What Rule 2a-7 Tells Us about Re-Proposed Rule 18f-4

My previous post tried to explain how Rule 2a-7 limits the "leveraging effects" of "firm commitments" made by money market funds. This post will add some important qualifications, compare the approach taken by Rule 2a-7 to the proposed approach in Rule 18f-4 and discuss the need to reconcile these rules.

Unintended Regulation

I did not mean to suggest in my previous post that the SEC limited the weighted average maturity ("WAM") of money market funds in order to limit their use of investment leverage. Rule 2a-7 limits a money market fund's WAM to 60 days in order to limit the volatility of its net asset value. The rule recognizes that firm commitments can contribute to such volatility and thus includes firm commitments in calculating the fund's WAM, which consequently limits the extent of the fund's commitments. Indeed, the measurement of leverage by WAM is inexact. The WAM calculation uses the sum of all of the fund's investments, including the firm commitment, as the denominator in weighting maturities. A more accurate measure of leverage would use the fund's net assets (thus excluding any firm commitments) as the denominator.

Resemblance to Proposed Rule 18f-4

Proposed Rule 18f-4 would resemble Rule 2a-7 in one important respect: it would seek to limit the volatility of a fund's shares (as measured by "Value at Risk" or "VaR") directly. Under proposed Rule 18f-4, a fund that engages in more than a limited amount of derivative transactions must limit the VaR of its "portfolio" to either 150% of the VaR of a designated index or 15% of its net assets. Rule 18f-4 would not discriminate between VaR attributable to derivatives transactions and VaR attributable to other types of investments. The rule would thus limit "leveraging effects" more comprehensively than Section 18, which limits the means, rather than the effect, of leverage.

How Should Proposed Rule 18f-4 Account for Money Market Funds?

The release proposing Rule 18f-4 explains that, even though unfunded commitments may not have "leveraging effects," they "could raise the asset sufficiency concerns underlying [Section 18]," namely:

A fund could be required to liquidate other assets to obtain the cash needed to satisfy its obligation under an unfunded commitment agreement if the fund did not have cash on hand to meet its obligation …."

Release 10666 required funds to segregate liquid assets as a means of addressing this concern (as well as potential leverage). Money market funds, which consist almost entirely of liquid assets, find it easy to segregate assets to cover their firm commitments and thus rely on Release 10666 to comply with Section 18(f). The SEC proposes to withdraw Release 10666 when they adopt Rule 18f-4. The proposed definition of "fund" in Rule 18f-4 would exclude money market funds. The combination of these reforms would thus leave money market funds without a means of entering into firm commitments in compliance with Section 18(f). I can think of two approaches to addressing the use of firm commitments by money market funds. The first would be to automatically include any registered open-end company that is regulated as a money market fund under Rule 2a–7 in the exemptions provided by Rule 18f-4(b). The portfolio liquidity requirements of paragraph (d)(4) of Rule 2a-7, particularly the requirement to hold 30% or more in weekly liquid assets, should address the asset sufficiency concern underlying Section 18, just as the 60-day WAM limit should address the leveraging concern. The other approach would be to not exclude money market funds from the definition of "fund" in Rule 18f-4. This would work if the final rule continues to include a carve out for "unfunded commitment agreements" of the type entered into by money market funds. It would also be important to confirm that money market funds would never become subject to the proposed VaR limitations. I cannot imagine how a fund designed to maintain a stable net asset value could ever have a VaR that approaches 15% of its net assets, so the proposed procedures for testing and monitoring the VaR of a money market fund would be a complete waste of resources.