Prime money market mutual funds are particularly susceptible to runs given the inherently short-term nature of their liabilities and the riskiness of their assets as compared with government funds. However, as discussed earlier, runs on such funds are unlikely to pose systemic risk concerns due to their role as funders of other financial institutions given the relatively low reliance large financial institutions have on funding from money market funds. But there is a legitimate concern, based on our experience in the crisis, that a run on money market funds could spark contagion in the rest of the financial system apart from any connectedness. Thus, it is a proper object of policy to minimize the possibility of prime money market fund runs. Accordingly, both FSOC and the SEC have addressed the issue with FSOC proposing recommendations to the SEC regarding money market mutual fund reform in 2012 and the SEC itself proposing rules in 2013.1 The SEC issued its final rule on money market mutual funds in October 2014, with which funds will have two years to comply.2
As previously recounted, the Federal Reserve and the US Treasury instituted a number of programs during the financial crisis to stem the contagion in the MMF industry. The Federal Reserve extended indirect access to the discount window to money market funds through a $150 billion Asset-Backed Commercial Paper Money Market Mutual Fund Liquidity Facility (“AMLF”),3 creating the $350 billion Commercial Paper Funding Facility (“CPFF”)4 and the Money Market Investor Funding Facility (“MMIFF”).5 The US Treasury provided an effective $3.2 trillion temporary guarantee of the liabilities of the money market funds through its Exchange Stabilization Fund.6
Despite the success of these programs, Dodd–Frank substantially curtailed the ability of the government to use similar tactics in the future to address contagion in the MMF industry. Future programs to inject Federal Reserve liquidity into the money market funds will require prior approval by the Secretary of the Treasury and must be programs that are applied on a broad basis only.7 Furthermore Congress has explicitly prohibited the Treasury from using funds from the Economic Stabilization Fund to conduct a similar temporary guarantee program in the future. 8
When evaluating the effectiveness of various MMF regulations, it is important to keep in mind that while regulations may be effective in combatting MMF contagion, the unintended consequence may be to push MMF activities into alternative, less-regulated sectors of the financial system. The MMF industry may be safer, but the financial system as whole would be no less vulnerable to contagion. Further any new proposals to address the contagion risks in the MMF industry should not simply be veiled attempts to eliminate money market funds by fundamentally compromising their business model. While many believe money market funds are the product of interest rate regulation under Regulation Q—and they may be right—this arbitrage has long ceased to exist. These funds serve an important function today as indicated by the fact that they have $2.7 trillion in assets.9 Indeed they allow large depositors, including major firms, to hold funds with a stable asset value, that can be used to make payments, that provide a return (not the case with transaction accounts in banks) and most importantly are less risky than holding uninsured funds in banks—due to the fact that MMF assets are much more liquid and generally carry less risk than bank assets.
The SEC’s approach to MMF reform incorporates three elements: (1) enhanced liquidity requirements, (2) a floating NAV requirement for certain classes of money market funds, and (3) the possibility of imposing liquidity fees and redemption gates on money market funds, which would limit rapid MMF creditor outflows in times of stress. It specifically rejected imposing a capital requirement on these funds. At the outset it should be clear that the concern with contagion should only be with prime money market funds and municipal funds, and not with government funds, which are all but immune from runs.
In February 2010 the SEC amended rule 2a-7 to significantly increase the liquidity of money market funds. The SEC reduced the maximum permitted weighted average portfolio maturity of money market funds from 90 days to 60 days. Additionally money market funds have to invest at least 10 percent of their portfolios in “daily liquid assets” (cash, US government securities, and other securities that provide the holder the right to demand payment within one day) and 30 percent of their portfolio in weekly liquid assets (same as above, US government securities maturing in 60 days or less, and other securities maturing within five business days). The rule also prohibits money market funds from investing more than 5 percent of the fund’s assets in illiquid securities. An illiquid security is any security that cannot be sold by the fund within seven days at approximately the value ascribed to it by the fund.10 According to a recent study by James Angel, due to the 2010 amendments a MMF can withstand redemptions of 10 percent of its assets in a single day or 30 percent of its assets in a week without having to “sell a single asset into a fragile market.”11 The 2010 amendments also adopted rule 22(e)(3), which permits a MMF’s Board of Directors to suspend redemptions when liquidating a fund. In a 2012 report the SEC argues that money market funds are more resilient under these new rules as funds are less likely to break the buck under the 2010 liquidity rules than under the previous rules.12 However, the report also concludes that the Reserve Primary Fund would still have broken the buck under the new 2010 rules.13
The floating NAV reform under the SEC’s 2014 rule will apply to prime institutional money market funds, as well as institutional municipal money market funds—often called tax-free money market funds, which invest in debt instruments issued by municipalities and state authorities14—that do not qualify for an exemption. A prime money market fund is any fund that invests more than 0.5 percent of its assets in private short-term debt. The final SEC rule exempts “retail” prime or municipal money market funds, as well as government funds, from the floating NAV reform, but applies the fees and gates reform to retail money market funds. A “retail” money fund is defined as a fund having “policies and procedures reasonably designed to limit all beneficial owners of the fund to natural persons.”15 Institutional prime money market funds were the only segment that suffered significant asset outflows during the 2008 financial crisis, with approximately $300 billion (or 14 percent of total assets) withdrawn during the week of September 15, 2008.16 While wholesale creditors that are better informed and more active in managing risk are more likely to run, retail investors might still run eventually as well—as noted before without the prompt adoption of Treasury guarantees, they might well have run in the crisis.
Government money market funds (regardless of whether investors are retail or institutional) are excluded from all of the reforms. To qualify as a government money market fund, a fund is required to invest at least 99.5 percent of its total assets in cash, government securities, and/or repurchase agreements that are collateralized exclusively by government securities or cash.17 This threshold replaced a less stringent 80 percent threshold under rule 2a-7.18 Again, the basis for the exemption is that runs on government funds did not happen in the crisis—there was indeed a run to such funds. And runs on such funds are highly unlikely in the future. We shall now look at these new rules in more depth.
Under the SEC’s final rule, institutional prime and municipal institutional money market funds are no longer eligible for two exemptions provided in rule 2a-7 that currently allow funds to maintain a stable NAV, namely to “sell and redeem [money market fund] shares at a stable share price without regard to small variations in the value of the securities that comprise its portfolio.”19 The first of these exemptions—amortized cost valuation—permits a fund to value its portfolio at cost, plus or minus adjustments for amortization of premium or accumulation of discount.20 The second—“penny-rounding”—is a method of pricing fund shares that allows NAV to be rounded to the nearest one percent (or one penny for funds targeting a $1.00 share price).21 Loss of these exemptions requires institutional money market funds to “sell and redeem shares based on the current market-based value of the securities in their underlying portfolios, rounded to the fourth decimal place (e.g., $1.0000),”22 or in other words, adopt a “floating” NAV, with daily share prices generally tracking the mark-to-market value of portfolio assets.
It is highly questionable whether a floating NAV offers a solution to contagion—even if it provides more transparency of pricing to investors. A floating NAV does not reduce the underlying risk of MMF investments, including interest rate risk, credit risk and liquidity risk. MMF investors will continue to need ready access to their cash and have a low tolerance for risk. During stress events, these risk-averse investors are still able to pull back quickly and are incentivized to do so. For example, according to the ICI, “French floating NAV dynamic money funds … lost about 40 percent of their assets over a three-month time span from July 2007 to September 2007.”23 It is true that under a fixed NAV, that overstates the true value of a fund, there is an incentive to withdraw early at par rather than to remain invested and suffer the actual losses. So if an investor can today withdraw for 100 when the true value is 98 he will do so. The floating NAV will mean the investor can only withdraw today for 98 but that will not stem withdrawals based on fears that the NAV will experience further declines (e.g., to 96). A floating NAV rule might address fairness among investors but early withdrawers from the funds will still get a better price since the fund will first liquidate its most liquid and best-priced assets to fund withdrawals—later withdrawals will be funded with the sales of less liquid and more poorly priced securities.24 In any event, a floating NAV will not stem contagion.
The SEC’s rule also permits the board of a nongovernment money market fund to impose liquidity fees and redemption gates triggered by a fall in a fund’s weekly liquid assets to below certain thresholds of the fund’s total assets.25 If a fund’s weekly liquidity assets falls below 30 percent of its total assets, investor redemptions from the fund could be subject to a fixed “liquidity fee” of 2 percent, except where the fund’s board of directors determines that imposition of such fee would not be in the best interests of the fund. And a fund’s board, including a majority of independent directors, would be authorized temporarily to suspend redemptions from the fund, known as a “redemption gate.” Any such gate must be lifted within 30 days, and no fund may institute a gate for more than 10 days in any 90-day period.26 In addition, if such a fund’s weekly liquidity assets fall below 10 percent of its total assets, the fund must impose a liquidity fee of one percent on all investor redemptions, unless the fund’s board of directors determines the imposition of such a fee would not be in its best interest.27 In the end all these decisions are left to boards and are not mandated.
The SEC’s trigger may prove to render the liquidity fee and redemption gate moot. According to data assembled by Fidelity Investments,28 money market funds currently hold liquidity in amounts far in excess of the requirements of either rule 2a-7 or the proposed 15 percent weekly liquid assets threshold. The Investment Company Institute (ICI) reports that, as of December 2013, prime money market funds held 23.30 percent of their portfolios in daily liquid assets and 36.29 percent in weekly liquid assets.29 Consequently it is unlikely the liquidity fee and redemption gates would ever be triggered
In terms of the overall effectiveness of liquidity fees and redemptions, some claim that a fee or an outright redemption restriction would improve the liquidity position of money market funds by reducing MMF investors’ incentives to flee, as they did when RPF broke the buck in September 2008. However, SEC Commissioner Kara Stein has emphasized that it is questionable whether the liquidity fee and redemption gate proposal would serve to check contagion, as the threat of such measures could conceivably accelerate redemptions as investors scramble to redeem their shares before the gates are lowered or a liquidation fee is assessed.30 In unstable market environments, investors may choose to redeem en masse in order to avoid the impending redemption restrictions. Indeed, even the SEC acknowledges that “the fees and gates proposal … would not fully eliminate the incentive to quickly redeem in times of stress, because redeeming shareholders would retain an economic advantage over shareholders that remain in a fund if they redeem when the costs of liquidity are high, but the fund has not yet imposed a fee or gate.”31 Furthermore BlackRock’s client research revealed that, if a portion of balances is held back for 30 days and subordinated, MMF investors would redeem even sooner, “at the slightest sign of nervousness in the markets.”32 According to BlackRock, the complexity of the redemption restriction model is a significant disadvantage in a crisis:
[W]e believe clients would not take the time to navigate the complex structure and would be more likely to redeem earlier—and in this model, 97% of balances are open for redemption. Rather than preventing runs, we believe this approach would act to accelerate a run.33
Prior to the adoption of federally insured deposits, withdrawal suspensions were commonly used to combat bank runs in the United States.34 While these suspensions were a response to fleeing depositors, they were also a cause of depositor flight. If past experience suggests that a bank or a bank regulator will limit withdrawals or redemptions, rational market participants will almost certainly attempt to withdraw their funds prior to their suspension, accelerating the run. While redemption restrictions on bank deposits, in the form of historical bank holidays, were somewhat successful during the Great Depression, they were also accompanied by deposit insurance, which likely achieved more in terms of reassuring depositors. Such public insurance is, of course, not currently available for money market fund investors, as previously discussed. Finally, recent research suggests that money market fund managers may be reluctant to impose fees or restrictions on redemptions for fear that this could be the death knell for the future of such funds. A study of hedge funds that imposed discretionary liquidity restrictions (DLRs) on investors during the 2008 crisis found that the use of the restrictions impacted fund family reputations so that after the crisis funds from DLR families faced difficulty raising capital and were more likely to raise their fees.35
Although not adopted—or even proposed—by the SEC, a further reform that has been suggested by FSOC36 and actually proposed for European funds37 by Michel Barnier, the EU’s European Commissioner for Internal Market and Services,38 is a requirement that money market funds hold a small capital buffer, such as the 3 percent of NAV proposed by Barnier, which would reduce the risk of a run on money market funds.39 Research suggests that such a requirement could contribute to financial stability by increasing the ability of MMFs to absorb losses.40 However, as former SEC Chairman Mary Schapiro suggested in Congressional testimony that “[t]he capital buffer would not necessarily be big enough to absorb losses from all credit events. Instead, the buffer would absorb the relatively small mark-to-market losses that occur in a fund’s portfolio day to day, including when a fund is under stress.”41 In this manner the capital buffer would not prevent substantial losses that would come with a major credit event and/or fire sale of assets, which could more than overwhelm a small capital buffer. To the extent that MMF investors may run to avoid such large losses, the capital buffer would not seem to solve the problem of contagion.
As previously noted, insurance for MMFs has also been proposed as a potential reform to prevent future contagion in the industry. In its October 2010 report on MMF reform, the President’s Working Group on Financial Markets (PWG) noted that some form of insurance for MMF investors may be useful in “mitigating systemic risk posed by MMFs ….”42 The report highlights the central role that the Treasury guarantee of MMFs played in stemming runs during the financial crisis and states that insurance would reduce the risk of future runs.43 Possible insurance programs may be provided by the private sector, the public sector, or a hybrid of the two.44 A more detailed discussion of MMF insurance is found in Part III, Chapter 11 B.
In recent months there has been concern that there may be a significant reduction in the assets of institutional prime MMFs. Certain large asset managers have indicated that they will further reduce their offerings of institutional prime MMFs. Fidelity recently announced that the firm plans to stop offering prime MMFs and instead offer government money market funds.45 Charles Schwab recently announced it will convert its institutional MMFs to retail funds and one of its prime funds to government.46 Furthermore asset managers, including Blackrock and Federated, have indicated that they are considering offering private MMFs that have stable NAVs and no redemption restrictions, as an alternative to SEC publicly registered institutional prime MMFs.47 Shifting from public to private funds does not eliminate such funds, in fact it makes them more risky since such funds would not even be subject to the 2010 SEC liquidity reforms. But a shift of these funds to government funds would eliminate the risk of runs in those funds; however, such risk would still exist for retail prime funds.
In my view, none of the existing reforms solves the problem of contagion. The options for doing so are: (1) prohibit prime funds, retail or institutional; (2) insure prime funds in some manner; or (3) provide strong lender of last resort to such funds.