Invesco

Invesco Advisers, Inc.
11 Greenway Plaza, Suite 1000
Houston, Texas 77046-1173
713 626 1919
www.invesco.com

January 13, 2016

VIA ELECTRONIC SUBMISSION
[email protected]
Brent J. Fields, Secretary
U.S. Securities and Exchange Commission
100 F Street, N.E.
Washington, DC 2549-1090
Re:

File Nos. S7-16-15 and S7-08-15
Comment on Open End Fund Liquidity Risk Management Programs and Swing Pricing
Proposals

Dear Mr. Fields:
Invesco Ltd. (“Invesco”) appreciates the opportunity to provide comments to the
U.S. Securities and Exchange Commission (“Commission”) in response to the Proposed Rule
published October 15, 2015.1 Invesco is a leading independent global investment manager
with approximately $775.6 billion in assets under management (“AUM”) as of December 31,
2015. We are pleased that the Commission, as the primary regulator for the asset
management industry, is addressing systemic risk and liquidity concerns and support its
efforts to reduce the risk that an open-end fund would not be able to meet its redemption
obligations. The Commission has the greatest understanding of the asset management
industry, a long history of prudent oversight and regulation of asset managers and our
products and activities, and understands how best to balance the potential benefits and
detriments of additional regulation to the U.S. capital markets, the capital formation
process, and the interests of investors. In light of Invesco’s 70 years of global industry
experience, dedication to our clients, and comprehensive range of investment capabilities,
we seek to provide you meaningful commentary in this letter (“Comment Letter”). 2

1

  Open‐End Fund Liquidity Risk Management Programs; Swing Pricing; Re‐Opening of Comment Period for 
Investment Company Reporting Modernization Release, Investment Company Act Release No. 31,835, 80 Fed.Reg. 
62,273 (proposed Oct. 15, 2015) (“Proposed Rule” or “Proposal”). 

2
 We appreciate the Commission’s thoughtful consideration of our relevant views expressed in our comments to 
the FSOC Notice in crafting the Proposed Rule, which are further explained in this Comment Letter.  See Financial 
Stability Oversight Council, Notice Seeking Comment on Asset Management Products and Activities, 79 Fed. Reg. 
77488 (Dec. 24, 2014) (“FSOC Notice”) and Comment Letter of Invesco Ltd. on the FSOC Notice (Mar. 25, 2015) 
(“Invesco FSOC Comment Letter”). 

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 2
Invesco is a global company focused solely on investment management, and our
services are provided through a wide range of product types. In particular, our subsidiaries
advise both mutual funds and open-end exchange-traded funds (“ETFs”) (excluding our
ETFs structured as unit investment trusts) for a broad client base with a combined AUM of
approximately $357 billion as of December 31, 2015. The 145 mutual funds Invesco
currently offers include a wide range of actively-managed, domestic, international/global,
specialty (including bank loan and alternative), and fixed income mutual funds to help
customize investors' portfolios to their unique needs. Invesco currently offers over 130
ETFs spanning seven broad categories: specialty, commodities and currencies, equity-based
resources, factor driven, alternatively weighted, income, and quantitative. Also relevant to
our discussion is our extensive experience and expertise with European investment funds.
Invesco manages more than 80 Ireland and Luxembourg domiciled UCITs that use swing
pricing across equity, fixed income, and alternative strategies offered to retail shareholders
with an AUM of over $54 billion USD as of December 31, 2015 (“Invesco UCITs”). Invesco
is well positioned to understand concerns regarding dilution of shareholder interests and
risks associated with open-end fund shareholder redemptions.
I.

Summary of Proposed Rule

The Proposed Rule would require each open-end mutual fund (except money market
funds) and open-end ETF (collectively, “open-end funds” or “funds”) to establish a tailored
liquidity risk management program (“LRM program”) under new rule 22e-4. The LRM
program would include an ongoing classification of each portfolio position (asset)3 into one
of six liquidity buckets, which are based on the number of days in which the position would
be converted to cash at a price that does not materially affect the value of that asset
immediately prior to the sale, i.e. convertible to cash within 1 business day, 2-3 business
days, 4-7 calendar days, 8-15 calendar days, 16-30 calendar days, and more than 30
calendar days. The classification would be based on a number of asset-level factors. 4
Funds would be required to disclose the classifications on proposed Form N-PORT, which
entails reviewing the classifications at least monthly. The information would be made public
for the last day of each quarter, on a 60-day delayed basis.

3

 References to “assets” should be read to include “positions” or “portions thereof,” as applicable. 

4

 Some of the factors a fund must consider, as applicable, about each asset when classifying the assets include:  
 existence of an active market for the asset, including whether the asset is listed on an exchange, as well as 
the number, diversity, and quality of market participants;  
 frequency of trades or quotes and the average daily trading volume;  
 volatility of trading prices;  
 bid‐ask spreads;  
 whether the asset has a relatively standardized and simple structure; 
 for fixed income securities, the maturity and date of the issue;  
 restrictions on trading and limitations on transfer of the asset;  
 the size of the fund’s position in the asset relative to the asset’s average daily trading volume and, as 
applicable, the number of units of the asset outstanding; and   
 relationship of the asset to another portfolio asset. 
(Collectively, “asset‐level factors”). 

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 3
The LRM program would also include a minimum percentage of each Fund’s net
assets invested in securities that can be converted to cash within three business days
without materially affecting the value of the asset immediately prior to sale (“three-day
liquid asset minimum”), which the Commission stated is the “cornerstone” of the program.5
If a fund falls below its three-day liquid asset minimum, it would be prohibited from
acquiring additional “less liquid” assets until the three-day liquid asset minimum is met. In
addition, the Proposal would codify the current Commission guideline that open-end funds
limit investments in assets that may not be sold or disposed of in the ordinary course of
business within seven calendar days at approximately the value ascribed to it by the fund
(“15% standard assets”) to not more than 15% of a fund’s total net assets (the “15%
illiquidity maximum”).
As part of the LRM program, each fund would need to assess and periodically review
the portfolio’s liquidity risk based on certain factors, such as size, frequency, and volatility
of historical purchases and redemptions of fund shares during normal and stressed periods
(“portfolio-level factors”). A fund’s board, including a majority of the fund’s independent
directors, would be required to approve the fund’s initial LRM program, including the fund’s
three-day liquid asset minimum, along with any material changes to the program. Large
entities, such as Invesco, would have 18 months from the effective date to comply with the
final rule.
The Commission also is proposing amendments to rule 22c-1 to permit a fund, under
certain circumstances, to use swing pricing. Under the Proposed Rule, open-end mutual
funds, except money market funds, would be permitted to adjust (or “swing”) the net asset
value (“NAV”) upward if there is a daily net purchase of fund shares (such that transacting
shareholders will bear the costs resulting from fund purchases of portfolio securities) or
downward if there is a daily net redemption of fund shares (such that transacting
shareholders bear the costs resulting from fund sales of portfolio securities). If the swing
threshold (the predetermined net purchase or redemption amount) is met, a fund would
adjust the NAV by a “swing factor,” which is the amount, expressed as a percentage of the
fund’s NAV, that takes into account (a) any near-term costs expected to be incurred
(including transaction costs and market impact); and (b) the value of assets purchased or
sold by the fund to satisfy net purchases or net redemptions that occur on the day the
swing factor is used to adjust the fund’s NAV. Only the adjusted NAV would be disclosed in
filings, performance reporting, and advertising materials. The board would be required to
approve the swing policies and procedures. The rule would be immediately effective after
the effective date of the final rules.
I. Executive Summary
In this Comment Letter, we discuss the following: (a) what we believe are important
overarching considerations; (b) our experience and understanding of liquidity practices in
open-end funds and swing pricing in UCITS; (c) our concerns regarding the proposed LRM
program and swing pricing; and (d) a possible alternative regulatory framework for
addressing liquidity that we believe would reduce liquidity risk without producing unintended
adverse consequences for shareholders and the industry. This Comment Letter is divided in
two main sections: the first section discusses liquidity risk management and the second
section discusses swing pricing.

5

 Proposed Rule at 145. 

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 4
A. Liquidity Risk Management Proposal
An overriding concern for Invesco is that the Proposed Rule is not solving
for measurable redemption risk and that it is neither necessary nor beneficial to
shareholders. We are unaware of investor protection concerns related to redemption risk,
liquidity risk, or systemic risk that justify prescriptive liquidity risk management rules. The
open-end fund industry poses minimal redemption risk or risk to the stability of U.S. or
global financial markets. While open-end funds are subject to liquidity risk, liquidity risk is a
basic element of investing. Prescriptive liquidity restrictions are not necessary for investor
protection, not necessarily in the best of interest of shareholders, nor have we found it part
of the Commission’s statutory mandate. We request that the Commission engage in further
study and analysis regarding the redemption and systemic risks posed by open-end funds
and the likely efficacy of the Proposed Rule before adopting liquidity risk management rules.
Invesco has successfully managed redemptions, even during extraordinary
periods of market stress, in the absence of prescriptive regulation. At the center of
our liquidity risk management practices are our portfolio managers, who analyze macroeconomic factors, asset-level markets, shareholder flows, and numerous additional factors
to understand and manage a fund’s liquidity risks. The portfolio managers, supported by
data and analytics, use a variety of ways to meet shareholder redemption requests. We do
not believe that portfolio managers rely most often on selling the most liquid assets first,
which appears to be the assumption behind many of the proposed LRM program
requirements.
Invesco believes in the effectiveness of our liquidity risk management
practices and supports a requirement for a strong and effective baseline LRM
program modeled upon the compliance program rule pursuant to Rule 38a-1,
promulgated under the Investment Company Act of 1940 (“1940 Act”). The LRM
program should be broadly required to address certain areas: the liquidity of the markets in
which the portfolio has significant exposure, the liquidity of portfolio assets, the portfolio’s
strategy and disclosed risk appetite, historical flows, and sufficiency of supplemental
liquidity tools in normal and stressed conditions. Invesco considers the required 15%
illiquidity maximum an important and instrumental element of an LRM program. Each
fund's board of directors should oversee the program by periodically reviewing the program,
changes to the program, its effectiveness, and its policies and procedures; however, the
board should not be required to approve a three-day liquid asset minimum because board
members do not have the resources or appropriate expertise to make such determinations.
The adviser should be free to determine the identity of the administrator of an LRM program
without restrictions; Invesco believes portfolio managers are in the best position to
administer the program with support from other departments due to their expertise and
ability to move quickly in response to shifting environments.
The proposed LRM program is not effective because of the prescriptive and
rigid classification system and three-day liquid asset minimum requirements,
which have the effect of unduly restricting certain investment strategies. As a
result of the position classification and three-day liquid asset minimum requirements,
portfolio compositions of funds will homogenize, regardless of whether the fund had
appropriate liquidity prior to the adoption of the Proposed Rule. This homogenization may
lead to diminished diversification among funds and tightened market liquidity.

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 5
The proposed position classification system, in particular, is overly
burdensome and ineffective. Invesco recommends a broad requirement to include
the liquidity of a portfolio asset as a factor in determining the overall liquidity of a
fund. We strongly urge the Commission to not require position classification because (a) a
security’s liquidity cannot be predicted with sufficient precision; (b) the criteria used to
classify positions are difficult to obtain, costly, or simply not available; (c) consistency of
classification amongst funds from different sponsors is highly unlikely and the inherent
subjectivity required in performing a position classification renders the classification
unhelpful to the Commission or shareholders; (d) the position classification potentially
exposes funds to unacceptable amounts of liability; (e) the frequency of reviewing portfolio
position liquidity is unreasonable and misguided in achieving objectives; and (f) the
significant labor and cost required is not justified.
The three-day liquid asset minimum proposal will lead to unintended
consequences; Invesco recommends, as an alternative, a required targeted range
of three-day and/or seven-day liquid assets. The three-day liquid asset minimum as
proposed will likely cause several unintended consequences: (a) increased portfolio
uniformity and decreased diversification within the asset management industry as funds
gravitate towards the same “liquid” assets; (b) tightened supply of eligible investments for
mutual funds; (c) reduced performance frustrating investor expectations as funds are
constrained from taking advantage of investment opportunities that might have a
meaningful impact on performance; (d) loss of assets from the industry as investors seek
alternatives to open-end funds; and (e) the inability of index funds to replicate the index to
the degree expected. Moreover, a three-day liquid asset minimum will probably not be
helpful for liquidity risk management or achieve its intended purpose in reducing
redemption risk because it will either not be needed or will not be sufficient. If the
Commission is determined to require a minimum amount of liquid assets in a fund, Invesco
recommends a requirement to maintain a target of three-day and/or seven-day liquid
portfolio assets within a reasonably narrow range. The use of a “target” range would allow
the fund the flexibility to adjust the minimum during changing market conditions, such that
the fund is not unnecessarily constrained and also allow enough diversification amongst
funds to avoid some of the potential negative consequences.
Invesco believes the depth and frequency of the proposed public disclosure
of position classifications are inappropriate. We agree that information about each
fund's LRM program and liquidity risks should be described and disclosed in such a way that
shareholders will find it useful in understanding the anticipated liquidity of the portfolio and
potential redemption risks. However, the proposed disclosure requirements pose the
following concerns: (a) we believe that the asset-level data proposed to be provided under
the Disclosure Proposal,6 Commission exams and reviews of the LRM program, and board
oversight obviates the need for position classification for Commission oversight purposes;
(b) the information is overly subjective and would be confusing to shareholders and
unhelpful to the Commission; and (c) disclosure would be harmful to the funds and advisers
because of the confidential and proprietary nature of the information.

6

 Investment Company Reporting Modernization, Investment Company Act Release No. 31,610, 80 Fed.Reg. 33,590 
(proposed June 12, 2015); Amendments to Form ADV and Investment Advisers Act Rules, Investment Advisers Act 
Release No. 4,091 (May 20, 2015), 80 Fed.Reg. 33,718 (proposed June 12, 2015) (the "Disclosure Proposal"). 
H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 6
Invesco recommends that the Commission consider the in-kind nature of
most ETF redemptions and exclude ETFs from the LRM program requirements. The
classification proposal is based on time to convert portfolio holdings to cash and assets are
not typically converted to cash by an ETF. When an authorized participant redeems in cash,
the variable transaction fee that an ETF may impose to offset transaction costs should
address both dilution and liquidity concerns. The three-day liquid asset minimum is
particularly challenging, if not impossible, for an index ETF to meet because of the ETF’s
need to track the index.
Finally, Invesco requests Commission assistance regarding interfund
lending, redemptions in-kind, and centralized trading to improve liquidity.
Interfund lending’s usefulness as an LRM tool in extraordinary circumstances could be
enhanced if some of the current restrictions imposed by exemptive orders are relaxed, such
as the ability to have an outstanding loan for more than seven days and curbing the effect
of a decline in NAV on an outstanding loan. We also welcome efforts by the Commission to
promote and facilitate funds’ ability to use redemptions in-kind with entities that are able to
receive securities. Invesco urges creating centralized trading platforms and other initiatives
to shorten settlement periods as a way of improving liquidity without harming shareholders,
markets, or funds.
B. Swing Pricing Proposal
Invesco agrees that partial swing pricing may help in mitigating potential
dilution of fund shareholders. Invesco believes swing pricing (a) must be
mandatory in order to have funds use it, (b) funds must receive investor flow
information in sufficient time for swing pricing to be operationally feasible, and (c)
there must be a safe harbor for good faith decisions and actions that shield funds
from potential liability. If these concerns are addressed, Invesco recommends that:


The calculation of the swing factor must be quantitative, automatable, and
transparent in order to avoid errors and potential liability;



The cost of market impact in meeting net redemptions or subscriptions is not a
quantifiable, automatable, or transparent amount so it should not be used in
calculating the swing factor;



It is inappropriate to include the adjusted NAV in the balance sheet, results of
operations and performance calculations in a fund’s financial statements or other
performance related disclosure documents;



The 2% variable fee cap should be lifted for ETFs (preferred option); or ETFs
should be permitted to use swing pricing;



Each fund’s board of directors should oversee the swing pricing program, but not
be required to approve the swing factor or threshold.

II. Comment on Liquidity Risk Management Proposal
Invesco generally supports the establishment of baseline LRM programs for open-end
funds; however we have concerns regarding various prescriptive elements of the Proposed
Rule. In this section of the Comment Letter, we discuss the following: (a) Invesco’s current
H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 7
liquidity risk management practices and our management of redemptions; (b) our thoughts
on the purpose and purported benefits of the Proposal; (c) specific concerns regarding, in
particular, the position classification system, the three-day liquid asset minimum and
proposed disclosure requirements; and (d) Invesco’s suggested recommendations for
alternatives or alterations to these elements.
A. Invesco Has Strong and Exemplary Liquidity Risk Management Practices.
1. Invesco Has Successfully Managed Redemptions, Even During
Extraordinary Events of Stress, Without Prescriptive Regulation.
Invesco has successfully managed redemptions, even during extraordinary events of
market stress and fund-specific stress, 7 in the absence of prescriptive regulation, such as
the Proposed Rule. Invesco’s funds have not experienced unmanageable strain in meeting
redemptions, even during periods of extraordinary market events. We reviewed daily flows,
redemptions, and asset changes in our high yield fixed income funds during the global
financial crisis, sovereign debt crisis, and other periods of market stress and found that
even funds that experienced their highest level of net outflows during those periods were
able to meet redemption requests capably.
Invesco would have to either (a) abandon our carefully tailored and robust liquidity
risk management practices in favor of the Commission’s untested, one-size fits all
proposals, which we find to be overly burdensome and not beneficial or (b) run two parallel
processes.
2. Invesco’s Liquidity Risk Management Practices Facilitated Our Ability
to Meet Shareholder Redemptions.
Invesco devotes a great deal of resources and time to the prudent assessment and
management of the different risks that affect a portfolio, including liquidity risk. Invesco
effectively manages liquidity using a three-pronged approach: (a) prudent risk management
by strong, well-resourced portfolio management teams and their CIOs (defined below); (b)
supplemental liquidity management tools and techniques; and (c) risk oversight and
governance. Our investment teams are supported by data and analysis provided by a
dedicated team of independent professionals.
Prudent Risk Management by Strong Portfolio Management Teams. Our portfolio
management (or investment) teams form the center of our practices. Invesco has more
7

 During the “Taper Tantrum” in 2013, open‐end municipal funds experienced the largest redemption event since 
1985 with over $50 billion in net outflows.  Invesco’s open‐end municipal funds experienced net outflows of $1.3 
billion during that same period and the Invesco High Yield Municipal Fund alone represented $1 billion of those 
outflows.  At all times the Invesco funds were able to meet their redemptions within standard market practices 
without taking any extraordinary measures. 

In late 2013, a high‐profile manager of Invesco’s two largest UK onshore products announced his departure from 
the firm; over the next eight months, one of the funds, with AUM of approximately $16 billion USD, experienced 
net outflows of over $6.9 billion USD.  The fund met all redemptions on a timely basis and shareholders were able 
to trade in and out of the fund without difficulty.  The fund’s portfolio managers maintained liquidity through 
prudent cash management and trading strategies and limited use of an overdraft facility provided by the fund’s 
custodian. 
H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 8
than 750 investment professionals worldwide in specialized investment teams who are
committed to our culture of strong risk management. Each team operates under the
supervision of its Chief Investment Officer (“ClO”) and Invesco’s executive management.
The investment teams have access to advanced investment technology, proprietary tools,
and investment platforms. Thus equipped, the portfolio managers actively monitor liquidity,
consider liquidity as a criterion in selecting instruments for their portfolios and make
frequent, typically daily, liquidity determinations, such as how much cash or cash
equivalents they should hold or whether to sell less desirable positions to meet shareholder
redemptions. Given the breadth of experience, education, information, expertise, and
investment tools available to our portfolio managers, Invesco believes that these individuals
are in the best position to evaluate and make the liquidity determinations for each of the
portfolios they manage. No one understands the assets and the markets for the assets as
well as the portfolio managers; therefore, we appropriately rely on their judgment to assess
liquidity risks and needs.
Liquidity Management Tools. Invesco’s portfolio management teams manage funds
for normal and foreseeable conditions and prepare for and consider stressed and
extraordinary circumstances. The investment teams may meet normal or foreseeable
redemptions in a number of ways. Normally, cash flows from sources such as gross
subscriptions (including new purchases and reinvested dividends on fund shares), dividend
and interest payments on portfolio securities and maturities of debt securities held in
portfolios, or cash in the portfolio are sufficient to meet normal redemption levels. When a
fund has large redemptions, the investment team, in discussion with back office support,
considers various options. Their decision is based on the best interest of the fund and
operational feasibility. This point is important because the three-day liquid asset minimum
and the position classification proposals appear to be based on the inaccurate assumptions
that (a) funds generally sell their most liquid assets to meet shareholder redemptions; and
(b) such redemptions cause harm. Some of the ways that our portfolio managers meet
redemptions are described below.


A fund may redeem in-kind (typically with respect to positions held by large
insurance company separate accounts in connection with the liquidation of the
account’s entire position pursuant to a substitution order). While this option is
not always available, Invesco has been able to redeem in-kind on numerous
occasions. We have found that certain security types are more amenable to
redemptions in-kind, such as domestic securities. We discuss some of the
challenges of redeeming in-kind later in this Comment Letter.



A fund may be able speed up settlement times. For instance, in
August/September 2011, one of our bank loan funds was able to accelerate loan
settlements during its heaviest period of redemption activity. Although a fund
cannot consistently rely on the ability to speed up settlements, it typically has
some control over the duration of the settlement period. Usually, a seller (a
fund) does not have incentive to speed up settlement because the seller collects
interest during the period. Thus, a buyer is often incentivized to accelerate
settlement.



Invesco’s portfolio managers, more often than not, opportunistically sell assets (if
they determine to sell assets to meet redemptions). Invesco portfolio managers
may seek to reposition portfolios and use redemptions as a catalyst for such
changes. Depending upon prevailing market conditions, prudent investment

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 9
managers, such as Invesco, consider remaining shareholders in meeting
redemptions. For example, in falling markets, portfolio managers may use
redemptions as an opportunity to sell assets such that the fund has less credit
risk.


Invesco may also sell assets pro rata. However, we have found that certain
situations are not appropriate for selling assets pro rata because of “practical
limitations on a fund’s ability to sell a pro rata slice of its portfolio, such as
minimum trade sizes, transfer restrictions, illiquid assets, tax complications from
certain sales, and avoidance of odd lot positions”8 or we found it unfavorable to
remaining shareholders. If portfolio managers determine to sell assets to meet
redemptions and the funds do not have these limitations, they tend to rely more
heavily on selling assets pro rata.



Invesco portfolio managers may sell the most liquid assets in a portfolio when
there is a need to raise cash (whether to fund redemptions or for other
purposes). The Proposed Rule is, in part, built on the theory that a fund usually
sells its most liquid assets first. Across asset strategies, we have not found that
to be the case. Even if a portfolio manager does decide to meet redemptions by
selling the most liquid assets first, it does not necessarily mean that the
remaining shareholders are worse off or that all of the most liquid assets are
sold. Selling the most liquid assets to meet redemptions may, in fact, benefit
remaining shareholders since more liquid assets are generally lower yielding.
Once these are sold, the remaining, relatively less liquid assets will increase the
earnings of the fund and help maintain its dividend rate. Furthermore, those
sales transactions can create tax loss carryforwards that may be used to offset
gains, avoiding or reducing the distribution of taxable capital gains to
shareholders. Moreover, it is usually easier and more cost efficient to repurchase
the most liquid assets, in the event the fund experiences inflows and the portfolio
manager chooses to repurchase such assets. Portfolio managers also have
deliberately improved the liquidity of a fund by focusing on more liquid assets or
even cash in anticipation of market stress. For example, many Invesco funds,
such as our municipal funds, gradually increased cash and liquid holdings
following the market disruptions of late 2008 and 2010.

Supplemental Tools and Techniques. Invesco has supplemental tools and techniques
upon which to rely to meet unanticipated levels of redemption. Invesco received exemptive
relief permitting interfund lending and borrowing for its long-term U.S. mutual funds. In
practice, however, Invesco has not used this relief since 2005. We are supportive of
Commission efforts to make interfund lending more useful in the future as discussed in
more detail later in this Comment Letter. Invesco has in place a line of credit for the
Invesco Floating Rate Fund9 and the PowerShares Senior Loan Portfolio (“BKLN”) (a listed
ETF investing in bank loans). These lines have been used to meet extraordinary redemption
8

 Proposal at 19, n.38. 

9

 The Invesco Floating Rate Fund, a fund that invests primarily in bank loans, used its line of credit as a preferential 
investment option to assist in meeting shareholder redemptions during periods of historically high outflows for 
bank loan funds.  We use this opportunity to correct an earlier statement in the Invesco FSOC Comment Letter: the 
Fund used the line of credit on more than two separate occasions for redemptions. 

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 10
activity on an infrequent basis. After a careful evaluation of our diverse product line-up, we
did not find that any other fund warranted the additional shareholder expense associated
with a committed line of credit in light of our comfort with our risk management practices
and historical abilities to meet redemptions, even under stressed conditions. Invesco has
not taken advantage of the statutory ability to extend settlement of mutual fund
redemptions to seven days, as permitted by Section 22(e) of the 1940 Act and disclosed to
shareholders; Invesco has always settled mutual fund redemptions within the applicable
market practice settlement period. Similarly, Invesco has never sought extraordinary relief
from the SEC to suspend mutual fund redemptions under Section 22(e)(2) of the 1940 Act.
Risk Oversight and Governance. Invesco’s executive management oversight and
governance includes the Invesco Global Performance and Risk Committee (“GPR
Committee”),10 which oversees Invesco investment teams, and reviews portfolio risk and
positioning. Invesco’s Global Performance Measurement & Risk group (“GPMR”) offers the
GPR Committee comprehensive, detailed information and analysis regarding investment
teams and the products they manage. Additionally, Invesco management carefully
considers portfolio construction during the design phase of new funds. For example, in
developing BKLN, Invesco used its extensive experience and expertise from managing daily
liquidity in senior loan registered investment companies since the mid 1990’s. We spent
over a year thoughtfully planning the launch and management of BKLN. At the forefront of
all our discussions was structuring a product that would provide investors access to the
private loan market in an exchange-traded fund format, treating liquidity with great
importance.
Independent Risk Data and Analysis. GPMR also supports the investment teams and
management by providing independent investment risk information, including with respect
to investment portfolio liquidity.11 GPMR provides analysis to help inform portfolio
managers’ understanding of risk in their portfolios and provide insight for investment team
ClOs and others who provide oversight. We consider the data and analyses provided by
GPMR as an important component of our risk oversight practices, but also recognize its
limitations and relative importance with respect to portfolio manager analysis and trading
insight. Therefore, GPMR’s data and analyses are not central to portfolio managers’
analyses.
GPMR provides information that can be used as an input for portfolio managers’
analysis of portfolio liquidity, but they do not direct the decision-making of any portion of
portfolio management. Investment team oversight at Invesco is the ultimate responsibility
of the Chief Investment Officer, facilitated by the GPMR teams. Through rigorous portfolio
analysis, the primary objective of the process is to monitor potential sources of risk and
help ensure adherence to stated mandates, process, and style. We believe that measuring
and defining liquidity risk itself is a multi-dimensional undertaking and recognize that
assessing risk requires human judgment with the aid of quantitative tools. GPMR assists
our portfolio managers and CIOs to better understand market risks in the funds that they
10

 The committee is chaired by the Chief Executive Officer and includes those Invesco Senior Managing Directors 
(executive officers reporting to the CEO) who oversee Invesco investment teams. 

11
 GPMR provides comprehensive, detailed information and analysis to the investment teams and internal groups.  
GPMR utilizes the best of breed analytical tools across all asset types underpinned by global data architecture and 
proprietary software applications for performance attribution and portfolio analysis that provides the framework 
for value added insight.   
H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 11
manage; however, Invesco also recognizes the need for flexibility in that analysis to account
for the different asset classes, products, and strategies Invesco deploys.
By way of example, our fixed income assessment of liquidity considers three levels of
liquidity analysis: market, sector, and fund-specific. Market level parameters are used to
gauge trends and as early warning signals. Market level factors used to determine liquidity
impact may include implied volatility, funding costs and risk aversion (among others).
Sector level technicals provide further insight; the factors to determine liquidity impact for
each asset class necessarily vary. Examples of factors in various sectors include new
issuance, trading volumes and sector level bid/ask spreads. Finally, GPMR analyzes the
liquidity risk profile of each fund in combination with stress testing. Examples of analysis
used to inform and challenge portfolio managers regarding the implications of current
positioning relative to varying market environments include (a) measuring trading volume
at the security level for equity portfolios; and (b) the ability to quantify cost to liquidate
using pro-rata, lowest cost and minimum market impact ordering methods for fixed income
portfolios.
Many portfolio managers find GPMR’s liquidity information and stress testing helpful
in the evaluation of risks relative to a fund’s investment objective while others find more
limited value with respect to their investment process. Portfolio managers have identified
certain common challenges regarding the current liquidity information provided by GPMR or
market vendors/suppliers, such as that the information may not reflect current market
dynamics, may be duplicative of what they already know, or does not consider the market
expertise of the traders or portfolio manager. Certain portfolio managers may not find as
much value in the stress testing because they feel that their daily, market-oriented, realtime management obviates the need for theoretical exercises whose results may correlate
poorly with actual market dynamics. As a result, Invesco continues to rely first and
foremost on the portfolio management team’s judicious use of such data and analysis.
The bucketing approach for categorizing the liquidity of portfolio holdings mentioned
in the Invesco FSOC Comment Letter consisted of breaking down each fund’s holdings into
liquidity buckets based on the cost of liquidation. The liquidity analysis, combined with
stress testing and captured within portfolio analytics tools, continues to be enhanced as the
markets and those tools evolve. Because the bucketing approach had several
disadvantages, including the inability to test stressed conditions and gaps or incompleteness
of certain data elements, we have since modified our approach. We supplanted bucketing
with a methodology that provides liquidation cost over several horizons and expanding
stress testing. The new framework continues to be based on liquidation cost rather than
days to liquidate.
We wish to also make clear that the bucketing or grouping approach mentioned in
the Invesco FSOC Comment Letter should not be confused with the Commission’s proposed
classification system. Invesco may indeed group elements of information that form our
liquidity view; however, we do not group or classify holdings by the number of days to
liquidate. Furthermore, that information is not disclosed publicly, which is also a critical
difference from the Proposal. As we explain in this Comment Letter, the validity and
scarcity of data, subjectivity of position-level classification determinations, and ephemeral
nature of the information (even the most readily available equity trading volume
information can change quickly) suggests that this information can be quite misleading for
investors and potentially opens the door for unnecessary and unproductive litigation.

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 12
B. We Have Concerns Regarding the Purpose and Benefit of the Proposed
LRM Program.
The Commission has stated that the goal of the Proposed Rule is to “promote
investor protection by reducing the risk that funds will be unable to meet their redemption
obligations, elevating the overall quality of liquidity risk management practices, and
mitigating potential dilution of existing shareholders’ interest.”12 We consider mitigating
redemption risk, i.e., the risk that a fund cannot meet its regulatory responsibility to meet
shareholder redemptions under normal or reasonably foreseeable stressed conditions, an
important goal of Commission regulation stemming from Section 22(e). We have not found
investor protection concerns related to redemption risk, liquidity risk, or systemic risk that
justify prescriptive liquidity rules.13 Our concern, thus, is that the Proposed Rule is not
solving for measurable redemption risk and is neither necessary nor beneficial.
1. Open-End Funds Pose Minimal Redemption Risk Such That Additional
Rulemaking is Not Necessary.
The open-end fund industry poses minimal redemption risk. While Invesco supports
carefully tailored rulemaking to reduce the risk that a fund cannot meet its redemption
obligations, we do not believe that open-end funds are exposed to sufficient redemption risk
to require a prescriptive LRM program in normal or foreseeable, stressed circumstances.
First, the industry has generally demonstrated its ability to meet shareholder redemptions
for 75 years (including through a global recession and other financial crises) through
prudent portfolio management in a solid regulatory framework.14 Second, asset managers
12

 Proposal at 275.   

13

 We suggest that liquidity risk and the Proposed Rule be reviewed holistically in light of the impact of other 
rulemakings addressing data reporting, derivatives, and transition planning.  The objectives of the rulemakings are 
interrelated.  Each potential regulation affects redemption risk and liquidity risk since they interact with other 
risks, regulatory requirements, and policies.  The holistic view is similar to the liquidity risk view of UCITS:  
“[l]iquidity risk for UCITS should be regarded in a holistic manner as it often interacts with other types of risk, with 
regulatory requirements as well as procedures and control activities.”  UCITS Liquidity Risk Management, Assoc. of 
the Luxembourg Fund Industry, 3 (Mar. 2013) (“ALFI Guidelines”).  For example, while still considering the 
implications of the Commission’s recently released proposal regarding the use of derivatives and leverage, we 
consider the prudent use of derivatives and leverage as significant considerations of whether a fund can meet 
redemption expectations.  The enhanced data reporting, including asset level reporting, that the Commission 
proposed in order to improve the quality and type of information provided, should also contribute to the oversight 
of redemption, liquidity, and systemic risks.  Furthermore, if a rule is finalized and implemented, the data received 
will be invaluably informative in better assessing the need for and construction of the liquidity rules.  Additionally, 
the risk that a fund is unable to meet redemption requests is directly related to the purposes of stress testing 
although we note that the Commission determined to not require stress testing for liquidity risk management.  See 
Proposed Rule at 109.  We also consider the discussion of liquidity fees and redemption gates to be tied to 
transition or resolution planning, as that is most likely the only situation in which such fees and gates would be 
introduced.   

14
 In a rare failure of a mutual fund, Third Avenue Focused Credit Fund (“Third Avenue FCF”) announced it was 
liquidating assets due to redemption requests and “general reduction of liquidity in the fixed income markets.”  
David Barse, Letter to Shareholders of Third Avenue Funds Focused Credit Fund (Dec. 9, 2015), 
http://thirdave.com/wp‐content/uploads/2015/12/FCF‐Shareholder‐Letter‐12‐2015.pdf.  Third Avenue FCF, 
Heartland Advisors’ three municipal bond funds in 2001, and two Schroder municipal bond funds in 2008 represent 
H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 13
have strong incentives, without prescriptive regulation, to manage redemption risk and
liquidity risk. The risk/return of the fund, reputation of the fund and the manager, and
income of the asset manager are directly affected by the prudent management of these
risks. For example, a fund (such as Third Avenue FCF) that cannot meet redemption
requests faces full closure and liquidation. Third, the Commission has already promulgated
guidance, rules, and regulations that have had the effect of helping to ensure sufficient
liquidity. Investors have benefited from the 15% illiquidity guideline, leverage limits, clear
and plain English shareholder disclosure requirements, asset concentration limits, and a
multitude of Commission pronouncements that have had the effect of ensuring sufficient
liquidity. This collection of federal rules and regulations, in part, enabled the proper
functioning of the funds during historic periods of stress. Finally, neither the Proposed Rule

the isolated failures in mutual fund history, all of which were idiosyncratic events.  Idiosyncratic risks and events do 
not justify imposing broad, prescriptive liquidity rules with substantial negative consequences for the entire 
industry. 

Based on a review of its disclosures, the Third Avenue FCF’s failure is perhaps not shocking.  Third Avenue FCF was 
an unusual fund and was not comparable to the investment strategies or portfolio holdings employed by more 
traditional high yield funds.  As evidenced by its disclosures, it had a significant focus in distressed debt and 
concentrated its investments in few names, i.e. the portfolio construction was extremely aggressive and took on 
significant amounts of credit, industry concentration, and diversification risk for a mutual fund.  “To be sure, Third 
Avenue’s portfolio was far riskier and more concentrated than the average junk‐bond mutual fund.  More than 
50% of its assets were unrated by credit agencies, while another 28% of the fund held bond issues rated CCC.  
That’s nearly triple the proportion held by high‐yield peers, according to S&P Capital IQ.  ‘These are the most 
illiquid bonds in an already illiquid market,’ says Morningstar senior fixed‐income analyst Sumit Desai.  ‘In 
hindsight, the strategy probably shouldn’t have been in a mutual fund wrapper.’”  Amy Feldman, Third Avenue 
Focused Credit Closes, Barron’s (Dec. 12, 2015), http://www.barrons.com/articles/third‐avenue‐focused‐credit‐
closes‐1449899465.  Most high yield and bank loan funds are far more diversified by issuer and hold a substantially 
greater amount of higher quality and more liquid assets.  Unlike other high yield and bank loan funds that met 
shareholder redemptions during extreme market stress and outflows in the 2008 global credit crisis or 2013 Taper 
Tantrum, Third Avenue FCF was launched in 2009 after the credit crisis when the fund’s strategy had been in the 
markets’ favor.  Until more recently, we did not observe Third Avenue FCF experiencing turbulent market 
conditions that tested its strategy.  The consequences of the Third Avenue FCF failure are material to the fund, its 
shareholders, and its sponsor, the chief executive officer departed, and the asset manager sustained a material 
reputational impact.  It was, otherwise, an idiosyncratic, not systemic, event that affected a limited number of 
entities with minimal impact on the broader financial system.  For instance, Invesco’s higher yielding fixed income 
funds did not experience significant redemption pressure (although there were higher than average redemptions 
for a few days) because the asset classes had natural buyers that softened any imbalance from the sellers, among 
other reasons.  Some lessons can, nevertheless, be gleaned from what we understand to date.  Despite an 
uncommitted line of credit and about 10% of assets in cash, Third Avenue FCF was unable to meet shareholder 
redemptions.  As discussed further below, we believe that a three‐day liquid asset minimum or a portfolio position 
classification system would not have changed the outcome.  Third Avenue FCF highlights the need for thoughtful 
portfolio construction.  In addition, we believe in limiting a portfolio’s 15% standard assets to the 15% illiquidity 
guidelines.  See Landon Thomas, Jr., A New Focus on Liquidity After a Fund’s Collapse, New York Times (Jan. 11, 
2016), http://www.nytimes.com/2016/01/12/business/dealbook/a‐new‐focus‐on‐liquidity‐after‐a‐funds‐
collapse.html?_r=1 indicating 15% standard assets may have been more than 15% of the portfolio.  Finally, Third 
Avenue FCF is a case study on transition planning, including the ability to suspend redemptions as part of a 
liquidation plan, rather than liquidity risk.   

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 14
nor the DERA study15 provide evidence of a significant redemption risk. Similarly, the
Commission has not provided evidence that the prescriptive elements of the Proposed Rule
(namely, the proposed classification system and the three-day liquid asset minimum) may
likely be effective particularly in light of their implementation cost.
2. Liquidity Risk is a Fundamental Element of Investing.
The Commission cited liquidity risk as a purpose of the LRM program and defined it
as “the risk that the fund could not meet requests to redeem shares issued by the fund that
are expected under normal conditions, or are reasonably foreseeable under stressed
conditions, without materially affecting the fund’s net asset value.”16 We do not believe that
mitigating liquidity risk is an appropriate goal for rulemaking. Liquidity risk is a
fundamental element of investing and liquidity restrictions are not necessary for investor
protection. As the Commission has explained to shareholders, all investments involve risk
and an investor with higher risk tolerance has greater potential rewards.17 In fact, liquidity
risk is embedded in the concept of a mutual fund, which is the sharing and pooling of risks,
expenses, and rewards. The risk and rewards associated with liquidity risk is a reason why
the absence of or mitigated liquidity risk does not necessarily equate to a better, more
desirable fund for shareholders. The robust disclosure regime for open-end funds has
helped shareholders understand these concepts. We note also that asset managers are
incentivized to prudently manage liquidity risks (just as redemption risk) without additional
prescriptive regulations. We also question whether mitigation of liquidity risk and potential
dilution are part of the Commission’s statutory mandate. We have not found that Section
22(e) and rule 22c-1 under the 1940 Act or any other regulation support rulemaking that
avoids “materially affecting the fund’s net asset value” or the mitigation of dilution risk.18
As discussed, the inherent nature of investing in open-end funds includes affecting the
fund’s net asset value, i.e. the fund’s value will change.
3. Open-End Funds Do Not Pose Systemic Risks.
The open-end fund industry poses minimal risk to the stability of U.S. or global
financial markets, such that we do not believe prescriptive liquidity rules are justified on the
15

 Paul Hanouna, Jon Novak, Tim Riley, Christof Stahel, SEC Division of Economic and Risk Analysis, Liquidity and 
Flows of U.S. Mutual Funds (Sept. 2015) (“DERA study”).  We note that the DERA study appeared limited to equity 
funds and extrapolations about municipal funds. 

16
 Proposal at 103.  We support the definition without the phrase “materially affecting the fund’s net asset value.” 

17
 See Commission Office of Investor Education and Advocacy, Beginners’ Guide to Asset Allocation, Diversification, 
and Rebalancing, https://www.investor.gov/sites/default/files/Beginners‐Guide‐to‐Asset‐Allocation.pdf (last 
visited Jan. 11, 2016).  Through a diligent regulatory disclosure system, which has long been the crux of the federal 
securities laws, investors understand that liquidity risk is a component of investment performance.  See The 
Investor's Advocate: How the SEC Protects Investors, Maintains Market Integrity, and Facilitates Capital Formation, 
http://www.sec.gov/about/whatwedo.shtml (last visited Jan. 11, 2016). 

18
 See also, Simpson Thacher Registered Funds Alert, SEC Proposes Minimum Liquidity Requirement for Open‐End 
Funds; Raises Questions Regarding the Relationship Between Liquidity and Valuation (Nov. 2015), 
http://www.stblaw.com/docs/default‐source/Publications/registeredfundsalert_november2015.pdf (discussing 
the Commissions’ authority to enact liquidity requirements).  As the LRM program is not apparently intended to 
mitigate dilution risk, we address dilution risk further as part of the discussion of the swing pricing proposal. 
H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 15
basis of systemic risk concerns. There is neither historical precedent nor empirical evidence
that open-end funds pose a threat to the U.S. or global financial systems, even if they are
unable to meet shareholder redemption requirements. We incorporate the explanations and
data provided in the Invesco FSOC Comment Letter and by the Investment Company
Institute (“ICI”) and the Securities Industry and Financial Markets Association (“SIFMA”)
that demonstrate the lack of a systemic problem with mutual fund liquidity.19 It is also
crucial to understand whether any systemic risks are associated with a fund’s failure to
meet redemptions before adopting rules intended to curb systemic risk. It is important that
the Commission create rules that address a real problem, not FSOC’s conjectural problems.
4. The Commission Should Engage in Further Study and Analysis.
We ask that the Commission engage in further study and analysis to determine
whether there actually is a problem, what that problem may be, and how it might be best
addressed. After a period of study by the Commission of the existence of redemption and
systemic risks, data from the Disclosure Proposal, and the application of the LRM programs,
the Commission may find that the LRM program requirements should evolve. The UCIT
industry in Europe necessarily evolved its LRM program due to the realities of and changes
in the market. Our own liquidity risk management practices have also evolved with
changing strategies, available information, and experience.
The Commission cited alternative and high yield fixed income strategies as a source
of concern and stated the “[DERA] study found that alternative strategy mutual funds had
cash flows that were significantly more volatile than other strategies, indicating that these
funds may face higher levels of redemption risk.”20 We believe that may be an
oversimplification. The DERA study offered refutable analysis on the effect of cash flows
and volatility on levels of liquid assets, but did not link cash flows and volatility to
redemption risk or even substantial liquidity risk in the portfolio. Without additional
information, it is not sufficient to support requiring position classification or a three-day
liquid asset minimum. We support Commission efforts to study the comparative volatility
and behavior of varying asset classes, particularly in times of stress, and its direct effect on
redemption risk. We agree that high yield and alternative strategies may be subject to
more volatility when the markets generally are stressed; however, there are times when
they experience less volatility than the overall markets. For instance, during the interest
rate volatility from the Taper Tantrum, we found that bank loan funds performed well and
were relatively stable while investment grade funds under-performed. In mid-December
2015, we found that loans were relatively insulated from broader market volatility (although
not immune) because the defensive nature of the asset class, i.e., its senior position in the
capital structure, secured status, and short duration served as a form of capital protection.
The differences in behavior amongst asset classes are a crucial reason for diversification in
an investor’s portfolio and why investors do not look to every fund for near-term liquidity
needs. The Commission should study whether particular strategies or asset classes pose a
risk of not meeting shareholder redemptions that must be addressed and determine the
unique cause of that risk. If the Commission finds, with evidentiary support, that certain

19

 Invesco FSOC Comment Letter; ICI, Comments on Notice (Mar. 25, 2015) (“ICI Comment Letter”); SIFMA, 
Comments on Notice (Mar. 25, 2015); Chamber, Comments on Notice (Mar. 24, 2015).   

20
 Proposal at 29.   

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 16
strategies or asset classes pose systemic risk or material redemption risk, then we
recommend tailored rulemaking to reduce such risk.
C. Invesco Supports a Strong and Effective Liquidity Risk Management
Program and Certain Required Elements of that Program for the
Protection of Shareholders.
1. Invesco supports a baseline liquidity risk management program.
Invesco supports the requirement for each fund sponsor and adviser to have a
baseline LRM program. Our liquidity management practices, described above, are an
integral part of our investment management services and we consider it to be part of our
responsibility to strive to maintain adequate liquidity to meet investor expectations. We
also recognize that the Commission found certain asset managers had “substantially less
rigorous”21 liquidity risk management practices.22 The failure of a fund affects the
reputation of the industry and may affect funds in similar asset classes. As such, we
appreciate that the Commission seeks to enhance liquidity risk management practices to
ensure all asset managers incorporate “an enhanced minimum baseline requirement for
fund management of liquidity risk.”23 To aid the effort, we recommend every adviser
establish and maintain a principles-based, written LRM program.
Invesco recommends a requirement to implement a principles-based LRM program,
comparable to the compliance programs under Rule 38a-1, for each open-end fund24
reasonably designed to manage redemption risk. Funds need the flexibility to establish
reasonable methods to implement such a program without overly prescriptive requirements
in order to avoid frustrating investor choice and expectation, significantly weakening
performance, tightening liquidity supply, and homogenizing and damaging the industry.
The program should “provide[] fund complexes with flexibility so that each complex may
apply the rule in a manner best suited to its organization.”25 Additionally, “funds and
advisers are too varied in their operations for the rules to impose of [sic] a single set of
universally applicable required elements. Each adviser should adopt policies and procedures
that take into consideration the nature of that firm's operations.”26 For similar reasons,
UCITS are required to establish, implement, and maintain an adequate and documented risk
management process that includes identifying the risks to which the UCITS are exposed,
21

 Proposal at 39. 

22

 We do not believe poor performance or using the compliance function to monitor the 15% illiquidity guideline 
are indications that a fund was in danger of not meeting redemptions; therefore, we believe that the LRM program 
is not intended to ensure funds meet redemption requirements. 

23
 Proposal at 39. 

24
 We agree that unit investment trusts, interval funds, money market funds, and closed‐end funds should be 
excluded for the reasons that the Commission set forth in the Proposed Rule. 

25
 Compliance Programs of Investment Companies and Investment Advisers, Investment Company Act Release No. 
26,299, 68 Fed.Reg. 74,714 (Dec. 24, 2003) (“Compliance Rule”). 

26
 Id. 

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 17
such as liquidity risk.27 Therefore, we submit that the Commission’s compliance program
rules (Rule 38a-1 under the 1940 Act and Rule 206(4)-7 under the Investment Advisers Act
of 1940) can provide an adaptable philosophical framework and model for a rule requiring
an LRM program. The Commission should require that advisers establish, adopt, and
implement written LRM programs that are reasonably designed to prevent violations of
Section 22(e) of the 1940 Act.
The LRM program should be broadly required to address certain areas: the liquidity
of the markets in which the portfolio has significant exposure, the liquidity of the portfolio
assets, the portfolio’s strategy and disclosed risk appetite, historical flows, and sufficiency of
tools in normal and stressed conditions. We agree that the portfolio-level factors in
proposed rule 22e-4(b)(2)(iii) are appropriate considerations for a liquidity risk assessment
of a fund. We recommend using the broad factor that requires the fund to review the
liquidity of each position instead of a classification system. The factors should be
sufficiently flexible to apply to differing strategies, funds, markets and other circumstances.
Therefore, we recommend that the Commission provide the portfolio-level factors as
guidance for fund managers and that funds maintain records of the factors that were
considered.
2. Invesco Supports a 15% Standard Asset Maximum.
We also agree that a principles-based program alone may not be sufficient to deter
bad actors or prevent failing to meet redemption requirements. Invesco considers the
required 15% illiquidity maximum an important and instrumental element of the LRM
program. We support the prohibition of acquiring 15% standard assets if immediately after
the acquisition the fund would have invested more than 15% of its total or net assets in
15% standard assets. The 15% illiquidity guideline serves the important purpose of
reducing redemption risk, can be applied with reasonable effort, works within the realities of
the markets, and will not have the unintended, harmful effects of the prescriptive
classification and three-day liquid asset minimum. The asset management industry has
worked very well with the definition of illiquidity based on the 15% illiquidity guidelines for
years and finds it sufficient.
3. Boards of Directors Should Oversee But Not Be Required to Approve
Certain Aspects of the LRM Program.
The board of directors of a fund serves an important independent oversight purpose.
The board, including a majority of its independent directors, should be required to adopt
written policies and procedures reasonably designed to prevent violations of Section 22(e)
of the 1940 Act. We support the board’s initial and annual review of a fund’s LRM program,
including any material changes to the program. The policies and procedures should be
reviewed annually to determine their adequacy and effectiveness and should consider the
need for interim reviews for significant changes in the markets, funds, or regulations. An
LRM program should require periodic reporting and reporting of any breaches to the board.
However, we do not believe that the board should approve individual and specific
details of an LRM program, such as the three-day liquid asset minimum (should the
minimum be required). Managing liquidity risk requires investment expertise and the time
and ability to use investment data that the board and individual directors may not possess.
27

 AFLI Guidelines at 5 (citing EU Directive 2010/43/EU). 

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 18
The board’s role is oversight and does not encompass processing data or making investment
judgments necessary for prudent liquidity risk management. Also, the board cannot
logistically move as quickly and nimbly as may be necessary to respond to rapidly evolving
market conditions.
Independent scrutiny by the board at the level suggested by the Commission is
unnecessary. With respect to risk, as explained above, the interests of the adviser are
aligned with those of the shareholder. Additionally, requiring the board to approve a LRM
program that is reasonably designed to manage redemption risk would provide sufficient
independent oversight. There is not a need for the board members to make judgments
about specific elements of a program that is beyond their expertise.
4. Each Adviser Should Determine the Appropriate Administrator of a
Fund’s LRM Program.
The Commission should not dictate who should or should not administer a fund’s LRM
program. Invesco believes that portfolio management is in the best position to administer a
LRM program with support from GPMR, the compliance department and other back office
support functions. The administrator must have a deep understanding of the markets,
portfolios, assets, strategies, and the program, which are skills resident within the
investment function. Moreover, the administrator must be able to move swiftly at times of
stress to appropriately manage redemptions and the portfolio management teams will be
the first to know of or expect such situations. An administrator independent from or
divorced from portfolio management would cause the liquidity assessment to be
compartmentalized away from the investment process, which we believe could be
disadvantageous to the fund and investors. Furthermore, since the interest of the
investment team is generally aligned with the need to provide adequate liquidity, a potential
conflict of interest is not likely to be a concern. The administrator should report to the
board regarding the LRM program in the context of regular board reporting on the risk and
performance of the fund. We believe our portfolio management teams and their CIOs are
best suited for this task.
D. Invesco Opposes the Asset Level Classification System Because it is
Unworkable and Unhelpful in Reducing Either Liquidity or Redemption
Risk.
Assessing the liquidity of assets is a significant consideration in prudent liquidity risk
management, but classification is not a suitable method to evaluate position-level liquidity.
Invesco portfolio managers firmly believe that the classification of assets will not be useful
in the reality of liquidity management across a variety of strategies. Liquidity is a fluid,
relative concept dependent on numerous considerations that does not lend itself to a
classification system. Moreover, the classification system in the Proposed Rule is overly
burdensome and ineffective because (a) liquidity cannot be predicted with sufficient
precision; (b) the criteria used to classify assets is difficult to obtain, costly, or simply not
available; (c) consistency of classification amongst funds is highly unlikely and subjectivity
renders classification unhelpful to the Commission or shareholders; (d) the position
classification potentially exposes funds to unacceptable amounts of liability; (e) the
frequency of classifying portfolio asset liquidity is unreasonable; and (f) the significant labor
and cost required is not justified. We recommend a broad requirement to include the
liquidity of a portfolio asset as a factor in determining the overall liquidity of a fund.

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 19
1. Position Classification is Inappropriate Because Liquidity Cannot Be
Predicted with Precision.
Liquidity cannot be predicted with sufficient precision to classify all of the funds’
holdings appropriately. An infinite number of unpredictable, ever-changing factors affect
liquidity, such as sector conditions, market structure, or fund ownership, and even a simple
announcement about asset purchases from the Federal Reserve such as the announcement
that threw the market into a Taper Tantrum. It is possible that a liquidity analysis was
accurate in the morning, but became inaccurate and dated in the afternoon. Regardless of
how much data, expertise, and analytics are thrown at the question, that degree of
precision and clairvoyance is not possible. An asset manager does not know the liquidity of
an asset with precision until they have tried to sell the asset.
Additionally, in the fixed income space, without a comprehensive source of data,28
liquidity is challenging to measure. The classification system is particularly unhelpful for
fixed income funds, such as corporate bond funds. As the DERA study noted, fixed income
instruments require high quality data, which is not readily available, and there is not a
common measure of liquidity available.29 The liquidity measures for fixed income “are
typically more complex and tailored to the data available for each [asset] class….In addition,
the infrequent trading of many fixed-income securities can introduce both stale and
inaccurate measures of liquidity into the calculation of a fund’s bottom-up liquidity.”30
These challenges are, in part, why Invesco relies on the qualitative opinion of our
experienced and knowledgeable portfolio managers, whose judgment incorporates even
unapparent information. Portfolio managers know the securities in their portfolios
intimately and do not find the proposed classification to be useful in their management of
fund liquidity. To the contrary, our portfolio managers find the proposed classification
system arbitrary and a potential unnecessary constraint on their investment discretion.
2. Position Classification is Inappropriate Because the Criteria for
Classification are Not Realistic.
The criteria for classification (i.e., “the number of days within which the fund’s
position in a portfolio asset (or portions of a position in a particular asset) would be
convertible to cash at a price that does not materially affect the value of that asset
immediately prior to sale”) are unworkable. First, the condition of not moving the market is
not realistic. There is a cost to liquidity that shareholders deliberately bear, for example,
when an investor chooses an investment grade fixed income fund over a money market
fund and is accordingly compensated for the increased risk. Liquidity analysis must allow
for reasonable price movement. Also, the price of a large position in any asset class will
28

 Some fixed income data is generally provided voluntarily by market participants or trading venues, but there is 
not a comprehensive source as with securities that have central clearing. 

29
 DERA study at 31‐32. 

30
 Id. at 32.  The challenges in the fixed income space highlight another concern about the proposed classification 
system.  We do not believe the asset‐level factors will fit for all funds; funds need the ability to view the liquidity of 
portfolio‐level assets more specifically to that fund’s objective and strategy.  Asset‐level analysis should be 
dependent on the fund’s strategy, among other factors.  For instance, a senior loan fund may need different types 
and more in‐depth analysis for each asset than a large cap equity fund. 

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 20
move in a normal market environment; not permitting a price movement band would make
it potentially questionable and certainly more difficult to maintain a large position in any
security even if such position is in the best interest of the fund. For such reasons, a large
fund may be perceived by shareholders as less liquid than a smaller fund because a portion
of its larger holding in the same asset held by the smaller fund will have a less liquid
classification compared to the smaller fund, even if the large fund is actually more liquid at
the portfolio-level. Similarly, for certain assets, such as those that are thinly traded, any
position size may move the market for that security. Second, the phrase “materially
affecting” is subjective and thus does not provide a reliable and common measure for funds
to use. Finally, the asset-level factors are not always necessary, feasible, or reasonable
considerations in determining the liquidity of an asset.
3. Position Classification is Inappropriate Because Subjectivity Renders
Classification Unhelpful or Harmful and Consistency is Improbable.
Consistency of liquidity classification practices and analysis amongst funds is highly
unlikely and, we believe that, the inherent subjectivity renders classification unhelpful to the
Commission or shareholders and exposes funds to inappropriate potential liability. We
appreciate that the Commission “recognize[s], and anticipate[s], that different funds could
classify the liquidity of identical portfolio positions differently, depending on their
analysis….”31 There are several subjective elements to the classification, such as the
analysis of the factors, the inclusion or exclusion of applicable factors, and the
determination of the “materially affecting” standard, that will lead to variations in analysis.
We anticipate that, in addition to differing good faith analysis, some asset managers may
use less conservative classifications based on claims of broader access to markets or
expertise in trading. The subjectivity also gives aggressive fund complexes the opportunity
to abuse the subjectivity, further exacerbating the variations and unreliability of the
analysis. The inherent subjectivity required to classify fund holdings makes the
classification an unhelpful view of comparative liquidity for shareholders and the
Commission. We believe that the Commission and shareholders will, therefore, be unable to
use the classification system as a common measure of liquidity.
Additionally, the subjectivity and estimation (e.g., forward-looking assessments) also
exposes funds to inappropriate levels of potential liability, even when such determinations
and decisions are made in good faith. Funds need a high threshold of intent before any
liability attaches to the LRM program; specifically, liability should require that conduct be at
least reckless or with malicious intent. Only the Commission should be able to enforce the
LRM regulation. There should not be a private right of action.
4. Position Classification is Inappropriate Because the Frequency of
Classification is Unreasonable.
The frequency of classifying portfolio asset liquidity is unreasonable. Even under
normal market conditions, liquidity is a constraint on the investment process that managers
address on a daily basis, and in some funds on a trade-by-trade basis. Managers
continually analyze liquidity, cash positions, fund flows, and other aspects of their assets,
funds, and markets. Although portfolio managers can be aware of and consider asset
liquidity in real-time, they cannot classify the assets in real-time. Classifying assets each
time the liquidity of an asset changes is not practical because the liquidity of an asset
31

 Proposal at 67‐68. 

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 21
changes constantly. The Commission indicated that classification should occur at least on a
monthly basis, but even monthly is not practical because of the amount of time and effort
involved. Moreover, monthly classification data is likely to be stale and unhelpful. The
classification system will be a significant burden that fails to produce an important benefit.
5. Position Classification is Inappropriate Because the Significant Costs
are Not Justified.
Given that the classification system does not fit into the realities of asset
management and does not provide a discernible benefit, the extensive burden on the
adviser is unwarranted. Invesco has more than 63,000 portfolio positions that would need
to be monitored and updated. Assuming we can automate the classification process and
based solely on preliminary inquiries, we estimate more than $2 million in initial costs and
more than $650,000 in annual recurring costs. However, we do not believe that one can
fully automate the proposed classification system. Since a manual decision needs to be
made by an investment professional on each position based on the complex classification
system, the information would be labor-intensive and very difficult (if not impossible) to
fully automate. Certain assets, such as over-the-counter (OTC) instruments, or factors,
such as the quality of market participants, will need to be analyzed manually, requiring
substantial time and resources. Additionally, reviewing the relationship of one asset to
another, such as linking the liquidity of a derivative, the collateral covering the derivative,
and the hedged security, will be a significant challenge because it will require an onerous
transaction-by-transaction classification, which is a departure from current industry
practice. The individuals in the best position to determine the liquidity of the assets (the
portfolio managers) would spend substantial time classifying positions rather than managing
their funds. We would need to hire additional investment professionals with asset class
specialties as well as additional GPMR and compliance support individuals for that task
alone. None of these individuals would be as knowledgeable as the portfolio manager.
Furthermore, third-party service providers that provide certain asset-level data may require
additional fees to permit the use of data for classification purposes if the classifications are
ultimately disclosed to the public. The Commission’s estimate of costs is a gross
underestimate of expenses32 and investors will bear the impact of the higher costs.
We agree that third party services can be used to assist the adviser in liquidity risk
management and that the responsibility is ultimately with the adviser and the fund for the
efficacy of the program. However, it is too early to determine whether third party solutions
will be available, viable, or widely acceptable in the industry, if their analysis is sufficient for
the funds or acceptable to the Commission, and how costly this will be for fund
shareholders. The manual, labor-intensive analysis will probably not be supported by any
third party service providers. We are also mindful of the lessons learned in using third party
credit rating agencies.

32

 Also, the Commission cited a lack of information multiple times in determining what the costs may be.  See e.g. 
Proposal at 318 (citing lack of information as one of the reasons why the Commission is unable to estimate 
potential costs).   

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 22
6. Invesco Proposes a Broad-Based Requirement to Review Portfolio
Asset Liquidity.
Rather than using the proposed classification system, Invesco recommends that
funds be required to evaluate portfolio asset liquidity as one factor in portfolio-level liquidity
management. Each fund manager would determine how to best assess asset liquidity for
each fund. Such a requirement would achieve the need to include portfolio asset liquidity in
determining the overall liquidity of the fund and ensure that shareholders are not exposed
to avoidable redemption risk.
While we do not agree that a position classification system is meaningful and believe
it to be overly burdensome and potentially harmful, if the Commission is determined to
require a position classification system, we suggest the following, which we believe is similar
to the alternative classification system proposed by SIFMA and one of the alternatives
proposed by ICI. Funds could provide to the Commission (1) the percent of portfolio assets
that can be converted to cash in one business day under normal market conditions; (2) the
percent of portfolio assets that can be converted to cash within two or three business days
under normal market conditions; (3) the percentage of 15% standard assets and (4) all
other assets. The proposed factors for evaluating an asset’s liquidity is more appropriately
issued as elective guidance rather than mandatory considerations because not every asset
will require an in-depth analysis (e.g., treasury securities), and some factors are not
feasible to incorporate without intensive manual involvement.
E. Invesco Opposes the Three-Day Liquid Asset Minimum as Potentially
Harmful and Unhelpful in Reducing Liquidity or Redemption Risk.
We fear that the three-day liquid asset minimum will lead to several unintended
consequences as well as be ultimately unhelpful. The three-day liquid asset minimum may
unnecessarily cause fund performance to decline.33 A prohibition on acquiring assets if the
three-day liquid asset minimum is not met may cause a fund that cannot acquire favorable
three-day liquid assets to be in the difficult position of selling favorable, less liquid assets in
the portfolio or acquiring less favorable three-day liquid assets. It may also cause the fund
to lose a valuable, time-sensitive investment opportunity, which may be particularly
unnecessary during times of liquid markets. Also, without a specific exemption, an index
fund will not be able to track the performance of the index it follows as expected by the
shareholders and the market. In any of these circumstances, the result is a decline in fund
performance and a limit on the risk/reward ratio of the average investor. Such a result is
contrary to investor expectations. Individual investors depend on open-end funds to meet
their financial needs in a cost effective manner. Mutual funds and ETFs permit investors to
access assets with high performance potential that may be considered less liquid, such as
bank loans and high yield bonds, in a cost effective manner.
Moreover, the requirement will probably not be helpful or achieve its intended
purpose because it will either not be needed or will not be enough. The amount of liquid
assets needed is dynamic – it changes depending on the facts and circumstances and
changes in the fund, shareholders, and markets. Unfortunately, an asset manager cannot
predict the conditions of an extraordinary (unforeseeable) market. In an extraordinary
33

 We anticipate that some asset managers will be incentivized to increase leverage in portfolios in order to 
recover lost returns if they need to carry higher levels of cash and/or three‐day liquid assets.   

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 23
market, liquidity may be unpredictable, fast moving, and so tight that no particular amount
of one-day liquid assets may be enough. In an unforeseeable stressed environment (i.e.,
an extraordinary environment) where redemption risk is present, the three-day liquid asset
minimum will likely be insufficient. Third Avenue FCF reportedly had a cash position of at
least 10% of assets, which was shown to be insufficient.34 In any normal or foreseeable
stressed environment, the three-day liquid asset minimum is not needed since portfolio
managers determine adequate liquid asset level requirements daily based on current facts
and circumstances.
As discussed in the next section, we also believe that the three-day liquid asset
minimum may cause decreased diversification across funds with similar strategies,
tightened supply of assets, and loss of assets from the industry as investors seek
alternatives to open-end funds.
In sum, the three-day liquid asset minimum may diminish diversification in open-end
funds, unnecessarily weaken performance, further concentrate portfolios into certain assets
(tightening liquidity), cause funds to not meet their principal investment strategies or
closely track their index, and cause investors to seek alternatives to open-end funds. If the
Commission is determined to require a liquid asset minimum, Invesco recommends a
requirement to maintain a “target” of three-day and/or seven-day liquid portfolio assets
within a reasonably narrow range. These assets could be converted to cash within three
days and/or seven days under normal market conditions, as determined by the adviser. A
fund could opt to maintain a certain percentage range in three-day assets and another
percentage range in seven-day assets, or just a percentage range in three-day assets or
seven-day assets. The fund would not be prohibited from acquiring any asset if the
minimum is not met, but must acquire a three-day or seven-day liquid asset within a
reasonable amount of time of dropping below the minimum. For example, an equity fund
could determine to maintain three-day liquid assets between 1% and 3% at all times.35 The
Fund could go above 3% at any time, but if the three-day liquid assets drop below 1%, the
fund must make reasonable efforts to increase to 1%. A different type of equity fund could
determine to maintain 3% to 5% in seven-day liquid assets at all times. The Fund could go
above 5% at any time, but if the seven-day liquid assets drop below 3%, the fund must
make reasonable efforts to increase to 3%. Instances and explanations of dropping below
the minimum range must be reported to the board, after-the-fact, and at least on an annual
basis. We recommend that the targeted range be disclosed to shareholders. The range
would allow the fund the flexibility to adjust the minimum on changing market conditions,
such that the fund is not unnecessarily constrained in highly liquid markets. The range
would also allow enough diversification amongst funds to avoid some of the potential harms
discussed above and the industry is less likely to homogenize. The fund could consider the
factors as Commission-issued guidance, but be permitted to consider any current or nearterm considerations. Problems with the portfolio-level factors include the fact that forwardlooking information is often unreliable. We suggest an option of three days or seven days
since Section 22(e) of the 1940 Act requires seven days to meet redemptions, not three
days; however, we appreciate the Commission’s argument that the Commission required
funds to meet broker-dealer redemptions in three days.

34

 See Joe Morris, Extraordinary Outflows Buried Focused Credit, Ignites (Dec. 18, 2015). 

35

 The percentages are meant for illustration and are not indicative of potential actual percentages. 

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 24
F. The Position Classification and Three-Day Liquid Asset Minimum
Requirements of the Proposed LRM Program Will Not Likely be Effective,
Carefully Tailored, or Preserve the Core Features of an Open-End Fund.
We ask that the Commission ensure that any rulemaking (a) is effective in
accomplishing the goal of minimizing liquidity risk to the extent that it poses a threat to a
fund’s ability to meet redemptions; (b) is carefully tailored to address that goal; (c)
preserves the utility and core features of an open-end fund (such as investor choice and
risk-based performance) in order to avoid harming the industry; and (d) minimizes other
unintended, potentially destabilizing consequences. We are concerned that the prescriptive
elements of the proposed LRM program do not meet these criteria.
As discussed in the previous two sections, we believe that the position classification
and three-day liquid asset minimum elements of the liquidity risk management proposal are
unlikely to achieve the proposed goal of minimizing a threat to a fund’s ability to meet
redemptions. Third Avenue FCF is illustrative. Neither the proposed classification system
nor the three-day liquid asset minimum would likely have affected the outcome with Third
Avenue FCF. Regardless of whether the asset manager misestimated how much to hold in
cash or cash-equivalent assets, the amount was insufficient due to market conditions and
shareholder redemption pressure overwhelmed the holdings. It is unlikely that a
requirement to have a three-day liquid asset minimum would have changed the course of
events. Even if Third Avenue FCF had increased the percentage of cash held, it is unclear
what particular amount would have been sufficient given their portfolio holdings and
strategy, outflows, market conditions, and poor performance. We also concluded that a
position classification system as proposed would not have caused shareholders to have
acted differently. While we do not make a judgment as to whether FCF’s disclosures were
sufficient, appropriate, or met regulatory requirements, we do note that the fund discussed
its high concentration of distressed debt and illiquid assets and liquidity risk and provided its
schedule of assets. Third parties, particularly sophisticated third parties, had sufficient
information to identify its vulnerabilities and understand its risks.
It is our view that the position classification system and three-day liquid asset
minimum will detrimentally impact investor expectations and change certain core features
of mutual funds and ETFs, which will lead to potential unintended consequences, as
discussed below.
1. The Position Classification and Three-Day Liquid Asset Minimum Will
Likely Diminish Diversification Among Open-End Funds Investing in
Similar Strategies.
The position classification and three-day asset minimum will likely diminish
diversification among open-end funds investing in similar strategies. The Proposed Rule
may cause funds with similar strategies to buy the same assets and increase their
investments in the same “liquid” investments resulting in similar portfolio compositions
amongst the funds. As a result, investors will have fewer truly distinct fund options
available to them. Furthermore, mutual funds and ETFs will be unable to compete with less
expensive and outperforming non-registered products (such as collective trusts) with similar
strategies that do not have to operate under the same restrictions. Also, competition
among fund families will diminish as funds with similar strategies will appear identical, which
in turn may cause funds to close and assets to move away from the industry. As an
example of the unfavorable effects of Commission rules that change core features of mutual
H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 25
funds, we can consider money market reform. Due to the July 2014 money market reforms
instituted by the Commission that require implementing floating NAVs, redemption gates,
and liquidity fees on prime money market funds, there have been significant money market
fund strategy changes and conversions from prime to government funds.36 In addition,
several fund families have announced their intention to close their money market fund
business altogether, in part because of the money market reforms, further homogenizing
the money market industry.37
2. The Increasing Correlation of Investment Portfolios May Lead to a
Liquidity Crunch.
The increasing correlation of investment portfolios may lead to tightened market
liquidity. A requirement to possess a defined amount of three-day liquid assets will cause a
high demand in three-day liquid assets. Further, because every fund has specific
investment objectives and restrictions, only certain three-day liquid assets are available to
each fund. Just as the three-day liquid assets will be in higher demand, the less liquid
assets will become even less liquid. The consequences of the demand are multi-fold. The
asset management industry is a crucial player in the buy-side of our capital markets,
providing vital benefits to capital formation. There will be less capital available to small or
other businesses that rely on buyers such as mutual funds as investments are diverted to
three-day liquid assets. Additionally, if the entire asset management industry is required to
concentrate a percentage of all assets in the same set of limited assets, then it could have a
tightening effect on the market of those assets, leave other assets without a market, and
exacerbate volatility. Furthermore, if one were to assume that the herd behavior and fire
sale theories promulgated by FSOC were true, the prescriptive limits and parameters of the

36

 A recent JP Morgan securities analyst estimated that $600 billion to $650 billion of assets of prime money 
market funds could convert to government money market funds (42%‐45% of all prime funds).  We also 
understand that intermediaries and investors will move more assets closer to the major money market reforms 
implementation date (October 2016). 

37
 “Mergers and acquisitions trimmed the money funds industry from 75 providers in the United States last 
summer to just 67 this year, according to Crane Data, an industry research service…So many large players have 
resisted getting out of the business until now, but it's just a matter of the costs and uncertainty of money fund 
reforms proving to be overwhelming to some players."  Trevor Hunnicut, Blackrock to Buy Bank of America’s $87 
billion Money‐Market Fund Business, Reuters (Nov. 3, 2015),  
http://www.reuters.com/article/2015/11/03/us‐bank‐of‐america‐blackrock‐funds‐
idUSKCN0SS1HJ20151103#vAgQ0dSujY8j1cBr.97 (internal quotations omitted).  For example, William Blair Funds is 
liquidating its sole money market fund as of November 2015, see William Blair Funds, Supplement to Prospectus 
(May 1, 2015); Huntington Asset Advisers, Inc. and Reich & Tang sold their money market businesses to Federated 
Investors, Inc. with about $5 billion in combined AUM (see Federated Press Release, Federated Investors’ Money 
Market Funds to Acquire Approximately $1 Billion in Assets from Huntington Money Market Funds (Sept. 9, 2015) 
and Federated Press Release, Federated Investors, Inc. Completes Transition of Shareholder Assets from Reich & 
Tang Money Market Funds (July 28, 2015)); and BofA Global Capital Management sold its $87 billion money market 
mutual fund business to BlackRock, Inc. (see Bank of America Press Release, Bank of America Corporation agrees to 
transfer investment management responsibilities of BofA Global Capital Management to BlackRock, Inc. (Nov. 3, 
2015)).  The ICI reported that the share of prime versus government funds moved to 47% and 44% respectively 
(from 54% and 37% respectively).  Press Release, ICI Reports Money Market Fund Assets, Investment Company 
Institute (Dec. 3, 2015). 

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 26
Proposed Rule would cause funds to behave similarly (creating herd behavior),38 which
would cause tightened supply (liquidity crunch) and fire sales in the event of stress. As an
example, the capital requirements imposed on dealers have caused tightened supplies and
increased treasury volatility, among other things.39 Additionally, money market reform has
led to an increased demand for government securities, at a time when supply was already
limited.
3. Fund Behavior Will Likely Change Regardless of Whether a Fund had
Appropriate Liquidity, Leading to Negative Effects.
The Commission argued that certain negative effects, such as diminished
diversification, will not occur because fund changes will occur only “if a fund were to
determine it needs to adjust….”40 We disagree. We believe that the composition of
portfolios will likely change as a result of the prescriptive elements of the proposed LRM
program regardless of the liquidity profile of the fund. Funds will change because of
competitive pressure. We expect that asset managers with a propensity for
conservativeness will have an unnecessary large three-day asset minimum and competitive
pressure will cause other funds to maintain a similar three-day liquid asset minimum. Many
funds may feel they are required to change the portfolio composition in order to satisfy the
Commission’s interest in aligning the classifications and practices of the funds and
demonstrate the effectiveness of the rulemaking. If the prescriptive elements of the
proposed LRM program do change fund behavior, then the industry and investors will face
diminished fund diversity, reduced fund performance, and other potential, unintended
market liquidity problems.
G. Invesco Finds the Depth And Frequency of the Proposed Disclosure
Requirements Inappropriate.
The depth and frequency of the proposed disclosure requirements are inappropriate
because (a) we do not believe that position classification is needed for the Commission's
oversight purposes; (b) the information would be confusing to shareholders; and (c) funds
would be harmed due to the confidential, proprietary nature of the information. However,
38

 FSOC and FSB have continued to speculate (without any evidentiary support) that asset management products 
pose a threat to financial stability because of herding behavior by investors, but the asset management industry 
has not found that to be reflective of how the markets behave.  See Juliet Samuel, Runs on Investment Funds Could 
Pose Systemic Risk, UK Regulator Says, Wall Street Journal (Dec. 8, 2015).  If, however, one is to accept that herd 
behavior and fire sales are existent among asset managers, then the prescriptive elements of the LRM program 
would exacerbate such problems.  We continue to maintain that investors in the registered fund space do not 
engage in herd behavior and fire sales is a negligible risk. 

39
 See Joe Rennison, Liquidity Deteriorates for U.S. Treasuries, Financial Times (Nov. 23, 2015) (liquidity is already 
tightening from lack of dealer support stemming from more rigid capital requirements).  “Higher bank capital 
standards increased the capital cost of trading books and securities financing activities, reducing their return on 
equity.  The Volcker Rule requires banks and their affiliates to refrain from proprietary trading, although the rule 
does not prohibit market making.  These regulatory reforms addressed critical systemic vulnerabilities revealed by 
the financial crisis; they have also affected trading and funding liquidity in securities market.”  Department of 
Treasury, Office of Financial Research, Financial Stability Report (Dec. 15, 2015). 

40
 Proposal at 317. 

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 27
we think that information about each fund's LRM program and liquidity risks should be
described in such a way that shareholders will find it useful in understanding the anticipated
liquidity of the portfolio and the redemption risks.
1. Position Classification Disclosure is Not Needed for the Commission’s
Oversight Purposes.
We believe that the Commission’s oversight of funds’ liquidity risk can be best
handled by reviews of the LRM programs through the Commission’s Staff in Investment
Management and the National Exam Program and supplemented by data analysis.41 Such
data should be helpful in ensuring that the industry maintains minimal redemption risk or
systemic risk (to the extent any exists). The information proposed to be disclosed and as
proposed to be amended by Invesco on Form N-PORT should provide sufficient data for the
Commission to identify the portfolio assets and the assets’ liquidity.42 For example, the
Commission will have such sufficient data from (a) the schedule of portfolio assets and (b)
certain metrics for debt and derivative instruments (both, as described in the Disclosure
Proposal with proposed amendments).43 If needed, the Commission can use its own
analytics44 or third parties for analysis, which would also ensure uniformity in review,
protect funds from liability of differing and subjective analysis, avoid shareholder
misunderstanding, and avoid competitive disadvantages. While we anticipate that the
current and proposed data to be collected by the Commission through proposed Form NPORT will be sufficient for the Commission’s oversight purposes, the Commission could
require additional data after the opportunity to test and analyze initial sets of data.
2. Position Classification Disclosure Would Be Confusing to
Shareholders.
Classifying and disclosing each position would be confusing to shareholders. The
Commission has worked for years at simplifying information provided to shareholders and
eliminating unnecessary information, but the classification proposal is directly contrary to
that movement. Liquidity analysis is inherently complex and cannot be simplified into a
classification system without being confusing.45 Many retail investors are not sufficiently

41

 In fact, the Commission’s 2016 exam priorities include review of “liquidity controls,” which “will include a review 
of various controls in these firms’ expanded business areas, such as controls over market risk management, 
valuation, liquidity management, trading activity, and regulatory capital.”  Commission National Exam Program, 
Office of Compliance Inspections and Examinations, Examination Priorities for 2016 (Jan. 11, 2016). 

42
 We incorporate our comment letter on the Disclosure Proposal.  Invesco Comment Letter to the Disclosure 
Proposal (Aug. 11, 2015). 

43
 Id. 

44
 We assume that Commission has the ability to analyze the massive data requested to make it useful in its 
oversight function. 

45
 Even the Wall Street Journal overly simplified illiquidity, misapplied equity concepts to fixed income, and 
“produced results that we consider erroneous and misleading to investors.”  See Andrew H. O’Brien, Bond Funds: 
Solid Answers to Key Liquidity Questions (Sept. 24, 2015), 
https://blog.lordabbett.com/blog/2015/09/bond_funds_solid_answers_to_key_liquidity_questions.html 
H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 28
sophisticated to understand how the classifications were determined and what they mean,
why one fund company has a different classification than another for the same asset, or
how classification fits into the liquidity of the portfolio. Investors engage asset managers
because understanding the liquidity of a collection of assets is challenging and must be
understood in the context of other risks and considerations, such as issuer/quality and
performance. Also, the lack of uniformity and standardization associated with the
classifications makes the public disclosure of this item not helpful to investors at best, and
potentially confusing at worst. Furthermore, with a two month time lag, the liquidity
determinations will be stale and unreliable since liquidity can change even intraday. Finally,
the classification system will likely have the effect of creating an over-emphasis on the
liquidity of portions of an asset position when there are numerous considerations that must
be analyzed in determining the liquidity of an entire portfolio.
Public disclosure of the classifications could have the additional unintended
consequences of stifling robust liquidity risk management processes as well as causing fund
complexes to use boilerplate liquidity determinations, such as from third party services.
The burden and cost to abide by the classification rule could be too harsh to allow for
multiple processes and would lead to less thoughtful liquidity determinations. For example,
a fund could currently have robust processes, but abandon them to abide by a classification
rule that is less effective but required by regulation. Additionally, the Commission also
proposed that Form N-1A include disclosures about cash redemptions, but disclosing ways a
fund may redeem would inappropriately limit the fund in its methods to meet redemptions
to those that were previously disclosed. We request the flexibility to use any and all means
available, without being limited to what may have been disclosed under different market
environments. Also, it does not serve a purpose for investors to know precisely how the
fund meets their redemption requests – so long as they receive their redemption proceeds
within the period prescribed by regulation.
3. Position Classification is Confidential and Proprietary Information and
Disclosure Would Be Harmful to Funds.
Disclosing our liquidity determinations and the analysis of factors used would be
releasing proprietary information to the public and competitors and be substantially harmful
to funds and their advisers. The disclosure requests are so invasive and detailed that funds
may lose some of their competitive advantage. Both position/asset-level and portfolio-level
liquidity analytical determinations consist of confidential financial information that is
commercially valuable and used extensively in our asset management services – our
business. For these reasons, we do not customarily release such information. For example,
a fund’s view of liquidity of an asset is a crucial part of its trading strategy and directly
affects the cost of the asset. If this is disclosed publicly, trading counterparties may charge
fees and other costs based upon those disclosed liquidity parameters even if they hold a
more or less aggressive view. Furthermore, competitors can benefit from the fund’s
substantial labor by mimicking the fund’s determinations or taking advantage of the fund’s
publicly disclosed determinations when the competitor later trades in the same asset. In
addition, providing the credit agreement for a fund’s line of credit will be disclosing
proprietary and competitive information, even if fee information is redacted. We are also
concerned about the security of the sensitive information filed with the Commission due to
recent high-profile cybersecurity breaches both in the governmental and private sectors.
(discussing Matt Wirz and Tom McGinty’s September 21, 2015 article titled The New Bond Market: Some Funds Are 
Not as Liquid as They Appear). 
H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 29

4. Certain Disclosures Regarding the LRM Program Should Be Required.
It is important for investors to understand the liquidity and the redemption risks of a
fund and information about the LRM program may aid in that understanding. Therefore,
funds should disclose a narrative of the program with information that an investor would
find helpful in each fund’s statement of additional information. We also recommend that the
prospectus have a statement about the liquidity risk appetite of each fund. However, if
position classification is required in any form, we adamantly request that it not be disclosed
publicly.
H. We Ask that the Commission Consider the In-Kind Nature of Most ETF
Redemptions and Exclude ETFs from the LRM Program Requirements.
ETFs that transact in-kind are the majority of ETF assets and should be wholly
excluded from the LRM proposal. Although the Commission has provided guidance that
funds “assess the liquidity characteristics of an ETF’s underlying securities, as well as the
characteristics of the ETF shares themselves, in classifying an ETF’s liquidity,” we do not
believe that to be workable guidance given that the liquidity classification is based on time
to convert to cash and these assets are not converted to cash by the ETF. The classification
also disregards the exemptive orders under which the ETFs operate and expect redemptions
in-kind whenever possible to minimize liquidity problems (among other reasons).
Additionally, the Commission has noted in the Proposed Rule that redemptions in-kind
minimize the risks with which they are concerned. Following that logic, these funds do not
pose the risks that the Commission is trying to mitigate with the LRM program. Alternately,
if ETFs are not excluded from the scope of the Proposed Rule, we recommend a
classification bucket for assets that are redeemed in-kind.
We also request an exemption for index funds. The three-day liquid asset minimum
is particularly challenging to impossible for an index ETF to meet. The investment team
does not have discretion regarding the assets it includes in the portfolio, if they follow a
replication strategy. Furthermore, the exemptive orders require a 5% maximum tracking
error, which may be easily exceeded depending on the level of the three-day liquid assets
contained in a particular index or size of the fund. We also have concerns regarding the
potential adverse ramifications on the index providers’ strategies if the Proposed Rule is
passed in its current form. Moreover, there has not been a documented problem concerning
the liquidity of ETFs that redeem in-kind.
III. Invesco Requests Commission Assistance Regarding Interfund Lending,
Redemptions In-kind, and Centralized Trading to Improve Liquidity.
A. Invesco Requests Commission Relief Regarding Interfund Lending.
Invesco believes that interfund lending could be a more useful tool to an LRM
program in extraordinary circumstances if some restrictions are relaxed. For instance, it
would be helpful if an interfund loan could be outstanding for longer than seven days.
Currently, the loan may only be outstanding for the lesser of the time required to receive
payment for securities sold, or seven days. In an extraordinary, stressed environment,
seven days may be insufficient time (otherwise redemption risk would not be present).
Even if the seven-day requirement is changed, the loaning fund still has the right to demand
repayment at any time. In addition, funds would benefit from less restrictive terms
H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 30
triggered by an NAV decline or redemptions. Currently, if the borrowing fund has more than
10% in aggregate outstanding loans because of a decline in NAV, redemptions, or similar
reason, then within one business day, the fund must repay the interfund loan, reduce all
borrowings below 10%, or secure the interfund loan with 102% margin until it is below
10%.
B. Invesco Favors Redemptions In-kind and Requests Commission
Assistance.
Invesco considers the ability to redeem in-kind an important tool in the proper
management of open-end funds. Each fund that uses redemptions in-kind should have
written procedures. The primary problem with using redemptions in-kind to meet large
redemptions is the willingness and ability of the redeeming entity to receive securities
instead of cash. While there are times when the receiving entity is unable to handle the
transfer of assets, other times the entity is simply unwilling to do so. We welcome efforts
by the Commission to aid in facilitating funds’ ability to use redemptions in-kind with
entities that are able to receive securities. Also, it would be clearer to eliminate Rule 18f-1
in order to clarify that all funds may redeem in-kind or in cash.
C. Invesco Urges Creating Centralized Trading Platforms to Shorten
Settlement Periods.
Certain market-wide solutions may improve liquidity in the capital markets without
causing unintended harmful consequences. For example, Invesco supports initiatives which
would shorten the settlement period of fixed income instruments (such as leveraged loans)
through standardization of terms and the automation of the settlement process and other
evolving industry initiatives. Similarly, Invesco supports the creation of centralized fixed
income trading platforms and standardization of fixed income issuance. It was not long ago
that equity securities settled on a T+7 basis rather than today’s T+3 standard and initiatives
are underway to shorten that time to T+2. Many futures contracts and derivative
instruments settle on a T+1 basis. A reduction in the settlement time frame of investment
instruments would increase the robustness of the capital markets generally while at the
same time mitigating liquidity risks.
IV. Comments on Swing Pricing Proposal
In theory, Invesco supports swing pricing as a tool that may help in mitigating the
potential dilution of fund shareholders due to large share purchases or redemptions.
However, we believe that there are fundamental issues and questions that must be
addressed prior to effective implementation of swing pricing in the U.S.
A. Partial Swing Pricing May Help in Mitigating Potential Dilution of Fund
Shareholders.
Currently, on those occasions when a fund incurs transaction costs to meet
shareholder subscriptions or redemptions, the subscribing or redeeming shareholders do not
bear those costs; rather, the long-term shareholders do. Swing pricing endeavors to shelter
those long-term shareholders from the impact of the actual cost of purchasing or selling the
underlying assets of a fund when those costs deviate from the carrying value of these
assets in the funds’ valuation due to commission charges, taxes, and any spread between
the buying and selling prices of the underlying assets (i.e. dilution). We have found partial
H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 31
swing pricing successful in meeting its stated objective in the Invesco UCITS, which have
used swing pricing since at least 2007.46 In measuring dilution since swing pricing was
introduced, we have generally found levying trading costs through an appropriate swing
factor has saved the funds notable transaction costs.47 Although certain investors who
redeem or purchase may have an extra price burden or windfall based on the actions of
other shareholders, our experience suggests that the value to the remaining shareholders
outweighs this concern. We are concerned that there may be investors who would prefer
not to use a vehicle with the potential for swing pricing, but we believe those would be a
minority of investors and the impact to the industry will be manageable.
While we concur that partial swing pricing would mitigate dilution, we have not found
that Section 22(e) and rule 22c-1 under the 1940 Act or any other regulation support
rulemaking that lessens dilution. Swing pricing, while useful in minimizing dilution, does not
affect the ability to meet redemptions. Additionally, we have observed that asset managers
already have a strong interest in minimizing dilution and thus take actions to reduce dilution
with tools currently available. Finally, we do not believe swing pricing is necessary for
investor protection, particularly as it is a disclosed cost of a fund’s liquidity.
B. We Identified Three Fundamental Issues with Partial Swing Pricing: It
Must Be Mandatory, Funds Must Receive Investor Flows in Sufficient
Time; and There Must Be a Safe Harbor from Potential Liability.
1. First And Foremost, Operationally Implementing Partial Swing Pricing
for U.S. Mutual Funds Is Not Feasible.
Operationally implementing swing pricing for U.S. mutual funds is not currently
feasible. As the Commission has noted, “swing pricing requires the net cash flows for a
fund to be known.”48 Our experience suggests that reasonably accurate net flow
information must be available to calculate the swing threshold sufficiently in advance of
calculating the swing factor and the net asset value (NAV). First, estimates based on
interim feeds received from the transfer agent or distributor prior to the NAV calculation are
insufficient. The net flow of a fund can vary wildly intraday making an estimate unreliable.
Additionally, a significant portion of transaction information comes in after the NAV is
calculated. Of the omnibus activity that we receive for the mutual funds (generally between
50%-80% of the daily volume), we ascertained that: about 23% of omnibus trades are
received before 3 p.m. CT; about 40% of trades are received between 3 p.m. CT and
7:30 p.m. CT; and 37% are received the next day at 5:30 a.m. CT. We receive the trade
information too late to be able to use partial or full swing pricing. Second, the application of
swing pricing and calculation of the swing threshold should be based on quantitative,
46

 Invesco opposes the ability to use full swing pricing.  Full swing pricing may introduce significant price volatility 
and be entirely unnecessary when funds do not incur trading costs to meet redemptions; partial swing pricing can 
meet the stated objective without introducing material price volatility or creating other material ancillary 
problems.  

47
 For instance, one Invesco UCIT had about $18 million in total dilution since implementing swing pricing resulting 
from approximately 400 swing events.  However, this fund was on the high side of the curve.  Several other funds 
had few to no swing events and/or very little dilution. 

48
 Proposal at 238.  Flow information is required for either partial or full swing pricing. 

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 32
automatable data, rather than unreliable estimates in order to alleviate potential errors and
certain liability concerns. Third, the consequence of wide-ranging estimates could result in
swinging the price the wrong way. Contrary to the U.S. mutual funds, Invesco UCITS
receive net flow information on a timely and reasonably accurate basis. These funds receive
transaction requests by noon local time and calculate net flows by 2 p.m. The NAV is
calculated at approximately 4 p.m. local time the same day. Therefore, the NAV is
calculated at a moment where the transaction volumes are certain.
The current regulatory requirements and industry practices preclude a fund from
receiving timely net fund flows. Accurate redemption and purchase information is not
received until after the net asset value is calculated each day. Each fund’s daily net asset
value is calculated at 4 p.m. (industry standard); however, shareholder transaction
information is received by the fund up until 10 a.m. the following day, as described above.
Furthermore, shareholders are required to receive the next calculated NAV after receipt of a
redemption or purchase order49 and industry practice allows intermediaries to submit orders
that that they have received from clients before 4:00 p.m. to the funds until 10 a.m. the
next day.
We request that the Commission create a regulatory obligation that intermediaries
provide trade information to fund sponsors on a time-table that allows all funds to use swing
pricing. Establishing an earlier cut-off for intermediaries to send transaction data would
resolve the operational challenge by assisting the funds in receiving flow information in
sufficient time and be aligned with global practices. The solution would be a fundamental
change for intermediaries and require significant system upgrades across the industry (both
for the fund industry and our intermediaries). The industry and our intermediaries are
unlikely to make these changes voluntarily. If the Commission permitted swing pricing
without enabling a solution to the operational challenges, then funds with shareholders who
invest through intermediaries could not reliably use swing pricing and it would create an
unfair advantage for funds that do not have shareholders through omnibus accounts,
wirehouses, or other such platforms that provide transaction information after the NAV
calculation.
2. Partial swing pricing must be mandatory across open-end mutual
funds.
Partial swing pricing must be mandatory across open-end mutual funds if it is to be
used effectively.50 Swing pricing will require a fund to take competitive and liability risks,
incur significant financial expenses to upgrade systems, make substantial changes to the
way of doing business, and cause business partners to also incur costs. Those funds that
choose not to implement partial swing pricing will have a competitive advantage,
particularly in short-term performance. While partial swing pricing has benefits to
shareholders, it may not outweigh these costs to most funds and will be a tool left unused.

49

 See Rule 22c‐1(a) of the 1940 Act. 

50

 For similar reasons, Invesco agrees that once the swing threshold is met, a fund must apply the swing factor.  
Making implementation optional would enable gaming and permit conflicts of interest.   
H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 33
3. Finally, Potential Liabilities Associated with Swing Pricing Are Too
Great to Bear.
The third fundamental concern is the potential liability associated with swing pricing.
The consequence and likelihood of making a decision that is not favorable to a particular
shareholder significantly discourages the use of swing pricing. The following scenarios pose
significant concerns: (a) individual shareholders who paid a higher price or sold at a lower
price than NAV because they were caught on a swing day; (b) good faith market impact
estimates used in determining the swing factor and net flow estimates used in calculating
the swing threshold that are later found to be inaccurate; (c) the determination to use
swing pricing when similar funds chose to or not to use swing pricing; and (d) information
received after implementing the swing factor that would have had a material effect on the
adjusted NAV. We believe that scenario (b) can be remedied by eliminating the use of
market impact from the calculation of the swing factor and (c) can be remedied by making
swing pricing required for all funds. The remaining scenarios, however, are inevitable
incidents that will occur without unreasonable action or error by the fund and/or the fund’s
administrator. We do not support swing pricing without safe harbors from liability in these
scenarios.
Mutual funds also need safe harbors based on non-material error or clear error (i.e.,
in the absence of recklessness or malicious intent) and a statement that there is not a
private right of action. The Proposed Rule, as drafted, exposes a fund and the fund’s
administrator to numerous other potential liabilities because of the onerous administrative
burden and the significant number of estimates and calculations that must be performed in
a short timeframe. Common industry and regulatory practice in Europe is to report
immaterial NAV errors and report and compensate for material NAV errors only. We agree
with that approach.
C. The Calculation of the Swing Factor Should Be Quantitative,
Automatable, and Transparent.
The calculation of the swing factor is an important element of the Proposal, so it
should be quantitative, automatable, and transparent.51 To decrease miscalculations from
manual calculation, subjectivity, and lack of time and to avoid potential conflicts of interest,
the calculation of the swing factor should be based on actual amounts and reasonable
estimates. To that end, Invesco supports the calculation of swing factor that includes
transaction fees and charges, under the proposed definition, i.e. brokerage commissions,
custody fees, and any other charges, fees, and taxes associated with portfolio purchases
and sales, and spread costs. These costs can be calculated using actual costs and
reasonable estimates.
Market impact and other amounts that the fund or fund administrator finds cannot be
reasonably estimated must be excluded from calculation or be permitted to be excluded

51

 The swing factor should be employed regardless of whether it is a downswing or an upswing.  Otherwise, the 
shareholders will not benefit from the use of their investment while purchasing, but must share in the cost of 
redemptions.  We have learned that it has been beneficial to swing either way in the Invesco UCITS. 

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 34
from calculation. We do not believe that “market impact”52 is an appropriate consideration
in the swing factor because it is subjective, difficult to quantify, and unpredictable;
therefore, a reasonable estimate is unattainable. Market impact can generally only be
determined after securities are traded. Invesco UCITS have not used a similar market
impact concept in calculations while successfully mitigating shareholder dilution; market
impact has not been found to be an important factor in meeting the objective. Additionally,
a fund should be able to determine that a cost cannot be reasonably estimated and thus
excluded from the swing factor calculation. For instance, if a foreign tax cannot be
reasonably estimated prior to the NAV publication, it should not be included in the
calculation. We would be willing to include market impact if we were granted a safe harbor
for using good faith estimates.
Calculating the swing factor using “information about the value of assets purchased
or sold by the fund to satisfy net purchases or net redemptions that occur on the day the
swing factor is used to adjust the fund’s NAV (if that information would not be reflected in
the current NAV of the fund computed on that day)”53 is reasonable. Unfortunately, such
information will not usually be readily available in the event that a fund does not receive
advance notice of shareholder activity and is required to engage in securities transactions
that day. A fund may only have minutes to determine and apply the swing factor. Funds
should endeavor to include such information if readily available or reasonably estimable at
the time of determining the swing factor, but not be required to do so. Invesco UCITS have
found that using actuals at the time of determining and applying the swing factor is
generally not possible. Instead, Invesco UCITS use a third party (overseen by Invesco
UCITS) to calculate the swing factor on a monthly basis, based on current market spreads
plus annualized average transactional charges incurred by the fund. The swing factor may
be re-calculated if needed to account for changing conditions.
D. Invesco Supports Certain Disclosures and Using the Unadjusted NAV in
Performance and Financial Statement Calculations.
The Proposal requests comment on several specific issues regarding swing pricing
financial statement disclosure considerations. First, we agree that the swing factor should
not be disclosed, except for the voluntary disclosure of the maximum swing factor, and the
swing threshold should not be disclosed. Disclosure of the swing factor or swing threshold
will enable gaming by large investors who could divide their trade to just below the
threshold amount, cause confusion to shareholders, and increase potential liability. We
support disclosing the swing price policy and swing price in the footnotes to the financial
statements.
We also believe that funds should be required to use the unadjusted NAV in their
performance calculations. Use of the adjusted NAV In total return calculations would not
present an accurate view of the fund’s performance because the swing price impacts only
those who purchased or sold on an adjusted NAV, and is not indicative of the performance
for the other shareholders in the fund. Performance calculated based on adjusted NAV lacks
comparability to the results of performance of other like managed strategies and introduces
52

 “Market impact cost cannot be calculated directly.  It can be roughly estimated by comparing the actual price at 
which a trade was executed to prices that were present in the market at or near the time of the trade.”  Proposal 
at 188, n.415. 

53
 Proposal at 221. 
H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 35
volatility in the performance that is misleading to shareholders that did not transact at the
adjusted NAV. For example, if a fund had an unadjusted NAV of $10.00 and adjusted NAV
as a result of net redemptions at the beginning of the period of $9.90 and adjusted NAV at
the end of the period of $10.10 and the unadjusted NAV remained constant the fund would
show a greater than 20% return when the actual return for all investors that did not trade
on the adjusted NAVs was 0%. Additionally, using the unadjusted NAV for performance
calculations promotes unfair competition for those funds and/or inappropriate swing
thresholds. Furthermore, shareholders will be able to estimate the swing factor on an index
fund by the tracking error. For example, if a fund had a lower price (swung price) at the
start of the period and then a larger price at the end of the period (swung price), it is a wide
spread for that particular shareholder’s return that is not an indication of the return for any
other shareholder in the fund. Therefore, the unadjusted NAV should be required for
standardized performance reporting. We believe that any required disclosure of
performance based on adjusted NAV should be disclosed in a format similar to disclosures of
after tax returns.
We believe that the financial statements, including the statement of assets and
liabilities, statement of operations, statements of changes and financial highlights should
reflect the actual result of operations and should not include the hypothetical results of an
adjusted NAV. For similar reasons cited above for standardized performance, we believe
that it is inappropriate for the financial statements to include the adjusted NAV in the
balance sheet and financial highlights. Funds should be required to disclose only the
unadjusted NAV in the financial statements, including the balance sheet and statement of
assets and liabilities and the financial highlights. Financial statements should reflect the
results of operations for the historical time period presented and should match the actual
performance and total return of the fund. The financial highlights are intended to show the
results of a per share investment held at the end of the period. We believe that the use of
the adjusted NAV at the beginning and end of period is inconsistent with the purpose of the
table. Financial statements should include the actual amounts of transactions executed at
adjusted NAV. Therefore, we agree that the dollar amount of purchases and redemptions
disclosed in a fund’s financial statements should be presented as the actual value received
by the fund or paid to shareholders (including the impact of swing pricing) that the
aggregate dollar amount of purchases and redemptions will also be reflected in the
statement of changes in net assets.
We do, however, support disclosing the impact of adjusted NAV shareholder activity
in financial highlights. It is appropriate for funds using swing pricing to report actual
performance versus hypothetical performance, and funds should be permitted to optionally
disclose the adjusted NAV in addition to the unadjusted NAV where appropriate. Since the
amounts received/retained by the fund as a result of implementing swing pricing are
intended to defray dilution from deploying or raising cash from shareholder activity we
believe the per share impact should be included in the results of operations section of the
financial highlights table.
E. ETFs Should Be Permitted to Use Swing Pricing or the 2% Variable Fee
Cap Should Be Lifted.
The Commission has proposed that partial swing pricing be permitted only for openend mutual funds and not for exchange-traded funds (ETFs). We agree that authorized
participants (APs) most often redeem in-kind, which avoids most shareholder dilution
concerns. We also agree that the variable transaction fee imposed to offset transaction
H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 36
costs should address dilution concerns when APs redeem in cash; however, the variable
transaction fee is capped at 2% under the terms of the ETF’s exemptive order. Where the
actual costs of execution and slippage exceed 2%, the redeeming ETF shareholders should
be asked to bear those costs for the same reasons open-end mutual funds should use swing
pricing. Furthermore, just as the Commission has appropriately not prescribed an upper
limit for the swing factor (“on account of the difficulty of establishing an appropriate acrossthe-board limit,” having “no wish to limit the extent to which swing pricing could mitigate
the dilution of existing shareholders,” and an upper limit “would not result in appropriately
high NAV adjustments” due to the safeguards in place),54 ETFs should not have a cap on the
variable transaction fee. Therefore, it is essential that ETFs be permitted (as opposed to
mandated) to use swing pricing or the 2% cap must be lifted. Invesco supports lifting the
2% cap and requiring swing pricing for open-end mutual funds, but not ETFs.
F. Each Fund’s Board of Directors Should Oversee the Swing Pricing
Program, but Not Be Required to Approve the Swing Factor or Threshold.
Invesco supports the funds’ boards of directors understanding and approving swing
pricing policies and procedures. The determination of the swing factor and threshold, on the
other hand, are clearly management decisions. It is inappropriate to substitute the
expertise and experience of a fund’s servicers for that of the fund directors. In Europe, the
relevant boards of directors approve the swing pricing policy, but not the swing threshold or
factor, for similar reasons.
V. Compliance Dates
Invesco requests an extended compliance date for the LRM program. The compliance
date should be 30 months for all funds, not just smaller complexes. Larger complexes will
likely have more challenges because of the tailoring and specificity of the program to each
fund, the larger range and diversity of funds, and larger number of individual securities
held. An adviser will not be able to use economies of scale. If position classifications are
required, then 30 months may not be sufficient time. Also if the Commission institutes a
number of other regulations, such as on derivatives and disclosures, around the same
timeframe, 36 months will be insufficient.
If (a) the Commission mandates swing pricing, (b) it can be operationally
implemented, and (c) safe harbors are provided, we request an industry-wide compliance
date. Considering the requirement of system upgrades, effects on third party relationships,
and policies and procedures changes, we believe that 24 months is a reasonable compliance
date.
VI. Conclusion
We appreciate and agree with the Commission’s goals to reduce redemption risk and
promote effective liquidity risk management practices. We believe that Invesco’s liquidity
risk management practices minimize redemption risk and appropriately manage liquidity
and dilution risks. We agree that all funds should have a liquidity risk management
program. An LRM program, along with the requirements to maintain below 15% standard

54

 Proposal at 226. 

H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16

Mr. Brent Fields
January 13, 2016
Page 37
assets and a minimum amount of three-day and/or seven-day liquid assets is sufficient to
effectively obtain those goals.
Position classification, however, does not benefit shareholders as a stand-alone
requirement or as part of an LRM program. Moreover, position classification carries
extraordinary costs and burdens to the funds. The classification system in the Proposed
Rule is overly burdensome and ineffective because (a) liquidity cannot be predicted with
sufficient precision; (b) the criteria used to classify assets is difficult to obtain, costly, or
simply not available; (c) consistency of classification amongst funds is highly unlikely and
subjectivity renders classification unhelpful to the Commission or shareholders; (d) the
position classification potentially exposes funds to unacceptable amounts of liability; (e) the
frequency of classifying portfolio asset liquidity is unreasonable; and (f) the significant labor
and cost required is not justified. We strongly urge the Commission to not require position
classification.
The three-day liquid asset minimum may diminish diversification in open-end funds,
unnecessarily weaken performance, further concentrate portfolios into certain assets
(tightening liquidity), cause funds to not meet their principal investment strategies or
closely track their index, and cause investors to seek alternatives to open-end funds. To
alleviate these concerns, Invesco recommends a requirement to maintain a “target” of
three-day and/or seven-day liquid portfolio assets within a reasonably narrow range.
Invesco is in favor of partial swing pricing, but requests that the Commission
facilitate resolution of the operational obstacles to its implementation, require that it be
mandatory, and provide certain safe harbors.
Invesco is a member of various associations that are submitting comment letters
addressing in more detail public policy considerations similar to those expressed herein.
These associations include the ICI, SIFMA, and Loan Syndications & Trading Association
(“LSTA”). 55 Based on drafts reviewed prior to submission and except in those very limited
circumstances this Comment Letter advocates a different view, Invesco endorses the
comments expressed in each of the ICI, SIFMA, and LSTA letters. Our overall experience as
an asset manager is consistent with the observations made in those letters.

55

   ICI, Comments on Proposed Rule (Jan. 13, 2015) (“ICI Comment Letter”); SIFMA, Comments on Proposed Rule 
(Jan. 13, 2015); LSTA, Comments on Proposed Rule (Jan. 13, 2015).  Invesco has had the opportunity to review 
each comment letter in draft form prior to its submission to the Commission; therefore, our endorsing comments 
are based upon our review of such pre‐submission drafts. 
H:\Documents\Zerr\SEC\Liquidity Comment Letter 2016.01.13.doc
13-Jan-16