds and their distinct risk profiles. As a result, FSOC in some instances proposes to recommend reforms broadly to all money market funds. In fact, there are four distinct types of money market funds—Treasury, government, tax-exempt, and prime funds—and each holds securities that trade in markets with varying degrees of liquidity, has somewhat different levels of default risk, and had distinct investor redemption experiences during the financial crisis of 2007–2008 and the events of 2011. Based on our analysis of the experience of each type of fund and the public record, it is abundantly clear that no case can be made for applying fundamental changes to Treasury, government, or tax-exempt money market funds. Even for prime money market funds, the measures FSOC proposes to recommend are wholly inappropriate and disproportionate to any theoretical threat (Section II.C.). We find it particularly troubling that FSOC—composed as it is of the heads of U.S. federal financial regulators—would see fit to propose drastic reforms for funds whose portfolios consist almost entirely of short-term Treasury and government securities. Absent implicit concerns about a default by the U.S. Government, these proposals seem wholly misplaced. If these proposals actually are motivated by such concerns, the implications for the financial system hardly can be confined to money market funds. Temporary Gates and Liquidity Fees Could Serve as Effective Tools to Address Redemption Pressures in Prime Money Market Funds. We do not believe the Report has made the case for further fundamental changes to money market funds. If, however, FSOC can demonstrate that changes are needed for prime money market funds, we would support FSOC’s recommending that the SEC propose requiring a prime money market fund to impose a liquidity “gate” if its “weekly liquid assets” fall to a specific, objective “trigger point,” set between one-quarter and one-half of the minimum weekly liquid asset level required by the 2010 amendments to Rule 2a-7 (i.e., weekly liquid assets at 7.5 percent to 15 percent of a fund’s assets).17 When a prime money market fund trips the trigger point, gates would automatically be imposed after the close of business to suspend redemptions received for processing the next business day. Money market fund boards then would be permitted to lift the gate and honor redemptions, provided that redeeming shareholders pay a nonrefundable liquidity fee to the fund equal to 1 percent of redemption proceeds—a level set to discourage redemptions, yet to allow investors truly in need of liquidity to have access to their funds. The redemption fee would benefit remaining 17 For a discussion of the 2010 amendments’ liquidity requirements, see Section II.B.1. below. Financial Stability Oversight Council January 24, 2013 Page 8 shareholders by mitigating liquidation costs and potentially rebuilding NAVs. We further suggest that prime funds be required to make frequent website disclosure of their mark-to-market share prices and weekly liquid asset levels to enhance transparency and encourage a highly conservative approach to portfolio management (Sections III.A.–B.). Our discussion addresses the experience of U.S. and European funds with redemption gates, tax and operational issues, and the impact of this proposal on certain types of money market fund transactions (Sections III.D.–G.). Our proposal differs from the Council’s MBR concept in that liquidity gates would not be imposed during “normal” market conditions, but only when a fund’s available weekly liquid assets fall to a specific threshold. In contrast to the MBR or FSOC’s other alternative recommendations, a liquidity- based trigger for gates aligns precisely with FSOC’s stated goal of stopping excessive or unexpected redemptions from a prime fund (in FSOC’s terminology, “runs”). This proposal has the immediate effect of suspending further redemptions and exacts a substantial cost for liquidity when liquidity is at a premium (Section III.C.). Requiring Floating NAVs Would Harm the Market. FSOC Alternative One would require all money market funds to let their NAVs float and to transact share purchases and redemptions at the portfolio’s daily mark-to-market value. ICI has maintained consistently since 2009 that proposals to force funds to float their NAVs reflect fundamental misunderstandings of the operation and role of money market funds, would increase systemic risk by driving investors away from money market funds to alternative products that strive to maintain stable values but that are not regulated under the Investment Company Act, and would disrupt well-established and efficient financing arrangements in the markets. FSOC’s proposal does not alleviate these problems. FSOC’s proposal would require money market funds to reprice their shares from $1.00 to $100.00 and would limit the use of amortized cost accounting for portfolio securities. These conditions, it contends, are “consistent with the valuation requirements that apply to all other mutual funds.” This statement is incorrect. FSOC’s recommendation in fact would require money market funds to comply with a pricing standard that is at least 10 times more onerous than the standard articulated by long- standing SEC accounting guidance. We question why sponsors would offer and investors would buy such funds (Section IV.A.). We then demonstrate, based on the experience of U.S. and European funds, that it is highly doubtful that forcing money market funds to float their NAVs would accomplish FSOC’s objective—inducing fund shareholders to refrain from reacting during periods of market stress (Section IV.B.). As FSOC acknowledges, forcing money market funds to float their NAVs would confront funds and investors with significant burdens in the tax and accounting treatment of gains and losses. While we offer suggestions for how the Internal Revenue Service (“IRS”), Treasury Department, SEC, and accounting authorities could alleviate these burdens, it is important to note that providing the specified relief would not cure FSOC’s proposal for floating NAVs of its significant shortcomings, nor justify FSOC’s recommending this alternative to the SEC. We stress that the necessary changes must be implemented before any floating NAV requirement takes effect. We also address the significant costs of operational and systems changes needed to implement a floating NAV, and the prospect that floating NAV funds would be unable to provide certain services to shareholders, including efficient processing of cash balances through sweep accounts (Section IV.C.). Financial Stability Oversight Council January 24, 2013 Page 9 In the face of these many burdens and barriers to the use of floating-NAV money market funds, the principal impact of FSOC Alternative One would be a major restructuring and reordering of intermediation in the short-term credit markets. It is very likely that institutional investors would continue to seek out diversified investment pools that strive to maintain a stable value. Most of these pools are not regulated under the Investment Company Act—and some of them lie beyond the jurisdictional reach of U.S. regulators. Regulatory changes that push assets from highly regulated, transparent products—i.e., money market funds—to less-regulated and less-transparent products arguably serve to increase systemic risk (Section IV.D.). Moreover, FSOC’s proposals for a transitional regime between stable value and floating NAV money market funds would be confusing and costly to investors; indeed, the transition itself could be destabilizing to the financial markets (Section IV.E.). The “Minimum Balance at Risk” Requirement Would Drive Investors and Intermediaries Away from Money Market Funds. FSOC Alternative Two proposes an untested experiment on $2.3 trillion in prime, tax-exempt, and government money market funds, requiring such funds, irrespective of current market conditions, to delay redemptions of a portion of shareholder accounts. We strongly oppose any sort of redemption restriction that would impair investor liquidity when liquidity is readily available within the money market fund. Alternative Two’s MBR restriction would impair a core mutual fund investor protection and reverse more than 70 years of SEC practice in fund regulation. Moreover, investor reaction to continuous redemption restrictions suggests that an MBR would greatly reduce investor use of money market funds. One survey of institutional investors indicates that institutional assets in money market funds would shrink by two-thirds if such restrictions were imposed. Although the Report asserts that an MBR would discourage shareholders from redeeming in times of stress, FSOC has not provided any data or analysis to support this assertion. To the contrary, discussions with investors indicate that shareholders would be more likely to redeem at the slightest sign of stress in the markets, given the punitive nature of the MBR (Section V.A.). An MBR also would create serious operational issues that would reduce or eliminate the usefulness of many services that money market funds and financial providers extend to investors. Drawing from a recent ICI study on the operational implications of an MBR-type proposal, our comments find that implementing FSOC’s proposed freeze on shareholders’ assets would require fund complexes, intermediaries, and service providers to undertake intricate and expensive programming and other significant, costly system changes.18 Our analysis indicates, however, that the costs of these changes could be so prohibitive that market participants are highly unlikely to undertake them, particularly if FSOC’s changes greatly curb investor interest in money market funds. FSOC’s proposal to exempt accounts with balances below $100,000 does not alleviate these burdens, instead, it would create an additional level of operational complexity and cost (Section V.B.). 18 See Investment Company Institute, Operational Impacts of Proposed Redemption Restrictions on Money Market Funds (2012) (“ICI Redemption Holdback Study”), available at http://www.ici.org/pdf/ppr_12_operational_mmf.pdf. Financial Stability Oversight Council January 24, 2013 Page 10 The likely consequences of an MBR requirement thus are mutually reinforcing. Fund complexes, intermediaries, and service providers would be hard-pressed to justify undertaking the significant costs of compliance with an MBR in the face of the rapid shrinkage of fund assets. We believe many intermediaries instead would make the business decision to migrate to unregulated or less-regulated money market investment vehicles or bank deposit products, disrupting short-term financing in the economy and increasing systemic risk. NAV Buffers and Capital Proposals Would Drive Sponsors from the Money Market Fund Product. FSOC Alternative Three contemplates that stable NAV money market funds would have a risk-based NAV buffer of up to 3.00 percent to provide an explicit loss-absorption capacity, potentially combined with other measures. A recent ICI study clearly shows the infeasibility of building capital at the levels suggested for either Alternative Two or Alternative Three, whether the capital is committed by fund advisers, raised in the market, or accumulated from fund income.19 Requiring money market fund advisers to commit capital to absorb possible future losses would alter fundamentally the business model of these funds, essentially requiring advisers to guarantee a portion of their funds. Rather than spreading small and infrequent losses across a large number of fund investors, an adviser-provided NAV buffer would concentrate losses on a single investor (the adviser) and on a small asset base (the NAV buffer). Fund advisers would require compensation for providing such guarantees, and the cost would be significant (Section VI.A.1.). Raising capital in the markets also faces formidable, if not insurmountable, hurdles. Working with capital market experts, ICI determined that adding subordinated debt or equity might require more than 560 individual money market funds to enter the market seeking to raise capital simultaneously. Small funds and small fund complexes likely would find it difficult and costly to issue and roll over subordinated securities, resulting in further industry consolidation and raising a barrier to entrants. Issuing subordinated debt also would add “rollover risk” to money markets funds, because investors in this class of money market fund shares might well be reluctant to roll over their investments in times of market stress. Thus, capital would disappear just when it might actually be needed—making such capital a source of instability in the markets (Section VI.A.2.). A third alternative—a within-fund capital buffer accumulated by retaining fund earnings—would be limited by legal and accounting considerations to 0.5 percent of fund assets. Capital at this level would not absorb large credit losses; at best it would provide funds somewhat greater flexibility in selling securities at a price below amortized cost. Even at that limited level, building such a buffer might take a typical prime fund 10 to 15 years (Section VI.A.3.). In sum, FSOC Alternative Three is a deeply flawed proposal. Its likeliest impact would be to impel money market fund sponsors to exit the business, thus depriving investors, issuers, and the economy of the benefits these funds provide. 19 See Investment Company Institute, The Implications of Capital Buffer Proposals for Money Market Funds (May 16, 2012) (“ICI Capital Buffer Study”), available at http://www.ici.org/pdf/ppr_12_mmfs_capital_buffer.pdf. Financial Stability Oversight Council January 24, 2013 Page 11 FSOC Failed to Meet Dodd-Frank or SEC Statutory Requirements for Economic Analysis. Under Section 120 of the Dodd-Frank Act, the Council must “take costs to long-term economic growth into account” when recommending new or heightened standards and safeguards for a financial activity or practice. Measured against this statutory mandate, the Report’s economic analysis has a number of significant shortcomings that exaggerate the potential benefits of the proposed reforms and may significantly underestimate their costs to the economy. The Report also fails to address—let alone satisfy—the SEC’s statutory and rulemaking requirements for analysis of the economic consequences of any eventual rule. We question why FSOC would use its Section 120 authority to propose recommendations without any consideration given to whether the recommended proposals will satisfy the SEC’s own governing statutes and other regulatory requirements. The Report’s discussion of the benefits of new regulations is flawed. The Council argues that its recommendations would reduce future outflows from money market funds during crises, which, in turn, would lower the probability and dampen the severity of any future crises. The Report, however, fails to show that the reforms it advocates would reduce risks; it ignores the salutary effects of the SEC’s 2010 amendments to Rule 2a-7; and it assumes that the regulatory system can ensure that investors in short- term markets will not react to vast, systemic events. The Council further assumes that money market funds have sufficient market power to compel fund investors (or, in some cases, issuers of short-term debt and intermediaries) to bear the costs and burdens of the Report’s recommended proposals. Given the numerous alternative products and services available to investors, particularly institutional investors, that is a wholly unrealistic assumption. As a result, the Report conveniently ignores the very high probability that its proposed fundamental changes will increase systemic risk by driving investors from money market funds into less-regulated, less-transparent cash management products (Section VII.A.). The Report’s analysis of the costs to long-term growth of its recommendations is highly speculative, perfunctory, and based on assumptions that are inconsistent with the Council’s assumed benefits. The Report asserts that its proposals’ costs to long-term economic growth are “very small.” These estimates are highly speculative and likely to be substantially understated. Curiously, this minimal estimate of the cost to long-term economic growth appears to contradict the Council’s own comments that money market funds “provide an economically significant service by acting as intermediaries between investors who desire low-risk, liquid investments and borrowers that issue short-term funding instruments.”20 The Council relies on models that fail to take seriously the role of financial intermediation—implicitly assuming that financial activity has no effect on real economic growth. Alternative estimates based on similar models yield estimates of costs to long-term economic growth that are seven or eight times greater than the Report’s figures. While these figures also are highly speculative, the absence of these alternative estimates from the Report suggests that FSOC was attempting to offer the lowest possible estimate of the cost, while ignoring the large uncertainties around its estimate. It is clear from this approach that the Council gave only the most perfunctory nod to its legal obligation under the Dodd-Frank Act to assess the cost of its proposals on long-term economic growth (Section VII.B.). 20 See Report, supra note 2, at 8. Financial Stability Oversight Council January 24, 2013 Page 12 II. Section 120 Requirements FSOC’s current action represents the first and only use of its authority under Section 120 of the Dodd-Frank Act, and thus merits particularly careful evaluation to ensure that it complies with the standards enacted by Congress. Section 120 imposes several statutory predicates for FSOC’s issuance of recommendations to financial regulators. Congress imposed these requirements to limit the exercise of the Section 120 process to specific circumstances and subject it to procedural safeguards intended to prevent it from being used arbitrarily. We submit that FSOC has failed to comply with all requirements necessary for a proper Section 120 recommendation. A. FSOC DOES NOT HAVE THE AUTHORITY TO MAKE RECOMMENDATIONS UNDER SECTION 120 REGARDING MONEY MARKET FUNDS FSOC’s authority to make recommendations regarding enhanced standards under Section 120 of the Dodd-Frank Act is expressly limited. It only may make recommendations with regard to a financial activity or practice conducted by “nonbank financial companies” or bank holding companies. FSOC has no basis to conclude, however, that money market funds are nonbank financial companies because a condition precedent—the adoption of definitional rules by the Board—has not occurred. Congress did not authorize FSOC to determine the requirements for a company to be deemed a nonbank financial company. Instead, Congress defined “nonbank financial company” generally to mean a company that is “predominantly engaged in financial activities.”21 Congress then expressly entrusted to the Board, in Section 102(b) of the Dodd-Frank Act, the responsibility to establish criteria necessary for applying this definition to specific companies: (b) DEFINITIONAL CRITERIA. – The Board of Governors shall establish, by regulation, the requirements for determining if a company is predominantly engaged in financial activities as defined in subsection (a)(6). (Emphasis added.) On February 11, 2011, the Board published for public comment a proposed rule that would establish the requirements for a determination that a company is predominantly engaged in financial activities (“Financial Activities Rule”).22 On April 10, 2012, it issued a supplemental notice of proposed rulemaking to clarify which activities the Board considers to be financial activities under Section 4(k) of the BHCA.23 To date the Board has not published a final rule. The protracted nature of this rulemaking, coupled with the critical nature of the comments received, indicates that it is far from clear which entities will meet this definition. 21 Congress established the general parameters for determining whether a company is predominantly engaged in financial activities: a company would be subject to such a determination if 85 percent of its gross annual revenues were derived from activities that are financial in nature as defined in Section 4(k) of the Bank Holding Company Act (“BHCA”), or 85 percent of its assets are related to activities that are financial in nature as defined in Section 4(k) of the BHCA. See Section 102(a)(6) of the Dodd-Frank Act. 22 76 FR 7731. 23 77 FR 21494. Financial Stability Oversight Council January 24, 2013 Page 13 Notwithstanding the fundamental importance of this rule in determining whether the government may exercise Section 120 authority, FSOC quite surprisingly made no reference in its Report to the Board’s pending Financial Activities Rule rulemaking proceeding, apparently substituting its judgment for that of the Board, contrary to Congress’s express intent in the Dodd-Frank Act. This is no mere technical deficiency. FSOC cannot make a legally valid determination that money market funds are nonbank financial companies for purposes of a recommendation under Section 120 unless the Board issues a final rule that would lead to the determination that money market funds are companies that are “predominantly engaged in financial activities.” In addition, as FSOC must be aware, commenters on the Board’s proposed Financial Activities Rule have pointed out that the Board has never determined for purposes of Section 4(k) of the BHCA that money market funds are engaged in a financial activity.24 Thus, the Board has not determined that money market funds are nonbank financial companies and may never reach that determination. In the Report, FSOC bases its authority to issue the recommendations on the following ground: The [FSOC] believes that [money market funds] are “predominantly engaged in financial activities” as defined in section 4(k) of the Bank Holding Company Act of 1956 and thus are “nonbank financial companies” for purposes of Title I of the Dodd-Frank Act.25 FSOC provides no explanation of how it arrived at its “belief.” FSOC merely includes a footnote that cites to four subsections of Section 4(k) of the BHCA without any discussion of how the cited subsections relate to the operations of the four types of money market funds that FSOC purports to cover with its proposed recommendations.26 By proceeding in this fashion, FSOC has acted both outside of its legal authority (by usurping rulemaking authority expressly assigned by Congress to the Board) and 24 See Letters from John D. Hawke, Jr., Arnold & Porter LLP, on behalf of Federated Investors, Inc. to Federal Reserve Board (March 30, 2011) and (May 24, 2012), available at http://www.knowledgemosaic.com/gateway/DoddFrankComments/FRB_SECRS_2011_April_20110401_R-1405_R- 1405_033011_69273_589557907011_1.pdf, at 7–9 and http://www.federalreserve.gov/SECRS/2012/October/20121017/R-1405/R- 1405_052412_107697_403310114866_1.pdf, at 4–6, respectively. 25 Report, supra note 2, at 15 (emphasis added, footnotes omitted). 26 Legitimate regulatory authority requires more than the mere recitation of statutory citations; it requires that the cited authorities actually stand for the proposition for which they are cited. Our analysis of these four subsections indicates that they do not support FSOC’s “belief.” Section 4(k)(1) refers to the Board’s authority to determine that an activity is financial in nature in consultation with the Secretary of the Treasury. The Board has not used this authority in regard to money market funds. Section 4(k)(4)(A) treats lending, exchanging, transferring, investing for others, or safeguarding money or securities as financial activities. FSOC does not explain which, if any, of these activities may be relevant to money market funds, or how Board precedent supports the application of such authority to money market funds. Section 4(k)(4)(D) contains authority to issue instruments representing interests in pools of assets permissible for a bank to hold. The Board has made it clear that it considers this provision as being directed at asset securitization transactions. See 77 FR 21494 at 21497, 21052. This authority does not relate to the operation of open-end investment companies, including money market funds. Finally, Section 4(k)(4)(H) provides merchant banking authority. FSOC provides no explanation as to why this authority is relevant to money market funds, nor does it make any suggestion that the Board has ever determined that the merchant banking authority encompasses the operation of a money market fund as a permissible financial activity under Section 4(k) of the BHCA, or that in any event, government debt securities would be subject to the merchant banking authority. Financial Stability Oversight Council January 24, 2013 Page 14 without the necessary explanation for a legally valid recommendation under Section 120 of the Dodd- Frank Act. As a result, it deprives the public and interested parties of a fair opportunity to provide meaningful comment regarding FSOC’s basis for its “belief.” In any reasonable administrative process, a clear explanation of an agency’s authority and rationale is fundamental to a fair process that can elicit meaningful comment. We submit that this threshold deficiency in FSOC’s use of its Section 120 authority is indicative more generally of the undue haste and lack of analytical rigor with which FSOC has approached this issue. In sum, FSOC has not provided an adequate basis to support a determination that money market funds would qualify as nonbank financial companies. It therefore lacks the authority to issue this recommendation proposing to treat money market funds as nonbank financial companies. Accordingly, to correct these substantive and administrative defects, we respectfully request that FSOC withdraw the proposed recommendations.27 B. FSOC’S DETERMINATION OF SYSTEMICALLY RISKY ACTIVITY UNDER SECTION 120 IS PREMISED ON MISLEADING AND INCORRECT STATEMENTS Under Section 120 of the Dodd-Frank Act, FSOC may issue a recommendation to a primary financial regulatory authority only if it determines that “the conduct, scope, nature, size, scale, concentration, or interconnectedness of [an] activity or practice [of a bank holding company or nonbank financial company] could create or increase the risk of significant liquidity, credit, or other problems spreading among bank holding companies and nonbank financial companies, financial markets of the United States, or low-income, minority, or underserved communities.” FSOC proposes to determine that money market funds’ activities and practices could create or increase these risks. Specifically, FSOC finds that the conduct and nature of money market fund activities and practices make them susceptible to “destabilizing runs” that can spread quickly among funds, impairing liquidity broadly and curtailing the availability of short-term credit. As support for this determination, FSOC points to various money market fund practices and activities; the funds’ interconnectedness with financial firms, the financial system, and the U.S. economy; “evidence” from the 2007–2008 financial crisis; and the limits of the SEC’s 2010 reforms. These assertions about money market funds distort the nearly 40-year record of their resilience, ignore the substantial impact of the 2010 SEC reforms, and exaggerate the impact of money market funds on the financial crisis. These misstatements are not merely incidental mistakes—they are the foundation of FSOC’s case for fundamental changes to money market funds. The Council’s determination also is based on the myopic premise that the features FSOC ascribes to money market funds as a product are unique, and not attributable to the functioning of cash- management products or the short-term markets generally. This premise is incorrect, and focusing 27 Notwithstanding our view that FSOC lacks the authority to issue a Section 120 recommendation for money market funds, we provide comments below on the rest of the Report. Financial Stability Oversight Council January 24, 2013 Page 15 attention on one product in the short-term markets will not address the broader systemic concerns identified by FSOC. 1. Conduct and Nature of Money Market Funds’ Practices and Activities Individuals and institutions have cash management needs that cannot entirely be met by insured bank deposits. At the same time, some borrowers have credit needs that are best fulfilled by issuing short-dated instruments. The money markets have long served as a means of bringing together cash investors and issuers of short-term debt instruments. Money market funds serve as one means of providing access to these markets, but are certainly not the only product or service to do so. For example, institutional investors often have cash holdings that exceed deposit insurance limits and therefore use money market instruments to invest their cash. These investors include large corporations, securities lending operations, bank trust departments, securities brokers, investment managers, and state and local governments. They use a variety of products and services in addition to money market funds to invest in the money markets, including uninsured bank deposits, trust accounts, separate accounts, short-term investment funds, local government investment pools, offshore money funds, and unregulated investment funds. Retail investors also invest in money market funds. These investors can access money market funds through brokerage accounts and pension funds, where insured bank deposits may not be options offered to them. Retail investors that have balances above the insured deposit limit also may choose to invest in money market funds rather than in a single, uninsured deposit at a bank or multiple insured accounts. In contrast to “posted” bank deposit rates that are set by bank personnel and tend to be far less responsive to market interest rates, money market funds provide retail investors with access to money market instruments with market-based yields. The Report identifies five activities and practices of money market funds that FSOC believes combine to create their alleged vulnerability to runs: attracting a base of highly risk-averse investors; investing in assets that are subject to interest rate and credit risk without having explicit loss-absorption capacity; relying on the amortized cost method of valuation and penny rounding to maintain a $1.00 per share price; offering shares that are redeemed on demand; and relying on discretionary support from sponsors. The Report generally fails to identify, however, how these activities and practices distinguish money market funds from other products or services that invest in money market instruments or the money markets themselves. This focus on one single product promises seriously flawed policy and will not address issues that are in fact features of the short-term markets or their investors in general. Money Market Fund Investors Are Not Uniquely Risk Averse. The Report states that money market funds’ successful track record of maintaining stable NAVs combined with the funds’ low risk investment strategies has attracted highly risk-averse investors. The Report does not provide any analysis demonstrating that investors in money market funds are unique or more risk averse than other investors in the money markets, or cash investors generally (including bank depositors). Given the nature of cash investors, and the wide range of products and services providing access to these markets, money market fund investors are likely to have the same risk tolerances as other investors in these markets. Indeed, cash Financial Stability Oversight Council January 24, 2013 Page 16 holdings continuously flow back and forth between money market funds and other types of cash management products and services. For example, the share of nonfinancial businesses’ short-term assets that are held in money market funds has fluctuated between 18 and 36 percent since the mid-1990s. As shown in Figure 1, at the end of the third quarter of 2012, nonfinancial businesses held $2.4 trillion in short-term assets, and money market funds managed 21 percent of these assets. State and local governments invest slightly more than $700 billion in short-term assets, and money market funds managed 18 percent of these cash assets. Thus, money market funds are only one product among many through which risk-averse investors can invest their short-term assets. A recent study by Treasury Strategies, Inc. further demonstrates that among institutional investors, money market funds are but one means of investing their cash.28 FIGURE 1 Money Market Funds’ Share of U.S. Nonfinancial Businesses’ Short-Term Assets1 Percent * Data for 2012 is through the third quarter. 1 U.S. nonfinancial businesses’ short-term assets consist of foreign deposits, checkable deposits, time and savings deposits, money market funds, repurchase agreements, and commercial paper. Source: Investment Company Institute and Federal Reserve Board   28 See Money Market Fund Regulation: The Voice of the Treasurer (April 2012) (“TSI Survey”), available at http://www.ici.org/pdf/rpt_12_tsi_voice_treasurer.pdf, at 34–35. ICI commissioned Treasury Strategies, Inc. to conduct a study to help understand the effects on money market fund investors of various SEC reform concepts. Treasury Strategies surveyed 203 unique corporate, government, and other institutional investors between February 13 and March 6, 2012, asking 31 questions regarding their cash pools, investment objectives, and three SEC concepts for money market fund reform—floating NAVs, capital NAV buffers, and redemption holdback restrictions. Treasurers and other institutional investors are significant users of money market funds: institutional share classes account for $1.7 trillion, or 65 percent, of the $2.7 trillion in U.S. money market fund assets. 18 21 22 21 28 28 23 21 21 23 28 36 30 24 22 21 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012* Financial Stability Oversight Council January 24, 2013 Page 17 Money market funds, therefore, are not unique in providing a means for risk-averse investors to invest in the short-term markets. In fact, as discussed in Section II.B.4., during the 2007–2008 financial crisis, other investors accounted for a larger and earlier pullback from the money markets than did money market funds. As a result, providing recommendations with respect to this product will do little to address FSOC’s concerns about investors in the money markets. Money Market Funds Can and Do Absorb Losses. FSOC asserts that money market funds do not have any formal capacity to absorb losses. FSOC, using a bank-centric approach, equates “loss absorbing capacity” with a capital buffer. According to FSOC, money market funds have no “loss absorbing capacity” because they are not required to have a dedicated capital buffer. This assertion is incorrect on several counts. First, money market funds, like all other mutual funds, absorb losses in the value of underlying portfolio securities through changes in funds’ share prices. If the losses on a money market fund’s portfolio exceed one-half cent per share ($0.005), those losses are reflected as a reduction in the fund’s share price. Thus, the reduction in share price is passed through to fund shareholders, who provide the “loss absorbing capacity.” This is exactly the same mechanism that is in place for all capital market securities (including long-term mutual funds): investors absorb losses. Indeed, despite its assertion that money market funds have no capacity to absorb losses, FSOC implicitly acknowledges that money market funds do in fact have such capacity. Alternative One in the Report would require money market funds to have a floating NAV but, importantly, no capital buffer. The lack of a capital buffer under Alternative One implies that FSOC believes changes in a fund’s share price transfer losses to shareholders, thus providing “loss-absorbing capacity.” Floating the NAV may change the timing of when such losses are absorbed by fund shareholders, but both stable and floating NAV money market funds use the same mechanism, and have the same capacity, to “absorb losses.” Second, this assertion implies that any set of rule changes or reforms can eliminate interest rate and credit risks of money market instruments or shares of money market funds. Money market funds invest in securities that have some degree of interest rate and credit risks. FSOC cannot completely eliminate credit losses or interest rate fluctuations on short-term investments, however, whether those instruments are held by money market funds, in another product or account, or directly. None of the FSOC recommendations will eliminate credit and interest rate risks from the market. Third, FSOC suggests that investors are seeking to invest in securities that have no interest rate or credit risks. Direct owners of money market instruments bear interest rate and credit risks, as do investors in short-term cash pools, and therefore interest rate and credit risks are not unique to money market funds. Rather, investors in the money markets generally will react to new information about changes in the riskiness of a debt issuer or group of issuers by selling or not rolling over securities issued by those issuers. Identifying concerns about risk as a unique feature of money market funds and crafting “reforms” that push money market fund investors into other products does not remove those investors’ exposure to interest rate and credit risks in the short-term markets. Financial Stability Oversight Council January 24, 2013 Page 18 The Report provides no evidence that investors believe they are investing in a riskless security. In fact, data from actual shareholders shows that investors understand that there is risk in money market funds.29 All mutual funds, including money market funds, are required to provide risk disclosures in their prospectuses. In addition to these risk disclosures, money market funds also must prominently disclose the following in their prospectuses and any advertisements: An investment in the [f]und is not insured or guaranteed by the Federal Deposit Insurance Corporation or any other government agency. Although the [f]und seeks to preserve the value of your investment at $1.00 per share, it is possible to lose money by investing in the [f]und. In light of money market funds’ experience during the financial crisis, the MMWG Report recommended that money market funds evaluate whether their disclosures, including advertising and marketing materials, and in particular their risk disclosures, fully capture the risks that money market funds may present and that these funds, if appropriate, revise their disclosures.30 Although many money market fund complexes voluntarily evaluated the adequacy of their own risk disclosures after the MMWG recommendation, the SEC inexplicably did not adopt this recommendation as part of the 2010 rule amendments. If regulators remain concerned about the level of investor understanding regarding the risks of investing in a money market fund and the potential for losses, despite clear indications that investors do understand these risks, the most direct remedy would be for regulators to focus on changing money market funds’ narrative risk disclosure requirements. The 2010 amendments also made money market fund portfolios far more transparent to both shareholders and regulators. Current disclosure standards require a fund to report details every month on every security it holds, every piece of collateral backing repurchase agreements, its mark-to-market NAV, and a wide range of other salient information on Form N-MFP.31 Institutional investors have found this data invaluable in monitoring holdings of their funds and encouraging those funds to minimize credit risks. This heightened scrutiny has at times led regulators and analysts to highlight 29 Notably, surveys have found, for example, that four out of five retail investors know that securities held by money market funds fluctuate in value, and 83 percent recognize that money market fund investments carry as much or more risk as bank accounts. See Letter from Scott C. Goebel, Senior Vice President and General Counsel, FMR Co., to Elizabeth M. Murphy, Secretary, Securities and Exchange Commission (April 26, 2012) (“Fidelity Survey”), available at http://www.sec.gov/comments/4-619/4619-170.pdf. In particular, the Fidelity Survey found that 81 percent of Fidelity retail customers with money market funds indicate that they understand that the securities held by these funds fluctuate up and down daily in value; 75 percent of Fidelity customers know that the money market funds they invest in are not guaranteed by the government; only 10 percent believe the government would step in to prevent money market funds from breaking a stable $1.00 share price; and the majority of customers do not favor further regulation of money market funds, but instead would support additional investor education. Of course, institutional investors—corporations, state and local governments, financial firms, retirement plans, and others—are well informed, given the key role money market funds play in their daily operations. 30 See MMWG Report, supra note 4, at 91–92. 31 Some funds voluntarily are providing more portfolio holdings and mark-to-market share value disclosure than what is required. See, e.g., Ronald D. Orol, Money funds moving to publish NAVs daily: Fidelity, Federated and Schwab plan to disclose daily NAVs shortly, MarketWatch (January 11, 2013), available at http://articles.marketwatch.com/2013-01- 11/economy/36274271_1_money-market-funds-money-market-frequent-disclosure. Financial Stability Oversight Council January 24, 2013 Page 19 potential risks in particular fund holdings. It also has led certain advisers to avoid investments that, although exhibiting stable credit fundamentals, may raise investor concerns.32 These observations are reflected in the SEC Staff Study, which asserted that “increased transparency, even if reported on a delayed basis, might dampen a fund manager’s willingness to hold securities whose ratings are at odds with the underlying risk, especially at times when credit conditions are deteriorating.”33 Thus, the discipline of far greater disclosure, consistent with the SEC’s historical, disclosure-based approach to protecting investors, in itself has had a strong palliative effect. We are deeply concerned that FSOC’s assertion that investors may not understand that money market funds could lose value reflects a view of money market funds predating the 2010 amendments, and does not demonstrate a complete understanding of the product and the vast array of information that is now available to investors and regulators. Amortized Cost Accounting and the Stable, Rounded NAV Per Share Do Not Give Investors a False Impression That Their Investment Is Guaranteed. The Report contends that the valuation and rounding methods used by money market funds “obscures the daily movements in the value of [a money market fund’s] portfolio and fosters expectation that [money market fund] share prices will not fluctuate.”34 Although various commentators have made this assertion, the claim has never been substantiated in any way. In fact, survey evidence submitted to the SEC35 and recent Senate testimony on money market funds demonstrates emphatically that investors do understand that fund share prices are not guaranteed. As the Maryland State Treasurer recently testified at a Senate hearing on money market funds, “[O]n behalf of many of the investors … [w]e do read the prospectus and we know it’s an investment. … So I think this treating us sort of like children is really not appropriate.”36 The Report’s assertion also overlooks the fact that the stability of a money market fund’s share price is largely attributable to the short-duration, high-quality nature of its portfolio securities and not to amortized cost accounting or rounding. Mark-to-market share prices typically do not deviate significantly from $1.00. From January 2011 to July 2012, for example, 96 percent of prime money market funds recorded an average absolute monthly change in their mark-to-market value of 1/100th of a cent (1 basis point, or $0.0001) or less.37 Indeed, even during the summer of 2011, when the short- term markets were buffeted by the eurozone sovereign debt crisis, the U.S. debt ceiling impasse, and the 32 See N. Flanders, G. Fink-Stone, and V. Baklanova, U.S. MMF Shadow NAV Volatility Declines Post-Crisis, Fitch Ratings (January 18, 2012) (“Fitch Ratings’ Special Report”). 33 See SEC Staff Study, supra note 10, at 38. 34 Report, supra note 2, at 19. 35 See Fidelity Survey, supra note 29. 36 See The Honorable Nancy Kopp, Treasurer, State of Maryland, Perspectives on Money Market Mutual Fund Reform, hearing before the U.S. Senate Committee on Banking, Housing and Urban Affairs (June 21, 2012). 37 See S. Collins, E. Gallagher, J. Heinrichs, and C. Plantier, “Money Market Mutual Funds, Risk, and Financial Stability in the Wake of the 2010 Reforms,” ICI Research Perspective (January 2013) (“ICI Research Perspective”), available at http://www.ici.org/pdf/per19-01.pdf. Financial Stability Oversight Council January 24, 2013 Page 20 downgrade of the U.S. government’s long-term debt, money market fund mark-to-market share prices did not deviate significantly from $1.00.38 Additional evidence that valuations determined from amortized cost accounting are frequently close to the market value of securities comes from actual sell transactions. Based on a sample of over 40,000 sell trades conducted by prime and government money market funds from January 2007 to October 2012,39 the market price obtained in a little over 53 percent of the trades was within one-half of a basis point (0.005 percent) of the amortized cost of the security (Figure 2). Deviations between market price and amortized cost reflect dealer transaction costs and changes in interest rates and credit quality. In non-stressed markets, dealer transaction costs tend to be small, usually within 1 to 3 basis points at an annual rate. For nearly 80 percent of the trades, the market price was within 2.5 basis points of amortized cost. Also, the data suggest that when money market funds need to sell securities, they sell securities that have appreciated in value. For 39 percent of the trades, the market price was greater than the amortized cost of the security by 0.5 basis points or more. FIGURE 2 Distribution of Deviations of Amortized Cost from Market Price Percentage of sell trades by taxable money market funds,* 2007–2012 * Number of trades = 40,057 Source: Confidential data from sample of taxable money market funds   38 See SEC Staff Study, supra note 10, at 33–34. 39 Confidential sell trade data were submitted to the Investment Company Institute by a sample of taxable money market funds comprising 45 percent of the industry’s total net assets. Trades include sales of commercial paper, Treasury and agency securities, certificates of deposit, and corporate notes. 0.2 0.2 0.1 0.4 0.8 1.5 1.6 3.2 53.2 15.5 6.1 7.9 4.2 2.1 1.2 1.7 0.3 0 10 20 30 40 50 60 Deviation (basis points) Financial Stability Oversight Council January 24, 2013 Page 21 Furthermore, and somewhat disingenuously, the Report fails to acknowledge that amortized cost accounting is a well-established valuation method for short-term securities generally.40 Under Generally Accepted Accounting Principles (“GAAP”), companies value cash equivalents at amortized cost. Cash equivalents are short-term, highly liquid investments that are both readily convertible to cash and so near their maturity that they present insignificant risk of change in value because of changes in interest rates. Generally, only securities with original maturities of three months or less qualify as cash equivalents.41 GAAP recognizes that for these short-term, highly liquid securities, amortized cost and fair value are substantially the same. Further, GAAP permits long-term securities to be valued at amortized cost, provided the company has the intent and ability to hold the security to maturity.42 Reflecting these standards, corporate issuers, including financial institutions such as banks and insurance companies, use amortized cost to varying degrees to value securities holdings. Even federal government agencies at times use amortized cost to value assets. For example, the Federal Reserve System uses amortized cost to value all of its holdings of Treasury and agency securities, and the Federal Deposit Insurance Corporation uses amortized cost to value securities held by the National Liquidation Fund. The Report also fails to acknowledge that investors always have some incentive to redeem out of or sell any financial product—be it a bank deposit, a stock, a bond, or any other instrument—if they fear losses. Indeed, as discussed in Section II.B.4., the driving factor in investors’ withdrawal from money market funds during the financial crisis was almost certainly the rapid and unprecedented deterioration of the banking system worldwide—and not the funds’ stable NAV pricing structure. In fact, the SEC Staff Study found that there are many possible explanations for the redemption activity during the 2008 financial crisis, including factors such as investors preferring the safety, liquidity, and transparency of government securities.43 With yields on Treasury bills falling and demand for longer-dated commercial paper collapsing on September 15, 2008, the financial markets showed a general flight to government securities before there were significant outflows from prime money market funds. Seeking to deflect attention from the unprecedented deterioration of the banking sector by pointing to the use by money market funds of a long-standing, commonly accepted accounting technique simply makes for bad policy. Today’s Money Market Funds Are the Most Liquid in History. The Report suggests that the liquidity, maturity, and credit transformation of money market funds is a cause for concern. But the degree of such transformation has always been modest and became even more modest as a result of the 2010 amendments.44 Today’s money market funds are stronger and more resilient than the funds that were available in 2008. FSOC must analyze these funds as they exist today, not through the outdated lens of 2008. The 2010 amendments directly and meaningfully addressed this “transformative” concern in two ways. 40 See generally D. Beresford, Amortized Cost Is “Fair” for Money Market Funds, Center for Capital Markets Competitiveness (Fall 2012), available at http://www.uschamber.com/sites/default/files/reports/money_market_funds_report.pdf. 41 FASB Accounting Standards Codification 305-10-20. 42 FASB Accounting Standards Codification 320-10-35. These “held to maturity” securities are subject to impairment testing. If the security is impaired, then its value will be decreased and the company will recognize a charge against earnings. 43 See SEC Staff Study, supra note 10, at 7-9. The 2010 reforms addressed concerns regarding both liquidity and transparency. 44 See Mark Hannam, Money Market Funds, Bank Runs and the First-Mover Advantage, Institutional Money Market Funds Association (January 2013), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2187818, at 7. Financial Stability Oversight Council January 24, 2013 Page 22 First, the amendments met the liquidity challenge faced by many money market funds during the financial crisis by imposing for the first time explicit minimum daily and weekly liquidity requirements. Under the new requirements, money market funds must maintain a sufficient degree of portfolio liquidity to meet reasonably foreseeable redemption requests. In addition, all taxable money market funds are required to hold at least 10 percent of their portfolios in assets that can be turned into cash within a day, and all funds must hold at least 30 percent in assets that are liquid within a week. The amendments also require funds, as part of their overall liquidity management responsibilities, to have “know your investor” procedures to help fund advisers anticipate the potential for heavy redemptions and adjust their funds’ liquidity accordingly, and to have procedures for periodic stress testing of their funds’ ability to maintain a stable NAV. The SEC Staff Study found that the new liquidity requirements have made money market funds “more resilient to both portfolio losses and investor redemptions.”45 In practice, prime money market funds have exceeded the liquidity minimums by a significant margin, and now hold twice as much in weekly liquid assets as the heaviest redemptions they faced in the worst week of the financial crisis in September 2008.46 Indeed, the ongoing fragility of the markets since the 2007–2008 crisis—attributable to a variety of factors, including regulatory uncertainty, the U.S. federal debt ceiling crisis in mid-2011, deteriorating conditions in European debt markets, the recent “fiscal cliff” negotiations to avoid sharp tax increases and cuts in government spending, and the U.S. government’s extension until the end of 2012 of unlimited deposit insurance on non-interest bearing checking accounts, which provided depositors a guarantee on business checking account balances held at banks47—has prompted many money market fund managers to hold larger amounts of liquidity as a way to mitigate risks.48 As Figure 3 shows, as of June 2012, 31 percent of the assets of prime money market funds were in daily liquid assets and 46 percent of their assets were in weekly liquid assets. In dollar terms, taxable money market funds held an estimated $1.38 trillion in weekly liquid assets, which includes an estimated $629 billion held by prime money market funds. In comparison, during the business week September 15–19, 2008 (the week Lehman Brothers Holdings Inc. (“Lehman”) failed), prime money market funds experienced estimated outflows of $310 billion. 45 See SEC Staff Study, supra note 10, at 37. 46 See PWG Report, supra note 7, at 12. 47 See Federal Deposit Insurance Corporation, Deposit Insurance Regulations; Unlimited Coverage for Noninterest-Bearing Transaction Accounts, 75 FR 69577 (November 15, 2010). As required by Section 343 of the Dodd-Frank Act, the unlimited insurance coverage became effective on December 31, 2010, and expired on January 1, 2013. We are pleased that this program expired as expected, as we view this type of program as having the potential to dislocate markets and increase systemic risk in times of market stress by creating an unlimited taxpayer-supported backstop for these transaction accounts. Programs that create and sustain such moral hazard have no place in our markets. See Letter from Karrie McMillan, General Counsel, Investment Company Institute, to Robert E. Feldman, Executive Secretary, Federal Deposit Insurance Corporation (October 10, 2010), available at http://www.fdic.gov/regulations/laws/federal/2010/10c48AD37p.PDF. 48 We note that the liquidity levels fund managers choose to hold ebb and flow based upon market conditions and that current liquidity levels may not be necessary or desirable in the future. Financial Stability Oversight Council January 24, 2013 Page 23 FIGURE 3 Liquid Assets for Taxable Money Market Funds Percentage of total assets, June 2012 1 Daily liquid assets include securities with a remaining maturity of one business day and Treasury securities. 2 Weekly liquid assets include securities with a remaining maturity of five business days or less, Treasury securities, and agency securities with a remaining maturity date of 60 days or less. Source: Investment Company Institute tabulations of SEC Form N-MFP data Moreover, the liquid assets that now make up much of prime money market funds’ portfolios are overnight repurchase agreements and Treasury and other government securities—exactly the types of securities that anxious investors want to buy in a crisis and the types of assets that government money market funds hold. As the Report acknowledges, for every dollar that flowed out of prime money market funds in September 2008, 61 cents went back into Treasury and government money market funds. In a future crisis, to match investors’ shifting demands, government money market funds and other investors would be ready buyers of many of the liquid assets that prime funds wish to sell. This is in sharp contrast to 2008, when prime money market funds held far fewer Treasury and agency securities and sought to sell commercial paper and similar assets that did not have a ready market in the wake of a wave of financial institution failures. In fact, many of the asset classes that make up prime money market fund portfolios today constitute “high-quality liquid assets” for purposes of the Basel III Liquidity Coverage Ratio (“LCR”). According to the Basel Committee on Banking Supervision, LCR-eligible assets have the following liquidity-related characteristics: (i) they are traded in active and sizeable markets; (ii) they have committed market makers; (iii) they have low market concentration; and (iv) they are “flight to quality” assets, i.e., “historically, the market has shown tendencies to move into these types of assets in a systemic 66 31 85 46 Government Prime Daily liquid assets Weekly liquid assets 1 2 30% weekly requirement 10% daily requirement Financial Stability Oversight Council January 24, 2013 Page 24 crisis.”49 While FSOC claims the liquidity of money market funds is a “concern,” the Basel Committee has proposed that internationally active banks be required to hold the same assets that money market funds hold to protect against illiquidity. Indeed, the overnight repurchase agreements and Treasury and other government securities that now make up much of prime money market funds’ portfolios are precisely the asset classes favored by the LCR framework.50 Second, in addition to the new liquidity requirements, the 2010 amendments require that a money market fund’s weighted average maturity (“WAM”) and weighted average life (“WAL”) cannot exceed 60 and 120 days, respectively.51 The SEC Staff Report found that the new maturity limits have “improved the resiliency of money market funds to interest rate shocks.”52 These requirements reduce liquidity and maturity transformation to very low levels. In practice, money market funds exceed these requirements. For example, although data on WALs before November 2010 are not publicly available, data disclosed since then suggest that the new WAL requirement likely has bolstered the resilience of funds. Figure 4 depicts the distribution of WALs for taxable money market funds as of June 2012. Although the maximum allowable WAL is 120 days, most funds are well below this, with the great majority having WALs in the range of 30 to 90 days. Only a very small proportion of funds have WALs in excess of 100 days. 49 See Basel Committee on Banking Supervision, Basel III: International Framework for Liquidity Risk Measurement, Standards and Monitoring 5 (December 2010), (“Basel III Liquidity Standards”) available at http://www.bis.org/publ/bcbs188.pdf (emphasis added). 50 Id. at 8 (permitting banking organizations to hold unlimited amounts of “Level 1” assets for purposes of the LCR, which includes claims on or claims guaranteed by sovereigns). 51 The introduction of a limit on money market funds’ WAL has strengthened the ability of money market funds to withstand shocks and meet redemption pressures. Unlike a fund’s WAM, a portfolio’s WAL is measured without reference to interest rate reset dates. The WAL limitation thus restricts the extent to which a money market fund can invest in longer- term adjustable-rate securities that may expose a fund to credit risk. 52 See SEC Staff Study, supra note 10, at 30. Financial Stability Oversight Council January 24, 2013 Page 25 FIGURE 4 Weighted Average Life (WAL) for Taxable Money Market Funds Percentage of funds, June 2012 Weighted average life in days Source: Investment Company Institute tabulations of SEC Form N-MFP data The third way in which the 2010 amendments addressed concerns about redeemable shares was through the creation of a powerful new tool for money market fund boards of directors. If a money market fund cannot meet redemptions without breaking the dollar, the 2010 amendments, through new Rule 22e-3 under the Investment Company Act, allow the fund’s board to liquidate the fund in an orderly manner—without a fire sale of portfolio securities or a first-mover advantage for early redeemers.53 In September 2008, the Reserve Primary Fund’s board did not have the ability to promptly suspend redemptions—leading to a chaotic response when the fund broke the dollar. Now, the SEC has given money market fund boards a mechanism that will, in the SEC’s own words, allow for the “orderly liquidation of fund assets” for a troubled fund and “reduce the vulnerability of investors to the harmful effects of a run on the fund, and minimize the potential for disruption to the securities markets.”54 To use this power, a board must decide to liquidate the fund. By suspending redemptions, the board helps protect all shareholders and ensures that “sophisticated” investors can’t exit first and inflict 53 The board continues to have the option to instead reprice the fund’s shares and allow the fund to remain open but with a floating NAV. 54 See 2010 MMF Reform Release, supra note 5, at 10088. Financial Stability Oversight Council January 24, 2013 Page 26 losses on those remaining behind. The new rule recognizes that a money market fund’s share price can decline in value, and provides for an orderly liquidation of the fund’s securities in a manner that best serves all of the fund’s shareholders.55 Again, FSOC’s assertion that these funds have “limited liquidity” to meet redemptions belies the current structure of these funds and the important reforms embodied in the 2010 amendments. As shown above, funds have significant levels of liquidity. Discretionary Sponsor Support Does Not Confuse Investors. The Report asserts that the use of discretionary sponsor support to maintain stable NAV prices has obscured some investors’ appreciation of money market fund risks and caused some investors to assume that money market fund sponsors will absorb any losses, even though they are under no obligation to do so.56 It contends that perceptions about sponsor support have contributed to uncertainty among market participants about who will bear losses when they do occur—likely making money market funds prone to large, unexpected outflows. These assertions are without support and contrary to the regulatory history surrounding sponsor support. Implicit in these assertions is the suggestion that investors do not understand or are oblivious to the risks, no matter how small, of money market fund investments. There is nothing in the Report, however, suggesting that FSOC (or the SEC for that matter) has surveyed money market fund investors, gauged their attitudes, or analyzed their behavior. On the other hand, as noted above, recent research indicates that investors are well aware of the risks associated with money market funds.57 Moreover, the SEC consistently has defended disclosure and investors’ understanding of money market funds—even when sponsor support for money market funds was at issue. In 1996, the SEC adopted Rule 17a-9 under the Investment Company Act, an exemptive rule permitting purchases of certain money market fund portfolio securities by affiliated persons under specified conditions.58 The rule codified a series of staff no-action letters in which the SEC staff agreed not to recommend enforcement action if affiliated persons of a money market fund purchased portfolio securities from the fund in order to prevent the fund from realizing losses on the securities. At the time, some commenters (including ICI) opposed the adoption of Rule 17a-9 because of concerns that the mere existence of such a rule would cause investors to expect a fund’s adviser to buy out troubled securities from the fund, thus “guaranteeing” that the fund will maintain a stable NAV. In response, the SEC stated in the 1996 adopting release that “existing rules applicable to money funds 55 If FSOC remains concerned that funds may not have sufficient liquidity to meet redemptions, it should give careful consideration to suggestions for “gating” (temporarily suspending redemptions) or liquidity fees, discussed later in Section III. 56 It is curious that the Report asserts that money market fund sponsors have provided “over 200” instances of discretionary sponsor support since 1989, yet only cites authorities that identify far fewer instances. See Report, supra note 2, at 20. 57 See Fidelity Survey, supra note 29. 58 See Revisions to Rules Regulating Money Market Funds, SEC Release No. IC-21837 (March 21, 1996), 61 FR 13956 (March 28, 1996) (“1996 MMF Reform Release”). Financial Stability Oversight Council January 24, 2013 Page 27 already address this concern by requiring money fund prospectuses and sales literature to disclose prominently that there is no assurance or guarantee that a fund will be able to maintain a stable net asset value of $ 1.00 per share. Moreover, the Commission believes it unlikely that the existence of an exemptive rule alone will create any investor expectations.”59 The SEC made the same arguments in 2010, two years after the Reserve Primary Fund broke the dollar. At that time, it amended Rule 17a-9, making it even easier for a sponsor to offer support by buying securities out of a money market fund portfolio. The SEC stated that the amendments would not “materially change shareholders’ perceptions about money market funds or the likelihood of sponsor support during times of market turmoil.”60 Rather, the SEC noted that affiliated sponsor support “transactions appear to be fair and reasonable and in the best interests of fund shareholders.”61 Given actual investor understanding and multiple findings by the primary financial regulatory agency for money market funds, there is no valid basis for FSOC to use alleged investor confusion about sponsor support as evidence for the notion that money market funds pose systemic risk. Money Market Funds Merely Reflect the Underlying Characteristics of Their Holdings. The Report concludes by suggesting that in combination, the activities and practices of money market funds tend to exacerbate each other’s effects and increase money market funds’ vulnerability to “runs.” It then suggests that policy responses that diminish these “interactions” hold promise for mitigating the risks that the Report alleges that money market funds pose. The Report fails to consider, however, that the money markets as a whole are highly interconnected with the rest of the financial system. For example, financial firms issue certificates of deposit and repurchase agreements that total $4.2 trillion into the money markets. In addition, nearly 80 percent or $800 billion of the commercial paper outstanding is issued by financial firms as financial commercial paper or asset-backed commercial paper (“ABCP”). “Fixing,” or shrinking, money market funds will not address challenges that arise from financial institutions’ dependence on the money markets because of the nature of these markets themselves. The Report fails to consider how stresses among a group of money market issuers, such as financial institutions in 2007–2008, can be transmitted to other sectors of the market. For example, two thirds of the primary dealers in the U.S. Treasury debt auctions have foreign parents. Primary dealers also are one of the principal groups of borrowers in the repurchase agreement market. Shocks to one or several of these dealers could significantly disrupt the operations of the repurchase agreement market. Money market funds are often cited for withdrawing their funding from the repurchase agreement market during the 2007–2008 financial crisis, but academic research has found that money market funds were not the source of contraction in these markets.62 Rather, the contraction was largely driven by dealers 59 Id. at 13974 (emphasis added). 60 2010 MMF Reform Release, supra note 5, at 10087. 61 Id. 62 See G. Gorton and A. Metrick, Who Ran on Repo? (2012), available at http://ssrn.com/abstract=2157174. Financial Stability Oversight Council January 24, 2013 Page 28 and other entities that were no longer willing to provide financing to one another. Again, focusing on a single product, rather than the money markets as a whole, will lead to a distorted policy response. As we discuss below in Sections IV, V, and VI, FSOC’s recommended policy alternatives ignore these issues and would not in fact address any of the risks attributed by the Report to the conduct and nature of money market funds’ practices and activities. We strongly object to FSOC’s proposed application of its Section 120 authority to money market funds when the true cause for concern lies in the money markets themselves. 2. Size, Scale, and Concentration The Report asserts that money market funds’ size, scale, and concentration increase both their vulnerability to large, unexpected redemptions and the damaging impact of such redemptions on short- term credit markets, borrowers, and investors. Specifically, the Report states that given the “dominant role” of money market funds in short-term funding markets, investor withdrawal from these funds can have “severe implications for the availability of credit and liquidity in those markets.”63 The premise of the Report is that money market fund investors are more likely to react in tandem and more negatively than other investors during stressed markets. The Report implies that if money market funds did not exist, markets would be more stable in future crises. As discussed in Section II.B.4, the evidence from the 2007–2008 financial crisis demonstrates that money market funds and their shareholders did not react more quickly or extremely than other investors. In fact, money market funds and their shareholders pulled back far less than other investors in the commercial paper market, particularly in 2007, and increased their lending in the repurchase agreement market when other investors retreated. Furthermore, shareholder reaction during the financial crisis varied considerably across different types of money market funds. The Report also fails to consider how the 2010 reforms have and will dampen the effects of shareholder outflows in the future. As we note in Section II.B.1, the minimum liquidity rules require funds to hold a sizeable portion of their portfolios in highly liquid securities that are most in demand and easiest to sell during periods of financial stress. These are the same types of securities that the Basel Committee on Banking Supervision requires large international banks to hold to protect against large reductions in funding. Indeed, during the summer of 2011, the ability of funds to sell or roll off these liquid securities worked as intended to minimize the effects of the funds’ outflows on credit markets, borrowers, and investors, contrary to the Report’s assertions.64 FSOC’s observations regarding money market funds’ size and concentration also contradict its findings regarding the economic impact of its proposed recommendations on long-term economic growth. Even as it highlights the “dominant role” of money market funds, the Report downplays these funds’ importance to the economy. According to the Report’s economic analysis, even if FSOC’s proposed recommendations lead to an increase in the cost of lending from money market funds, the 63 Report, supra note 2, at 22. 64 See ICI Research Perspective, supra note 37, at 37-44. Financial Stability Oversight Council January 24, 2013 Page 29 result would be only a “very small increase in the weighted-average cost of credit for U.S. businesses, households, and state and local governments, with commensurately small potential costs to long-term economic growth.”65 FSOC cannot have it both ways. 3. Interconnectedness The Report contends that money market funds’ extensive interconnectedness with financial firms, the financial system, and the U.S. economy can create a “significant” threat to broader financial stability because the shocks from large, unexpected outflows from money market funds can rapidly propagate to other entities throughout the financial system. The Report implicitly—but inaccurately—assumes that market shocks originate with money market funds and then spread to other sectors. Actual experience from the 2007–2008 crisis shows just the opposite. Shocks occurred outside of money market funds and for that matter outside the money markets; money market funds actually acted as shock absorbers to the rest of the markets. As discussed in Section II.B.4., the financial crisis swept across banks and other intermediaries, and the money markets and money market funds absorbed large shifts in investor cash movements. It was not until the collapse of Lehman and American International Group, Inc. (“AIG”) created the widespread expectation in the markets that other major commercial and investment banks could fail that the money markets finally reacted. Once again, the Report fails to explain why trying to “fix” just this one aspect of the complex, large, and interconnected money markets will address challenges that exist in the market as a whole. Further, the Council’s exclusive focus on 2008 ignores the many changes that have occurred since then, including not only the 2010 money market fund amendments but also the broader reforms of the Dodd-Frank Act and the Basel Committee for Banking Supervision,66 which are designed to enhance the stability and resiliency of financial firms and our financial system as a whole. To provide but a few examples, the Dodd-Frank Act creates a new regulatory architecture for the over-the-counter derivatives markets, with the aim of reducing market opacity and counterparty risks, and addresses an array of other markets and practices, ranging from securitizations and credit rating agencies to executive compensation and corporate governance. The Act also imposes enhanced prudential standards on large banking organizations, including new capital, liquidity, risk management, stress testing, and resolution planning requirements—reforms supplemented by numerous measures devised at an international level by the Basel Committee. By enhancing ex ante regulation of key financial firms and markets, these and other changes enhance the stability of the money markets and the financial system as a whole. In addition, other changes to the regulatory framework will reduce the impact of a firm’s failure, should one occur. Most importantly, Title II of the Dodd-Frank Act establishes a mechanism for the orderly liquidation and wind-down of major financial firms. Other changes, such as enhancements to 65 See Report, supra note 2, at 67. 66 See generally Basel III Liquidity Standards, supra note 49; Basel III: A Global Regulatory Framework for More Resilient Banks and Banking Systems (Revised June 2011) (“Basel III Capital Standards”), available at http://www.bis.org/publ/bcbs189.pdf. Financial Stability Oversight Council January 24, 2013 Page 30 bank lending limits and the imposition of single counterparty credit limits, reduce overall credit exposures between firms and significantly lessen the spillover effects of the failure of any single firm. In sum, the regulatory framework has changed significantly since 2008 in ways that reduce both the probability and potential impact of the failure of a major financial firm; the Council must not view money market fund reform in isolation and without taking these other changes into account. 4. “Evidence” from the 2007–2008 Financial Crisis The Report asserts that the financial crisis demonstrated how money market funds can “interact and amplify the transmission of risk of significant liquidity and credit problems in the financial system.” As support for this assertion, the Report points to outflows from prime money market funds in the days that followed the Reserve Primary Fund’s announcement that it would break a dollar due to losses on the Lehman securities the fund owned. The Report does not posit any reason for those redemptions other than the travails of the Reserve Primary Fund, which it suggests provoked a “run” on the entire industry in isolation. Conspicuous by its absence is any mention of myriad other adverse financial market developments that both preceded and surrounded the events of that difficult week. Money market funds were not the cause of the financial crisis, but were directly affected by its enormous scale and duration, and by the lack of coherent, consistent government policy responses.67 Money Market Funds Were Not the Source of Growth in the Build-Up to the Financial Crisis. Some regulators and commenters have stated that money market funds provided a key source of financing to the short-term markets and indirectly to the mortgage market in the years prior to the financial crisis, helping to fuel the bubble that burst in 2007 and 2008.68 As shown in Figure 5, the increase in taxable money market funds’ holdings cannot credibly be deemed sufficient to be responsible for the expansion in the money markets. From the beginning of 2000 to mid-2007, the money markets expanded by $4.5 trillion, while assets of taxable money market funds increased by only $299 billion. In addition, money market funds financed, at most, 6 percent of home mortgage borrowing over this period.69 67 For a timeline of major developments in the financial crisis, see Appendix B. This appendix tracks movements in the commercial paper market, and clearly demonstrates that the market reacted to Lehman’s failure before Reserve broke a dollar. 68 Adair Turner, Shadow Banking and Financial Instability, Cass Business School (March 14, 2012), available at http://www.fsa.gov.uk/static/pubs/speeches/0314-at.pdf. 69 This estimate is an upper bound because it assumes that 100 percent of the increase in money market funds’ holdings of ABCP, corporate notes (not shown in Figure 5), and agency securities was ultimately funneled to home mortgage lending. Financial Stability Oversight Council January 24, 2013 Page 31 FIGURE 5 Selected Money Market Instruments Billions of dollars Total Outstanding Taxable Money Market Fund Holdings 2000:Q1 2007:Q2 Change 2000:Q1 2007:Q2 Change Repurchase agreements $1,470 $3,904 $2,434 $163 $435 $272 Large time deposits $966 $2,150 $1,184 $159 $217 $58 ABCP $547 $1,179 $632 $63 $256 $193 Other CP $923 $931 $8 $596 $408 -$188 Treasury bills $753 $866 $113 $61 $63 $2 Short-term agency securities $491 $606 $115 $177 $139 -$38 Total $5,150 $9,636 $4,486 $1,219 $1,518 $299 Source: Investment Company Institute, Federal Reserve Board, U.S. Treasury Department, Fannie Mae, Freddie Mac, Federal Housing Finance Board, Federal Reserve Bank of New York, and iMoneyNet Investors Other Than Money Market Funds Were the Initial Source of Pressure in the Money Markets. In August and September 2007, investors rapidly pulled back from the commercial paper market, with outstanding commercial paper falling a little over $300 billion or 14 percent. During this same period, money market funds’ holdings declined by $15 billion, accounting for only 5 percent of the sharp drop in total commercial paper. This experience demonstrates that investors other than money market funds can and do trigger abrupt pullbacks in short-term markets. Money market funds were the net recipients of assets flowing out of other short-term investment vehicles. Over the 13 months from the end of July 2007 through August 2008, money market funds absorbed about $900 billion in new cash, boosting the size of the money market fund industry by more than one third. About 80 percent of this vast inflow (around $700 billion) was directed to institutional share classes from investors such as corporate cash managers and state and local governments. The experience of money market funds during this time was in sharp contrast to the difficulties experienced by other types of investment pools that held subprime mortgages, or ABCP issued by structured investment vehicles backed by subprime mortgages. For example, on August 14, 2007, an unregistered commodity cash pool managed by Sentinel Management Group, Inc., erroneously described by CNBC as a “money market fund,” halted redemptions and failed within a week. In the coming weeks, other short-term, unregistered cash-like pools, frequently but incorrectly described by the press as “money market funds,” also failed. The difficulties that these pools encountered were the harbingers of a broader crisis. In the 12 months between the collapse of the Sentinel pool and the failure of Lehman, at least 13 major institutions in the U.S. and abroad went bankrupt, were taken over, or were rescued. Financial Stability Oversight Council January 24, 2013 Page 32 Investors Lost Confidence in Financial Institutions and Government Policy in the Fall of 2008. The financial crisis reached a critical stage, characterized by severely impaired liquidity in the global credit markets and insolvency threats to numerous investment banks and other financial institutions, during September 2008. Lehman’s failure was a shock for the market because it represented an abrupt reversal by the U.S. government from its previous decisions to intervene and rescue the smaller Bear Stearns and—just a week prior to Lehman’s failure—Fannie Mae and Freddie Mac.70 The widespread uncertainty about the government’s stance towards other troubled institutions had severe impacts on markets and market participants. Certain money market funds and many other money market participants were hit by a liquidity freeze. Banks, seeking to preserve their own liquidity, refused to lend to one another. Immediately after Lehman filed for bankruptcy, concerns spread in the financial markets that the debt of other large investment banks (The Goldman Sachs Group, Inc. and Morgan Stanley) and certain large commercial banks (Wachovia Corporation, Washington Mutual, and Citigroup) presented much greater risk than previously thought. The government’s decision to let Lehman fail—upsetting widespread market expectations of a rescue—also caused significant confusion. Reflecting these concerns, the cost of insuring against defaults by these institutions rose dramatically and deepened the credit freeze. Federal Reserve officials seem to have been surprised by the severity of the market’s reaction. For example, in congressional testimony on September 23, 2008, Federal Reserve Chairman Ben Bernanke noted that: [t]he failure of Lehman posed risks. But the troubles at Lehman had been well known for some time, and investors clearly recognized—as evidenced, for example, by the high cost of insuring Lehman’s debt in the market for credit default swaps— that the failure of the firm was a significant possibility. Thus, we judged that investors and counterparties had had time to take precautionary measures. While perhaps manageable in itself, Lehman’s default was combined with the unexpectedly rapid collapse of AIG, which together contributed to the development … of extraordinarily turbulent conditions in global financial markets.71 Chairman Bernanke also characterized the market events in the fall of 2008 to the Financial Crisis Inquiry Commission as “the worst financial crisis in global history, including the Great Depression.”72 He went on to say that “[i]f you look at the firms that came under pressure in that period … only one … 70 One day after Lehman was allowed to fail, the government again switched course and agreed to lend AIG up to $85 billion and to take a nearly 80 percent stake in the company, reversing an earlier indication that it would not participate in a rescue of the insurance giant. 71 Statement of Chairman Ben S. Bernanke, before the U.S. Senate Committee on Banking, Housing, and Urban Affairs (September 23, 2008), available at http://banking.senate.gov/public/index.cfm?FuseAction=Files.View&FileStore_id=bbba8289-b8fa-46a2-a542- b65065b623a1, at 3. 72 U.S. Financial Crisis Inquiry Commission, The Financial Crisis Inquiry Report (January 2011) (“FCIC Report”), available at http://www.gpo.gov/fdsys/pkg/GPO-FCIC/pdf/GPO-FCIC.pdf, at 354. Financial Stability Oversight Council January 24, 2013 Page 33 was not at serious risk of failure. So out of maybe the 13, 13 of the most important financial institutions in the United States, 12 were at risk of failure within a period of a week or two.”73 A variety of market participants pulled back their exposures to financial institutions, particularly banks, during the fall of 2008. Borrowing from the Federal Reserve’s discount window, excluding the commercial paper programs and lending associated with AIG and Bear Stearns, rose from $170 billion as of September 10, 2008, to $587 billion as of December 17, 2008, and remained at that level through the end of 2008.74 Much of this increase was through the Term Auction Facility, which held biweekly auctions of term funds to depository institutions against collateral that could be used to secure loans at the discount window. At the same time, interbank lending by commercial banks fell more than 30 percent, or nearly $145 billion, on a seasonally adjusted basis. The spread between the three-month London Interbank Offered Rate (“LIBOR”) and the overnight index swap (“OIS”) rate reflected the stress in the banking industry; this spread jumped from less than 100 basis points on September 12 to nearly 370 basis points one month later (Figure 6).75 The LIBOR-OIS spread is generally viewed as an indicator of the banking industry’s financial health, and a widening of the spread can be interpreted as a reluctance or unwillingness by banks to lend to other banks because of an increase in credit risk. 73 Id. 74 See Federal Reserve Statistical Releases (September 11, 2008) available at http://www.federalreserve.gov/releases/h41/20080911/ and (December 18, 2008), available at http://www.federalreserve.gov/releases/h41/20081218/. 75 In an effort to demonstrate financial strength, certain banks may have kept their LIBOR submissions artificially low during the crisis. See, e.g., Non-Prosecution Agreement between the United States Department of Justice and Barclays Bank PLC, Statement of Facts (June 26, 2012), available at http://www.justice.gov/iso/opa/resources/9312012710173426365941.pdf, at 15-22. This may have suppressed LIBOR during this period, in which case the true LIBOR-OIS spread would have been even higher than reported. Financial Stability Oversight Council January 24, 2013 Page 34 FIGURE 6 Spread Between Three-Month LIBOR and Overnight Index Swap Rate* Basis points, daily * 90-day LIBOR less the 90-day Overnight Index Swap (OIS) rate. An OIS is an interest rate swap with the floating rate tied to an index of daily overnight rates, such as the effective federal funds rate. At maturity, two parties exchange, on the basis of the agreed notional amount, the difference between interest accrued at the fixed rate and interest accrued by averaging the floating, or index, rate. Source: Bloomberg To be sure, the events of 2007–2008 were highly unusual. Even in these extreme conditions, however, investors remained invested in money market funds. As the Report acknowledges, investors shifted their assets from prime money market funds, which held financial institutions’ securities, to Treasury and government money market funds, which did not. About $310 billion flowed out of prime money market funds; for every dollar that left these funds, however, 61 cents flowed into Treasury and government funds. Indeed, investors did not abandon money market funds; like other participants in the money markets, they reacted to their concerns about the financial health of banks, the U.S. government’s unpredictable responses to financial institutions’ collapses, and concerns about whether prime money market funds could continue to sell assets into a commercial paper market that was essentially frozen. 0 50 100 150 200 250 300 350 400 J a n 2 0 0 7 M a r 2 0 0 7 M a y 2 0 0 7 J u l 2 0 0 7 S e p 2 0 0 7 N o v 2 0 0 7 J a n 2 0 0 8 M a r 2 0 0 8 M a y 2 0 0 8 J u l 2 0 0 8 S e p 2 0 0 8 N o v 2 0 0 8 J a n 2 0 0 9 September 15 Financial Stability Oversight Council January 24, 2013 Page 35 Following these events, the Federal Reserve and U.S. Treasury Department announced a series of broad initiatives designed to stabilize the money markets, which had ceased to function even for very short-term, high-credit securities. One of these programs was the Temporary Guarantee Program for Money Market Funds.76 Although steps taken by the Federal Reserve and the Treasury Department helped to stabilize the commercial paper market and thereby moderate outflows from money market funds, many investors— both those using money market funds and those invested in other instruments—continued to pull back from riskier credits and sought refuge in the U.S. Treasury market. Yields on four-week and three- month Treasury bills remained well under 1 percent on most days during the first half of October. Money Market Funds Were Not the Primary Source of Pressure in the Commercial Paper Market. The Report also suggests that money market funds were the primary source of pressure in the commercial paper market during the crisis. As we have noted, other investors accounted for most of the decline in commercial paper outstanding in 2007, and the data simply do not support FSOC’s conclusion for 2008 either. In fact, pressures in these and other short-term markets were driven by the rapid retreat of a wide range of investors, not just money market funds. Because of their transparency and regulatory oversight, however, money market funds were simply the most visible and easily observable market participants. The commercial paper markets began to seize up before prime money market funds experienced significant outflows and continued to suffer lack of liquidity long after those outflows abated. On September 15, when Lehman announced its bankruptcy, commercial paper markets were hit hard. Lehman had been one of the largest commercial paper dealers, and its bankruptcy eliminated a key source of liquidity in the market. Merrill Lynch also was a large commercial paper dealer, and its emergency sale to Bank of America negatively affected the market. On the day of Lehman’s bankruptcy, investors began pulling back from longer-dated paper. They did not come back to the market until after the Federal Reserve launched the Commercial Paper Funding Facility program in late October. From the middle of September through late October, commercial paper market issuance was heavily weighted to paper with four days or less to maturity. Financial issuers of commercial paper were particularly hard hit, and most issuers were unable to issue paper with maturities extending much beyond a month. For example, in the four weeks after Lehman collapsed, on average, only 14 issues of financial paper with maturities beyond 40 days reached the market each day, compared with a daily average of 140 in early September. The daily dollar volume of new financial paper issuance with these maturities was equally impaired, averaging $152 million, compared with $2.9 billion during the first half of September. Prime money market funds sold 76 See Appendix B. No claims were made on this program and taxpayers received an estimated $1.2 billion in premiums. The program expired on September 18, 2009. Financial Stability Oversight Council January 24, 2013 Page 36 commercial paper to meet redemption requests, but the amount of outstanding commercial paper had begun to fall prior to those sales and continued to fall through late October.77 By the end of September, outstanding commercial paper had declined by $185 billion.78 ICI data show that money market funds reduced their holdings of commercial paper by $164 billion in September, but $152 billion of that decline reflected sales to the AMLF. Hence, money market funds’ net reduction (after adjusting for sales to the AMLF) amounted to $12 billion. By creating the AMLF, the Federal Reserve effectively increased demand for commercial paper. Sales to this program did not reduce the overall demand for commercial paper, and therefore did not contribute to the contraction in outstanding commercial paper. Since money market funds’ $12 billion net reduction in commercial paper holdings amounted only to about 6 percent of the total decline, other investors clearly had to account for more than 90 percent of the $185 billion decline in this market. Data for other investors is not available specifically for September, but the Federal Reserve’s Flow of Funds Accounts show that funding corporations, foreign investors, state and local governments, and the household sector (which includes hedge funds and nonprofit organizations) were significant sellers of commercial paper in the third quarter of 2008.79 It would appear that much of the selling by these investors occurred during September.80 Furthermore, prime money market funds became net buyers of commercial paper in October, and by the end of that month had increased their holdings by $43 billion. Again, factoring in the AMLF program, the $250 billion decline in commercial paper outstanding in September and October resulted from other investors reducing their holdings and economic contraction caused by the onset of the “Great Recession.” Through the end of 2008, prime money market funds steadily increased their holdings of commercial paper and time deposits as inflows to these funds lifted total net assets by $412 billion. Aftermath. The U.S. government’s programs were eventually highly successful in shoring up confidence in financial markets generally and money market funds specifically. By mid-October 2008, the assets of prime money market funds began to grow. They continued to grow into 2009, indicating a return of confidence by institutional investors in these funds. During this same time period, assets of 77 Data from iMoneyNet show that money market fund holdings of commercial paper contracted before the Asset-Backed Commercial Paper Money Market Mutual Fund Liquidity Facility (“AMLF”) program began during the week of September 22, and a special survey by the Financial Crisis Inquiry Commission of money market funds shows that money market funds reduced their holdings of commercial paper during the first week of the crisis. See FCIC Report, supra note 72, at 358. We do not dispute the fact that money market funds contributed to the contraction of the market during the week of September 15, 2008. We note, however, that the data clearly show that money market funds were not the primary cause of the contraction in the commercial paper market in September. 78 Federal Reserve Commercial Paper Statistical Release and historical data, available at http://www.federalreserve.gov/releases/cp/. 79 Data from the Flow of Funds Accounts (not seasonally adjusted) show that these sectors, combined, reduced their commercial paper holdings on net by $131 billion in the third quarter of 2008. 80 Confidential data submitted to ICI show that stock, bond, and hybrid mutual funds lowered their holdings of commercial paper by $10 billion in September. Financial Stability Oversight Council January 24, 2013 Page 37 Treasury and government money market funds also continued to grow, although at a much reduced pace. By January 2009, although assets of prime money market funds had not returned to the level seen at the beginning of September 2008, they had regained much ground. Perhaps more importantly, total assets of money market funds had achieved an all-time high of $3.9 trillion by January 2009, reflecting the renewed confidence in money market funds among both retail and institutional investors.81 The events of 2007–2008 are in stark contrast to those of 1994—the only other time a money market fund broke a dollar.82 At that time, the financial system was not in cataclysmic disarray. The 1994 incident had no “systemic” consequences. In fact, money market fund assets grew during the month after that fund broke a dollar. At that time, there was no reason for investors to lose confidence in the assets their funds were holding or in the financial system at large, as there was in 2008. In contrast, the Reserve Primary Fund’s break-the-dollar incident in 200883 followed an unprecedented, worldwide series of financial institution failures going back to the middle of 2007 and bewildering, inconsistent responses to these events by the U.S. and other governments. Our observations also are consistent with the SEC staff’s characterization of events during the 2008 financial crisis. The SEC Staff Study found that there are many possible explanations for the redemption activity during the 2008 financial crisis.84 Importantly, their findings also suggest that idiosyncratic portfolio losses caused by interest rate changes, issuer defaults, and credit rating downgrades that can lead to significant valuation losses for individual funds do not appear to cause systemic problems, because such events do not cause abnormally large redemptions in other money market funds.85 5. The 2010 Reforms86 The Report asserts that although the SEC’s 2010 reforms were important, they did not address certain activities and practices of money market funds that continue to make the funds vulnerable to large, unexpected outflows. As discussed above in Section II.B.1., the 2010 amendments to money market fund regulation have made these funds even more stable, liquid, and transparent than ever 81 Notably, any investments made to money market funds after September 19, 2008 were not covered by the Treasury’s Temporary Guarantee Program for Money Market Funds. Therefore, this growth cannot be attributed directly to this program. 82 Community Bankers U.S. Government Money Market Fund broke a dollar in September 1994 and ultimately paid investors $0.96 per share. The fund had a large percentage of its assets in adjustable-rate securities (inverse floaters) that did not return to par at the time of an interest rate readjustment. Since that time, the SEC has prohibited a money market fund from investing in an adjustable-rate security if its interest rate readjustment formula does not ensure that the market value of the security will return to par once a readjustment occurs. 83 Reserve Primary Fund ultimately paid investors $0.99 per share. 84 See SEC Staff Study, supra note 10, at 7-9. 85 Id. at 15. 86 For a detailed analysis of the effectiveness of the SEC’s 2010 reforms, see generally ICI Research Perspective, supra note 37. Financial Stability Oversight Council January 24, 2013 Page 38 before. Indeed, recent events in the financial markets underscore the effectiveness of the 2010 amendments. In 2011, money market funds weathered two financial market shocks attributable in large measure to government gridlock: the looming U.S. federal debt ceiling crisis in mid-2011 and deteriorating conditions in European debt markets throughout the year. Money market funds also had to contend with historically low interest rates and the U.S. federal government’s extension of unlimited deposit insurance on non-interest bearing checking accounts.87 FIGURE 7 Prime Money Market Funds Accommodated Large Outflows During U.S. Debt Ceiling and Eurozone Debt Crises Total net assets, billions of dollars, weekly, 2011 Source: Investment Company Institute Reflecting these circumstances, investors withdrew $216 billion from prime money market funds over the six-month period from June 2011 to November 2011 (Figure 7). To be sure, these outflows were smaller in dollar and percentage terms than the flows prime funds experienced during the worst months of the financial crisis in September and October 2008. Nevertheless, they were quite large, totaling 13 percent of the assets of prime money market funds as of May 2011. Moreover, the bulk of 87 See supra note 47. $1,400 $1,450 $1,500 $1,550 $1,600 $1,650 $1,700 Jan-5 Feb-4 Mar-6 Apr-5 May-5 Jun-4 Jul-4 Aug-3 Sep-2 Oct-2 Nov-1 Dec-1 Week ended June 8, 2011 Week ended August 3, 2011 Financial Stability Oversight Council January 24, 2013 Page 39 these outflows occurred in a very short time (the weeks ended June 8, 2011 to August 3, 2011) as the U.S. federal debt ceiling crisis came to a head. Over that eight-week period, outflows totaled $172 billion, or 10 percent of prime money market fund assets as of May 2011. Outflows in the month of June 2011 were the second largest monthly total on record, totaling $86 billion. Prime money market funds accommodated these sizable outflows in an orderly manner. Funds had plentiful liquidity to meet redemptions. This was confirmed by the SEC Staff Study, which found that unlike in 2008, money market funds in 2011 had “sufficient liquidity to satisfy investors’ redemption requests.”88 As of the end of May 2011, prime money market funds held an estimated $625 billion in weekly liquid assets, far more than needed to meet the outflows experienced over the next several months. Moreover, the large outflows in the second half of 2011 had only a small impact on funds’ liquid asset ratios, which remained well above the required minimum levels of 10 percent and 30 percent for daily and weekly liquid assets, respectively (Figure 8). FIGURE 8 Liquid Asset Ratios of Prime Money Market Funds, March 2011 to March 2012 Percentage of prime fund assets Source: Investment Company Institute tabulations of SEC Form N-MFP data 88 See SEC Staff Study, supra note 10, at 34. 22 25 23 24 27 27 26 25 27 30 26 28 29 39 40 38 40 42 45 44 46 46 47 42 45 44 Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar 2011 2012 Sum of Daily liquid assets Sum of Weekly liquid assets Weekly liquid assets Daily liquid assets Financial Stability Oversight Council January 24, 2013 Page 40 In addition, despite the outflows and stresses in the market, money market funds’ per-share market values were extremely stable. The average change in the mark-to-market value of prime funds between May and September 2011 was less than 1 basis point.89 These findings are consistent with the findings of other analysts who note that the variability of prime money market funds’ per-share market values has declined significantly since the 2007–2008 financial crisis, a decline which they attribute in large measure to the revisions to Rule 2a-7 that went into effect in May 2010. 90 FSOC’s concerns about money market funds’ ability to meet large-scale redemptions unquestionably reflect an out-of-date view of the industry that wholly ignores the 2010 amendments. We strongly object to FSOC taking the drastic step of using its Section 120 authority based on faulty assumptions or data that does not reflect the current regulatory regime or actual market experiences of money market funds. C. FSOC’S PROPOSED DETERMINATION IS OVERLY BROAD FSOC’s proposed determination is overly broad because it is not based upon nuanced and thoughtful analysis of the various types of money market funds and their distinct risk profiles. As a result, FSOC proposes to recommend reforms broadly applicable to all money market funds without regard for these differences.91 Money market funds generally are categorized into four different types based on their investment strategies: Treasury, government, tax-exempt, and prime. These four types of funds hold securities that trade in markets with varying degrees of liquidity, have somewhat different levels of default risk, and had distinct investor redemption experiences during the financial crisis. The failure of FSOC to segment the market is puzzling and constitutes a fundamental omission in its analysis. Based on our study of money market funds (and the great weight of evidence in the public record), it is abundantly clear that no case can be made for applying fundamental changes to Treasury, government, or tax-exempt money market funds. Even for prime money market funds, the measures FSOC proposes to recommend are wholly inappropriate and disproportionate to any theoretical threat.92 1. Treasury and Government Money Market Funds Treasury and government money market funds have significantly different portfolios from those of prime money market funds. Treasury money market funds invest primarily in U.S. Treasury obligations and repurchase agreements collateralized with U.S. Treasury obligations. Government money market funds are defined as those that invest primarily in U.S. Treasury obligations, U.S. government agency securities, and repurchase agreements collateralized with U.S. Treasury or agency obligations. 89 See ICI Research Perspective, supra note 37, at 9. 90 See Fitch Ratings’ Special Report, supra note 32. 91 FSOC would not apply Alternatives Two and Three to Treasury funds. 92 See generally Robert Comment, Do Money Market Funds Require Further Reform? (December 31, 2012), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2174125. Financial Stability Oversight Council January 24, 2013 Page 41 FSOC generally supports imposing the same reforms on government money market funds as it does on prime funds. It contends that “[g]overnment [money market funds] … may pose the same structural risks [as prime money market funds], in that the funds’ investors would have an incentive to redeem if they feared even small losses”93 and that “government [money market funds] also can be vulnerable to runs.”94 It also recommends a floating NAV for Treasury money market funds despite acknowledging that these funds are “unlikely to suffer credit events; tend to experience net inflows, rather than net redemptions, in times of stress; and may be more likely to maintain a stable value during times of market stress, when Treasury securities generally maintain their values.” The Report fails to note, however, that in these respects Treasury and government money market funds are exactly the same. The Report also fails to acknowledge that both Treasury and government money market funds hold securities that trade in markets that are more liquid and have reliably higher credit quality than non-government securities. Further, both Treasury and government funds have had similar shareholder redemption experience during periods of financial stress. Both Treasury and government agency markets are deep and liquid, accommodating significant trading volume. Among the primary dealers alone, daily trading volume in the past year averaged around $225 billion a day for short-term U.S. Treasury securities and $380 billion a day for agency mortgage- backed securities and short-term agency securities.95 Significant interest rate movements due to impairments in market liquidity, therefore, are highly unlikely. Furthermore, interest rate risk in money market funds is already highly constrained by Rule 2a-7. Given the short duration of money market fund portfolios, any interest rate movements have a modest and temporary effect on the value of the funds’ securities. Although all money market funds are required to hold securities with minimal credit risk, credit losses would occur in Treasury or government funds only if the U.S. government failed to repay its maturing debt in full or allowed a federal agency to collapse precipitously, causing the agency to default on its outstanding short-term debt. We assume FSOC would agree that these events are extremely unlikely and, were they to occur, would have a broad, global market impact far beyond money market funds. We find it extremely troubling that the regulators serving on the Council would even suggest—by applying their proposed alternatives so broadly—that markets and investors should anticipate the occurrence of these events. Importantly, during periods of financial stress, the behavior of investors in prime funds and Treasury and government funds differs substantially. For example, Treasury and government funds saw substantial inflows during September 2008. The Report acknowledges that government money market funds attracted inflows of $192 billion during the week following the Lehman bankruptcy, and “did not face similar run vulnerabilities at the time because they had significantly different portfolio holdings 93 See Report, supra note 2, at 26. 94 Id. 95 Transactions by primary dealers of U.S. Treasury bills, U.S. Treasury securities due in three years or less, discount notes and coupon securities due in three years or less issued by federal agencies and government-sponsored enterprises, and mortgage- backed securities averaged over the year 2012 from Federal Reserve Bank of New York, Weekly Release of Primary Dealer Positions, Transactions, and Financing, Table I, available at http://www.newyorkfed.org/markets/statrel.html. Financial Stability Oversight Council January 24, 2013 Page 42 than the distressed prime funds and many government [money market fund] instruments were appreciating in value.”96 The Report points only to the Treasury Department’s assistance with the liquidation of the Reserve Fund’s U.S. Government Fund and outflows from government money market funds in July 2011 (during the debt ceiling crisis and under the threat of a U.S. government default) as supporting its case that government money market funds can be vulnerable to runs. These events do not remotely support FSOC’s assertion that government money market funds pose a “run” risk. The Problems of Reserve Government Fund Were Addressed in the 2010 Amendments. The challenges that the Reserve Government Fund faced in 2008 were in large measure the result of the fund’s affiliation with the Reserve Primary Fund. After the Reserve Primary Fund announced it was breaking the dollar, the assets of all of the Reserve funds fell dramatically. The Reserve Government Fund also faced challenges meeting outflows and maintaining a fixed $1.00 NAV because it held large amounts of long-dated adjustable-rate government securities that posed significant interest rate risk. These securities had historically maintained market values similar to short-term fixed-rate securities during normal markets. During the crisis, however, some of these securities proved difficult to sell. The Report fails to acknowledge, however, that the 2010 amendments directly addressed the portfolio risks inherent in the Reserve Government Fund. Indeed, under the 2010 amendments, a money market fund can no longer be structured like the Reserve Government Fund. The Reserve Government Fund’s WAL would have exceeded 230 days—almost twice the current 120-day limit. As discussed in Section II.B.1., the amendments added a WAL requirement that effectively restricts the extent to which a money market fund can invest in longer term adjustable-rate securities whose prices may depart from amortized cost. Furthermore, because of its heavy investment in floating-rate agency securities, which would not have counted toward the current liquidity requirements, the Reserve Government Fund held less than 5 percent of its assets in securities that would be defined as daily and weekly liquid assets under revised Rule 2a-7. Thus, the SEC’s 2010 reforms fully addressed the portfolio risks that undermined the Reserve Government Fund. In sharp contrast to the Reserve Government Fund, other government money market funds did not experience difficulties but instead saw substantial inflows at that time. Figure 9 below shows the distribution of the change in assets of government money market funds from September 2 to September 30, 2008.97 Three-quarters of funds received inflows (yellow line), and half (green line) had inflows of at least 13 percent of total assets during September as investors moved to the safety and security of U.S. Treasury and government agency debt. Twenty-five percent of the funds did have outflows for the month, but the outflows were quite modest for most of the funds, and no other fund had to suspend redemptions. Even the Lehman Institutional Government Reserve Fund (Figure 9, red line) 96 See Report, supra note 2, at 26. 97 Each fund’s level of total net assets is indexed to 100 on September 2, 2008. For each fund, the index value on September 30, 2008, is indicative of the percentage change in the fund’s assets from September 2, 2008. The cross-sectional distribution is determined by the fund’s index values on September 30, 2008. For example, the maximum index value represents the fund with the largest percentage change in assets over the period September 2, 2008, to September 30, 2008. Financial Stability Oversight Council January 24, 2013 Page 43 accommodated the redemption requests of its shareholders, which began early in September. FSOC disingenuously and opportunistically seeks to magnify an isolated problem at one fund—and a problem that under current regulations will never be repeated—into an argument for applying a systemic risk remedy to all government money market funds. FIGURE 9 Distribution of Assets of Government Money Market Funds1 Index = 100 on 9/2/2008, daily, September 2008 1 There are 104 funds in the sample. Data exclude the Reserve U.S. Government Fund, which suspended redemptions on September 17, 2008; funds with missing data during the September period; and funds with less than $50 million in assets. Source: Investment Company Institute tabulations of iMoneyNet daily data Government Funds Met Redemptions Without Incident During the Summer of 2011. The Report also cites the experience during the summer of 2011 as evidence that government money market funds are vulnerable to runs, focusing on the last three days of July 2011. Although some funds had significant contractions in total net assets, these redemptions were not problematic. Indeed, a careful look at the historical pattern of flows from government money market funds demonstrates that the outflows in July 2011 were not particularly widespread, damaging to other funds, or disruptive to the markets. During the last week of July 2011, 62 percent of government money market funds had outflows. It is not unusual, however, for a majority of government funds to have outflows in any given week, and these funds easily accommodate changing investor cash needs as part of their normal operations. As shown in Figure 10, from January 2008 to December 2012, on average 58 percent of government funds had outflows in any given week, only slightly less than the percentage in late July 2011. Of particular note are other weeks when a higher percentage of government money market funds had outflows. In the 0 50 100 150 200 250 9/ 2 9/ 3 9/ 4 9/ 5 9/ 8 9/ 9 9/ 10 9/ 11 9/ 12 9/ 15 9/ 16 9/ 17 9/ 18 9/ 19 9/ 22 9/ 23 9/ 24 9/ 25 9/ 26 9/ 29 9/ 30 99th percentile 25th percentile 95th percentile 5th percentile 75th percentile Lehman Govt Reserve fund Median Financial Stability Oversight Council January 24, 2013 Page 44 spring of 2009, for example, roughly 70 percent of government funds each week had outflows, and in the week of May 26 to June 1, 2011, 75 percent of government money market funds had outflows. These occurred without impact on other funds or the markets. FIGURE 10 Percentage of Government Money Market Funds with Outflows Weekly, January 2008 to December 2012 Source: Investment Company Institute asset data based on iMoneyNet designation of government/agency money market funds The data for the entire month of July 2011 (Figure 11) also does not indicate that government money market funds experienced widespread, destabilizing outflows. Half of government money market funds had inflows (green line). An additional 25 percent of funds (yellow line) experienced a small (2.7 percent or less) contraction in total net assets during the month. These data make it clear that investors in government money market funds did not redeem en masse over this period. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% Ja n- 08 M ar -0 8 M ay -0 8 Ju l-0 8 Se p- 08 N ov -0 8 Ja n- 09 M ar -0 9 M ay -0 9 Ju l-0 9 Se p- 09 N ov -0 9 Ja n- 10 M ar -1 0 M ay -1 0 Ju l-1 0 Se p- 10 N ov -1 0 Ja n- 11 M ar -1 1 M ay -1 1 Ju l-1 1 Se p- 11 N ov -1 1 Ja n- 12 M ar -1 2 M ay -1 2 Ju l-1 2 Se p- 12 N ov -1 2 July 28–August 3: Debt ceiling negotiations D ec -1 2 Financial Stability Oversight Council January 24, 2013 Page 45 FIGURE 11 Distribution of Assets of Government Money Market Funds1 Index = 100 on 7/1/11, daily, July 2011 1 There are 82 funds in the sample. Data exclude funds with missing data during the given period and funds with less than $50 million in assets. Source: Investment Company Institute tabulations of iMoneyNet daily data The Report also raises concerns about the total dollar volume of outflows from government funds during the last week of July 2011. As was the case with the number of funds, the amount of the outflows during this time does not provide evidence that investors were reacting en masse or disrupting the markets and affecting other funds. For the week of July 28 to August 3, 2011, total net assets of government money market funds declined by $20 billion, or 4.3 percent of the previous week’s assets. In 2009 and 2010, however, there were five one-week periods in which outflows from government money market funds, as measured by the change in total net assets, exceeded 3 percent of assets. Most of these periods had total outflows that exceeded the dollar amount of the outflows that occurred in the last week of July 2011.98 Even outflows for the three-week period ended August 3, 2011, when assets of government money market funds fell by $32 billion or 6.7 percent, were less than for the three-week 98 For the week ended June 17, 2009, outflows from government money market funds totaled $29 billion or 3.5 percent of assets. For the week ended October 14, 2009, outflows totaled $28 billion or 3.8 percent of assets. For the week ended January 20, 2010, outflows totaled $25 billion or 3.7 percent of assets. For the week ended March 17, 2010, outflows totaled $22 billion or 3.8 percent of assets. For the week ended April 14, 2010, outflows totaled $20 billion or 3.7 percent of assets. Source: Internal tabulations based on confidential data submitted to the Investment Company Institute as part of its weekly survey of money market fund assets. 40 60 80 100 120 140 160 7/ 1 7/ 5 7/ 6 7/ 7 7/ 8 7/ 11 7/ 12 7/ 13 7/ 14 7/ 15 7/ 18 7/ 19 7/ 20 7/ 21 7/ 22 7/ 25 7/ 26 7/ 27 7/ 28 7/ 29 99th percentile 25th 95th percentile 5th percentile 75th Minimum Median Financial Stability Oversight Council January 24, 2013 Page 46 period ended March 17, 2010, when assets declined $49 billion or 7.9 percent.99 In both of these instances, the outflows occurred without impact on other funds or the markets. The magnitude of the outflows from government money market funds in and of itself is completely irrelevant, however; the systemic issue for FSOC to consider is whether the outflows had a significant impact on the broader markets. As previously noted, the markets for Treasury and U.S. government agency securities are liquid and deep. The $20 billion in net redemptions in the week ended August 3, 2011 amounted to only 0.6 percent of the $3.5 trillion in short-term Treasury and U.S. government agency debt outstanding.100 In addition, the tri-party repurchase agreement market had an additional $1.3 trillion in outstanding repurchase agreements collateralized with Treasury and agency securities in mid July 2011. Further, the average mark-to-market share price for government money market funds decreased on average 1 basis point between the end of June and the end of July, again demonstrating that there was no spillover to investors who remained invested in the funds or to other funds from the outflows in late July 2011. Finally, the Report suggests that investors in money market funds have become more likely to react to market events than in the past. Historically, investors have moved into government money market funds during turbulent times. As discussed in Appendix C, this tendency for government money market funds to receive net inflows during periods of market stress has remained strong since the financial crisis of 2007–2008, again countering FSOC’s unsubstantiated assertion that these funds are susceptible to “runs.” To meet the exacting standard set forth in Section 120 of the Dodd-Frank Act, FSOC must demonstrate more than the fact that funds have redemptions. Rather, it must show that those redemptions also “create or increase the risk of significant liquidity, credit, or other problems spreading among bank holding companies and nonbank financial companies, financial markets of the United States, or low-income, minority, or underserved communities.” FSOC’s evidence falls far short of meeting that test. We find it particularly troubling that FSOC—composed as it is of the heads of U.S. federal financial regulators—would see fit to propose drastic reforms for funds whose portfolios consist almost entirely of short-term Treasury and government securities. Absent implicit concerns about a default by the U.S. Government, these proposals seem wholly misplaced. If these proposals actually are motivated by such concerns, the implications for the financial system hardly can be confined to money market funds. 99 Indeed, the decline in total net assets for the three-week period ended March 17, 2010, was part of a larger $80 billion or 12.5 percent decline over the six-week period ended March 24, 2010. 100 Short-term Treasury and U.S. government agency debt is defined as marketable Treasury securities held by the public due to mature by the end of July 2012 and debt issued by Fannie Mae, Freddie Mac, and the Federal Housing Finance Agency due to mature by the end of June 2012 (category excludes agency-backed mortgage pools). Financial Stability Oversight Council January 24, 2013 Page 47 2. Tax-Exempt Money Market Funds The Report also does not provide a basis for including tax-exempt money market funds in its Section 120 determination. Tax-exempt money market funds did not suffer heavy redemptions during the financial crisis (Figure 12). For the month of September 2008, nearly 25 percent of tax-exempt money market funds (blue line) experienced inflows and another 25 percent of funds had outflows amounting to 7 percent of total net assets or less.101 In the aggregate, tax-exempt money market funds had net redemptions of $38 billion or 7.5 percent of previous month-end total assets for the month of September 2008. A closer examination of those outflows (Figure 12) demonstrates how limited they were. Many investors in tax-exempt money market funds remained calm even when Reserve suspended redemptions at all 14 of its tax-exempt funds on September 17, 2008, and a Lehman-sponsored money market fund—the Neuberger Berman Tax-Free fund (red line)—had two thirds of its total net assets redeemed.102 Redemption pressures at those funds associated with Reserve and Lehman had no ripple effect on other tax-exempt funds or the broader municipal market, which remained stable. 101 Total net assets of the median fund declined by 7 percent in September 2008. 102 Another Lehman fund, the Neuberger Berman NY Municipal Fund had half of its total net assets redeemed by the end of September. Financial Stability Oversight Council January 24, 2013 Page 48 FIGURE 12 Distribution of Assets of Tax-Exempt Money Market Funds1 Index = 100 on 9/2/08, daily, September 2008 1 There are 232 funds in the sample. Data exclude Reserve’s 14 tax-exempt money market funds, which suspended redemptions on September 17, 2008, funds with missing data during the September period, and funds with less than $50 million in assets. 2 Neuberger Berman funds were sponsored by Lehman in September 2008. Source: Investment Company Institute tabulations of iMoneyNet daily data Historically, another experience in which market events could have affected investor behavior involved the default by Orange County on December 9, 1994 and the disclosure the previous week that the county’s investment fund had lost $1.5 billion.103 One fund that was heavily invested in securities issued by California municipalities had about 30 percent of its assets redeemed from November 30, 1994, to January 4, 1995 (Figure 13, red line). Other California tax-exempt funds also tended to have outflows over this period. But these redemptions did not trigger broad-based, destabilizing outflows from all tax-exempt money market funds. To the contrary, by early January 1995, half of all tax-exempt money market funds had inflows over the period (green line). As a whole, tax-exempt money market funds had inflows of about $400 million over this five-week period. 103 See Sallie Hofmeister, “A Default by Orange County,” New York Times (December 9, 1994), available at http://www.nytimes.com/1994/12/09/business/a-default-by-orange-county.html. 0 20 40 60 80 100 120 140 9/ 2 9/ 3 9/ 4 9/ 5 9/ 8 9/ 9 9/ 10 9/ 11 9/ 12 9/ 15 9/ 16 9/ 17 9/ 18 9/ 19 9/ 22 9/ 23 9/ 24 9/ 25 9/ 26 9/ 29 9/ 30 99th percentile 25th percentile 95th percentile 5th percentile 75th percentile Neuberger Berman Tax-Free fund Median 2 Financial Stability Oversight Council January 24, 2013 Page 49 FIGURE 13 Distribution of Assets of Tax-Exempt Money Market Funds1 Index = 100 on 11/30/94, weekly, November 30, 1994–January 4, 1995 1There are 201 funds in the sample. Data exclude funds with missing data during the given period and funds with less than $50 million in assets. Source: Investment Company Institute The Report’s proposed determination that tax-exempt money market funds pose systemic risks also fails to take into account the impact of the 2010 amendments to Rule 2a-7. Tax-exempt funds now have weekly liquidity far in excess of the 30 percent required under Rule 2a-7. As of June 2012, tax- exempt funds had $217 billion in weekly liquidity, amounting to 81 percent of total assets. Just as with Treasury and government money market funds, FSOC has failed to demonstrate that tax-exempt money market funds meet the test for a Section 120 recommendation. To the contrary, the few examples where isolated tax-exempt money market funds experienced redemption pressure clearly demonstrate that tax-exempt funds do not present the risks hypothesized in the Report. III. Temporary Gates and Liquidity Fees The Report recognizes that there may be other money market fund changes that FSOC should consider. In particular, the Report requests comment on measures that would operate only during times of market stress, and that would not change the fundamental nature of money market funds under normal circumstances. These include: • Temporary restrictions on redemption, or “gates,” that when triggered would prohibit investors from redeeming and would provide time for the fund to restore share value. 60 80 100 120 140 160 180 11/30/94 12/7/94 12/14/94 12/21/94 12/28/94 1/4/95 99th percentile 25th 95th percentile 5th percentile 75th Minimum Median Financial Stability Oversight Council January 24, 2013 Page 50 • Liquidity fees that when triggered would charge redeeming shareholders to compensate the fund and remaining investors for the potential cost of the withdrawal to the fund. As members of the Council are aware, throughout 2011 and 2012, ICI and its money market fund members weighed numerous potential reforms that might improve upon the 2010 SEC amendments while ensuring a continued robust and competitive money market fund sector. Among these possible changes was the imposition of temporary gates and liquidity fees for prime money market funds. We discuss this option below. In our judgment, these are tools that would be effective should a prime money market fund face large and unexpected redemptions. As explained below, we do not believe that the availability of these tools would accelerate outflows in times of stress. Redemption restrictions and gates in fact were used to good effect by U.S. and European funds during the financial crisis. Before formally proposing these alternatives, however, regulators must consider the tax and operational implications, as well as the impact on certain transaction types, of imposing gates and liquidity fees on prime money market funds. A. OBJECTIVE TRIGGER FOR GATES AND FEES We do not concede that the Report has made the case for further reform. If, however, FSOC can demonstrate that changes are needed for prime money market funds, we would support FSOC’s consideration of recommending that the SEC propose requiring a prime money market fund to impose liquidity gates if its “weekly liquid assets” (as defined under Rule 2a-7)—after accounting for unsettled portfolio trades—fall to a specific, objective “trigger point.” This is in contrast to the MBR concept in FSOC’s Alternative Two that would add continuous redemption holdbacks to money market funds, thus forcing these funds’ investors to pay a premium for liquidity under all market conditions. Under the approach we recommend, liquidity gates would not be imposed during “normal” market conditions, but only when a fund’s available weekly liquid assets fall to a specific threshold. That threshold or “trigger point,” would apply to all similarly situated funds, should be set at a level that is high enough to ensure that the fund still has some liquidity remaining, but low enough to ensure that the trigger point likely would not be reached during normal, or even somewhat stressed, market conditions. For this purpose, we suggest a trigger point when weekly liquid assets fall to between 7.5 percent and 15 percent of total fund assets; that is, between one quarter and one half of the current minimum weekly liquid asset level required under Rule 2a-7. When a prime money market fund trips the trigger point, gates would automatically be imposed after the close of business to suspend redemptions received for processing the next business day. Money market fund boards then would be permitted to lift the gate and honor redemptions, provided that redeeming shareholders pay a nonrefundable liquidity fee to the fund equal to 1 percent of redemption proceeds. A liquidity fee set at this level would discourage redemptions, but allow the fund to continue to provide liquidity to investors. Insofar as investors choose to redeem, the fee would benefit remaining shareholders by mitigating liquidation costs and potentially rebuilding NAVs. Investors truly in need of liquidity would have access to it, but at a pre-determined cost. Other investors would be able to preserve the full value of their shares by maintaining their position in the fund until such time as liquidity is restored and the gate is lifted. Financial Stability Oversight Council January 24, 2013 Page 51 Funds would be required to make prompt disclosure to the SEC, their shareholders, and the public whenever imposing redemption gates and liquidity fees. This should include disclosure of the fund’s plan to lift the gate or to liquidate the fund. Importantly, the use of gates and fees would be a temporary expedient (lasting, for example, no longer than 30 days) designed to compel a prime money market fund to address its portfolio liquidity issues (without having to rapidly sell assets in a fire sale manner) or start an orderly liquidation process. At the same time, of course, a prime money market fund’s board would retain its existing authority to employ other measures that best serve the interests of the fund and its investors, including the authority under Investment Company Act Rule 22e-3 to suspend redemptions and liquidate the fund. B. ENHANCED DISCLOSURE Importantly, liquidity gates and fees would be coupled with measures to enhance still further the transparency of prime money market fund portfolios. Such measures would benefit both investors and regulators. Currently, money market funds are required to disclose their mark-to-market share price every month, but with a 60-day lag; they are not required to make any disclosure of their weekly liquid assets.104 As part of the measures outlined above, prime money market funds could be required to make frequent public disclosure (via their websites) of both their mark-to-market share price and their weekly liquid asset levels.105 This kind of portfolio transparency would encourage a highly conservative approach to the management of prime money market fund portfolios. Investors would have far greater insight than is available today into the current holdings and liquidity of all prime money market funds— those in which they have invested, as well as any that may be experiencing difficulties. This would permit investors to determine more readily if such difficulties are idiosyncratic to particular funds, thus minimizing the prospect of redemption pressures on funds not similarly impacted. C. TEMPORARY GATES AND LIQUIDITY FEES SPECIFICALLY ADDRESS SYSTEMIC CONCERNS The Report expresses concern that temporary gates and liquidity fees may not adequately address—and in fact may further increase—the potential for widespread large and unexpected outflows in times of stress. Specifically, the Report suggests that gates may increase the risk of preemptive redemptions by investors who could be motivated to sell fund shares before a gate is triggered. The Report also expresses concern about contagion risk, because the triggering of gates in one money market fund could encourage shareholder redemptions in other money market funds. We take issue with these concerns. In contrast to FSOC’s three alternative recommendations, a liquidity-based trigger for gates aligns precisely with the goal of stopping large, unsustainable redemptions: it has the immediate effect of suspending further redemptions. Also, unlike a trigger based 104 Although funds are not required to report weekly liquid assets, analysts can calculate the information from tabulations of individual securities holdings submitted as part of Form N-MFP filings. 105 Some funds already are moving in this direction voluntarily. See supra note 31. Financial Stability Oversight Council January 24, 2013 Page 52 on a fund’s mark-to-market NAV, a liquidity-based trigger for fees introduces immediate “redemption frictions”: it exacts a substantial cost for liquidity when liquidity is at a premium. Some investors might be motivated to redeem in anticipation of any gate being imposed. Nonetheless, once imposed in a time of market stress, the gate would halt promptly any outflows from the fund, providing time for the fund to rebuild its liquidity as the short-term, high-quality instruments in its portfolio reach maturity. A fee would provide a strong disincentive for investors to make further redemptions by causing them to choose between paying a premium for current liquidity or delaying liquidity and benefitting from the fees paid by redeeming investors. In contrast, as discussed below, FSOC suggests that requiring shareholders to have a 30-day MBR provides a disincentive for investors to redeem during times of stress. It has not demonstrated, however, why investors that are worried about the markets in general would not start moving away from money market funds earlier (i.e., so as to start their 30-day clock), as all losses would be concentrated on shareholders who redeemed in the 30 days before losses were realized. A trigger based on specified liquidity levels also would encourage funds to police themselves—for example, by maintaining a more diversified client base and/or higher liquidity levels than those currently required. It likewise would provide very strong incentives for fund managers to deal with potential problems promptly and aggressively, so as to avoid triggering the gates. FSOC itself acknowledges this as a specific benefit. Conditioning fund manager behavior in this way would make it less likely that prime money market funds would edge toward reduced levels of liquidity. D. U.S. AND EUROPEAN EXPERIENCES WITH SUSPENSION OF REDEMPTIONS AND GATING In the United States, experience with gating and redemption tools is limited, especially for investment products registered under the Investment Company Act. Other U.S. products, such as unregistered cash pools and hedge funds, however, have made use of gating and suspension of redemptions. Outside the United States, the availability and use of redemption and gating tools for registered funds are more common. Nevertheless, these kinds of tools are generally viewed as measures to be used only during crises. Conversations with industry experts indicate that such tools, when used, can be both effective and efficient. They have helped certain funds (in some cases money market funds and in other cases long- term funds) by: (i) providing breathing room to assess market conditions; (ii) creating a “circuit breaker” against cascading shareholder redemptions and, in combination with a redemption fee, allowing some funds to reopen in due course; or (iii) achieving a more orderly liquidation of the fund if ultimately necessary. 1. U.S. Fund Experiences Although there are few experiences of U.S. mutual funds implementing redemption restrictions, one case during September 2008 may be instructive. Putnam Prime Money Market Fund, a sizable institutional fund ($17.4 billion on September 12, 2008) experienced a 30 percent decline in assets over September 15 and 16. On September 17, the fund’s board voted to close and liquidate the fund. The Financial Stability Oversight Council January 24, 2013 Page 53 board apparently took this action on the basis of significant redemption pressure and dwindling liquidity in the money markets. Subsequently, a purchaser was identified, and on September 24, the fund’s assets were transferred in an in-kind transaction to Federated Investors, Inc., which merged those assets and fund shareholders into its Prime Obligations Fund. The board’s decision to suspend redemptions allowed for an orderly transition and was made in the best interest of the fund’s remaining shareholders. In a non–mutual fund context, on November 29, 2007, the Florida Local Government Investment Pool (“Florida LGIP”), a private cash pool offered to cities, counties, school districts, and other local and state agencies for investing money on a short-term basis, suspended redemptions in the face of declining asset quality, $3.5 billion in redemption requests, and an inability to sell assets to meet redemptions. Suspending redemptions gave the pool breathing room to later institute a mandatory redemption fee of 2 percent to encourage investors to stay invested. Although some investors still chose to redeem, many remained. Over the course of the year, most of the pool’s underlying securities matured, which eliminated the need for ongoing redemption fees. The Florida LGIP is still in business. 2. European Fund Experiences Under the European Union’s Undertakings for Collective Investment in Transferable Securities (“UCITS”) directive, European UCITS (mutual funds) have the ability to “temporarily suspend” redemptions. This authority is intended to be used only in “exceptional cases … and where suspension is justified having regard to the interests of unit-holders.”106 “Gating,” which in Europe is different from suspensions, is generally possible if provided for in the fund’s prospectus. Gating generally allows for partial redemptions on a pro-rata basis per shareholder. Beyond that, operation of temporary suspensions and gating may be controlled by rules in individual countries. For example, in the United Kingdom, a fund must review the suspension of redemptions at least every 28 days and must terminate the measure as soon as practicable. During the 2007–2008 financial crisis, a number of European-domiciled funds suspended redemptions.107 For example, on August 7, 2007 three BNP Paribas “Dynamique” money funds suspended redemptions, reportedly following illiquidity in the U.S. subprime market on August 6, 2007. These funds held U.S. subprime asset-backed securities, which became very difficult to value. Rather than sell at fire sale prices, the funds elected to suspend redemptions temporarily. The three funds apparently reopened for purchases and redemptions on August 28, although each of the funds reportedly took small losses. According to the funds’ manager, “the slight drop in value of the three 106 See, e.g., UCITS Regulations, Part 11, Section 104(2)(a)(ii). 107 European funds are not required to and do not provide the public with the same level of disclosure as U.S. funds. As a result, obtaining more detailed or precise information about how these funds fared during the crisis is difficult. Much of what we know about the experience of European money market funds during the crisis is anecdotal. Financial Stability Oversight Council January 24, 2013 Page 54 funds since 7 August … confirmed that the decision to suspend valuations on a temporary basis played its expected protective role for investors.”108 Three non-U.S. money market funds advised by Lehman also suspended redemptions.109 On September 19, 2008, Lehman suspended redemptions on its Euro Liquidity Fund, Sterling Liquidity Fund, and U.S. Dollar Liquidity Fund. The decision apparently reflected higher than normal redemptions, market-wide liquidity issues, and, concerns about the Lehman name. The Euro and Sterling funds were liquidated on January 29, 2009, and the U.S. Dollar Liquidity Fund on March 30, 2009. Investors in all three funds ultimately received full value. In this case, the ability to suspend redemptions did not save these funds, but it did allow for a more orderly liquidation process and helped investors receive full value back, which would not have occurred if the funds had been forced to sell at fire sale prices. Most importantly, all shareholders were treated equitably. In the United Kingdom, a number of longer-term funds, apparently real estate-related, suspended redemptions in 2008. Some of these funds ultimately reopened. Others, primarily because of a large decline in the value of their assets, were liquidated. E. U.S. TAX IMPLICATIONS A liquidity fee could be assessed either by reducing the gross proceeds paid out to the investor upon the redemption, or by reducing the remaining balance in the investor’s account. In either case, a liquidity fee likely would be characterized for tax purposes as either: (i) a capital contribution by the shareholder to the money market fund;110 or (ii) a fee that is an expense to the shareholder and income to the fund. Capital Contribution. If the fee were a capital contribution to the fund, it would increase the shareholder’s basis in its remaining shares in the fund. Thus, the shareholder’s basis would increase above $1.00 per share for those shares. If the shareholder later sold any of those shares at $1.00, the shareholder would have a capital loss because the shareholder’s basis for those shares would be higher than $1.00. This effectively would break the dollar with respect to those shares and trigger tax reporting obligations for funds, intermediaries, and investors. Treating the fee as a capital contribution to the fund would have no tax consequences to the fund, and the money market fund could use 100 percent of the fee to stabilize and rebuild the NAV. Fee: Expense to Shareholder, Income to Fund. If the fee were treated as an expense to the shareholder, it would affect neither the shareholder’s gain or loss, nor the basis of the remaining shares. The fee could be deducted by the shareholder as an investment expense, subject for individual taxpayers to a floor of 2 percent of adjusted gross income. If the fee were an expense to the shareholder, it generally 108 See BNP Paribas press release, Background information on suspension and reopening of ABS funds in August, available at http://media-cms.bnpparibas.com/file/76/1/5761.pdf. 109 These funds adhered to the Institutional Money Market Funds Association’s Code of Practice, which provides a framework that is similar to the standards required by Rule 2a-7. For more information about IMMFA funds, see http://www.immfa.org/. 110 Capital contribution treatment assumes that the shareholder has shares remaining in the fund when making a redemption; this treatment would not apply if the shareholder redeemed all shares in the fund. Financial Stability Oversight Council January 24, 2013 Page 55 would constitute ordinary income to the fund. As such, it would be subject to the distribution requirements of the Internal Revenue Code under both Subchapter M and the excise tax on regulated investment companies under Section 4982. More specifically, if the fund did not distribute this extra income, the fund would be subject to corporate level income tax and a 4 percent excise tax on the amount retained. Although the fund could use some portion of the fee to rebuild its NAV, it would not receive 100 percent of the benefit.111 Proposed Solution. The best solution for both the money market fund and the shareholder would be for the liquidity fee to be treated as a fee to the shareholder—and thus possibly deductible— and capital gain to the fund. Given the uncertainty in this area, the IRS and the Treasury Department would need to issue some type of formal guidance (though not necessarily regulations) permitting capital gain treatment to the fund. This outcome would be consistent with existing case law and long- standing tax policies of matching income with expenses. F. OPERATIONAL IMPLICATIONS 1. Difficulties of Intraday Gating Whether a fund reached the trigger point would need to be determined after the close of business and before the next day’s opening to allow sufficient time for all transactions to be processed overnight. This is imperative for a number of compelling reasons. Liquidity Calculations. Determining the moment during the day that a fund trips the liquidity trigger is not possible without substantial and costly systems modifications to provide functionality that only would be used on extremely rare occasions. Most money market fund investors purchase shares or maintain their accounts with intermediaries, such as broker-dealers. Intermediaries’ systems do not submit shareholder transactions on a real-time basis to funds. Most intermediaries aggregate trade orders, which are sent to the fund on a periodic basis through batch processing cycles (some intraday, but many after the close of business). Intraday liquidity calculations would require enhancements to portfolio management, transfer agent, and intermediary systems to get updated portfolio and shareholder purchase and redemption information for calculation purposes on a more real-time basis throughout the day. Obtaining this information on direct fund transactions flowing from multiple streams and intermediary transactions would be extremely challenging, if not impossible. Indeed, significant shareholder activity is conducted through intermediaries “away from the fund” and the nature of that activity is not known to the fund until after close of business, when the fund receives 111 An argument could be made that the fee received by the fund should be treated as capital gain, rather than ordinary income, because it is being used to offset capital losses incurred by the fund on its portfolio to pay the redeeming shareholder. Because the capital gain would be offset by the capital loss, the money market fund would not have an additional distribution requirement. See Arrowsmith et al v. Commissioner of Internal Revenue, 344 U.S. 6 (1952) (stockholder was required to treat a judgment paid as a capital loss, not an ordinary loss, because the amount was paid in connection with the liquidation of the corporation, which generated capital gains). The IRS took this position in Revenue Procedure 2009-10 with respect to amounts paid to a money market fund by the fund adviser to prevent the money market fund from breaking the dollar. The Revenue Procedure provided only temporary guidance, however, so funds and their legal advisors likely would be reluctant to rely upon this argument without further guidance from the IRS. See Rev. Proc. 2009-10, 2009-2 IRB 267. Financial Stability Oversight Council January 24, 2013 Page 56 batch files from intermediaries overnight. Intermediaries likely would move to alternative products rather than make significant systems modifications to capture, segregate, and forward shareholder trades to the fund or transfer agent intraday. Notification and Processing Concerns. Intraday gating also would add layers of complexity with respect to notification of intermediaries and shareholders, accurate processing of shareholder transactions intraday (applying the appropriate cutoffs), and significant labor-intensive manual intervention that would be expensive and fraught with risk. Funds would need to make costly enhancements to various systems—including portfolio accounting, transfer agency, and ancillary transaction systems—to be able to impose a redemption gate intraday. Intermediaries also would have to implement changes to various systems that are far more complicated than funds’ systems to accommodate intraday gating. Notification of intraday gating to thousands of intermediaries in any quick and comprehensive method would be challenging. There is no existing centralized information conduit for real-time messaging amongst fund industry participants, so imposition of a gate would be communicated through web-postings, blast emails, faxes, and telephone calls. This would be a very manual, time-consuming, and labor-intensive process for both funds and intermediaries to inform shareholders that a fund has been gated intraday and that redemption transactions received after the gated time would be rejected. The current processing environment would present difficulties for intermediaries and funds in appropriately segregating orders to ensure only those redemption orders received prior to the imposition of the gate are processed and investors are treated fairly. Intermediaries currently have no need to segregate trade orders received intraday. Intraday gating would require substantial system changes by intermediaries that would have to be triggered at a moment’s notice to segregate orders (received from various processing streams) intraday.112 Advantages of End-of-Day Gating. Since a redemption gate imposed after the close of business is similar to an unscheduled market close, fewer system enhancements are required. The processes and procedures for managing unscheduled market closures exist today, suggesting that both funds and intermediaries would be more likely to continue sponsoring and using money market funds with end-of- day gates that would only be triggered in prescribed and very unusual circumstances. 2. Systems Modifications for End-of-Day Gating Temporary gating would require fund transfer agent and intermediary system providers to ensure their systems can suppress redemption activity while supporting all other transaction types. In some 112 Moreover, the fund would not have any transparency regarding the time individual investor orders were received or executed by the intermediary, or that a proper cutoff of redemption transactions has occurred (for files submitted to the fund after the gated time). The fund would be completely dependent on the intermediary’s ability to segment orders by time to properly implement an intraday redemption gate. This reliance on intermediaries to apply the appropriate intraday cutoffs would likely require modification to agreements, additional certifications, and changes to prospectus disclosure (all costly endeavors). Financial Stability Oversight Council January 24, 2013 Page 57 instances, system providers use redemption-like functions to facilitate non-redemption transactions;113 such instances will require system modification. The ability to implement gating quickly—before the processing of transactions begins for a given business day—would be essential. Although fund transfer agent and intermediary system providers can assess fees on mutual fund redemptions today, applicable systems would require modification to handle a proposed temporary liquidity fee for money market funds. The nature of the liquidity fee would require support for a separate fee type, for reporting to investors and to isolate the fee for appropriate tax treatment. These system modifications are far less onerous and costly, however, than the intricate and expensive programming and other system changes necessary to apply continuous redemption restrictions accurately and consistently across all investors as contemplated by an MBR, as discussed in Section V. G. IMPACT ON CERTAIN TRANSACTION TYPES Sweep vehicles use money market funds at the end of the business day to invest available cash held in customer accounts. This cash is intended primarily to support trading activity conducted in investor accounts during the business day. Because intermediaries would not know whether their overnight sweeps would be subject to a temporary gate until after the daily investment is made, requiring prime money market funds to use temporary gates may cause sweep vehicles to seek alternative investment products. Similarly, sponsors of 401(k) and other retirement plans may need to reevaluate whether a money market fund with a temporary gate requirement continues to be an appropriate product that meets their plans’ needs for ready liquidity. These decisions may turn on intermediaries’ assessments about the likelihood of the funds triggering the liquidity gates. IV. FSOC Alternative One: Floating NAV FSOC Alternative One would require all money market funds to have a floating NAV instead of a stable NAV. It would implement this change by “removing the special exemption that currently allows [money market funds] to utilize amortized cost accounting and/or penny rounding to maintain a stable NAV.”114 Under this proposal, each money market fund would reprice its shares to $100.00 and reflect the actual market value of the underlying portfolio holdings, “consistent with the valuation requirements that apply to all other mutual funds.” A. A $100 PRICE IS ARBITRARY AND WITHOUT PRECEDENT No current laws or regulations require any investment company registered under the Investment Company Act to offer its shares at any particular price. Nonetheless, FSOC has proposed that money market funds reprice their shares from $1.00 (the price that money market funds customarily seek to maintain) to $100.00 per share, so that the funds’ NAVs would be more sensitive to fluctuations in the 113 At least one fund transfer agent system utilizes special redemption processing to honor investor requests to direct dividends earned in one CUSIP for reinvestment into a different CUSIP. 114 See Report, supra note 2, at 30. Financial Stability Oversight Council January 24, 2013 Page 58 value of the portfolios’ underlying securities. Under Accounting Series Release No. 219 (“ASR 219”),115 which predates the adoption of Rule 2a-7, the SEC permits mutual funds, including money market funds, to use amortized cost to value securities, provided the securities have maturities of 60 days or less. ASR 219 also provides that the appropriate standard of materiality when calculating a fund’s NAV is 1/10th of 1 percent. Thus, for mutual funds that typically offer their shares at an initial price of $10 (i.e., funds that are not money market funds), this equates to $0.01; accordingly, a NAV per share of between $9.995 and $10.005 would round to $10.00. Rule 2a-7, as adopted in 1983, permits a money market fund to use “penny rounding” on a $1.00 NAV; as a result, a money market fund NAV of between $0.995 and $1.005 would round to $1.00. This penny-rounding convention, and the use of amortized cost to value all of its securities, is available to a money market fund provided that it complies with the risk-limiting provisions of the rule. ASR 219, therefore, authorizes funds to use amortized cost to a limited extent outside of Rule 2a-7. Further, we believe virtually all mutual funds round their NAV to the nearest penny, regardless of share price, when calculating NAV for purposes of processing trades in fund shares.116 By contrast, FSOC’s proposal would seem to require money market funds to comply with a pricing standard that is at least 10 times more onerous than the standard articulated in ASR 219. Thus, for example, under ASR 219, an ultrashort bond fund manager could initially price the fund’s shares at $10 and structure its portfolio so that its NAV fluctuates less than a floating money market fund with the proposed $100 NAV (e.g., by creating a very liquid portfolio with a majority of its securities maturing in 60 days or less). We question why sponsors would offer and investors would buy a $100 NAV money market fund that potentially floats more than an ultrashort bond fund, especially if, as discussed below, tax, accounting, and operational complexities associated with a floating NAV product destroy the convenience and simplicity of money market funds for investors. B. A FLOATING NAV WILL NOT PREVENT INVESTOR RUNS FSOC suggests that requiring money market funds to float their NAVs will reduce the tendency of money market funds to experience large redemptions during periods of financial stress because “regular fluctuations in [money market fund] NAVs likely would cause investors to become accustomed to, and more tolerant of, fluctuations in NAVs.”117 Evidence from products with floating NAVs suggests this 115 See Valuation of Debt Instruments by Money Market Funds and Certain Other Open-End Investment Companies, SEC Release No. IC-9786, Accounting Series No. 219 (May 31, 1977), 42 FR 28999 (June 7, 1977). If amortized cost does not represent fair value due to an impairment of the creditworthiness of an issuer, ASR 219 requires the fund to consider the impairment in determining the fair value of the security. 116 Rule 2a-4 under the Investment Company Act provides support for this practice. In particular, Rule 2a-4(b) provides that daily accrual of fund expenses, dividends receivable, and interest income need not be reflected in NAV so long as they do not cumulatively amount to as much as 1 cent per outstanding share. 117 See Report, supra note 2, at 30. Financial Stability Oversight Council January 24, 2013 Page 59 conclusion is incorrect. Investors may still choose to redeem fund shares whether the NAV of a money market fund is floating or stable.118 For example, while ultrashort bond funds are not required to follow Rule 2a-7, they do invest in a portfolio of relatively short-dated securities. In contrast to money market funds, however, the NAV of an ultrashort bond fund fluctuates. Beginning in the summer of 2007, the average NAV on these funds began to fall (Figure 14). In February and March 2008, several ultrashort bond funds posted significant declines in their NAVs, and the average NAV of these funds fell about 2 percent. This preceded a large outflow of assets from such funds; during a four-week period ending in early April 2008, these funds experienced cumulative outflows of 15 percent of their assets. By the end of 2008, assets of these funds were down about 60 percent from their peak in mid-2007. 118 See D. Blackwell, K. Troske, and D. Winters, Money Markets Funds Since the 2010 Regulatory Reforms: More Transparency, Increased Liquidity, and Lower Credit Risk, Center for Capital Markets Competitiveness (Fall 2012), available at http://www.uschamber.com/sites/default/files/reports/FinalpaperwithCover_smalltosend.pdf, at 36 (“a floating NAV does not change investors’ incentives to remove their money quickly when they believe there has been a change in the riskiness of the fund. In other words, [money market funds] reporting floating NAVs can still experience runs.”). Financial Stability Oversight Council January 24, 2013 Page 60 FIGURE 14 Weighted Average NAV and Net New Cash Flow of Ultrashort Bond Funds Weekly Sources: Investment Company Institute and Morningstar -5 -4 -3 -2 -1 0 1 2 3 8.6 8.8 9.0 9.2 9.4 9.6 9.8 10.0 2004 2005 2006 2007 2008 Dollars Percentage of total net assets Weighted average NAV Net new cash flow Financial Stability Oversight Council January 24, 2013 Page 61 The experience in Europe of certain money funds likewise demonstrates that floating NAV funds also can face strong investor outflows during periods of market turmoil. For example, French floating NAV dynamic money funds (or trésorerie dynamique funds), lost about 40 percent of their assets over a three-month period from July to September 2007.119 Thus, it is highly doubtful that floating the NAV of money market funds would accomplish FSOC’s objective: inducing fund shareholders to refrain from reacting during periods of market stress. It could, of course, prompt investors to abandon money market funds in favor of alternative products that seek to maintain a stable NAV. In that case, whatever risks FSOC sees in stable NAV Rule 2a-7 money market funds simply would shift to a less-regulated, more opaque part of the market. C. INVESTORS WOULD SEEK STABLE NAV ALTERNATIVES One very significant concern is whether investors would continue to use money market funds were the stable NAV eliminated. For investors, the stable NAV money market fund provides a host of tax, accounting, recordkeeping, and operational benefits that are not currently available for floating NAV mutual funds. Citing these benefits, a wide range of businesses, state and local government entities, financial services companies, and consumer organizations have argued that a floating NAV would destroy the convenience and simplicity of money market funds for investors, and compromise an important source of financing for many segments of the U.S. economy.120 Also weighing in against a floating NAV are many individual investors who strongly oppose changing the fundamental nature of money market funds. Furthermore, surveys of money market fund investors indicate clearly that most do not want and would not use a floating NAV product. For example, a survey of corporate treasurers and other institutional investors indicated that nearly 80 percent of respondents would either decrease their use of money market funds or discontinue using them altogether if money market funds are required to have a floating NAV. Based on this response, more than 60 percent of corporate money market fund assets would move to other investments if this concept were adopted.121 A survey of retail money market fund investors commissioned by T. Rowe Price and conducted online by Harris Interactive indicated much the same response.122 Two-thirds of retail investors surveyed found the idea of a floating NAV money market fund unfavorable. Among those who reacted to the 119 For a more detailed discussion of the experience of certain money and bond funds in Europe, see MMWG Report, supra note 4, at 106–107. 120 For examples of businesses, governments, financial services, and consumer organizations that have voiced support for maintaining the stable NAV for money market funds, see http://www.preservemoneymarketfunds.org/what-others-are- saying/. 121 See TSI Survey, supra note 28. 122 Based on a study commissioned by T. Rowe Price and conducted online by Harris Interactive from August 31 to September 7, 2010, of 413 adults aged 35–75 who own money market funds outside of a retirement plan, who also own at least one long-term mutual fund, who invest directly with a mutual fund company, do not rely solely on the advice of an investment adviser, and have $100,000 or more in investable assets. The data are weighted to be representative of the adult population with $100,000 or more in investable assets. A full methodology is available upon request. Financial Stability Oversight Council January 24, 2013 Page 62 concept unfavorably, 72 percent indicated that they would use the product less, and that their most likely response would be to close their money market fund accounts (29 percent), decrease their money market fund balances (33 percent), or execute fewer money market fund transactions (10 percent). A third survey, conducted among both retail and institutional shareholders by Fidelity Investments, found much the same result.123 This survey found that institutional investors overwhelmingly (89 percent) indicated a preference for keeping the stable NAV and more than half (57 percent) indicated they would use money market funds less or not at all if faced with the prospect of a floating NAV. Retail investors also disliked the floating NAV concept. Seventy-four percent of the retail investors surveyed favored keeping the stable NAV and 47 percent of those surveyed said they would move all or some of their assets out of money market funds if funds changed to a floating NAV. In short, data on the subject demonstrate that investors do not want and likely would reject a floating NAV money market fund, especially if, as discussed below, the tax, accounting, and operational implications of a floating NAV are not resolved first. 1. Tax Implications FSOC acknowledges that a stable NAV offers significant convenience in terms of tax compliance. The stable NAV relieves shareholders, funds, and intermediaries from having to track capital gains and losses and determine for every redemption which share was redeemed, the tax basis (generally, the acquisition cost) of that share, and whether the holding period of that share was long term or short term.124 Because a money market fund’s shares are typically bought and sold at the same price, shareholders do not realize capital gain or loss upon a redemption; thus all of the fund’s returns are ordinary income to its shareholders. The stable NAV also eliminates the need to consider the timing of sales and purchases of fund shares (i.e., wash sale tax rule considerations).125 To be sure, investors already 123 See Fidelity Investments, The Investor’s Perspective: How Individual and Institutional Investors View Money Market Mutual Funds and Current Regulatory Proposals Designed to Strengthen Money Funds (December 2, 2011). 124 When an individual shareholder redeems shares from a mutual fund, the tax laws require the fund to send the shareholder and the IRS an IRS Form 1099-B. The fund must report on this form the shareholder’s gross proceeds from the sale and, for shares acquired after January 1, 2012, the shareholder’s cost basis in the shares redeemed, to permit the shareholder (and the IRS) to calculate the shareholder’s capital gain or loss. For shares acquired in 2012 and thereafter, the fund also must report to the shareholder whether such capital gain or loss is long-term or short-term. Forms 1099-B must be sent by February 15 of the year following the year in which the sale transaction took place. These rules also apply to shareholders that are partnerships or S corporations. Certain other shareholders, however, including corporations, financial institutions, retirement plans, and other tax-exempt entities, are “exempt recipients” to which information reporting is not required. Corporations and other exempt recipients, however, would still be responsible for tracking their cost basis and reporting any gains or losses. Money market funds that maintain a stable NAV under Rule 2a-7 currently are exempt from these information reporting rules. 125 The Internal Revenue Code imposes rules intended to prevent taxpayers from recognizing capital losses on securities if the taxpayer has not truly liquidated its position in that security. The “wash sale” rule thus prevents taxpayers from recognizing losses on the sale of securities if, within 30 days before or after such sale, the taxpayer purchased substantially identical shares. In the money market fund context, the wash sale rule poses particular problems in connection with a floating NAV. Many money market fund investors automatically reinvest their dividends. Money market funds typically declare dividends daily and pay them monthly, so a redemption from the fund would almost always be within 30 days of a dividend reinvestment. The wash sale rule thus would prevent a shareholder who redeems shares from the fund from ever recognizing a capital loss, until the shareholder completely liquidates its position. Any disallowed loss on a redemption would be added to the basis of Financial Stability Oversight Council January 24, 2013 Page 63 face these burdens in connection with investments in mutual funds with floating NAVs. But most investors make fewer purchases and sales from such mutual funds because they are used for long-term investing, not cash management. And in any case, many purchases (or exchanges) in floating NAV funds are made within tax-advantaged accounts (e.g., 401(k) plans) where such issues do not arise. Thus, if money market funds are required to float their NAVs, current law would require these funds to maintain and report cost basis information for redemptions that occur throughout the year. Floating NAV money market funds also would be subject to the transfer reporting requirements, which require brokers and funds to transfer cost basis information if an account moves between brokers or between brokers and funds. These new requirements present significant operational and recordkeeping burdens and costs. • Mandatory cost basis reporting became effective for mutual fund shares acquired beginning in 2012. Implementation of these rules has been a costly and labor-intensive exercise for the industry, one that took several years to execute. Because money market funds were exempt from these rules, fund complexes and intermediaries have not included money market funds in the systems necessary to capture, report, and transfer cost basis. Implementing cost basis reporting for money market funds would require additional time, effort, and costs, which likely will be borne by investors. Further, these burdens will be amplified because of the sheer volume of transactions conducted by shareholders in money market funds. These costs must be understood and their impact evaluated as part of any floating NAV cost-benefit analysis. • Brokers and fund sponsors typically offer investors a range of features tied to their money market funds, including automated teller machines (“ATM”) access, checkwriting, electronic check payment processing services and products, and Fedwire transfers. If the current information reporting requirements are applied to money market funds, a fund would be required to send a Form 1099-B for every sale transaction that occurs during the year (though only to nonexempt recipients). For retail investors, every check written or ATM withdrawal, for example, would constitute a reportable sale.126 The costs of this compliance would similarly need to be part of any cost-benefit analysis done on a floating NAV proposal. Fully implemented relief for shareholders, fund sponsors, and intermediaries from additional tax burdens of moving from a stable NAV to a floating NAV product would be an essential first step to gaining any investor support for this new type of money market fund. FSOC has indicated that the Treasury Department and the IRS may provide relief regarding basis reporting and wash sale issues; the newly purchased shares. Under the cost basis reporting rules, funds and brokers would be required to track these wash sales for any shareholders that are not exempt recipients and adjust their cost basis accordingly. Corporations, which do not receive cost basis information from the funds, would have to track these wash sales and basis adjustments themselves. 126 From the shareholder standpoint, a floating NAV could mean that the shareholder must recognize gain or loss (including minute variations that cancel out over time) for tax purposes for each money market fund redemption. From a tax reporting standpoint, an individual shareholder would receive in January or February a Form 1099-B from the fund for each redemption, though all of the information may be consolidated into a single statement or mailing. The individual shareholder would be able to rely upon these forms when completing his or her own tax return, but would still face the inconvenience of receiving and reporting such a large quantity of tax information each year. Financial Stability Oversight Council January 24, 2013 Page 64 however, the Report fails to provide any details on the relief they will provide. We offer the following initial suggestions for possible ways such relief might be achieved. Cost-Basis Reporting • Netting of Transactions. The IRS could provide guidance allowing a money market fund to net all of an investor’s transactions throughout a calendar year. The money market fund then would send only one Form 1099-B, to the extent that the shareholder has a net gain or loss in the fund for the year.127 This would simplify reporting by the fund or intermediary, as well as the investor. As noted above, this solution would solve the problem of multiple Forms 1099-B, but the money market fund still would have to calculate and maintain the shareholder’s cost basis. The IRS would have to be comfortable that it has regulatory authority under the tax code to allow such netting.128 This also would require funds and intermediaries to implement systems, processing, and reporting changes to net transactions. • No Reporting for De Minimis Amounts. The IRS also could provide that funds are not required to send Forms 1099-B for capital gains or losses that are of a de minimis amount.129 If applied in conjunction with the netting approach discussed above, a de minimis exception could reduce further the amount of reporting by shareholders and funds. Although the money market fund would not have to send Forms 1099-B for any sale resulting in gains or losses that do not exceed the de minimis threshold, the fund still would have to track all of the transactions and have some mechanism in place to initiate reporting when a gain or loss exceeded the de minimis amount. Wash Sales • One solution for resolving the wash sale issue would be to exempt from the rule either (i) automatic dividend reinvestments or (ii) money market funds entirely. Either such exemption likely would require a statutory change. From a tax policy standpoint, Congress and the IRS might be reluctant to permit such exceptions because they could create opportunities for abuse and would negate the purpose of the wash sale rule. One could argue, however, that money market funds are not the type of investment at which the wash sale rule is aimed. The rule is intended to prevent a taxpayer from generating current losses in a security that the taxpayer continues to own and expects to rise in value. Investors do not expect any capital appreciation in a money market fund, so it is unlikely that money market fund shareholders would abuse 127 Short-term and long-term capital gains must be reported separately. Therefore, if the shareholder has both long- and short- term capital gains or losses for the year, the fund would send two Forms 1099-B. 128 Section 6045(a) of the Internal Revenue Code, which provides the rules for gross proceeds and cost basis reporting, does seem to provide the Treasury Department some regulatory authority to determine when reporting is required. 129 Under current law, payors are not required to send Forms 1099-B for dividends or interest to a payee if the aggregate amount paid to that payee is $10 or less per calendar year. The IRS could apply a similar rule for money market fund redemptions, such that aggregate gains or losses for a calendar year that fall below a certain amount would not be subject to reporting. The $10 de minimis exception for dividends and interest is statutory, but Section 6045(a) may grant the Treasury Department authority to implement a similar rule for capital gains and losses. Financial Stability Oversight Council January 24, 2013 Page 65 the rule. Further, any dividends paid and reinvested likely would be small, meaning that any losses disallowed due to the wash sale rule also would be small. The nature of the product and the amount of potential losses may be sufficient justification for an exception from the wash sale rule. It is important to note that providing the specified relief would not cure FSOC’s proposal for floating NAVs of its significant shortcomings, nor justify FSOC’s recommending this alternative to the SEC. 2. Accounting Implications FSOC also acknowledges that there may be accounting implications related to floating NAV money market funds. For example, it is unclear whether a floating NAV money market fund would still meet the characteristics of a “cash equivalent” under relevant accounting guidance. U.S. GAAP currently includes investments in money market funds as one of three examples of a cash equivalent (along with Treasury bills and commercial paper). From a corporate investor’s perspective, investments in floating NAV money market funds may not be considered cash equivalents under GAAP. If so, they could not be presented in the “cash and cash equivalents” line item in the balance sheet.130 Investors in corporate securities like to see cash on the balance sheet as a measure of financial strength. Thus, although changing the balance sheet presentation of investments in money market funds from “cash and cash equivalents” to “investments” would not affect the economic substance of the corporation’s financial position, it may nevertheless act as a strong disincentive for corporations to use money market funds for cash management purposes. If corporate investments in money market funds are not cash equivalents, they would instead be considered investment securities held for trading purposes under GAAP. Investment securities are marked-to-market on the balance sheet, and gains and losses (both realized and unrealized) flow through to earnings. Although gains and losses on investments in money market funds likely would be immaterial to the corporation’s earnings, they still would need to be tracked for both book and tax purposes. The burden associated with tracking gains and losses would act as a further disincentive to corporate investment in money market funds and may cause these investors to migrate to other products. Similar to the tax implications of moving to a floating NAV, these costs must be understood and evaluated as part of any floating NAV cost-benefit analysis. If regulators proceed with the floating NAV reform option, it would be important for the Financial Accounting Standards Board (“FASB”) or the SEC to issue prior guidance indicating that money market funds continue to qualify as cash equivalents under GAAP, notwithstanding a floating NAV. We believe such guidance would be appropriate given that money market funds’ investments would continue to be subject to the risk-limiting provisions included in Rule 2a-7. 130 GAAP defines cash equivalents as short-term, highly liquid investments that are both (i) readily convertible to known amounts of cash, and (ii) so near their maturity that they present insignificant risk of changes in value because of changes in interest rates. Generally, only investments with original maturities of three months or less qualify under that definition. Original maturity for this purpose means original maturity to the entity holding the investment. Financial Stability Oversight Council January 24, 2013 Page 66 3. Specialized Business Applications and Automated Systems For many investors, a variety of business applications involving automated and specialized systems has made the $1.00 per share pricing vitally important to the usefulness of money market funds. For example, same-day settlement (T+0 processing) is important to many investors.131 Many money market funds utilize systems that support the same-day settlement transaction process, in which proceeds are remitted through the Fedwire system throughout the day at various cutoff times (e.g., 2:00 p.m., 4:00 p.m., 5:00 p.m.). This is possible because the stable NAV allows sponsors to calculate each fund’s NAV at the appropriate cutoff times using amortized cost, absent a material credit event during the day that moves the NAV below $0.995 or above $1.005. Elimination of the stable NAV for money market funds likely would force intermediaries and fund sponsors to consider how or whether they could continue to provide such services to shareholders requiring same-day settlement. Floating NAV money market funds may not be able to ascertain intraday market prices for securities held in the fund (pricing vendors typically provide prices for money market securities only once a day at 4:00 p.m.). Therefore, to continue to offer same-day settlement for investors several times a day, funds not only would need to obtain intraday data for securities from pricing vendors, but also would need to make significant systems modifications to support the NAV calculation process. Otherwise the valuation process generally would occur only at the end of the day and the Fedwire system either would have to accommodate large amounts of redemption activity and related settlements near the end of the day, or would have to remain open later.132 The use of amortized cost accounting and a stable NAV also allow the efficient processing of cash balances through cash sweep programs, in which all customer cash balances are “swept” into investments in shares of money market funds that are owned by the customers but transacted through fund accounts registered to a broker-dealer or a bank. Examples of specialized systems that use stable NAV money market funds to hold short-term liquidity include trust accounting systems at bank trust departments, corporate payroll processing, corporate and institutional operating cash balances, federal, state and local government cash balances, municipal bond trustee cash management systems, consumer receivable securitization cash processing, escrow processing, custody and investment manager cash balances, employee pension benefit plan processing, broker-dealer and futures dealer customer cash balances, and cash management–type accounts at banks and broker-dealers.133 131 Approximately two-thirds of money market fund assets are in institutional share classes that primarily use same-day settlement for their money market fund transactions. 132 The Report notes that the DWS Variable NAV Money Fund conducts same-day settlement. See Report, supra note 2, at n. 79. Discussions with the fund’s management, however, reveal that it is only able to accommodate same-day settlement because it currently receives very few redemption requests per day. The small volume can be processed after the market closes and settled same-day prior to that day’s close of the Fedwire system. 133 For a detailed description of each of these specialized systems that use stable NAV money market funds to hold temporary liquidity balances, see Letter from John D. Hawke, Jr., Arnold & Porter LLP, on behalf of Federated Investors, Inc., to The Honorable Mary L. Schapiro, Chairman, Securities and Exchange Commission (November 2, 2012), available at http://www.sec.gov/comments/4-619/4619-274.pdf. Financial Stability Oversight Council January 24, 2013 Page 67 Requiring money market funds to float their NAVs would impose significant operational and systems changes and related costs. Both intermediaries and investors will evaluate these burdens in addition to the effect on product utility (e.g., no longer a stable NAV), and many may determine (as investor survey research has demonstrated) that a floating NAV money market fund is not a viable cash management tool. 4. Statutory Prohibitions and Investment Restrictions Many institutional investors also face legal or other constraints that preclude them from investing their cash balances in pools that do not maintain a stable NAV. For example, many state laws and regulations authorize municipalities, insurance companies, and other state-regulated entities to invest in stable NAV funds, sometimes explicitly including funds operating in compliance with Rule 2a-7. Corporations also may have board-approved policies permitting them to invest operating cash (balances used to meet short-term needs) only in pools that seek to maintain a stable NAV. Indentures and other trust documents may authorize investments in money market funds on similar grounds. Thus, absent a stable NAV, many state and local governments, corporations, and securities issuers no longer would be able to use money market funds to help manage their cash.134 D. FLOATING THE NAV WOULD HARM THE MARKET Assets in money market funds now total $2.7 trillion. As discussed above, money market fund investors of all types are unlikely to use a floating NAV product, especially if the tax, accounting, and operational complexities of moving to a floating NAV are not fully addressed first. Requiring money market funds to float their NAVs thus would risk precipitating a vast outflow of assets from money market funds to other products. As of November 2012, money market funds held more than one-third of corporate commercial paper and about three-quarters of state and local government short-term debt. Shrinkage of money market fund assets would significantly disrupt the flow of short-term financing within the U.S. economy. The principal impact of a floating NAV for money market funds, therefore, would be a major restructuring and reordering of intermediation in the short-term credit markets. If assets move to less- regulated and less-transparent products or structures, risks in the financial markets will increase. This was the primary concern cited by SEC Commissioner Luis Aguilar in his decision not to support Chairman Schapiro’s reform agenda last summer: “I remain concerned that the Chairman’s proposal will be a catalyst for investors moving significant dollars from the regulated, transparent money market fund market into the dark, opaque, unregulated market.”135 Requiring money market funds to float their NAVs assuredly would shift credit intermediation from one type of product to others. There are a number of alternative products that money market fund investors could use, including enhanced cash pools, local government investment pools, and other vehicles that seek to maintain a stable unit price but are not regulated under the Investment Company 134 See Appendix D of the MMWG Report, supra note 4. 135 See Aguilar Statement, supra note 8. Financial Stability Oversight Council January 24, 2013 Page 68 Act.136 Regulatory changes that push assets from highly regulated, transparent products (i.e., money market funds) to less-regulated and less-transparent products arguably serve to increase systemic risk. These products had their own difficulties during the financial crisis.137 Many intermediaries already have the ability through banks to select among various sweep arrangements that seek to offer a stable unit value, such as money market fund sweeps, repurchase agreement sweeps, commercial paper sweeps, and, importantly, sweeps into offshore (non-money market fund) accounts (e.g., Eurodollar sweeps).138 If money market funds no longer provide a stable NAV, investors can and will migrate to these other sweep arrangements, including some (e.g., Eurodollar sweeps) largely beyond the jurisdictional reach of U.S. regulators. Although banks are one option for investors, corporate cash managers and other institutional investors do not view an undiversified holding in an uninsured (or underinsured) bank account as having the same risk profile as an investment in a diversified short-term money market fund subject to the risk-limiting conditions of Rule 2a-7. Such investors would continue to seek out diversified investment pools. Further, if investors did shift their liquid balances to conventional bank deposits, the federal government’s potential insurance liability would increase, as would moral hazard—a development that would surely increase systemic risk. These observations are consistent with the SEC Staff Study, which found that a shift by money market fund investors to bank deposits would “increase reliance on FDIC deposit insurance and increase the size of the banking sector, which raises additional concerns about the concentration of risk in the economy.”139 In addition, a shift to traditional banks would result in a significant reduction in the supply of short-term credit to corporate America unless banks raised significant amounts of capital to support their expanded balance sheets. Even if they could raise the capital to support this expansion, the market would be less efficient and the cost of short-term credit would rise.140 Furthermore, municipalities would lose an important source of financing in the short-term markets because banks cannot pass through tax- exempt income and simply could not replace tax-exempt money market funds. In sum, the principal impact of a floating NAV for money market funds likely would be a major restructuring and reordering of intermediation in the short-term credit markets. Some of this impact may be lessened, however, if the complexities noted above involving tax, accounting, and operations are resolved by putting appropriate legislation, rules, or guidance in place before mandating a move to a floating NAV structure. 136 For an overview of some of these alternatives, see MMWG Report, supra note 4, at 41-46; SEC Staff Study, supra note 10, at 38-46. 137 See MMWG Report, supra note 4, at 62–64; Appendix B. 138 For a general discussion of overnight sweep arrangements, see MMWG Report, supra note 4, at 43–44. 139 See SEC Staff Study, supra note 10, at 45. 140 See, e.g., Remarks by Carol DeNale, CVS Caremark, stating that “[a]pproximately 40 percent of my outstanding CP [commercial paper] at any given time is owned by a 2a-7 fund. That’s an amazingly important part of our capital structure, and one that will not easily be replaced.” See SEC Roundtable, supra note 7. Financial Stability Oversight Council January 24, 2013 Page 69 E. TRANSITION AND GRANDFATHER ISSUES FSOC’s proposal suggests that existing money market funds could be grandfathered in and allowed to maintain a stable NAV for a phase out period, potentially lasting five years. Under the proposal, the SEC would prohibit any new share purchases in the grandfathered funds after a predetermined date. Any new money invested after that date would go into floating NAV money market funds. On the surface, this dual strategy seems investor-friendly; however, the reality is that such an approach would be confusing and costly to investors, funds, and intermediaries. For example, funds and intermediaries would need to create and hold two positions for each investor—one in the stable NAV fund and one in the floating NAV fund. Transfer agent recordkeeping and servicing charges to the fund typically entail relatively fixed costs associated with maintaining each shareholder position on its books and records. These costs, which are ultimately borne by investors, typically are billed on a per-account basis. As a result, the expenses incurred for investor recordkeeping during the grandfathering period likely would double under the proposal. Disparate cost basis reporting requirements between stable and floating NAV accounts also would create confusion for investors and investor servicing. Systematic pre- scheduled transactions, such as retail investor systematic withdrawal plans (“SWPs”) or automatic investment plans (“AIPs”), could conceivably occur on different accounts, requiring transitions of all SWP or AIP transactions to the floating NAV account at the end of the grandfathered period. Such transitions are not typically done systematically but require manual intervention. Significant additional legal, accounting, compliance, and operational costs would be incurred to wind down stable NAV funds and create new floating NAV fund structures. Indeed, FSOC’s concern about disruption around the transition from stable to floating NAV is well-founded, as the transition, in and of itself, could be destabilizing to the financial markets. It could require money market funds to shed hundreds of billions of dollars of commercial paper, bank certificate of deposits, Eurodollar deposits, repurchase agreements, and other assets, as their investors redeem in favor of other products. Even assuming the calmest of financial market conditions, this would be an unsettling and difficult process. During a period of stress in the money market, such a transition could set off the very kind of systemic event that FSOC seeks to avoid. F. REMOVING EXEMPTIONS UNDER THE INVESTMENT COMPANY ACT IS NOT NECESSARY The Report states that if money market funds were required to float their NAVs, the SEC would need to rescind two exemptive rules: Rule 22e-3, which allows a fund board to suspend redemptions and begin an orderly liquidation if the fund has broken or is about to break the dollar; and Rule 17a-9, which allows money market fund affiliates to purchase portfolio securities from a fund for a variety of purposes, including to help the fund maintain a stable NAV. The Report asserts that because a floating NAV money market fund is designed to fluctuate in value, these types of exemptions no longer would be necessary. The Report does not acknowledge or address, however, the possibility that the elimination of these rules might make episodes of heavy redemptions more likely or more severe. Rule 22e-3 facilitates the orderly disposal of assets in a troubled fund in a manner that protects the interests of all shareholders—making it possible to avoid a fire sale of portfolio securities or a first-mover advantage for early redeemers. As discussed above, the experience of products with a floating NAV Financial Stability Oversight Council January 24, 2013 Page 70 indicates that variable NAVs do not preclude the possibility of substantial shareholder outflows. Indeed, FSOC acknowledges that outflows from a money market fund might occur even if the fund had a floating NAV. The exemptive rule, therefore, provides needed flexibility for emergency situations: it is difficult for the SEC to provide individual exemptive orders as quickly as a fund’s board would be required to react. Similarly, the SEC adopted Rule 17a-9 in 1996 (and then expanded in the 2010 amendments) to codify a series of no-action letters in which the SEC staff agreed not to recommend enforcement action if affiliated persons of a money market fund purchased portfolio securities from the fund to prevent the fund from realizing losses on the securities that may otherwise have caused it to break the dollar. As noted above in Section II.B.1., the SEC adopted the rule because in its experience these type of transactions appeared to be “fair, reasonable, in the best interests of fund shareholders, and consistent with the requirement that money market funds dispose of a defaulted security in an orderly manner as soon as practicable.”141 Moreover, sponsors engage in affiliated support transactions for a variety of reasons, often having little to do with any risk that the fund might break the dollar. For instance, a fund sponsor may buy downgraded securities from a fund’s portfolio to maintain the fund’s AAA credit rating. FSOC provides no basis for suggesting these types of transactions would not continue to be in the best interests of shareholders in a floating NAV money market fund, or that an ad hoc exemptive order process, which requires valuable SEC staff time, is better than an exemptive rule.142 V. FSOC Alternative Two: NAV Buffer and Minimum Balance at Risk Under FSOC Alternative Two, stable NAV money market funds no longer could use the amortized cost method or penny rounding to maintain a stable NAV. Instead, the proposal would require stable NAV money market funds to maintain a risk-based NAV buffer of up to 1.00 percent (in excess of assets needed to maintain a $1.00 share price).143 In addition to this NAV buffer, the proposal would require a money market fund, irrespective of current market conditions, to delay redemptions of a portion of a shareholder’s account. This portion, the “minimum balance at risk” or MBR, would consist of 3 percent of a shareholder’s highest account value in excess of $100,000 during the previous 30 days. If a money market fund were to suffer losses that exceed its NAV buffer, these losses would be borne first by the MBRs of shareholders who had recently redeemed. The proposal would not apply to Treasury money market funds, and the MBR requirement would not apply to investors with account balances below $100,000. 141 See 1996 MMF Reform Release, supra note 58, at 13974; 2010 MMF Reform Release, supra note 5, at 10088. The rule imposes strict conditions designed to assure that any such transactions do not provide the opportunity for abusive conduct by fund affiliates. 142 Indeed, this reflects the SEC’s own views on the existence of an exemptive rule for these types of transactions. “[W]e believe that the alternative of funds obtaining no-action assurances from the Commission staff for these transactions, particularly during times of market stress, is time consuming and inefficient.” See 2010 MMF Reform Release, supra note 5, at 10087–10088. 143 In Alternative Two, the Report estimates that the average NAV buffer would be 0.84 percent for prime funds, 0.80 percent for tax-exempt funds, and 0.70 percent for government funds. Our discussion of Alternative Two’s proposed NAV buffer is included in Section VI. Financial Stability Oversight Council January 24, 2013 Page 71 The Report asserts that the NAV buffer and the MBR would reduce a money market fund’s susceptibility to runs by allowing a fund to absorb day-to-day fluctuations in the value of its portfolio securities, providing a disincentive for shareholders to redeem in times of stress, and allocating more fairly the costs to the fund that can result when shareholders do redeem. The hypothesis is that the MBR would prevent or mitigate redemption pressure by removing investors’ incentives to be among the first to redeem (the so-called first-mover advantage), while also making explicit the fact that money market funds entail risks to their investors. In our judgment, Alternative Two is deeply flawed, and its likeliest impact will be to drive investors as well as intermediaries away from money market funds. A. EFFECTS ON INVESTOR LIQUIDITY AND MARKET SIZE Importantly, an MBR restriction would impair a core mutual fund investor protection and reverse more than 70 years of SEC practice in fund regulation. Under the Investment Company Act, one hallmark feature of mutual funds, including money market funds, is that they issue “redeemable securities,” meaning that the fund stands ready to buy back its shares at their current NAV. Section 22(e) of the Investment Company Act generally prohibits funds from suspending the right of redemption and from postponing the payment or satisfaction upon redemption of any redeemable security for more than seven days, except under extraordinary circumstances that are delineated in the statute or determined by SEC rule. Under this authority, in 2010, the SEC adopted Rule 22e-3, which exempts money market funds from Section 22(e) to permit them to suspend redemptions and postpone payment of redemption proceeds—but only in very limited circumstances, i.e., in order to facilitate an orderly liquidation of the fund.144 By contrast, the MBR would permanently alter the ability of money market fund investors to redeem all of their shares on a daily basis.145 ICI strongly opposes any sort of redemption restriction that would impair investor liquidity when liquidity is readily available within the money market fund. If ultimately adopted, an MBR limitation on redemptions represents an experiment on the $2.3 trillion prime, government, and tax-exempt money market fund industry that could have harmful consequences for the broader financial markets, including financing for businesses and state and local governments. Moreover, although the Report asserts that an MBR would provide a disincentive for shareholders to redeem in times of stress, FSOC has not provided any data or analysis supporting this assertion. In contrast, based on discussions with investors, our members have indicated that an MBR would increase a shareholder’s likelihood of redeeming during a 144 When it adopted Rule 22e-3, the SEC noted that the rule “is intended to reduce the vulnerability of investors to the harmful effects of a run on the fund, and minimize the potential for disruption to the securities markets.” 2010 MMF Reform Release, supra note 5, at 10088. The SEC recognized, however, that permitting suspension of this statutory protection should be limited to extraordinary circumstances, stating: “Because the suspension of redemptions may impose hardships on investors who rely on their ability to redeem shares, the conditions of the rule limit the fund’s ability to suspend redemptions to circumstances that present a significant risk of a run on the fund and potential harm to shareholders. The rule is designed only to facilitate the permanent termination of a fund in an orderly manner.” Id. 145 Although the “gating” concept described in Section III would impair an investor’s right to redeem, it only would be used for a temporary period of time under tightly prescribed and very unusual circumstances. Financial Stability Oversight Council January 24, 2013 Page 72 financial crisis. Indeed, many have suggested that shareholders would be more likely to redeem at the slightest sign of stress in the markets, given the punitive nature of the MBR. Investor reaction to continuous redemption restrictions, such as the MBR, also suggests that imposition of an MBR would greatly reduce investor use of money market funds. In a survey of corporate treasurers and other institutional investors, 90 percent of these investors indicated that they would reduce their usage of money market funds, or stop using them altogether, if MBR restrictions were put in place.146 Calculations based on these investors’ responses suggest that institutional assets in money market funds would shrink by two-thirds if the restrictions were imposed. Investors that hold accounts directly with funds may choose alternative products that are less regulated and more opaque, but would far better meet their liquidity needs. This movement would seem unlikely to reduce systemic risk and, indeed, would be more likely to increase risk. B. OPERATIONAL IMPLICATIONS An MBR also would create serious operational issues that would reduce or eliminate the usefulness of many services that money market funds and financial providers extend to investors. ICI recently issued a paper that focuses on the operational implications of a concept similar to the MBR.147 As discussed in our study, throughout the 40-year history of money market funds, investors have benefited from the convenience, liquidity, and stability of these funds. Individual or retail investors use money market funds as a tool that provides a current money market rate of return on cash that is awaiting investment or other disposition, that is held as savings, or that constitutes the stable value component of an investment or retirement portfolio. Institutional investors—which for these purposes include corporations of all sizes, state and local governments, securities lending operations, bank trust departments, sweep programs, securities brokers, and investment managers—use money market funds as a cost-effective way to manage and diversify credit risk, while providing same-day liquidity with market- based yields. To meet these various objectives, funds, intermediaries, service providers, and investors have developed a variety of arrangements for distributing and using money market funds efficiently. Investors 146 See TSI Survey, supra note 28. BlackRock Inc., in separate interviews of its institutional money market fund shareholders, found “virtually without exception” that redemption restrictions in the form of a “minimum account balance” “would cause [shareholders] to abandon [money market funds].” See BlackRock ViewPoint, Money Market Funds: The Debate Continues—Exploring Redemption Restrictions, Revisiting Floating NAV (March 2012), available at https://www2.blackrock.com/webcore/litService/search/getDocument.seam?venue=PUB_IND&source=GLOBAL&cont entId=1111160117; see also Letter from Barbara G. Novick, Vice Chairman, and Richard K. Hoerner, Managing Director, Head of Global Cash Management, BlackRock, to the Financial Stability Oversight Council (December 13, 2012) at 18. 147 See ICI Redemption Holdback Study, supra note 18. Unlike the concept explored in our study, the MBR requirement would not apply to investors with account balances below $100,000. Although clearly a benefit for smaller accounts, the $100,000 exemption actually entails an additional level of operational complexity and cost that would further dissuade intermediaries from offering and investors from utilizing money market funds. It also would create significant servicing challenges for funds and intermediaries to ensure that investors (i) understand the implications of the threshold and (ii) not work around the requirement—e.g., by carefully allocating investments among multiple accounts or funds in amounts below the threshold to preserve flexibility in meeting cash needs. Financial Stability Oversight Council January 24, 2013 Page 73 can purchase and redeem money market fund shares directly from fund sponsors or through a wide array of platforms, portals, and financial intermediaries such as broker-dealers and retirement plans. Money market funds are the primary investment for sweep accounts offered by broker-dealers and financial advisers. Investors also benefit from the convenience of check-writing or debit-card access to their money market funds. These offerings depend critically on an intricate and complex operational infrastructure created by the industry that allows investors to transact smoothly and efficiently, often with same-day settlement. Implementing FSOC’s proposed freeze on shareholders’ assets would require changes to myriad complex systems that extend well beyond those under the control of the funds themselves.148 Fund complexes, intermediaries, and service providers have developed these systems to communicate and process significant volumes of money market fund transactions on a daily basis through a variety of mechanisms on behalf of investors. To apply continuous redemption restrictions accurately and consistently across all investors in money market funds, each of these entities, including a host of intermediaries, would need to undertake intricate and expensive programming and other significant, costly system changes. Our analysis indicates that the costs of these changes could be prohibitive, particularly if FSOC’s changes greatly curb investor interest in money market funds, as numerous surveys clearly indicate they will. The MBR requirement, in itself, would remove money market funds as a viable option in many instances. Fiduciaries, such as retirement plans, trustees, and investment advisers, may be legally prohibited from using money market funds with constant redemption restrictions for their clients, because such restrictions would impair clients’ liquidity and be punitive in nature. Sweep programs, which rely upon the ability to move 100 percent of an investor’s available cash on a daily basis, would not be able to employ money market funds if they are subject to a constant holdback of investor assets. Retail investors’ ability to access their money market fund balances in excess of $100,000 through check writing and other redemption delivery methods also would be impaired. In other uses, funds, intermediaries, and institutional investors conceivably could restructure and reprogram operational systems to incorporate daily redemption restrictions. ICI’s Redemption Holdback Study provides an overview of the systems and processes that would require modification by thousands of institutional investors, funds, intermediaries, and service providers. Based on ICI’s cost- benefit analysis of a prior rule proposal requiring extensive systems and operational changes, it is reasonable to expect that requiring money market funds to adopt an MBR would cost the industry 148 The Report states that “[Money market funds] would be required to apply the MBR requirement to each of their recordholders. This would include recordholders that are financial intermediaries, such as banks or broker-dealers that hold shares on behalf of their customers, unless the intermediaries provide the [money market fund] sufficient information to apply the MBR requirement to the intermediaries’ individual accounts directly.” Report, supra note 2, at 44. It would be extremely burdensome and cost prohibitive for funds to undertake shadow recordkeeping of underlying shareholder activity in intermediary accounts for the purposes of applying the MBR at the investor level for customers that are being serviced exclusively by intermediaries. Applying the MBR at the intermediary level also is unworkable for a number of reasons. See ICI Redemption Holdback Study, supra note 18, at 34–35. Financial Stability Oversight Council January 24, 2013 Page 74 hundreds of millions of dollars.149 These costs are largely fixed and not scalable to the size of the asset base. It would be difficult for intermediaries, in particular, to justify such expenses even if money market fund assets were to remain at their current level. The likely consequences of an MBR requirement thus are mutually reinforcing. Fund complexes, intermediaries, and service providers would be hard-pressed to justify undertaking the significant costs of compliance with the restrictions in the face of the rapid shrinkage of money market fund assets. We believe many intermediaries would make the business decision to migrate to unregulated or less- regulated money market investment vehicles or bank deposit products where possible, in lieu of implementing costly changes to their systems in order to continue to offer money market funds to a dwindling shareholder base. The total effect would be to drive users away from money market funds, disrupt short-term financing for the economy, and increase the use of less-regulated, less-transparent alternatives. VI. FSOC Alternative Three: NAV Buffer and Other Measures Alternative Three contemplates that stable NAV money market funds would have a risk-based NAV buffer of up to 3.00 percent to provide an explicit loss-absorption capacity that could be combined with other measures to “enhance the effectiveness of the buffer and potentially increase the resiliency of [money market funds].”150 These other measures could include more stringent investment diversification requirements, increased minimum liquidity levels (e.g., 20 percent daily and 40 percent weekly liquid assets as a share of total fund assets), and more robust disclosure requirements. To the extent these additional measures, either alone or in combination with other measures, reduce the vulnerabilities of money market funds, the Report indicates FSOC could include these additional measures in its final recommendation and reduce the size of the NAV buffer by an unstated amount. Like Alternative Two, this proposal would not apply to Treasury money market funds. Importantly, the Report indicates that the other measures that would accompany the NAV buffer in Alternative Three would be unlikely to provide the same protections that FSOC claims would result from the MBR in Alternative Two. Therefore, the NAV buffer contemplated in Alternative Three would serve as the primary tool to increase the resiliency of money market funds and reduce their vulnerability to runs. Accordingly, this NAV buffer must be “significantly larger” than the NAV buffer in Alternative Two. The Report estimates that the average NAV buffer under Alternative Three would be approximately 2.51 percent for prime funds, 2.39 percent for tax-exempt funds, and 2.10 percent for government funds. 149 Two years ago, ICI conducted a cost-benefit analysis of proposed changes to Rule 12b-1 under the Investment Company Act that would have required extensive systems and operational changes. The estimated costs for these changes were $231 million for fund complexes only, not including additional costs that would have been incurred by intermediaries. See Investment Company Institute, Cost-Benefit Analysis of SEC Rule 12b-1 Reform Proposal (December 1, 2010), available at http://www.ici.org/pdf/10_12b1_sec_cba.pdf, at 11, Figure 4. We believe the costs of the changes required to implement FSOC’s redemption restrictions easily could meet or exceed this prior estimate for fund complexes, without even beginning to take into account the costs for thousands of intermediaries and money market fund service providers to transition to an MBR requirement. 150 Report, supra note 2, at 51. Financial Stability Oversight Council January 24, 2013 Page 75 As discussed below, Alternative Three likewise is a deeply flawed proposal. Its likeliest impact would be to impel money market fund sponsors to exit the business, thus depriving investors, issuers, and the economy of the benefits these funds provide. A. NAV BUFFER In a recent ICI study, we analyzed the likely outcomes of a NAV or “capital buffer” for the money market fund industry.151 Our study considered several variations on the NAV buffer idea, including requiring money market fund advisers to commit capital, requiring funds to raise capital in the market, or requiring funds to build a NAV buffer inside the funds from fund income. Our analysis clearly shows the infeasibility of building capital at the levels suggested for either Alternative Two or Alternative Three. A summary of our findings, as well as an analysis of the NAV buffers contemplated in the Report, is provided below. 1. Requiring Fund Advisers to Commit Capital Proposals requiring money market fund advisers to commit capital to absorb possible future losses in their funds would alter fundamentally the money market fund business model. A money market fund, like every other mutual fund, provides investors a pro rata interest in the fund, whereby fund investors share in the risks and rewards of the securities held by the fund. All of the fund’s shares are equity capital. The default risk of the highly diversified, short-term portfolios held by money market funds is very low, and is shared equally by all fund investors, so the likelihood that an individual investor will experience a sizeable loss, or any loss at all, is remote. Imposing NAV buffer requirements on a fund adviser would transform the essential nature of a money market fund by interposing the adviser between the fund and its investors, essentially requiring the adviser to guarantee a portion of the fund. Currently, fund advisers do not allocate capital to absorb losses because, as with all securities products, investors bear the risks of investing in funds. To be sure, some money market fund advisers have at times voluntarily supported their funds. But these advisers did so as a business decision. Requiring all fund advisers to take a first-loss position would be a radical departure from the current agency role that fund advisers play. The mutual fund structure, including that of money market funds, is designed so fund advisory fees compensate the adviser for managing the fund as a fiduciary and agent and for providing ongoing services that the fund needs to operate. Advisers are not compensated for bearing investment risks of the fund. Shifting investment risks from fund investors to advisers would require advisers to dedicate capital to absorb possible losses of the funds that they manage. Some advisers would have to raise new capital in the market. Others could perhaps shift capital from other parts of their businesses. Either way, all advisers would have to earn a market rate of return on such capital. If they cannot earn that rate of return, they would find better business alternatives, such as seeking to move investors to less-regulated cash management products where investors still must bear the risks of investing. 151 See ICI Capital Buffer Study, supra note 19. Financial Stability Oversight Council January 24, 2013 Page 76 While the potential for losses is remote, the cost of providing a NAV buffer likely would be significant. Under money market funds’ current structure, small and highly infrequent losses are spread across a large number of fund investors and a large asset base. Under the structure being contemplated, small losses would be concentrated in a single investor (the adviser) and across a small asset base (the value of the NAV buffer). The adviser could face large percentage losses on its NAV buffer investment and thus would require a compensatory rate of return. In theory, advisers could seek to pass along to investors the cost of providing the NAV buffer to absorb investment risks. As a practical matter, however, we doubt this is possible. Because of the very low interest rate environment, advisers at present have no ability to pass along cost increases; doing so would raise fund expense ratios, dropping net returns below zero. Even in a more normal interest rate environment, advisers would have difficulty passing the cost of the required capital on to fund investors. Rule 2a-7’s risk-limiting provisions effectively place a ceiling on what a prime money market fund may earn. Yields on Treasury funds set a floor on the yields that prime funds may return to investors after expenses, which in turn limits the fees that prime funds may charge. No rational investor would hold a prime money market fund that offered a return below that of a Treasury fund. In addition, any proposed increase in a fund’s advisory fees must be put to a shareholder vote. Shareholder votes can be costly to undertake and outcomes by no means would be guaranteed. Even if shareholders accepted a fee increase, the necessary increase could be so large as to reduce the net yield on a prime fund below that of a Treasury money market fund. All else being equal, an increase in a fund’s advisory fee will lower the fund’s net yield. Any desire to offset the effect on the fund’s yield by holding riskier and therefore higher yielding securities would be constrained by the risk-limiting provisions of Rule 2a-7 and would run directly counter to the goals of regulators. By far the most likely outcome is that advisers wishing to continue to sponsor money market funds would have to absorb the cost of providing the NAV buffer. Although outcomes depend on the particulars of any proposal, a net present value (“NPV”)152 analysis of the NAV buffers proposed in Alternatives Two and Three indicate advisers would reconsider the money market fund business model.153 The NPV calculations shown in Figures 15 through 17 plot 152 Net present value is a capital budgeting technique that uses a criterion for deciding whether to undertake an investment or project. In brief, a project’s NPV is calculated as the present value of future cash returns discounted at an appropriate interest rate, less the present value of the initial cost of the investment. If the NPV is positive, the project adds value to the firm and is viewed as “acceptable.” If the NPV is negative, the project erodes the value of the firm and is viewed as “unacceptable.” If the NPV is equal to zero, the firm is indifferent to the project because it neither adds to nor subtracts from the value of the firm. 153 In the ICI Capital Buffer Study, supra note 19, we performed an internal rate of return (“IRR”) analysis on the net cash flows generated from an upfront provision of a 3 percent NAV buffer with a 10-year fixed stream of additional fee income to “repay” the adviser for the buffer “contribution.” An IRR analysis and an NPV analysis performed on net cash flows under such a structure will produce equivalent results. This is not the case for net cash flows derived under FSOC’s Alternatives Two and Three. Under these scenarios, funds are given a transition period (two years for Alternative Two and six years for Alternative Three) to build up the applicable NAV buffer. This structure generates net cash flows that, for the most part, are negative (additional fee income less outlay for NAV buffer) for the first two years under Alternative Two and for the first six years under Alternative Three and then positive (additional fee income) for the next eight years under Alternative Two and for the next four years under Alternative Three. An NPV analysis on these net cash flows will produce more accurate results than an IRR analysis. Financial Stability Oversight Council January 24, 2013 Page 77 the NPV for different fee levies based on the Report’s estimated NAV buffer ratios.154 Under Alternative Two, FSOC estimated NAV buffers of 0.84 percent for prime funds, 0.70 percent for government funds,155 and 0.80 percent for tax-exempt funds. Under Alternative Three, FSOC estimated NAV buffers of 2.51 percent for prime funds, 2.10 percent for government funds, and 2.39 percent for tax- exempt funds. Prime Funds: Under Alternative Three, advisers of prime funds would outlay a total of $37.3 billion156 for the NAV buffer in equal installments over a six-year transition period. At a discount rate of 5 percent, the NPV is negative until fees are increased by 27.5 basis points—the point at which the NPV is zero (Figure 15). At higher fee levels, the NPV turns positive as additional fee income reduces the net amount of the NAV buffer provided by the adviser in the first six years and increases the payments to advisers in the final four years. Under Alternative Two, advisers would outlay a total of $12.5 billion157 for the NAV buffer in two equal annual installments and the NPV is negative until prime money market fund fees are increased by almost 10 basis points. 154 The NPV measures the return in billions of dollars to the money market fund industry from providing the NAV buffer over the transition period and receiving a fixed stream of additional fee income over a 10-year period. The analysis assumes that money market fund assets remain constant, and that the funds incur no losses that would require a payment to funds over the 10-year period. The net cash flows are discounted at a 5 percent rate. 155 Consistent with the Report, government money market funds are defined as those that invest primarily in U.S. Treasury obligations and U.S. government agency securities. This definition excludes Treasury money market funds—those funds that invest primarily in U.S. Treasury obligations and repurchase agreements collateralized with U.S. Treasury obligations. 156 Calculated by the Investment Company Institute as 2.51 percent of $1.5 trillion in total net assets of prime funds as of January 2, 2013. 157 Calculated by the Investment Company Institute as 0.84 percent of $1.5 trillion in total net assets of prime funds as of January 2, 2013. Financial Stability Oversight Council January 24, 2013 Page 78 FIGURE 15 Net Present Value (NPV)1 of NAV Buffer: Prime Money Market Funds Billions of dollar s 1 Ten-year stream of net cash flows discounted at a 5 percent rate Government Funds: Under Alternative Three, advisers to government funds would outlay a total of $10.3 billion158 for the NAV buffer in equal installments over a six-year transition period. At a discount rate of 5 percent, the NPV is negative until fees are increased by 23 basis points—the point at which the NPV is zero (Figure 16). Under Alternative Two, advisers would outlay a total of $3.4 billion159 for the NAV buffer in two equal annual installments and the NPV is negative until fees are increased by a little more than 8 basis points. 158 Calculated as 2.10 percent of $490 billion in total net assets of government funds as of the end of November 2012. See Report, supra note 2, at 9. 159 Calculated as 0.7 percent of $490 billion in total net assets of government funds as of November 2012. See Report, supra note 2, at 9. -40.00 -30.00 -20.00 -10.00 0.00 10.00 20.00 30.00 40.00 0 5 10 15 20 25 30 35 NPV with 2.51% NAV buffer NPV with 0.84% NAV buffer Increase in fees (basis points) NPV = 0 NPV = 0 Financial Stability Oversight Council January 24, 2013 Page 79 FIGURE 16 Net Present Value (NPV)1 of NAV Buffer: Government Money Market Funds Billions of dollars 1 Ten-year stream of net cash flows discounted at a 5 percent rate Tax-Exempt Funds: Under Alternative Three, advisers of tax-exempt funds would outlay a total of $7 billion160 for the NAV buffer in equal installments over a six-year transition period. At a discount rate of 5 percent, the NPV is negative until fees are increased by 26 basis points—the point at which the NPV is zero (Figure 17). Under Alternative Two, advisers would outlay a total of $2.3 billion161 for the NAV buffer in two equal annual installments and the NPV is negative until fees are increased by almost 10 basis points. 160 Calculated by the Investment Company Institute as 2.39 percent of $291 billion in total net assets of government funds as of January 2, 2013. 161 Calculated by the Investment Company Institute as 0.8 percent of $291 billion in total net assets of government funds as of January 2, 2013. -10.00 -5.00 0.00 5.00 10.00 15.00 0 5 10 15 20 25 30 35 NPV with 2.10% NAV buffer NPV with 0.70% NAV buffer Increase in fees (basis points) NPV = 0 NPV = 0 Financial Stability Oversight Council January 24, 2013 Page 80 FIGURE 17 Net Present Value (NPV)1 of NAV Buffer: Tax-Exempt Money Market Funds Billions of dollars 1Ten-year stream of net cash flows discounted at a 5 percent rate Based on these calculations, it is foreseeable that many, if not most, fund advisers would make the business decision to change their cash management offerings, particularly with respect to Alternative Three. (We believe the MBR aspect of Alternative Two alone will cause advisers and intermediaries to cease offering money market funds.) It is questionable whether advisers have the ability to pass along to shareholders the estimated increases in fees that would be necessary to recoup the costs of providing the capital buffer (i.e., the increase in fees at which the NPV = 0). Under Alternative Three, these fee increases range from 23 to nearly 28 basis points for government, tax-exempt, and prime money market funds. The current low interest rate environment and the regulatory requirement to obtain an affirmative shareholder vote to raise fees make this alternative completely unworkable. Even if advisers were able to pass along to investors the cost of the capital buffer, the necessary rise in fund fees would reduce the net yields of prime and government funds below those of Treasury money market funds and push net yields on tax-exempt funds even further below Treasury funds. Investors would not invest in these funds when Treasury funds would yield a superior tax-adjusted return. Although the estimated fee increases are lower under Alternative Two (between 8 and 10 basis points), the same arguments stated for Alternative Three still hold—low interest rates provide little room to raise fees; shareholder votes are costly and outcomes are not guaranteed; and the reduction in yields relative to a Treasury money market fund reduces the demand for prime, government, and tax- exempt money market funds. As a result, some advisers may simply liquidate their funds and not offer alternative products. Others may refocus their efforts on alternative cash-like products that are less regulated and less transparent, thereby increasing risks in the financial markets. -8.00 -6.00 -4.00 -2.00 0.00 2.00 4.00 6.00 8.00 0 5 10 15 20 25 30 35 NPV with 2.39% NAV buffer NPV with 0.80% NAV buffer Increase in fees (basis points) NPV = 0 NPV = 0 Financial Stability Oversight Council January 24, 2013 Page 81 For many fund sponsors, adviser-provided capital would raise difficult financial accounting issues as well. Under GAAP, a money market fund adviser that commits first-loss capital to its funds probably would be required to consolidate the financial positions of its funds on the adviser’s books. Bank-related advisers may be forced to hold “double capital” against their funds—once to meet any NAV buffer requirement, and the second time to comply with Basel capital standards.162 The amount of additional capital banks would have to hold is uncertain, as it would depend on how regulators interpret and apply Basel capital standards in this case. Fund advisers indicate that the additional capital that banks might need to hold to meet Basel capital requirements could total between 0.25 percent and 4.0 percent of their money market fund assets. 2. Requiring Funds to Raise Capital in the Market As an alternative to requiring fund advisers to commit capital, the Report suggests that the NAV buffer could be funded by raising capital in the market. ICI engaged capital markets experts to help study this approach in depth.163 We ultimately concluded, for several reasons, that market-provided capital is not a feasible option for the money market fund industry. Adding subordinated debt or equity would turn a rather simple product—the money market fund—into a considerably more complex offering. Small funds and small fund complexes likely would find it difficult and costly to issue and roll over subordinated securities, resulting in further industry consolidation and raising a barrier to entrants. The approach also would potentially create tension between the subordinated investors’ desire to avoid losses and the tolerance of senior shareholders (i.e., traditional money market fund investors) for taking greater risks for greater yields. Finally, issuing subordinated debt would add “rollover risk” to money markets funds, because investors in this class of money market fund shares may well be reluctant to roll over their investments in times of market stress. Thus, the capital would disappear just when it might actually be needed. A market-raised capital buffer would reduce the yield available to senior shareholders, and subordinated investors would have a highly levered—and hence potentially volatile—investment.164 The compensation subordinated investors would demand for assuming such volatility would reduce the yield available to the senior share class. Another aspect of a market-raised capital buffer is that the smaller the capital buffer, the larger the potential losses to the subordinated investors. While the fund would be required to raise less capital, the resulting subordinated securities would be more levered, more volatile, and therefore more expensive and difficult to sell. Capital market experts identified several other issues that could complicate the use of this structure. To be marketable, the subordinated securities would need to obtain a credit rating (and thus be structured as debt), but for various reasons, credit rating agencies would not be likely to treat the 162 See Basel III Capital Standards, supra note 66. 163 ICI engaged Barclays Capital, PricewaterhouseCoopers, and Sidley Austin LLP to analyze the potential for funds or advisers to raise capital through the capital markets. 164 As previously discussed, concentrating losses to a smaller investor base and smaller asset base results in the subordinated investors taking on the potential for large percentage losses on their investments. They would demand a compensatory rate of return. Financial Stability Oversight Council January 24, 2013 Page 82 securities as debt. The legal structure of the subordinated securities—whether they are issued by the fund or issued by a special purpose bankruptcy remote entity—also would pose challenges. In addition, while in theory capital could be raised more quickly in the markets than through retained earnings, launching a new form of security is likely to be a complex and time-consuming process, possibly requiring more than 560 individual money market funds to enter the market seeking to raise capital simultaneously. Finally, it is unclear how well this structure would protect senior share class investors during times of market stress. 3. Requiring a Within-Fund Capital Buffer Building a within-fund capital buffer would align the costs of the buffer more directly with the fund’s beneficiaries: fund shareholders. Legal and accounting considerations, however, would limit a within-fund capital buffer to 0.5 percent of a fund’s total assets.165 Capital at this level would not absorb large credit losses; at best it would provide funds somewhat greater flexibility in selling securities at a price below amortized cost. Also, because of tax and economic considerations, a fund likely would need many years to build such a buffer.166 As the analysis in the ICI Capital Buffer Study shows, under plausible assumptions, building such a buffer might take a typical prime fund 10 to 15 years. The exact horizon depends on whether short-term interest rates rise somewhat more quickly than is currently expected, on how investors respond to a buildup of a within-fund capital buffer, and on the willingness of advisers to continue to absorb the cost of maintaining large fee waivers. B. OTHER MEASURES As discussed above, we have significant objections to imposing a NAV buffer requirement on money market funds. Sponsor-provided capital is beyond the reach of most fund sponsors. Market- provided capital is completely unknown to investors and likely a source of instability. Shareholder- provided capital could not be accumulated at any appreciable rate in the current near zero interest rate environment. In any form, therefore, a NAV buffer for money market funds would do little to reduce systemic risk—but would reduce choice and competition. As with the MBR in Alternative Two, this option is not a viable answer response to the money market fund reform debate. 165 Under SEC rules and GAAP, a fund can accumulate a capital buffer of no more than 0.5 percent of net assets. The fund’s NAV would remain fixed at $1.00 only until the fund’s mark-to-market value rises to $1.0050. At that point, the fund would have to adjust its NAV upward to $1.01, breaking the dollar on the upside. Any buffer in excess of $0.0050 would require regulatory relief from the SEC and FASB. 166 The Internal Revenue Code limits the speed at which a within-fund capital buffer can be built. In any given year, mutual funds, including money market funds, are required to pay out to shareholders at least 90 percent of their annual earnings. Otherwise they must pay corporate income tax on all their earnings (and tax-exempt money market funds would lose their ability to flow through tax-exempt income to their shareholders). Failing to meet this test would impose significant double taxation on fund shareholders: once when the fund pays income tax on its earnings, and a second time when the investor pays income tax on earnings received. To avoid this significant double taxation, a money market fund could at most set aside 10 percent of its annual earnings (assuming it has earnings) toward a capital buffer. In addition, if a money market fund were to retain some of its income to build a capital buffer, it would have to pay corporate income tax, presumably at a 35 percent rate, on the amount retained. This would reduce the amount that a money market fund could set aside to 6.5 percent or less of its income in any given year. Financial Stability Oversight Council January 24, 2013 Page 83 FSOC Alternative Three also contemplates adding measures “to enhance the effectiveness of the buffer and potentially increase the resiliency of [money market funds.]”167 We offer the following cautions and observations about making additional changes to the risk-limiting provisions of Rule 2a-7. 1. More Stringent Investment Diversification Requirements Rule 2a-7 requires a money market fund’s portfolio to be diversified, both as to the issuers of the securities it acquires and to the guarantors of those securities. Generally, money market funds must limit their investments in the securities of any one issuer (other than government securities) to no more than 5 percent of fund assets and their investments in securities subject to a demand feature or a guarantee from any one provider to no more than 10 percent of fund assets. Due to unprecedented market conditions and consolidations since the financial crisis, however, the universe of institutions issuing or providing guarantees or liquidity for eligible money market securities has become extremely limited. Further restricting the diversification limits may only heighten this problem by potentially forcing money market funds to invest in less creditworthy issuers to meet new diversity requirements. It also could materially reduce the amount of funding that money market funds provide to larger (and potentially safer) issuers. This could have the effect of actually increasing the risk within money market funds’ portfolios, rather than decreasing it. 2. Increased Minimum Liquidity Requirements Increased minimum liquidity levels may limit money market funds’ investment risks and increase their ability to meet heightened redemption requests without selling portfolio securities. Nevertheless, seeking to increase funds’ liquidity levels and thus reduce funds’ WAMs raises other concerns. One significant issue is whether there is a sufficient supply of money market securities at the very short end of the yield curve. In particular, market participants have voiced concerns about the limited availability of short-dated Treasury securities. Another concern is that particular issuers of money market securities may have a preference for funding at the longer-end of the money market maturity spectrum. For instance, an issuer might, in October, wish to lock in short-term financing over the calendar year-end in order to avoid rollover risk. Further limiting the ability of money market funds to provide such financing (e.g., three-month commercial paper) could be detrimental. 3. Shareholder Transparency One measure that we encourage regulators to pursue is additional “know your investor” requirements to provide money market funds with increased visibility into omnibus accounts, portals, sweep arrangements, or other trading platforms. Such requirements would improve money market funds’ ability to understand their shareholder base and to predict investors’ redemption activity. Money market funds, however, are not in a position independently to obtain this information. We instead recommend that each relevant member of FSOC consider rules to provide a clear legal mandate that, upon request of a money market fund, intermediaries under its jurisdiction must furnish sufficient 167 See Report, supra note 2, at 51. Financial Stability Oversight Council January 24, 2013 Page 84 investor information to aid the fund’s efforts to enhance its existing know your investor programs. For example, intermediaries, upon request, could provide funds with investor-specific data related to trading activity over a specified period or investor data related to holdings of a certain percentage. Such data would assist the fund’s adviser and board in monitoring a fund’s investor profile and adjusting liquidity accordingly. Actual investor names and other proprietary data, however, would not need to be provided. We note that in previous rulemakings the SEC has imposed an obligation on funds, but not on intermediaries, to obtain similar information from intermediaries (e.g., Rule 22c-2 under the Investment Company Act, concerning redemption fees, and Rule 204-2(a)(18) under the Investment Advisers Act of 1940, known as the adviser “pay-to-play” rule). Getting this information, however, has proven to be quite difficult, burdensome, and costly for funds when intermediaries do not have any legal obligation to provide it. In the pay-to-play context, the SEC staff had to grant relief precisely because funds and their advisers were unable to obtain the information mandated by the rule.168 By imposing an affirmative legal requirement on intermediaries, FSOC members can ensure that more complete information is provided to funds. VII. Economic Impact of Proposed Recommendations Under Section 120 of the Dodd-Frank Act, FSOC is required to “take costs to long-term economic growth into account” when recommending new or heightened standards and safeguards for a financial activity or practice. Measured against this statutory mandate, the Report’s economic analysis has a number of significant shortcomings that exaggerate potential benefits of its proposed reforms and may significantly underestimate their costs to the economy. 168 For example, Rule 204-2(a)(18)(i)(B) under the Investment Advisers Act requires all investment advisers to maintain a list of all government entities that are or were an investor in any covered investment pool to which the investment adviser provides or has provided investment advisory services. Because many of these government accounts are held in omnibus positions on the fund’s books and records, and because the fund and its adviser would not have information on the shareholders in these accounts, the rule requires fund advisers to obtain shareholder information from the omnibus accountholders. Omnibus accountholders receiving these requests, however, have no legal obligation to provide such information to the funds requesting it. Recognizing that the lack of transparency in omnibus accounts impeded the ability of advisers to comply with the rule, the SEC staff issued no-action relief permitting advisers to satisfy the recordkeeping requirement of Rule 204-2(a)(18)(i)(B) through an alternative means that does not require advisers to pierce their omnibus accounts or seek information from their omnibus accountholders in order to comply with the rule. See Investment Company Institute, SEC No-Action Letter (September 12, 2011). Financial Stability Oversight Council January 24, 2013 Page 85 Further, while the purpose of the Report is to recommend specific rules for the SEC to promulgate, the Report’s economic analysis fails to address—let alone satisfy—the SEC’s statutory and rulemaking requirements for analysis of the economic consequences of any eventual rule. When the SEC engages in rulemaking, it is required to consider “in addition to the protection of investors, whether [the rule proposal] will promote efficiency, competition, and capital formation.”169 In considering these factors, the SEC is obliged to “determine as best it can the economic implications of the rule it has proposed”170 and assess the baseline of benefits provided under existing regulations to “determine whether, under the existing regime, sufficient protections existed” so as to “reduce the need for, and hence the benefit to be had from” the proposed rule.171 It is notable that the Report nowhere discusses how the Council’s proposals would impact efficiency, competition, and capital formation. Nor, as discussed below, does it take into account whether sufficient protections exist under current SEC regulations—particularly Rule 2a-7 as amended in 2010—to reduce the need for additional reform. If FSOC hopes that its recommendations actually might come to fruition, then we question why FSOC would use its Section 120 authority to propose recommendations without giving any consideration to whether the recommended proposals will satisfy the SEC’s own governing statutes and other regulatory requirements. A. FSOC’S ESTIMATED BENEFITS OF PROPOSED RECOMMENDATIONS ARE ILLUSORY As justification for new regulations, FSOC provides an argument that the Report’s recommendations would bolster the resilience of money market funds by reducing future outflows during crises, which, in turn, would lower the probability and dampen the severity of any future crises. It arrived at this conclusion by reasoning that outflows from prime money market funds led to a reduction in the overall supply of short-term credit and significantly worsened the 2007–2008 financial crisis. The Council states that “expected reductions in the probability or severity of crises associated with [money 169 See Section 2(c) of the Investment Company Act. This analysis must be “substantially complete” at the proposing stage. See Division of Risk, Strategy, and Financial Innovation and Office of General Counsel, Securities and Exchange Commission, Current Guidance on Economic Analysis in SEC Rulemakings (March 16, 2012), available at http://www.sec.gov/divisions/riskfin/rsfi_guidance_econ_analy_secrulemaking.pdf, at 16. The United States Court of Appeals for the District of Columbia Circuit has emphasized repeatedly how important it is for the SEC to consider the costs regulated entities would incur in order to comply with a rule. See, e.g., Business Roundtable and Chamber of Commerce v. Securities and Exchange Commission, No. 10-1305 slip op. (D.C. Cir. July 22, 2011) (In vacating the SEC’s proposed proxy access rules, the Court noted that “[h]ere the Commission inconsistently and opportunistically framed the costs and benefits of the rule; failed adequately to quantify the certain costs or to explain why those costs could not be quantified; neglected to support its predictive judgments; contradicted itself; and failed to respond to substantial problems raised by commenters.”); American Equity Investment Life Insurance Company v. Securities and Exchange Commission, Case No. 09-1021 (July 21, 2009) (finding that the SEC’s analysis of effects on efficiency, competition, and capital formation in adoption of rules related to indexed annuities was arbitrary and capricious, and remanding the matter to the SEC for reconsideration); and Chamber of Commerce v. Securities and Exchange Commission, 412 F.3d 133, 144 (June 21, 2005) (“Uncertainty…does not excuse the Commission from its statutory obligation to do what it can to apprise itself—and hence the public and the Congress—of the economic consequences of a proposed regulation before it decides whether to adopt the measure.”). 170 See Chamber of Commerce, 412 F.3d at 143. 171 See American Equity, 613 F.3d at 178–79. Financial Stability Oversight Council January 24, 2013 Page 86 market fund] reform would imply a sizable net benefit in terms of higher expected economic growth, given the very large costs of financial crises on economic output.”172 Certainly, financial crises inflict large costs on economic growth, and measures that could demonstrably reduce the risks of such crises would provide commensurate benefits. To attribute those benefits to a particular set of reforms, however, regulators must demonstrate that those reforms can credibly reduce the risks of future crises. In this case, however, the Report fails to show that the reforms it advocates would reduce risks. The Report merely asserts this—and thus fails to clear even the lowest threshold for analyzing the benefits of its proposed regulatory policy. FSOC’s analysis is flawed in several ways: • The Report’s “baseline” for the alleged risks of money market funds is the 2007–2008 financial crisis, completely ignoring the effects of the SEC’s 2010 reforms to Rule 2a-7. The Report thus significantly overstates the potential risks money market funds might pose in a future crisis. • The Report’s analysis implicitly assumes that the regulatory system can ensure that investors in short-term markets will not react to vast, systemic financial events. This is an unrealistic basis for evaluating these (or any other) regulatory reforms. • The Report’s analysis assumes that money market funds exercise sufficient market power to compel fund investors and issuers of short-term debt to bear the costs and burdens of the Council’s proposed reforms. Given the wide range of alternatives available to cash investors, this assumption, too, is unrealistic. • The Report thus fails to take seriously the likelihood that its proposals would merely drive investors into less-regulated, less-transparent products and increase the risks and severity of a future financial crisis. The Report simply mentions this possibility in passing, observing laconically that FSOC and its members can address this possibility through further regulation “where appropriate and within their jurisdictions.”173 First, by basing its conclusion in the Report’s account of the 2007–2008 financial crisis,174 the Report does not address how the 2010 reforms already have made money market funds more resilient. As discussed in Section II.B.1., these reforms will make any future shifts by investors between prime and Treasury or government funds more fluid and less disruptive to the markets. Money market funds now must observe minimum liquidity standards to accommodate investor outflows more easily. Among the assets that meet these requirements are Treasury securities, certain agency securities, and repurchase agreements. In aggregate, Treasury and agency securities and repurchase agreements (which are typically backed by Treasury and agency securities) now make up more than one-third of prime money market 172 See Report, supra note 2, at 72. 173 Id. 174 This account itself is flawed and exaggerates the role of money market funds in spreading or accelerating the crisis, as we demonstrate in Section II.B.4. Financial Stability Oversight Council January 24, 2013 Page 87 funds’ assets, three times the level that these funds held before the 2007–2008 financial crisis.175 To the extent that flows out of prime funds occur, funds can meet redemptions with these liquid securities, which have proven to be in high demand during periods of financial stress. The Report does not address how these salutary changes, which have already proven their value in the events of 2011, have changed the role prime money market funds might play in a future crisis, relative to the risks that FSOC attributes to prime funds during the 2007–2008 crisis. Second, the Report exaggerates the effect that its proposals could have on investor behavior in a future financial crisis. By their very nature, financial crises are systemic events, and the crisis of 2007– 2008 was no exception. As the Report acknowledges, and as discussed more fully in Section II.B.4., Appendix B, and the SEC Staff Study, financial markets were already under extreme stress prior to Lehman’s default. The Report fails to acknowledge, however, that the default of Lehman itself was a highly destabilizing event, particularly given the already fragile state of the global financial system and the sudden reversal of the government’s prior course of rescuing “too big to fail” institutions. In this financial maelstrom, investors in prime funds redeemed shares and moved into other financial products, including Treasury and government money market funds. Given the broad systemic forces in play, these actions by money market fund investors—mirroring the actions of many other participants all across the short-term markets—were rational and predictable responses. Indeed, as the SEC Staff Study notes, there were “many possible reasons for the shift from prime money market funds toward Treasury funds” during that month.176 The Council’s efforts appear to be aimed at achieving a money market fund product that will ensure that investors and portfolio managers never react to bad news. This is evident in the Report’s suggestion that outflows from money market funds during the twin debt crises in the summer of 2011 demonstrate the need for further reforms. In fact, this modulated investor reaction, which occurred over several months, should stand as an example of the market discipline that policymakers should encourage. Instead, FSOC proposes to mute investors’ and managers’ response to systemic events. Its Alternative Two seeks to impede the ability of money market fund investors to react to bad news by imposing a 30-day MBR. Its Alternatives One and Three suggest that investors should become wholly insensitive to such information, either because floating money market funds’ NAVs would inure them to further changes in value, or because these funds would carry a capital buffer so high that it could be assumed to cover most losses a fund may incur. The Council offers no evidence that its proposals would reduce systemic risk. Indeed, as discussed above, some of the Council’s proposals would accelerate investors’ flight to quality and safety when the financial system shows even faint signs of distress. 175 See ICI Research Perspective, supra note 37. 176 See SEC Staff Study, supra note 10, at 7–8. Financial Stability Oversight Council January 24, 2013 Page 88 Third, FSOC assumes that money market funds have sufficient market power to compel fund investors (or, in some cases, issuers of short-term debt and intermediaries) to bear the costs and burdens of the Report’s proposals. The floating NAV would eliminate the stability, cash equivalency, and tax, accounting, and operational convenience that investors currently enjoy. The MBR would impose an above-market price for liquidity on money market fund investors under all market conditions, no matter how placid. Capital buffers would significantly reduce investor returns. Given the numerous alternative products and services available to investors, particularly institutional investors, it is wholly unrealistic to assume that investors would simply accept these costs and burdens and continue to use money market funds as they do today. Money market funds’ share of the short-term cash market already fluctuates.177 To the extent that investors have alternative means of investing in the money markets that allow them to avoid the costs and burdens of these regulations, money market fund investors will shift to less costly forms of intermediation, and borrowers will follow the money to seek their financing through these alternatives. Many of these products are less regulated and less transparent than Rule 2a-7 money market funds. Thus, the Report ignores the very high probability that its proposed fundamental changes will increase systemic risk by driving investors from money market funds into less-regulated, less-transparent cash management products. FSOC indirectly acknowledges the risk of investor flight to alternative products when it suggests, in the last sentence of the Report, that its member regulators will “use their authorities where appropriate and within their jurisdictions, to reduce or eliminate regulatory gaps to address any risks to financial stability that may arise from dissimilar standards for other cash-management products.”178 The Report does not consider, however, where such gaps currently exist, nor does it provide any holistic approach for addressing market risks that clearly extend beyond money market funds, and even beyond U.S. borders. Without a comprehensive approach, reforms targeted to one product out of many cannot address the broader systemic risks that clearly manifested themselves during the last financial crisis in less-regulated or unregulated products or in the short-term markets generally. Whatever “benefits” the Council might expect from its proposed recommendations cannot be realized if the assets in money market funds are driven into products beyond the reach of U.S. regulations. For these reasons, FSOC failed to demonstrate that its proposed recommendations would change overall investor behavior in the short-term markets and thereby reduce the probability or severity of a future financial crisis. The sweeping benefit the Council claims from its proposed recommendations is based on assertions, not evidence. FSOC also failed to consider the likelihood that its ideas could increase systemic risk and therefore produce negative benefits. 177 As demonstrated for nonfinancial businesses in Figure 1 above. 178 See Report, supra note 2, at 72. Financial Stability Oversight Council January 24, 2013 Page 89 B. FSOC’S COST ANALYSIS IS NARROW AND HIGHLY SPECULATIVE Under Section 120 of the Dodd-Frank Act, the Council is required to “take costs to long-term economic growth into account” when recommending new or heightened standards and safeguards for a financial activity or practice. The Report’s analysis of the costs to long-term growth of its recommendations, however, is highly speculative, perfunctory, and based on assumptions that are inconsistent with the Council’s assumed benefits, indicating that FSOC has failed to take seriously its obligations under the Dodd-Frank Act. FSOC’s Cost Analysis Is Narrow. The Council’s cost analysis is narrow in a fundamental aspect: it focuses only on the costs to long-term growth of requiring funds to hold a capital buffer. It does not even attempt to analyze the costs to long-term economic growth of requiring funds to adopt floating NAVs or MBRs. Given that floating NAVs and MBRs constitute key elements of FSOC’s proposed reforms, the lack of cost analysis on these aspects is puzzling. FSOC’s analysis is narrow in another fundamental aspect: how it defines “cost.” The Council defines “cost” as the “reduced level of spending that may accompany higher costs of financing investments and other outlays.”179 According to its cost analysis, the Council assumes that issuers of short-term securities that obtain financing from money market funds would, under a capital buffer requirement, incur higher financing costs because funds would pass along the costs of the capital buffer to issuers. This narrow definition, however, is inconsistent with the Council’s approach to determining the benefits of its proposed reforms. There, the Council suggests that the economy would benefit considerably from its proposed changes because these measures would reduce the probability and severity of future financial crises. As we have noted throughout this letter, however, the Council’s proposed changes might cause a shift of assets from money market funds to less-regulated products, leading to an increase in systemic risk as these assets move outside the reach of Rule 2a-7 and its risk- limiting provisions. This could offset any benefits the Council presumes might arise from its proposed fundamental changes to money market funds. The Council does not even acknowledge this potential effect, let alone attempt to address this issue in its cost analysis. FSOC’s Cost Analysis Is Highly Speculative. The Council seeks to analyze the costs of its proposed money market fund reforms using a methodology that produces estimates that are highly speculative. The Council seeks to estimate the cost of its proposals to long-term economic growth by assessing the reduction in U.S. gross domestic product (“GDP”) that could result from higher costs that issuers might incur to obtain financing. Using a particular methodology, FSOC estimates that a 3 percent capital buffer on money market funds would permanently reduce the level of GDP by 0.005 percent, which translates to about $682 million per year in real GDP. The Council states that this cost is “very small.”180 That assessment, 179 Id. at n.118. 180 See Report, supra note 2, at 70. Financial Stability Oversight Council January 24, 2013 Page 90 however, is questionable. In present value terms, this cost amounts to $23 billion (discounting a perpetual annual cost of $682 million per year at a 3 percent rate, which is on the high end of expectations for long-term growth in real GDP). Moreover, the Council’s “very small” estimate is derived from simulation models that arguably are unsuited to the task of assessing the effects of financial market regulation on long-term economic growth. FSOC relies on Bank for International Settlements (“BIS”) model simulations (provided by central banks around the world) that the BIS used to assess the economic effects of increased capital standards on banks.181 It is far from clear, however, whether the underlying simulation models are suited to measuring the long-term effects of a change in bank capital standards—let alone quite different changes in money market fund regulation. These models, termed “dynamic stochastic general equilibrium models” (“DSGE models”), rarely take seriously the role of financial intermediaries. In these models, borrowing constraints, amounts and types of collateral, leverage, maturity of assets and liabilities (i.e., maturity transformation), credit analysis, and the instruments used by firms for financing (e.g., money market instruments versus bonds versus equities) generally are missing or are treated as irrelevant to the economy. Indeed, DSGE models arose from earlier “real business cycle models” that explicitly assumed that the financial sector has no long-term effect on the real economy.182 Given that background, it is hardly surprising that the Council finds that its proposed money market fund recommendations have little effect on real economic growth. Curiously, this minimal estimate of the cost to long-term economic growth appears to contradict the Council’s own comments that money market funds “provide an economically significant service by acting as intermediaries between investors who desire low-risk, liquid investments and borrowers that issue short-term funding instruments.”183 Moreover, even if one is willing to accept the use of DSGE models for assessing the long-term growth effects of financial market regulation, FSOC’s estimates of the long-term cost to economic growth of its proposals are highly uncertain. The Council acknowledges this, but seeks to dismiss the uncertainty as unimportant because, in its view “the overall effects remain modest across the range of assumptions.”184 That is incorrect. The range of estimates provided in the BIS study that FSOC relies on is wide, implying much larger cost estimates than indicated by the Council. Indeed, the degree of 181 The Council relies on estimates reported in Bank for International Settlements, Interim Report: Assessing the Macroeconomic Impact of the Transition to Stronger Capital and Liquidity Requirements, Macroeconomic Assessment Group (August 2010) (“BIS Interim Report”). 182 See, e.g., Andrew Haldane, Executive Director, Financial Stability, Bank of England, What Have the Economists Ever Done for Us? (October 1, 2012), available at http://www.voxeu.org/article/what-have-economists-ever-done-us (“Because [] [DSGE] models were built on real-business-cycle foundations, financial factors (asset prices, money and credit) played distinctly second fiddle, if they played a role at all.”) See also Camilo E. Tovar, DSGE Models and Central Banks, BIS Working Papers, 258 (September 2008), available at http://www.bis.org/publ/work258.pdf, at 5 (“Possibly the main weaknesses in current DSGEs is the absence of an appropriate way of modeling financial markets.”). 183 See Report, supra note 2, at 8. 184 Id. at n.124. Financial Stability Oversight Council January 24, 2013 Page 91 uncertainty is so large as to put the Council’s single “very small” estimate of 0.005 percent of GDP in the realm of the highly speculative. FSOC easily could have provided some sense of the uncertainty of its cost estimates, but chose not to. The same BIS report that the Council relies on provides alternative estimates of the GDP losses that could arise from imposing higher bank capital standards in the United States.185 These alternative estimates are derived from DSGE and other models that at least make crude initial attempts to integrate the characteristics of the financial sector and also are based on a model of the United States only. The expected losses to GDP from these models are about seven times higher than the single estimate used by the Council.186 A study by the Institute for International Finance (“IIF”), which independently sought to confirm the BIS results, found estimated GDP losses from imposing higher bank capital standards to be eight times higher than the BIS estimate used by FSOC.187 These alternative estimates illustrate the highly speculative nature of the Council’s estimate of the cost of its proposed money market fund reforms to long-term economic growth. For example, using the alternative BIS estimate (seven times higher than the base case), the Council’s approach would permanently reduce U.S. output (i.e., GDP) by $4.8 billion per year. In present value terms (again, discounting in perpetuity at a 3 percent growth rate), these costs amount to $160 billion. This figure also is speculative, given that it too is based on a model that does not, at its core, take seriously the interaction between financial intermediaries and the real economy. But it illustrates that the range of uncertainty in the Council’s cost estimates is meaningfully large. The absence of these alternative estimates from the Report also suggests that FSOC was attempting to offer the lowest possible estimate of the cost, while ignoring the large uncertainties around its estimate. It is clear from this approach that the Council gave only the most perfunctory nod to its legal obligation under the Dodd-Frank Act to assess the cost of its proposals on long-term economic growth. In light of the untested nature of the proposed recommendations and the highly speculative nature of any benefits and costs, this experiment with a core sector of our financial system is unwarranted. 185 The Council’s 0.005 percent estimate of the long-term reduction in the level of GDP is based on the median “standard approach” BIS model, which is a median estimate of the effect of higher bank capital standards estimated from models used by the central banks of 17 countries, including the United States, China, South Korea, India, and Russia. The relevance for the United States of long-term cost estimates derived for these other countries is highly questionable. 186 For example, the Council relies on a “standard approach” median estimated GDP loss that is -0.12 percent of GDP at a horizon of 18 quarters. See BIS Interim Report, supra note 181, at 19. Alternative estimates for the United States, which are based on the Federal Reserve Board’s FRB/US model, are -0.79 percent or -0.89 percent depending on the precise model specification; 6.6 and 7.4 times higher than the GDP cost estimates used by the Council. Id. at 25–26. 187 The BIS Interim Report indicates that its “median estimate of the GDP impact is roughly one eighth the size of the estimate computed recently by the Institute of International Finance.” Id. at 4. Regarding its approach to estimating the cost to long-term growth of bank capital standards, the IIF has the advantage of being “rooted in data [that] … takes the current reality as the starting point [rather than being] a theoretical analysis of the long-run steady state.” See Institute of International Finance, Interim Report on the Cumulative Impact on the Global Economy of Proposed Changes in the Banking Regulatory Framework (June 2010), available at http://www.ebf-fbe.eu/uploads/10-Interim%20NCI_June2010_Web.pdf, at 27. In other words, the IIF’s estimate has the advantage that it is not based on a DSGE model. Financial Stability Oversight Council January 24, 2013 Page 92 VIII. Conclusion ICI and its members appreciate the opportunity to comment on FSOC’s proposed recommendations to the SEC. We remain firmly committed to working with policymakers to further strengthen money market funds’ resilience to severe market stress. * * * * If you have any questions regarding our comments, please feel free to contact me at (202) 326- 5901, Karrie McMillan, General Counsel, at (202) 326-5815, or Brian Reid, Chief Economist, at (202) 326-5917. Sincerely, /s/ Paul Schott Stevens Paul Schott Stevens President and CEO Investment Company Institute cc: The Honorable Elisse B. Walter The Honorable Luis A. Aguilar The Honorable Troy A. Paredes The Honorable Daniel M. Gallagher Norm Champ Director Division of Investment Management Securities and Exchange Commission A p p e n d i x A Money Market Funds: Select ICI Materials, 2009–2013 RESEARCH • Money Market Mutual Funds, Risk, and Financial Stability in the Wake of the 2010 Reforms January 15, 2013 Available at http://www.ici.org/pdf/per19-01.pdf • Operational Impacts of Proposed Redemption Restrictions on Money Market Funds June 20, 2012 Available at http://www.ici.org/pdf/ppr_12_operational_mmf.pdf • The Implications of Capital Buffer Proposals for Money Market Funds May 16, 2012 Available at http://www.ici.org/pdf/ppr_12_mmfs_capital_buffer.pdf • Money Market Fund Regulations: The Voice of the Treasurer April 19, 2012 Available at http://www.ici.org/pdf/rpt_12_tsi_voice_treasurer.pdf • Pricing of U.S. Money Market Funds January 25, 2011 Available at http://www.ici.org/pdf/ppr_11_mmf_pricing.pdf • Report of the Money Market Working Group March 17, 2009 Available at http://www.ici.org/pdf/ppr_09_mmwg.pdf TESTIMONY • Perspectives on Money Market Mutual Fund Reforms (Senate Banking Committee) June 21, 2012 Available at http://www.ici.org/pdf/12_senate_pss_mmf_written.pdf • Oversight of the Mutual Fund Industry: Ensuring Market Stability and Investor Confidence (House Financial Services Committee, Subcommittee on Capital Markets) June 24, 2011 Available at http://www.ici.org/pdf/11_house_mf_oversight_tmny.pdf • SEC Roundtable on Money Market Funds and Systemic Risk (ICI Chief Economist Brian Reid, panelist) May 10, 2011 Full transcript available at http://www.sec.gov/spotlight/mmf-risk/mmf-risk-transcript- 051011.htm Financial Stability Oversight Council January 24, 2013 Page 94 SPEECHES • Clouds Overhead: Financial Regulation After the Crisis March 19, 2012 Available at http://www.ici.org/policy/current_issues/12_km_mfim_conf • Preserving the Value of Money Market Funds for Investors and the Economy March 8, 2012 Available at http://www.ici.org/pressroom/speeches/12_pss_mmx • Do Money Markets Pose Systemic Risk? February 16, 2012 Available at http://www.ici.org/pressroom/speeches/12_pss_blmbg_mmf • The Seventh-Inning Stretch: The State of Play for Money Market Funds July 26, 2010 Available at http://www.ici.org/policy/current_issues/10_crane_pss_keynote • Weathering the Worst: Making Money Market Funds Even Stronger March 15, 2010 Available at http://www.ici.org/policy/current_issues/10_mfim_conf_pss_spch • One Year Later: Some Lessons from the Financial Crisis September 24, 2009 Available at http://www.ici.org/policy/trading/domestic/09_capmktsconf_stevens_spch • “These Great Masses of Money”: Making Money Market Funds Even More Resilient May 21, 2009 Available at http://www.ici.org/mmfs/resources/speeches/09_immfa_stevens_spch COMMENT LETTERS • ICI Letter on FSB Consultation on Securities Lending and Repo Markets January 14, 2013 Available at http://www.ici.org/pdf/26871.pdf • ICI Letter Regarding Banking Organizations’ Regulatory Capital Proposals October 22, 2012 Available at http://www.ici.org/pdf/26597.pdf • ICI and the Retirement Plan Community Letter to the SEC Regarding Proposed Money Market Fund Reforms August 21, 2012 Available at http://www.ici.org/pdf/12_joint_mmf_ltr_retirement.pdf Financial Stability Oversight Council January 24, 2013 Page 95 • ICI Letter to Treasury Secretary Geithner on the 2012 Annual Report of the Financial Stability Oversight Council July 25, 2012 Available at http://www.ici.org/pdf/12_geithner_pss_ltr.pdf • ICI Letter to the OCC Regarding Proposal to Revise Requirements for Short-Term Investment Funds June 8, 2012 Available at http://www.ici.org/pdf/26226.pdf • ICI Letter on FSB Report on Securities Lending and Repos May 29, 2012 Available at http://www.ici.org/pdf/26196.pdf • ICI Letter on IOSCO Money Market Fund Systemic Risk Analysis and Reform Options May 25, 2012 Available at http://www.ici.org/pdf/12_iosco_mmf_com_ltr.pdf • ICI Letter on Proposed Enhanced Prudential Standards for “SIFIs” and Large Bank Holding Companies April 30, 2012 Available at http://www.ici.org/pdf/26118.pdf • ICI, ICI Global, EFAMA, and IMMFA Joint Letter to IOSCO Regarding Money Market Fund Reform February 16, 2012 Available at http://www.ici.org/pdf/25936.pdf • ICI Letter on FASB Consolidation Proposal February 15, 2012 Available at http://www.ici.org/pdf/25914.pdf • ICI Submits Information to IOSCO on Money Market Fund Reform February 7, 2012 Available at http://www.ici.org/pdf/25877.pdf • ICI Letter on FSOC Proposal Regarding “SIFI” Designations December 19, 2011 Available at http://www.ici.org/pdf/25729.pdf • ICI Letter Concerning Nationally Recognized Statistical Rating Organizations August 8, 2011 Available at http://www.ici.org/pdf/25403.pdf Financial Stability Oversight Council January 24, 2013 Page 96 • ICI Letter on the Financial Stability Board’s Note on Shadow Banking June 3, 2011 Available at http://www.ici.org/pdf/25258.pdf • ICI Letter on FASB and IASB Draft Proposal That Would Eliminate “Cash Equivalents” from Financial Reporting May 31, 2011 Available at http://www.ici.org/pdf/25231.pdf • ICI Letter on Federal Reserve Board Proposal to Repeal Regulation Q May 13, 2011 Available at http://www.ici.org/pdf/25193.pdf • ICI Letter on SEC’s Proposal to Remove NRSRO References from Rules 2a-7 and 5b-3 April 25, 2011 Available at http://www.ici.org/pdf/25144.pdf • ICI Letter on President’s Working Group Report on Money Market Fund Reform January 10, 2011 Available at http://www.ici.org/pdf/11_sec_pwg_com.pdf • ICI Letter on CFTC Proposal Restricting Investments of Customer Funds in Money Market Funds December 6, 2010 Available at http://www.ici.org/pdf/24761.pdf • ICI Letter on FASB Accounting for Financial Instruments Proposal September 24, 2010 Available at http://www.ici.org/pdf/24561.pdf • ICI Statement for SEC Investor Advisory Committee Regarding Money Market Funds May 11, 2010 Available at http://www.ici.org/pdf/24289.pdf • ICI Letter on Basel Committee’s Proposed Standards Regarding a Global Framework for Banks’ Liquidity Risk Management April 16, 2010 Available at http://www.ici.org/pdf/24245.pdf • ICI Letter on FASB Proposal to Defer Adviser Consolidation of Funds Advised January 6, 2010 Available at http://www.ici.org/pdf/24062.pdf Financial Stability Oversight Council January 24, 2013 Page 97 • ICI Letter on CESR’s Consultation on a Common Definition for European Money Market Funds December 23, 2009 Available at http://www.ici.org/pdf/24043.pdf • ICI Letter on SEC’s Proposed Amendments to Money Market Fund Regulation September 8, 2009 Available at http://www.ici.org/pdf/23769.pdf • ICI Letter on CFTC’s Consideration of Changes to Permitted Investments July 21, 2009 Available at http://www.ici.org/pdf/23643.pdf ICI VIEWPOINTS Since 2011, ICI has published 75 posts regarding money market funds at its ICI Viewpoints blog. These posts are available at http://www.ici.org/viewpoints/mmfs. ICI’s Money Market Fund Resource Center Key comment letters, research, press releases, fact sheets, and other material can be found at ICI’s online resource center devoted to money market funds, available at http://www.ici.org/mmfs. A p p e n d i x B Timeline of Major Developments in the Financial Crisis1 I. Market Events Leading Up to September 2008 JUNE 2007 • Two Bear Stearns Companies Inc. hedge funds suspended redemptions in the face of deteriorating investments in securities backed by subprime mortgages. SUMMER AND FALL OF 2007 • A number of additional short-term investment pools (e.g., unregistered “enhanced cash” funds, liquidity pools run by municipalities, and offshore funds) began to fail after investing in securities backed by subprime mortgages. • BNP Paribas, France’s largest bank, temporarily froze redemptions of three investment funds that operated in a manner similar to European variable net asset value (“NAV”) money funds. • An unregistered commodity cash pool managed by Sentinel Management Group, Inc., erroneously described by CNBC as a money market fund, halted redemptions and failed within a week. • Local government investment pools run by King County, Washington, and the State of Florida experienced difficulties due to structured investment vehicle and asset-backed commercial paper (“ABCP”) investments. King County intervened to buy the troubled securities, and the Florida pool experienced a cascade of redemptions, until it froze withdrawals in November. AUGUST 2007 TO MARCH 2008 • A number of major financial institutions in the U.S. and Europe failed, including American Home Mortgage Corp.; HomeBanc Corp.; Sachsen Landesbank; Northern Rock, plc; Financial Guaranty Insurance Company; and Countrywide. Others, such as Citigroup, Inc. and the monoline insurers Ambac Financial Group, Inc. and MBIA, Inc. needed significant help (both government and private) to survive. • The auction rate securities market froze as securities for sale exceeded demand, auction agents refused to take the excess supply on their balance sheets, and the auctions failed en masse. 1 Much of the information in this appendix was drawn from the Investment Company Institute, Report of the Money Market Working Group (March 17, 2009) (“MMWG Report”), available at http://www.ici.org/pdf/ppr_09_mmwg.pdf and the U.S. Financial Crisis Inquiry Commission, The Financial Crisis Inquiry Report (January 2011), available at http://www.gpo.gov/fdsys/pkg/GPO-FCIC/pdf/GPO-FCIC.pdf. Financial Stability Oversight Council January 24, 2013 Page 99 • During this time, the money market2 continued to exhibit considerable stress. For example, spreads between yields on one-month ABCP and Treasury bills widened dramatically, reaching nearly 400 basis points at one time. • Outstanding commercial paper declined a little over $300 billion, a 14 percent drop, over August and September 2007. As money market funds reduced their commercial paper holdings by only $15 billion, other investors accounted for the bulk of the decline. WEEKEND OF MARCH 15 AND 16, 2008 • The federal government orchestrated a rescue of Bear Stearns, allowing JPMorgan Chase & Co. to purchase Bear Stearns, with the federal government guaranteeing up to $30 billion in potential losses. Under this transaction, Bear Stearns’ shareholders suffered very significant losses but its debt holders were unharmed. As of May 31, 2007, Bear Stearns’ assets were 31 times its shareholder equity. APRIL 2008 • Wachovia Corporation amassed a first-quarter loss of $350 million. JULY 14, 2008 • The Office of Thrift Supervision closed IndyMac Bank, making it the largest-ever thrift to fail. JULY 22, 2008 • Washington Mutual reported a $3.3 billion loss. Depositors withdrew $10 billion during the next two weeks. • Wachovia amassed an $8.9 billion second-quarter loss. II. Key Market Events—September 2008 WEEKEND OF SEPTEMBER 6 AND 7 • The government placed the nation’s two largest mortgage finance companies, Fannie Mae and Freddie Mac, in conservatorship, with a plan to provide financial support to the agencies through the purchase of senior preferred stock and the extension of short-term secured loans. 2 In the United States, the market for debt securities with a maturity of one year or less is generally referred to as “the money market.” For an overview of the money market, including its structure and participants and the key characteristics of money market funds, see Appendix A of the MMWG Report, supra note 2. Financial Stability Oversight Council January 24, 2013 Page 100 WEEK OF SEPTEMBER 8 • Long-circulated rumors about the financial stability of Merrill Lynch & Co., Inc., American International Group, Inc. (AIG), and Lehman Brothers Holdings Inc. gained traction. WEEKEND OF SEPTEMBER 13 AND 14 • Bank of America Corporation agreed to buy Merrill Lynch for $50 billion. • The future of AIG, one of the largest underwriters of credit default swaps, remained highly uncertain, as credit rating agencies threatened to downgrade the company’s debt, a move that would have prompted counterparties to make margin calls on their contracts that would be in excess of AIG’s available liquidity. • The U.S. Treasury Department and the Federal Reserve tasked CEOs of major Wall Street firms to come up with a private sector solution to prevent a Lehman bankruptcy. MONDAY, SEPTEMBER 15 • Lehman, lacking a buyer and failing to obtain government assistance, declared bankruptcy. • As with Bear Stearns, the viability of Lehman had been questioned for several months. Nevertheless, Lehman’s failure was an especially difficult shock for the market because it represented an abrupt reverse in direction by the U.S. government from its previous decisions to intervene and rescue Bear Stearns (an investment bank smaller than Lehman), Fannie Mae, and Freddie Mac. Lehman was a large dealer of fixed-income securities, including commercial paper. • The collapse of Lehman on September 15 triggered a severe credit freeze in the short-term markets, as investors pulled back from lending to financial institutions and rushed to buy short-dated Treasury securities. • Yields at the short-end of the Treasury market traded down sharply, with four-week bills trading at 0.28 percent, down from 1.35 percent on September 12 and 1.51 percent on September 11. • At the same time, investors retrenched from the commercial paper market. Issuance at the longer end of the market fell sharply. Issuers had difficulty attracting investors to paper with maturities beyond the end of the week. Issuance volume on commercial paper with maturities beyond four days dropped to $23 billion on September 15 from $51 billion on September 12. • In the afternoon, AIG was downgraded by S&P, Moody’s, and Fitch Ratings, triggering billions of dollars in additional cash collateral calls on AIG’s credit default swaps. Financial Stability Oversight Council January 24, 2013 Page 101 • On September 15, 2008, prime money market funds had outflows of $63 billion, of which a large fraction likely represented normal outflows associated with tax payments. In the previous four years, outflows from prime money market funds averaged $20 billion on September tax payment days. After accounting for estimated outflows related to tax payments and outflows from the Reserve Primary Fund, non-tax-related outflows for all prime money market funds totaled approximately $31 billion or 1.5 percent of total net assets.3 Government money market funds had inflows of $2 billion on September 15. TUESDAY, SEPTEMBER 16 • The Treasury bill market continued to be swamped by heavy demand as investors sought the safety of short-term U.S. government securities. The four-week bill traded at 0.23 percent and the three-month bill traded at 0.84 percent. Stresses in the commercial paper market increased as issuers continued to have difficulty attracting investors beyond the very short end of the market. Issuance beyond four days dropped to $20 billion. • Outflows from prime money market funds began to pick up as some investors in these funds, like other investors, began to seek the safety of government securities. Outflows from prime funds totaled $32 billion, while inflows to government money market funds were $33 billion. • After the markets closed, Reserve Primary Fund announced that it would no longer redeem shares at $1.00. The fund held about 1.2 percent of its assets in Lehman debt. • BNY Institutional Cash Reserves’ (a securities lending collateral pool that was not a money market fund) share price fell to $0.991 from $1.00. Lehman debt represented 1.13 percent of its portfolio. • Late in the evening after the markets were closed, the Federal Reserve announced that it had agreed to lend AIG up to $85 billion. The U.S. government took a nearly 80 percent stake in the company. WEDNESDAY, SEPTEMBER 17 • Other money market funds with exposure to Lehman also experienced difficulties. Nevertheless, all money market funds, with the exception of the Reserve Primary Fund, maintained their $1.00 NAV. • Investors continued to flee to the Treasury bill market for safety. Four-week bills traded at 0.07 percent and three-month bills were at 0.03 percent. Meanwhile, the credit squeeze in the commercial paper market continued: issuance beyond four days fell to $18 billion, with 40 3 Data for September 15 includes estimated redemptions of $11.6 billion processed by the Reserve Primary Fund on September 15. As of September 12, the Reserve Primary Fund had $62.6 billion in total net assets. As of the close of business on September 15, the Reserve Primary Fund had approximately $51 billion in total net assets. The fund was effectively frozen at this level until it starting making distributions to shareholders beginning October 30. See http://www.primary-yieldplus- inliquidation.com/pdf/PressReleasePrimDist2008_1030.pdf. Daily data for all other money market funds are from iMoneyNet. Financial Stability Oversight Council January 24, 2013 Page 102 percent of that issuance between five and nine days. Outstanding commercial paper was down $51 billion from a week earlier, or about 3 percent. • Colorado Diversified Trust, a local government investment pool (“LGIP”) (not a money market fund) transferred its assets to another LGIP to maintain its rating (the pool held 1.8 percent of its portfolio in Lehman paper). The trust served as a cash pool for more than 60 local government entities in Colorado. • Inflows to government money market funds rose to $49 billion and prime money market fund investors redeemed, on net, $106 billion. THURSDAY, SEPTEMBER 18 • Short-term markets continued to trade with difficulty due to investors’ preferences for government securities. Demand for Treasury bills kept yields well below their prior week levels, with the four-week bill yield at 0.25 percent and three-month bills yielded 0.23 percent. • Commercial paper issuance beyond four days remained depressed at $24 billion. Investors’ deep concerns about the viability of banks and other financial institutions around the world, and about the willingness and wherewithal of their governments to support them, constricted the access of these firms to funding in the short-term markets. For example, financial firms were only able to place 11 issues of commercial paper with maturities beyond 40 days, compared with 149 issues on September 12. • For a third day, money market fund investors mirrored behavior in the broader markets, as investors sought the security of government securities. Inflows to government money market funds totaled $58 billion, and outflows from prime funds were $94 billion. • Putnam Investments announced in the morning that it was closing the Putnam Prime Money Market Fund. The fund had no exposure to Lehman or other troubled issuers, but had experienced significant redemption pressures from its concentrated institutional investor base. The fund determined to close rather than sell portfolio securities into a liquidity- constrained market; this action allowed the fund to treat all shareholders fairly. On September 24, the fund merged with Federated Prime Obligations Fund at $1.00 per share; shareholders did not lose any principal. All pending redemptions were processed coincident with the merger within 7 days. FRIDAY, SEPTEMBER 19 The Federal Reserve and Treasury announced a series of broad initiatives designed to stabilize the market, which, as demonstrated above, had ceased to function even for very short-term, high-credit securities. • The Asset-Backed Commercial Paper Money Market Mutual Fund Liquidity Facility (AMLF) provided non-recourse loans at the primary credit rate to U.S. depository Financial Stability Oversight Council January 24, 2013 Page 103 institutions and bank holding companies to finance purchases of high-quality ABCP from money market funds. • Treasury announced its Temporary Guarantee Program for Money Market Funds, which temporarily guaranteed certain account balances in money market funds that qualified for and elected to participate in the program. ICI worked with Treasury and other regulators to limit the reach of the Treasury Guarantee Program, urging that the guarantee be limited and temporary. The program expired on September 18, 2009. No claims were made on the Guarantee program, and no amounts were paid out. Instead, Treasury and, as a result, taxpayers, received an estimated $1.2 billion in premiums paid by participating money market funds. • Pressures in the Treasury market eased somewhat after the announcement of these programs. The yield on the four-week bill rose to 0.75 percent, and three-month bill yields were at 0.99 percent. Commercial paper markets remained under pressure, however, with only $25 billion in new issuance beyond four days. • Money market fund flows returned to the level and pattern seen on September 16. Outflows from prime funds totaled $36 billion, and inflows to government money market funds were $47 billion. III. Events of Late September 2008 to October 2008 Although the steps taken by the Federal Reserve and Treasury helped to stabilize the commercial paper market and thereby moderate outflows from prime money market funds, further developments added to investor concerns about overall stability of the global financial markets. These events unfolded through September and into October. SEPTEMBER 21 • The Federal Reserve Board approved the applications of The Goldman Sachs Group, Inc. and Morgan Stanley to become bank holding companies. SEPTEMBER 25 • After nearly two weeks of speculation about the future of Washington Mutual, Inc., the FDIC officially placed it in receivership. A credit downgrade on September 15 sparked a run and caused investors to pull $16.7 billion in assets, or 9 percent of its June 2008 deposits, from the bank. 4 The FDIC subsequently sold the savings bank to JPMorgan. 4 See Office of Thrift Supervision Fact Sheet on Washington Mutual Bank (September 25, 2008), available at http://www.files.ots.treas.gov/730021.pdf. Financial Stability Oversight Council January 24, 2013 Page 104 SEPTEMBER 28 • The governments of the Netherlands, Belgium, and Luxembourg rescued Fortis Bank. SEPTEMBER 29 • The British government rescued Bradford & Bingley plc, a mortgage lender. Iceland nationalized Glitnir Bank. SEPTEMBER 30 • The governments of Belgium, France, and Luxembourg rescued Dexia SA, a major European banking group. • Wachovia, trustee of the $40 billion Commonfund, a bank common trust fund sold to charities and colleges as a cash management vehicle (but not a Rule 2a-7 money market fund), announced that it commenced a liquidation of the fund in response to the credit markets’ reaction to “the failure of Lehman and Washington Mutual Bank, the nationalization of American International Group and the failures of Congress to pass legislation.”5 • Outstanding repurchase agreements fell $400 billion during the month of September 2008. Investors other than money market funds accounted for more than all of this decline. Money market funds increased their holdings of repurchase agreements by a little over $90 billion in September 2008. SEPTEMBER 22 THROUGH SEPTEMBER 30 • Money market fund investors continued to shift their holdings from prime funds to government money market funds. Outflows from prime funds totaled $103 billion and inflows to government funds were $146 billion. OCTOBER 2 • The President of Ireland signed legislation guaranteeing Irish banks. OCTOBER 3 • Congress passed and President George W. Bush signed the Emergency Economic Stabilization Act of 2008, which included the Troubled Asset Relief Program (“TARP”). TARP allowed Treasury to purchase assets and equity from banks. The FDIC approved Wells Fargo’s offer to buy Wachovia, reversing an earlier offer by Citigroup to purchase the banking firm. 5 Important Announcement Regarding Commonfund Short Term Fund (September 30, 2008), available at http://www.commonfund.org/Templates/Generic/RESOURCE_REQUEST/target.pdf?RES_GUID=E770A035-E068-49B4- 8E60-612CE9E2DCB5. Financial Stability Oversight Council January 24, 2013 Page 105 OCTOBER 7 • The Federal Reserve authorized the Commercial Paper Funding Facility (CPFF). The program provided a backstop to U.S. issuers of commercial paper through a special purpose vehicle that would purchase three-month unsecured commercial paper and ABCP directly from eligible issuers. • Icelandic bank Landsbanki was placed into receivership. OCTOBER 8 • Icelandic bank Kaupthing was nationalized. OCTOBER 13 • Treasury invested $125 billion from TARP in preferred shares of nine large commercial banks. The Federal Reserve, the Bank of England, the European Central Bank, the Bank of Japan, and the Swiss National Bank announced a coordinated program “to provide broad access to liquidity and funding to financial institutions.” • The British government rescued the RBS Group by taking an ownership stake in the financial firm. WEEK OF OCTOBER 7 TO OCTOBER 14 • Money market funds became net buyers of commercial paper, acquiring $17 billion. OCTOBER 27 • The Federal Reserve launched the CPFF. A p p e n d i x C Regression Analysis of Government and Tax-Exempt Money Market Fund Flows The Report suggests that investors in money market funds have become more likely to react to market events than in the past and that government6 and tax-exempt money market funds are not immune to destabilizing redemptions. We empirically investigate through regression analysis whether investors in government or tax-exempt money market funds are sensitive to changes in financial conditions that reflect market anxiety or uncertainty and whether this sensitivity to uncertainty has changed since the 2007–2008 financial crisis. Our analysis finds that investors turn to government money market funds and tax-exempt funds when market uncertainty increases and the tendency for this behavior has not changed since the 2007– 2008 financial crisis. We test two measures of market uncertainty: (i) the percent change in the implied volatility of S&P 500 options as measured by the VIX and (ii) the change in the spread between the yields on Moody’s Baa corporate index and the 10-year constant maturity U.S. Treasury security. A summary of the regression results is provided below. • Flows of institutional government money market funds7 are positively related to the change in the VIX and the Baa spread. The strength of this relationship is about the same between the post-crisis and pre-crisis periods for the change in the VIX and somewhat weaker in the post- crisis period for the change in the Baa spread. • Flows of retail government money market funds8 are positively related to the change in the Baa spread. The strength of this relationship is about the same between the post-crisis and pre-crisis periods. The percent change in the VIX has little to no impact on retail government fund flows in either period. • Flows of tax-exempt funds9 exhibit the same relationship to the market uncertainty variable as flows of retail government funds. They are positively related to the change in the Baa spread and the strength of this relationship is about the same between the post-crisis and pre-crisis periods. The percent change in the VIX also has little to no impact on tax-exempt fund flows in either period. 6 Consistent with the Report, government money market funds are defined as those that invest primarily in U.S. Treasury obligations and U.S. government agency securities. This definition excludes Treasury money market funds—those funds that invest primarily in U.S. Treasury obligations and repos collateralized with U.S. Treasury obligations. 7 Share classes of institutional government money market funds are identified from iMoneyNet and matched to weekly asset data submitted to the Investment Company Institute. Weekly flow data are aggregated across share classes. 8 Share classes of retail government money market funds are identified from iMoneyNet and matched to weekly asset data submitted to the Investment Company Institute. Weekly flow data are aggregated across share classes. 9 Share classes of tax-exempt funds that submit weekly asset data to the Investment Company Institute. Weekly flow data are aggregated across share classes. Financial Stability Oversight Council January 24, 2013 Page 107 GOVERNMENT MONEY MARKET FUNDS Historically, investors have moved into government money market funds during turbulent times. Our results show that this tendency for government money market funds to receive net inflows during periods of financial stress has remained strong since the financial crisis of 2007–2008, countering the belief that these funds are now susceptible to runs. Institutional Funds In general, we find that weekly flows to institutional government money market funds are related to the previous week’s flows, the level of contemporaneous short-term yields, the previous week’s short- term yields, as well as to a contemporaneous measure of long-term yields. This relationship is represented by the regression equation shown below.  !!"#! =  !   + !!!!"#!!! +  !!!"#$%! +  !!!"#$%!!! +  !!!!! +  !!!!!!! +  !!!"#$! +  !"#!$%#&!+ !!   where, !!"#! =  percent change in aggregate weekly total net assets of institutional government money market fund share classes, !"#$%! =  7-day simple average net yield of institutional government/agency money market funds, !!! = federal funds rate, !"#$! =  yield on 10-year constant maturity U.S. Treasury security, and !"#!$%#&! = dummy variables for March 15, June 15, and September 15 weeks in which corporate tax payments are due. In order to gauge whether institutional investors might be expected to use government money market funds when market anxiety increases, we added measures of “uncertainty” to the model shown above. We tested two market uncertainty variables: (1) !"#! (the weekly percent change in the VIX index) and (2) !""#$%&"'! (the change in the spread between the yields on Moody’s Baa corporate index and the 10-year constant maturity U.S. Treasury security). Tables C.1 and C.2 below show the results of OLS regressions over the pre-crisis sample period (January 2000 to August 2007), the post- crisis sample period10 (October 2008 to December 2012), and the entire sample period (January 2000 to December 2012) for each of the uncertainty measures. If institutional investors use government money market funds when market uncertainty rises, we would expect to see a positive coefficient on the uncertainty variables. In addition, if institutional investors since the 2007–2008 financial crisis are less prone to use government money market funds 10 Post-crisis for money market funds is determined to be after the Temporary Guarantee program and the AMLF program were announced and put in place. Financial Stability Oversight Council January 24, 2013 Page 108 when market uncertainty rises, we would expect to see a smaller coefficient on the uncertainty variables in the post-crisis period than in the pre-crisis period. TABLE C.1 Coefficient Estimates for Government Institutional Money Market Fund Flows with VIX Variable Pre-Crisis (Jan. 2000–Aug. 2007) Post-Crisis (Oct. 2008–Dec. 2012) Entire Period (Jan. 2000–Dec. 2012) ! 1.546* (1.91) 1.259*** (2.60) 0.803** (2.53) !!"#!!! 0.069 (1.37) -0.044 (-0.68) 0.156*** (4.19) !"#$%! -3.835* (-1.70) -7.57 (-1.51) -3.151* (-1.89) !"#$%!!! 5.091*** (2.93) 9.804** (2.08) 4.558*** (3.08) !!! -1.276 (-1.41) -0.946 (-0.48) -2.267*** (-3.14) !!!!! 0.266 (0.33) 1.427 (0.78) 1.157* (1.68) !"#$! -0.370** (-2.20) -0.612*** (-3.52) -0.255** (-2.56) !"#! 0.016** (2.17) 0.020** (2.14) 0.024*** (4.16) R2 0.14 0.27 0.21 Notes: *, **, and *** denote significance at the 10 percent, 5 percent and 1 percent level respectively. T-statistics are shown in parentheses. Coefficient estimates on seasonal dummies are omitted for brevity. As shown above in Table C.1, the estimated coefficient on !"#! is positive and statistically significant in all three time periods. This result can be interpreted as the higher the increase in market uncertainty, the more money institutional investors tend to invest in government money market funds. For example, based on the post-crisis regression, a one standard deviation (11.6 percent) increase in the VIX is expected to lead to an additional 0.23 percent in weekly total net assets of government money market funds. In the post-crisis period, the estimated coefficient (0.020) is only marginally larger than the estimated coefficient (0.16) in the pre-crisis period, and the two coefficients are not significantly different from each other.11 This result implies that for the same change in market uncertainty 11 The results presented in this appendix are robust to the inclusion of controls for the level of the uncertainty variable. For example when we include the level of the VIX index in the regression shown in Table 1, the coefficient on the change in the VIX is roughly 0.016 in both the pre- and post-crisis periods. Financial Stability Oversight Council January 24, 2013 Page 109 institutional investors made approximately the same use of government money market funds in the post-crisis period that they did in the pre-crisis period.12 Table C.2 below reports the regression results when the weekly change in the Baa spread, !""#$%&"'!, is used as a proxy for market uncertainty. As can be seen, the regression results are roughly similar to those in Table C.1. The estimated coefficients on the change in the Baa spread are consistently positive across the sample periods. The coefficients in the pre-crisis and entire sample periods are statistically significant, but insignificant in the post-crisis period. While institutional investors of government money market funds appear to have become less responsive to the change in the Baa spread since the financial crisis, the estimated coefficients between the pre-crisis and post-crisis period are not significantly different from each other. TABLE C.2 Coefficient Estimates for Weekly Government Institutional Money Market Fund Flows with Change in Baa Spread Variable Pre-Crisis (Jan. 2000–Aug. 2007) Post-Crisis (Oct. 2008–Dec. 2012) Entire Period (Jan. 2000–Dec. 2012) ! 1.515* (1.87) 1.218** (2.50) 0.776** (2.44) !!"#!!! 0.075 (1.49) -0.062 (-0.95) 0.137*** (3.65) !"#$%! -4.373* (-1.93) -9.319* (-1.79) -3.600* (-2.16) !"#$%!!! 5.434*** (3.11) 11.396** (2.33) 4.826*** (3.25) !!! -1.197 (-1.32) -0.501 (-0.26) -2.218*** (-3.07) !!!!! 0.370 (0.46) 0.933 (0.50) 1.270* (1.85) !"#$! -0.366** (-2.18) -0.585*** (-3.27) -0.242** (-2.44) !""#$%&"'! 0.029** (2.43) 0.015 (1.25) 0.033*** (4.23) R2 0.14 0.26 0.21 Notes: *, **, and *** denote significance at the 10 percent, 5 percent and 1 percent level respectively. T-statistics are shown in parentheses. Coefficient estimates on seasonal dummies are omitted for brevity. 12 As an additional robustness check we ran the same regression during the crisis period (not reported). Flows to government money market funds showed high positive auto-correlation. Nevertheless, the effect of the VIX index remained positive and statistically significant and the estimated coefficient was insignificantly different from those estimated in the pre- or post-crisis period. Financial Stability Oversight Council January 24, 2013 Page 110 Retail Funds In general, we find that weekly flows to retail government money market funds are related to the previous week’s flows, and the spread between the yields on retail government money market funds and money market deposit accounts (MMDA). This relationship is represented by the regression equation shown below.  !!"#! =  !   + !!!!"#!!! +  !!!!"#! + !"#!$%!"# + !!   where, !!"#! =  percent change in aggregate weekly total net assets of retail government money market fund share classes, !!"#! = 7-day simple average net yield of retail government/agency money market funds less the 7-day MMDA rate, !"#!$%#&! = dummy variables for last two weeks of March, all weeks of April, and first two weeks of May to account for federal and state tax payments. As shown in Table C.3 below, flows of government money market funds appear to be positively and statistically significantly related to the change in the Baa spread.13 As with institutional investors, retail investors tend to invest more money in government money market funds when there is an increase in market uncertainty. Market uncertainty, in this case, is measured by the Baa credit spread. Based on the post-crisis regression, a one standard deviation (9 basis points) increase in the Baa spread is expected to lead to an additional 0.15 percent in weekly total net assets of government money market funds. The estimated coefficients in the pre-crisis and post-crisis periods are virtually the same, implying that for a given change in the Baa spread retail investors made approximately the same use of government money market funds in the post-crisis period that they did in the pre-crisis period. 13 The regression results for the percent change in the VIX are omitted because the estimated coefficients were close to zero and insignificant in all the three time periods. Financial Stability Oversight Council January 24, 2013 Page 111 TABLE C.3 Coefficient Estimates for Weekly Retail Government Money Market Fund Flows with Change in Baa Spread Variable Pre-Crisis (Jan. 2000–Aug. 2007) Post-Crisis (Oct. 2008–Dec. 2012) Entire Period (Jan. 2000–Dec. 2012) ! -0.060 (-0.871) -0.055 (-0.990) -0.026 (-0.489) !!"#!!! 0.188*** (3.749) 0.416*** (11.383) 0.372*** (10.261) !!"#! 0.089*** (2.946) -0.353 (-0.994) 0.070** (2.508) !""#$%&"'! 0.016** (2.259) 0.017*** (3.651) 0.021*** (4.350) R2 0.09 0.52 0.22 Notes: *, **, and *** denote significance at the 10 percent, 5 percent and 1 percent level respectively. T-statistics are shown in parentheses. Coefficient estimates on seasonal dummies are omitted for brevity. Tax-Exempt Money Market Funds In general, we find that weekly flows to tax-exempt government money market funds are related to the previous week’s flows, the MMDA spread, and the gross yield on tax-exempt money market funds. This relationship is represented by the regression equation shown below.  !!"#! =  !  + !!!!"#!!! +  !!!!"#!  +  !!!"#$%! + !"#!$%#&! + !! where, !!"#! =  percent change in aggregate weekly total net assets of tax-exempt government money market fund share classes, !!"#! = 7-day simple average net yield of tax-exempt market funds less 7-day MMDA rate, !"#$%! = 7-day simple average of the net yield of tax-exempt money market funds, !"#!$%#&! = dummy variables for last two weeks of March, all weeks of April, and first two weeks of May to account for federal and state tax payments. As shown in Table C.4 below, flows of tax-exempt money market funds appear to be positively and statistically significantly related to the change in the Baa spread.14 Similar to the government retail results, investors tend to invest more money in tax-exempt funds when there is an increase in the Baa 14 The regression results for the percent change in the VIX are omitted because the estimated coefficients were close to zero and insignificant in all the three time periods. Financial Stability Oversight Council January 24, 2013 Page 112 spread. Based on the post-crisis regression, a one standard deviation (9 basis points) increase in the Baa spread is expected to lead to an additional 0.08 percent in weekly total net assets of tax-exempt money market funds. The estimated coefficients in the pre-crisis and post-crisis periods are identical, implying that for a given change in the Baa spread investors made approximately the same use of tax-exempt money market funds in the post-crisis period that they did in the pre-crisis period. TABLE C.4 Coefficient Estimates for Weekly Tax-Exempt Money Market Fund Flows with Change in Baa Spread Variable Pre-Crisis (Jan. 2000–Aug. 2007) Post-Crisis (Oct. 2008–Dec. 2012) Entire Period (Jan. 2000–Dec. 2012) ! -0.041 (-0.511) 0.077 (1.537) -0.006 (-0.146) !!"#!!! 0.110** (2.270) 0.205*** (3.179) 0.296*** (8.135) !!"#! 0.057 (1.042) 1.238*** (5.029) 0.201*** (4.158) !"#$%! 0.018 (0.406) -0.453*** (-3.957) -0.045 (-1.504) !""#$%&"'! 0.009** (2.344) 0.009*** (3.084) 0.012*** (4.692) R2 0.07 0.55 0.24 Notes: *, **, and *** denote significance at the 10 percent, 5 percent and 1 percent level respectively. T-statistics are shown in parentheses. Coefficient estimates on seasonal dummies are omitted for brevity.