Rare Coins: A Distinct and Attractive Asset Class By Robert
A. Brown, Ph.D., CFA
Dear
Fellow Numismatist and Coin Person,
Every now and then an article
is published related to the subject
of rare coin investing that deserves
a closer look. This is one of those
articles.
I invite questions and dicsussions one
on one afterward.
Abstract
It has been suggested that the decade
ahead may offer relatively unattractive
asset class returns for the most traditional
investment categories such as large-cap
domestic equities. The resulting debate
has promoted interest in such less traditional
asset categories as venture capital,
hedge funds, commodities, timber, real
estate, energy, works of art, and collectibles
on the part of both institutional and
individual investors. It is this last
category of rare collectibles, specifically
coins, which I examine within this paper.
According
to the Financial Times, over the past
50 years art as an investment has shown
returns in line with the S&P 500
Index, according to the Mei/Moses Fine
Art Index, which tracks 6,300 works
sold at auction in New York. From 1954
to the end of 2003, art returned +12.6
percent, according to Mei/Moses (versus
+11.7 percent for the S&P 500).
Under this
broader rare art and collectibles universe,
similarly attractive performance can be
found in the rare-coin market, which,
excluding silver coins and bullion-based
gold, exceeds $40 billion. This paper
offers an unbiased assessment of the inherent,
fundamental, and relatively permanent
risk, return, and diversification properties
of the robust collectables sub-segment
called numismatics and is based on rare-coin
pricing data spanning a 62-year period.
The
data accumulated, analyzed, and presented
here suggests a highly attractive complementary
asset category that even with a small
allocation provides good diversification
for a well-diversified portfolio, and
whose returns look impressive when compared
against the fixed-income category and
as an inflation hedge. In addition,
capital and economic market conditions
suggest that rare coins could exhibit
above-normal returns compared with underperforming
recent years.
Allocation
of a portion of an individual's portfolio
to rare coins must be pursued with full
appreciation for the non-income generating
nature of this asset category and recognition
that no profit-generating corporate
entity stands behind it. Moreover, rare
coins exhibit a relative lack of liquidity
and potentially high transaction costs.
With
the failure of the S&P 500 to reach
lows that would have brought the majority
of fundamental valuation measures back
to their historic norms, many of our
industry's more serious thinkers, those
with real gravitas, have suggested that
we'll experience long-term investment
challenges ahead. Investment luminaries
such as Robert Arnott (editor, Financial
Analysts Journal, Peter Bernstein (consulting
editor, Journal of Portfolio Management),
Bill Gross (PIMCO), John Templeton (Franklin/Templeton),
Bill Bonner (The Daily Reckoning), Jim
Grant (Grant's Interest Rate Observer),
Jeremy Grantham (GMO), and Stephen Roach
(Morgan Stanley) have suggested over
the last two years that the decade ahead
may offer relatively unattractive, if
not painful, asset class returns for
the most traditional and vanilla investment
categories such as large-cap domestic
equities. The resulting industry-wide
debate has prompted a serious reexamination
of less traditional asset categories
such as venture capital (and its associated
neighbors and offspring within the private
investment arena), hedge funds, commodities,
timber, real estate, energy, and rare
art or collectables. It is this last
category of rare collectibles that I
examine within this paper—focusing
specifically on the subcategory of rare
coins or numismatics.
Stephen
Roach, among others, believes that domestic
stocks, bonds, and residential real
estate are simultaneously and significantly
over-valued as a result of excess liquidity
having been pumped into our economy
through actions by the U.S. Federal
Reserve and by the monetary, savings,
and investment behaviors of foreign
nations. If true, as financial planners
we have the responsibility to seek less
traditional asset categories that remain
predominantly unaffected by today's
extreme overabundance of liquidity.
Rare
coins is one such nontraditional asset
category. But why consider rare coins
as a viable component of your client's
total wealth management solution? As
will be demonstrated below, rare coins
have historically offered fairly consistent
and attractive levels of return, risk,
and diversification. Over the last 62
years, this asset category has delivered
an arithmetic mean annual return (before
inflation) of 10.5 percent with a relatively
low standard deviation of 12.3 percent.
But the real benefit derived from this
asset category lies in its diversification
power. Some of its best relative performance
is delivered during those periods when
long-term bonds perform particularly
poorly.
Moreover,
it is not uncommon for high net worth
clients to already be comfortable with
and have significant holdings in the
rare art or rare coin markets. Nevertheless,
investment in the asset category of
rare coins entails a series of distinct
challenges. These potential impediments
include transaction costs (bid/ask spread),
transaction time, specialized knowledge,
proven investment properties, market
depth, lack of dividend or interest
income, and recognition that no profit-generating
corporate entity underlies the asset
category.
Clearly,
rare coins are a highly distinct asset
category, offering a remarkably different
set of investment characteristics and
properties. But the impediments recognized
in the preceding paragraph must be overcome.
Much as with physical real estate (so-called
"bricks and mortar"), the
sale and purchase of rare coins entails
incurring a measurable bid/ask spread.
This obstacle is significant, but with
reduced trading, investors can satisfactorily
overcome turnover through the attractive
long-term fundamental returns accruing
to this asset category. Once again,
just like physical real estate, participants
in the rare coin market should allow
significant time to complete acquisitions
or sales. Nevertheless, the speed with
which coin prices evolve is sufficiently
slow and gradual to accommodate the
more extended time required for transactions.
Finally,
just as with physical real estate, specialized
knowledge is well developed and readily
available at a reasonable cost and is
delivered by a wide range of professional
dealers in the marketplace. The challenge
of verification is no longer an issue,
as all investment quality coins are
authenticated and encapsulated by two
primary service bureaus: Professional
Coin Grading Service and Numismatic
Guaranty Corporation of America. Valuation
is well handled by the ready availability
of recent transaction prices (providing
investors with accurate bid/ask spreads)
and through the advice of professional
dealers—much as is the case with
the private real estate market. Safekeeping
(storage and insurance), however, remains
an ongoing expense—once again,
not dissimilar from bricks-and-mortar
real estate.
The
last two potential impediments—proven
investment properties and market depth,
are the primary subjects of this article.
As described below, returns in both
an absolute standalone basis and, more
importantly, in a well-diversified portfolio
context, have been highly attractive
over the last 62 years. Similarly, the
market for rare coins is deep, well
developed, and improving.
Art
and Collectibles Universe
I begin my examination of rare coins
with a quick high-level review of the
far broader rare art and collectibles
universe. If the investment merits of
rare coins are real and substantive,
then we would expect to find equally
attractive performance within the more
broadly defined arena of art and collectibles.
As reported
in the January 27, 2004, London Financial
Times, the Artprice Global Index rose
1.5 percent in 2003. This index includes
the prices of 30,000 works of art sold
at 2,900 auction houses around the world
in 2003 and is considered to be a reasonable
measure of the state of the aggregate
global high-end collectables marketplace.
In 2004, the Artprice Global Index rose
24.4 percent. Another index, the Mei/Moses
Fine Art Index—All Art, showed
gains of 21.7 percent and 13 percent
in 2003 and 2004, respectively. Over
the last 50 years (1955–2004,
inclusive) this index appreciated 10.5
percent annually versus 10.9 percent
for the S&P 500, 6.6 percent for
10-year U.S. Treasury bonds, and 5.4
percent for short-term U.S. Treasury
bills. This index is noteworthy because
it tracks over 8,000 works sold at auction
and is therefore based on real-time
investor-realized pricing. Fernwood
Art Investments LLC, a New York-based
company, founded recently by former
Merrill Lynch & Co. Inc. veteran
Bruce Taub, began offering art-based
funds in early 2005 as an alternative
asset class. Funds and investment partnerships
are being formed to offer shared interests
in portfolios of paintings by old masters,
impressionists, and others. Fernwood
states "Art is like what real estate
was 30 years ago, it's ripe for investment."
Rare
collectibles as an investment and as
a personal pursuit have been with us
for centuries. But one of the oldest
and most well-developed segments of
the collectables asset category is that
of numismatics or rare coins. In 1912,
the U.S. Congress created the American
Numismatic Association (ANA) to promote
popular interest in the study of numismatics.1
Barry S. Stuppler, a member of the ANA's
nine-person board of governors, estimates
that the total rare coin market experienced
domestic sales approximating $2 billion
in 2003. Moreover, he estimates that
the total supply (market value) of domestic
rare coins is in the neighborhood of
$40 billion (and this figure does not
include bullion-based gold and silver
coins that would bring the total market
value up considerably).
The
depth, vitality, and consistency of
the numismatic marketplace are exemplified
by numerous anecdotes.
In
February 1989, Kidder Peabody launched
a $40 million rare coin investment
fund, later increased to $110 million,
named the American Rare Coin Fund.
A
year later, in February 1990, Merrill
Lynch underwrote a $50 million fund
called the NFA World Coin Fund Limited
Partnership. United Bank of Switzerland
(UBS) maintains a separate and independent
division, UBS Gold & Numismatics,
designed to provide an ultra high
level of professional advice by experienced
experts in the field of numismatics
to its private client base.
In
a July 2002 Sothebys/Stack's auction
a 1933 $20 double-eagle gold piece
sold for $7,590,020. In 1960, this
same piece traded for $25,000—a
14.6 percent annual rate of return
over 42 years.
In
an August 1999 Bowers & Merena
auction, an 1804 proof silver dollar
sold for $4,140,000. In 1960, this
same piece traded for $30,000—representing
a 13.5 percent annual rate of return
over 39 years.
In
an early 2004 Blanchard and Company
auction, a 1913 Liberty nickel sold
for $3 million. In 1960, this same
piece traded for $50,000—a 9.8
percent annual rate of return over
44 years.
Finally,
in a May 1999 Goldberg Coins auction,
a 1907 ultra high-relief $20 double-eagle
gold piece sold for $1,210,000. In
1960, this same piece traded for $20,000—an
11.1 percent annual rate of return
over 39 years.
It should
be emphasized that the returns stated
above are gross of appraisal, insurance,
storage, and other transactions costs
(similar to but higher than what one would
experience in the private real estate
industry).
In the
context of these examples and data,
this paper's objective is to offer an
unbiased assessment of the inherent,
fundamental, and relatively permanent
risk, return, and diversification properties
of the robust collectables sub-segment
called numismatics.
Methodology
Rare coins may be stratified along numerous
dimensions; however, one convenient
classification divides the universe
by metal type: copper, silver, gold,
and other precious metals. The behavior
of the precious metals (gold, platinum,
and palladium) spot and futures markets
has significantly affected the return
behavior of rare gold coins during periods
of extreme precious metals price volatility.
So as not to confuse and otherwise intermix
the behavior of precious metals and
numismatics, I have excluded gold and
other precious metals (platinum and
palladium) coins from this paper's analysis.
Moreover, for consistency and reliability
I have restricted the analysis to U.S.
coins and pricing data has been drawn
exclusively from domestic sources.
All
rare coin pricing data was provided
by the following two sources: Handbook
of United States Coins with Premium
List (the Blue Book) for the annual
copyrights 1942 through 1951, and A
Guide Book Of United States Coins (Red
Book) for 1950 through 2004.2
Coin-price
data provided from these publications
was used to build unique equal-weighted
(by dollar value invested) portfolios
of coins for each individual year. These
portfolios were then tracked to the
subsequent year and an annual return
was calculated. The composition of each
year's portfolio was uniquely tailored
in order to most evenly distribute holdings
across the entire available domestic
universe at that time. Selections were
evenly spread across all copper-based
and silver-based U.S. coins. For publication
copyrights 1942–1960, portfolios
consisted of 600 individual coins. For
copyrights 1960–present, portfolios
consisted of 650 unique rare coins.
The higher 650-coin count was used during
the later years with the objective of
better representing the breadth and
depth of the more modern rare coin market
relative to the somewhat narrower pre-1960
marketplace.
Minimum
coin condition was generally held to
the highest available as reported in
the then-current Whitman publications.
In the initial years, such as 1942,
"fine" was the minimum condition
included within the 600-coin portfolios.
The maximum condition (MS-65) appearing
across all portfolios was similarly
defined by the reporting limits of the
Whitman publications.
As an
example, a portfolio of 650 copper and
silver coins was created and based on
the pricing appearing in the 2004 edition
of A Guide Book of United States Coins.
Coins were selected with the objective
of evenly dispersing the holdings across
the entire available universe as reported
in the 2004 Red Book. Portfolio holdings
were equally weighted (by dollar amount
invested in each coin). The 2005 edition
was used to reprice the original portfolio
and determine an annual rate of return,
(+16.39 percent in this instance.)
A
Critical Question
A critical question concerns the exact
start and end dates for the annual 12-month
investment windows. For example, the
2004 edition has a 2003 copyright and
was first released to the public in
late July 2003. The start date for the
subsequent 12-month investment window
depends on when the coin price data
was collected for this publication.
This paper's analysis is based on a
specific estimate for each annual return's
inception date. The following methodology
was employed to make this estimate.
A series
of 62 annual returns were calculated
based on the Blue Books and Red Books
spanning 1942 through 2004. This series
was compared against annual returns
for 13 other unique and fully differentiated
asset classes (spanning both fixed-income
and equities). The annual rare coin
return series was next shifted one month
at a time, both forward and backward.
This generated a series of 31 alternate
comparisons (that is, no time shift,
1-month through 15 months shifted backward,
and 1-month through 15-months shifted
forward). An evaluation was completed
for each of these 31 return series.
Specifically, I calculated the correlation
coefficients between the rare coin series
and each of the 13 other asset categories.
A median correlation (across the 13
alternate asset categories) was identified
for each of these comparisons. I then
identified the single rare coin time
series (from among the 31 alternate
shifted series) that delivered the highest
median correlation coefficient. The
answer turned out to be the rare coin
series that had its series shifted backward
in time by one month. In other words,
I found that by assuming that pricing
data taken from the 2004 edition of
the Red Book with copyright 2003 was
actually obtained from the marketplace
on November 30, 2002, then the correlation
coefficients with 13 other common asset
categories were maximized. This result,
in effect, amounts to a shift backward
in time by one month (that is, from
December 31, 2002 back to November 30,
2002).
Underlying
this methodology are the following premises.
First, coin prices appearing in a 1990
edition with copyright dated 1989 will
have been collected either very early
in 1989 or sometime in late 1988. Second,
asset classes have a strong and fundamental
tendency to move in tandem with the
passage of significant time. Thus, they
generate positive correlation coefficients.
Third, the most realistic assessment
as to the time positioning of the annual
rare coin series can be determined by
positioning the annual window so as
to achieve a maximum median correlation
coefficient. The statistical result
identifying an initial start date of
November 30th of the year preceding
the annual copyright date is intuitively
appealing. It suggests that the data
reflected in the 1990 edition with copyright
1989 was current and immediate in the
rare coin marketplace as of November
30, 1988. The November 30, 1988, date
is also logically consistent with the
public availability of the final publication—that
is, on or about July 31, 1989. This
would allow approximately eight months
to collect the rare coin data, interpret
it, update the guidebooks, send them
for printing, and distribute them to
retailers for public sale.
Results
Table 1 provides this paper's estimates
for rare coins' annual returns spanning
the period November 30, 1941, through
November 30, 2003. Returns ranged from
a low of –7.46 percent (1981/1982)
to a high of +48.08 percent (1971/1972),
with a median annual return equal to
+6.26 percent. During the last 62 years,
rare coins delivered a positive return
84 percent of the time (52 of 62 years
as based on annual returns) and an annual
return greater than 10 percent 40 percent
of the time (25 of the years from 1941
through 2003, inclusive).
Table
2 provides the nominal (not inflation-adjusted)
summary statistics for this series.
Relative to 13 other popular asset categories,
rare coins appear to offer relatively
higher average annual returns with commensurate
to moderate levels of return volatility.
Over the last 62 years, rare coins offered
an average annual arithmetic mean return
of 10.46 percent and an average annual
geometric mean return of 9.83 percent,
with a standard deviation of 12.3 percent.
This compares quite favorably against
growth stocks over this same 62-year
period, which delivered average annual
arithmetic and geometric returns of
12.35 percent and 11.02 percent, but
with a significantly higher standard
deviation of 17.1 percent.
It
should be noted that Table 2 reports
on all total return series for which
monthly return data is available spanning
the entire time period indicated (November
30, 1941, through November 30, 2003).
Data for asset classes other than rare
coins were taken from the Ibbotson Associates
databases. Standard deviations were
calculated on annual returns with years
starting and ending on November 30 as
described in the data methodology in
preceding sections of this article.
It is believed that the standard deviations
are directly comparable across all of
the asset classes appearing in Table
2, including rare coins, because each
return series is based on actual transaction
data as opposed to valuation estimates
(as frequently occurs with private real
estate).
Table
3 provides the same statistics but in
real (fully inflation-adjusted) terms.
A relevant observation for rare coins
is highlighted through a comparison of
Tables 2 and 3. Observe how, after adjustment
for inflation, rare coins' return-per-unit-of-risk
declines from 0.80x to 0.46x. Yet, intermediate
U.S. Treasury bond's return-per-unit-of-risk
declines by a far greater amount, falling
from 0.99x to 0.24x. This difference is
a result of rare coins' propensity to
deliver a degree of protection during
inflationary environments. This observation
is supported by the clearly superior correlation
coefficient between coins and inflation
(0.12) relative to the correlation coefficient
between the S&P 500 and inflation
(–0.31).
Table
4 reports the correlation coefficients
between rare coins and each of the 13
primary asset categories (plus CPI inflation)
based on nominal returns (non-inflation
adjusted). Rare-coin correlations with
equities tend to be higher than with
fixed income due to fixed income's poor
performance during periods of rising
inflationary expectations. Note rare
coin's correlations of 0.12 and –0.36
with the S&P 500 Index and intermediate
U.S. Treasury bonds, respectively, offering
powerful diversification potentials.
More
Powerful Diversification Role
Using real returns (after adjustment
for consumer price inflation), we obtain
the correlation coefficients appearing
in Table 5. Comparing data appearing
in Tables 4 and 5, we note that correlations
increase significantly across all asset
categories moving from Table 4 to Table
5. But the correlations for rare coins
increase less than any of the other
asset categories (that is, rare coins'
average increase amounts to a lesser
+0.112 versus the other 13 asset categories
average +0.153 increase). This smaller
change results from coins' superior
behavior during inflationary environments
and allows numismatic investments to
potentially play a more powerful diversification
role within a portfolio.
The
practical benefits or attraction of
rare coins as an asset category are
best demonstrated by examining the cross-time
performance characteristics of portfolios
with and without their inclusion. The
bid/ask spread associated with buying
and/or selling numismatic coins tends
to be quite high. In a cross-time portfolio
context, this has an important practical
implication; specifically, regular periodic
rebalancing of one's allocations to
coins in a portfolio is impractical.
Table
6 provides a historical view of portfolio
performance with and without the inclusion
of rare coins. But this data assumes
an initial allocation to coins of 0
percent, 5 percent, 10 percent, 15 percent,
and 20 percent, with no rebalancing
during the 62 years covered by the analysis.
By avoiding periodic rebalancing, this
paper delivers a more realistic view
of how investors actually construct
and maintain exposure to this attractive
asset category.
Notice
how a portfolio with an initial 15 percent
allocation to rare coins delivers a
geometric mean return 1 basis point
lower than that same portfolio with
a 0 percent allocation to coins. But
the standard deviation falls from 10.7
percent to 8.7 percent after the inclusion
of rare coins, and the worst single
annual return improves from –20.3
percent (without rare coins) to –16.6
percent (after inclusion of rare coins)
But
no asset category should be evaluated
solely on the basis of simple summary
statistics spanning some long-term time
period. Asset categories have a propensity
to deliver performance attuned to the
current-period capital market and economic
environments in which they operate. During
the 18-year period ending November 30,
1999, for example, rare coins returned
an unattractive 0.0 percent a year.
The
causality underlying this painful time
period is not difficult to identify.
First, during the prior 18-year time
period (1963–1981) coins returned
+14.2 percent annually, generating a
level of over-excitement that probably
drove rare coin prices to a transitory
level from which prices needed to fall
back. Second, during the nine years
ending November 30, 1999, the S&P
500 returned +21.01 percent a year.
This abnormally high stock market return
spanning an unusually long time period
had the effect of pulling assets either
out of or away from investment in numismatic
coins. Third, inflation (as measured
by the CPI) was running at a rate of
+12.74 percent for the year ending November
1979. Just seven years later, this same
inflation rate had fallen to +1.29 percent
(year ending November 1986). This remarkable
fall in the rate of inflation removed
a key motivational factor that had previously
been driving coin prices during the
18-year period ending November 30, 1999.
Improving
Coin Environment
Today it would appear that the environment
is changing. Capital market and economic
dynamics are potentially evolving in directions
that will result in below-average stock
market returns while at the same time
delivering a period of rising inflation.
If we experience such a capital market
environment, rare coin investments may
deliver above-average returns.
Graph
1 provides a historical cross-time view
of the path followed by inflation-adjusted
rare coin prices since 1941 plotted
on exponential scale (natural log scale).
By using an exponential scale, equal
vertical distances correspond to identical
percentage rates of return independent
of the current absolute level of coin
prices. The solid line identifies the
growth of one dollar invested in numismatic
coins over the 62-year time period.
This line shows an average annual inflation-adjusted
geometric mean return of 5.5 percent.
The dashed line shows the constant rate
of return path determined by applying
a best-fit analysis. This second, perhaps
more realistic, characterization of
history shows an average annual geometric
inflation-adjusted return of 5.6 percent
for numismatic investments.
Since 1941,
rare coin returns have followed a reasonably
constant return path—exhibiting
occasional periods of out- and under-performance.
But as of November 30, 2003, rare coin
prices stood 131 percent below their trend
line path. This level of below-path pricing
is most likely a function of the recent
below-normal return experience being driven
by low inflation in combination with considerably
above-normal stock market returns. Of
special note, coin returns began a marked
turnaround starting November 30 2001 having
risen a total of +25.2 percent (unannualized)
since that date.
A
different method for exploring the environment-dependent
nature of rare coin returns is demonstrated
by comparing fixed-income portfolios constructed
with and without numismatic investments.
Table 7 shows the 62-year summary statistics
for these alternate fixed-income portfolios
in real terms (that is, after adjustment
for CPI inflation.) By making these comparisons
after inflation adjustment, we are able
to emphasize how relatively small allocations
to rare coins can deliver radical improvements
to a fixed-income portfolio's inflation-fighting
performance.
Table
7 presents a series of six portfolios
with initial allocations to numismatic
investments of 0 percent, 5 percent,
10 percent, 15 percent, 20 percent,
and 25 percent. As before, the allocations
between rare coins and the other asset
categories are not rebalanced with the
passage of time. The fixed-income portion
of these six portfolios consists of
a mixture of long- and intermediate-term
U.S. Treasury bonds which are rebalanced
once each year to maintain a constant-percentage
allocation.
The fixed-income
portfolio without numismatic investments
experiences a standard deviation of 7.3
percent. The same portfolio with an initial
10 percent allocation to rare coins delivers
a lower standard deviation of 7.2 percent
and a higher geometric mean return of
2.67 percent (versus 1.54 percent without
coins). Similarly, note how the bond portfolio
without rare coins suffers a worst single
year of –14.4 percent. In contrast,
the bond portfolio with an initial 25
percent allocation to numismatics experiences
a far less severe worst single year performance
of –10.2 percent. These data suggest
that investors seeking the most conservative
portfolios (that is, bond portfolios)
may be better served by incorporating
small allocations to numismatic investments
as a method to significantly enhance the
inflation-protection characteristics of
their otherwise attractive fixed-income
securities.
Investability
Three issues generally define the question
of just how viable rare coins or numismatics
are as an investable asset category for
small portfolio allocations. These three
factors are the bid/ask spread, liquidity,
and asset class-specific expertise. Bid/ask
spreads in the rare coin arena are consistent
with those found in other sub-segments
of the rare art/collectables marketplace.
At the high end, bid/ask spreads can approach
40 percent. But the rapid growth of both
the electronic and physical auction market
and the robust development of authentication
and grading of rare coins are serving
to reduce these spreads across the numismatic
spectrum. The best-developed, commonly
traded, and standardized (through third-party
authentication) coins may trade at bid/ask
spreads approaching 10 percent.
Liquidity
concerns the time it takes to execute
a transaction, whether a buy or a sell.
It also affects the size of the transaction
and indirectly the bid/ask spread that
is paid. In the field of numismatics,
the transaction time period is properly
measured in months as opposed to minutes
or hours, such as one would find in
trading large-capitalization shares
on the New York Stock Exchange.
In general,
the rare-art marketplace requires significant
knowledge, experience, current information,
and close contacts with key participants
to be fruitfully exploited from an investment
standpoint. The field of numismatics
is no different. But one of the reasons
that a bid/ask spread exists is to compensate
the rare-coin dealer for the advice
and consultation he or she provides.
Yes, part of the bid/ask spread is compensation
for performing the function of, in effect,
marketmaker. But in today's numismatic
market, there exists a wide cross section
of highly professional coin dealers
who view their buyers as long-lived,
permanent clients for whom they deliver
expert advice and consultation on how
best to assemble a portfolio of numismatic
items designed to achieve a long-range
objective. Such dealers have intimate
knowledge of markets and the longevity
of participation to properly assess
the current environment. So as with
many other investment asset categories
such as venture capital, buyout, hedge
funds, and real estate, one pays for
the required (and desired) expertise
via the direct or indirect bid/ask spreads
reflected in the investment vehicle.
Rare coins are no different.
Conclusions
As ANA's governor Barry S. Stuppler has
suggested, the marketplace for numismatics
is both broad and deep, having now reached
an aggregate market value in the vicinity
of $40 billion (and this figure does not
include bullion-based gold and silver
coins that would bring the total market
value up considerably). It is an active
market where single isolated auctions
can experience transaction volume over
$30 million. Professionalism and initial
efforts towards institutionalization have
been demonstrated by numismatic investment
funds launched by Kidder Peabody and Merrill
Lynch and by United Bank of Switzerland's
UBS Gold & Numismatics professional
consulting division.
Pricing
data drawn from the last 62 years of
history (1941–2003) suggest an
attractive and potentially highly complementary
asset category—particularly relative
to fixed-income securities and as an
offset to inflation. Over this period,
rare coins, the S&P 500, and long-term
U.S. Treasury bonds have delivered geometric
mean returns of 5.52 percent, 7.86 percent,
and 1.38 percent, respectively, after
adjustment for inflation. But after
reflecting their differential standard
deviations, we find that rare coins,
the S&P 500, and long-term U.S.
Treasury bonds have delivered return-per-unit-of-risk
of 0.46x, 0.46x, and 0.18x, respectively
after inflation adjustment. Rare coins'
greatest benefit may be one of diversification.
The S&P 500 typically correlates
with other common asset categories at
a level of between 0.28 and 0.97 (inflation
adjusted). In contrast, numismatic investments
have delivered correlations across similar
asset categories ranging from a far
lower –0.21 to 0.29.
Robert
A. Brown, Ph.D., CFA serves as Chairman-Investment
Management Executive Committee and Chief
Investment Officer for GE Private Asset
Management, Inc., located in Encino, California.
Endnotes 1. The American Numismatic
Association (ANA) maintains the distinction
of being one of the very few organizations
in the United States to operate under
a charter. The purpose of the organization,
as stated in its federal charter, is
to advance and promote the study of
coins, paper money, tokens, medals,
and related numismatic items as a means
of recording world history, art, economic
development and social changes, and
to promote greater popular interest
in the field of numismatics. On April
10, 1962, the Congress passed a second
act making the ANA a permanent and perpetual
nonprofit, educational organization.
Today, the ANA maintains a library with
more than 50,000 reference materials
available for loan to members free of
charge.
Handbook of United States Coins with
Premium List was written by R.S. Yeoman,
Lee F. Hewitt, and Charles E. Green.
A Guide Book of United States Coins
was written by R.S. Yeoman and edited
by Kenneth Bressett. Whitman Publishing,
LLC of Atlanta, Georgia, produces both
of these publications. The Handbook
of United States Coins focuses more
on wholesale prices. A Guide Book of
United States Coins focuses primarily
on retail pricing.