WhiteCapability
MFS
Paper Series
Focus
Month
May 2016*
2012
®
Authors
CAPACITY MANAGEMENT:
A KEY RISK MEASURE
IN BRIEF
Michael T. Cantara, CFA
Head of Global Client Group
•C
apacity management is an integral component of the
investment process.
• P reserving alpha-generating capability for clients is an overarching
guiding principle in the review of capacity at MFS®.
•C
apacity is considered under the rubric of risk and evaluated in a
systematic semiannual process along with other portfolio risks.
Joseph C. Flaherty Jr.
Chief Investment Risk Officer
• T he firm’s approach to capacity management includes quantitative
analysis and qualitative assessments as well as additional non-portfolio
considerations.
• T he steps taken to manage the growth of assets in MFS’ Global Equity
strategy since 2006 present a clear case study on capacity management
decisions.
John E. Stocks, CFA
Quantitative Research Analyst
Ravi B. Venkataraman, CFA
Global Head of Consultant Relations
Success in asset management leads to a well-known conundrum:
how to best manage increases in assets under management (AUM)
while continuing to generate value-added alpha for clients.
A growth in AUM may make it more difficult to implement a strategy by imposing
certain costs and impediments, such as liquidity constraints, potentially higher
transaction costs and client-servicing requirements, as well as the need to ensure
adherence to the strategy. Business diversification across multiple strategies and
investment teams is another important consideration when evaluating the capacity
of individual investment strategies. This is key to ensuring a sustainable business
model, which in turn impacts individual product performance.
Capacity can be managed in various ways, including with the implementation of
product closures, which are designed to protect the interests of existing clients by
limiting further inflows. In this case, we are referring to both clients in the strategy
of interest as well as clients more broadly in strategies that may overlap with the
* A version of this paper was first published in May 2014.
MAY 2016 / CAPACITY MANAGEMENT
product in question. Preserving alpha-generating capability for
existing clients is paramount in our view, and is at the heart
of capacity management. It is in this context that capacity
management is viewed as an integral component of risk
management at MFS.
the MFS risk review process. In addition, capacity is examined
at the firm level on a periodic basis. Product risk profiles and
investment style are monitored on an ongoing basis to
ascertain whether a growing asset base is leading to shifts in
a product’s risk profile or style. Trading patterns and
transaction costs are also monitored to determine whether
rising assets are making a strategy more difficult and
expensive to implement.
While there is general agreement among asset managers and
their clients that products need to be closed for capacity
reasons, there is little consensus on how capacity should be
measured. Various academic and industry studies have offered
a number of quantitative tools to help determine product
capacity; however, these approaches are sensitive to the
underlying assumptions made and none is definitive. Not only
is capacity hard to measure, it is also a function of current
market conditions and the characteristics of a given strategy.
Exhibit 1: MFS’ approach to capacity management
Non-portfolio
considerations
Certain asset classes are inherently more capacity constrained
than others. Portfolios invested in large-cap US equities have
significantly more capacity than portfolios invested in either
small-cap or emerging market equities. Highly concentrated
portfolios (e.g., 20-stock portfolios) generally have less
capacity than more diversified portfolios, depending on the
liquidity of stocks included in the portfolio. Portfolios with
high turnover require greater market liquidity and therefore
have less capacity than portfolios with lower turnover that
can patiently trade over longer holding periods. One should
bear in mind that all these parameters interact with one
another and market conditions change over time, so portfolio
characteristics must be fully examined before applying
generalizations about capacity.
Quantitative
model
CAPACITY
REVIEW
Qualitative
assessment
Guiding principles
The firm’s approach to capacity management is guided by
these important principles:
Protecting clients’ interests – Capacity management is
an integral part of MFS’ client-centered strategy. We are
committed to continually monitoring and prudently
managing product capacity consistent with the firm’s
commitment to maintaining product integrity and adding
sustainable value for our clients. At the core of the firm’s
strategy is a commitment to ensuring that existing clients are
not adversely impacted by the rising costs and declining
performance that can result from capacity constraints.
In this paper, we outline MFS’ approach to managing capacity
in equity portfolios in some detail to illustrate the firm’s
philosophy and considered methodology. We also provide
information on the product restriction decisions made with
regards to the Global Equity strategy as a case study. The
way capacity is considered from a fixed-income portfolio
perspective is addressed briefly on page 8.
Aligning compensation – Portfolio managers at MFS are not
MFS’ approach to capacity management
paid based on AUM but rather on their contribution to the
firm, viewed through the prism of long-term investment
performance and collaboration with colleagues. This ensures
that the portfolio manager incentive structure is aligned with
clients’ interests and the long-term performance of portfolios.
This alignment of interests is an important principle in our
view as it ensures that portfolio managers are not incentivized
to take on more assets than may be prudent, but rather
the reverse.
MFS employs a combination of quantitative analysis and
qualitative assessments to help measure and manage capacity,
and also takes other non-portfolio considerations into
account such as idea generation, business diversification and
portfolio manager non-investment responsibilities (see Exhibit
1). Discussions regarding capacity focused on the individual
strategy level take place formally every six months as part of
2
MAY 2016 / CAPACITY MANAGEMENT
Considering capacity management a risk measure – At MFS,
The two primary factors we use to set ownership limits at the
security level are:
capacity management is evaluated under the rubric of risk and,
as such, is a component of the semiannual risk review process,
which involves an in-depth examination of portfolio risks with
senior management. Capacity management is reviewed in this
context along with other risk measures. The semiannual risk
review ensures there is a systematic process to monitor and
measure capacity for every equity strategy at MFS.
•percentage ownership of average days trading volume (ADV)
•percentage ownership of the shares outstanding of
a company
Capacity model detailed
The capacity model includes the following steps:
Preserving alpha-generating capability for existing
clients is paramount in our view, and is at the heart
of capacity management. It is in this context that
capacity management is viewed as an integral
component of risk management at MFS.
1. Determine security/issuer level ownership constraints
for the entire firm for all securities held within the
portfolio being measured
Quantitative capacity model
• Percentage of shares outstanding = MFS aggregate
shares/total shares outstanding
Ownership constraints are set by the percentage of
shares outstanding and the percentage of average days
trading volume (ADV).
The capacity management process begins with a quantitative
framework, in which product capacity forecasts are generated
based on factors such as share ownership and trading
volume, and considering the holdings overlap across multiple
MFS products. Sensitivity analysis is performed around all of
the assumptions inherent in the quantitative analysis to
generate a multidimensional grid of capacity estimates.
The calculation of the percentage of shares outstanding
takes place at the equity issuer level and accounts for
multiple listings, share classes, ADRs/GDRs, etc. We model
this limit at a 10% threshold for all securities, although
we may own up to approximately 15% in some stocks.
• Percentage of ADV = MFS aggregate shares/90-day
average trading volume
The basis of the capacity model is to establish the constraints
on replicating an existing portfolio at increasing levels of
AUM, i.e., determining the aggregate assets that can be
managed in the strategy as it is positioned at the time. In the
model, we do not attempt to adjust for market appreciation,
make portfolio positioning assumptions or account for how
managers might react in the face of capacity constraints.
We use three base scenarios to assess the liquidity of
ownership: 5 days, 10 days and 15 days of ADV. The
number of days trading in this calculation refers to the
total shares owned over the period at 100% of volume.1,2
2. Calculate the number of incremental shares available
to MFS
For each security in the portfolio, we calculate the
difference between the aggregate shares currently
owned by MFS and the total shares set at each ownership
constraint level defined above. We call this the incremental
shares available for purchase.
Ownership limits are initially set for the overall organization at
the security and issuer level. As a multistrategy firm with
cross-ownership among strategies, exposure needs to be
monitored across the complex. This leads to the need for
certain product share assumptions. The product share figures
are then applied to incremental available securities or the
modeled growth in assets.
T his should not be confused with the number of days we expect it to take to liquidate a position. In other words, when we set the limit at 10 days volume, it does not mean we
are setting a limit of 10 days to liquidate a position. In order to measure liquidity on the basis of time, we would need to add an estimate of a given participation rate. Typical
participation rates on larger orders range from 10% to 25% of ADV but may be significantly larger or smaller depending on the market conditions for a security on a given day.
2
In addition, we make numerous adjustments to get accurate estimates of the trading volumes. This includes, for example, aggregating the trading volumes across multiple listings
such as ADRs, GDRs, dual-listed Canadian stocks, etc., in order to capture the liquidity that is actually available to our traders. We also source volume for certain securities from
nonstandard sources. For example, when Russian and Thai securities report no volume for a particular share class, we have a process for identifying these holdings and sourcing
the correct volume from Bloomberg.
1
3
MAY 2016 / CAPACITY MANAGEMENT
5. Run the analysis for the various scenarios of product
share and percentage of ADV
For example, if MFS currently owns 100,000 shares of ABC
company and the minimum of 10% shares outstanding or
10 days trading volume is 250,000 shares, as a firm we
may now purchase 150,000 more shares before hitting
our limit.
The model then runs an iterative process to find the
AUM, when 20% of the positions by weight are in
securities which have breached the assigned incremental
shares. The output is a matrix of AUM capacity estimates
for each scenario.
3. Assign a range of product share assumptions and
calculate shares available to purchase for this single
strategy
An example of the capacity model output is provided in
Exhibit 2 below. In this case, a portfolio with a current
AUM of $10 billion was considered. Assuming a 30%
product share for this portfolio, the capacity model
suggests that assets could range from $18.8 billion under
a 5-day trading volume constraint to $55 billion if the
trading volume constraint is set to 15 days. For a given
number of days trading volume, increasing the product
share percentage also increases the capacity estimates,
though to a lesser degree than when one varies days
trading volume. This highlights the point that capacity
estimates are always a range and never a single number.
The table and chart below illustrates the way in which the
model produces various capacity ranges based on the
various inputs and constraints imposed in the model.
We will assign a range of product share assumptions,
which are generally related to the current average
product share. These are sometimes adjusted to take
account of perceived changes in the product opportunity
going forward.
In each case, we determine the number of shares available
for the particular strategy to own given the overlap and
internal allocation with other strategies. Following on from
the example above, if the firm can buy 150,000 more
shares of ABC company, in a 25% product share scenario,
we would calculate that this strategy could purchase
37,500 incremental shares in ABC company’s stock.
4. Determine what percentage of the portfolio to replicate
without breaching any of these ownership “limits”
We can approach this in a couple of different ways. One
way is to ask ourselves how much of a portfolio we are
willing to own in securities where MFS has a significant
ownership in terms of average trading volume or shares
outstanding. The other way is to say that we want to be
able to immediately replicate a certain percentage of the
portfolio so that when future assets come in the door, the
portfolio managers have some implementation flexibility,
but not to the point of deviating from the strategy.
Exhibit 2: Example of capacity model output matrix for
commingled portfolio
Assumptions:
Replication of portfolio: 80%
Max: MFS ownership % shares outstanding: 10%
Strategy current AUM: $10 billion
Days Trading Volume (ADV)
Product share
n 20%
n 30%
n 40%
We have been using an 80% immediate replication rate,
which means that we allow for 20% substitution. In
practice, MFS equity portfolios have been replicated at
much higher rates and typically fully. Further adjustments
are made to the model depending on whether the assets
are held in commingled or separate accounts. Commingled
accounts, which benefit from the base of existing assets,
will have greater capacity for incremental flows than a full
funding of a separate account under this approach.
5 days
10 days
15 days
17.8
18.8
19.6
30.0
35.5
40.0
43.5
55.0
66.0
AUM
70.0
50.0
30.0
10.0
40%
Source: MFS.
4
30%
Product share
20%
5 days
10 days
ADV
15 days
MAY 2016 / CAPACITY MANAGEMENT
Quantitative example
Exhibit 3: Replication positions at varying AUM for
Portfolio A
The sensitivity of the quantitative model outlined to changes
in the input variable assumptions can be illustrated using the
following example of a large-cap US equity portfolio (Portfolio
A). The first test involves replicating Portfolio A at increasing
asset levels, then examining trading volume and percent of
outstanding shares owned for each position in the portfolio.
Replication positions
100%
In this case, we limit aggregate MFS ownership of an
individual stock to 10 days trading volume and 10%
of the company’s outstanding shares, and assume that
Portfolio A’s product share percentage is 30% (see Exhibit 3).
The chart shows the proportion of Portfolio A that can be
replicated as assets grow under the constraints stated. In this
case, the 80% replication level is reached when assets are
approximately $35 billion.
90%
80%
Assumptions:
Ownership % = 10%
Trading volume = 10 days
Product share = 30%
70%
60%
15
20
25
30
35
40
Future assets ($ billions)
45
50
55
60
Source: MFS.
Exhibit 4: Replication positions at varying trading
volume constraints for Portfolio A
In the second test, Exhibit 4, we assume that Portfolio A is
fixed at assets of $35 billion, product share allocated to the
portfolio is fixed at 30% and ownership of outstanding
shares is limited to 10%, while we reduce the number of
days trading volume we are willing to hold. In the graph
below, we show that the percent of Portfolio A that can
be replicated decreases as the days trading volume constraint
is tightened.
100%
Replication positions
80%
In the third test, Exhibit 5, we keep the size of Portfolio A
fixed at $35 billion, the days trading volume constraint fixed
at 10 days and the ownership of outstanding shares limited
to 10%, while varying the product share constraint. In this
example, only 85% of the portfolio can be replicated even if
Portfolio A absorbs all available liquidity, i.e., if it assumes
100% of the product share.
60%
Assumptions:
Ownership % = 10%
Product share = 30%
Future assets = $35 billion
40%
20%
19
17
15
13
11
9
Days trading volume
7
5
3
Source: MFS.
Exhibit 5: Replication positions at varying product
share constraints for Portfolio A
Qualitative review
100%
Replication positions
The capacity analysis described can potentially lead to a
wide range of capacity estimates largely because capacity
models tend to be driven by the assumptions employed
(i.e., number of days to establish or exit a typical position,
strategy overlap, etc.). In addition, to arrive at more accurate
estimates of trading volume we make adjustments to
standard reported volumes by, for example, taking account
of multiple listings, the fragmentation of the European
trading market and volumes traded on alternative lesstransparent trading platforms.
80%
60%
40%
Assumptions:
Ownership % = 10%
Trading volume = 10 days
Future assets = $35 billion
100
80
60
Product share
Source: MFS.
5
40
20
1
MAY 2016 / CAPACITY MANAGEMENT
For these reasons, qualitative input from portfolio managers
and the trading desk regarding the ease or difficulty
associated with implementing the strategy and maintaining
the investment style is extremely important.
Overlap in strategies – The capacity considerations related to
strategy overlap across the firm is also part of the risk review
process. The product share assumptions used in the capacity
model may be revisited in this context.
The capacity model is discussed as part of the semiannual
risk review process at MFS, and it is in this context that a
qualitative review of the matrix of capacity estimates takes
place. The broader risk measures prepared for the review
contain additional information that helps put the portfolio
in perspective in terms of the risk profile and portfolio
characteristics, as well as trading costs and performance.
Trading costs, turnover, style consistency and overlap are
some of the elements related to capacity that are examined in
detail in the risk review alongside the capacity model output.
Capacity estimates are always a range
and never a single number.
Non-portfolio considerations
The capacity review comprises the quantitative and qualitative
portfolio considerations outlined above as well as additional
non-portfolio factors such as those listed below.
Idea generation
At MFS, the global research platform forms the backbone
of the firm’s investment idea generation. Analyst opinions,
expressed in the form of a ratings system, continue to be the
dominant source of new ideas for all our client portfolios,
regardless of asset class or style. The platform has been
strengthened and expanded in the past several years by
additional resources and the opening of new investment
offices in Hong Kong, São Paulo, Sydney and Toronto in
order to exploit the growing universe of investment
opportunities globally. We continue to be cognizant of
investment idea generation as we evaluate capacity decisions,
given that expanded research coverage allows for greater
firmwide capacity.
Trading costs – Trends in trading costs can be an important
source of information with regard to how the AUM base
may be impacting transaction costs and potentially reducing
the manager’s ability to add alpha. However, it should be
noted that trading costs are also a function of the market
environment, tend to vacillate for a variety of reasons and
may not be indicative of capacity constraints. In general, the
centralized risk management platform employed at MFS
provides a broad, effective lens with which to evaluate trading
costs and other risk measures.
Turnover – The longer-term investment time horizon at MFS
means that turnover for the firm’s strategies is generally on
the low end of the spectrum. This is as true for the strategies
with low AUM as it is for those with higher AUM. Turnover is
also viewed in concert with liquidity risk, in that lower
turnover strategies can take on more liquidity risk. For
example, even at a higher cost per trade, incrementally higher
trading costs may not have a significant impact on a portfolio
with turnover of 15% compared with a portfolio with
turnover of 100%.
Business diversification
Diversifying assets across multiple strategies is another
important consideration in evaluating the capacity of
individual investment strategies. MFS believes that
diversification translates into long-term stability for the firm,
in part by also allowing for broader investment talent
contributions, and for these reasons actively manages the
business risk associated with product concentration.
Style consistency – This encompasses a number of elements
Portfolio manager time constraints
Portfolio manager non-investment responsibilities are also a
consideration when capacity is under review. While we seek
to maximize value added for our clients by using institutional
portfolio managers and investment product specialists to
support the client base so that portfolio managers are able
to focus on managing investments, new business often
brings with it additional portfolio manager marketing and
service requirements.
like name count, active share, average market capitalization
relative to an index, growth/value style drift and any other
changes in portfolio investment characteristics. In general, any
significant changes that cannot be rationally explained by
market dynamics would come under the spotlight as part of
the risk review. It should be noted, too, that style consistency
is examined as a key risk measure on an ongoing basis.
6
MAY 2016 / CAPACITY MANAGEMENT
GLOBAL EQUITY CASE STUDY
In April 2006, the decision was made to soft close the Global Equity strategy to new separate accounts. Pooled
vehicles for new and existing institutional investors remained open until December 2011, when institutional pooled
funds were closed to new accounts. In March 2013, further restrictions for separate account clients were announced,
with contributions from then-current separate account clients only being accepted until September 2013. The vehicles
designed for individual investors have remained open.
Exhibit 6: Global equity capacity management decisions
Apr 2006
Soft close —
No new separate
accounts
Dec 2011
Closed institutional
pooled funds
to new accounts
2006
Oct 2013
Contributions from
current separate account
clients no longer accepted
2011
2013
The rationale for closing Global Equity was that it would slow the growth of assets for the following reasons:
•Protect the interests of existing and future clients by preserving investment flexibility for portfolio managers such
that they are able to potentially deliver performance in line with client expectations
•Preserve capacity for other existing strategies as well as the prudent launch
of new investment strategies driven by client demand
•Minimize the non-investment responsibilities of the portfolio managers
by limiting large mandates that usually require significant marketing and service requirements to allow the existing
client portfolios to receive the requisite focus
Portfolio characteristics
The portfolio managers have been able to continue to pursue active alpha-generating investment opportunities in
line with the investment philosophy and process for the strategy. Portfolio characteristics have remained relatively
consistent over time.
7
MAY 2016 / CAPACITY MANAGEMENT
Fixed-income capacity management
MFS has imposed restrictions on strategies other than
Global Equity in keeping with the approach outlined in this
paper. Currently, the list of strategies subject to a restriction
of some kind includes European Research, European Smaller
Companies, European Value, Global Concentrated, Global
Value, International Small Cap, International Value and US
Large Cap Value, in addition to Global Equity. This list is an
indication of the commitment MFS has made to a rigorous
capacity risk review process and, in doing so, protecting
clients’ interests.
As we mentioned in the introduction, this paper is largely
focused on outlining our approach to managing capacity in
equity portfolios. Our capacity methodology in the case of
fixed income is similar in certain respects to equities; however,
there are also some key differences. The overall architecture
of the capacity review is comparable in that they both
comprise three key elements: quantitative model, qualitative
assessment and non-portfolio considerations. The nature of
the securities and the markets means that the fixed-income
quantitative model differs in the following regards:
Conclusion
• Trading volume is not considered in the fixed-income
model.
MFS employs both quantitative analysis and qualitative
assessments to help measure and manage capacity. Given
how important capacity management is to the overall
investment process, we consider capacity under the rubric
of risk and review capacity within the context of the semi­
annual risk review process. Idea generation, business
diversification and portfolio manager non-investment
responsibilities are additional considerations taken into
account in capacity decisions.
• Replication constraints are placed at both the bond issue
and the issuer level.
• Only non-government issues are considered in the
replication process.
Product closure decisions
Based on the various factors described above, product
closures are implemented at appropriate asset levels to
protect the interests of our clients. The interests of existing
and future clients are protected by preserving investment
flexibility for our portfolio managers. The MFS goal is to put
each portfolio manager in the best potential position to
generate superior risk-adjusted performance for our clients.
The firm has taken steps to close a number of products in line
with the capacity management approach outlined in this
paper. This includes the soft close imposed on Global Equity
in 2006, which has been followed by additional restrictions
on asset inflows into the strategy.
In general, we are committed to managing capacity to
preserve alpha-generating capability for clients. We believe
that we can build trust and credibility with clients by working
in their interests, and simultaneously create a business model
that is sustainable in the long term.
The firm has tended to adopt a staged approach to
managing capacity in the past, with soft close decisions
preceding hard closure of the strategy, in part to minimize
business disruptions for clients invested in the strategy.
This is illustrated in the timeline provided for the capacity
management decisions made with regard to the Global
Equity strategy (see page 7: Global Equity Case Study).
When circumstances change, we have also made the
decision to reopen strategies or roll back the soft close.
A differentiated approach is often adopted for the retail
versus institutional markets to account for the fact that it is
much easier to accommodate inflows of smaller amounts of
retail assets than it is to absorb larger pools of funds in
institutional separate accounts. Also, it is worth noting that a
proportion of institutional account redemptions take place on
a regular basis for reasons outside the control of the
investment manager concerned. In this context, the inflow of
retail assets can be seen as replacing some of these
institutional outflows.
8
The views expressed are those of the author(s) and are subject to change at any time. These views are for informational purposes only and should not be
relied upon as a recommendation to purchase any security or as a solicitation or investment advice from the Advisor.
Unless otherwise indicated, logos and product and service names are trademarks of MFS® and its affiliates and may be registered in certain countries.
Issued in the United States by MFS Institutional Advisors, Inc. (“MFSI”) and MFS Investment Management. Issued in Canada by MFS Investment Management
Canada Limited. No securities commission or similar regulatory authority in Canada has reviewed this communication. Issued in the United Kingdom by MFS
International (U.K.) Limited (“MIL UK”), a private limited company registered in England and Wales with the company number 03062718, and authorized and
regulated in the conduct of investment business by the U.K. Financial Conduct Authority. MIL UK, an indirect subsidiary of MFS, has its registered office at
One Carter Lane, London, EC4V 5ER UK and provides products and investment services to institutional investors globally. This material shall not be circulated
or distributed to any person other than to professional investors (as permitted by local regulations) and should not be relied upon or distributed to persons
where such reliance or distribution would be contrary to local regulation. Issued in Hong Kong by MFS International (Hong Kong) Limited (“MIL HK”),
a private limited company licensed and regulated by the Hong Kong Securities and Futures Commission (the “SFC”). MIL HK is a wholly-owned, indirect
subsidiary of Massachusetts Financial Services Company, a U.S.-based investment advisor and fund sponsor registered with the U.S. Securities and Exchange
Commission. MIL HK is approved to engage in dealing in securities and asset management-regulated activities and may provide certain investment services
to “professional investors” as defined in the Securities and Futures Ordinance (“SFO”). Issued in Singapore by MFS International Singapore Pte. Ltd., a private
limited company registered in Singapore with the company number 201228809M, and further licensed and regulated by the Monetary Authority of Singapore.
Issued in Latin America by MFS International Ltd. For investors in Australia: MFSI and MIL UK are exempt from the requirement to hold an Australian financial
services license under the Corporations Act 2001 in respect of the financial services they provide. In Australia and New Zealand: MFSI is regulated by the U.S.
Securities and Exchange Commission under U.S. laws, and MIL UK is regulated by the U.K. Financial Conduct Authority under U.K. laws, which differ from
Australian and New Zealand laws.
MFSE-CAPAC-WP-5/16
30201.4