When a company heads toward bankruptcy, many investors want out.
Helping a company avoid bankruptcy or nursing it through bankruptcy is
a particular skill that many investors do not have. The skill involves spe-
cialized legal expertise combined with negotiating strategies and corporate
management skills that are a far cry from those required of normal stock
and bond investors.
There are basically two kinds of distressed investors: (1) One will try
to become part of the bankruptcy proceedings and thereby inļ¬‚uence the
outcome. Such an investor may try to obtain as many of the outstanding
shares or loans as possible so he can control the vote on any issues before
100 ALTERNATIVE ASSET CLASSES
the court. (2) Other distressed investors avoid becoming part of the bank-
ruptcy proceedings, partly because of the processā™s heavy demands on their
time and expense, but more often in order to retain ļ¬‚exibility. Such an in-
vestor can sell at any time, while one involved with the bankruptcy court
becomes an insider and, in effect, has his investment locked up until the
courtā™s ļ¬nal resolution.
The proliferation of LBOs and high-yield bonds in recent years
promises to provide a continuing supply of distressed securities, as a cer-
tain portion of these investments will predictably get into trouble. Well-
managed distressed security funds can provide good returns and useful
diversiļ¬cation to our portfolio.
For the patient investor, timber offers outstanding diversiļ¬cation beneļ¬ts
and the prospect of moderately high long-term returns. Timber is one of
the better inļ¬‚ation hedges, and its returns have had a very low correlation
with those of stocks and bonds. Forecasts of net long-term real rates of re-
turn (net of inļ¬‚ation) from timberland range upwards from 6%.
Patience is necessary because of the cyclicality of timber values and be-
cause active management of timberland takes years to pay off. But depend-
ing on the price of timberland, the case for timber is fairly persuasive:
Despite the use of wood substitutes and the impact of electronic com-
munications on the printed page, the demand for timber should con-
tinue to increase in the years ahead, especially as living standards rise
in the developing countries of the world.
Increasingly, timber will have to come from timber farms, because nat-
ural forests have been cut so heavily, and remaining natural forests are
gaining more and more environmental protection.
The percentage of the worldā™s timber that today is provided from tim-
ber farms is very small. As demand continues to rise, supply is likely to
be constrained. Creating a new timber farm that is ready to harvest
takes at least 20 years.
Weā™re not likely to be surprised by a sudden increase in the supply of
timber, as the supply for the next 15 to 20 years is pretty well known.
Itā™s already in the ground and growing.
Timber is a commodity that does not have to be harvested at any one
time. Each year, a tree continues to grow, and it becomes more valu-
Private Asset Classes
able per cubic foot until it reaches some 30 years of age (depending on
the kind of tree). Such in-growth amounts to 6% to 8% per year.
Some of the best places for growing timber are in the southern hemi-
sphere, such as New Zealand, Australia, Chile, Brazil, and South
Africa. Many trees will grow twice as fast there as in the United States,
where in turn, trees grow faster than in Canada or northern Europe.
Over the past century, the real price of timber (net of inļ¬‚ation) has
ļ¬‚uctuated a great deal, but overall it has risen by some 1% per year.
Investments in timberland programs are long-term investments, but
unless liquidity is particularly important to our fund (it normally shouldnā™t
be), a few percent of our portfolio in timberland seems to make a lot of
senseā”especially if spread over a range of investment years.
Oil and Gas Properties
Oil and gas properties are a volatile but diversifying investment for a
taxfree fund. They offer better inļ¬‚ation protection than most asset classes,
and their returns have had a meaningfully negative correlation with returns
on stocks and bonds. They also provide strong cash ļ¬‚ow.
Returns on oil and gas properties are highly dependent on energy prices,
which have deep cycles that can last for decades. They are also impacted, but
to a lesser extent, by the accuracy of estimates of a wellā™s reserves.
Because of the uncertainty of energy prices, oil and gas properties are
usually priced to provide double digit real returns, assuming the real (inļ¬‚a-
tion-adjusted) price of oil and gas doesnā™t change. During the 1980s and
the ļ¬rst half of he 1990s, double digit real returns (or even nominal re-
turns) were a pipe dream, as a result of weak energy prices. During the
1970s, however, oil and gas properties were the place to be.
Participations can be bought in producing wells, development drilling,
and exploratory drilling. Other oil and gas investments encompass the
gamut of oil service companiesā”from drilling supplies to pipelines to gas
storage salt minesā”and these can be effective and diversifying investments
in a private energy portfolio.
The costs of investing in oil and gas, including hidden costs, can be
high. And we must invest in such a way as not to be an oil and gas opera-
tor, or else our taxfree fund will be subject to Unrelated Business Income
Tax. Every time I have ventured into the oil patch, I have felt a bit like the
city slicker waiting to be ļ¬‚eeced. In short, I want an extremely competent
102 ALTERNATIVE ASSET CLASSES
and reliable manager who really knows his way around the oil patch. If we
have such a manager, however, we should ļ¬nd it worthwhile long-term to
invest several percent of our total assets in oil and gas properties.
Alternative investmentsā”anything other than marketable stocks, bonds,
and cashā”can add valuable diversiļ¬cation to our portfolio. More and
more investment funds are allocating a growing portion of their assets to
Because these asset classes are especially dependent on the skill of the
investment manager, we need to invest with only the best managers.
This chapter is included as a reference, so we can recognize these asset
classes if our adviser should raise some of them for discussion.
Selecting and Monitoring
nce we have developed our fundā™s objectives and its Policy Asset Alloca-
O tion, we must decide who will manage the investments in each asset class.
What should be our overriding goal? We should strive to obtain the
best possible managers in each individual asset classā”the managers who
are most likely to produce the best future performance.
Such perfection is obviously unattainable. No one can realistically evalu-
ate all managers in the world. And simply chasing managers with the best
track record is a losing game, because all managers have hot and cold streaks.
Also, some of the best managers wonā™t accept our money. Finally, no one can
come even close to being a perfect judge of the future performance of an in-
vestment manager. But the best possible managers should still be our goal.
That goal implies:
No constraints or preferences as to geography or kind of manager
(small, large, here, there, bank, independent ļ¬rm, etc.).
A commitment to objectivity. That does not mean relying only on num-
bers and ascertainable facts. Ultimately, these decisions come down to
judgments. But we should make decisions as dispassionately as possible.
THREE BASIC APPROACHES
There are three basic ways for an investment fund to go about investing:
1. Index funds
2. In-house (do-it-ourselves)
3. Outside managers
104 SELECTING AND MONITORING INVESTMENT MANAGERS
For stocks or bonds, an index fund should not only be our benchmark.
It should also be our investment vehicle of choice unless we can ļ¬nd a
manager in that asset class who we are conļ¬dent will do betterā”net of
all fees and expenses. The case for index funds is persuasively articulated
by Jack Bogle in Common Sense on Mutual Funds (John Wiley & Sons,
Letā™s consider the Wilshire 5000 index initially. This is a capitalization-
weighted index of all stocks traded in the United States. It is a truism that
the average active investor has to underperform the Wilshire 5000 in his
U.S. investments. The investor has trading costs and investment manage-
ment fees that in combination can equal 0.5% to 1.5% per year, whereas
we can invest in an index fund that closely matches the Wilshire 5000 in-
dex with minimum cost.1 So the odds are against active investing.
The most widely used index fund is one that replicates Standard &
Poorā™s 500 index, which is very heavily weighted toward the largest U.S.
stocks and has a growth-stock bias. These are widely researched stocks. It
is difļ¬cult for any investor to get an information advantage over other in-
vestors in large U.S. stocks. As a result, the pricing of large U.S. stocks is
often thought to be very efļ¬cient. That means that if a good active investor
stays within the S&P 500 stock universe, it is difļ¬cult for him to produce
net returns that outperform that index over the long term. If we choose an
active manager over an index fund for these stocks, we must be arrogant
about our ability to choose active managersā”and then we must prove our
right to be arrogant.
As we consider active managers, we should be aware of the fact that
most active managers not only should underperform the broad indexes
theoretically but also have underperformed:
āIn the 25 years ending with 1997, on a cumulative basis, over three-
quarters of professionally managed funds underperformed the S&P
500,ā according to Charlie Ellis of Greenwich Research Associates.2
In the year 2000, ādata from Morningstar show that of 5,253 domes-
tic, non-index, equity mutual funds, 769 have performance records of
The Vanguard 500 Index fund has an expense ratio of 0.18%, and with enough
size, the cost of an S&P 500 index fund can get as low as 0.01%.
Charles Ellis, Winning the Loserā™s Game: Timeless Strategies for Successful Invest-
ing (McGraw-Hill Professional Publishing, 1998).
Three Basic Approaches
10 years or more. Of these 769 funds, only 195 [25%] have generated
annualized returns greater than that of the Wilshire 5000 index over
the past 10 years, after accounting for the impact of fees and sales
loads,ā according to Mark Armbruster of WealthCFO.
Is an S&P 500 index fund therefore a no-brainer? Well, intuitively,
does it make a lot of sense to increase the weighting of a stock in our port-
folio as its price goes up, and vice versa (as an index fund implicitly does)?
That would make sense if the change in this yearā™s price is a good predictor
of next yearā™s price, but we know that isnā™t true. In fact, contrarian in-
vestors have long known that a pervasive general trend in the investment
world is reversion to the mean.3
The active investor has another advantage: He is able to invest outside
the index that is used for his benchmark. For instance, from a practical
standpoint, a large-stock investor may be able to invest in a 700-stock uni-
verseā”not just the 500 stocks included in the S&P 500. In fact, most ac-
tive investors periodically do go outside their benchmark universe. That is
undoubtedly why active investors as a group will outperform the S&P 500
for a period of years, and then underperform it for another period of years.
Weā™ve been talking mainly about large stocks. One can also invest in a
smaller stock index fund such as a Russell 2000 index fund in the U.S.
Would that make just as much sense?
Because smaller stocks are not as widely researched, it is possible for a
good investor who digs hard enough to gain an information advantage on
other investors and therefore outperform a small-stock index by a wider
margin than a good large-stock investor can outperform the S&P 500. But
if the opportunity is greater with small stocks, the reverse is true as well.
One can really get bagged with small stocks. In short, if we are careful in
āReversion to the meanā is the tendency for the price of an asset (or an asset class)
that has greatly outperformed or underperformed the average of other assets (or as-
set classes) to revert over time toward the average. That tendency, which in general
has been well documented, also makes some intuitive sense.
For example, if a company is earning a particularly high rate of return in a
line of business, its high earnings will attract competitors to that line of business.
The competitors will challenge the pricing ļ¬‚exibility of the company and limit its
subsequent returns. Conversely, a company that is performing poorly attracts
takeover bids from other managements who believe they can squeeze more value
out of the company.
106 SELECTING AND MONITORING INVESTMENT MANAGERS
selecting managers, I believe for most investment funds active management
of small stocks can make more sense than a Russell 2000 index fund.
Viable index funds are available for large stocks in all the developed
countries. Since there are so many countries with different dynamics, it
would intuitively seem that an active investor should be able to add a lot of
value through country allocation alone. But that hasnā™t proved easy to
doā”unless the manager was smart (or lucky) enough to underweight
Japanese stocks starting in 1990, and to fully weight them in the 1980s,
when at one point Japanese stocks accounted for more than 60% of the
Reliable index funds for investment-grade U.S. bonds are also avail-
able and compete very well with active bond managers. In selecting an ac-
tive ļ¬xed income manager, it is equally important to ask ourselvesā”do we
really have sound reason to believe that, net of fees and expenses, this man-
ager can meaningfully outperform an index fund? In short, index funds are
a very viable bond alternative. Yes, it is entirely possible to do better, but
how much better?
Returns in excess of an index fundā”always hard to achieveā”have
been more readily achievable in some asset classes than in others. In small
stocks, U.S. or non-U.S., and in emerging markets stocks, ļ¬rst quartile
managers were able to add alpha of more than 3% per year during the 10
years ending in mid-2000. Meanwhile, in large U.S. growth stocks ļ¬rst
quartile managers were able to add less than 1%/year and in high-grade
bonds less than 0.5% per year.4
Index funds can be managed inexpensively in-house, but fees are so low
for outside-managed index funds, it is hard to justify in-house manage-
ment. Hence, in our discussion of in-house management, we shall focus
on active management.
Many sponsors of large investment funds actively manage all or a por-
tion of their assets in-house. They avoid the high fees charged by active
managers by hiring a staff to buy and sell the assets themselves. If we are
large enough, is this the way to go?
Letā™s go back to our original goalā”to have the best possible man-
agers in each asset class. With respect to whatever asset class we are talk-
Source: BARRA RogersCasey.
Criteria for Hiring and Retaining Managers
ing about, can we objectively convince ourselves that we can put together
a management team that, net of all costs, can match or exceed the net re-
sults of the best managers we could hire outside? If so, then in-house is
the way to go.
In-house management, however, faces serious challenges. For example,
does our compensation schedule enable us to attract some of the best pos-
sible investment managers? And if we hire some smart young people and
are lucky enough to grow them into the best, can we keep them? The best
managers tend to be entrepreneurial people who want ownership in their
own ļ¬rm. Even if we insulate our investment management team from the
rest of our bureaucratic organization, can we realistically aspire to hire and
retain the best?
And if our in-house hires donā™t ultimately challenge the best we can
hire outside, we have the unpleasant task of putting them out on the street.
Thatā™s a lot tougher than terminating an outside manager.
In each individual asset class we should seek the best manager (or man-
agers) we can get, regardless of geographic location. With literally thou-
sands of managers to choose from, our adviser should recommend ones for
each asset class. What kind of questions should we ask about those recom-
mendations? Why does the adviser believe his candidate is the best we can
get in that asset class? And how does the candidate meet our criteria for
the hiring and retention of managers?
CRITERIA FOR HIRING AND RETAINING MANAGERS
Our criteria should be the same for both new and existing managers:
1. Character. Integrity and reliability. Can we give this manager our
2. Investment approach. Do the assumptions and principles underlying
the managerā™s approach make sense to us?
3. Expected return. The managerā™s historic return, net of fees, overlaid by
an evaluation of the predictive value of that historic return, as well as
other factors that may seem relevant in that instance and may have
108 SELECTING AND MONITORING INVESTMENT MANAGERS
4. Expected impact on the fundā™s overall volatility. Two facets:
a. Expected volatilityā”the historic volatility of the managerā™s invest-
ments overlaid by an evaluation of the predictive value of that his-
toric volatility, as well as a recognition of the historic volatility of
that managerā™s asset class in general.
b. Expected correlation of the managerā™s volatility with the rest of
5. Liquidity. How readily in the future can the account be converted to
cash, and how satisfactory is that in relation to the fundā™s projected
needs for cash?
6. Control. Can our organization, with the help of our adviser, ade-
quately monitor this investment manager and its investment program?
7. Legal. Have all legal concerns been dealt with satisfactorily?
A common mistake is to be mesmerized by great past performance,
without looking further. In a study of 613 U.S. common stock managers,
BARRA RogersCasey found little predictive value in their performance
during the ļ¬ve years 1991ā“1995 for the ensuing ļ¬ve years 1996ā“2000 (see
Table 6.1). Moreover, BARRA RogersCasey found similarly discouraging
predictive value in ļ¬ve-year returns for prior intervals.
It is crucial to focus on predictive value. What do we mean by āpredic-
tive value of historic returns?ā The whole selection process has to do with
predicting future performance. Past performance is irrelevant except as it
may have predictive value. Whatā™s done is done. The only thing we can im-
TABLE 6.1 Study of the Predictive Value of Past Performance
Subsequent 5-Year Results Subsequent 5-Year Results
for 1st Quartile Performers for 4th Quartile Performers
during the 5 years 1991ā“95 during the 5 years 1991ā“95
1st Quartile 4th Quartile 1st Quartile 4th Quartile
Large value 24% 32% 30% 16%
Large growth 37 22 20 34
Core large cap 38 31 20 33
Small value 23 23 45 27
Small growth 29 24 35 15