Skip to main content

Investment Company Liquidity Risk Management Programs Frequently Asked Questions

Nov. 15, 2017

The staff of the Division of Investment Management has prepared the following responses to questions related to the investment company liquidity risk management (“LRM”) program requirements adopted in October 2016 and expects to update this document from time to time to include responses to additional questions.  These responses represent the views of the staff of the Division of Investment Management. They are not a rule, regulation, or statement of the Commission, and the Commission has neither approved nor disapproved this information. The Adopting Release for investment company LRM programs (“Adopting Release”) is available at: https://www.sec.gov/rules/final/2016/33-10233.pdf. We define “rule” as Rule 22e-4 under the Investment Company Act.

Sub-Advised Funds

1. Q. The rule defines “person(s) designated to administer the program” as “the fund or In-Kind ETF’s investment adviser, officer, or officers (which may not be solely portfolio managers of the fund or In-Kind ETF) responsible for administering the program and its policies and procedures . . . .” May the program administrator delegate responsibilities to a sub-adviser under the fund’s liquidity risk management program, subject to appropriate oversight?

A. Yes. The rule requires a fund to adopt and implement a written LRM program, with a board-approved “person(s) designated to administer the program” (“program administrator”). The Adopting Release explains that the term “adviser” as used in the Adopting Release and the rule generally refers to any person, including a sub-adviser, that is an “investment adviser” of an investment company.[1] A fund’s sub-adviser could thus be designated as program administrator.[2]

The Commission also recognized in the Adopting Release that sub-advisers could serve as third parties who could assist the program administrator in managing liquidity risk.[3] Thus, the staff believes that the program administrator also could delegate specific responsibilities to a sub-adviser under the fund’s LRM program (e.g., providing liquidity classifications for the fund’s investments). Neither the rule nor the Adopting Release prescribes whether or how a program administrator could delegate responsibilities—either for administering the entire LRM program or for handling discrete responsibilities under the fund’s LRM program. Therefore, the staff believes that, subject to appropriate oversight, a program administrator has flexibility regarding delegation, provided that each responsibility is delegated to, and assumed and handled by, an appropriate entity. We note, however, that the fund at all times retains ultimate responsibility for complying with the rule, and the program administrator is responsible under the rule for the administration of the LRM program. A fund may wish to implement policies and procedures regarding the scope of and conditions on permitted delegation of responsibilities by the program administrator, and the program administrator may in turn wish to implement policies and procedures to oversee those to whom it has delegated responsibilities under the LRM program, including sub-advisers if applicable to ensure that those responsibilities are appropriately fulfilled. Similarly, a fund’s policies and procedures may permit an entity with delegated responsibilities to appoint sub-delegates and task them with handling certain portions of its delegated responsibilities, provided that the entities with sub-delegated authority are appropriately supervised.

2. Q. Does an adviser (including a sub-adviser) have an independent obligation to adopt and implement a liquidity risk management program?

A. No. The rule requires funds—not advisers—to adopt and implement LRM programs. However, the staff believes that the rule and Adopting Release clearly contemplate a role for advisers and their personnel in handling responsibilities under funds’ LRM programs. For example, as discussed in response to Question 1, this role could take the form of administering funds’ LRM programs or handling specific program responsibilities delegated by the program administrator.[4]

3. Q. In the staff’s view, may an adviser (including a sub-adviser) administer, or have specific delegated responsibilities under, multiple fund LRM programs that differ from one another?

A. Yes. The staff recognizes that (i) an adviser (including a sub-adviser) may provide advisory services to multiple funds (including funds in multiple fund complexes), and (ii) funds, advisers, and sub-advisers currently assess, manage, and review liquidity risk using a diverse set of practices. The Adopting Release acknowledges the need for a fund’s LRM program to be appropriately tailored to the fund’s risks and circumstances, and the staff believes the rule’s framework accommodates, and is fully consistent with, continued diversity and evolution of liquidity risk management practices.[5] An adviser (including a sub-adviser) may have responsibilities under multiple fund LRM programs that differ from one another (including multiple programs within the same fund complex), and the staff believes that such an adviser is under no obligation to reconcile the elements of those programs; the programs’ underlying methodologies, assumptions, or practices; or the program outputs (e.g., liquidity classifications of fund investments).

4. Q. If an adviser (including a sub-adviser) has responsibilities under multiple fund LRM programs, either in its role as the designated program administrator or under specific delegated responsibilities, which program should control for a particular fund?

A. Notwithstanding the diversity of fund LRM programs under which an adviser (including a sub-adviser) may have responsibilities, each fund’s board-approved LRM program will control how an adviser or sub-adviser carries out any of its responsibilities under the rule (e.g., liquidity classification of the fund’s investments).

5. Q. May different funds classify the same investment differently?

A. Yes. As the Commission stated in the Adopting Release, the Commission recognizes that different funds may classify the liquidity of similar investments differently based on the facts and circumstances informing their analysis.[6] This could result in classifications of the same investment that vary from fund to fund. Under the rule, a fund must take into account “relevant market, trading, and investment-specific characteristics” in classifying its portfolio investments’ liquidity, as well as the investments’ market depth, based on trades of the size that a fund would “reasonably anticipate trading.” Funds—including funds in the same complex—could use differing methodologies and assumptions with respect to the market, trading, and investment-specific characteristics, as well as market depth and reasonably anticipated trade size, and thereby appropriately arrive at different classifications for the same instrument.

6. Q. If a fund’s adviser and sub-adviser (assuming it has one) have differing conclusions regarding an investment’s appropriate liquidity category, how should the fund’s liquidity program policies and procedures address those differences?

A. In the staff’s view, a fund’s program administrator could delegate responsibility for the classification of investments to either its adviser or sub-adviser, in which case that entity’s decisions would control how the fund classifies its investments. Alternatively, the program administrator could adopt an approach where the adviser and sub-adviser(s) each have some input into the fund’s liquidity classifications. In the latter case, and for a variety of reasons (including those set forth in the response to Question 5 above), a fund’s adviser and sub-adviser may reach differing conclusions regarding an investment’s appropriate liquidity category. Therefore, the staff believes that the fund’s LRM program policies and procedures could address how the fund would resolve those differences. For instance, the staff believes that a fund’s policies and procedures could stipulate that, in the event of a classification difference, a specified party’s determination (e.g., program administrator, adviser, or sub-adviser) would control, or establish another method (e.g., a factor analysis or hierarchy or adopt the most conservative (i.e., least liquid) classification) for resolving classification differences. These examples are meant to be illustrative only, and fund LRM programs could employ other ways of addressing any differences.

7. Q. Suppose that (i) a fund has multiple sub-advisers, each with discretion to manage its own distinct “sleeve” (a “manager-of-managers” structure), (ii) the fund’s sub-advisers have been delegated responsibility under the liquidity program for classifying the investments in the sub-advisers’ respective sleeves, and (iii) more than one sleeve holds the same investment. If each sub-adviser classifies the commonly-held investment differently, in the staff’s view must the program administrator reconcile those differences for purposes of compliance with the rule?

A. Much like in the case of an adviser and sub-adviser discussed in question 6, in a manager-of-managers structure, the staff recognizes that multiple sub-advisers may have differing views regarding the appropriate liquidity category for their investments. Sub-advisers may manage their respective sleeves autonomously, with no reference to (or visibility into) the other sleeves. Consequently, just as funds (including funds within a complex) may classify the same investment differently, so too may sub-advisers of the same fund, because they may be guided by differing assumptions and methodologies. If such classification differences were to arise, the staff believes that neither the fund, program administrator, nor the adviser nor the sub-advisers with delegated LRM responsibilities would be under any obligation to resolve these differences for compliance purposes (e.g., in connection with monitoring for compliance with the fund’s highly liquid investment minimum (if applicable) and the 15% limit on illiquid investments). The staff believes, however, that a fund’s policies and procedures could have a process for resolving these differences, similar to those discussed in Question 6.

As discussed below, however, a fund must reconcile such classification differences for purposes of reporting on Form N-PORT.

8. Q. If a fund that operates under a manager-of-managers structure classifies the same investment held in multiple sleeves differently as outlined above for compliance purposes, how should the fund report the investment’s liquidity classification on Form N-PORT?

A. Form N-PORT does not permit a fund to report more than one liquidity classification of a single investment. If, as outlined above, a fund classifies the same investment held in multiple sleeves differently for purposes of compliance with the rule, the staff believes that its policies and procedures should have a process for selecting a single classification for that investment for purposes of Form N-PORT reporting.

When determining the overall liquidity classification of an entire investment position classified differently by multiple sub-advisers, the staff believes that a fund may use any reasonable method for resolving this difference, so long as the fund applies that method consistently. The staff believes that the fund could, for example: (i) adopt the classification of the sub-adviser with the largest position in the investment; (ii) calculate a weighted average (based on each sub-adviser’s classification and its respective position size) and round to the nearest classification; or (iii) use the most conservative (i.e., least liquid classification). These examples are meant to be illustrative only, and fund LRM programs could employ other ways of determining a single classification for purposes of Form N-PORT reporting.

The staff encourages a fund that has diverging liquidity classifications for an investment for compliance purposes but a single classification for reporting that investment on Form N-PORT to note this in the Form’s Explanatory Notes section. For example, a fund could describe which of the methods described above the fund used to resolve differences in the classification of a single investment. The staff notes that any information included in the Explanatory Notes section is non-public if it is related to a non-public reporting item, such as the item requiring a fund to report its monthly position-level liquidity classification information.[7]

ETFs

9. Q. In its discussion of circumstances under which an In-Kind ETF may use cash to meet redemptions, the Adopting Release states that such an ETF’s use of cash may “correspond to uninvested cash in the fund’s portfolio (which, to the extent that this amount of cash equals the fund’s cash position in the portfolio, would be an ‘in-kind’ redemption) . . . .” The Adopting Release then states that “if an In-Kind ETF were to use more than a de minimis amount of cash (as determined in accordance with its written policies and procedures) to meet redemptions (for any of the reasons discussed above or otherwise), it would not qualify as an In-Kind ETF . . . .” For purposes of defining and testing compliance with its de minimis cash amount, in the staff’s view may an ETF exclude the amount of cash in redemption proceeds that is proportionate to the ETF’s uninvested portfolio cash, irrespective of amount?

A. Yes. The Commission specifically recognizes in the Adopting Release that to the extent the amount of cash (i.e., U.S. currency) in a redemption equals the fund’s cash position in the portfolio, the redemption is “in-kind” and is not considered “cash” that is subject to the de minimis amount.[8] Therefore, the staff believes that an ETF may exclude cash in redemption proceeds that is proportionate to the ETF’s uninvested portfolio cash for purposes of defining and testing compliance with its de minimis cash amount.

10. Q. In the event that an ETF is no longer able to qualify for the In-Kind ETF exception, must a fund immediately come into compliance with the classification and highly liquid investment minimum requirements of the LRM program?

A. The staff understands there are practical considerations that would prevent a fund that loses its status as an In-Kind ETF from coming into immediate compliance with these requirements of the LRM program. Therefore, the staff would not recommend enforcement action if such an ETF comes into compliance with these requirements as soon as reasonably practicable after the ETF no longer qualifies for the In-Kind ETF exception. The staff notes that an ETF must indicate in Item E.5 of Form N-CEN its current status as an In-Kind ETF for the reporting period.

11. Q. According to the Adopting Release, an In-Kind ETF’s policies and procedures “would determine the amount of cash and the types of transactions that it will treat as de minimis” for purposes of qualifying as an In-Kind ETF. Notwithstanding the lack of a bright line test in the rule or Adopting Release for what constitutes a de minimis amount of cash, is there a de minimis cash amount that the SEC staff would view as reasonable?

A. The Adopting Release states that “ETFs that redeem more than a de minimis amount in cash can have substantially similar liquidity risks as mutual funds. . . .”[9] Those ETFs are, as stated in the Adopting Release, subject to the same LRM program requirements as mutual funds, including the rule’s investment classification and highly-liquid investment minimum requirements. Defining what constitutes a de minimis amount of cash in its policies and procedures is a determination that the rule and Adopting Release leave to each ETF, and therefore what is de minimis may differ among ETFs.[10] The staff believes that it would be reasonable for an In-Kind ETF to determine that if the percentage of its overall redemption proceeds paid in cash does not exceed 5% (subject to permissible exclusions, such as set forth in Question 9), such use would be de minimis.[11]

The staff further believes that an In-Kind ETF may determine that cash use of more than 5% in redemptions is de minimis. In making such a determination, the staff believes that an ETF should evaluate its particular facts and circumstances, including the ETF’s LRM program and whether a redemption(s) in cash in excess of 5% could give rise to liquidity risks substantially similar to those of mutual funds.

However, the staff believes that, if an ETF’s percentage of overall redemption proceeds paid in cash exceeds 10% (subject to permissible exclusions, such as set forth in Question 9), it would be unreasonable (subject to permissible exclusions, such as set forth in Question 9), to consider it a de minimis amount of cash for purposes of qualifying as an In-Kind ETF. The staff notes that this framework for assessing de minimis use of cash in the context of fund redemptions would not necessarily be representative of a de minimis amount in the context of other requirements under the federal securities laws.

12. Q. In its discussion of the “In-Kind ETF” definition and what would constitute a de minimis amount of cash thereunder, the Adopting Release notes that there may be circumstances under which an In-Kind ETF may use cash to meet redemptions. The Adopting Release then states, “[b]y way of example, an ETF that normally redeems in-kind, but delivers all cash to a single authorized participant [an “AP”] that elects to receive cash, would not be an ETF that uses a de minimis amount of cash.” In the staff’s view, does this statement indicate that an ETF engaging in a single redemption transaction consisting entirely of cash is necessarily precluded from qualifying as an In-Kind ETF under the rule?

A. In the staff’s view, where an ETF has agreed to accommodate an AP’s election to receive cash as a standard practice, that ETF may be exposed to the risks that the rule is generally designed to address, and thus would not be able to qualify as an In-Kind ETF. However, if the delivery of all cash to a single AP is at the ETF’s discretion (and not the election of the AP to receive cash as a standard practice), then the choice to provide cash or in-kind redemptions would be under the ETF’s control, and thus the ETF would not be exposed to the risk that it would need to sell investments to make a cash redemption in adverse liquidity situations. Accordingly, the staff believes that a redemption transaction consisting entirely of cash does not necessarily preclude an ETF from qualifying as an In-Kind ETF so long as such a redemption transaction as a proportion of the ETF’s aggregate redemption transactions does not exceed the de minimis amount of cash defined in the ETF’s policies and procedures, and the AP who receives the cash redemption is not an AP who has elected to receive cash redemptions as a standard practice.

13. Q. What methods may an In-Kind ETF use to test whether its cash use is de minimis?

A. Although the rule and Adopting Release do not prescribe a precise methodology for how an ETF should test to determine whether it continues to qualify as an In-Kind ETF (thus leaving this determination to each ETF), the staff believes that an In-Kind ETF may take a variety of reasonable approaches to determine whether its cash use is de minimis, so long as the approach the ETF selects is consistently applied. For example, the staff would not object if an In-Kind ETF were to determine that a reasonable approach might include (i) testing each individual redemption transaction, to ensure that each has no more than a de minimis cash amount, or (ii) testing its redemption transactions in their totality over some reasonable period of time to ensure that, on average, its aggregate redemption transactions have no more than a de minimis cash amount.[12] The staff believes that a reasonable period of time for a fund with frequent redemption basket activity might be a day or a week, while a reasonable period for a fund with less frequent redemption basket activity may be up to a month. The staff does not believe that using a period of time over a month would be reasonable. Regardless of the method used, the staff believes that the ETF’s policies and procedures should describe the ETF’s approach for testing compliance and the time period used, and they should be applied consistently.

14. Q. The Adopting Release also states that an ETF that has lost its status as an In-Kind ETF “may be able to conclude that it qualifies as an In-Kind ETF in later years if such circumstances are not repeated.” Does it follow from this that an ETF must wait, in all circumstances, at least two years before it can re-qualify as an In-Kind ETF?

A. No. The staff believes that there is no specific period of time that a fund must wait before it can determine that it once again qualifies as an In-Kind ETF. As noted in question 10 above, an In-Kind ETF that loses its status should come into compliance with these requirements as soon as reasonably practicable after the ETF no longer qualifies for the In-Kind ETF exception. However, the staff believes that the determination of whether an ETF qualifies as an In-Kind ETF after such status has been lost is a facts and circumstances determination. The staff believes that an ETF that loses its status as an In-Kind ETF should consider whether it is it appropriate to avail itself of the In-Kind ETF exemption, in light of the statement in the Adopting Release that an In-Kind ETF’s written policies and procedures should describe how it analyzes its ability to redeem in-kind in all market conditions such that it is unlikely to suddenly fail to qualify for the exception.[13] Nonetheless, the staff believes that an ETF that lost its status as an In-Kind ETF could potentially avail itself of the In-Kind ETF exception again if it makes a reasonable determination, based on its particular facts and circumstances, that the event that caused it to lose its status as an In-Kind ETF was an extraordinary one-time event that is unlikely to occur again. The staff believes that such an ETF should describe the factors it generally would consider in making this determination in its policies and procedures.[14]

15. Q. May an ETF, including an ETF with little or no operating history, consider factors other than its redemption history in determining whether it qualifies as an In-Kind ETF?

A. Although the staff believes that an ETF’s backward-looking redemption history is ordinarily relevant in determining whether it qualifies as an In-Kind ETF, it is not, by itself, dispositive. The staff believes that an ETF’s “in-kind” analysis also may have a forward-looking component. For example, the staff believes that a new ETF could conclude that it qualifies as an In-Kind ETF based on an analysis of its policies and procedures and its expected redemption practices.

In addition, the staff believes that an ETF with an operating history could consider material changes to its policies and procedures and redemption practices and their anticipated effects. For instance, the staff believes that if an ETF changes its policies and procedures to restrict its ability to meet redemptions using cash, and if the ETF otherwise reasonably concludes that its use of cash to meet redemptions is likely to decline to a de minimis level and reasonably expects to maintain such levels going forward, depending on the circumstances it may determine that it qualifies as an In-Kind ETF, notwithstanding recent redemption transaction history, based on an analysis of material changes made to its policies and procedures to maintain compliance.


[1] See Adopting Release at footnote 279 and accompanying text.

[2] See Adopting Release at footnote 810 and accompanying text.

[3] See Adopting Release at footnote 818 and accompanying text.

[4] See e.g., Adopting Release at footnote 66 and accompanying text.

[5] See Adopting Release at footnote 180 and accompanying text.

[6] See Adopting Release at footnote 311 and accompanying text.

[7] See General Instruction F of Form N-PORT.

[8] See Adopting Release at footnote 852 and accompanying text.

[9] See Adopting Release at footnotes 842-845 and accompanying text.

[10] Id. (“An In-Kind ETF generally should describe in its policies and procedures…the circumstances in which the In-Kind ETF may use a de minimis amount of cash to meet a redemption and what amount of cash would qualify as such.”)

[11] See question 13 on methods an In-Kind ETF may use for this calculation.

[12] See Adopting Release at footnote 854 and accompanying text.

[13] See Adopting Release at footnote 854 and accompanying text.

[14] See Adopting Release at footnote 854 and accompanying text.

Return to Top