Overview
James H. Gilkeson, CFA
Associate Professor of Finance
University of Central Florida
Orlando, Florida
Managing private client assets is barely different
from managing institutional assets—just add a few
tax issues, throw in a bit of investor behavior talk, and
you are set, right? If only that were true. In practice,
investment professionals who work with high-net-
worth individuals and families may feel as if they
have developed split personalities—part securities
specialist, part tax expert, and part therapist.
As the presentations at the AIMR conference
“Private Wealth Management: New Developments
in Investing and Advising,” held in Atlanta on 17–18
March 2003, make clear, working with private clients
does have many facets. But the skills a private client
manager needs in addition to a thorough grounding
in investment management are a general under-
standing and appreciation of the tangential issues of
income and estate tax planning, business succession
planning, tax-efficient investment strategies, and so
on, not expert status in each of them. The presenta-
tions in this conference proceedings provide valuable
insights into each of the three key facets of investing
for high-net-worth clients: the characteristics of new
investment opportunities, tax complexities, and
client needs.
New Investment Opportunities
Investment professionals need to move beyond a
single-period framework (i.e., invest today and liqui-
date in 20 years) to a life-cycle approach, according
to Zvi Bodie. Private clients acknowledge the cyclical
pattern of earning, spending, and leisure, as well as
a concern for wealth accumulation during their life-
time. This new approach requires more complex
modeling of portfolio construction and of risk. It also
implies a need for new, more sophisticated invest-
ment products and more cost-efficient channels of
distribution. As an example, Bodie points out that
two common life-cycle needs of investors are long-
term-care insurance and life insurance. Bundling
these together creates a product with risk character-
istics that better meet client needs and are easier for
issuers to manage.
Hedge funds are increasingly being included in
the portfolios of high-net-worth individuals. Jean
Brunel explains that it is a mistake to conclude that
©2003, AIMR®
hedge funds are a separate asset class. Simply put,
the hedge fund universe is not homogeneous.
Instead, Brunel argues, hedge funds are an extension
of active management that represents many different
asset classes. The key to allocating them properly in
a private client portfolio is to understand how they
change the risk profile of the portfolio. The key issues
are liquidity and manager risk. Hedge funds are
much less liquid than traditional actively managed
investments (mutual funds) but allow much greater
manager flexibility.
Tax Complexities
Whereas Brunel concentrates on the unique invest-
ment characteristics of hedge funds, Edward Dough-
erty explains their tax effects. He provides a primer
on the differing tax concerns of individuals, corpora-
tions, and “tax-exempt” organizations and how a
hedge fund’s structure affects an investor’s tax bur-
den. Critical structural decisions include corporation
versus partnership or limited liability partnership,
onshore versus offshore, and master-feeder versus
side-by-side. It is important that portfolio managers
and advisors understand the tax effects of each of
these decisions before placing private client assets
with a hedge fund manager.
Of course, taxes are important for all manner of
private client portfolios, not just hedge funds. As
Christopher Luck explains, there are a number of
ways to create tax alpha by more efficiently manag-
ing portfolios. In a world of uncertain outcomes,
however, determining which strategies will be most
effective is more difficult. As a solution to this prob-
lem, Luck shows how simulations can be used to
judge the relative effectiveness of various approaches
to tax management.
Christine Todd takes a fresh look at tax-exempt
securities. In recent years, this asset class has pro-
vided surprisingly strong returns with little risk.
There is, however, a concern lurking on the horizon
for an increasing number of high-net-worth (and not
so high-net-worth) investors—the alternative mini-
mum tax (AMT). Although a relatively minor per-
centage of the tax-exempt securities in the market are
subject to AMT, Todd explains that these securities
www.aimrpubs.org • 1
Investment Counseling for Private Clients V
often provide the highest yield but are often avoided
because more and more taxable investors are nearing
proximity to AMT liability. Todd also explains that
the muni market’s volatility is increasing as non-
traditional (taxable) investors are beginning to par-
ticipate in the market through crossover strategies—
investing in munis when they look cheap relative to
taxable bonds and reversing the trade when the yield
relationship reverts to a more normal spread.
Client Needs
Aside from taxes, the primary difference between
working with institutional investors and working
with private clients is dealing with behavioral issues.
When working with high-net-worth individuals and
families, it is as crucial to understand each client’s
distinct point of view as it is to understand the char-
acteristics of the financial marketplace.
As Marty Carter explains, an individual’s
approach to investing is driven by his or her learned
attitudes toward wealth, saving, and spending.
Wealth can critically impact—often negatively—
personal and family relationships. Although it may
seem odd to suggest that an investment professional
should be concerned if a large family or a couple
cannot communicate effectively about wealth-
related issues, it will be next to impossible to work
with them if these sensitive issues are not tackled.
Carter recommends involving an expert facilitator to
help open the doors of honest communication before
proceeding with an investment plan that lacks client
buy-in.
Scott Welch points out the growing need for
private client advisors who can bypass the conflicts
2 • www.aimrpubs.org
of interest, biases, and other problems frequently
associated with traditional investment channels. He
explains that private clients seek six traits from an
advisor: objectivity, expertise, discipline, access, edu-
cation, and service. Welch suggests that advisors can
particularly demonstrate their value to clients
through manager search and selection and by creat-
ing performance reports that are easily understood
and address each client’s unique concerns.
Thomas Giachetti also emphasizes clear commu-
nication with clients. He focuses on the regulatory
examination process but makes it clear that the keys
to avoiding client complaints (as well as surviving
regulatory examinations) are creating and following
specific procedures, making repeated and complete
disclosures to clients, and maintaining good record
keeping. Clients who are informed do not get sur-
prised, and clients who do not get surprised are likely
to remain clients.
Conclusion
Understanding client parameters is only the begin-
ning of the story, of course. As Zvi Bodie observes,
“Investment decision making that is dynamically
tailored to the needs of an individual is highly com-
plicated.” Indeed, the considerable difficulty of inte-
grating client needs and goals into a comprehensive,
effective investment strategy is why clients hire
managers and advisors in the first place. The presen-
tations in this proceedings offer valuable insights for
investment professionals who seek to put their wis-
dom and expertise to better use for their clients.
©2003, AIMR®
Applying Financial Engineering to Wealth
Management
Zvi Bodie
Professor of Finance and Economics
School of Management, Boston University
Boston

The traditional investment paradigm, the Markowitz mean–variance model, focuses

only on total wealth at the end of a single (long) period. The assumption is that time
diversification will make equities “safe.” A more appropriate approach is the
Samuelson–Merton life-cycle paradigm, a multiperiod model that stresses the need for
hedging and insurance in addition to precautionary saving and diversification. A
corollary of this new model is that the investment industry ought to provide affordable
products tailored to suit the average investor’s means and needs.

rises give rise to opportunities,” as the old
“C saying goes. Now is the time that the services
of investment professionals are probably most
needed by private clients. During the bull market,
investors could do no wrong. Now, they desperately
need professional help.
During the past 50 years, investment manage-
ment has become a science in the same way that med-
icine is a science. In both cases, some practices long
predated the development of the underlying science.
In medicine, although disagreements may exist about
what treatments are appropriate for certain types of
problems, most people can distinguish between a
well-trained doctor and a quack. In investment man-
agement, the distinction is not always so clear.
I have written extensively about the fallacy of the
long run—the belief that holding extremely risky
stocks for a long time makes them safe.1 This assertion
has made me rather unpopular with practitioners
because I am challenging an almost universally held
belief on their part. Certainly, such a rift has existed
between what academics think and what investment
managers and investors believe for the more than 30
years since Samuelson and Merton wrote ground-
1
Zvi Bodie, “On the Risks of Stocks in the Long Run,” Financial
Analysts Journal (May/June 1995):18–22.
breaking articles on this subject.2 A number of the
concepts described in this presentation have come
from my collaboration with Robert Merton.3
The modern science of investment management
(and finance in general) can help practitioners add a
tremendous amount of value for their clients. Much
value can come from creating new types of invest-
ment products and educating practitioners in their
use and how to explain them to their clients. The time
has come to move beyond the rather simplistic
choices of which mutual fund to buy and of the right
mix of stocks, bonds, and cash for a client’s portfolio
to a discussion of investment products designed to
meet specific goals of ordinary investors. The chal-
lenge is to use financial engineering to produce a new
generation of user-friendly investment products that
can be customized at reasonable cost.
2
See Robert C. Merton, “Lifetime Portfolio Selection by Dynamic
Stochastic Programming: The Continuous Time Case,” Review of
Economics and Statistics (August 1969):247–257 and Paul A. Samuel-
son, “Lifetime Portfolio Selection by Dynamic Stochastic Program-
ming,” Review of Economics and Statistics (August 1969):239–246.
3
For further discussion of some of the topics in this presentation, see
Zvi Bodie, “Thoughts on the Future: Life-Cycle Investing in Theory
and Practice,” Financial Analysts Journal (January/February
2003):24–29 and Robert C. Merton, “Thoughts on the Future: Theory
and Practice in Investment Management,” Financial Analysts Journal
(January/February 2003):17–23.

©2003, AIMR® www.aimrpubs.org • 3

Investment Counseling for Private Clients V
Shortcomings of Current
Investment Products
Virtually no one can understand today’s investment
choices, whether the source of the information is an
online advice engine or a brochure from a mutual
fund company. Such concepts as mean–variance
analysis or probability of loss/gain mean nothing to
most people. I am often approached by fellow profes-
sors who are quite intelligent yet do not have a clue
about what this information means or how they
should use it. Investors are struggling with the deci-
sion of how to allocate among various investment
options that currently exist, yet the mutual fund
industry continues to add more products. If investors
do not know how to allocate their funds among three
choices, offering 30 choices will not help them. A
good example of this is the explosion of investment
choices offered by the Teachers Insurance and Annu-
ity Association—College Retirement Equities Fund
(TIAA-CREF).
The investment management industry must
rethink the choices they give to investors. Today’s
offerings are like a restaurant that requires diners to
choose from a menu of eggs, flour, butter, and milk,
rather than a list of prepared desserts. Investors, like
diners, want to choose from a list of completed prod-
ucts, not a list of ingredients.
Such a menu should be tailored to the goals and
needs of individuals and should enable them to make
informed choices. Consider the way computers are
sold online. The major manufacturers offer standard
configurations tailored to home or business users.
Using a menu-like interface, customers can click on
items to learn what various features do and thus can
easily customize the standard configurations to their
specific needs and desires.
The investment management business should use
a similar approach. Such an approach could not have
been followed 30 years ago, but because of advances
in financial engineering, it can be accomplished today.
Financial engineering is used extensively by banks,
pension funds, and other financial institutions. It is
used to manage currency, interest rates, and other risk
exposures. To date, however, it has had little impact
on investment management decisions.
Today, financial engineering is commonly used
in the world of private banking. Multimillionaires
use custom-tailored derivatives to help them manage
their risk. For example, they might desire a contract
that puts a collar around a large exposure (such as an
undiversified equity holding), but they have to pay
dearly for this service. These kinds of private bank-
ing services are by far the most lucrative of a bank’s
activities. But the costs are far beyond the means of
4 • www.aimrpubs.org
ordinary people who are saving for retirement, their
child’s education, or other common goals. The
investment management industry should strive to
provide this type of financial engineering to average
investors—but on a mass scale.
Insights of Modern Finance
Five key insights of modern investment science
should be considered in redesigning standard invest-
ment products. First, a person’s welfare depends not
only on end-of-period wealth—what they will own in
a year or 30 years or whatever their “end point” is—
but also on consumption of goods and leisure over a
lifetime. An investment portfolio should be tailored
to the investor’s personal objectives. An individual
who is concerned about a certain minimum standard
of living in retirement is extremely risk averse, and
the way to achieve his or her goal is not through
diversification but through hedging. Such a person
should invest all of his or her money in inflation-
indexed Treasury securities (commonly referred to as
“TIPS”) because doing so will lock in a real rate of
return that provides a real standard of living.
The majority of personal investment advisors
say that the most basic principle of investing is diver-
sification. For example, as an experiment, I tried a
large number of online investment advice engines.
For each one, when answering the set of questions
that helped determine my risk tolerance, I made it
perfectly clear that I am extremely risk averse. Nev-
ertheless, every engine recommended that I hold at
least 30 percent of my money in equities. Why?
Because I have a long time horizon. No matter how
risk averse I made myself appear, these advisors still
wanted me to put one-third or more of my wealth
into the riskiest class of securities. Advisors still cling
to the mistaken idea that over a long-enough time
period, the magic of diversification will somehow do
away with risk.
Actually, a risk-averse investor with a long time
horizon should hold long-term U.S. Treasury bonds
that are inflation protected. One of the first discus-
sions in basic finance courses on the term structure of
interest rates is on determining the preferred habitat
of the investor. An investor with a short horizon
should focus on short-term bonds; a long-horizon
investor should focus on long-term bonds. This prin-
ciple is not reflected in any of the advice engines that
I found on the Internet.
Second, when thinking about investment risk
and the creation and preservation of wealth, it is
extremely important to consider what is likely to
happen to real rates of return in the future. For exam-
ple, consider a simple choice. Which portfolio would
©2003, AIMR®

you prefer at retirement: $5 million in cash or $10
million in cash? The obvious answer is $10 million.
But what if you could have $5 million with a real
interest rate of 5 percent per year or $10 million with
a real interest rate of 1 percent per year? A real
interest rate of 1 percent a year on $10 million gener-
ates a perpetual annual income of $100,000 a year in
real terms, whereas 5 percent on $5 million generates
$250,000 per year. Most people care less about the size
of their retirement portfolio than the standard of
living that the portfolio can produce to sustain them
in retirement. Most people would prefer the $5 mil-
lion with the 5 percent real interest rate. Of course,
time horizon (or life expectancy) affects the optimal
choice. The greater wealth wins in the short term; the
greater rate of return wins in the longer term.4
This approach to investment management dif-
fers from the standard approach, which measures
risk in terms of ending wealth rather than interest
rate (earning power) uncertainty. Virtually all of the
quantitative tools currently used for asset allocation
measure risk in terms of ending wealth; interest rate
(earning power) uncertainty is ignored.
Third, how should investors deal with multiple
periods? The Markowitz mean–variance model
developed in the 1950s, which is at the analytical core
of every advice engine being used today, is a single-
period model designed to optimize end-of-period
wealth. In the Markowitz model, lengthening the
time period always results in a higher allocation to
equities (risky securities). In 1969, the Markowitz
model was superseded in the academic literature by
the Samuelson–Merton life-cycle model. The life-
cycle model produced some interesting results that
contradict the single-period Markowitz model. In
particular, what happens as the length of the plan-
ning horizon gets longer and longer? Samuelson
demonstrated that length of time horizon has no pre-
dictable effect on the optimal fraction of assets that
should be invested in equities. The Markowitz model
relies on time diversification to manage market risk;
what is really needed is multiperiod hedging. The
life-cycle model shows that diversification alone is
insufficient to ensure the investor’s goals. Invest-
ments that limit downside risk must be held in the
portfolio, regardless of the investor’s time horizon.
Fourth, the value, riskiness, and flexibility of a
person’s labor earnings are of first-order importance
in determining optimal portfolio selection at each
stage of the life cycle. The rule of thumb among
financial planners and personal investment advisors
4
Robert Merton, “Thoughts on the Future: Theory and Practice in
Investment Management,” Financial Analysts Journal (January/
February 2003):17–23.
©2003, AIMR®
Applying Financial Engineering to Wealth Management
is that the fraction of an individual’s portfolio that
should be allocated to equities is 100 minus the per-
son’s age. A 30-year-old investor should have a 70
percent allocation to equities, but a 70-year-old inves-
tor should have a 30 percent allocation to equities.
Although simplistic, this rule may be appropriate for
the average individual. However, consider those
who work on Wall Street or are entrepreneurs start-
ing new businesses. They already have a huge expo-
sure to the market by virtue of the risk to their human
capital. That is, if the market takes a dive, not just
their wealth but also their earning power may be
damaged. An analogous and more common situation
is an individual who holds the stock of his or her
employers (perhaps in retirement accounts). The
additional exposure to the market is undesirable
because the employee already has a large exposure
to the risk of the company as a result of working for
it. In fact, the company-specific exposure from
employment alone is probably larger than is optimal,
so perhaps the best strategy is to short the stock of the
employer.
I am not suggesting that if you work in the invest-
ment management industry and your future income
depends on the stock market, you should immedi-
ately short the market. Buying some puts on the S&P
500 Index, however, might not be a bad idea. Gener-
ally, some people, especially those in their 20s or 30s,
have exposure to the stock market as a whole or to
the stocks in a particular industry. Depending on the
occupation and the industry, that equity exposure
may be great or small. But whatever the degree of
exposure, it should be an important determinant of
the composition of the individual’s investment port-
folio. Why, out of thousands of choices, are there no
mutual funds configured for people who work in
certain industries?
Fifth and finally, habit formation is a basic aspect
of human nature and can give rise to a demand for
guarantees against a decline in investment income.
People want to maintain the standard of living to
which they have become accustomed. A need exists
for ratcheting or escalating products. These products
rise in value when the market does well and are stable
when the market drops. Life insurance companies sell
some products of this nature, but they are incredibly
expensive. The major problem is selling costs. If these
products were mass produced, they could be offered
at much lower margins. If professional investment
advisors and financial planners respond positively to
these products and recommend them to clients, it will
help lower the huge marketing costs associated with
selling them. Image problems must also be overcome.
For example, variable annuities are a terrific idea in
concept but have acquired a pejorative meaning
www.aimrpubs.org • 5
Investment Counseling for Private Clients V
because of their high fees. Although most variable
annuity products are offered by life insurance com-
panies, they are essentially mutual funds in various
wrappers (with a few added features).
Comparing the Samuelson–Merton
and Markowitz Models
Exhibit 1 compares the two paradigms of life-cycle
investing. The old paradigm is the Markowitz model,
which currently dominates the world of personal
investment management. The new paradigm is the
Samuelson–Merton model.
Welfare. What is the best measure of welfare?
In the old paradigm, it is wealth at the end of the
planning horizon. In the new paradigm, it is lifetime
consumption (or some component of consumption,
such as consumption during retirement). A key to the
new paradigm is that welfare is measured by the flow
of consumption over time, not the stock of wealth at
a particular point in time. To understand the differ-
ence between the two concepts, remember my exam-
ple of whether you would rather have $5 million
earning 5 percent or $10 million earning 1 percent.
Time Frame. With the old paradigm, a single-
period model, the time horizon could be 1 year or 30
years. When a user of an online financial engine
performs a Monte Carlo simulation for his or her
chosen time horizon, no dynamic rebalancing is
involved. In effect, an investor makes a single deci-
sion and runs with it. The structure of the model is
such that the longer the time horizon, the greater the
equity allocation.
The results are quite different for a multiperiod
model. In a multiperiod setting, investors make
dynamic adjustments depending on unfolding cir-
cumstances. They may change their asset allocation
as markets rise or fall or as their income rises or falls.
Exhibit 1. Comparison of Life-Cycle Paradigms
Feature Old Paradigm
Measure of welfare Wealth
Time frame Single period (stocks seem
safe in the long run)
Risk management techniques Precautionary saving
Diversification
Retail financial products Cash
Insurance policies
Mutual funds
6 • www.aimrpubs.org
They may even change their long-term life plan. If
the market does particularly well, they may choose
to retire early. If it does particularly poorly, they
may have to work longer than expected. Dynamic
adjustments are also at play throughout the retire-
ment period, because investors may alter the speed
with which they draw down their wealth in
response to changing market conditions and per-
sonal circumstances.
Risk Management Techniques. The old para-
digm rests on two risk management techniques: pre-
cautionary saving and diversification. The wealthy
are able to build a large cash reserve to protect against
the occurrence of a downside event (precautionary
saving). For most, however, it is a challenge to save
the small amount that they do, and then in an attempt
to better the return on those savings, invest some of
it, if not all, in the market. Fortunately, or so the story
goes, diversification provides an acceptable approach
to risk management for these investors—over a long-
enough time horizon, risky stocks become magically
safe. In the old paradigm, diversification is the touch-
stone.
The new paradigm also includes precautionary
saving and diversification as risk management tech-
niques but adds two more important, fundamental
risk management techniques: hedging and insuring.
Hedging involves a conscious decision to forgo some
of the upside of a risky investment in order to elimi-
nate the downside. For example, an investor might
decide to lock in a return, maybe using TIPS. Cur-
rently, the 30-year TIPS is paying a real rate of interest
of 2.4 percent. (A few years ago, investors could lock
in a real rate of return of 4 percent.) The price paid
for this protection against falling stock or corporate
bond markets is forgoing the ability to participate in
rising markets.
New Paradigm
Lifetime consumption
Many periods (stocks are risky in short
and long run)
Precautionary saving
Diversification
Hedging
Insuring
Targeted savings accounts (e.g.,
tuition-linked certificates of
deposit)
Structured standard-of-living
contracts
©2003, AIMR®

Insuring means paying a premium today to elim-
inate the future downside—but with the ability to
retain the benefit of the upside. For example, buying
a put option on a stock portfolio is insuring against a
price decline below the exercise price of the option.
This insurance has a cost, but the investor retains the
upside less the cost of the put. Most of the new
products I envision, designed for ordinary people
with relatively limited wealth, would incorporate
some kind of hedging or insurance features to protect
against downside risk. Of course, investors would
have to pay for this protection, either with an insur-
ance (put) premium or with less upside participation.
Retail Financial Products. The retail financial
products in the old paradigm are cash, traditional
insurance policies, and mutual funds. In the Markow-
itz world, client goals have a limited impact on asset
allocation recommendations. The goals are only incor-
porated in determining the time horizon; if the time
horizon is long, the standard formulation (i.e., a large
equity allocation) is considered optimal. As the hori-
zon moves shorter, the allocation weights are changed
to favor cash and bonds. This approach is demonstra-
bly wrong. If the strategy is driven only by age and
has nothing to do with expected market performance
over the planning horizon, the result may be anything
but optimal. I call the conventional model “faith-based
investing” because it is based on the idea that the stock
market will always rise. The stock market, however,
can decline, and the conventional model does not use
dynamic feedback to guide the investor’s decision
making if, and when, the market does decline.
For the new paradigm, I envision structured sav-
ings and investment accounts that are designed to lock
in specific client goals over the client’s life cycle. TIAA-
CREF (of which I am currently a member) offers vari-
able annuity payouts that can increase or decrease
over time. I like the increasing part, but I do not want
to experience the decreasing part during my retire-
ment! Because the annuity provides income that lasts
for life, the purchaser of the annuity is insured against
living too long, not against a decline in the market or
a decline in interest rates. The same is true of most
annuity products offered today. It is important that
new standard-of-living contracts, or variable annu-
ities, include provisions for escalation. Ultimately, the
goal should be that the standard form of retirement
annuity will be an annuity that can only increase.
Consider an example of an escalating life annuity
that provides a minimum standard of living (indexed
for inflation) that increases if the stock market per-
forms well. A 65-year-old investor with $1 million in
savings puts $900,000 into an inflation-proof real
annuity paying $55,000 a year and invests the other
$100,000 in equities for potential growth in real
income. Each year, she transfers some of the money
©2003, AIMR®
Applying Financial Engineering to Wealth Management
from the risky equity portfolio to buy an additional
life annuity with an escalation. The upside leverage of
this strategy could be increased by investing the
$100,000 in a series of equity call options maturing in
each of the next 10 years. If, on the annual expiration
date, that year’s call is in the money, the proceeds
would be used to increase the guaranteed income
floor (the lifetime annuity). If the call is out of the
money, the floor remains unchanged for another year.
We can take this a step further and consider
another product. People worry about two things in
their old age: outliving their resources (assuming they
are healthy) and needing assisted living or nursing-
home care. Life annuities are available to take care of
the first problem, and insurance products are avail-
able to provide for long-term care. But problems exist
with these two products. On the one hand, many
people do not qualify for long-term-care insurance
because they are too sick or their health history puts
them in a high-risk category. On the other hand,
companies that sell life annuities have to protect
themselves against adverse selection; that is, only
healthy people will buy life annuities because they are
the ones for whom it makes sense actuarially (so the
price of lifetime annuities is too high for all but the
healthiest). A bundled-risk product that provides a
lifetime annuity and long-term-care insurance will
help solve both problems. A healthier person will
receive the lifetime annuity payment for a longer time
but will not use the long-term-care protection for
many years. A less-healthy person will need long-
term care sooner but is unlikely to receive annuity
payments for as long. From the insurance company’s
point of view, the adverse selection problems have
cancelled out, and the investor receives desired pro-
tection from both the risk of living too long and the
risk of needing expensive long-term care.
Conclusion
The old investment paradigm, the Markowitz mean–
variance model, is a single-period model in which a
longer investment time horizon invariably leads to a
higher equity (risky security) allocation. The focus of
the model is end-of-period wealth. It mistakenly
assumes that time diversification will make equities
safe enough for all investors, no matter how risk
averse they are. The new paradigm, the Samuelson–
Merton life-cycle model, is a multiperiod model that
stresses the need for hedging and insurance in addi-
tion to precautionary saving and diversification. The
new focus is on the investor’s income, consumption,
and leisure patterns throughout life. For investment
advisors and managers, a primary implication of this
new paradigm is the need for affordable structured
products that are tailored to fit the average investor’s
goals and needs.
www.aimrpubs.org • 7
Investment Counseling for Private Clients V
Question and Answer Session
Zvi Bodie
Question: What are the educa-
tional implications of this new par-
adigm for investment managers
and clients?
Bodie: We will not see radical
changes in or rewriting of books
intended for public consumption.
In a way, the literature developed
during the bull market is all wish-
ful thinking. Glassman and Has-
sett, two reputable market
commentators, wrote a book called
Dow 36,000: The New Strategy for
Profiting from the Coming Rise in the
Stock Market, in which they said
stocks are safer than inflation-
indexed Treasury securities.5 No
one believes that premise any
longer because it was predicated
on the notion that stocks can never
go down.
The views I expressed in my
presentation derive from main-
stream academic thinking. The
Investments textbook I co-authored
with Kane and Marcus makes the
same points in several places, and
all of the top 30 Business Week
5 James K. Glassman and Kevin A. Hassett,
Dow 36,000: The New Strategy for Profiting
from the Coming Rise in the Stock Market
(Three Rivers, MI: Three Rivers Press, 2000).
8 • www.aimrpubs.org
schools use this text.6 So, my views
are not marginal in the academic
community. Among practitioners,
however, these ideas are consid-
ered quite controversial.
Question: Could you comment
on the lack of growth in the reverse
mortgage market relative to your
claim that new products are needed
to meet life-cycle investing goals?
Bodie: The problem with the
reverse mortgage market is high
cost. The hedging products that
reduce volatility are quite trans-
parent in terms of cost. In contrast,
one of the nice features (from the
seller’s viewpoint) of equity
mutual funds is that fees are buried
in the volatility, so the investor
doesn’t notice a 200 bp fee in the
midst of 10, 15, or 20 percent
annual volatility. But when the
investor is earning 2.4 percent or
paying the interest rate on a
reverse mortgage, the fee is notice-
able. The fee is a fixed sum, so the
big challenge is lowering the cost,
which is an engineering problem
(hence my use of the term “finan-
cial engineering”).
6
Zvi Bodie, Alex Kane, and Alan Marcus,
Investments (New York: The McGraw-Hill
Companies, 1999).
The same is true with escalat-
ing annuities. The challenge is to
use new technology to make the
escalation option more accessible
and affordable. Professional finan-
cial planners and investment advi-
sors have a huge role to play in
addressing this challenge. Perhaps
10 percent of sales will be direct
sales. The rest will be intermedi-
ated because most investors are
uncomfortable buying these prod-
ucts on their own. The stakes are
too high. They want validation
from a trained professional. Thus,
the practitioner community needs
to educate itself.
I don’t understand why ordi-
nary people, unless they’re inter-
ested in the intricacies of investing,
should have to become educated
about it any more than I see why
ordinary people should have to
become educated about cars. One
needs to know the basics of car
maintenance, but one should not
have to be knowledgeable enough
to diagnose and repair car mal-
functions. Car mechanics should
do that for us. Investment decision
making that is dynamically tai-
lored to the needs of an individual
is highly complicated.
©2003, AIMR®
A New Perspective on Hedge Funds and
Hedge Fund Allocations
Jean L.P. Brunel, CFA
Managing Principal
Brunel Associates, LLC
Edina, Minnesota

The hedge fund universe is not homogeneous and should not be considered a coherent
asset class. Rather, hedge fund management is an extension of traditional active
management, with some critical differences. Thus, if investors seek to place portfolio
constraints on hedge fund allocation, they should do so not to limit general exposure to
an asset class but rather to achieve a certain level of exposure to manager and liquidity risk.

edge funds are not a homogeneous universe;
H therefore, characterizing hedge funds as a
generic asset class is an error. Hedge funds are simply
an extension of traditional forms of active manage-
ment, with several notable differences. In this presen-
tation, I will share an approach I have recently
developed to help managers make decisions about
optimal hedge fund allocations.
By definition, any statistical analysis that is per-
formed on historical manager results is subject to a
number of caveats, particularly survivorship bias.
The data used in this presentation are almost cer-
tainly affected by this problem. But to paraphrase
Winston Churchill’s famous statement on democ-
racy, this was the absolutely worst possible approach
I could take—except for all the others.
The Problem with Traditional
Thinking
The main problem confronting portfolio managers
when making asset allocation decisions about hedge
funds is assigning them the right strategic weight.
Because hedge funds are viewed as a different type of
investment vehicle, constraints are often imposed on
allocations to them. Some of these constraints are wise,
but frequently, they are highly arbitrary. And whether
the constraints are informed or not, the typical result
is that managers do not focus on what makes a hedge
fund truly different from other types of assets.
Hedge funds are a highly heterogeneous uni-
verse, not a single asset class with homogeneous char-
acteristics. Panel A of Figure 1 plots (for the five-year
period ending 31 December 2002) the risk and return
for indexes representing all of the distinct hedge fund
strategies covered by the Hedge Fund Research (HFR)
database, composites (some of which are provided by
HFR and some of which I calculated), the traditional
asset classes used in strategic asset allocation (i.e., U.S.
Treasury bills, Salomon Brothers Broad Investment-
Grade [BIG] Index, high-yield bonds, corporate
bonds, S&P 500 Index, Europe/Australasia/Far East
[EAFE] Index, and Russell 2000 Index), and an index
of market-neutral strategies. Panel A illustrates why
categorizing hedge funds as a separate asset class is a
mistake; the hedge fund universe comprises a wide
range of strategies and outcomes.
As evidence that the wide range of outcomes
shown in Panel A is not period specific, Panel B plots
risk and return for the 10-year period ending 31
December 2002. Panel B also shows that hedge funds
have provided a higher level of return than tradi-
tional asset classes for any level of risk. Nevertheless,
the differences between various hedge fund strate-
gies are so great that to think of them as a single
aggregate category makes no sense. But if hedge
funds are not a single asset class, then what are they?
Figure 2 separates various parts of the hedge
fund universe into clusters of commonality. Panel A
shows that the so-called market-neutral strategies
share risk and return outcomes that are much closer
than those outcomes for the full hedge fund universe.
Note that the range of risk and return levels in Panel
A of Figure 2 is much smaller than in Panel A of

©2003, AIMR® www.aimrpubs.org • 9

Investment Counseling for Private Clients V

Figure 1. Risk–Return Scattergrams of Hedge Funds and Major Indexes,

December 1992–December 2002
A. Most Recent 5 Years
Return (%)
20

15

HF Sector Total
10 BA MN Index
JPM EMBI+
Salomon
BIG Composite MN Strategies HF Equity Unhedged

5 HF Fixed Total Composite EM Equities and Bonds

U.S. HF EM Total
T-Bills

Merrill HY S&P 500 Russell 2000

0
MSCI EAFE MSCI EM

5
0 5 10 15 20 25 30 35
Risk (%)
Hedge Fund Strategy Composite Traditional Asset Class

B. 10 Years
Return (%)
25

20

Composite MN Strategies HF Sector Total

15
BA HF Equity Unhedged
MN Index
HF EM Total

10 JPM EMBI+
HF S&P 500 Russell 2000
Fixed Total
Salomon Merrill HY
5 U.S. BIG MSCI EAFE
MSCI EM
T-Bills Composite EM Equities and Bonds

0
0 5 10 15 20 25 30
Risk (%)
Hedge Fund Strategy Composite Traditional Asset Class

Note: BA MN = Brunel Associates market-neutral composite; EM = emerging markets; HF = hedge fund;

HF EM = hedge fund emerging markets; JPM EMBI+ = J.P. Morgan Emerging Markets Bond Index Plus;
Merrill HY = Merrill Lynch High-Yield Index; MN = market neutral; MSCI EM = MSCI Emerging
Markets Index.

10 • www.aimrpubs.org ©2003, AIMR®

 A New Perspective on Hedge Funds and Hedge Fund Allocations

Figure 2. Risk–Return Scattergrams for Various Strategies, 10-Year Period Ending 31 December 2002
A. Market-Neutral Strategies B. Fixed-Income Strategies
Return (%) Return (%)
16 11
14 Event Driven 10 FI Mortgage Backed
12 Relative-Value Arbitrage Convertibles
Distressed
9 Total
10 Convertible Arbitrage Merger Arbitrage
8
8 Equity Market Neutral Statistical Arbitrage Salomon BIG FI High Yield
Fixed-Income Arbitrage 7
6 Arbitrage
Benchmark (60% U.S. T-Bill, 40% BIG) 6 Merrill Lynch High Yield
4
5
2
0 4
0 1 2 3 4 5 6 7 8 0 2 4 6 8 10 12 14 16
Risk (%) Risk (%)

C. Equity Strategies D. Sector-Specific Strategies

Return (%) Return (%)
18 25

16 Equity Hedge (Long/Short) Energy Technology

20 Financial Sector
14 Equity Unhedged
Healthcare and Biotech
15 Sector Total
12
Nasdaq Miscellaneous
10 S&P 500 S&P 500 Nasdaq
Equity Market Neutral Russell 2000 10
Russell 2000
8 Real Estate
6 5
Short Sellers
4
0 5 10 15 20 25 30 35 0
0 5 10 15 20 25 30 35
Risk (%)
Risk (%)

E. Emerging Market Strategies

Return (%)
16
14 EM Latin America
JPM EMBI+ EM Global
12 EM Total
10
8 EM Asia
6
Benchmark (50% EM Debt, 50% EM Equity)
4
MSCI EM-Free
2
0
0 5 10 15 20 25 30
Risk (%)

Note: EM = emerging markets; FI = fixed income; JPM EMBI+ = J.P. Morgan Emerging Markets Bond Index Plus; MSCI EM-Free = MSCI
Emerging Markets Free Index.
Source: Based on data from HFR.

©2003, AIMR® www.aimrpubs.org • 11

Investment Counseling for Private Clients V
Figure 1. Fixed-income strategies, as shown in Panel
B, are less closely related than market-neutral strate-
gies but are still more closely related than the uni-
verse as a whole. This wider range is understandable,
given that various fixed-income strategies differ
greatly in their exposure to equity prices, interest rate
volatility, and so on.
Equity strategies, as shown in Panel C, are rela-
tively well clustered but with a higher than expected
level of dispersion. The “equity market-neutral”
index, the market-neutral universe, appears in Panel
A but clearly does not belong in the equity strategy
world, despite its name. The higher level of disper-
sion evident in Panel C is likely the result of returns
being above the long-term norm for the last 10-year
period. And as would be expected in such an envi-
ronment, short sellers had unusually poor perfor-
mance.
Sector-specific equity strategies, as shown in
Panel D, share little commonality. In my view, these
Figure 3. Risk–Return Scattergram Composite of All Strategies, 10-Year
Period Ending 31 December 2002
Return (%)
25
20
15
10
0
0 5 10
Market Neutral
Sector Specific
Emerging Markets
Source: Based on data from HFR.
12 • www.aimrpubs.org
strategies lack similarity in their risk–return profiles
because no single tendency dominates their perfor-
mance. Finally, Panel E shows emerging market strat-
egies, which exhibit a relatively high level of
commonality. Differences in this universe are most
pronounced between region-specific strategies, such
as Latin America and Asia.
Now, consider these strategy clusters in a com-
posite view, as shown in Figure 3. Most strategies fall
within well-defined groups with common risk–
return characteristics, although there are exceptions.
As noted earlier, the sector-specific cluster, denoted
by the dotted line, is extremely large—stretching
from the lower left to upper right of the scattergram—
and exhibits only a minimal degree of commonality.
In contrast, the market-neutral and fixed-income
clusters have relatively tight dispersions. The obvi-
ous conclusion is that hedge funds constitute not a
homogeneous universe but rather a conglomeration
Nasdaq
S&P 500
Russell 2000
15 20 25 30 35
Risk (%)
Equity
Macro and Market Timing
Fixed Income
©2003, AIMR®
 A New Perspective on Hedge Funds and Hedge Fund Allocations
of distinct strategies whose common characteristics
lead to similar risk and return outcomes, with wide
dispersion at times when the features of the strategies
differ dramatically. Thus, the best approach in an
asset allocation framework that includes hedge funds
is to think of them as a set of distinct manager strate-
gies rather than an asset class.
A Different Approach
What makes hedge funds different from more tradi-
tional active strategies such as mutual funds? I pro-
pose that we think of all active strategies as having
two components: the risk–return profile of the base
asset class and the overlay of the active strategy of the
manager.
Hedge funds are simply an extension of the active
management continuum. At one end of the contin-
uum is the strategy of buying the index—a totally
passive strategy. Managers move along the contin-
uum away from a passive approach either to reduce
risk versus the index or increase return versus the
index. Traditional active managers and hedge fund
managers take much the same approach with the
fundamental asset—but with two notable differences.
First, the structure of the investment, typically a lim-
ited partnership with an initial lock-up period, means
that hedge funds are considerably less liquid than
traditional investment structures. Second, both tradi-
tional and hedge fund managers can buy attractive
securities, but hedge fund managers can also short
unattractive securities. As they manage the balance
between attractive and unattractive securities, they
can also manage their net exposure to the market
index. Some manage the exposure in a systematic,
top-down manner; others take a bottom-up approach.
For example, a market-neutral strategy can be
viewed as a cash or short bond portfolio to which the
manager adds security selection alpha. This analogy
is attractive both intuitively and theoretically. The
approach taken to eliminate market risk in a market-
neutral strategy is effectively a hedge. Imagine a long
General Motors/short Ford Motor Company pairs
trade. The alpha the manager expects to capture is not
the excess return that General Motors will earn rela-
tive to the equity market or the auto sector as a whole
but simply its performance relative to Ford. Another
market-neutral strategy is a convertible arbitrage
trade, in which a manager trades the difference
between the expected return on a company’s convert-
ible bond and its stock. Yet, another market-neutral
strategy is a capital structure arbitrage in which the
manager might be long a company’s high-yield bond
and short its underlying equity.
©2003, AIMR®
Regardless of the market in which the market-
neutral strategy is pursued, the strategy encompasses
the two components I outlined earlier: the risk–return
profile of the base asset class and the risk–return
profile of the active strategy. The manager uses the
active strategy to generate alpha. In the case of an
arbitrage strategy, the manager can generate alpha if
she accurately judges that the spread between the
expected returns of two assets varies from the norm,
she locks in that spread, and the spread does, in fact,
revert to a more normal relationship over a reasonable
time period. For example, over the past five years,
fixed-income arbitrage has proven to be an extremely
poor strategy because fixed-income investors were
unwilling to take risk during that time, preferring
high-quality issuers. Thus, although fixed-income
arbitrageurs saw abnormally wide spreads that were
expected to revert to historical levels, spreads actually
grew wider by the day. Recently, spreads have started
to shrink, but the damage has been done. The point is
that a manager’s alpha depends on the ability to find
an attractive spread, trade on it, and be right.
Consider a few other strategies in light of this
perspective. In a hedged equity strategy, the portfolio
manager has a long equity portfolio and a short
equity portfolio, with some net exposure to the mar-
ket if the market value of the two portfolios differs.
The average net exposure for long–short equity man-
agers is about 35 percent net long over time, suggest-
ing some directional bias. The manager’s alpha
(added value) is driven by his ability to correctly pick
winners and losers as well as on the portfolio’s net
exposure to the market. That net exposure may
depend on the available opportunities (a bottom-up
approach) or the manager’s view of the market (a top-
down approach).
Fixed-income arbitrage strategies also require
the manager to pick winners and losers. In addition,
however, they involve decisions on how much inter-
est rate risk to take and how much exposure to other
market features, such as convexity, or prepayment
risk in a mortgage portfolio, is appropriate given
anticipated market conditions.
Once the base asset class that best represents a
hedge fund manager’s strategy is identified, the next
step in the forecasting process is twofold—(1) forecast
return and risk for the asset class and (2) estimate
expected manager alpha and tracking error, as shown
in Table 1. (Note that this capital market forecast is
not intended to be definitive. The point is to illustrate
the process, not to argue the accuracy of the esti-
mates.) The method for generating these forecasts
should be consistent with those used for other, non-
hedge fund strategies, and it is the same approach that
should be used for a traditional large-cap equity fund,
www.aimrpubs.org • 13
Investment Counseling for Private Clients V

Table 1. Example Forecasts for Asset Class Return, Manager Alpha, and Manager Tracking Error
Expected Return Expected Risk
Manager Manager
Strategy Base Asset/Strategy Total Base Alpha Total Base Tracking Error
Hedge fund strategy
Various arbitrage Short bonds 9.5% 3.5% 6.0% 4.5% 2.0% 4.0%
Event driven Short bonds 10.5 3.5 7.0 6.3 2.0 6.0
Fixed income Bonds 9.0 5.0 4.0 7.2 4.0 6.0
Equity long–short 35% equity 13.1 5.1 8.0 10.5 5.4 9.0
U.S. equity Equity 12.0 8.0 4.0 17.0 15.0 8.0
Emerging markets Emerging market composite 14.0 10.0 4.0 21.5 20.0 8.0
Equity sectors Equity 13.0 8.0 5.0 19.2 15.0 12.0
Macro/market timing Short bonds 11.5 3.5 8.0 10.6 3.5 10.0
Managed futures Cash 5.0 1.0 4.0 8.1 1.0 8.0

Traditional strategy
U.S. fixed income Bonds 5.5 5.0 0.5 4.1 4.0 1.0
U.S. large-cap equities
Indexed Equity 7.8 8.0 –0.3 15.0 15.0 0.5
Core Equity 9.0 8.0 1.0 15.1 15.0 2.0
Active Equity 11.0 8.0 3.0 16.2 15.0 6.0

incorporating forecasts of the equity market return
and the particular manager’s alpha and tracking error.
Comparing Hedge Funds and
Mutual Funds
To further evaluate the difference between hedge fund
and traditional managers, I performed an analysis of
tracking error—the measure that managers have been
taught is their risk—for equity and fixed-income
mutual funds using the Morningstar database. The
ideal data for this analysis would have been the
monthly return stream of every mutual fund in that
universe. Although the information is available, the
amount of data (for about 1,700–2,000 funds) would
be overwhelming for a firm as small as mine, which
employs only two people. I therefore chose a simpler
approach that I will explain.
The Morningstar database provides not only the
statistics for each mutual fund but also its best-fit
index (i.e., the index that best correlates with the
manager’s strategy). With a best-fit index and an R2
between the performance of the manager and the
best-fit index, we could infer the relative tracking
error of the mutual fund; a higher R2 implies a lower
tracking error.
E q u i t y F u n d s . For U.S. large-cap equity
funds, we divided the universe into four broad
categories—traditional index funds, active/core
funds, concentrated portfolios, and mandates—as
shown in Table 2. The first category, traditional index

Table 2. Index Correlation and Risk–Return Analysis for Traditional

U.S. Large-Cap Equity Mutual Funds for Two Periods Ending
31 December 2002
Correlation with
Number S&P 500 5 Years 10 Years
Fund Type of Funds R2 >= R2 < Return Risk Return Risk
Traditional index 199 0.99 1.00 –1.55% 19.34% 8.61% 16.84%
Active/core
Subgroup 1 292 0.95 0.99 –1.78 17.05 7.12 15.43
Subgroup 2 437 0.90 0.95 –1.76 17.37 7.07 15.55
Subgroup 3 470 0.80 0.90 –1.93 18.53 7.23 15.88
Subgroup 4 382 0.65 0.80 0.02 16.91 8.32 15.27
Concentrated 102 0.50 0.65 1.61 17.05 10.16 16.09
Opportunistic 70 0.00 0.50 2.45 17.19 11.27 16.82
Source: Based on data from Morningstar.

14 • www.aimrpubs.org ©2003, AIMR®

 A New Perspective on Hedge Funds and Hedge Fund Allocations

funds, has an R2 of 0.99 or more. The next category,
active/core funds (which is made up of four
subgroups based on their observed R2), includes
managers who seek to match the index or beat it by
up to 2 percent, with a tracking error usually less than
4 percent. The third category, funds classified as
“concentrated” (with an R2 between 0.65 and 0.5),
only holds about 40–45 stocks, a relatively concen-
trated portfolio when the fund’s goal is to match or
beat the S&P 500.
Finally, the “mandates,” or opportunistic funds,
category is a residual group, with an R2 of less than
0.5. Remember, all the funds selected for this analysis
had a large-cap focus and had the S&P 500 as their
best-fit index (according to Morningstar). The funds
in this fourth category have a low correlation with
the S&P 500 because they might be weighted 80
percent in technology one year and 80 percent in
energy the next year. Their managers are the ulti-
mate stock pickers.
In Panel A of Figure 4, this universe of mutual
funds is presented on a risk–return basis, with the
three equity hedge fund strategies (hedged [long–
short], unhedged, and market neutral) from Panel C
of Figure 2 added for comparison. For the five-year
period ending December 2002, the unhedged equity
hedge fund strategy, the opportunistic mutual funds,
and the concentrated mutual funds have similar risk–
return profiles. The opportunistic and concentrated
mutual fund strategies exhibited roughly the same
level of risk as the unhedged equity hedge fund, with
a positive, but much lower, return.
The 10-year picture shown in Panel B of Figure 4
better illustrates the similarity in risk and return of
the concentrated and opportunistic mutual funds to
the unhedged equity hedge fund strategy. Over the
period, all the active mutual fund managers closely
tracked the equity index mutual fund in terms of risk.
And four of the mutual fund strategies actually had
a lower return than the equity index mutual fund. In
my view, this analysis shows that equity hedge
funds, in exhibiting similar risk–return profiles to
traditional active equity managers, are simply an
extension of the active equity management universe,
not a separate asset class.
Fixed-Income Funds. In Table 3, the same cat-
egorizing analysis using R2 values is applied to the
Morningstar fixed-income universe. Over the five-
year period ending December 2002, as shown in
Panel A of Figure 5, the best a fixed-income hedge
fund manager could hope for was to produce the
same return as a traditional manager but with sub-
stantially higher risk. Fixed-income arbitrage was
clearly the worst strategy to follow during this
period. Remember, I have defined a fixed-income
hedge fund manager as one who looks for an attrac-
tive spread—for example, between two credit
issues—that is expected to change in the future.
Given this definition, the story depicted in Panel A
makes sense because, over the past five years, the
major feature in fixed-income markets has been the
continual widening of credit spreads. Any manager
who bet that spreads would narrow back to tradi-
tional levels would have performed poorly.
The 10-year results shown in Panel B of Figure 5
reveal a different story. Fixed-income portfolios
range from short to long durations so that the fixed-
income indexes that are available in terms of “best
fit” are quite limited. During this period, the “highest
tracking error” fixed-income mutual funds were
those with short- to intermediate-term portfolios. The
active management contribution to the fixed-income
manager’s performance was actually a bit misleading
in that the manager’s relative performance was not
greatly affected by security-selection skills. Instead,
the duration of the portfolio made the greatest differ-
ence in performance.

Table 3. Index Correlation and Risk–Return Analysis for Traditional U.S.

Fixed-Income Mutual Funds: Two Periods Ending 31 December 2002
Search Criterion 5 Years 10 Years
Number
Fund Type of Funds R2 >= R2 < Return Risk Return Risk
Traditional index 6 0.99 1.00 7.16% 3.58% 7.29% 3.95%
Active/core
Subgroup 1 86 0.95 0.99 6.66 3.69 6.75 4.07
Subgroup 2 153 0.90 0.95 6.35 3.59 6.62 3.96
Subgroup 3 51 0.80 0.90 6.27 3.27 6.19 3.49
Subgroup 4 29 0.65 0.80 6.11 2.95 6.14 2.87
Concentrated 8 0.50 0.65 6.52 2.87 7.37 3.06
Opportunistic 4 0.00 0.50 5.73 2.27 5.84 2.62
Source: Based on data from Morningstar.

©2003, AIMR® www.aimrpubs.org • 15

Investment Counseling for Private Clients V

Figure 4. Risk–Return Scattergrams for Various Equity Hedge Fund

Strategies and Traditional Mutual Funds, December 1992–
December 2002
A. Equity Hedge Fund Strategies (5 years)
Return (%)
14

12 Equity
Hedged
10

8 Equity
Equity Unhedged
Market
6
Neutral
4

2 Equity Highest
Equity Concentrated Tracking Error
0

2 Equity Active/Core Equity

Index Fund
4
0 5 10 15 20 25
Risk (%)

B. Equity Hedge Fund Strategies and Traditional Mutual Funds (10 years)
Return (%)
18

16 Equity
Equity
Hedged
14 Unhedged

12 Equity Highest Tracking Error

Equity
Market Nasdaq
10 S&P 500 Equity Concentrated
Neutral
Equity
8
Salomon Index Fund
BIG
6 Equity Active/Core
EM Debt/Equity Composite
4

2
0
0 5 10 15 20 25 30 35
Risk (%)
Note: EM = emerging markets.
Source: Based on data from Morningstar.

An important caveat, as I mentioned at the begin-
ning of the presentation, is that all of my analysis is
based on historical data. Thus, I have no idea whether
my findings will hold in the future. The differences
between the 5- and 10-year periods suggest that the
patterns I have discussed are not always stable. Per-
haps an even longer time period would provide more
convincing evidence of the similarity in hedge fund
and other active management strategies.
Another caveat is that Morningstar includes only
those managers with a complete 10-year perfor-
mance history. Managers who went out of business
during the period are not included. A manager who
takes on more business risk (in this case, more poten-
tial for tracking error) relative to his competitors is
more likely to “guess wrong” and consequently go
out of business. His poor performance will not be
included in the analysis of the 10-year period. This

16 • www.aimrpubs.org ©2003, AIMR®

 A New Perspective on Hedge Funds and Hedge Fund Allocations

Figure 5. Risk–Return Scattergrams for Various Fixed-Income Hedge Fund

Strategies, December 1992–December 2002
A. Recent 5 Years

Return (%)
8
Fixed-Income
7 Index
Fixed-Income Concentrated
6 Fixed-Income
Fixed-Income Highest Fixed-
Fixed-Income Mortgage Backed Income Total
Tracking Error
5 Active/Core

3
Fixed-Income
2 Arbitrage

0
0 1 2 3 4 5 6 7
Risk (%)

B. 10 Years
Return (%)
12

10 Fixed-Income Mortgage Backed

Fixed-Income Total
Fixed-Income
8
Index
Fixed-Income Concentrated
Fixed-Income
6 Fixed-Income Highest Arbitrage
Tracking Error Fixed-Income
Active/Core
4

0
0 1 2 3 4 5 6
Risk (%)

Source: Based on data from Morningstar.

means that my analysis of the 10-year period is likely
biased in favor of higher-risk strategies. Because the
sample includes all winners but not all losers, it show
a higher expected return and a lower risk than would
have been experienced by someone invested in the
average manager.
given asset class, what are the fundamental differences
between hedge funds and other types of strategies?
Essentially, the differences are the two that I mentioned
earlier—manager risk (which comes in part from the
ability to sell short and from a tendency to hold fewer,
more concentrated positions) and lower liquidity.

Manager Risk. The ability to sell short and,

Manager Risk and Liquidity more generally, the freedom to pursue a wider variety
If hedge funds truly represent nothing more than an of strategies mean that hedge fund and mutual fund
extension of the active management spectrum for a managers take on different kinds of risk. Consider the

©2003, AIMR® www.aimrpubs.org • 17

Investment Counseling for Private Clients V

return distribution statistics for 1990 through 2002, as
shown in Table 4. (Note that this 12-year period is not
selected to be convenient for my purposes; it is the
longest monthly data series I could obtain for hedge
funds.) Two differences are apparent between the
hedge and nonhedge strategies. First, the return per
unit of risk tends to be higher for hedge fund strate-
gies than for the various indexes. Second, and more
significant, hedge fund strategies exhibit substan-
tially more kurtosis than the indexes (i.e., their return
distributions have fat tails).
Consider the return distributions of the lower-
volatility hedge fund strategies shown in Figure 6
compared with the Salomon BIG Index. Using the
returns for the 12-year period, I let the computer
decide what buckets should be used to create a his-
togram based on the distribution of the Salomon BIG
Index. Then, I forced all the other indexes (hedge and
nonhedge) into those buckets and graphed them
together. I have two observations to make. First, I
created an average of all market-neutral strategies
because I wanted to prove a point—for a large
enough number of managers and a large enough
number of strategies, the high kurtosis may be
reduced. (I will return to this thought later.) Second,
the other two strategies graphed in the histogram—
equity hedge and fixed-income average—exhibit
huge tails, which means that accidents will happen
more frequently than one might anticipate.
Now consider my point about normality and the
relationship between kurtosis and manager risk—the
traditional way to measure manager risk is tracking
error to the index. In fact, the more decisions a

Table 4. Selected Hedge Fund Return Distribution Statistics, 1990–2002

Market-Neutral Equity Fixed-Income Equity Salomon
Measure Strategies Composite Hedge Average Sector Total Nonhedge BIG S&P 500 Nasdaq
Average return 0.93% 1.44% 0.67% 1.57% 1.28% 0.67% 0.87% 1.11%
Standard deviation 0.92% 2.70% 1.18% 4.15% 4.29% 1.09% 4.41% 7.52%
Return per unit of risk 1.009 0.533 0.570 0.378 0.297 0.616 0.197 0.147
Skew –1.522 0.147 –0.705 0.051 –0.490 –0.234 –0.458 –0.372
Kurtosis 7.198 1.152 3.036 2.654 0.421 0.099 0.445 1.076
Source: Based on data from Morningstar.

Figure 6. Distribution of Hedge Fund Monthly Returns, 1990–2002

(lower-volatility strategies)
Frequency
50

40

30

20

10

10
e
28

82

90

43
21
48

94

41

87

33

75

36

or
1.

1.

2.

3.
0.
2.

1.

1.

0.

0.

0.

2.

Return Basket (% range)

Average Market-Neutral Strategies Equity Hedge

Fixed-Income Average Salomon BIG

Source: Based on data from HFR.

18 • www.aimrpubs.org ©2003, AIMR®

 A New Perspective on Hedge Funds and Hedge Fund Allocations
manager makes, the more likely it is that he or she will
make both extremely good and extremely bad deci-
sions. In concentrated portfolios, the manager makes
fewer decisions but each decision is critical and thus
that much more important. Therefore, in concentrated
portfolios, extreme outcomes are more likely. It is also
likely, however, that more concentrated traditional
mutual funds would also exhibit a significant level of
kurtosis.
Liquidity. Hedge fund investments are consid-
erably less liquid than traditional mutual funds.
Hedge funds require investors to provide advance
notice of intended withdrawal. Withdrawals are usu-
ally allowed only once per quarter and sometimes
only once per year. Hedge funds also typically
impose lock-up periods, meaning that investors must
stay in the fund for a certain amount of time before
they can withdraw their investment.
In a traditional mean–variance framework, an
asset class is defined by three sets of important data—
return expectations, expected volatility, and expected
covariances (or correlations) with other asset classes.
In the same traditional framework, an investor’s pref-
erences, defining a preferred risk–return trade-off
and investment horizon, are used to select a point on
the efficient frontier (the set of portfolios that pro-
vides maximum expected return for each level of risk)
that specifies the appropriate portfolio allocation.
In practice, the selection of a portfolio is subject
to many other influences. For example, “agency
risk”—the possibility that the manager may care
more about job sustainability than the welfare of the
client—is a concern. The investor may also impose
constraints that reflect biases and preferences
described by behavioral finance. Furthermore, inves-
tors are exposed to what I call decision risk, which is
the risk of not unwinding a position at the point of
maximum pay. More important, the traditional
framework does not allow a real investigation of the
exposure to overall manager risk and overall portfo-
lio liquidity. This is true because the traditional
framework does not incorporate a distinction
between market and manager risk: All asset classes
or strategies are described in terms of total-strategy
expected return and risk. Thus, an investor cannot
explicitly trade lower manager risk in one strategy
for higher manager risk in another.
Measuring Manager Risk and Liquidity. I have
seen investors set an arbitrarily low allocation to
hedge funds and yet place their entire equity alloca-
tion with a single highly concentrated mutual fund.
Such a decision makes no sense because they are
accepting an extremely high level of manager risk in
©2003, AIMR®
the traditional world while being unwilling to accept
any manager risk in the hedge fund world. The only
reasonable explanation would be that mutual funds
trade daily and thus provide much greater liquidity
than hedge funds. But I do not believe these investors
are motivated by liquidity concerns. Rather, I believe
the process is driven by an incomplete understanding
of the world of hedge funds and thus by prejudice
and unhealthy labeling.
An alternative approach is to evaluate each strat-
egy for its exposure to both expected market and
manager risk. The idea is to evaluate the combined
components of base asset class and manager strategy.
To determine how much market risk is taken relative
to manager risk, consider the expected risk columns
in Table 1.
Table 5 shows the squares of the two risk fore-
casts that provide a breakdown of the relative expo-
sures. For two assets, the variance of a portfolio
composed of those two assets is equal to the
weighted-average sum of the variances of each of the
assets plus an additional component that is calcu-
lated by multiplying the correlation coefficient
between the two assets, their portfolio weights, and
their respective standard deviations. I have made one
critical assumption for the sake of simplicity, but it is
not necessary for the method to produce correct
results. I have assumed that no correlation exists
between market and manager risk. If the extent to
which a manager outperforms his or her market is
independent of the market’s rise or fall, the correla-
tion part of the equation is eliminated and the vari-
ance of the manager’s return is equal to the sum of
the squared volatility of the asset class plus the
square of the manager’s tracking error.
The “various arbitrage” strategy provides a good
example of this calculation. The strategy has a base
(asset class) risk of 2 percent and a manager expected
tracking error of 4 percent. The square of these is
0.0004 and 0.0016, respectively. Thus, I calculate that
the strategy’s risk allocation is 20 percent market/80
percent manager. Table 5 shows that hedge funds
tend to have more manager risk than market risk,
whereas the reverse is true for traditional strategies.
A similar approach can be used with respect to
liquidity, as shown in Table 6. Gauging liquidity
requires estimating the time needed to liquidate the
portfolio. The most important information in Table 6
is the column labeled “Effective Liquidity Period,”
which I define as the average number of days needed
to find liquidity without causing a market impact.
Hedge funds have a much longer effective liquidity
period than traditional funds.
www.aimrpubs.org • 19
Investment Counseling for Private Clients V

Table 5. Estimated Market vs. Manager Risk

Expected Risk Strategy Risk Allocation
Manager
Tracking Tracking
Strategy Base Asset/Strategy Total Base Error Base Risk Error Market Manager
Hedge fund
Various arbitrage Short bonds 4.5% 2.0% 4.0% 0.0004 0.0016 20.0% 80.0%
Event driven Short bonds 6.3 2.0 6.0 0.0004 0.0036 10.0 90.0
Fixed income Bonds 7.2 4.0 6.0 0.0016 0.0036 30.8 69.2
Equity long–short 35% equity 10.5 5.4 9.0 0.0029 0.0081 26.5 73.5
U.S. equity Equity 17.0 15.0 8.0 0.0225 0.0064 77.9 22.1
Emerging markets Emerging markets composite 21.5 20.0 8.0 0.0400 0.0064 86.2 13.8
Equity sectors Equity 19.2 15.0 12.0 0.0225 0.0144 61.0 39.0
Macro/market timing Short bonds 10.6 3.5 10.0 0.0012 0.0100 10.9 89.1
Managed futures Cash 8.1 1.0 8.0 0.0001 0.0064 1.5 98.5

Traditional
U.S. fixed income Bonds 4.1 4.0 1.0 0.0016 0.0001 94.1 5.9
U.S. large-cap equities
Indexed Equity 15.0 15.0 0.5 0.0225 0.0000 99.9 0.1
Core Equity 15.1 15.0 2.0 0.0225 0.0004 98.3 1.7
Active Equity 16.2 15.0 6.0 0.0225 0.0036 86.2 13.8

Table 6. Estimated Hedge Fund Liquidity

Required Notice Liquidation Effective Liquidity
Fund Type Period Period Period Liquidity Factor
Hedge fund
Absolute return Quarter Annually 180 50.7%
Semidirectional Quarter Annually 180 50.7
Directional Quarter Annually 180 50.7
Fund of funds Quarter Quarterly 135 63.0

Traditional
Individual portfolios
U.S. fixed income None 1 Week 7 98.1
U.S. large-cap equity None 1 Day 1 99.7
U.S. small-cap equities None 1 Week 7 98.1
EAFE equities None 1 Day 1 99.7
Emerging markets equity None 1 Week 7 98.1
Commingled funds Two weeks None 15 95.9
Mutual funds None 1 Day 1 100.0
Note: The “Liquidity Factor” is calculated by dividing the number of days required to liquidate the
portfolio by the number of days in a year and can thus be seen as the fraction of any year during which
liquidity is lost.

Optimizing. In addition to the tracking error,
return, and covariance, I now have a vector that
shows exposure to manager and market risk and an
estimate of expected liquidity. With such informa-
tion, an optimizer, whether a mean–variance or a
more complicated mathematical form, can be used to
calculate the whole portfolio’s exposure to manager
and market risk as well as expected liquidity for any
portfolio. Constraints can be set on maximum allow-
able manager risk and required minimum liquidity
(rather than setting an arbitrary limitation on hedge
funds), requiring that optimal portfolios must have
a manager risk exposure lower than the level
established by the investor and higher liquidity than
the minimum set by the investor. Hedge fund invest-
ment will still be limited by these constraints, but this
decision-making approach correctly focuses on the
investor’s goals and needs rather than simply being
an arbitrary decision.

20 • www.aimrpubs.org ©2003, AIMR®

 A New Perspective on Hedge Funds and Hedge Fund Allocations

When a traditional optimizer is applied with the
liquidity and manager risk constraints, it leads to a
barbell allocation. As shown in Figure 7, the portfolio
will seek liquidity from traditional sources and take
on manager risk from other sources (including hedge
funds) to provide the highest possible alpha. If the
binding constraint is manager risk, most of the equity
exposure will be in index or index-like strategies and
most of the active portfolio management will be in
hedge fund-type strategies. Sometimes, problems of
liquidity arise, in which case the binding constraint
that limits the hedge fund allocation will be required
liquidity rather than maximum manager risk. Interest-
ingly, this is the same structure that I encountered
earlier in trying to construct tax-efficient portfolios
using a traditional approach focused on after-tax
return and risk—a barbell structure that promoted
high tax efficiency at the expense of manager alpha
and high alpha strategies at the expense of tax effi-
ciency. Both approaches thus converged on the con-
clusion that tax-efficient and manager-risk/liquidity-
efficient portfolios are often best achieved through
barbell structures.
Conclusion
The hedge fund universe is broadly viewed as a
homogeneous asset class for purposes of asset alloca-
tion. But hedge fund management is an extension of
traditional active management with several impor-
tant differences—less liquidity; the ability to short
and to have higher levels of portfolio concentration
(in terms of securities, sectors, or other risk character-
istics), resulting in higher levels of manager risk; and
manager-risk-induced kurtosis. Constraints on the
allocation to hedge funds in the asset allocation pro-
cess should be set not because hedge funds are a
separate asset class, which, in my view they are not,
but rather to achieve the desired portfolio exposure
in terms of manager risk and the investor’s liquidity
needs, together with the usual concern for efficiency
in the trade-off between after-tax return and risk for
the whole portfolio.

Figure 7. Optimal Asset Allocation with Manager Risk and Liquidity

Constraints
Frequency
60

50

40

30

20

10

0
e
66
4

51

65

81

12

27
97

82

96

or
.4

.2

.1

3.

1.

0.

2.

7.

9.
14

12

10

7.

5.

4.

Return Bucket (% range)

Equity Nonhedge S&P 500

Sector Total Nasdaq

©2003, AIMR® www.aimrpubs.org • 21

Investment Counseling for Private Clients V
Question and Answer Session
Jean L.P. Brunel, CFA
Question: Given the problems
with hedge fund reporting, how
can private wealth managers have
confidence that reported results
are credible?
Brunel: Take that question and
substitute corporate financial
reporting for hedge fund report-
ing. I don’t mean to be flippant,
because this issue is truly impor-
tant. My point is that reporting
problems are not limited to hedge
funds.
I tend to serve families that are
extremely wealthy. My average
family has close to $1 billion and
yet, as for most families, we are
constrained to using only funds of
funds. I know that it is extremely
painful to pay 1 percent in addition
to what the hedge fund managers
are charging, but I see that 1 per-
cent as a reasonable insurance pol-
icy. This approach, however, will
not avoid all catastrophes.
For example, Morgan Stanley
has a team with a long history of
investing in hedge funds. They
came from the Weyerhaeuser pen-
sion fund, which until 1999, I think,
was the best-performing pension
fund in the United States for the
previous 15 years and was also
invested entirely in alternative
22 • www.aimrpubs.org
assets. Yet, this team managed to
stumble twice over the last year.
One position basically blew up and
the other, whose problems have
just been discovered, has not yet
produced audited returns for 2001
(and we’re in 2003).
Mistakes will occur, but by
investing in funds of funds and
doing a serious amount of due dil-
igence, one can rely on the assump-
tion that the reported returns make
some sense. If you can understand
what a hedge fund does and
develop a good level of comfort
with the fund-of-funds manager, I
believe the risks are manageable.
A large amount of discipline
and quite a bit of homework go a
long way toward providing confi-
dence. In this case, as in many other
situations, remember that if it looks
too good to be true, it probably is.
Be prepared to invest in, say, mar-
ket-neutral funds that generate a
long-term return of about 10–11
percent. If a fund says it has pro-
duced much more, ask yourself
how it could continue to do so over
the long term.
Question: What situations do
you see as red flags when you are
considering a new manager?
Brunel: One well-known fund
uses extremely high diversifica-
tion, with more than 100 securities
plus leverage. I do not understand
how this fund has continued to
generate a high double-digit
return. Even last year, the fund was
in the 15 percent range. I will not
invest in this fund, despite the fact
that I cannot point at anything it
does wrong. I am simply not confi-
dent that the fund can continue to
produce such results. If you stay
away from the funds with extraor-
dinary returns and focus more on
the funds with sustainable long-
term, low-double-digit returns,
you will probably do well.
Being cautious, however, can
lead to its own problems. I recently
made a big mistake in this regard.
A mortgage arbitrage fund gave me
all the access I wanted and allowed
me to spend a half-day sitting at its
desk so that I could see not only the
whole portfolio but also all of the
trade variations it uses. But I am an
equity jockey by trade and because
I simply could not understand the
nature of these backward, forward,
interest-only, and so forth transac-
tions, I could not be convinced to
pull the trigger. Last year, this fund
had a 22 percent return.
©2003, AIMR®
Tax Implications of Hedge Fund
Investments
Edward H. Dougherty
Senior Tax Manager
Deloitte & Touche LLP
New York City
Different investment vehicles have different implications for individual, corporate, and
tax-exempt investors. Hedge funds in particular pose significant tax considerations. In
order to achieve the benefits of tax planning for U.S.-based investors, a thorough
understanding of the major differences in tax treatment of three common hedge fund
structures (U.S. master-feeder, offshore master-feeder, and side-by-side) and private
foreign investment companies is necessary.
edge fund investments have varying tax impli-
H cations for their investors, depending on
whether the investor is an individual, a corporation,
or a tax-exempt entity, such as a pension fund. Each
investor has a unique tax situation. In turn, the struc-
ture of the hedge fund itself has tax ramifications that
can affect each type of investor differently. A hedge
fund can be structured as a partnership, a limited
liability company (LLC), or a foreign corporation.
Accordingly, for prospective and current hedge fund
investors, understanding the potential tax impact of
an investment is imperative.
My presentation begins with a brief review of the
fundamental tax rules governing investors. I will
then examine the after-tax performance reporting
requirements for hedge funds and mutual funds and
provide a comparison of the types of funds in terms
of tax implications and other issues. This analysis will
lead into a comparative discussion of the partnership
versus corporation hedge fund structures and an
explanation of the significant features of the three
primary multi-entity hedge fund structures (U.S.
master-feeder, offshore master-feeder, and side-by-
side). Finally, I will describe a variety of ways that
hedge funds can engage in tax planning for their
investors.
Tax Fundamentals of Investors
In the United States, an individual’s tax rate varies
with the type of income being taxed. Ordinary
©2003, AIMR®
income and short-term capital gains are currently
taxed at 35 percent.1 Long-term capital gains (hold-
ing period greater than one year) are taxed at a pref-
erential rate of 15 percent.2 Individuals cannot carry
back capital losses to previous years but have an
unlimited carryforward period. Throughout the pre-
sentation, when I talk about tax-advantaged income,
I usually mean long-term capital gains, but other tax
attributes are important to investors as well. For
example, although U.S. government bond interest is
not exempt from federal income tax, it usually con-
fers a benefit for state and local income tax purposes.
In the United States, a Subchapter C corporation
pays tax at the marginal rate of 35 percent on all
income and has a three-year capital loss carryback
and a five-year capital loss carryforward. In contrast
to the flow-through entities of a Subchapter S corpo-
ration, LLC, or partnership (whose individual items
of income or loss, as well as deductions, flow through
to the individual shareholder or partner for reporting
and taxation and are not taxed at the entity level), a
1 Note that the Jobs and Growth Tax Relief Reconciliation Act of
2003 (the “2003 Tax Act”) was passed by the U.S. Congress and
signed by President George W. Bush in May 2003. The Act reduced
the top marginal tax rate for individuals to 35 percent; at the time
this presentation was made, the top marginal rate was 38.6 percent.
The reduction in the ordinary tax rate is retroactive to 1 January
2003 and is scheduled to “sunset” after 2010.
2
The 2003 Tax Act reduced the top tax rate applicable to long-term
capital gains from 20 percent to 15 percent, effective for gains
recognized after 6 May 2003. The reduction in the long-term capital
gains rate is scheduled to sunset after 2008.
www.aimrpubs.org • 23
Investment Counseling for Private Clients V
C corporation is taxed on its net income at the entity
level and makes distributions of this net income to its
owners in the form of a dividend, which is taxed
again when reported on the shareholder’s tax
return.3
A tax-exempt entity in the United States, such as
a pension fund or an endowment, contrary to the
implication of its name, is subject to tax on unrelated
business taxable income (UBTI), which can arise from
two sources. The first is income from a trade or a
business unrelated to an entity’s tax-exempt purpose.
The second is income generated through the use of
leverage in a trading portfolio. Therefore, before a
tax-exempt entity invests in a hedge fund, it should
find out whether the fund expects income to be gen-
erated from either of these sources, and if so, the
choice of investment vehicle needs to be reconsid-
ered. As I will explain later, a tax-exempt entity gen-
erally would prefer to invest in a hedge fund
structured as a non-U.S. corporation because the for-
eign corporation would serve as a “blocker” for
UBTI.
Fundamentals of Hedge Fund
Structures
As I mentioned earlier, the income of a partnership
or LLC is not taxed at the entity level; rather, the entity
is treated as a conduit, and all of the income or loss it
generates flows through to the owners (or investors,
in the case of hedge funds) in the year of recognition.
(Typically, LLCs are treated as partnerships for tax
purposes.)
The big draw for hedge fund managers is the
performance fee they can generate for themselves. The
partnership flow-through structure is particularly
valuable in this regard. Typically, in a partnership
fund structure, a hedge fund manager will charge
two types of fees: a management fee (which is a
percentage of assets under management) and an
incentive or performance fee (which is based on any
positive return the manager generates for the portfo-
lio). Although performance fees are standard in the
hedge fund arena, some portfolio managers are
required to achieve a predetermined return, com-
monly referred to as a “hurdle rate,” before they can
collect the performance fee.
3
The 2003 Tax Act also reduced the tax rate applicable to “Qualified
Dividend Income” (QDI); prior to passage of the 2003 Tax Act,
dividends were taxed at the same rate as ordinary income (i.e., 38.6
percent). Under the new law, dividends that meet certain tests will
be considered QDI and will be taxed at the long-term capital gains
rate of 15 percent. This provision of the law is scheduled to sunset
after 2008.
24 • www.aimrpubs.org
The partnership structure gives fund managers
(general partners) the ability to take performance fees
as a profit allocation rather than as fee income. A
hedge fund’s return contains tax-advantaged compo-
nents, such as long-term capital gains and unrealized
gains, which retain their character when they are
passed through in a profit allocation. Structuring the
performance fee as a profit allocation within a part-
nership also reduces the investors’ adjusted gross
income (AGI). This can be an advantage because
itemized deductions on the investors’ individual
returns are limited at higher levels of AGI.
If a hedge fund is structured as a corporation, the
performance fees earned by the portfolio managers
(along with fund expenses) are simply a deduction in
determining the net income, or return, due the share-
holders, or investors, in the fund. Such fee income is
taxable to the fund manager as ordinary income
taxed at the highest marginal rate. If the corporation
is an offshore corporation, some opportunity exists
to defer recognition of the income from performance
fees for the portfolio managers. This practice is com-
mon in the hedge fund industry.
After-Tax Performance Reporting
and Tax Planning
In February 2002, the U.S. SEC began requiring
mutual funds to report their performance on an after-
tax basis for 1-, 5-, and 10-year periods. This after-tax
performance must be calculated for the highest indi-
vidual tax rate (currently 35 percent).
At the present time, hedge funds are not simi-
larly required to report performance on an after-tax
basis. After-tax reporting, as well as tax planning,
would be extremely cumbersome for a hedge fund
because it would entail accounting for all the differ-
ent stakeholders in the fund—namely, high-net-
worth (taxable) individual investors, corporations
that may be invested as general partners or limited
partners, and tax-exempt investors. In the hedge
fund world, the two most common ways of measur-
ing the after-tax performance of an investor are the
current tax liability divided by economic income and
the taxable income divided by economic income. For
individual investors, the first method makes more
sense because it takes into account the preferential
rate for long-term capital gains and other tax-
advantaged income. Neither method takes into
account the future tax liability (benefit) potentially
arising from unrealized gains (losses) in the fund.
A hedge fund can incorporate tax planning as
part of its overall investment strategy, but to do so
imposes two types of costs on the fund: transaction
costs and opportunity costs. For example, if a hedge
©2003, AIMR®
 Tax Implications of Hedge Fund Investments

fund manager chooses to execute a basket, or total Figure 1. U.S. Domestic Master-Feeder Hedge
return, swap to avoid triggering capital gains in the Fund Structure
portfolio, the investment bank with whom the swap
U.S. Investors Non-U.S. Investors
is structured is going to want a piece of the action— and U.S. Tax-Exempt
a transaction cost for the hedge fund. And because Investors
this tax-planning activity, the basket swap, reduces
the time that the manager would otherwise spend
working to maximize the portfolio’s market-related U.S. Cayman
performance, an opportunity cost is also incurred by Partnership Corporation
(Feeder) (Feeder)
the hedge fund. Do investors really want their port-
folio managers distracted from managing the portfo-
lio on an economic basis so that they can engage in
tax planning instead?
Whether hedge funds will continue to pursue
more aggressive tax-planning opportunities is an
open question. More fundamentally, should they U.S.
engage in tax planning in the first place? In my view, Partnership
(Master)
the answer has to be yes. But given that an opportu-
nity cost is involved, the best solution is for someone
other than the portfolio manager to handle tax plan-
ning. This person can be either internal or external, Figure 2. Offshore Master-Feeder Hedge Fund
perhaps a consultant, and should not be responsible Structure
for executing tax-related transactions. Rather this U.S. Investors Non-U.S. Investors
person should be charged with the task of determin- and U.S. Tax-Exempt
ing what tax-related transactions are feasible and Investors
how they should be implemented.

Hedge Fund Structures
Three common types of hedge fund structures exist—
the U.S. master-feeder, the offshore master-feeder,
and the side-by-side structure. The tax implications
differ for each depending on the type of investor. I
will briefly describe these differences and the reasons
a portfolio manager would choose one structure ver-
sus another.
U.S. Master-Feeder. The standard U.S. master-
feeder hedge fund structure is shown in Figure 1.
Two feeder funds exist—a U.S. partnership or LLC
that accepts investments from taxable U.S. investors
and an offshore corporation or partnership (often
domiciled in the Cayman Islands) that accepts invest-
ments from tax-exempt U.S. investors and non-U.S.
investors. Both feeders make investments in the mas-
ter fund, which is a U.S. partnership or LLC. All
trading activity occurs at the master level. Income
earned by the master fund is allocated to the two
feeder funds. U.S. source dividends earned by non-
U.S. investors in the offshore feeder are subject to a
30 percent U.S. withholding tax.
Offshore Master-Feeder. The offshore master-
feeder structure, as shown in Figure 2, is similar to
the U.S. master-feeder, except that the master fund is
domiciled offshore, often in the Cayman Islands. As
U.S. Cayman
Partnership
(Feeder)
Corporation
(Feeder)
Cayman
Partnership
(Master)
with the U.S. structure, all trading takes place at the
master level and income is allocated between the two
feeder funds. In a typical offshore fund, the master
fund is structured as a corporation under local law,
but an election in the U.S. tax code allows investors
to treat the structure as a partnership. This is called
the “check-the-box election” and is made by most
U.S. investors (for reasons that I will explain shortly).
Offshore master-feeder structures are subject to the
same withholding issues as U.S. structures, that is to
say that U.S. source dividends earned by non-U.S.
investors in the offshore feeder are subject to a 30
percent U.S. withholding tax.
Side-by-Side (Mirror). The side-by-side fund,
also called a “mirror” or “clone fund,” is shown in

©2003, AIMR® www.aimrpubs.org • 25

Investment Counseling for Private Clients V
Figure 3. In this structure, the trading activity takes
place in two different vehicles. Typically, the invest-
ment vehicle for U.S. taxable investors is a U.S. part-
nership or LLC, whereas the investment vehicle for
tax-exempt U.S. investors and non-U.S. investors is
an offshore corporation.
Comparison of Structures
Multiple considerations, both tax and nontax, are
involved in choosing the structure a hedge fund
wishes to use, as shown in Exhibit 1. Tax planning
and tax-sensitive investments are facilitated with the
side-by-side structure. When the master-feeder struc-
ture is used, however, a conflict arises between the
tax-planning needs of taxable U.S. investors and the
lack of such a need on the part of both non-U.S. and
tax-exempt U.S. investors. For example, under the
new U.S. dividend tax rules, U.S. taxable investors
would prefer their stock not be loaned so they can
potentially earn qualified dividend income (QDI),
whereas non-U.S. and tax-exempt investors, who do
not qualify to earn QDI, would prefer their stock be
loaned to generate additional income. So, the funda-
mental question is: Should a master fund engage in
tax planning, even though such planning offers no
benefit to non-U.S. and tax-exempt investors? In a
side-by-side structure, targeted tax planning can be
done in the domestic partnership or LLC without
affecting non-U.S. and tax-exempt U.S. investors.
All three structures provide incentive-fee flexi-
bility to the fund manager. Incentive fees can be taken
as a profit allocation from the master fund, which is
a partnership. More commonly, however, the fees are
taken from the feeder funds, where they are struc-
tured as a profit allocation from the domestic partner-
ship and as a straight fee from the offshore
corporation. The latter approach is preferred because
the investment manager wants to retain the ability to
Figure 3. Side-by-Side (Mirror) Hedge Fund Structure
Investment
Manager LLC
U.S. Limited Partners
Cayman Corporation
U.S. Domestic (Investment Vehicle)
Limited Partnership
(Investment Vehicle)
26 • www.aimrpubs.org
make elections that allow the manager to defer recog-
nition of the income and thus the payment of the tax
liability associated with the performance fees earned.
All three structures provide for the ability to
eliminate UBTI and passive foreign investment com-
pany (PFIC) issues for U.S. tax-exempt and taxable
investors. As I stated earlier, tax-exempt investors
typically invest in the offshore (Cayman) corporation
of the offshore master-feeder because the corporate
structure blocks the flow-through of tax attributes to
the investors. Thus, if a hedge fund manager uses a
leveraging strategy, the offshore corporation blocks
the UBTI-tainted income that would otherwise flow
to the tax-exempt investor. A similar advantage is
provided to U.S. taxable investors who can invest in
the U.S. partnership of the offshore master-feeder
and thus avoid dealing with the complex PFIC rules.
Two non-tax-related business issues arise with
the side-by-side structure that generally do not occur
with either of the master-feeder structures. The first
issue is trade allocation. In a master-feeder structure,
all of the trading activity takes place in the master
fund, so no trade allocation issues are involved. The
economic income is allocated to the feeder funds at
the end of the accounting period based on initial
investment levels, which is the job of the administra-
tor or the fund accountants. But in a side-by-side
structure, at the end of each day, the trades made by
the investment manager must be allocated between
the domestic fund and the offshore fund. The second
issue is that side-by-side structures do not provide for
economies of scale in terms of account aggregation.
PFICs
Under U.S. tax law, a foreign corporation will be
treated as a PFIC if either 75 percent of its income or
50 percent of its assets are deemed to be passive.
U.S. Tax-Exempt
Investors Non-U.S. Investors
©2003, AIMR®

Exhibit 1. Tax and Nontax Considerations of
Different Hedge Fund Structures
Non-U.S.
Tax/Nontax Side- Master- U.S. Master-
Consideration by-Side Feeder Feeder
Target tax efficiency Yes No No
Make tax-sensitive
investments Yes No No
Incentive-fee flexibility Yes Yes Yes
Avoid UBTI and PFIC
issues for investors Yes Yes Yes
Avoid trade allocation
issues No Yes Yes
Account aggregation No Yes Yes
Passive income includes interest, dividends, capital
gains, rents, and royalties, and passive assets are those
that generate such passive income. Most offshore
hedge funds structured as corporations are PFICs.
Prior to 1986, high-net-worth individuals could
put their money in an offshore corporation and avoid
paying tax on any income from the investment until
they disposed of it. This changed in 1986, when PFIC
rules were enacted to establish what is essentially an
antideferral regime. These rules are designed to dis-
suade investors from deferring recognition of the
income earned in a passive investment vehicle.
The income earned by an investor in a PFIC can
be taxed in one of three different ways.
Qualified Electing Fund. The first (and most
common) way that the income from a PFIC invest-
ment is taxed is under the qualified electing fund
(QEF) regime. In this case, a U.S. investor makes an
election on his or her income tax return to pay tax on
a current basis on the ordinary income and net capital
gains from the offshore corporation (as if the corpo-
ration were a partnership).4 In other words, the
investor does not take advantage of any deferral
opportunities that are provided because the foreign
entity is structured as a corporation. This election can
only be made, however, if the offshore fund issues to
each U.S. investor an annual information statement
that details the investor’s share of ordinary earnings
and net capital gains generated by the fund during
the year. The election is made at the U.S. investor
level and cannot be made by the offshore corporation
itself. Without such a statement, the QEF election is
not permitted and taxes cannot be paid on a current
basis. This statement must also include an agreement
by the fund to allow the investor and the U.S. IRS to
inspect its records so that the income can be verified.
4
Note that capital gains and losses do not flow through to the
investor as they do in a partnership—retaining their individual
character—but are instead netted.
©2003, AIMR®
Tax Implications of Hedge Fund Investments
Excess Distribution. The second way that
income from a PFIC can be taxed is under the excess
distribution regime. In this case, investors are taxed
on a PFIC investment when they receive a distribu-
tion in the form of a dividend or when they receive
cash from the redemption or sale of shares. Because
an implicit deferral has occurred (the investor has not
been declaring the income on a current basis), there
is an interest charge in addition to the tax liability.
The rules are extremely complex, but in short, they
are intended to determine over the investor’s holding
period when and how much tax the investor would
have paid on a current basis. Because of these com-
plex calculations, investors will generally make the
QEF election if they can get a PFIC annual informa-
tion statement from the fund.
Mark to Market. The investor can also make a
mark-to-market election with respect to his or her
PFIC stock if the stock is traded on an organized
exchange. Thus, an investor who is not issued an
annual information statement can still be currently
taxed on the PFIC investment. Unrealized gain is
treated as ordinary income and unrealized loss is
treated as an ordinary loss.
Mutual Funds vs. Hedge Funds
Mutual funds and hedge funds vary in their tax
implications for investors, as shown in Exhibit 2. The
first column in Exhibit 2 represents mutual funds, the
second and third columns represent trader hedge
funds and investor hedge funds, respectively, and the
fourth column represents PFICs. The difference
between a trader hedge fund and an investor hedge
fund is determined by the strategy of the fund—that
is, whether the fund is focused on active trading or
on making investments for long-term capital appre-
ciation. Most tax issues are the same for both types of
hedge funds.
Current Taxation. A shareholder of a mutual
fund generally pays tax on a current basis on his or
her share of all the taxable income that is earned by
the mutual fund. Similarly, an investor in a hedge
fund structured as a partnership will pay tax on a
current basis on his or her pro rata share of income
as reported on his or her Schedule K-1. For a PFIC,
current taxation depends on whether the PFIC inves-
tor makes a QEF election. If the election is made, tax
is paid on a current basis. If no election is made, there
is no current tax liability; however, when a dividend
is paid or when an investor redeems shares, taxes
plus an interest charge will be due.
www.aimrpubs.org • 27
Investment Counseling for Private Clients V

Exhibit 2. Comparison of Tax Attributes for Mutual Funds and Hedge Funds
Trader Hedge Investor Hedge
Tax Attribute Mutual Fund Fund Fund PFIC
Current taxation Yes Yes Yes Depends
Mandatory cash distribution Yes No No No
Flow-through of tax attributes Modified Yes Yes Modified
Capital gains flow-through Yes Yes Yes Modified
Foreign tax credit flow-through Maybe Yes Yes No
Expenses reduce AGI Yes Yes No Yes
Incentive fee as profit allocation No Yes Yes No
Blocking of UBTI taint Yes No No Yes
Partial sale taxable Yes No No Yes
Securities distributions for tax
planning No Yes Yes No
Flow-through of losses No Yes Yes No

Mandatory Cash Distribution . The only one
of these four investment vehicles that provides for
mandatory cash distributions is a mutual fund. This
distinction is important. Prospective hedge fund
investors should be aware that, although they must
pay tax each year on the allocated income from the
hedge fund reported on their Schedule K-1, they will
not necessarily receive a cash distribution with which
to pay the associated tax liability. Some partnership
agreements, however, do stipulate that the hedge
fund will make a cash distribution just for this pur-
pose. Other partnership agreements stipulate that to
get cash out of the investment, an investor must
request a redemption 45 or 90 days in advance. In
many cases, the partnership agreement gives the gen-
eral partner the discretion to deny the withdrawal
request. Before investing in a hedge fund, investors
need to satisfy themselves that they will have the
liquidity necessary to pay taxes on the hedge fund
investment.
Flow-Through of Tax Attributes . The flow-
through of tax attributes is limited for mutual funds
because it only applies to three types of income:
ordinary earnings (including net short-term capital
gains), net long-term capital gains, and now QDI.
Under certain circumstances, U.S. government inter-
est is also specified as a pass-through item. For a
hedge fund structured as a partnership, all of the tax
attributes flow through to the investor. Similar to the
treatment of a mutual fund, a PFIC reports only
limited income information—the investor’s share of
ordinary earnings and net capital gains; other tax
attributes do not pass through to the PFIC investor.5
5
It is not clear at this time whether QDI income will retain its
character when earned by a PFIC. Under the 2003 Tax Act, divi-
dends paid by a PFIC will not be treated as QDI for a U.S. individ-
ual.
Capital Gains Flow-Through. All of these four
structures provide for the pass-through of long-term
capital gains. To determine the amount of long-term
gains that pass through to a PFIC investor, however,
net short-term capital losses are netted against long-
term capital gains rather than against ordinary
income.
Flow-Through of Losses. Losses generated by
mutual funds and PFICs do not flow through to the
investor except as a reduction to ordinary income
(short-term losses) or capital gains (long-term losses).
In contrast, losses generated by a partnership (trader
or investor hedge fund) do flow through to the inves-
tor in the period in which they are generated and, in
retaining their character as losses, are deductible on
the investor’s individual income tax return, whether
on page 1 of Form 1040 or on Schedule A (Schedule
of Itemized Deductions) of Form 1040.
Foreign Tax Credit Flow-Through . Foreign
withholding taxes (that is, foreign from the stand-
point of a U.S. investor) can only be passed through
a mutual fund to investors if 50 percent or more of the
assets in the mutual fund are international in nature.
In contrast, all foreign withholding taxes paid by a
partnership flow through to the partner (investor) to
be claimed as a credit on the partner’s Form 1040.
Foreign withholding taxes do not flow through to the
investor in a PFIC, but are an adjustment to ordinary
earnings—a deduction before the income is reported
to the investor. This treatment does not provide a
benefit to the taxpayer equal to that of a tax credit.
Expenses Reduce AGI . Expenses in a mu-
tual fund reduce the ordinary dividend that the fund
pays to its investors, thus reducing the individual
investor’s AGI. Similarly, with a trader hedge fund,
all fund expenses reduce the investor’s AGI because
they are treated as associated with the trade or

28 • www.aimrpubs.org ©2003, AIMR®


business of the hedge fund and are reportable as a
reduction to AGI on page 1 of the investor’s individ-
ual income tax return, Form 1040. But an investor in
an investor hedge fund must report the expenses of
the fund on Schedule A of Form 1040 rather than as a
reduction to his or her AGI. Such expenses are treated
as a miscellaneous itemized deduction, meaning that
these expenses can be deducted for most high-net-
worth investors only if they exceed 2 percent of AGI.
So, for an investor hedge fund, the deduction of
expenses is effectively lost because the investor’s
share of the expenses is seldom large enough to over-
come the 2 percent hurdle. Expenses do reduce the
ordinary earnings that flow through to the investor of
a PFIC, so they effectively result in a reduction in the
investor's AGI.
Incentive Fee as a Profit Allocation. An in-
centive fee is not typically taken by a mutual fund,
but if it were, it would be treated as an expense of the
fund without the favorable tax attributes associated
with the profit allocation treatment of a partnership.
As I stated earlier, the incentive fee associated with
hedge funds is the main attraction from the point of
view of the hedge fund manager. And the fact that
incentive fees can be treated as a profit allocation in
a trader hedge fund or an investor hedge fund, either
of which can be structured as a partnership, makes
these structures particularly appealing to managers.
Likewise, any fees generated by a PFIC are consid-
ered to be like any other expense and cannot be
treated as a profit allocation. As with the other
expenses of a PFIC, the performance fee reduces the
ordinary earnings that flow through to the investor
of a PFIC, so they effectively result in a reduction in
the investor’s AGI.
Blocking UBTI Taint. If a mutual fund or a PFIC
generates UBTI, a tax-exempt investor will not suffer
any negative results because the corporate shell
blocks the pass-through of UBTI. For a hedge fund
with a partnership structure, however, the UBTI taint
passes through to the investor so that tax-exempt
investors would then be liable for tax on the tainted
income.
Partial Sales Taxable . Selling shares in a
mutual fund, whether a partial or a complete posi-
tion, is a taxable event. Usually, the investor has
received dividend distributions and capital gains dis-
tributions over the investment period, which (if rein-
vested) increased his or her basis in the shares, so the
capital gains realized at liquidation and thus the tax
due at liquidation are not likely to be great. A partial
redemption of a partnership interest in a hedge fund,
however, is subject to fundamentally different
©2003, AIMR®
Tax Implications of Hedge Fund Investments
income recognition rules. Under U.S. tax law, a cash
withdrawal from a partnership is not a taxable event
unless the withdrawal exceeds the investor’s tax
basis in the investment. If cash withdrawn exceeds
basis, gain is recognized. In no event is a loss gener-
ated on a partial redemption. The partial or complete
sale of a PFIC investment is a taxable event, much like
it is for a mutual fund.
Securities Distributions for Tax Planning.
Technically, using securities distributions to fund
withdrawals is allowed for mutual funds, but for a
large mutual fund, distributing securities to satisfy
withdrawals is not practical. The practice is common
in hedge funds that are structured as partnerships,
and it provides some tax-planning opportunities,
notably as a tool to take highly appreciated securities
out of the fund. This type of distribution is typically
not done by PFICs.
Other Business Issues. The other business
differences between hedge funds and mutual funds
are summarized in Exhibit 3.
■ Short selling and leverage. Hedge funds are
free to engage in short selling, whereas mutual funds
typically may do so only in a limited manner. Also,
most mutual funds have only a limited ability to use
leverage or margin borrowing, whereas hedge funds
have a much greater ability to use leverage to gener-
ate additional returns for investors.
■ Incentive fee income/ability to defer. The mutual
fund fee structure usually does not include an incen-
tive or performance fee component—a standard in
the hedge fund industry. The ability of managers
under a partnership structure to take profit alloca-
tions for their performance fees allows hedge fund
managers to share in tax-advantaged income, such as
long-term capital gains, but always on a current-
period basis. The only structure that provides a fund
manager with the ability to defer recognition of fee
income is the PFIC structure.
■ After-tax performance and reporting. Currently,
a mutual fund is the only vehicle required to provide
after-tax performance reporting to the SEC, although
in the future, certain hedge funds may also be
required to follow suit.
■ U.S. tax filing/privacy. No U.S. tax filing is
required by a PFIC. A PFIC may provide an annual
information statement to its investors (in order for
investors to make the QEF election), but this state-
ment is not required to be filed with the IRS. In
contrast, mutual funds, as well as hedge funds struc-
tured as partnerships, are closely monitored by the
IRS through required annual filings. The superior
privacy offered by PFICs explains why most non-U.S.
www.aimrpubs.org • 29
Investment Counseling for Private Clients V

Exhibit 3. Comparison of Business Differences between Mutual Funds and

Hedge Funds
Trader Hedge Investor Hedge
Business Practice Mutual Fund Fund Fund PFIC
Short selling allowed Limited Yes Yes Yes
Leverage/borrowing Limited Yes Yes Yes
Manager earns incentive fee No Yes Yes Yes
Manager earns management fee Yes Yes Yes Yes
Manager can defer incentive fee No No No Yes
Manager can take fee as profit
allocation No Yes Yes No
After-tax reporting required by
SEC Yes No No No
Privacy maintained No No No Yes
U.S. tax filings Yes Yes Yes No

investors prefer to invest through a PFIC or a corpo- include tax-lot identification, use of broad-based
rate investment vehicle. stock index options, constructive-sale avoidance
through total return swaps and long-term hedging,
the doubling up of a position to avoid the wash-sale
Tax Planning for Hedge Fund rules, forward conversion transactions, collars, and
Investors special allocations to partners (investors) withdraw-
Different types of investors have different tax consid- ing from the partnership.
erations, as shown in Exhibit 4. High-net-worth indi- Tax-Lot Identification. The most basic form of
viduals and corporations are interested in tax-
tax planning is tax-lot identification. Generally, if a
planning arrangements that defer the recognition of
hedge fund sells a partial position and it has multiple-
gains or accelerate the recognition of losses. High-
cost lots of the stock, the first lot purchased is the one
net-worth individuals who invest in an investor
reported as the sale. Unfortunately, the first lot gen-
hedge fund rather than a trader hedge fund are con-
erally has the lowest cost basis and thus triggers the
cerned about the deductibility of expenses. Tax-
highest capital gain. At Deloitte & Touche, we advise
exempt entities typically make investments through
our clients to minimize the gain or maximize the loss
offshore vehicles and thus are interested in tax plan-
ning that minimizes withholding taxes on dividends by designating the lot(s) with the highest cost basis
(and UBTI, if they are not investing through a foreign as the lot(s) being sold.
corporate vehicle). Only high-net-worth individuals Broad-Based Stock Index Options. Broad-
care about transactions that convert ordinary income based stock index options, such as the S&P 500 Index
and short-term capital gains into long-term capital
contract, are referred to in the tax law as Section 1256
gains, because they are the only investors who enjoy
contracts and are taxed at a preferential rate. By
the commensurate benefit.
statute, 60 percent of any gain from one of these
Several common and relatively straightforward
contracts is treated as long-term, even if the contract
tax-planning techniques used by hedge funds
was held for only a day, and the remaining 40 percent
is treated as short-term regardless of the holding
period. Therefore, an index fund interested in tax
Exhibit 4. Tax-Planning Considerations for efficiency for the benefit of its investors may opt to
Different Hedge Fund Investors use 1256 contracts rather than buying all of the secu-
High-Net- rities in the index to take advantage of the preferen-
Tax-Planning Worth Tax-Exempt tial tax treatment.
Consideration Individual Corporation Entity
Gain deferral Yes Yes No Total Return/Basket Swaps. A total return
Loss acceleration Yes Yes No swap is typically not subject to the constructive-sale
Expense deductibility Yes No No rules (which address gain acceleration) and the wash-
Withholding taxes No No Yes sale and straddle rules (which address loss deferral).
Leverage/UBTI issues No No Yes For example, when a stock is sold at the end of the
Gain conversion (to year for the express purpose of harvesting losses as
long-term capital) Yes No No part of the fund’s tax planning and the fund

30 • www.aimrpubs.org ©2003, AIMR®


purchases the same stock within 30 days of, either
before or after, the date of sale, the wash-sale rule will
cause the realization of the losses to be deferred until
the next lot of the same stock is sold. The solution is
a basket swap designed to provide a return that is
expected to emulate the return of the stock that was
sold as if the stock had been repurchased by the fund.
For tax purposes, this arrangement is typically not
treated as a wash sale; thus, deferral of the loss is not
required.
Doubling Up. Doubling up is another planning
strategy that is typically not subject to the wash-sale
rules. For example, suppose that a hedge fund is long
100 shares of General Motors (GM) with an unreal-
ized loss in the position. The manager wants to real-
ize the loss but also wants to continue to hold GM
from an economic standpoint. The doubling-up tech-
nique would involve buying another 100 shares of
GM and selling the original lot 31 or more days after
the date of the second purchase to trigger the loss.
Forward Conversion Transaction. If the expo-
sure to the extra 100 shares of GM for the 31-day
period is unacceptable, another approach that is not
typically subject to wash-sale treatment is a forward
conversion transaction. The investor buys a put and
writes a call at the same strike price on the second lot
of GM stock that is purchased. Economically, the
result is the same as if the second block of stock was
simultaneously bought long and sold short. The net
exposure is still only 100 shares of GM. After 31 days,
the original lot is sold (triggering the tax loss) and the
option positions are liquidated. The change in the
value of the stock, the call, and the put over the 31-
day period will essentially cancel each other out
(ignoring transactions costs and time decay in the
options).
Collar. Another common tax-planning strategy
is a collar transaction, which is used to hedge a signif-
icantly appreciated stock without triggering gain rec-
ognition. In this case, the investor writes a call and
©2003, AIMR®
Tax Implications of Hedge Fund Investments
buys a put. The income from writing the call pays for
(or helps to pay for) the cost of the put. As long as
there is a difference between the strike prices of the
call and the put (we typically recommend at least a 20
percent difference), this transaction will not trigger
gain recognition under the constructive-sale rules.
Partnership Special Allocations. Prospective
investors in hedge funds should carefully read the
partnership agreements that govern the hedge fund
investment they are contemplating in order to under-
stand how the economic and tax allocations to the
partners are handled. For example, many partner-
ship agreements provide that partners who with-
draw some or all of their capital get a special
allocation of income to eliminate the difference
between the book basis and the tax basis of their
partnership interest. The effect is to give more income
to the partner who is withdrawing and reduce
income allocated to the continuing partners. Of
course, disposing of the entire interest in a partner-
ship is always a taxable event. The opportunity for
tax planning is in determining the kind of income that
is allocated to the withdrawing partner—long-term
capital gains or a mix of income types. Investors need
to focus on these provisions in the partnership agree-
ment, particularly if they are not likely to be in the
partnership for the long term.
Conclusion
The tax implications of a particular investment
vehicle—in this case, mutual funds, hedge funds, and
PFICs—differ for individual, corporate, and tax-
exempt investors. As hedge fund investing increases
in popularity with all types of investors, a thorough
understanding of the major differences in the tax
treatment of three common hedge fund structures
(U.S. master-feeder, offshore master-feeder, and side-
by-side) as well as PFICs is needed to achieve the
benefits of tax planning under current U.S. tax law.
www.aimrpubs.org • 31
Investment Counseling for Private Clients V
Question and Answer Session
Edward H. Dougherty
Question: What would a capital
loss carryback mean for a corpo-
rate investor?
Dougherty: If a corporation
experienced a capital loss in the
current year, the corporation could
use the loss to offset income in the
prior three years and potentially
get a tax refund. Of course, the cap-
ital loss carryback is available only
for Subchapter C corporations. In
contrast, a Subchapter S corpora-
tion is a pass-through entity; any
losses, whether capital or operat-
ing, flow through to the owners of
the S corporation (who can only be
individuals) and are thus taxed on
the shareholder’s individual
return. The loss carryback is not
applicable for a capital loss in the
case of an S corporation because
individuals have no carryback
ability.
Question: Is it possible to trigger
the wash-sale rule twice?
Dougherty: The wash-sale rule is
a loss-deferral rule. In effect, the
deferred loss adjusts the tax basis
of the new security. When the new
security is sold, the loss becomes
deductible (unless you buy back
32 • www.aimrpubs.org
the security within the 30-day
period).
This question may refer to the
desire to convert long-term losses
into short-term losses. On an indi-
vidual’s Form 1040 income tax
return, long-term gains are netted
against long-term losses and short-
terms gains are netted against
short-term losses and then net
long-term and net short-term are
netted together. If the ultimate
result is a net long-term capital
gain, then the preferential tax rate
of 20 percent (15 percent for sales
after 6 May 2003) applies. So, all
things being equal, individuals
would prefer to have long-term
gains and short-term losses.
We do use a technique for con-
verting long-term losses to short-
term losses. It works as follows:
You sell a long-term position (with
a holding period longer than one
year) for a loss. You buy a call
option on the same security, which
triggers the wash-sale rule. The
loss is disallowed, thus adjusting
the basis of the option. Then, you
exercise the option and take deliv-
ery of the underlying stock.
According to the way the tax rules
are written, the loss is embedded in
the stock position, which has a new
holding period. You can dispose of
the stock the next day and generate
a short-term loss.
Question: Does the Cayman
Islands impose a withholding tax
on assets domiciled there?
Dougherty: No. The Cayman
Islands imposes no withholding
taxes that are applicable to hedge
funds, which is one of the reasons
why so many hedge funds are
domiciled in the Cayman Islands.
Question: In light of new mutual
funds billing themselves as hedge
funds, what investment restric-
tions do registered funds have that
private funds do not?
Dougherty: The most significant
difference is that most registered
funds are limited in their ability to
engage in short selling and thus
cannot use that method to manage
risk. Mutual funds might, in cer-
tain circumstances, be able to use
derivatives to achieve the effect of
short sales, but rarely do they
engage in outright short selling
because of investment limitations;
this limits their risk management
opportunities.
©2003, AIMR®
Capturing Tax Alpha in the Long Run
Christopher G. Luck, CFA
Partner
First Quadrant, LP
Pasadena, California

For taxable investors, tax alpha (the value added from tax management) is far more
valuable than pretax alpha. The benefits of tax management strategies, such as loss
harvesting and HIFO (highest in, first out) accounting, can be quantified by using Monte
Carlo simulations that are based on various assumptions about tax rates, return,
volatility, and so on. Regardless of the market environment, a portfolio managed in a tax-
efficient manner should consistently outperform a buy-and-hold portfolio.

onte Carlo simulations are an excellent tool for
M analyzing strategies designed to maximize
risk-adjusted after-tax return. Although this presen-
tation focuses on equity portfolios, the principles are
applicable to any asset class. In essence, our approach
at First Quadrant is simple. Instead of using historical
data to evaluate a strategy, we make assumptions
about various parameters—return, volatility, tax
rates, dividends, and so forth—and apply these
assumptions in Monte Carlo simulations to deter-
mine the strategy’s after-tax performance under a
wide range of possible outcomes.
Tax Alpha
For taxable clients, the goal is to maximize risk-
adjusted after-tax return. Unfortunately, the focus
(both for mutual funds and separate accounts) is
typically on pretax returns. For active managers, the
real challenge is balancing uncertain stock alpha ver-
sus relatively certain tax alpha.
Tax alpha is the value added from tax manage-
ment. In other words, it is the difference between a
manager’s pretax added value and after-tax added
value. There are three value-adding components to
any process focused on taxable management. First,
managers can use the traditional method to generate
pretax excess return through superior stock selection.
Second, they can use risk management to minimize
Editor’s note: This presentation uses the tax rates and assumptions
of a U.S. investor.
tracking error and extreme outcomes. Third, they can
use tax management to minimize or eliminate taxes.
Tax management itself has three main compo-
nents: managing taxes on dividends, deferring capi-
tal gains taxes, and harvesting losses. Currently,
dividend management is a relatively unimportant
source of tax alpha because dividends tend to be low
in both the U.S. and non-U.S. equity markets. Because
the greatest source of negative tax alpha tends to be
realization of capital gains, gain deferral is a very
important source of tax alpha. The opposite of realiz-
ing capital gains is realizing capital losses, loss har-
vesting, which is also a significant source of tax alpha.
At First Quadrant, we use Monte Carlo simulations
to evaluate the benefits of all three sources of tax
alpha.
Pretax alpha is difficult to achieve. Even an
extremely skilled manager has only a relatively mod-
est advantage over the market. Generating tax alpha,
however, is a far easier task. The reason is that divi-
dend rates and unrealized capital gains and losses are
known with almost complete precision. Thus, esti-
mating the tax hit or the tax advantage of a particular
trade is straightforward.
This observation is not intended to be an argu-
ment against active management. At First Quadrant,
we believe in active management. But we also recog-
nize that adding value through stock selection is far
more difficult than adding value through tax man-
agement. Furthermore, we understand that once pre-
tax excess return is achieved, the challenge becomes
keeping it in the portfolio. Tax management is the key
to meeting this challenge.

©2003, AIMR® www.aimrpubs.org • 33

Investment Counseling for Private Clients V

Managing Taxable Assets total portfolio “utility,” including risk (and stock
selection, if appropriate).
There are two complementary approaches to tax
management: the accounting approach and the
investing approach. The accounting approach—the Monte Carlo Simulations
highest in, first out (HIFO) method—involves evalu- The Monte Carlo simulations discussed in this pre-
ating the tax impact of every trade by looking at the sentation are also presented in an article co-authored
applicable short- or long-term capital gains tax rate by several investment professionals at First Quad-
and picking the cost lot that offers the best tax out- rant.1 In a Monte Carlo simulation, assumptions
come. This is an accounting decision: Given multiple about the mean and standard deviation of a variety
blocks of the same stock with different costs pur-
of inputs are made. For each simulation, the com-
chased at different times, which one should be iden-
puter randomly generates values for each input
tified as having been sold? Thus, this approach does
based on the assumptions. The outcome of the simu-
not alter the underlying portfolio management. The
lation (in this case, the pre- and after-tax returns) is
second approach involves delaying the sale of a stock
determined based on the randomly generated inputs,
solely to avoid recognizing a capital gain (gain defer-
our fixed assumptions about the world (such as tax
ral) or selling a stock that has dropped in price solely
rates and dividend yields), and the particular strat-
to use the loss to negate a realized capital gain (loss
egy being followed. For our study, we ran a large
harvesting). In short, this approach means managing
number of simulations, and we examined the range
the portfolio (making trading decisions) so as to min-
of outcomes to determine how a particular strategy
imize the tax burden.
performs relative to other strategies.
At first glance, using the accounting approach
For the base case in our simulations, we assumed
seems to be an obvious choice. After all, any manager
an average annual benchmark return of 8 percent
who works with taxable clients should choose
with a standard deviation of 15 percent for the equity
accounting conventions that maximize tax savings,
market, consistent with historical norms. Part of that
particularly when this choice does not alter any port-
total return comes from a fixed annual dividend yield
folio management decisions. Applying the HIFO
of 1.44 percent (i.e., 0.12 percent a month, which is
method is essentially Pareto optimal; that is, it almost
close to the actual average dividend). We assumed a
always produces a tax cost that is less than or equal
tax rate of 35 percent; in other words, the assumption
to that of any other method.
was for a corporate entity, not an individual who
In the case of the investing approach, altering
faces both long- and short-term capital gains tax
investment decisions to minimize taxes is a funda-
rates. We created 300 months of results for each sim-
mental change for many managers. Deferring gains
ulation. The investment universe consisted of the 500
and harvesting losses clearly provides a tax benefit to
stocks in the benchmark. The return for each stock
the client, but it also introduces risk in the form of
was assumed to be the benchmark return plus a
tracking error. First, selling stock for the sole purpose
normally distributed random variable with a mean
of harvesting losses incurs transaction costs, which
of zero and a standard deviation of 9 percent, which
reduce pretax portfolio returns. Tracking error is typ-
again is consistent with the long-term mean and stan-
ically measured by comparing the pretax returns of
dard deviation of individual stocks in the S&P 500
the portfolio and its benchmark, so a reduction in the
Index. This means that each stock had the same
portfolio return becomes a component of tracking
expected return as the benchmark but a larger stan-
error. Second, from an active management point of
dard deviation. We assumed that each month, one
view, loss harvesting may lead to stock-selection
new company was added to the benchmark and one
shortfall over the 31-day wash-sale period (when the
was removed. Transaction costs were assumed to be
manager has to be out of the stock in order to be able
zero, and no wash-sale rule was in effect. The portfo-
to recognize the loss). The advantage, of course, is
that it can maximize client after-tax return. To maxi- lio was rebalanced monthly, and the tax alpha and
mize both tax savings and after-tax return, the loss dividends were reinvested in the portfolio quarterly.
harvesting/gain deferral approach requires a contin- No cash contributions or withdrawals were allowed
uous, disciplined, systematic process. To achieve after the initial cash contribution. We ran each Monte
maximum tax savings, a manager should dispose of Carlo simulation 500 times in an effort to show the
all stocks for which the tax benefit of the loss exceeds range of likely outcomes.
the related transaction costs. This discipline should 1
Andrew L. Berkin and Jia Ye, “Tax Management, Loss Harvesting,
be used continuously, not only at year-end or quar- and HIFO Accounting,” Financial Analysts Journal (July/August
ter-end. Furthermore, it should take into account the 2003).

34 • www.aimrpubs.org ©2003, AIMR®

 Capturing Tax Alpha in the Long Run

We used the base-case simulations to evaluate Figure 1. Cumulative Alpha for HIFO and Loss-
three tax management strategies: Harvesting Strategies
• Strategy 1. Buy-and-hold portfolio with average- A. HIFO
cost accounting. Cumulative Alpha (%)
• Strategy 2. Buy-and-hold portfolio with HIFO
16
accounting. 14
• Strategy 3. Tax-managed portfolio with HIFO 12
accounting and loss harvesting. 10
All three strategies resulted in a portfolio that fully 8
6
duplicated the benchmark at the end of each month. 4
The difference between Strategies 1 and 2 is purely 2
an accounting convention. The difference between 0
Strategies 2 and 3 is that in Strategy 3, when a stock 0 50 100 150 200 250 300
had to be sold (because it had left the benchmark) and Months
the sale resulted in a capital gain, a search was made 25th Percentile 50th Percentile
for other stocks that currently had unrealized losses. 75th Percentile
If available, these other stocks were sold, the loss was
realized, and the stock was immediately replaced at B. Loss Harvesting
its current market price. (Recall that we assumed zero Cumulative Alpha (%)
transaction costs and no wash-sale rule.) 180
We measured the value of the portfolio each 160
140
month in both gross and net terms. The gross value 120
was the market value of the shares held. The net value 100
was the gross value less any capital gains taxes that 80
60
would be paid if the portfolio was liquidated imme- 40
diately. 20
0
Interpretation of the portfolio values is easy. In 0 50 100 150 200 250 300
short, subtracting Strategy 1’s value from Strategy 2’s
Months
value shows the benefit of HIFO accounting. Sub-
25th Percentile 50th Percentile
tracting Strategy 2’s value from Strategy 3’s value
75th Percentile
shows the benefit of loss harvesting.
The 500 Monte Carlo simulations using the
parameters I have described produced the following
results: The median cumulative (over 300 months) lowing the index is apparent. Remember that this
tax benefit from HIFO accounting is 8.6 percent simulation is based on a monthly turnover rate of 2.5
before liquidation taxes and a little more than 4 per- percent, which is consistent with historical turnover
cent after liquidation taxes. The median cumulative in the S&P 500. The reason the curves rise over time
tax benefit from loss harvesting is 115 percent before is compounding; the tax benefit is invested back into
liquidation taxes and 58 percent after liquidation the portfolio.
taxes. Figure 2 shows that most of the benefit of loss
Although the use of HIFO accounting is essen- harvesting occurs early on. The reason is that we
tially Pareto optimal, it confers only a slight advan- assumed the benchmark (and each individual stock)
tage over using a non-tax-sensitive accounting increased in market value at an average rate of 8
method. If the management approach is changed in percent a year. Initially, our assumption that individ-
favor of gain deferral/loss harvesting, the tax benefit ual stocks had an independent standard deviation of
is much greater. Of course, these data are for the
9 percent over and above the standard deviation of
median of the 500 simulations. Panel A of Figure 1
the benchmark meant that quite a few stocks (out of
shows the cumulative benefit over time for the 25th,
50th (median), and 75th percentile simulation for the 500) would experience a few bad months in a row
the HIFO accounting strategy, and Panel B shows or even a bad year or two. Such losses can be har-
the same information for the tax-managed (loss- vested. After a few years, however, these initial losses
harvesting) strategy. will have been harvested and (because our assump-
In all cases, the tax-managed strategy provides a tion was a rising market) most other stocks will have
much larger benefit than the tax-accounting strategy. risen above their initial values. The portfolio will
The benefit of harvesting losses and not blindly fol- contain fewer and fewer unrealized losses.

©2003, AIMR® www.aimrpubs.org • 35

Investment Counseling for Private Clients V
Figure 2. Alpha of the Loss-Harvesting Strategy
before Liquidation Taxes
(annualized)
Annualized Alpha (%)
10
8
6 200
4 150
2 100
0
0 50 100 150 200 250
Months
25th Percentile 50th Percentile
75th Percentile
Because loss-harvesting activity will diminish
over time, its marginal added value will diminish.
After three years, its benefit is less than 2 percent a
year, and after five years, its benefit is less than 1
percent. Interestingly, after 25 years, the benefit is still
positive, although small. A manager will still be har-
vesting losses after 25 years, even though the bench-
mark is growing at 8 percent a year, for two reasons.
First, the dividends that flow into the portfolio every
year will be reinvested, creating blocks of each stock
that will have a generally higher cost basis. Even if,
on average, the cost basis of a particular stock is less
than its current market value, individual lots pur-
chased with dividends may be underwater. Second,
the benchmark has turnover. When turnover occurs,
the proceeds from the required sale are reinvested in
the new stock at its current market price. Some of
these new purchases will generate losses that can be
harvested.
Just as the tax benefit is greatest in the beginning,
turnover is also highest in the early years—more than
100 percent in the first year, as shown in Figure 3.
Clearly, little difference exists in turnover between
the 25th, 50th, and 75th percentile of the Monte Carlo
simulation distribution.
Stress Testing
After running the simulations with the base case, we
decided to examine how our results would change
when individual assumptions changed. In other
words, we wanted to know what would be the impact
of changing market conditions on after-tax liquida-
tion values.
First, we increased stock-specific volatility. One
could argue that the 9 percent stock-specific volatility
36 • www.aimrpubs.org
Figure 3. Turnover of the Loss-Harvesting
Strategy
(annualized)
Annualized Turnover (%)
350
300
250
50
0
300
12 60 108 156 204 252 300
Months
25th Percentile 50th Percentile
75th Percentile
in our base case is not representative of recent expe-
rience. We chose 9 percent because it is the average
for the S&P 500 over the past 20 years. When we
increased individual stock volatility to 10, 12, or 14
percent, after-tax liquidation values improved. The
reason is that significant loss harvesting continued
for a longer time. Increasing volatility resulted in a
wider distribution of stock returns and, correspond-
ingly, stocks owned at a larger amount of unrealized
loss—and thus more opportunities for tax mitigation.
Because greater stock-specific volatility leads to bet-
ter after-tax performance, tax management is more
valuable in such an environment. Tax management
matters less during periods of low stock-specific vol-
atility.
A second test was to assume that the range of
outcomes at the individual stock level would remain
unchanged but the benchmark itself would become
more volatile. In other words, cross-sectional stock
volatility held steady but total market volatility
increased. This assumption had virtually no effect on
after-tax performance. For purposes of tax manage-
ment, what matters is how volatile stocks are relative
to each other, not market volatility per se.
As I explained earlier, the base case assumed an
8 percent annual market return (1.44 percent from
dividends and 6.56 percent from capital gains). When
we increased the market return by increasing the
capital gains component, keeping the dividend rate
the same, the opportunities for loss harvesting dimin-
ished and the value of capital gains deferral
increased. Nonetheless, the compounding of loss-
harvesting benefits at a higher rate of return over 25
years led to superior performance. An important
point to note here is that the investment time horizon
©2003, AIMR®

matters for tax-advantaged strategies; the longer the
horizon, the greater the value of tax management.
Testing the sensitivity of tax alpha to the level of
market return produced some of our more interesting
results, as shown in Figure 4, which indicates the tax
alpha (after liquidation taxes) for an average market
return of 11, 8, and 5 percent. (Note that the data are
for the 50th percentile, or median, of the 500 simula-
tions run for each market return assumption.) In the
first 10 years, little difference in cumulative alpha is
apparent. At the 10-year point, however, the com-
pounding effect took over and the benefit of a higher
market return becomes clear.
Figure 4. Impact of Market Return: Median
Cumulative Alpha after Liquidation
Taxes
Cumulative Alpha (%)
140
120
100
80
60
40
20
0 greater the advantage for tax management. Interest-
0 50 100 150 200 250
Months
11% Average Market Return
8% Average Market Return
5% Average Market Return
Note: The data are for the 50th percentile, or median, of the 500
simulations run for each market return assumption.
We tried both increasing and decreasing the
benchmark turnover rate and observed relatively lit-
tle impact on after-tax returns, although higher
benchmark turnover, not surprisingly, provided a
slight benefit. The advantage is slight because higher
turnover leads to more gain realization, with the
offset that replacing high-gain stocks with new stocks
allows for more ability to harvest losses.
Increasing the dividend return while holding the
total benchmark return constant led to slightly higher
after-tax performance. The dividend return gener-
ated a little more cash, which was invested in the
portfolio at a current-cost basis year after year, pro-
viding more opportunities for loss harvesting. The
advantage was slight because the change in assump-
tion was slight—from 1.4 percent to 1.8 percent. A
larger increase would have led to a greater after-tax
advantage.
Allowing cash contributions over time had a
slightly positive effect on after-tax performance.
©2003, AIMR®
Capturing Tax Alpha in the Long Run
Receiving (or increasing) cash contributions is equiv-
alent to earning a higher dividend. Larger cash
inflows provide greater scope for loss harvesting. But
because additional cash also went into the passive
benchmark, the impact was extremely slight. Simi-
larly, allowing cash withdrawals over time also had
only a slight negative impact, leading to lower after-
tax performance. Using HIFO accounting to meet
withdrawals, however, always led to improved per-
formance relative to using average-cost accounting.
Of course, increasing the amount of withdrawals
would make an increasingly more significant differ-
ence because doing so would reduce the effective
time horizon, thus leading to a greater similarity
between the actively managed tax-aware portfolio
and the buy-and-hold strategy.
Finally, adding (or increasing) both contribu-
tions and withdrawals (i.e., a net neutral effect) led to
slightly higher after-tax performance. Interestingly,
the difference was greater than was the case for alter-
ing either contributions or withdrawals alone. Rather
than canceling out each other, these effects appear to
be additive.
Increasing the tax rate led to significantly higher
after-tax performance compared with the simple
buy-and-hold strategy; the higher the tax rate, the
300
ingly, the improvement was most pronounced in
moving from a low to a medium tax rate, as shown
in Figure 5. The benefit of tax management when
moving from the 20 percent to the 35 percent tax rate
is clearly greater than when moving from the 35
percent to the 50 percent tax rate. We tested only these
three tax rates, but we believe that they adequately
describe the range of investor exposure to tax liabil-
ity. A number of corporate institutions are subject to
a 20 percent rate. The 35 percent rate applies to other
Figure 5. Impact of Tax Rate: Median Cumulative
Added Value before Liquidation Taxes
Cumulative Value Added (%)
180
160
140
120
100
80
60
40
20
0
0 50 100 150 200 250 300
Months
50% Tax Rate 35% Tax Rate
20% Tax Rate
www.aimrpubs.org • 37
Investment Counseling for Private Clients V
corporate entities and is fairly descriptive of federal
individual rates. The 50 percent rate is relevant when
high state and local rates are added to federal rates.
Conclusion
The past three years in the equity markets have not
produced an environment conducive to superior tax
alpha. But when equity market returns are positive,
a tax-managed portfolio should consistently outper-
form a buy-and-hold portfolio. At First Quadrant,
we have published several articles finding that tax-
oblivious active management would have to deliver
tremendous alpha to overcome the after-tax hurdle
presented by the simple buy-and-hold approach.2
2 Robert D. Arnott, Andrew L. Berkin, and Jia Ye, “How Well Have
Taxable Investors Been Served in the 1980s and 1990s?” Journal of
Portfolio Management, vol. 26, no. 4 (Summer 2000):84–94; Robert
D. Arnott and Robert H. Jeffrey, “Is Your Alpha Big Enough to
Cover Its Taxes?” Journal of Portfolio Management, vol. 19, no. 3
(Spring 1993):15–26.
38 • www.aimrpubs.org
Tax management can also add a lot of value relative
to a simple buy-and-hold strategy but is often over-
looked.
Taxes matter greatly, especially over long time
horizons. Taxes amount to a heavy transaction cost—
whether a 35 percent short-term capital gains rate, a
35 percent corporate rate, or a 15 percent long-term
capital gains or qualifying dividend rate.
Unfortunately, most managers focus solely on
beating the benchmark on a pretax basis, which is an
uncertain game to play because consistently generat-
ing this sort of alpha is very difficult to do. Taxes are
perhaps the one aspect of investment management
that can be known with virtual certainty. With such
information, anyone who uses proper discipline can
dramatically reduce the impact of taxes. For taxable
clients, the focus should be on after-tax return.
Whether active or passive management is applied,
tax management can provide a tremendous benefit.
©2003, AIMR®

Question and Answer Session
Christopher G. Luck, CFA
Question: Your base-case sce-
nario assumes a positive market
return. Are there benefits from loss
harvesting in a negative market
return environment?
Luck: In the past three years, few
equity managers have had diffi-
culty avoiding capital gains. But
even when the market is down,
loss harvesting makes sense. The
benefit of loss harvesting, how-
ever, depends critically on the abil-
ity to balance a loss against a
realized gain. The constraint is the
client who insists that she doesn’t
need any more losses because her
portfolio has no gains to protect.
But if the client invests in hedge
funds, which seem to generate a lot
of short-term gains, she may well
benefit from consistent loss har-
vesting, even in down markets.
Overall, although a negative mar-
ket affords more opportunities for
loss harvesting, the compounded
benefit over a long-term horizon is
not particularly high if the market
is declining year after year.
Question: Is it better to harvest
tax losses throughout the year
rather than only at year-end?
Luck: Harvesting only at year-
end is better than doing no harvest-
ing at all, but if clients can actually
use losses, the opportunity set is
much larger if loss harvesting is
done throughout the year. For
example, think back to 1999, to a
market in which stock prices were
rising rapidly. A tax-oblivious
manager with a lot of capital gains
to offset might have decided to har-
vest losses at year-end but then
found few positions with losses
that could be harvested. In con-
trast, a manager who harvested
losses throughout the year would
have been better able to offset gains
with realized losses.
When the market is declining,
year-end versus constant harvest-
©2003, AIMR®
ing does not make much differ-
ence. When the market is rising, it
can make a dramatic difference.
Question: Is the major risk of loss
harvesting having to be out of the
security for 31 days, and have you
attempted to quantify this risk in
your simulations? Also, do you use
a market index to reduce tracking
error?
Luck: Switching from a different
process to active tax management
clearly introduces risks, particu-
larly higher transaction costs and
greater tracking error. Recognizing
a loss means being out of the stock
for 31 days before you can buy it
back, which tends to introduce
some tracking error for the portfo-
lio, especially for highly volatile
stocks.
When we test our portfolios,
we look at the tracking-error
impact of loss harvesting and try to
juggle the risk. For example, imag-
ine that you are holding Microsoft
Corporation at benchmark weight
and it is at a loss. Harvesting the
entire loss will provide a tax bene-
fit, but it will also introduce 50 bps
of tracking error. We probably
would start slowly reducing the
Microsoft position over time
because, although we do not want
to add 50 bps of tracking error, we
are willing to add 10 bps of track-
ing error in return for picking up
some of the tax benefit. You also
might buy calls and puts on
Microsoft to hedge your position
during the wash-sale period.
We tend not to buy index
options because it means buying
the index. Of course, you could buy
Standard & Poor’s Depositary
Receipts or derivatives. Such meth-
ods, however, do not offer a good
hedge against idiosyncratic risk
when selling Microsoft or some
other stock.
Capturing Tax Alpha in the Long Run
Question: At what level of turn-
over in a portfolio does tax man-
agement stop adding alpha, and if
it always adds alpha, at what level
are the expenses greater than the
value added?
Luck: Obviously, as portfolio
turnover increases, so do transac-
tion costs. In the base case I
described, we make an unrealistic
simplifying assumption—zero
transaction costs. So, in that sense,
the benefit I’ve reported is over-
stated. In actuality, managers have
to figure out the costs of a trade. For
us, it is a mathematical question.
We set aside the stock-selection
alpha, although we believe in it,
and focus on the tax benefit. For
example, if I sell a stock at a loss,
calculating the tax impact is easy.
The tax benefit may then be com-
pared with the incremental risk
and transaction costs imposed by
selling the stock and buying it back
(including the fact that I have to
buy another stock for the portfolio
to remain fully invested). For us, if
the tax benefit exceeds twice the
transaction cost, plus the risk
impact times a penalty function
(representing my dislike for a
higher tracking error), we will do
the trade—assuming that the client
can use the tax benefit.
Whether turnover is 80 percent
or 10 percent does not matter
because we look at the tax benefit
of each individual trade. In a rising
market, however, turnover levels
for a tax-managed portfolio are
low because half of the battle is
deferring gains. For example, in
the late 1990s, our turnover was
about 20–25 percent a year. In the
past few years, our turnover has
been almost 100 percent a year
because the market has been down
and opportunities to harvest losses
have been numerous.
www.aimrpubs.org • 39
Strategic and Tactical Allocation of
Municipal Bonds in Private Client Portfolios
Christine L. Todd, CFA
Senior Vice President
Standish Mellon Asset Management
Boston

In the current market, municipal bonds offer attractive after-tax yields combined with
high credit quality and low price volatility, but investors’ heavy use of munis in short-
term tactical allocations (such as crossover trades) has added volatility to the market in
recent years. Given the probability of yield-curve flattening and a return to a more normal
short-term/long-term yield spread, investors should consider a barbell strategy in which
the portfolio is divided approximately equally between long-term and short-term
maturities. Investors also should consider the implications of the alternative minimum
tax, which will affect an increasing percentage of taxpayers if the tax code is not repealed
or modified.

few years ago, municipal bonds, federally tax-
A exempt bonds issued by states and municipal-
ities in the United States, were underappreciated by
investors and managers alike. In the past few years,
however, their stature has appreciated along with
their market value. Municipal bonds have performed
quite well on an absolute and relative basis, espe-
cially considering their high credit quality, low vola-
tility, excellent after-tax equivalent yield, and
tremendous diversification benefits. In addition to
the strategic importance of municipal bonds for tax-
able investors, this presentation will concentrate on
certain tactical maneuvers that can be used within the
municipal bond (“muni”) market as well as between
munis and other fixed-income markets. Finally, I will
examine the significance of the alternative minimum
tax (AMT) for the tax-exempt bond market.
Advantages of Municipal Bonds
The tax-exempt bond sector is known for its high
credit quality, as shown in Figure 1. Even in today’s
environment, with so many headlines about states
under fiscal pressure because of a weak economy, this
market remains strong, especially when compared
with the early 1990s.
A comparison of default rates for municipal ver-
sus corporate bonds is telling. Table 1 compares the
default rates of municipal and corporate bonds 15
years after issue. The municipal market simply has a
higher standard of quality. The default rate in AA
and AAA municipal bonds is zero; furthermore, the
recovery rate in the event of a skipped payment is 100
percent. The same cannot be said of corporate bonds.
In fact, BBB municipal bonds have a default rate
essentially equal to AAA corporate bonds.
Munis have about half the volatility of U.S. Trea-
sury bonds, as shown in Figure 2. Stock market inves-
tors who see Treasuries as a flight-to-quality vehicle
may want to consider munis instead. Of course,
munis have lower volatility in part because they are
not as liquid as Treasuries. An investor who believes
that interest rates might rise should consider buying
munis rather than Treasuries because munis lag Trea-
suries not only in a rally but also in a sell-off. So,
munis are a more defensive high-quality, fixed-
income vehicle than Treasuries.
The after-tax yield advantage of munis versus
taxable bonds is quite attractive in today’s environ-
ment, as shown in Table 2. (Note that this is calculated
for the 38.6 percent marginal tax bracket. The after-tax
yield for an AMT filer is shown in Table 3.) In only
one case are munis less attractive than other fixed-
income vehicles—the two-year maturity for A rated
corporates. The difference of 17 bps, however, is not
sufficient to compensate a fixed-income investor for
the significantly lower quality in the corporate market
(see Table 1).

40 • www.aimrpubs.org ©2003, AIMR®

 Strategic and Tactical Allocation of Municipal Bonds in Private Client Portfolios

Figure 1. Upgrades vs. Downgrades for Tax-Exempt Bonds, 1991–2002

Number
1,000

900

800

700

600

500

400

300

200

100

0
91 92 93 94 95 96 97 98 99 00 01 02
Upgrades Downgrades

Source: Based on data from Standard & Poor’s.

Table 1. Standard & Poor’s Cumulative Default Munis are cheap not only in today’s market but
Rates for Municipal vs. Corporate also on a historical basis. As of January 2003, at cer-
Bonds, 1986–2000 tain longer maturity points along the yield curve,
Rating Municipal Corporate specifically the 30-year maturity, the yield of munis
AAA 0.00% 0.51%
even exceeds the yield of comparable-maturity Trea-
suries. Panel A of Figure 3 shows the yield on 10-year
AA 0.00 1.07
AAA munis as a percentage of the yield on 10-year
A 0.16 1.83
Treasuries. The higher the ratio, the closer the muni
BBB 0.52 4.48
yield is to the Treasury yield (i.e., the higher the ratio,
BB 1.34 16.36
the cheaper munis are). Compared with the 1993–97
Note: Rolling 15-year periods, 1981–2000. period, munis have become extraordinarily cheap
Source: Based on data from Standard & Poor’s. versus the Treasury market; in fact, farther out the
yield curve, past the 10-year point, munis are even
cheaper than shown in Panel A. The reverse situation
Table 2. Municipal After-Tax Yield Advantage has occurred in the corporate market, however, as
over Taxable Bonds, as of 20 February shown in Panel B of Figure 3. Corporate yields have
2003 increased considerably relative to muni yields, as
(non-AMT taxpayer; bps) corporates have shown themselves to be of lower
Bond Sector 1 Year 2 Year 5 Year 10 Year quality than munis. As a result, munis are less attrac-
Treasury 34 34 78 135 tive (not as cheap) on a pure yield measure in com-
Agency 30 30 60 105 parison with the corporate market than in
Asset-backed 14 13 45 92 comparison with the Treasury market.
Corporate A 21 –17 17 61 Panel A of Figure 3 shows that, as of 31 January
Note: After-tax yield advantage calculated for the 38.6 percent 2003, the yield of 10-year AAA rated munis was 96
marginal tax bracket. percent of the 10-year Treasury yield. This means that
Source: Based on data from Municipal Market Data and J.P. Morgan any investor with a marginal tax rate of 4 percent or
Chase & Company. more should own munis rather than Treasuries

©2003, AIMR® www.aimrpubs.org • 41

Investment Counseling for Private Clients V

Figure 2. Volatility for the S&P 500, 30-Year Municipal Bonds, and 30-Year
Treasury Bonds, 1999–2003
(12-week moving average)
Moving-Average Volatility (%)
50

40

30

20

10

0
99 00 01 02 03
30-Year Treasury S&P 500 30-Year Municipal

Note: Data through 7 February 2003.

Source: Based on data from Bloomberg and Thomson Financial.

Figure 3. 10-Year AAA Municipal Bond Yields as a Percentage of Treasury

and Corporate Bond Yields, 1 December 1993 to 31 January 2003
A. Versus 10-Year Treasury Bonds
Percent

90

75

60
4

2
3

/9

/9

/9

/9

/9

/9

/0

/0

/0
/9

ec

ec

ec

ec

ec

ec

ec

ec

ec
ec

D
D

1/

1/

1/

1/

1/

1/

1/

1/

1/
1/

B. Versus 10-Year AAA Rated Corporate Bonds

Percent

85

70

55
4

2
3

/9

/9

/9

/9

/9

/9

/0

/0

/0
/9

ec

ec

ec

ec

ec

ec

ec

ec

ec
ec

D
D

1/

1/

1/

1/

1/

1/

1/

1/

1/
1/

Note: Merrill Lynch Municipal Bond Index for 7- to 12-year maturities; Merrill Lynch U.S. Corporate
Bond Index for 7- to 10-year maturities.
Source: Based on data from Municipal Market Advisors and Merrill Lynch & Company.

42 • www.aimrpubs.org ©2003, AIMR®

 Strategic and Tactical Allocation of Municipal Bonds in Private Client Portfolios

because the after-tax equivalent yield will be higher
from the muni investment. The historical average
yield relationship over the last 10 years is only 80
percent—the current level is three standard devia-
tions above this average.
Munis are also a great diversifier. Consider the
correlations with the following indexes for the period
from 1 January 1990 to 31 December 2002: Nasdaq
Composite Index, 0.03; Dow Jones Industrial Aver-
age Index, 0.10; Wilshire 5000 Index, 0.11; and S&P
500 Index, 0.14. The correlation of municipal bonds
with equity markets is negligible. And in fact, munis
have exhibited negative correlation with equities in
2003. This low level of correlation is driven partly by
emotional factors and partly by technical factors.
Many retail investors who fear further stock market
sell-offs are selling equity mutual funds and buying
muni mutual funds. Nonetheless, these low correla-
tions are based on more than just the poor equity
market performance of 2000–2002.
Tactical Allocation
From a risk–return standpoint, munis are the best
place to be. Figure 4 shows the after-tax risk and return
(based on a 28 percent marginal federal tax bracket) of
four fixed-income indexes with comparable durations
for the 10 years ending 31 December 2002—Merrill
Lynch Agency 5–7 Year Index, Merrill Lynch Corpo-
rate 5–7 Year Index, Lehman 5-Year Muni Index, and
the Merrill Lynch 5-Year Treasury Index. During this
period, munis provided about as high a return or
higher with far less price risk. Given that a strategic
allocation to munis makes sense, what can managers
do tactically to add value to fixed-income portfolios
for high-net-worth (i.e., taxable) clients?
Crossover Allocation. Crossover allocation
means using all fixed-income sectors to maximize
after-tax returns. At Standish Mellon Asset Manage-
ment, when we see that munis are becoming rich
(although, of course, the opposite has been true
recently), we crossover into the taxable markets.
The starting point for determining the appropri-
ate crossover portfolio weighting is the client’s income
tax situation. The trade-off between tax-exempt and
taxable bonds will vary according to investment
guidelines, market conditions, relative value, and
liquidity.
The excess yield provided by munis versus tax-
able bonds is typically lower in the short end of the
curve, where the excess yield advantage of munis
tends to be much greater. Thus, a tactical crossover
allocation to taxable bonds makes sense in the short
end of the curve.
Managers need to know two things: the histori-
cal after-tax yield of each taxable bond sector as a
reference point for relative value versus munis and
the likely direction of the price relationship between
munis and the other sectors going forward. Even if a
manager decides that corporates are cheap at the
present time and their after-tax yield is attractive, it
still might be dangerous to crossover. For example,
during the past two years, while corporates offered
compelling after-tax yield versus munis, the prices of
corporates kept depreciating versus munis as quality
concerns plagued the sector. In this environment, the
crossover strategy would not have accomplished the

Figure 4. After-Tax Risk and Return for Municipal, Agency, Corporate, and
Treasury Bonds, 10 Years Ending 31 December 2002
Annualized Return (%)
6.5

Merrill Lynch Agency 5−7 Year Index

Merrill Lynch Corporate 5−7 Year Index
Lehman 5−Year Muni Index
5.5
Merrill Lynch Treasury 5-Year Index

4.5
2.5 3.5 4.5 5.5 6.5
Standard Deviation (%)

Note: Taxable returns netted down at top federal tax rate (28 percent).
Source: Based on data from Lehman Brothers and Merrill Lynch & Company.

©2003, AIMR® www.aimrpubs.org • 43

Investment Counseling for Private Clients V

desired outperformance. In other words, it is Table 3. Municipal After-Tax Yield Advantage vs.
extremely important for managers to assess the Taxable Bonds, as of 20 February 2003
future price direction of the relative sectors as well as (AMT taxpayer; bps)
the simple after-tax yield advantage of doing the Bond Sector 1 Year 2 Year 5 Year 10 Year
crossover trade.
Treasury 21 17 47 94
Figure 5 shows the ratio of the muni yield to the
Agency 16 12 26 59
pretax yields of three taxable bond sectors for the 10-
Asset-backed –2 –7 9 43
year period ending December 2002. This crossover
Corporate A 6 –42 –24 8
historical yield comparison shows munis becoming
Note: After-tax yield advantage calculated at 28 percent tax rate.
cheaper relative to Treasuries and agency debt. The
opposite has been true for corporate bonds since the Source: Based on data from Municipal Market Data and J.P. Morgan
Chase & Company.
credit crisis began in late 1998. In December 1999,
corporate bond yields were below the AMT
breakeven point at 72 percent (i.e., a marginal tax rate
it is possible to capture a bit more after-tax yield by
of 28 percent; the non-AMT breakeven point for a 39.6 crossing over into the corporate market, but this pat-
percent federal taxpayer is 61.4 percent). Although tern does not hold as one goes further out the curve.
this might have appeared to be a good opportunity It may appear that two-year A rated corporates—
to add after-tax yield by selling munis and buying with a 42 bp after-tax yield advantage—are attractive
corporates, as I explained earlier, it would have been versus munis, but given the greater volatility of the
a bad trade because corporates cheapened even fur- corporate market, this yield advantage alone is not
ther. Late in 2002, however, if a taxable investor made enough to compensate investors for lower quality
a crossover trade into corporates, it would have been and possibly lower liquidity.
a good trade because the investor would have cap-
tured both the favorable after-tax yield and the prin- The Yield-Curve Trade. Although a crossover
cipal appreciation from the relatively greater price trade might not be best pursued in the current envi-
appreciation of the corporate sector from that point ronment, the yield-curve trade absolutely will be the
forward. As with every other market, timing and trade of the coming year, second half 2003 to first half
relative value cannot be ignored in the muni market. 2004, because the yield curve is so steep. At 346 bps,
Consider the after-tax yield advantage of munis the slope of the curve (measured as the difference
versus taxable bonds for an AMT filer, as shown in between the yields of the 30-year and 2-year maturi-
Table 3. As in Table 2, at the short end of the curve, ties) is more than double its 8-year average of 175

Figure 5. Crossover Historical Yield Comparison for 10-Year Maturities,

December 1993–December 2002
Muni/Taxable Yield (%)
100
95
90
85
80
75
70 AMT
65
60 Non-AMT
55
50
12/93 12/94 12/95 12/96 12/97 12/98 12/99 12/00 12/01 12/02
Treasury Agency Corporate

Note: AMT/Non-AMT = after-tax breakeven point for individual investor.

Source: Based on data from Merrill Lynch & Company and Bloomberg.

44 • www.aimrpubs.org ©2003, AIMR®

 Strategic and Tactical Allocation of Municipal Bonds in Private Client Portfolios

bps, as shown in Figure 6, and also two-and-a-half
standard deviations above that average. The 2-year-
to-10-year slope is also remarkably steep at 245 bps
relative to, for example, the flat slope of 32 bps on 31
December 2000 just prior to a major rally in the short
end of the yield curve and to the average of 100 bps.
At the front end of the curve, a lot of incremental
yield can be gained without adding much duration by
taking advantage of the steepness in intermediate
maturity muni yields. For example, in Figure 7, the
bars represent the incremental yield pickup of moving
from one maturity to the next in AAA rated munis.
The greatest incremental yield pickup for extending
maturity is in maturities of less than 10 years. The yield
curves shown in Figure 7—the dotted line at the begin-
ning of 2003 and the solid line at mid-February—show

Figure 6. Slope of Municipal Bond Yield Curve for 2- to 30-Year Maturities,

1995–2002
Yield (bps)
400

350

300

250

200
Average

150

100
95 96 97 98 99 00 01 02

Source: Based on data from Thomson Financial.

Figure 7. Municipal AAA Yield-Curve Comparison, 31 December 2002 vs.

18 February 2003
Yield Pickup (bps) Yield (%)
5.0
60

50 18/Feb/03
4.0
42 43
38 37
40 31/Dec/02
30 3.0
30 25
23
20 17
12 12 2.0
10 10 9
10 8 7
0 0
0 1.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 20 30
Maturity (years)

Source: Based on data from Thomson Financial.

©2003, AIMR® www.aimrpubs.org • 45

Investment Counseling for Private Clients V
that yields have remained unchanged at both the long
and short ends of the yield curve but that the market
sold off in the belly of the curve.
Hedge funds often put on the crossover trade
from the taxable bond sectors to the muni sector to
exploit the relative value of markets. Believe it or not,
the muni market has become extremely interesting to
hedge funds, as well as to the broader investing
public, because of its cheapness relative to the other
liquid fixed-income sectors. The muni sector has seen
a lot more issuance than other sectors, the deal size is
much bigger, and credit quality has remained high.
These factors have supported the liquidity of the
muni sector. Thus, the potential exists for a bit more
profitability in arbitrage situations.
When munis were trading close to 100 percent of
Treasury yields, hedge funds bought tax exempts
aggressively, focusing on the intermediate part of the
curve because of the lack of callability in those matu-
rities and the number of deals of large size being
issued with intermediate maturities. Once demand
drove muni yields down to, say, 95 percent of inter-
mediate Treasury yields, hedge funds dumped their
munis, leading to a sell-off in the market. This cross-
over trade was later repeated after the sell-off suc-
cessfully returned munis to a position of relative
value versus the taxable bond sectors. It is likely that
the muni sector will become more volatile (but more
efficient) as it becomes increasingly driven by large
arbitrage accounts. The traditional muni investor
needs to be aware of this potential toward greater
volatility to avoid getting caught on the wrong end
of the trade. The slope of the yield curve is important,
but how it can be manipulated by crossover investors
is becoming even more important.
Panel A of Figure 8 shows the annualized total
return for municipal bonds at different points on the
yield curve for the 1990–2000 period. Not surpris-
ingly, total return increases as maturity is length-
ened. Since 31 December 2000, an anomaly has
occurred, however, with short-term bonds outper-
forming long-term bonds, as shown in Panel B.
Reversion to the mean would suggest that long-term
muni bonds now hold greater relative total return
potential—in a stable interest rate environment or in
the long run—than muni bonds at other parts of the
curve because they have underperformed short-term
bonds versus historical experience.
Looking ahead, should the fear of rising interest
rates encourage the selling of long-term bonds and the
reinvesting of those proceeds in short-term bonds?
Two arguments contradict such a view. The first is
reversion to the mean, as I have just described. The
second is that the muni yield curve is so steep that the
opportunity cost of owning short-term bonds can be
46 • www.aimrpubs.org
Figure 8. Annualized Historical Yield-Curve
Return for Municipal Bonds
A. 1 January 1990 to 31 December 2000
Return (%)
8 7.91 8.00
7.43
7
6.36
6
5
5 10 20 Long
Maturity (years)
B. Since 31 December 2000
Return (%)
8 7.50
6.78 7.01 6.99
7
6
5
5 10 20 Long
Maturity (years)
Note: Panel B data through 31 January 2003.
Source: Based on data from Lehman Brothers.
punishing. The recent rally in interest rates has had
little effect on the long-term end of the yield curve. As
of the end of the first quarter of 2003, to forgo a 30-year
maturity for a 2-year maturity would mean a give-up
of 340 bps in yield, and to forgo a 10-year maturity for
a 2-year maturity would mean a give-up of about 225
bps. This give-up in yield represents a great deal of
money for a sizable fixed-income portfolio, especially
given today’s historically low yields.
I keep focusing on 31 December 2000 because it
predates the significant yield-curve shift shown in
Figure 9. This dramatic change in the shape of the
yield curve between then and now is called a “snap-
down” in short-term yields, which pivoted off the
intermediate part of the curve. This event was driven
by the U.S. Federal Reserve Board’s surprisingly
aggressive easing in January 2001, a policy it has so
far maintained. The result was a significant steepen-
ing of the yield curve because the yields at the long
end of the curve remained virtually unchanged. The
curve will have to flatten at some point and return to
a more normal shape, and so this snap-down will be
reversed by a “snap-up” should again pivot off the
intermediate part of the curve. This means that short-
term yields will rise relatively more than long-term
yields. Current real yields in the municipal sector are
favorable to those in the Treasury market, which also
©2003, AIMR®
 Strategic and Tactical Allocation of Municipal Bonds in Private Client Portfolios

Figure 9. Decline of Short-Term Municipal Bond Yield, 31 December 2000 vs.

31 December 2002
Yield (%)

5.00
4.50 31/Dec/00
4.00
3.50
3.00 31/Dec/02
2.50
2.00
1.50
1.00
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 20 30
Maturity (years)

Source: Based on data from Municipal Market Data.

minimizes the risk of remaining invested at the long
end of the muni market.
The challenge is in structuring a municipal bond
portfolio to take advantage of the anticipated flatten-
ing of the yield curve. Figure 10 shows two extremes:
a bullet approach and a barbell approach. The bullet
is a concentrated exposure at the intermediate part of
the curve, which would capture more yield than the
barbell approach but would fail to take advantage of
the expected flattening of the yield curve. The barbell
provides broader exposure along the yield curve,
with a concentration in the maturity buckets that will
offer the best roll if the curve flattens as expected.
Consider two market scenarios. In the first sce-
nario, the yield curve flattens, returning to its average
175 bp slope between 2-year and 30-year maturities.
In the second scenario, the yield curve exhibits only
modest flattening, returning to a slope of 205 bps. In
the first scenario, the barbell strategy will outperform
the bullet by slightly more than 100 bps. In the second
scenario, the barbell’s excess return over the bullet is
90 bps. The reason the barbell outperforms in both
scenarios is its exposure to the long-term end of the
curve, whereas the bullet is concentrated in the inter-
mediate range of the curve, the pivot point for yields.
The AMT
The AMT poses a problem for taxable investors in the
United States. Table 4 shows that by 2010, if tax law
remains unchanged, the proportion of taxpayers sub-
ject to the AMT will grow significantly. Political pres-
sure, however, may lead to a change in the law before
such a scenario develops. The AMT is not indexed for

Figure 10. Duration Bucket Exposure: Barbell vs. Bullet Approach

Portfolio Allocation (%)
80

60

40

20

0
0−2 2−3 3−4 4−5 5−6 6−7 7−8 8−9 9−10 10+
Duration (years)
Barbell Bullet

©2003, AIMR® www.aimrpubs.org • 47

Investment Counseling for Private Clients V
Table 4. Estimated Increase in Filers Subject to
AMT
(percent)
Adjusted Gross Income per Year
Year $75–$100K $100–$200K $200–$500K
2002 3 11
2010 79 94
Source: Based on data from Lehman Brothers.
inflation, so each year an increasing number of tax-
payers becomes subject to the tax. This is a challenge
for private wealth management because both manag-
ers and clients must be prepared if the law does not
change and, at the same time, be prepared if the law
does indeed change.
Issuance of bonds subject to the AMT accounts
for about 10 percent of total issuance in the muni
market, as shown in Figure 11. So, it is quite easy to
find muni bonds that are not subject to the AMT for
those investors who currently are subject to the AMT.
Unfortunately, some of the higher-yielding sectors,
such as industrial development bonds, housing
bonds, high-quality intermediate-term structures,
and student loan bonds, are also subject to the AMT,
so avoiding them may mean sacrificing yield.
The steady supply of bonds subject to the AMT
combined with decreasing demand for them (as more
investors anticipate becoming liable for the AMT) has
caused yield spreads on bonds subject to the AMT to
widen versus yields on muni bonds that are not
subject to the AMT (such as insured bonds), as shown
in Figure 12. Even though the spread between yields
on AMT-exempt and AMT-subject bonds is wider
Figure 11. Issuance of Bonds Subject to AMT vs. Total Municipal Bond
Issuance, 1989–2002
Issuance ($ billions)
500
400
300
200
100
0
89 90 91 92
Source: Based on data from Thomson Financial.
48 • www.aimrpubs.org
now than it has been in recent years, the spread will
continue to widen as more investors avoid buying
AMT-subject bonds as they anticipate AMT liability.
Figure 13 illustrates that the pickup in yield that
AMT-subject bonds provide is not sufficient to offset
paying the AMT on the coupon and that, net–net, the
36
investor is better off in all maturities to buy AMT-
97
exempt AAA munis or Treasuries.
The challenge for a portfolio manager working
with an AMT-subject client is to mitigate the impact
of the AMT. The most important way to do this is to
avoid buying municipal bonds that are subject to the
AMT. Unfortunately, many managers do not have
good control over the AMT status of bonds that are
put into their clients’ portfolios and, in many
instances, learn only at the time of tax filing that their
clients owe the AMT on some of their municipal
bonds. A wide variety of tax management strategies
that are beyond the scope of this presentation can also
be used to mitigate the effects of the AMT. For exam-
ple, because the exercise of stock options is consid-
ered regular income, which is subject to the AMT,
AMT-subject taxpayers need to carefully time their
exercise of stock options. Two deductions that are
still allowable in the calculation of the AMT are char-
itable giving and home mortgage interest. Maximiz-
ing these deductions is a tax-saving measure that can
mitigate the impact of the AMT.
Conclusion
In the current market, municipal bonds typically offer
higher after-tax yields than other fixed-income sec-
tors. Furthermore, these attractive yields are bundled
with high credit quality (lower levels of default than
corporate bonds at every credit rating) and low price
93 94 95 96 97 98 99 00 01 02
Total AMT
©2003, AIMR®
 Strategic and Tactical Allocation of Municipal Bonds in Private Client Portfolios

Figure 12. 10-Year Yield Spread for Insured client’s portfolio is approximately equally divided
Municipal Bonds vs. AMT-Subject between long-term and short-term maturities.
Municipal Bonds, July 1994–July 2002 And the AMT lurks on the horizon as a thorn in
the side of taxable investors. Will it or will it not be
Yield Spread (bps)
repealed or adjusted? In the meantime, steady issu-
33 ance of AMT-subject bonds and a lessening of demand
28 for them are causing spreads on AMT-subject bonds
to widen but not enough to compensate for the AMT
23
if owed. All told, the muni market offers many poten-
18 tial rewards for taxable investors, as well as a few
13
wooden nickels.

8
94

95

96

97

98

99

00

01

02
Figure 13. Yield–Tax Trade-Off for AMT-Subject
7/

7/

7/

7/

7/

7/

7/

7/

7/
and AMT-Exempt Bonds
Note: Insured muni curve is virtually AMT free.
Source: Based on data from Bloomberg. Yield (%)
5

4
volatility. Their exceptional characteristics make them
an attractive fixed-income vehicle for investors in 3
almost any tax bracket.
2
Munis can also be used in short-term tactical
allocations. In a crossover trade, taxable investors can 1
move from munis to Treasuries to corporates and
back to capture superior after-tax yields in anticipa- 0
2 4 10 20 30
tion of changes in the relative yields of each sector.
Maturity (years)
This type of activity has added volatility to the muni
market in recent years. In anticipation of a flattening AAA AMT-Exempt Muni Yield
of the yield curve and a return to a more normal After-Tax Treasury Yield
short-term/long-term yield spread, the trade that After-Tax AAA AMT-Subject Muni Yield
would best position investors to take advantage of
such a flattening is a barbell strategy in which a Source: Based on data from Bloomberg and Thomson Financial.

©2003, AIMR® www.aimrpubs.org • 49

Investment Counseling for Private Clients V
Question and Answer Session
Christine L. Todd, CFA
Question: Would you combine
mutual funds with individual
bond selection in managing a muni
portfolio for a private client?
Todd: I would avoid buying
mutual funds. With the decline in
interest rates, they contain a lot of
accumulated capital gains and
they do not allow investors the
ability to control their tax situation.
I would buy separate holdings of
munis, even for a small portfolio.
Question: For active manage-
ment, what is the minimum num-
ber of issues you would buy in
order to achieve adequate diversi-
fication, and what would be the
shortest and longest maturities?
Todd: We strongly believe that a
$2.5 million to $3.0 million portfo-
lio with 30 or 40 issues can be
actively managed—although not
full-blown active management on
the order of 30–40 percent annual
turnover because position size
would be too small. A $5 million
portfolio is more suitable for full-
blown active management because
it allows 3 percent positions that
are large enough to be liquid and
the ability to do relative-value
trading—concentrated sector over-
weights, state overweights, and
yield-curve overweights. The ideal
is a portfolio large enough for
positions of 250,000 par value, but
positions of 100,000 par value
should still provide reasonably
good liquidity.
With a $1 million portfolio,
you should buy 10 high-quality,
noncallable bonds with maturities
from two to seven years. You
should search for anomalies that
enable you to gain an extra 10 bps
in yield because of excess supply in
the issuance of a particular state or
through odd-lot trading at a
regional brokerage firm. Buy
bonds with those anomalies and
hold them. Collect your coupons
50 • www.aimrpubs.org
until they mature and reinvest
with a similar strategy. Active
trading of a $1 million portfolio
would cause transaction costs to
erode returns, so we don’t recom-
mend active management for such
a small portfolio.
Question: What is the current
yield-curve slope telling us about
the economy?
Todd: The current steep slope is
saying that the Fed is expected to
keep short rates extremely low. We
don’t believe that another easing is
likely, but the Fed could further
lower short-term interest rates to
support the economy if unfortu-
nate geopolitical events were to
occur. Of course, the Fed has other
means to boost the economy with-
out lowering short-term rates.
Such a disparity between long-
term and short-term rates can
mean that the market is anticipat-
ing a heating up of inflation in the
future, and in certain sectors, infla-
tion has shown signs of bubbling
up. Not long ago, however, when
the curve was just as steep as it is
today, people were talking about a
deflationary scenario. So, I don’t
believe that the current steepness is
signaling inflation but rather
investor concerns about the econ-
omy and weak consumer confi-
dence. A flight to quality has
occurred and no more so than at
the short end of the yield curve
because that is where volatility is
the lowest.
Question: Why is monthly cor-
relation with other asset classes
important for a long-term investor?
Todd: The lack of correlation
between munis and other financial
asset classes means that the total
portfolio will perform better on a
risk-adjusted basis. For example,
when equities are selling off, your
allocation to munis will stabilize
the portfolio, providing a better
total portfolio return.
Question: Given the deteriorat-
ing budgetary outlook for state gov-
ernments, what is your outlook for
the supply of both high- and low-
quality munis over the next year?
Todd: We’re seeing a lot of
headlines about bonds being
issued to bail out certain states and
thus more issuance than was antic-
ipated. The wider spreads have
been driven by an emotional reac-
tion to the glut of supply in the
marketplace. Thus, we are not
afraid of these credits whose
yields have widened out. At
Standish Mellon, we see this situ-
ation as a tremendous opportunity
for relative-value trading. These
state governments are well run
and, in many cases, have zero-
deficit requirements. They also
have a great record with not a sin-
gle default, so we see a strong
recovery rate.
Question: What is your opinion
on the relative quality of general
obligation bonds and revenue
bonds?
Todd: The market has tradition-
ally considered general obligation
bonds to be of higher credit quality
than revenue bonds because of the
taxing power of the issuer. At
Standish Mellon, we don’t take
that view. We actually believe that
revenue bonds are a better source
of quality because they’re easier to
analyze. In the past 18 months, the
market has seen that event risk
applies to general obligation bonds
because legislation to increase
taxes or freeze tax decreases is hard
to pass. Many states now regret tax
reductions they made late in the
economic boom. Fortunately, some
states had sizable rainy-day funds.
States that are experiencing greater
deficits or greater revenue reduc-
tion will have to rely more heavily
©2003, AIMR®
 Strategic and Tactical Allocation of Municipal Bonds in Private Client Portfolios
on new issuance, which will cause
their spreads to widen relative to
other states. Investors have already
seen a dramatic example of this
effect in New York and California.
Question: What is your opinion
of the credit quality of munis issued
by New York and California?
Todd: Our belief is that the Cali-
fornia economy is now stronger
than it was in the early 1990s. The
budget situation is a bit more
severe, but fixes are in place, and
not just one-time fixes, but fixes
©2003, AIMR®
that will correct matters for the
long term. The yield spread
between California and general
market AAA munis is 20 bps wider
now than it was when Orange
County filed for bankruptcy. The
wide spreads indicate that the
combination of the heavy issuance
to bail out the state and the recent
utility crisis is driving spreads
beyond what credit factors would
justify. After all, California is an A
rated credit and the sixth largest
economy in the world.
For states with heavy issuance,
that issuance is causing spreads to
widen between the yields of those
states and the yields of other com-
parable quality states. For example,
after September 11, New York City
took all of its issuance off the table.
Now, the pent-up supply is burst-
ing through, bringing a lot of issu-
ance from both New York state and
New York City. So, their bonds are
trading 1 percent cheaper than oth-
ers of similar maturity and quality.
www.aimrpubs.org • 51
The Psychology of Wealth
Marty Carter
Family Communications Advisor
Charles D. Haines, LLC
Birmingham, Alabama
Investment advisors tend to focus on the “hard” side of managing private client wealth,
but they should give equal attention the soft side (i.e., the psychological factors).
Understanding the psychology of wealth can help advisors build more effective
relationships with their clients. Advisors who do not understand their clients’ needs and
objectives are unlikely to serve them effectively.
dvisors who try to understand the psychology
A of wealth, rather than ignore it, will have more
productive relationships with their clients. Although
a “soft” issue, the psychology of wealth is a major
factor in family dynamics, especially money conflicts,
that can have important implications for those
charged with managing a family’s wealth. I will
review the most common challenges facing wealthy
clients and the psychological roadblocks that inhibit
their ability to manage their assets. I will also discuss
how, in certain circumstances, a facilitator can help to
break barriers and build more effective client rela-
tionships. At my firm, Charles D. Haines, LLC, for
example, we have adopted a multidisciplinary, col-
laborative approach that compels clients to establish
meaningful goals and devise appropriate solutions.
Common Wealth Issues
A wealthy individual faces many challenges peculiar
to his or her status, whether he or she is ultrawealthy,
a “millionaire next door,” or an aspiring millionaire.1
Typically, the wealthy talk about money a great deal
but rarely talk about the purpose of money or how it
might affect their relationships. Some issues affect
certain wealth levels more than others, but recogni-
tion of certain factors can help advisors gain insight
into what is going on in their clients’ lives.
1
For a discussion of the “millionaire next door” type, see Thomas
J. Stanley and William D. Danko, The Millionaire Next Door: The
Surprising Secrets of America’s Wealthy (Atlanta, GA: Longstreet
Press, 1998).
52 • www.aimrpubs.org
Culture. One of the most significant issues is the
culture of wealth. In many families, the descendents
of those who made the money tend to assume that
the well will never run dry. As a result, parents, who
perhaps themselves never acquired sound money
practices, do not instill appropriate attitudes and
behaviors in their children, which may engender a
sense of entitlement. For example, the children might
think their parents are spending their inheritance,
even though the money is not yet theirs to spend.
Seeing the family name over the door of a new art
center can create at best a selfless spirit of giving back
to the community and at worst a sense of entitlement.
Those who feel entitled to their wealth tend to be
cheap and expect discounts and immediate service.
In my experience, the wealthiest people are often the
most reluctant to pay for what they want.
In addition, wealthy people are often placed on
pedestals and receive deferential treatment in their
communities. They are viewed as possible donors or
potential clients, and such community adulation,
although welcomed by some individuals, causes
many to retreat.
Problem Values. For ultrawealthy families,
inculcating in their children a sense of responsibility
about money is a significant challenge. In short, par-
ents who do not set boundaries and clear expectations
tend to rear children who do not have a strong sense
of responsibility or self-worth. If the message is that
the money will last forever and no one in the family
needs to worry about money, then the children have
little or no incentive to do well in school and prepare
for a job.
©2003, AIMR®

Self-Esteem. Wealthy individuals are often
uncertain whether they are liked for who they are or
for how much money others think they have. A lack
of self-esteem is thus often exacerbated by wealth.
Some individuals feel undeserving or self-conscious
of their inherited wealth or isolated because of the
deferential attention they receive. A common com-
plaint among young wealthy individuals in particu-
lar is that they are often expected to pick up the tab.
A 10-year-old child recently told me about an inci-
dent that will likely affect him for years to come. A
classmate borrowed his tennis racket and returned it
damaged. When he asked the classmate to fix it or
buy a new one, the response was, “Buy yourself a
new tennis racket. You’re rich.”
Another stumbling block to a healthy sense of
self-esteem occurs in families in which success is
measured by money. Little attention is given to other
accomplishments, such as intellectual ability, the
capacity to help others, or commitment to family and
friends.
Money Beliefs. Everyone has what I call
“money beliefs,” or attitudes and behaviors about
money that stem from childhood. Helping clients to
identify the messages they have received about
money from their families provides insight into their
current attitudes about money. These beliefs also
play a significant role in determining how they will
relate to an advisor.
I have heard every cliché imaginable about
money from clients when I ask them about their
beliefs: for example, “money doesn’t grow on trees,”
“money is the root of all evil,” “marry money, don’t
bury it,” or “marry for love or money, but mostly
money,” all of which are messages that have been
forever ingrained in clients’ psyches. One of the most
common answers, however, is that their families sim-
ply do not talk about money, which is a message that
can be particularly detrimental to effective invest-
ment management.
Money Personality Polarities. In order to
establish a rapport and build a successful working
relationship with clients, advisors may find it useful
to identify a client’s “money personality.” For exam-
ple, people tend to be either generous or miserly,
straightforward or secretive, cautious or impulsive,
intrusive or respectful. People also tend to be more
or less dictatorial, more or less financially informed,
and more or less risk averse.
Lifestyle. Surprisingly, many wealthy individ-
uals live beyond their means. One of my clients will
be insolvent by the third quarter of 2003 if she does
not drastically cut her spending. For years, her advi-
©2003, AIMR®
The Psychology of Wealth
sors have been warning her about the possibility of
running out of money, but she has refused to heed
their advice. She recently purchased a villa in Rome
and continues to make unsustainable philanthropic
commitments. She can choose to sell some of her
assets or have a family member bail her out, as she
has done before, but she seems to be incapable of
altering her extravagant lifestyle.
Many people are ashamed of their wealth, how-
ever, particularly if it was inherited, so they hide how
much money they have. Others flaunt their wealth for
the sake of appearances, which often leads to partic-
ularly lousy financial decisions. In New Orleans, for
example, families will take out a second mortgage to
pay for a daughter’s debutante ball. The cost of a
wedding pales in comparison to that of a debutante
ball, but appearances mean a lot to some people.
Relationships. When spouses are from differ-
ent socioeconomic backgrounds, many problems can
arise. The person who grew up without much money
may continue to be unnecessarily frugal or, at the
opposite extreme, become a spendthrift. One daugh-
ter of a wealthy family married a man who had no
income. Worried about the balance of power in the
relationship, the family gifted the groom $1 million.
Unfortunately, he developed a sense of entitlement
to the money. He spent nine months sailing and is
now building an airplane and merely engaging in the
proverbial eating of bonbons on the couch. Needless
to say, parents of newlyweds are often concerned that
the spouse of the inheritor will take advantage of the
newfound access to money.
One of the factors that most negatively affects
relationships is the reluctance to talk constructively
about money. I was recently asked to help complete
a prenuptial agreement—on a Tuesday before a Sat-
urday wedding—because the couple had reached an
impasse. The problem was that, even though both
the bride and groom were from extremely wealthy
families and knew they needed an agreement, nei-
ther one had been asked by his or her attorney to
specify what was necessary and doable. Once I was
able to get them talking, we constructed a satisfac-
tory agreement within 45 minutes and sent it to their
respective attorneys.
Relationships are inevitably affected by power
and control issues associated with money. I know a
woman who was worried about her grandson’s poor
academic performance. Without checking with the
child’s parents, she hired a tutor for him, an act that
created resentment in the parents. Jealousy between
family members who have varying levels of wealth
can also have a significant impact on relationships.
Furthermore, those who are wealthy sometimes feel
put upon by family members who are less well-off.
www.aimrpubs.org • 53
Investment Counseling for Private Clients V
And wealthy people are frequently puzzled by rela-
tives who refuse to ask for financial help, even for a
worthy cause.
Financial Literacy. Contrary to popular belief,
wealthy individuals are not always financially literate
nor do they always recognize the need to have a long-
term strategy. Many people hire financial advisors but
take little or no responsibility for managing their own
money. Some of my clients have never possessed a
checkbook because they rely on the family office to
pay their bills. As James Hughes stresses, the danger
of such an attitude lies in going from shirtsleeves to
shirtsleeves in three generations: The first generation
toils, the second generation is frugal (although it has
access to money the first generation did not have), and
the third generation depletes the rest of the fortune.
By the fourth generation, the family is back to nothing
but shirtsleeves. Therefore, to preserve wealth, fami-
lies should consider long-term planning, even for a
time frame of 100 years or longer.2
My campaign to help families become financially
literate and talk about money stems partly from per-
sonal experience. My daughter was accepted at two
colleges, an A-level school and a B-level school. Only
later did we learn that she chose the B-level school
primarily because of a scholarship the school offered.
I regret that I somehow had given her the message
that we could not afford a top-notch school. To avoid
such an issue, parents should begin discussing finan-
cial issues when children are young. For example,
children should be encouraged to do their own due
diligence in regard to paying for their education.
Psychological Roadblocks
Once advisors understand these common wealth-
related issues, they can help clients move past the
psychological roadblocks that inhibit the ability to
manage money effectively. Advisors need to ask
probing questions in order to delve behind the num-
bers and better understand what is going on in their
clients’ lives. Such a strategy helps to foster long-term
relationships that will steer clients toward thinking
about the “why” of investing rather than merely the
“how” of investing.
Attitudes and Behaviors. Many factors gov-
ern client attitudes and behaviors in relation to
money, but anxiety about money is an overriding
concern, no matter how wealthy someone is. Wealthy
parents fear their children will be lazy, or if they think
they have too little money, they worry about running
2 James E. Hughes, Family Wealth: Keeping It in the Family (Hughes
and Whitaker, 1997).
54 • www.aimrpubs.org
out of money. Advisors can make a significant con-
tribution by allaying clients’ fears and letting them
know how long they can expect their money to last
given all the variables, such as current spending
plans and stated goals.
Advisors can also help clients overcome their
fear of conflict, which is one of the most significant
factors that prevents people from talking about
money. Some people simply refuse to share financial
information with the next generation. Instead, their
strategy is to let everyone fight about the money after
they are gone. But the fear of conflict often leads to
conflict. Consider the case of Sarah, who hid from her
husband that she earned more money than he did.
Her assumption that he would be uncomfortable
with this fact led to her assumption that he did not
need to know anything about her spending habits.
When her husband had to sign their joint income tax
return, however, he discovered her secret and was
not happy about being duped. In time, I found out
that Sarah’s father used to harangue her mother
about expenses. In order to avoid conflict, she and her
mother grew accustomed to hiding their purchases.
Clearly, Sarah was repeating the same pattern with
her husband and needed to alter her behavior.
Client Capacity. Another roadblock is client
capacity. The habit of buying something as a kind of
emotional panacea is a major hurdle for many clients
to overcome. Others struggle with disinterest; they
have cultivated the idea that they do not need to
worry about money. The desire to use money as a
weapon, to get even, is another limitation. A couple
in the midst of household renovation argued so much
about the amount of money to spend that the hus-
band refused to spend another cent. Furious, the wife
got even by “forgetting” to write their quarterly esti-
mated income tax payments to the U.S. government.
Keep in mind as well the psychological impact
of the recent bear market, particularly for clients with
liquidity concerns or for those who are financially
overextended. Clients are no longer as interested in
philanthropy and gift giving as they were in the
1990s. Anxious about the economy, clients may
retreat from interaction with their investment man-
ager because they do not want to hear bad news, or
they may decide to micromanage their investments
or second-guess their advisor’s recommendations.
Furthermore, a client’s low risk tolerance may reflect
not only values learned at home but also concern over
significant downturn in the value of a portfolio.
Age is also an important factor in a client’s will-
ingness to listen to advice. Young adults are generally
not interested in planning for retirement or setting
long-term goals. Nonetheless, advisors can help by
©2003, AIMR®

encouraging parents to establish financial plans for
their adult children.
Advisor Attitudes. Another roadblock, which
is often unrecognized, is the effect that a client’s
wealth may have on an advisor. Some advisors are
envious of their clients’ wealth. I admit that hearing
wealthy clients complain about their cruise ships
being too crowded grates on my nerves. Nonetheless,
advisors need to be aware of their own attitudes and
prejudices in this regard and keep their personal
judgments to themselves.
Solutions
My discussion of the “soft” side of money will make
the most sense to so-called “right brainers.” People
who use the left side of their brain analyze information
by breaking it down into units and examining causes
and effects (e.g., financial analysts), whereas right
brainers look at the big picture by synthesizing infor-
mation and looking for patterns, relationships, and
connections (e.g., psychologists). This preference for
using one side of the brain more than the other directly
affects the way a person makes decisions; challenging
this natural preference can be disconcerting.
Because of this dichotomy in approach, advisors
should consider hiring a facilitator to help advise
certain clients on the soft issues. Clients who are
analytical, logical, and think in terms of the bottom
line are likely to be easy to deal with, but not all clients
fit that mold. If an advisor needs to delve into soft
issues with a client who is a left brainer, for instance,
the client may become uncomfortable and accuse the
advisor of being “touchy feely,” thus putting the
relationship at risk.
Having a facilitator involved in the communica-
tions process can be helpful in managing the client–
advisor relationship. Advisors should never open a
wound they cannot close. If, for example, a couple
fights in the advisor’s presence or a client is secretive
or appears to be troubled or hurt, seek professional
help in facilitating. And do not worry that a facilitator
will try to take away clients. When I first started
working with advisors outside my firm, many finan-
cial professionals were afraid I would steal their cli-
ents. Only gradually did they come to trust that I was
interested only in making their client relationships
run more smoothly and effectively.
Breaking Barriers. Advisors can help clients
move beyond their psychological roadblocks by lis-
tening and paying attention to the client’s stories
about money. In addition, advisors should seek cli-
ents’ feedback, a lesson I learned from the following
example. A couple hired my firm to update their
©2003, AIMR®
The Psychology of Wealth
financial plan because they were selling part of their
business. The husband asked for a plan that
addressed the worst-case scenario. The wife was not
happy with the resulting plan. Although she had
never voiced her opinion in my presence, I eventually
learned that she thought the plan meant that, even
though she now had gazillions of dollars, she was
going to have to shop at Kmart. Clearly, advisors
need to specifically ask whether clients are comfort-
able with their recommendations. The worst out-
come possible—to have no action taken after a plan
is devised—is bound to occur if client feedback is
insufficient.
Both the head (business factors) and the heart
(emotional factors) should be acknowledged as hav-
ing equal weight in financial decisions. Advisors are
trained to encourage clients to make sound business
decisions, but they need to recognize that, occasion-
ally, emotional factors interfere. For example, some
clients ignore the advice not to accelerate their mort-
gage payments and insist on paying off their mort-
gage. They are more interested in the psychological
satisfaction of ownership than the benefit of low
interest rates.
Advisors need to take the time to find out about
the life experiences that influence their clients’ choice
of investments, their priorities, and how much they
save or spend. For example, one wealthy couple
could not meet their financial goals, and only through
a story the husband told was I able to discover why.
When I asked the husband what word came to mind
when I said the word “money,” he said, without
hesitating, “food.” He was the oldest of five brothers.
His father had died when he was 11, and he always
worried whether the family would have enough food
on the table. He did not think this association influ-
enced his relationship to money, but his wife dis-
agreed, suggesting that his reaction was to give his
daughters everything they wanted and more.
Although he had never made the connection before,
he sheepishly agreed that his experience had led him
to indulge his daughters to the point of hampering
his own lifestyle.
Although left brainers may cringe, advisors need
to encourage clients to ask themselves the following
questions to ascertain their relationship to money:
What is meaningful in my life? What are the five most
important things in my life? What do I want my
money to do? How can I help my children and grand-
children learn to be responsible with money?
Sample Model. The multidisciplinary financial
planning model Charles D. Haines, LLC, has adopted
is not based on rocket science, but it does require an
investment of time and money. Our clients are enthu-
siastic about our collaborative approach and benefit
www.aimrpubs.org • 55
Investment Counseling for Private Clients V
from the extensive relationships with many team
members (investment advisors, financial planners, a
philanthropy advisor, a communications advisor to
businesses and foundations, and a psychologist). We
also work closely with clients’ own advisors, invest-
ment managers, estate attorneys, accountants, and
insurance representatives.
Before drawing up a financial plan, we require
that clients take a test to determine their goals and
priorities. The first draft of the plan is a reality check
to determine whether the plan goals are doable. My
favorite part of the process is the two-hour meeting I
conduct with each client. I spend a small amount of
time ensuring that the plan is doable and that we are,
in fact, meeting client expectations. But the majority
of time is spent learning the client’s story and deter-
mining how it affects his or her relationship to
money. I want to know how financial decisions are
made in the family and whether one person bears
most of the fiscal responsibility. And given that all
families fight about money, I want to gain insight into
how the person’s family fights about money.
Throughout the meeting, I try to resolve any partic-
ular psychological roadblocks I encounter. Thus,
while others create the financial plan and put
together all the financial pieces, I help individuals
understand their relationship to money so they can
adopt prudent attitudes and achieve their goals.
After discussing with key family members their
concerns about money and their goals, I facilitate a
series of family meetings focusing on family history,
attitudes and behaviors about money, and financial
education for both children and adults. I encourage
all family members to participate in these meetings.
At the first meeting, I draw a family tree to encourage
storytelling. A grandfather in a recent meeting
shared a family secret that his Uncle Harry was a
horse thief. His revelation led to a lively discussion
about the values passed down by that ancestor. By
taking notes and composing the stories on large
sheets of paper around the room, various patterns
begin to emerge (e.g., the family does not talk about
money, or women are not supposed to know any-
thing about money). We then work on eliminating
undesirable patterns or attitudes about money.
The second family meeting (what I call the
money meeting) goes into more depth, although I
have designed it to be engaging, not intimidating. I
remember my first such meeting. The family had 16
members, with ages ranging from 8 to 74, and the
unspoken rule in this family despite its enormous
amount of wealth was not to talk about money. I
composed 150 questions about money and put the
questions in a hat so they could be passed around. If
56 • www.aimrpubs.org
someone picked a question he or she did not like,
another question could be chosen.
Imagine the silence in the room when the 8-year-
old drew the question, “Grandma, what is the dumb-
est thing you have ever done with money?” Equally
provocative was the question, “What has money
done for our family, whether good or bad?” The
longest silence was produced by the question, “In our
family, are there strings attached to money?” I had
them write their answers on a piece of paper so I
could look at them later. As a facilitator, I do not
tolerate confrontation. I prefer to work with individ-
uals behind the scenes. In this case, I discovered that,
indeed, in this family, many strings were attached to
money.
The next task is to identify the patterns that need
to be changed in order to develop positive behaviors
and attitudes about money. I track these on a chart,
identifying benchmarks that will be used to measure
progress.
I later meet individual family members with
their investment manager, estate attorney, or whom-
ever, depending on the topic that needs to be
addressed. I also work with any young adults who
have, or will soon have, access to a trust fund to help
that young adult (the beneficiary) and the trustee to
agree on their relationship. I follow the recommen-
dation that Hughes makes in his book to clearly
define the roles, rights, and responsibilities of each
party. Even though young adults do not tend to think
about long-term goals and are not likely to read
Hughes’ book, two questions invariably pique their
interest: How is their portfolio doing, and what are
their family members’ attitudes and behaviors about
money?
Inheritance can be a touchy subject. While many
parents are comfortable sharing information about
their estate plan because they appreciate the value of
sharing the information so that the next generation
can develop its own strategies, other clients refuse to
discuss it. I recommend treating this topic gingerly
by acknowledging and honoring client wishes in this
regard. Sometimes, I have to tell potential inheritors
that no information about their future inheritance is
available, and I help them deal with that frustration.
For families that are philanthropically inclined,
family meetings are a perfect opportunity to focus on
the importance of charitable giving. To hear about a
grandmother’s experience serving as a volunteer fire-
woman can be important for a child’s self-esteem.
The discussions are also designed to help the family
identify its philanthropic interests and create vehicles
for charitable grant making.
With young children, I hold special meetings to
discuss allowances, spending plans, wish lists,
©2003, AIMR®

delayed gratification, and how to be successful with
money. I encourage the idea of investing together as
a family because it teaches concepts and skills to
young family members that they might not learn
otherwise. My partner, Charlie Haines, remembers
being invited to his mother’s investment committee
meeting at the age of 14. Inspired, he purchased two
shares of stock. Unfortunately, they were shares in
Pan Am Corporation. Nevertheless, the committee
members guided him by emphasizing what he
gained from the experience of losing money. Letting
children make mistakes with money is an important
learning tool.
A family has hired me and a colleague, Charles
Collier, a senior philanthropic advisor at Harvard
University, to coach its next generation of nine chil-
dren on how to be responsible with money. Through
cash flow analysis, goal setting, and benchmarks, we
are trying to impart a sense of responsibility to them.
To do this, we created a series of financial education
programs with one of the family’s investment man-
agers, and this approach has worked well.
One of the children, a 21-year-old trust fund
beneficiary who has been kicked out of three private
preparatory schools, initially said the purpose of his
trust fund was to allow him to buy anything he
wanted. I asked him to be more serious, but he had
no other response. I knew I had my work cut out for
©2003, AIMR®
The Psychology of Wealth
me because he did not say it allowed him to do any-
thing he wanted. In time, after several meetings in
which I managed to get him to talk about his goals,
his values about money, and his family’s values
about money, he decided he wants to do things in a
different way from his family. He is wondering how
many houses, cars, and boats he really needs. I have
also worked with him on changing the nature of his
relationship with the family trustee from that of
“money cop” to valued resource.
Conclusion
By giving equal consideration to the soft side of
money as well as to the hard side, advisors can build
more effective relationships with their clients. Advi-
sors should stress to their clients that money is merely
the means to an end. If goals are unclear or nonexist-
ent, the money serves no purpose. Recall the Cheshire
cat’s response to Alice in Alice’s Adventures in Wonder-
land when she is lost in the garden and asks the cat
which way she should go. “That depends a good deal
on where you want to go,” says the cat. When Alice
responds that she does not care, the cat replies, “Then
it does not matter which way you go.” Likewise, if an
advisor does not know which way a client wants to
go, then the advisor cannot optimally manage that
client’s money.
www.aimrpubs.org • 57
Investment Consulting and Its Evolving
Role in Advising Family Offices
(as corrected September 2003)

Scott D. Welch
Managing Director
Lydian Wealth Management
Rockville, Maryland

The rise of “insti-viduals”—individuals whose needs are virtually institutional in their

complexity—means that the demand for investment consultants is likely to grow. But it
also means that consultants must meet the requirements of ever more sophisticated
clients. Indeed, the key to success for consultants may be the ability to differentiate
themselves not through quantitative performance but through the qualitative aspects of
the services they provide.

ore and more family offices are relying on
M investment consultants to provide a variety of
services, from investment plan development and
asset allocation to manager selection, performance
measurement, and more. The impetus behind this
presentation was a posting that was made to the
“member channels” of the website for the Institute for
Private Investors (IPI), a noncommercial consortium
of wealthy families and advisors.1 Based in New York
City, IPI has between 200 and 300 members, approxi-
mately 40 percent of whom have $200 million or more
in assets. The following comment struck a particular
chord with me and led to a rather lengthy “e-mail
thread” on the member channels:
Most of my colleagues at the IPI with whom I am
friendly do not use advisors/consultants for one or
more of the following reasons. I would very much
appreciate any comments on these reasons:
• lack of value added,
• lack of “wattage,”
• not being “on the line,” and
• lack of candor in advising.
This comment brings to mind the old saying,
“Other than that, Mrs. Lincoln, how was the play?”
It prompted much discussion within my firm (and, I
am sure, at other consulting firms) as to how we can
1
The channels, available only to members of the IPI, can be
accessed at www.memberlink.net.
Editor’s note: Lydian Wealth Management was formerly CMS
Financial Services.
prove our worth. To that end, I will discuss several
important trends that are making the investment
consulting business more vital than ever before. In
addition, I will list the attributes that consultants
should bring to the table, illustrated by our perspec-
tive at Lydian. In particular, consultants’ value added
can be discerned by their clients through the man-
ager’s selection process that the consultants follow
and the consultants’ performance reporting capabil-
ities. Finally, although identifying the right consult-
ant is not easy, I will offer some suggestions on how
a family office might go about finding the best con-
sultant for its needs.
Industry Trends
A generally accepted hypothesis is that asset alloca-
tion and discipline in adhering to an investment plan
are the greatest contributors to long-term investment
success. In my view, a number of trends in the private
wealth management industry are making consult-
ants instrumental in achieving these goals. As long as
consultants can prove to their clients that they are
doing a good job designing allocations and maintain-
ing discipline, they should be able to justify their fees.
In the words of Dr. Alan Starkie (a consultant
specializing in placing business development officers
and client relationship managers for high-net-worth-
focused firms), wealthy families are “insti-viduals”—

58 • www.aimrpubs.org ©2003, AIMR®

 Investment Consulting and Its Evolving Role in Advising Family Offices
individuals who have institutional needs in terms of
complexity and sophistication.2 As a result, more of
these families are forming family offices, more family
offices are starting or joining multifamily offices
(MFOs), and more family offices and MFOs are look-
ing to outsource expertise—all of which bodes well
for the investment consulting business.
Family Offices. Many families retain extraor-
dinary wealth, despite the recent market downturn,
and many of them are forming or have formed family
offices. According to recent data, the number of fam-
ilies with intergenerational wealth is growing at 12
percent a year and more than 3,500 dedicated family
offices now exist in the United States.3 Furthermore,
based on IPI’s annual survey of member families, the
percentage of respondents who have a family office
rose from 62 percent in 1999 to 80 percent in 2000 (the
most recent data available).4
Alternative Advice. Families are not only more
willing to form family offices; they are also more
likely to seek advice from alternatives to such tradi-
tional service providers as banks, trust companies,
and brokerage firms. The reasons for this transition
are numerous and include the following:
■ Perceived conflicts of interest. Media reports
about conflicts of interest between the research and
investment banking divisions at major brokerage
firms have been discouraging.
■ Captive product mix. Many firms claim to offer
a broad mix of investment alternatives to their clients
through an “open architecture” platform but do not.
■ Constant acquisitions. Constant acquisitions
tend to create uncertainty and a lack of continuity
within an organization, both of which make clients
uncomfortable.
■ High turnover rate. The turnover rate for those
involved in relationship management is high in the
corporate world. The problem is that just as someone
becomes adept at understanding client needs, he or
she moves on to a different job, either within the same
organization or with a rival firm.
■ Poor investment performance. Poor investment
performance is one of the main reasons clients are
seeking alternatives to traditional service providers.
More than any single factor, the recent market down-
2
Alan Starkie, “The Myth of the Million Dollar Revenue Producer,”
Private Asset Management (4 February 2001).
3
Based on data from the Merrill Lynch/Gemini Consulting World
Wealth Report 2000 (www.foxexchange.com/public/fox/news/
industry_trends/world_wealth_report_2000.pdf) and the Family
Office Exchange (www.familyoffice.com).
4 6
IPI Year 2000 Member Profile Survey. This survey is available only
to members of IPI and can be accessed at www.memberlink.net.
©2003, AIMR®
turn has focused investors’ attention on the subopti-
mality of using only captive or internal asset
managers.
■ Wrong business model. The salesman/portfolio
manager model—whereby a business development
officer gathers clients and assets while a portfolio
manager manages the assets and serves as the primary
contact for clients—has not proven to be successful.
When clients have questions or concerns, especially in
times of poor market performance, portfolio managers
may not be the best professionals to turn to for
answers. (Put less delicately, the very characteristics
and personalities that make someone a good portfolio
manager often make that same person a less than
optimal contact point for concerned clients.) Accord-
ing to some independent studies, a relationship man-
ager/support team model is better in terms of both
gathering assets and retaining clients.5 A relationship
manager remains with a client throughout the invest-
ment decision-making process and maintains knowl-
edge of the client’s investment needs and preferences.
A team of specialists in such areas as portfolio man-
agement, concentrated wealth, trusts and estates, and
accounting provides the necessary support. The pre-
viously mentioned surveys indicate that clients much
prefer this business model, as they have a consistent
contact person who understands their needs and
objectives and who can bring to bear the necessary
experts within the organization to address specific
needs.
■ Sellers of products. Many clients perceive tra-
ditional service providers as sellers of (often captive)
products rather than trusted providers of advice.
Outside Consultants. As another indication of
a trend toward the use of consultants, according to
IPI member surveys, the number of families retaining
outside consultants increased from 35 percent in 2001
to 42 percent in 2002.6 A majority of families still do
not use consultants, but survey respondents are, by
definition, self-selected, so the results need to be
viewed in context. I suspect we will see an increase
in the use of outside consultants as the following
factors come into play:
■ Buy-or-build decision. When high-net-worth
investors form family offices, a decision has to be
made whether to “buy” the necessary support struc-
ture (i.e., enlist outside experts) or to build it internally.
5
Based on data from studies conducted by Tiburon Strategic Advi-
sors and the Creating Equity Group. These surveys can be accessed
at www.tiburonadvisors.com and www.cegworldwide.com.
IPI Year 2002 Member Profile Survey. This survey is available only
to IPI members and can be accessed at www.memberlink.net.
www.aimrpubs.org • 59
Investment Counseling for Private Clients V
Buying the support structure through a consultant is
often the best choice. Building the structure can be
quite a lengthy and expensive process. For example, it
has taken Lydian 10 years to build the business we
now have.
■ Generational changes. As generational changes
occur within families, family members may become
less attached to their historical providers of advice
and less interested in directly managing their assets.
If the third, fourth, or sixth generation has no interest
in running a family office or even being a part of it,
then hiring a consultant is an obvious alternative.
■ Acquisitions. Consider some noteworthy
events of the past few years: SunTrust Bank acquired
Asset Management Advisors, Wilmington Trust Cor-
poration acquired Balentine & Company, and Atlan-
tic Trust Group acquired Pell Rudman & Company
(prior to being acquired itself by Invesco). All of the
acquired firms were independent investment con-
sulting and wealth management firms prior to being
bought. These are just three examples of a trend in
which more and more trust companies seem to be
acquiring consulting firms. One possible explanation
for these acquisitions is that the trust companies—all
of which had historically captive investment man-
agement products—realized the “flight risk” associ-
ated with younger beneficiaries who are seeking the
optimal (i.e., objective) investment solution. Given
the choice of building a business internally or buying
one, buying one made more sense. As an alternative
to the acquisition model, however, I will note that
Lydian has successfully partnered with several
regional bank and trust companies that sought to
expand their wealth management offering without
necessarily acquiring another company.
■ Specialization. No firm can possibly excel in
all areas. Yet, some firms that brand themselves as
MFOs attempt to position themselves as being capa-
ble of providing “best-of-breed” performance across
multiple services: investment management, wealth
management, and family office services (such as tax
return preparation, accounting, and record keeping),
which implies they can perform these services better
than an outsourced solution. True consulting, how-
ever, entails relying on specialists—the best providers
for whatever the service may be—rather than main-
taining all services in-house.
■ Lack of omniscience. People often ask whether
the past three years have been difficult with respect
to maintaining existing client relationships and
developing new ones. The answer, at least for Lydian,
is no. The past three years were not nearly as difficult
as the period from 1996 to 1999, when many inves-
tors, confusing a bull market for investment bril-
liance, thought they did not need professional
60 • www.aimrpubs.org
assistance. Over the past three years, however, the
concepts of objectivity, discipline, diversification,
and asset allocation have regained popularity, which
has been good for the consulting business.
■ Independent expertise. Clients have come to
recognize the need for independent expertise and
access to the “best and brightest.”
■ Improved technology. Technological advances
have made “virtual” family offices a reality. Family
members can communicate directly with each other
and run a successful operation without having to be
in the same location.
MFOs. More families are either joining or form-
ing MFOs. They are being launched by both individ-
ual family offices and institutional firms (banks and
brokerages)7 for a variety of reasons:
■ Outside investors and profitability. As more
families start hedge funds internally, discover hedge
funds, or create funds of funds within their own
family offices, they find they can attract new clients
with this expertise and increase profitability.
■ Economies of scale and cost reduction. Another
driver has been the need to contain costs. Sometimes,
a family office’s fixed costs are too high to efficiently
serve only its dedicated family. Accepting addi-
tional clients allows the office to leverage its struc-
ture and hired expertise, reducing the dedicated
family’s costs.
■ Consolidated buying power. Consolidated buy-
ing power allows consultants to negotiate better
deals for their clients. The goal is to have large firms
treat all clients of a firm as one big client. Placing $500
million in collective assets with one manager, for
example, should reduce costs for all of the individual
clients that make up that $500 million.
■ Threat of stagnation. MFOs can help individ-
ual family offices retain and attract top talent. Often,
once the investment plan is in place and the portfolio
is implemented, talented managers get bored. Open-
ing the doors to other families creates new challenges,
which create a more dynamic workplace.
What Consultants Should Bring to
the Table
To meet the demands of these trends, consultants
need to offer objectivity, expertise, discipline, access
(to managers who might otherwise be off-limits and
to reduced investment minimums), education, and
7
For more information on the involvement of banks and broker-
ages, see “Family Office Roundtable,” Private Asset Management (17
September 2001).
©2003, AIMR®
 Investment Consulting and Its Evolving Role in Advising Family Offices
service. The most critical component is service, but
everything a consultant does should be driven by all
six characteristics.
An issue that arises, however, is how to measure
a consultant’s performance. Consultants, after all, are
not sellers of products but rather providers of advice.
How can they prove they are adding value and pro-
viding “wattage”? How can they demonstrate to cli-
ents that they are “on the line” with respect to results
and being candid in their advice? To answer these
questions, I have identified two tangible (i.e., measur-
able) areas in which consultants can add value: man-
ager search and selection and performance reporting.8
Manager Search and Selection. If consult-
ants can deliver a portfolio construction approach
that integrates and optimizes the performance of
managers and these managers consistently add alpha
and offer tax efficiency and reduced fees, then con-
sultants will be adding value and providing wattage.
Therefore, when vetting managers, consultants
should do the following:
• objectively and proactively monitor the universe
of managers,
• provide access to any manager (not restricted to
legacy, wrap, or platform managers),
• use consolidated assets to negotiate lower fees
(manager, custodial, and trading) and greater
access (reduced minimums and closed managers),
• evaluate return in a risk-adjusted context,
• evaluate managers using benchmarking and
asset class, peer, and universe comparisons, and
• evaluate the performance of the entire portfolio
as well as each individual component.
Most people in the industry would agree that the
previous summary describes the baseline for compe-
tent consulting. At Lydian, investment goals and
asset allocation come first in the manager selection
process. In other words, the process is driven by the
objectives of the investment policy. When a new
client asks the classic question, “What should I do? I
have $25 million and a wife and two kids,” the correct
answer is “I have no idea.” Before recommending a
particular allocation strategy, the consultant should
find out the client’s lifestyle needs, philanthropic
goals, generational transfer issues, and so on. More-
over, even though we conduct separate manager
searches for each asset class, we are careful to view
the portfolio as a whole, not as a collection of individ-
ual managers. A portfolio construction approach
8 Formore information on these topics, see Scott Welch and Jamie
McIntyre, “Show Time,” Bloomberg Wealth Manager (April 2003)
and Scott Welch, “The Role of the Investment Consultant in Advis-
ing Family Offices,” Bloomberg Wealth Manager (July 2003).
©2003, AIMR®
allows us to evaluate each manager in isolation and
then put together a portfolio that considers manager
correlations both within and across asset classes.
Rather than trying to pick the best manager, the
goal is to select the best combination of managers in
terms of optimal risk–return characteristics. Manager
performance should be monitored relative to bench-
marks and peer groups, keeping in mind that good
asset allocation combines art and science. In other
words, trust the math but use common sense. For
example, if a mathematical allocation model with no
investment constraints was used to construct a hypo-
thetical portfolio, the result could be a 100 percent
allocation to an absolute-return hedge fund strategy
because of its consistent expected return and low
volatility. The math, of course, does not mirror the
intuitive reasoning of an experienced advisor. Such a
portfolio might be appropriate for a few clients but
not for the majority.
We advocate a “core-satellite” approach because
certain asset classes seem to be more efficient than
others. We favor tax-sensitive indexing managers for
the asset classes we deem to be efficient and active
managers for those we believe are less efficient. The
performance of the managers we select is far more
important to us than the performance of those we do
not select, no matter how stellar. We concentrate on
making sure that our managers are doing the job we
hired them to do. We focus on long-term results, not
quarterly performance, and try to be unemotional;
how much we personally like a manager does not
override his or her results. Ironically, in a couple of
cases, families who had long-standing relationships
with managers they wanted to fire, but were uncom-
fortable doing so, hired us to “do the dirty work”;
they hired us to fire them.
Finally, the following quantitative and qualita-
tive factors are important in the manager selection
process:
■ Quantitative factors. On the quantitative side,
look for long-term, risk-adjusted outperformance rel-
ative to the index. We take into consideration all the
usual statistics, such as alpha, beta, standard devia-
tion, Sharpe ratios, and upside and downside capture
ratios. We are not interested in managers who alter-
nate between hitting home runs and striking out. We
prefer managers who hit singles and doubles year
after year and consistently beat their benchmarks in
both up and down years.
■ Qualitative factors. On the qualitative side,
consider how managers generated their performance
(was it on the broad portfolio or because they picked
one or two stocks that drove the performance?),
whether they can repeat their performance, and how
www.aimrpubs.org • 61
Investment Counseling for Private Clients V
they fit with other managers in the portfolio in terms
of holdings and investment style. Another important
factor is whether the current team managing the
assets is the same team that was responsible for the
historical track record. We do not change managers
often and are fairly loyal to the managers we select,
but if someone who was responsible for a manager’s
success leaves, that manager could be fired; likewise,
if a manager claims to be a value manager but
includes growth stocks in a portfolio for no justifiable
reason (i.e., inexplicable style drift), that manager is
likely to be fired.
Other decision criteria include who owns the
firm (particularly whether the manager is an owner)
and how quickly the firm has grown (or not grown).
We are not likely to be interested in a successful
manager who has gathered a lot of assets quickly
because the larger the amount of assets, the more
difficult it can be to replicate historical results.
As a general rule, we tend to select boutique man-
agers or asset class specialists more often than large
multistrategy firms. Interestingly, investment man-
agement divisions of banks, brokerage firms, and
insurance companies are rarely selected. And many of
the best managers neither belong to a wrap program
nor have any interest in being in one. They do not want
to cut their fees for the sake of gathering assets. They
are not necessarily interested in having billions of
dollars under management. They simply want to do
what they do best and make money. No single wrap
program contains all the best managers, a point con-
sultants should make when they talk to families.
Performance Reporting. Basically, consult-
ants can prove they are “on the line” for results and
being candid with respect to advice if they are man-
aging portfolios in accordance with investment
plans; exercising discipline in rebalancing and
adjusting portfolios as required to bring them into
alignment with the plans; providing objective results
via consolidated performance reports; documenting
results in a comprehensive, segmented manner;
meeting with clients at least quarterly; and commu-
nicating when portfolio changes are necessary.
■ Value. Often overlooked by consultants is the
fact that the quarterly performance report is the single
most tangible piece of information the client receives
from the consultant. A performance report serves
several purposes. It is the best tool for comparing
actual performance to the original investment plan
and for objectively assessing how the plan is perform-
ing. It also provides a framework for determining the
factors driving portfolio performance. A good report
helps to identify the need for manager changes or a
re-evaluation of risk–return parameters as well as
62 • www.aimrpubs.org
highlighting new opportunities. Performance reports
are thus more than numbers on a page; they are the
most tangible product a consultant has to offer.
A performance report is also the best tool for
evaluating whether a consultant is adding any value
with respect to discipline, costs and fees, tax manage-
ment, and performance. Note that performance is last
on the list. As I mentioned earlier, good performance
is expected; otherwise, a consultant would not be
hired in the first place. Furthermore, consultants can-
not control the overall performance of the market, so
focusing on performance sets false expectations as to
what the consultant actually can control and dimin-
ishes the focus on those areas where they can add
tangible value—taxes, costs and fees, access to man-
agers, and overall service.
■ Format. A comprehensive performance report
should analyze all of a client’s investment assets
(regardless of custodian), not just those under the
consultant’s advisement. Clients should also be able
to view the performance of each entity, as well as the
family office as a whole, because a family office typi-
cally has multiple investment entities, including
trusts, family limited partnerships, and cross-
generational structures (i.e., the consultant should be
able to deliver a performance report for any and all of
the client’s “composite” portfolios). For example, one
of our clients is a family office with 54 family members
across three family branches and about 150 trust struc-
tures. We provide quarterly reports for each of them
as well as for the aggregate family portfolio.
In addition to being quarterly, reports should be
as user-friendly as possible and follow a “hierarchy”
that allows the investor to drill down to his or her
desired level of information. We begin with a one-
page portfolio summary that describes the portfolio
value at the beginning of the quarter, contributions,
withdrawals, fees incurred, portfolio value at the end
of the quarter, and why the values changed. Many
clients find that information to be sufficient (i.e., all
they really want to know is how much they are worth,
how much that value has changed, and why), but
other clients want to know how each asset class per-
formed or what their hedge fund exposure is. Some
clients also want to know the performance of each
asset style or which manager added the most value
or caused the most drag. And although we do not
encourage clients to ask for security-specific report-
ing, we can provide the data for each security that was
bought or sold by each manager within the portfolio.
By segmenting the performance report into these sec-
tions, our report allows clients to review areas in
which they have interest, depending on their objec-
tives, knowledge base, and level of sophistication.
Note that in our performance report, the performance
©2003, AIMR®
 Investment Consulting and Its Evolving Role in Advising Family Offices
of individual managers is fairly low in the “drill down
hierarchy.” This is intentional; it is typical for individ-
ual investors to focus on how specific managers are
performing rather than on their portfolios as a whole.
By placing this information fairly deep into the report,
we are signaling (reinforced whenever we speak with
the client) that the real focus should be on how the
portfolio is performing, not on any single component
within that portfolio.
Most important, clients should be able to see not
only how the investment plan is performing but also
what is driving that performance. They should be able
to compare the portfolio results with the plan that was
originally developed to see the sources of any devia-
tion and whether they are temporary or permanent.
They need to decide whether any adjustments are
needed to bring the portfolio back in line with the
plan. Each manager’s performance should therefore
be reported net of fees, relative to benchmarks, rela-
tive to peers, and relative to weighted-index results to
determine how much alpha is being added. (After-tax
performance is also important but is beyond the scope
of this presentation.) Reconciliation of all cash inflows
and outflows should be clear because clients often
lose track of how much money they are withdrawing
to finance their lifestyle. Finally, time- and dollar-
weighted returns should appear on the report for
multiple time periods (quarterly, year-to-date, three-
year, five-year, and since inception).
Identifying the Right Consultant
Identifying the right consultant is not easy for families.
The problem is that most competent consultants offer
similar core services—investment plans, asset alloca-
tion recommendations, portfolio construction, man-
ager search and selection, and performance reporting
(although not all offer consolidated reporting on all
investments)—and generate similar investment
results for a given level of risk over a reasonable time
horizon. Thus, distinguishing one consultant from
another by focusing solely on performance is difficult.
One solution is to differentiate among consult-
ants according to the services they offer. The best
consultant for a specific family office is the one that
offers the specialized services best matched to the
family’s needs. For example, in addition to core
investment management products, a family might
want a specialist in concentrated wealth strategies,
consolidated performance measurement, trust ser-
vices, or alternative investments. Families should also
consider the other services a consultant can provide,
such as consolidating management fees, managing
tax liability, or providing reduced investment mini-
©2003, AIMR®
mums and access to managers who would not be
available otherwise. Once the family determines its
specific needs and goals, it can make decisions accord-
ingly and identify the most appropriate consultant.
One of the most difficult challenges facing true
consultants is in differentiating themselves and
breaking through the “noise” of other, larger firms
that claim to offer independent and objective advice
but really do not. A family office should be highly
wary of advertising campaigns from organizations
offering a wide array of products and should clarify
whether a consultant truly offers an open architec-
ture platform, not just on the management side but
also on the full spectrum of family office services—
accounting, estate planning, alternative investments,
life insurance, mortgage financing, and so forth. For
example, one national firm claims to be an “open
architecture provider with proprietary asset class
products,” while another, also purporting to be an
open architecture provider, claims that it will not use
its own internal managers—a true case of being
“reverse conflicted.”
Service Differentiators. To distinguish among
consultants, families should first consider which
wealth management services they need, whether
concentrated wealth strategies, access to collateral-
ized lending, or creative financing strategies for such
items as mortgages, luxury goods, collectibles, or
travel expenses. Family office services, such as trust
capabilities, bill paying, record keeping, and family
governance, are also important considerations.
Another distinguishing feature is whether the con-
sultant is client centered or investment centered. A
study by John Bowen, principal of the Creating
Equity Group, found that during the recent market
decline, investment-centered firms focused on
investments, portfolio construction, manager selec-
tion, and performance improvement, whereas client-
centered firms focused on communication—making
sure the client understood the investment plan, fos-
tering a client perspective that extended beyond the
current market cycle, and, in essence, doing serious
“hand holding” as investors faced declining portfo-
lios.9 Client-centered firms had a much higher client
retention rate as well as greater success in attracting
new clients than investment-centered firms. For
firms that focus on investment results, this may be
difficult to accept, but the data support it.
Segmentation. Families should be aware of
how the consulting business has been segmented into
various (unofficial) niches, each of which offers a
9 This survey can be accessed at www.cegworldwide.com.
www.aimrpubs.org • 63
Investment Counseling for Private Clients V
slightly different array and level of services. Some
firms follow the external chief investment officer
model. The historical business of such firms was the
institutional market, with later expansion into the
high-net-worth market. These firms typically limit
their services to investment management-related
areas: investment policy statements, asset allocation,
portfolio construction, manager search and selection,
and performance reporting.
Wealth advisory firms or wealth managers (such
as Lydian) offer services that go beyond basic invest-
ment management consulting. They offer estate plan-
ning and insurance coordination, philanthropic
planning coordination, wealth management (e.g.,
concentrated wealth strategies and access to loans),
tax planning and management, and special projects
management (concierge services).
MFOs frequently build on a wealth management
platform by offering administrative functions, such
64 • www.aimrpubs.org
as bill paying, cash management, tax management,
and family governance.
Conclusion
The family office market is rapidly evolving, with
more family offices, more MFOs, more demands on
providers of services, and more outsourcing. To keep
pace and take advantage of the myriad opportunities,
good consultants need to differentiate themselves in
the industry through their objectivity, specialized ser-
vices, product and service mix, and technological
sophistication. Rather than focusing on performance,
they should concentrate on providing a level of service
commensurate with the demands of “insti-viduals.” If
they fail to do this, the perception will remain that
consultants lack value added and wattage, are not “on
the line” for results, and are not candid in their advice.
©2003, AIMR®

Question and Answer Session
Scott D. Welch
Question: Why don’t you
choose managers from the invest-
ment divisions of banks, brokerage
firms, or insurance companies?
Welch: The bottom line is that,
typically, they don’t meet our
selection criteria. It is nothing per-
sonal. We don’t care where a man-
ager works or how he or she runs
the business, as long as he or she
produces good numbers.
Question: What is the Sortino
ratio?
Welch: The Sortino ratio is a vari-
ation of the Sharpe ratio that was
developed to differentiate between
“good” and “bad” volatility. It
attempts to adjust for the fact that
what clients really don’t like is
downside volatility—they typi-
cally don’t mind upside volatility
at all.
Question: What is your opinion
of MyCFO?
Welch: MyCFO was a well-
financed concept that was perhaps
ahead of its time. As an Internet-
based business, however, the
bursting of the Internet bubble
doomed it. The firm hired a lot of
people quickly who commanded
large compensation packages and
wanted major responsibilities, but
family office services typically are
fairly low margin. People don’t
want to pay much for record keep-
ing, bill paying, and tax prepara-
tion. The firm offered far more than
those services, of course, but its
cost structure relative to its ability
to gather assets was simply not via-
ble. When the firm was launched in
©2003, AIMR®
Investment Consulting and Its Evolving Role in Advising Family Offices
1999, a couple of our large tech
clients were interested in it because
they knew the founder, Jim Clark,
or some of the other initial inves-
tors in the firm. But we never lost a
client to MyCFO and never felt
direct competition from it.
Question: Does Lydian plan to
ever report, as individual manag-
ers do, a composite performance
number?
Welch: We won’t report com-
posite numbers, because each allo-
cation is unique to the client. At the
family office level, unlike the insti-
tutional level, allocations vary
widely. New clients frequently ask
what the “typical” performance of
our portfolios was for the past year,
but there is no typical perfor-
mance; performance depends on
the individual allocation. Instead,
we determine an appropriate allo-
cation for the client and then pro-
vide either historical data or
references to existing clients with
similar portfolios. Clients can
judge for themselves how well the
managers we have selected have
done. Our clients are our best mar-
keting assets.
Question: Over a long period,
do client-centered firms have bet-
ter investment performance than
investment-centered firms?
Welch: I don’t know. The study
I mentioned did not address per-
formance, but I believe that any
competent consulting firm will
have good performance numbers.
Some variation from year to year
will occur among consultants and
advisors, but if two firms reach
similar conclusions about the risk
tolerance, goals, and objectives of a
particular client, their performance
should be similar. The point made
in the study is that with client-
centered firms, clients have a better
understanding of their overall
investment plan and they’re more
comfortable with the process, even
in periods when their portfolio has
underperformed.
Question: How does Lydian
provide investment performance
data for noncustodied assets?
Welch: I should make it clear
that we don’t have a “master”
custodian—our clients and manag-
ers can custody assets wherever
they please. We have our prefer-
ences, but at last count, we were
currently working with around 90
different custodial relationships.
About half of our staff is dedi-
cated to our technology reporting
and reconciliation group. Our pref-
erence for certain custodians is
based only on a couple of criteria.
First, a custodian should offer
favorable custody and trading fees.
Second, it should have reliable
electronic feeds into our perfor-
mance system so that we can run
reports as efficiently as possible. Of
course, clients often have their own
preferences. We accept that as part
of the family office business, even
if it entails obtaining duplicates of
monthly brokerage statements and
manually inputting them into our
performance system.
www.aimrpubs.org • 65
The Importance of Proactive Compliance
Thomas D. Giachetti
Chairman, Securities Practice Group
Stark & Stark Attorneys at Law
Princeton, New Jersey

For financial advisors and investment managers, taking the proper precautions to ensure
regulatory compliance should be viewed not as an inconvenience or disservice to clients
but as an essential protection against punitive action. Firms need to understand the
regulatory examination process, how to handle complaints, and how to make proper
disclosures to their clients. By being proactive and taking the right steps today in three
key areas—preparing for compliance examinations, averting complaints, and handling
disclosures—firms can avoid serious problems in the future.

his presentation will focus on three topics: first,
T how to prepare for a compliance examination;
second, how to avert customer and regulatory com-
plaints; and third, how to ensure that disclosures are
handled appropriately. My central point is that finan-
cial advisors and investment managers should be
proactive in all of these areas; taking the right steps
today will avoid serious problems in the future.
Preparing for a Regulatory
Examination
The scope of the examination process has definitely
changed in the past two years. It has become much
more comprehensive, requiring me to spend a great
deal of time traveling throughout the country helping
our clients at Stark & Stark prepare for examinations.
Ten steps in particular are crucial to improving exam-
ination readiness.
Obtain the Current U.S. SEC Examination
Checklist. The first step is to get a copy of the cur-
rent U.S. SEC examination checklist. Be warned that
the format of the checklist may differ according to
which branch office of the SEC will perform your
firm’s examination. The Miami office uses a different
format from the one the New York office uses, and so
forth. Most of the questions, however, are the same.
And on any given day, the questionnaire provided
might be one they used four years ago rather than the
questionnaire that they are using now. Unfortu-
nately, the version a firm gets depends on what the
SEC happens to send.
The SEC may give 48 hours notice of an exami-
nation or none at all. If a firm is close to a branch
office, the SEC examiners are likely to provide no
notice. Firms farther away will most likely get some
notice (typically, on Thursday for an examination on
Monday). Either way, a firm will have little time to
prepare; thus, advance preparation is important. The
best way to be ready is to know the questions on the
checklist and the answers to those questions. A firm
that is prepared will be able to provide better service
to its clients during the examination, rather than
focusing most of its energy on gathering information
for the examiners. A prepared firm will also have
more appropriate responses to the concerns of exam-
iners. For example, unprepared firms often supply
documentation that was not requested, which is a
waste of everyone’s time.
Review Prior Examination Deficiency
Letters. Firms should review their prior examina-
tion deficiency letters. The examiners will start with
previous deficiencies. The key is to make sure that
any earlier deficiencies have been corrected. If a firm
indicated in its deficiency response letter that it
would take certain steps to ameliorate a deficiency, it
would be wise to implement those changes before
examiners return. A firm should have a good reason
for any failure to make the promised reforms.
Maintain Good Record Keeping. Good record
keeping is essential. By necessity, large institutions
generally understand the importance of good record
keeping and assign that responsibility and oversight

66 • www.aimrpubs.org ©2003, AIMR®


to specific professionals within the firm. Although
smaller firms have fewer resources, being able to
retrieve and produce documents for the examiners
sends a message about how well prepared the firm is
and how well its operations are run; it also improves
the odds for an examination that is concluded in a
timely fashion. A firm should not entrust record keep-
ing to a part-time employee or an employee who does
not have its complete confidence.
When asked, firms need to be able to go straight
to the right file and find the requested documents.
Furthermore, files should have a consistent method
of organization. The examiners will ask to see client
files in a certain format (as indicated on the checklist).
Typically, they will ask for about half a dozen,
depending on the size of the firm. The number of files
they will actually investigate can vary. If the first few
files are organized without rhyme or reason, the
examiners might review a few more. If the files are
organized in the same manner and kept in the proper
format and the examiners can readily understand
them (because the firm has explained its record-
keeping system), the examiners most likely will not
spend an entire day inspecting files.
Insist on an Entrance Interview. A firm should
insist on an entrance interview because it is an oppor-
tunity to tell the examiners what the firm does and,
perhaps more important, does not do. Remember, at
the time of examination, two years will have passed
since a firm filed its last written disclosure statement
(i.e., Part II and Schedule F) with the SEC. Thus, the
examiners do not have a clue about what a firm has
been doing during the past two years, except for the
Form ADV filing in the Investment Advisor Registra-
tion Depository (IARD). Although the IARD was sup-
posed to improve reporting, I admit that I still do not
understand Form ADV. In short, examiners probably
need a concise explanation of the nature of the firm.
The most valuable information to give examiners
is what the firm is not. This information narrows the
scope of the examination. The goal is to remove cer-
tain potential conflicts of interest (such as soft dollars,
referral fees, relationships with broker/dealers, and
so forth) that examiners may be interested in and to
dispel any preconceived notions they might have
about the firm. Also, firms should tell examiners that
the concerns raised by the SEC during the previous
examination have been addressed in the manner the
firm described in its response letter. Firms that take
this step are likely to find that their examination will
go much more quickly. They will earn the respect of
the examiners and will be able to carry on with nor-
mal business during the period that the examination
is ongoing.
©2003, AIMR®
The Importance of Proactive Compliance
Use a Point Person. Firms should name a
point person responsible for all direct interaction
with the examiners. A primary reason for having a
point person is to prevent the examiners from using
a divide-and-conquer strategy, asking the same ques-
tion of three or four different people and possibly
getting as many different answers. This does not
occur because of problems in the firm but because
different people have different levels of understand-
ing about various practices and procedures. When I
meet with client firms, I usually arrange a quick staff
meeting to tell them how they should address the
examiners and that they should not engage the exam-
iners in conversation. At some point, people may be
asked questions that are important in ways they do
not understand.
The key is to make it clear to the examiners during
the entrance interview that the point person is the
only one who will respond to substantive questions
or requests for copies of any documents. The message
should be conveyed in a clear, nonconfrontational
manner. Effective use of a point person can give a firm
some control over the examination process.
Involve Legal Counsel in the Examination
Process. Firms should have legal counsel involved
in the process. We usually call our clients two or three
times a day during an examination—to help them
manage the process, to help them solve issues that
have been raised by the examiners, and so forth. If an
examiner says a firm should not be engaging in a
certain activity, legal counsel can help the firm decide
how to respond. If the examiners are correct, a
prompt solution may be possible.
Correct Deficiencies during the Examination
Process. Firms should not wait until they get a
deficiency letter; they should correct as many defi-
ciencies as possible during the examination process.
The challenge is finding out what are the defi-
ciencies. Firms cannot always trust examiners to
bring deficiencies to their attention. The point person
should talk to the examiners two or three times a day:
“How are things going? Anything we need to
address? Anything you want to bring to my atten-
tion?” This communication is critical because they
usually will tell the point person what they have
found. Some issues can be addressed immediately,
eliminating examiners’ concerns.
They will bring other issues to the firm’s atten-
tion. After discussion with counsel, the firm can then
respond to concerns or, in some cases, not respond—
if the issues raised are not deemed to be legitimate.
Maintain Document Control. Examiners are
likely to request documentation in addition to what
www.aimrpubs.org • 67
Investment Counseling for Private Clients V
is indicated on the checklist. Firms should maintain
a detailed list of all documents that are provided.
Firms should not give examiners access to their
files. The point is not to tell examiners that they do
not have access to files but to avoid the problem.
Bring them the documents they request. Make copies
if they want copies. It is to a firm’s advantage to know
what examiners are looking at and what they are
concerned about. The point person will better under-
stand the process and the issues raised and be able to
address concerns as they arise.
Insist on an Exit Interview. Firms should
insist on an exit interview, with an eye toward know-
ing what examiners will probably identify in a defi-
ciency letter. The point person should not be the only
one involved. One or two other people should take
copious notes because the point person needs to lis-
ten. When appropriate, the point person should
emphasize that concerns raised during the examina-
tion have already been resolved by the firm. All firms
get deficiency letters. The goal is to narrow the scope
of the letter.
If the process has been managed effectively, the
exit interview should hold no surprises. If new issues
are raised, at least the exit interview can reveal them
in a timely manner.
Firms may not be able to address all deficiencies
during the examination. Some issues may be too
broad or may involve enforcement. The SEC has
changed in recent years. Matters that I formerly could
resolve with a phone call or a quick meeting now may
go straight to enforcement. For this reason, firms
need to be prepared for this eventuality.
Respond to Deficiency Letters Carefully and
in a Timely Fashion. Few firms escape receiving a
deficiency letter that raises multiple issues. If a letter
cites 10 deficiencies, 3–5 of them may be inaccurate
and others may be without merit. The examiners may
have misconstrued some of the information provided
to them or found it to be insufficient. Inexperienced
examiners may not understand the particular way
that a firm does business. In some cases, examiners
do not understand the SEC’s own questions.
When a firm’s leaders receive a deficiency letter,
they should not respond hastily. They need to under-
stand how to respond. They should not say they will
do something that is contrary to their best interests if
it is not required by statute or regulation. This requires
a clear understanding of the statutes and regulations
the examiners are quoting. Firms should develop a
deep appreciation as to what their response may com-
mit them to do. They should never put anything in
writing until they understand its implications.
68 • www.aimrpubs.org
Firms should respond to a deficiency letter in a
timely manner, usually within 30 days. An extension
can be requested but usually should not be needed.
Firms should also follow up on their response. The
examiners will not send a letter indicating that every-
thing is fine. After the examination is over and before
sending a deficiency letter (or even after sending it),
examiners may ask for additional information. Once
in a while, they will raise questions about aspects of
the deficiency letter response, an occurrence that has
happened much more often in the past year or two
than was the case five years ago.
Disaster Preparedness
In addition to the steps I have discussed, firms should
have disaster procedures and policies. Although
disaster provisions are not required, they are recom-
mended. This topic is now included in every SEC
examination checklist, and the examiners want to see
such policies. As a result of September 11, we now
prepare a disaster policy for all our clients in support
of their fiduciary duty to their clients. The most likely
scenario is not a major calamity but a smaller, localized
event. If firms lose power or their computer system or
access to their facilities, they need a comprehensive
policy to address the problems that will arise.
Averting Regulatory and Client
Complaints
Averting regulatory and client complaints is an issue
of critical importance. Over the past year and a half,
we have seen a tremendous influx of complaints
against investment advisors. Broker/dealers have
always been popular targets of complaints from
retail customers, but never before have we seen these
types of complaints against investment advisors.
This change in circumstance is a result of the severity
of recent losses. The legal theory behind the com-
plaints seems to be: “I gave you money; you lost
money. You now have to pay me money.” Whether
the investment was consistent with client objectives
seems to be beside the point. Still, firms can take
certain compliance and disclosure actions to help
avoid such complaints.
When a client complains to the SEC, the typical
result is an SEC request for information. The SEC will
rarely take action against a firm on behalf of a disgrun-
tled client of that firm unless the regulators believe the
complaint rises to the enforcement level; the SEC is
not the proper forum for dealing with damages from
market losses. Nor will state examiners get involved
in market-loss cases, although they may investigate
whether a firm was acting in a manner contrary to its
©2003, AIMR®

accepted procedures or policies, which then exacer-
bated or contributed to the client’s loss.
The greatest concern arising from most market-
loss complaints is a lawsuit. Right now, we are defend-
ing cases in probably two-dozen states throughout the
country, ranging from small to multimillion-dollar
requests for judgment.
How do advisors deal with this threat? First,
advisors should distinguish between macro and
micro events. In the case of a macro event, such as a
major terrorist attack, advisors should send letters to
their clients explaining the firm’s response to the
event and how the event could affect the clients’
investments.
Micro events involve issues concerning a particu-
lar client or portfolio. A micro event should not be
hidden away in the dark like a mushroom. The advisor
should be proactive. If the problem is not addressed,
a minor matter can turn into a liability claim. In deal-
ing with a micro event, a letter will not suffice. The
advisor needs to talk directly to the client—personally,
not through a subordinate.
Advisors inadvertently do certain things to get
themselves in trouble. Typical areas of vulnerability
arise from such actions as administering poorly
defined risk-tolerance questionnaires, disclosing per-
formance inadequately, and using canned docu-
ments. But if advisors understand the areas in which
firms are vulnerable to complaints and take the nec-
essary steps to safeguard themselves, many com-
plaints can be avoided.
Administer Well-Defined Risk-Tolerance
Questionnaires. Many firms administer risk-
tolerance questionnaires to their clients. These ques-
tionnaires often use terms that may not be well
defined. For example, the questionnaire might refer to
aggressive versus moderate risk. My advice to firms
is to pinpoint what the different risk categories mean
for their clients by tying them to particular indexes.
For example, indicate that a certain risk level is com-
mensurate with, say, the Russell 1000 Index and pro-
vide the index’s average annual return—the highs, the
lows, and so forth—for a reasonably long history
before asking the client to answer the question.
Adequately Disclose Performance. Consider
clients who, in 1999, were shown only the past four
years of equity performance. Did the advisor do an
adequate job in terms of disclosure? If a four-year
window of performance was all that was shown to
clients, should the advisor be held liable for losses?
Were clients counseled properly? These are the sorts
of issues that might be scrutinized during SEC exam-
inations and are also often the cause of lawsuits. One
of the most common ways for advisors to get them-
©2003, AIMR®
The Importance of Proactive Compliance
selves in trouble is by hanging themselves with their
own documents. If a client responds to a question-
naire by saying on one page that he or she wants a
10–12 percent annual return but on another that the
maximum acceptable loss is 5 percent a year, there is
an obvious problem. Examiners or lawyers (and, ulti-
mately, courts) might later take an interest in how this
contradiction was addressed.
Avoid Canned Documents. Firms should
avoid “canned” documents purchased from a third
party. Keeping such documents in a file as proof that
a policy exists is not an effective procedure in the eyes
of the SEC. Believe it or not, firms are not currently
required to have a compliance manual, but canned
compliance manuals are being marketed to firms as
necessary tools. I am not a proponent of comprehen-
sive manuals unless a firm is actually reading them
and implementing them effectively. The worst sce-
nario is to have a lawsuit expose that the firm is not
doing half of the things it says it should do in its own
manual.
Understand the Importance of Client Intake
Documents. Probably the most important docu-
ment in the client’s file is the investment policy state-
ment (IPS), investment profile, or client
questionnaire. The professionals at the firm should
read each client’s IPS and understand the obligations
that are entailed. The IPS is not a rudimentary, rou-
tine document. In fact, having an IPS can give a firm
a false sense of security. The IPS is sometimes the best
place for the SEC to assess liability. For example, it is
not unusual for the IPS to state that the firm will not
change a client’s portfolio weights (or anything else)
unless the request from the client is in writing but, in
practice, changes frequently will be made to the port-
folio in the absence of a written request in direct
contradiction to the IPS.
I developed a one-page letter for a firm to give its
clients, a mini-IPS. It was essentially a form letter
stating that the firm would allocate client assets
among specified asset classes in accordance with the
client’s designated investment objectives, and
requesting that the client advise the firm of any
change in his or her financial or personal situation or
of a desire to impose new restrictions on the manage-
ment of the account. Every document sent to clients
should have such a disclosure in it, putting the onus
on clients to inform the firm if something has changed
that may require a reassessment of the investment
management process, consulting arrangements, or
financial planning recommendations. This approach
prevents clients from claiming that the firm never
contacted them because the firm will have ample
documentation to refute the complaint. Attention to
www.aimrpubs.org • 69
Investment Counseling for Private Clients V
such details can provide considerable protection for a
firm. The IPS is not a document that is cast in stone.
Like compliance, maintenance of the IPS is an ongoing
process. Firms should not get locked into an IPS that
can cause problems for them later. These documents
should be reviewed regularly.
Another important document is the investment
management agreement. Investment management
agreements serve one purpose—to protect advisors
and managers, not to give the client warm feelings
about the advisor or manager. Out of 100 clients, 98
will sign the agreement. Clients have been signing
brokerage agreements for years, and if they will sign
those, they will sign virtually anything. The agree-
ments should tell the client what the manager will do
and will not do for them. If a firm does not offer
certain services (such as financial planning, estate
planning, and insurance planning), its investment
management agreement should declare that the firm
does not offer these services. Clients need to under-
stand the scope of the engagement because if the
client dies, the heirs may blame the investment man-
ager, complaining that the proper estate or insurance
planning was not in place. The manager needs to be
able to document that clients were informed that such
advice and services would not be provided.
The financial planning agreement, also an
important document to have in the client’s file,
should say that the engagement does not include
ongoing investment implementation, monitoring or
reviewing, supervision, or management services. It
should say that clients who want other services must
contract them pursuant to the terms and conditions
of a separate engagement, for which the client will
pay a separate, additional fee.
A good disclosure statement specifies what the
firm does and does not do. It states that services are
limited to certain activities and describes how they
will be executed. The point is to protect the firm.
Managers are being sued every day, and most of them
cannot point to documents indicating that clients
were notified of the issues or concerns that gave rise
to the suit. A firm that receives a complaint should be
able to respond that it gave notice not only in the
initial agreement but also in disclosures made in
invoices, correspondence, and newsletters. For those
who truly hate to give disclosure statements to clients
unless requested, the minimum is to give them a
complete disclosure every year, not every two to four
years. Documentation of regular disclosure is the best
defense against charges of conflicts of interest or com-
plaints about how the advisory business is run. Firms
should not change their agreements and disclosures
to accommodate the wishes of a few individuals. In
fact, losing such clients is probably desirable because
70 • www.aimrpubs.org
they will become a problem sooner or later. Agree-
ments are designed to protect advisors and managers,
as are disclosures. Most people do not even read such
documents and sign them without a quibble.
When the market takes a turn for the worse, firms
should remind clients that their assets have been
allocated in a manner that is consistent with their
own objectives, that the advisor does not know what
the market will do in the future, and that the advisor
is not the cause of market behavior. It is fair and
appropriate to remind clients that, although the advi-
sor may have helped them understand their objec-
tives and allocate their assets properly, what goes up
may eventually come down (or at least stop going up)
and that the market is no exception. By the way, a
primary function of advisors is to help clients define
reasonable objectives. Clients who insist on unrea-
sonable objectives should be avoided because they
eventually will come back to bite the advisor.
Recognize the Risks of Firm Newsletters.
Many firms send newsletters to clients, and these
newsletters pose serious risks. Unless a firm’s only
business is writing a newsletter, it should avoid
forward-looking statements. It should not make
declarative statements about the future. No matter
how great their expertise or their previous success,
firms do not know what will happen next. And any
newsletter or marketing letter should contain disclo-
sures for everything. For example, I am defending a
case in Wisconsin in which an advisor is being sued
for $3.5 million. For two years, the firm told its clients
that the technology sector would return to its near
former levels. Month after month, the firm repeated
this same refrain. I do not know how to defend the
case, because the firm clearly told its clients to stay put.
Review Liability Coverage Exclusions.
Another important issue is not having sufficient
insurance liability coverage. Firms need to read their
insurance policies. Premiums are going sky-high, but
coverage may be lacking. Firms should focus not on
what is included in the coverage, which usually
amounts to only a few lines of text, but on what is
excluded. Most policies have several pages of exclu-
sions. I review all of the policies for my clients
because I want to know that they have appropriate
coverage for the activities in which they engage. It is
important to have an insurance carrier that respects
an investment manager’s fiduciary duty. If a client
loses money because the manager made a mistake,
the manager has a fiduciary duty to inform the client.
But such a disclosure may be treated by the carrier as
a breach of the terms of the policy. Managers need to
review their policies in this regard, question their
agents, and know who is underwriting the policy.
©2003, AIMR®

Most important, when a client makes a claim and
demands to be compensated, the manager should take
absolutely no action before reporting the claim to the
insurance carrier. Managers should do nothing to
jeopardize their coverage because, in its current con-
dition, the insurance industry is eager to avoid liabil-
ity. I recommend letting the manager’s legal counsel
report the claim to the carrier. The policy may or may
not allow the manager to appoint counsel, and this
step will allow the manager to have greater influence
on the decision regarding appointment. If the man-
ager’s counsel has insurance defense experience and
is willing to work at the rates the insurance carrier will
pay, the manager’s appointment may be accepted.
This is an important concession because the manager’s
counsel is focused on the manager’s interests. Counsel
appointed by the insurance company may be more
focused on the carrier’s interests than the manager’s.
Disclosures
Firms should not use canned agreements or canned
disclosure documents. Using canned documents is
like trying to put a square peg in a round hole. Dis-
closure documents should specify how each service
is handled, from asset allocation to block trades. They
should not contain language that is inappropriate to
the particular firm. Using a canned document may
mean one day being held to a standard that the
manager does not understand or one that is not appli-
cable to the particular manager’s business. Most of a
manager’s activities require disclosures, but some
activities have fallen under stronger scrutiny in
recent years, including IPOs, client-directed trades,
websites, client-directed brokerage, and referrals.
IPOs. IPOs are a subject of particular interest to
examiners. If a firm plans to get involved with IPOs,
it should establish a policy before beginning to allo-
cate IPOs to its clients. During the market bubble,
many investment management firms were buying
IPOs on a regular basis, including smaller firms that
got access to the deals through their clearing arrange-
ments. Instead of having a policy in writing or a
procedure for how to allocate these IPOs, they were
giving them to their best clients only and not rotating
the allocation. Such discrimination is inappropriate
and can get a firm in real trouble.
Generally, there are three types of IPO policy;
one of the three should be added to the client’s doc-
uments. The first type says, in effect, “We do not
recommend IPOs, but if we get an allocation through
our clearing arrangement and you ask for an alloca-
tion and we agree that it is suitable, you will get an
allocation.” The second type says, “We recommend
IPOs, and we have a procedure for allocating them.”
©2003, AIMR®
The Importance of Proactive Compliance
The third type says, “We purchase IPOs for our pro-
prietary hedge funds only.” Regardless of the policy
type to which a firm adheres, all firms need to have
a definite IPO policy.
Client-Directed Trades. Clients may call man-
agers and ask them to buy shares of a particular stock.
Most managers will oblige, but they should use an
accommodation letter stipulating that the security
was purchased as an accommodation, that the man-
ager prefers that the client use a separate retail
account for such acquisitions, and that the manager
did not choose the security and will neither be
responsible for its performance nor monitor the secu-
rity as part of assets under management.
During the bull market, many of my clients were
inundated by calls from their clients begging them to
buy certain stocks. When the stock prices later
plunged, the managers were often blamed. The solu-
tion is to set up an accommodation or courtesy
account for such transactions and to document how
the matter was and will be handled. In such cases,
clients may get preferential trading prices through
the manager’s clearing firm, but the manager does
not take responsibility for the investment.
Similarly, clients often ask managers to handle
stocks they already own or that they inherited. Man-
agers should use an excluded assets letter that stipu-
lates that such stocks are not part of assets under
management and that the manager will be available
to consult with the client but will not be responsible
for making decisions regarding the holding. The goal
is liability prevention. Firms will find it much more
difficult to protect themselves after a problem has
occurred.
Websites. Many firms have their own websites,
which are often purchased as canned products from
outside providers. A website requires comprehen-
sive disclosures. Clients may have access to such
third-party links as economic calculators and news-
letters from other sources. Disclosures should be all
over the website.
The disclosures should be accurate. For example,
many websites have disclosures that say, in effect, “I
cannot do business in your state unless I am regis-
tered.” This claim is not true. An SEC-registered
advisor can have a website in any state and can have
up to five clients in almost any state without filing
notice with that state. Only four states have no
de minimus exemption: Vermont, New Hampshire,
Texas, and Louisiana. So, an advisor with even one
client domiciled in any of these four states must file
with the respective state. (Most states require filing if
the firm maintains a place of business in the state.)
All firms should regularly review their clients’ state
www.aimrpubs.org • 71
Investment Counseling for Private Clients V
of residence to make sure that they have made all
necessary regulatory filings.
Client-Directed Brokerage. Client-directed
brokerage is a serious topic. Some clients request that
a manager direct the client’s brokerage to a particular
broker. I am involved in an enforcement matter that
centers on directed brokerage. Although the SEC has
not yet provided concrete guidance regarding this
issue, it will soon do so.
In such cases, managers need to tell the client in
writing—ideally in a disclosure statement in the
investment management agreement—that the client
has directed the manager to use a particular brokerage
and that this direction may result in higher transaction
fees or commissions than may have occurred with the
clearing arrangement available through the manager.
Referrals. Firms should beware of receiving
referrals from broker/dealers. Directing clients to
those broker/dealers is a problem. It is a problem
even if the client directs the manager to use the
broker/dealer. The SEC is saying that brokerage
directed to the referral source is a significant conflict
of interest. The firm must tell the client that using the
broker that referred the client may not be in the
client’s best interest. Firms need to negotiate best
execution because if referred clients are paying up for
brokerage and disclosures are inadequate, there is
real potential for trouble.
Solicitors or third-party referral agents need to
understand that the rules differ from state to state.
Rule 206(4)-3 under the Investment Advisers Act of
1940, “Cash Payments for Client Solicitations,” states
72 • www.aimrpubs.org
that if the manager is going to pay a referral fee, there
must be a written agreement between the manager
and the solicitor.1 At the time that the solicitor intro-
duces the manager to the prospective client, the client
must receive not only a disclosure statement from the
manager but also a statement from the solicitor dis-
closing the solicitor’s relationship with the manager.
The client must be told that if the client engages the
manager, the solicitor will receive a certain fee. Even
if the manager takes such precautions, the manager
must ascertain whether the solicitor has any individ-
ual registration requirements for the state in which
clients are solicited. Few people understand the
scope of the solicitor rules. The rules differ signifi-
cantly from state to state. If four examiners from a
state are asked the same question about solicitor
requirements, it would not be unusual for them to
provide four different answers.
Conclusion
A financial advisor or investment manager’s job is not
to be a hero for clients but to allocate client assets in a
manner that is consistent with designated investment
objectives. Taking the proper precautions should be
viewed not as an inconvenience or disservice to clients
but as an essential protection against regulatory or
legal action. Firms need to understand the regulatory
examination process, how to handle complaints, and
how to make proper disclosures to their clients.
1
Rule 206(4)-3, “Cash Payments for Client Solicitations,” can be
accessed at www.sec.gov/rules/extra/iarules.htm#20643.
©2003, AIMR®