invest mainly in stocks; others invest in the gamut of assets, including
Most share one thing in commonā”high fees, often a ļ¬xed fee equal to
1% of asset values each year plus 20% of all net proļ¬ts. We should not be
surprised that many of the very best investment managers become hedge
fund managers, because their compensation can be astronomical. And if the
manager is good enough, the high fees can be worth our paying. After all,
the only thing that counts for the investor is long-term returns net of fees,
and certain hedge funds have provided some of the best returns available.
The problem is that the high fees of a hedge fund donā™t necessarily
mean high returns. There have been well-publicized instances of investorsā™
value in a hedge fund being completely wiped out by the managerā™s specu-
lation. We therefore must have extraordinary conļ¬dence in a hedge fund
manager in order to agree to his high fees.
Another drawback of hedge funds is that they have not usually been
very forthcoming about the composition of their portfolios. Institutional
investors may ļ¬nd it difļ¬cult to plug the composition of hedge funds into
their fundā™s overall asset allocation. Transparency has improved in recent
years, however, under continuing pressure from investors.
It is true that some of the best investment managers of our time are
hedge fund managers, and they have made their investors rich. Is our ad-
viser up to identifying who they are?
One way to approach such hedge funds (or market neutral funds) is
through a fund of funds. Yes, that adds another heavy layer of fees. But the
fees may be worth it if the fund of funds (a) has long experience in identi-
fying the best funds, (b) is investing in many hedge funds that are closed to
new investors (because wise hedge fund managers recognize that size can
limit their returns and therefore their incentive fees), and (c) has a compe-
tent staff to follow closely all of the funds in its portfolio.
Illiquid investments are private investments that may be difļ¬cult to sell
within a year, or perhaps impossible to get out of for the next 5, 10, or 15
years. Illiquid investmentsā”usually limited partnershipsā”include:
Real estate funds
Venture capital funds
Other private corporate investments, such as buy-in or buy-out funds
Distressed securities (most of which are illiquid)
Oil and gas properties
92 ALTERNATIVE ASSET CLASSES
Characteristics of Illiquidity
Are illiquid investments prudent? Yes, provided we have enough mar-
ketable securities that we can convert to cash in time to meet our fundā™s po-
tential payout requirements.
Most investment funds hold far more liquid assets than they need, and
by doing so, they may be incurring a material opportunity cost. As a gen-
eral rule, the more marketable an asset, the higher its price is bid up, and
therefore the lower the return we can expect from it. Thatā™s why prices of
the largest, most active stocks generally carry a āliquidity premiumā over
prices of less actively traded stocks. We pay a price for liquidity.
Conversely, prices of illiquid investments should be lower. There should
be an āilliquidity premiumā to the net return on private, illiquid investments.
The word āpremiumā is in quotes because caveat emptor applies especially
to private, illiquid investments, which come with fees far higher than fees on
normal common stock accounts. But if we can invest intelligently in a diver-
siļ¬ed group of private, illiquid investments, we should expect a somewhat
higher return per unit of risk than on marketable securities.
āActive managers willing to accept illiquidity achieve a signiļ¬cant edge
in seeking high risk-adjusted returns,ā writes David Swensen of the Yale
endowment fund. āBecause market players routinely overpay for liquidity,
serious investors beneļ¬t by avoiding overpriced liquid securities and locat-
ing bargains in less widely followed, less liquid market segments.ā6
Because of the difļ¬culties of valuing private investments, and because
a private investment purchases its assets over time and then sells them
over time, time-weighted rates of return have little meaning. We should
evaluate the performance of a private investment by dollar-weighted rates
of return (internal rates of return, or IRR). And again, because valuations
are so suspect, the only solid performance ļ¬gure of a private investment
is its dollar-weighted rate of return on our contributions, calculated from
our ļ¬rst contribution to the last payout we receiveā”net of all fees and
expenses, of course.
Because private investments funds are illiquid, our investment fund should
hold only a small percentage of its assets in any one private investment
David F. Swensen, Pioneering Portfolio Management (The Free Press, 2000), p. 56.
Private Asset Classes
fund. If we would like a meaningful allocation to private investments, we
should build a portfolio of diverse private investments funds over time. We
want diversiļ¬cation by kind and by time. Time diversiļ¬cation is important
in private investments because there are common factors that impact re-
turns to partnerships of each vintage year, and it is close to impossible to
divine up front which vintage yearā™s partnerships will be most successful.
If we are a small endowment fund, say, less than $50 or $100 million
in assets, we might be better off sticking with liquid assets. If we do go into
private investments, we would probably want to use a fund of funds,
which can give us instant diversiļ¬cation. But, of course, the fund of funds
charges a high fee on top of the high fees charged by sponsors of private in-
Also, it is even more important than with marketable securities to be
with the very best. Median returns of private investment funds have not
been very attractive. This is best illustrated by the long-term results of ven-
ture capital funds, where there is a wide difference in performance between
the better and the poorer funds. Among venture capital partnerships
started between 1980 and 1995, the difference in IRR between the ļ¬rst-
quartile performer and the median was nearly 9 percentage points per year,
and between median and the third-quartile performer, some 8 percentage
points per year.7 Those are humongous differences, especially when we
consider that 25% did better than ļ¬rst quartile and 25% were lower than
Perhaps the leading manager of funds of venture capital funds, whose
every fund from its ļ¬rst in 1985 has had ļ¬rst-quartile returns, achieved a
net IRR of 29% over the 19 years through 2003. By comparison, the me-
dian venture capital fund returned only 9%, which means that half the
venture capital funds returned 9% or less. Investors in the majority of ven-
ture capital funds could not have felt well rewarded!8
PRIVATE ASSET CLASSES
Of all private, illiquid investments, real estate funds are the asset class most
widely used. Real estate is truly a major asset class, since close to half of
Per Venture Economics, IRRā™s are through year-end 1999.
94 ALTERNATIVE ASSET CLASSES
the worldā™s wealth may lie in real estate. Moreover, real estate values have
had a relatively modest correlation with those of stocks and bonds, mak-
ing real estate a useful diversiļ¬er. Even if our portfolio includes REITs
among its liquid asset classes, private real estate funds may be worth con-
sidering as well.
Real estate is often viewed as an inļ¬‚ation hedge. While far from per-
fect, I believe real estate is a better long-term inļ¬‚ation hedge than most
other asset classes. Few asset classes have investment returns more highly
correlated with inļ¬‚ation.
As always, diversiļ¬cation counts, and itā™s helpful to diversify real estate:
1. By type. Mainly ofļ¬ce (downtown and suburban), retail (major
malls and strip centers), industrial parks (warehouses and light in-
dustrial), apartments, and perhaps single family residential, hotels,
and raw land.
2. Geographically. The various parts of the country, such as Northeast,
Southeast, Midwest, Southwest, Mountain States, and Paciļ¬c. (For this
purpose, the United States is often divided into economic zones that
have the lowest correlations with one another.)
3. By size of property. Such as properties valued at less than $15 million,
those between $15 and $75 million, and those valued at more than
Diversiļ¬cation is advantageous because it lowers the volatility of our
real estate portfolio. For example, ofļ¬ces in one area may get overbuilt,
with values thereby declining, while apartment vacancies may fall unusu-
ally low in another area, resulting in premium rents and prices for apart-
ments there. These individual cycles are additional, of course, to the overall
cycles in commercial real estate.
What rates of return can one expect from core real estateā”from prop-
erties purchased with the intention of holding them for the long term? The
NCREIF index (National Council of Real Estate Investment Fiduciaries) is
the best index of U.S. commercial real estate returns available. It shows
that over the 20 years ending in 2003 aggregate net total returns (if we as-
sume about 1% per year in management fees) were just over 61/2% per
year (about 31/2% real returns, net of inļ¬‚ation), with a standard deviation
of 31/2% per year and a correlation with the S&P 500 of about zero. Given
these ļ¬gures, what should we expect of real estate?
The above return ļ¬gures include the years of 1989ā“1993 when com-
mercial real estate went through its worst depression since the 1930s. I ex-
Private Asset Classes
pect that real returns in a more normal interval would be higher than the
31/2%. For that 1982ā“2003 interval real estate returns were 61/2 percentage
points per year lower than those of the S&P 500. While that difference
should narrow dramatically, I still donā™t believe long term we can expect
quite as high returns from core real estate as from common stocks.
Volatility of real estate is not nearly as low as the 31/2% NCREIF
ļ¬gure. This reļ¬‚ects the fact that appraisals mask much volatility in
the real estate market. Even so, I believe volatility is lower than for com-
The one ļ¬gure I largely believe is the correlation with the S&P 500 of
essentially zero. Although private real estate and common stocks are both
impacted by economic factors, they are impacted by different factors and
at different times. While the stock market was enjoying an historic boom
during the decade 1989ā“1998, returning over 19% per year, private real
estate barely scratched out 5% per year. Then real estate did relatively well
during the stock market debacle of 2000ā“2002. A key advantage of real es-
tate is that it is a good portfolio diversiļ¬er.
Venture Real Estate An approach to real estate investing that I like bet-
terā”especially if we also invest in REITsā”is what I call āventure real es-
tate.ā Simplistically, itā™s where a manager buys a property to which he can
add material value (as through construction or rehabilitation), then adds
that value in a timely manner and promptly sells the property to someone
who wants to buy some good core real estate.
The approach is far more management-intensive than core real estate
and requires greater expertise. Thatā™s why I refer to it as venture real estate.
If we pursue this approach, we should be sure we have especially com-
petent managers, and we should target net investment returns that are
higher than those on common stocksā”at least 8% real (in excess of inļ¬‚a-
tion). In fact, I prefer to wait until we ļ¬nd exceptional real estate funds
from which we can expect net IRRs of at least 10% to 12% real, or 15%
nominal, and then invest in a diversity of such funds.
Individually, venture real estate projects may be more volatile than
core real estate, but it is possible, through commingled funds, to build a
highly diversiļ¬ed portfolio of venture real estate. I am not convinced that
such a diversiļ¬ed venture portfolio is materially more volatile than core
real estate. Nor do I think the correlation with common stock returns is
any higher. I believe there is a higher degree of ādiversiļ¬able riskā in ven-
ture real estate (as opposed to āsystematic risk,ā which cannot be diversi-
96 ALTERNATIVE ASSET CLASSES
The best approach for investing in private real estate is, in my opinion,
commingled funds, since they can provide immediate diversiļ¬cation. The
challenge is to ļ¬nd the best-managed commingled funds and then to nego-
tiate a partnership agreement that aligns the ļ¬nancial motivations of the
manager with those of the investors.
In-House Management of Real Estate Some of the larger investment funds
invest in their our own portfolio of properties. They thereby retain maxi-
mum control. They themselves decide which properties to buy, how to
manage them, whether to leverage them, and when to sell. And they may
avoid the substantial fees that are built into commingled funds and REITs.
Drawbacks are that managing such a portfolio takes a lot of real estate
expertise and time on the part of staff or an outside adviser, and it is difļ¬-
cult to get as broad portfolio diversiļ¬cation.
According to the Commonfund study, nearly a quarter of all endowment
funds larger than $50 million invest in venture capital.
There are broad and narrow deļ¬nitions of venture capital, and here
weā™ll use a narrow deļ¬nition: investment in private start-up companies,
mostly high-tech companies. These companies may be as early-stage as
an idea and a business plan, or as late-stage as a private company that
is already producing a product, needs expansion capital, and may be
preparing to go public (make an initial public offering of its stock). As
thus deļ¬ned, venture capital is probably the riskiest of investments.
Most start-up companies fail to survive, and only a small percentage
become highly successful. How can our fund invest prudently in such
Fortunately, there has developed in the United States the worldā™s most
effective environment for funding and nurturing start-up businesses. Its
venture capital industry is one of the United Statesā™ real competitive advan-
tages. A key to this process, and one that provides a sensible way to invest
prudently in start-up businesses, has been the development of sophisticated
venture capital investment ļ¬rms.
A venture capital investment ļ¬rm consists of a small group of experi-
enced people who have become expert at evaluating start-up enterprises,
identifying the most promising, investing in the best of them, taking seats
on their boards, putting them in touch with those who can provide exper-
tise they happen to need, and advising them on business strategy and rais-
Private Asset Classes
ing capital. These venture capital ļ¬rms form limited partnerships and raise
funds from wealthy persons and institutional investors.
Each partnership may invest its capital over three to ļ¬ve years in some
25 different startup companies, and the fund may take some 15 to 20 years
before it is able to convert the last of its investments into cash through ei-
ther acquisition or an initial public offering (IPO)ā”or must write them off
Despite its managerā™s expert winnowing and nurturing, many ventures
are losers. Most of the rest earn only a modest rate of return. A few home
runs make or break the fund. To illustrate, one of the best managers of
funds of venture capital funds invested indirectly in 2,637 different ven-
tures between 1985 and 2001. Just 25 home runsā”less than 1% of those
venturesā”provided 43% of all returns, and the 8% that repaid at least 10
times their cost provided 72% of all returns. Yet these would be above av-
Venture capital investing is very labor-intensive, and the fees charged
by such venture capital partnerships are very highā”typically 21/2% per
year of an investorā™s commitment to the fund, plus 20% of cumulative
Net internal rates of return (IRRs) to the investors over the life of a
venture capital partnership range from ā“10% per year to +40%, and in
some cases more extreme. Thatā™s a far narrower range than for individual
start-up companies, but itā™s still an extraordinarily wide range of results
from an investment with an average duration of seven or eight years.
The task for us investors is to diversify among the best venture capital
partnerships or enter a fund of venture capital funds, then dollar-average
into additional partnerships over time in order to reduce the range of our
aggregate net IRR expectations. Time diversiļ¬cation is very important in
venture capital because there are common factors that impact returns to
partnerships of each vintage year, and it is close to impossible to divine up
front which vintage yearā™s partnerships will be most successful.
The median venture capital partnership started in the 1980s provided
a disappointing return, while the median partnership begun in 1990ā“1995
achieved nearly 23%, and those that started in the next few years earned
far higher. Then in 1999ā“2000 nearly three times as many venture-backed
companies were formed as in 1995ā“1997.9 This number declined to a low
98 ALTERNATIVE ASSET CLASSES
level by 2002, but returns on the companies formed during that bubble
have been disappointing.
Well diversiļ¬ed, a good venture capital partnership should earn a net
long-term IRR of 15% to 20%ā”well worth an allocation of several per-
cent of our assets. But these returns come in a lumpy fashion, and most
partnerships have negative or ļ¬‚at returns in their ļ¬rst few years, as fees are
greater than returns.
Buy-in funds (like venture capital funds) also invest in private compa-
nies, but companies that are more established, usually ones that are or
have been proļ¬table. Such companies need capital for expansion, for ac-
quisitions, or perhaps even for turn-around. The fund buys privately is-
sued common stock or convertible securities or a combination of bonds
Abroad, both buy-in and buy-out funds are usually referred to as ven-
ture capital funds, but few invest in start-up companies.
The risk of investing in a going concern is clearly less than investing in
a start-up company, but the opportunity to earn 10 or 20 times our invest-
ment is also much lower.
Investors pay fees to the manager of the buy-in fund that in total, in-
cluding performance fees, are much higher than for a typical common
stock program. Hence, the investor should demand a premium return be-
cause of both illiquidity and increased risk. The investor should be highly
convinced that, net of all costs, he can realistically expect to earn that pre-
Whenever the buy-in fund makes an investment, it should have an exit
planā”a process and time line for getting its money out, often through an
assurance that a public stock offering can be held by a given date, or a
promise that the company will be willing to buy back the securities at a
certain price by a certain date (a put).
Buy-Out Funds and LBOs
A buy-out fund purchases the whole company instead of simply providing
a portion of the companyā™s capital. In the process, it either gives strong
Warrants are options to buy stock at a certain price up to a certain date.
Private Asset Classes
support to the existing management team or installs new senior manage-
ment. In either case, it usually ensures managementā™s sharp focus by pro-
viding lucrative stock options to the senior executives.
Sometimes, mainly in the United States, the company is acquired
chieļ¬‚y with debtā”in what is known as a leveraged buy-out (LBO). LBO
funds typically ļ¬nance some 60% or more of an acquisitionā™s purchase
price with debt. The strategy takes advantage of the fact that interest on
debt is tax deductible, which sharply reduces the income tax bill previously
faced by the company. (Without the advantage of tax deductibility of inter-
est on debt, I doubt that there would be such a thing as LBOs. There would
be buyouts, but not leveraged buyouts.) An LBO leverages the investment
for the relatively small number of outstanding common shares. Hence, if
the company is moderately successful, the share owners can realize internal
rates of return above 50% and possibly above 100%.
Such leverage also incurs high risk, of course. A modest decline in
the companyā™s fortunes can leave it unable to meet its debt obligations
and thereby lead to bankruptcy. In that case, the common stock in-
vestors (and sometimes the high-yield bond investors as well) may lose
their entire investment.
Hence, a very high premium is placed on the competency of the man-
agement of the buy-out fund in its selection of appropriate companies to
buy and in its ability to install excellent management in the companies
once it buys them.
Another alternative asset class consists of the vultures of the investment
worldā”distressed security funds. Just as vultures contribute to the ecology
by cleaning up the carrion, distressed security funds buy loans or securities
that other investors no longer want.