Which Coins Are the Best Investment?

A potential new client recently asked me a basic but interesting question: which gold coins are the best investments? As those of you who know me realize, I don’t really tout coins as an “investment.” But I want my customers to make money on the coins that they purchase from me and I try to steer them towards pieces that I think will appreciate in value over the course of time. In my opinion, there are a few factors that make specific coins a good investment and which should perform well. Some of these factors are as follows:

1. Liquidity: Does a specific type of coin sell very quickly when I list it on my website or does it tend to languish for an interminable period of time? I notice that certain coins are consistently good sellers. Generally speaking, they fall in the “sweet spot” pricing range of $2,500 to $7,500. They are usually coins with an interesting history and pieces with good aesthetic appeal. As a rule of thumb, the more expensive a coin is the less liquid it becomes (although, in the last three to five years, very expensive coins and ultra-rarities have been remarkably liquid). As I told the gentleman who inquired about coins as investments, the pieces that are the most liquid are the best to own.

The “quality of liquidity” is important as well. Will the person most likely to buy your coin(s) be a specialized collector or a dealer? I personally like coins that I can sell to end-users as opposed to dealers who will look at them primarily as commodities and be more conscious of price than quality.

2. Popularity: Popularity and liquidity are not the same thing. A coin can be liquid but be a part of a not especially popular series (an 1838-C half eagle is an example of coin that is very popular but it is from a series—Charlotte half eagles—that I would not describe as being immensely popular) while a popular coin can be relatively illiquid (an example of this would be an unappealing, lower grade High Relief which most collectors would probably pass on and spend a bit more money to acquire a nice piece). An excellent collection could be created out of nothing but very popular coins—pieces like 1861-D gold dollars, 1839-O quarter eagles, 1838-C and 1839-D half eagles, 1838-D and 1839-D half eagles, 1838 eagles, 1861-O double eagles, etc. I refer to issues like this as the “Krugerrands of Rare Date Gold” as they are coins that are almost like cash.

In the same vein, I am an advocate of “absolute rarity” as opposed to “condition rarity.” A coin like an 1841-O eagle is rare in all grades and I will buy any example I can find, unless it has been harshly cleaned or damaged. An 1843-O eagle is not a rare coin in lower grades and I will generally not purchase a piece unless it grades at least AU55. Give the choice of owning a nice EF40 1841-O eagle or an AU55 1843-O eagle, I would personally rather have the former.

3. Rarity: It would seem obvious that the rarer the coin, the better the investment it is. This is actually not always the case. If a coin is very rare but it is part of a series that is not popular and/or readily liquid, then it may not necessarily be a good investment. An 1846 eagle is a genuinely rare coin that is nearly impossible to find in any grade higher than EF45 to AU50. If I were offered a nice, original AU55 I would certainly purchase it. But this is a coin that I would not expect to sell quickly and it might actually take me a number of months to move it. The problem with this coin is that it is a member of a series (Liberty Head eagles) that does not have many specialists and it is a Philadelphia issue.

The perfect “investment quality” coin is one that is not only rare but which is popular and liquid. A coin that scores highly on all three of these fronts is one that should perform well.

4. Historic Price Performance: With the advances in price dissemination available to collectors, it is easier than ever to track how coins have performed over the past three to five years. We have been in the midst of what is ostensibly the greatest sustained bull market in numismatic history and if a specific coin hasn’t done well since the early 2000’s, than the chances are good it isn’t going to do very well when this market finally cools off.

By the same token, an investor wants to avoid a coin that is currently at an all-time high in price. If you look at the price levels for a coin like a 1911-D quarter eagle in Uncirculated, you can see that it was selling for considerably more money by the beginning of 2006 than it had at any time since the halcyon days of the late 1980’s. It didn’t take a rocket scientist to figure that the 1911-D had probably risen to the point where it was no longer a good value. And this is exactly what has happened, as the value of this issue has dropped 10-20% in the past few months.

If you are a bottom-line oriented coin buyer, avoid issues which appear to be at a market peak. Conversely, being a total contrarian might not be a great idea either. The perfect coins to buy are those that have shown some price increases in recent years but whose price levels still make sense, considering their rarity and grade.

Rarity and Mintage Figures

Many collectors believe that a coin is automatically rare because it has a low mintage figure or that it is common because it has a comparatively high original mintage. This is most definitely not the case, especially in the field of American gold coinage. There are a number of factors that make a coin potentially rarer (or less rare) than its original mintage figure. Some of these are as follows:

1. Acts of Legislation:

The 1933 eagle is a very rare coin as a direct result of the passage of the Act of April 1933 which ended the production of gold coinage in the United States. Banks and individuals were required to turn in their gold coinage and many of these pieces were melted. Just a small number of 1933 eagles were rescued from the melting pot. The same holds true for certain late date St. Gaudens double eagles such as the 1929, 1930, 1930-S, 1931, 1931-D and 1932. All of these coins are much rarer than their original mintage figures would suggest due to heavy meltings that were a direct result of an act of legislation.

2. Contemporary Economics:

In times when the economy was strong, people collected and saved coins. For instance, high grade gold dollars from the 1880’s are more common than their low mintage figures suggest because dealers, collectors and hoarders, during a prosperous era, saved large quantities of these coins. The same is also true, although not to as great an extent, for Three Dollar gold pieces from this era.

3. Exportation:

Once the United States began to actively trade with Europe, Latin America and South America, large numbers of American gold coins were sent overseas in order to pay off foreign trade debts. This seems to have begun in earnest around 1878. Today, quantities of United States gold coins are still being found in foreign banks and other sources. This means that some United States gold coins from the 1880’s and 1890’s are more available in the lower Uncirculated grades than their mintage figures might suggest. It also makes for an unusual grade distribution for certain issues. As an example, the New Orleans eagles from the 1890’s are rarely seen in grades below AU55 but almost never above MS62. This is indicative of issues that were probably sent overseas soon after they were struck but which were very roughly handled and are now heavily abraded.

4. Incorrect Mint Records:

In the 18th and early 19th century, the Mint often kept somewhat sketchy records. There are certain early gold issues which are more common than their mintage figures suggest while others are far rarer. In some cases this has to do with the fact that the original mintage figures are incorrect.

Sometimes, even more modern records appear to be incorrect. In 1910 the mintage figure for Proof Indian Head quarter eagles was recorded as 682. This is nearly triple the highest recorded figure for any other date of this design and from the number of pieces known to exist, it seems likely that this figure is incorrect. The same holds true with the other Proofs of this year and makes us wonder, are the records simply wrong or were far too many Proofs ordered to be struck and were most of these later melted?

5. Hoards:

The rarity of certain United States gold coins has been greatly influenced in recent years by the discovery of hoards. A classic example of this is the S.S. Central America shipwreck which made the 1856-S and 1857-S double eagles very common in the higher Uncirculated grades. More recently, the S.S. Republic shipwreck contained a number of New Orleans eagles from the 1840’s and 1850’s as well as large numbers of Philadelphia double eagles from the early to mid-1860’s. Other smaller hoards have been found as well. As an example, a little-known group of 1840 quarter eagles was found in the mid-1990’s. It contained less than ten pieces but included at least four or five coins that graded in the MS63 to MS65 range. This was a very important group for collectors of Liberty Head quarter eagles as this date had been essentially unknown in Uncirculated before this.

6. Mint Meltings:

This category is especially applicable to Proofs. These coins have always been made specifically for collectors. In the 19th and early 20th century, the Mint would simply melt those Proofs that were not purchased. As an example, in 1859, the Mint struck 80 gold Proof sets. This was far more than needed and it is likely that the vast majority were melted. Today, all 1859 Proof gold coinage is very rare and the larger denomination issues, specifically the eagle and the double eagle, have as few as four or five survivors.

7. Restrikes:

There are certain instances where it is clear that a coin is rarer than its original mintage figure. Two excellent examples of this are the Proof-only 1875 and 1876 Three Dollar gold pieces. These have original mintage figures of 20 and 45, respectively. The PCGS and NGC populations for each of these issues are actually greater than the original mintage figures. Some of these have to be discounted as multiple submissions but clearly the original mintage figures for these two issues have to be called into question. It is my feeling that both of these were restruck (probably in the same year in which they were issued) by the Mint to fill a demand by contemporary collectors and dealers. Before you determine how rare a coin is, there are many factors to consider. The seven mentioned above are important and I would suggest that there are even more that await the curious collector.

"Fundamental" vs. "Grade" Rarity

When discussing “rarity” it is important to realize that there are two different types of rarity. These are “fundamental rarity” and “grade rarity.” Fundamental rarity refers to the overall rarity of an issue. A coin like a 1796 No Stars quarter eagle is fundamentally rare in any grade. The value of such a coin is due primarily to its rarity and not its grade. There is a high level of demand for this issue, even in very low grades. If you have a 1796 No Stars quarter eagle that looks like it was run over by a train, someone is still going to buy it.

A coin that is a grade rarity is one that is considered rare solely because of its high grade. A classic example of this is a 1920 Saint Gaudens double eagle. This is a common issue in the lower Uncirculated grades and it is only moderately scarce and desirable in MS63 to MS64. But it is extremely rare in MS65 and there is a dramatic jump in price between MS64 and MS65.

From my personal viewpoint, I like coins that are fundamental rarities. As someone who has a significant amount of money invested in coins (i.e., my inventory) I feel a lot better buying a coin that is considered desirable by virtue of its rarity in any grade than one that is desirable ONLY because of its high grade.

Some coins are “hybrids” which can be considered both fundamental rarities and condition rarities. As an example, a Proof Barber quarter is a relatively scarce coin with most issues having mintages figures below 1,000. Even an Impaired Proof has a degree of collector value. But at what point in the grading continuum does a Proof Barber Quarter morph from a fundamental rarity to a condition rarity? I’m not certain that I know the answer but I do know that once Proofs of this design are graded PR67 or higher, they become less interesting to me.

Another factor to consider is this. What if third-party grading ended tomorrow and slabs no longer had any relevance? In this scenario, a coin like an 1854-D Three Dollar gold piece would still be considered a desirable numismatic item with many potential buyers. But would a PR68 Barber quarter be considered four or five times “better” than a pretty, lightly hairlined piece that had previously been in a PR64 slab? I’m not sure I know the answer but…

These are interesting questions that the collector needs to ponder as he decides what his next purchase will be.

How Rare Are Rare Date Coins?

In the past few years, there seems to have been an explosion in the availability of rare date United States gold coins. Many issues that were considered very rare a few years ago now seem almost commonplace. Is this actually the case--or is this perception totally incorrect? The correct answer lies somewhere in the middle. A number of interrelated events have occurred during the past few years which have significantly altered the traditional relationship between supply and demand that has dictated the rare date gold market for the better part of three decades. All of these events are important and each needs to be analyzed individually to make any "big picture" conclusions in regards to this market.

I. Rising Prices Have Brought Many Coins Onto The Market

Ten to fifteen years ago, the prices for many better date United States gold coins were extremely low. It was possible to purchase outstanding pieces in the $500-$2000 range. Today, many of these same coins are worth five to ten times (or even more) their original purchase price. Many collectors have decided to take their profits. After all, it's hard to pay $7,000 for another AU-50 1840-C half eagle when the AU-50 you bought fifteen years ago was priced at $750.

II. Grading Standards Have Changed Considerably In The Past Few Years

Population explosions seem to be greater for high grade gold coins than for low and middle grade pieces. There is a good reason for this. The grading standards of the late 1980's and the early 1990's have loosened considerably. The very same coin that PCGS or NGC graded Extremely Fine-45 in 1989 is almost certainly an About Uncirculated-53 to About Uncirculated-55 today. And many of 1989's AU-55 and AU-58 coins are today's Mint State-60, Mint State-61, Mint State-62 or even Mint State-63 pieces. The result of this "gradeflation" is that many dates that once appeared to be very rare in high grade now seem much more readily available. The truth of the matter, however, is that many of these Uncirculated coins are the exact same pieces that were considered About Uncirculated a decade ago.

III. A Number Of Exceptional Hoards Have Come Onto The Market

The last ten years have seen some amazing groups of gold coins enter the market. Some, like the Brother Jonathan hoard of 1861-1865 San Francisco double eagles and the S.S. Central America hoard of 1857-S double eagles and territorial issues, are well known. Others, like the group of Uncirculated 1856-D half eagles discovered in South Carolina a few years back or numerous groups of key date eagles and double eagles found in Europe or other overseas locations, are not as well known. All of these have made certain formerly rare issues much more available in high grades than in the past.

IV. The Market Is Now Dominated By Investors

Until recently, issues such as Charlotte and Dahlonega gold were strictly collector-oriented. Investors were led to believe that these coins were "bad investments" because they were not available in Mint State-65 or because they were "too esoteric." After years of bad performance by investor series such as silver commemoratives or modern type, the same investment gurus who had badmouthed Charlotte and Dahlonega gold coins were now jumping on the bandwagon. The result was a major disturbance in the supply/demand ratio for such coins.

The most immediate result was a quick run-up of prices. Suddenly, a lot more people wanted high quality Charlotte and Dahlonega coins. Many of the Condition Census quality pieces that had previously been owned by serious collectors were now in the hands of investors. This became a negative factor for the short and medium range health of the Charlotte and Dahlonega market.

Collectors are great for markets because they tend to hold their coins for a long period of time. In order for Charlotte and Dahlonega prices to have risen in an orderly fashion, as they did through the 1970's and the 1980's, the coins had to appear to be rare (i.e., the demand had to outstrip the supply). One of the reasons why it has become so hard to find nice medium grade (i.e., Very Fine-35 to Extremely Fine-45) Charlotte and Dahlonega coins is that these are "collector grades" and the people who own these coins hold them for ten to fifteen years.

Investors are bad coin owners because they tend to sell very quickly. They will buy a high grade piece and sell it six months to a year later. This creates a situation where supply suddenly outstrips demand. Even more hazardous is the fact that the perception of rarity for these coins becomes altered. Coins that were once believed to be rare were now trading on a regular basis. While it might have been the same two examples of a specific date that traded over and over, the casual observer was now led to believe that a supposedly rare issue was more available than he had been led to believe.

The fallout from this has been very interesting. As this article is being written, I find that the demand for affordable, lower grade (i.e., Very Fine and Extremely Fine) Charlotte and Dahlonega gold coins, as well as most other branch mint issues, is extremely high. But coins like this are very hard to find as they tend to be in the hands of serious, long-term collectors who do not wish to sell them. I now find myself in the curious position that when someone calls me looking for a $1,500 version of a certain Charlotte or Dahlonega issue, I have to tell him that I do not have this coin but I do have a $10,000 version. The market now seems to be saturated with hard-to-sell expensive coins while the demand is for inexpensive coins.

V. Too Much Information Leads To Disinformation

In the 1970's and 1980's there was very little easily accessible information about United States gold coins. Today, there is an incredible amount of information. For the sophisticated collector or dealer, this data is a godsend. For the neophyte, this information can be confusing or even misleading.

As an example, let's look at the PCGS and NGC population reports. These monthly publications, which list the number of coins graded by each service, have become widely-recognized sources of information. According to the most recent reports (May 2000 for PCGS and April 2000 for NGC), there have been 21 1859-D gold dollars graded in Mint State by PCGS and another 16 have been graded by NGC for a total of 37 coins.

In my 1997 book "Gold Coins of the Dahlonega Mint, 1838-1861" I estimated that 10-12 Uncirculated examples of this date were known. Was my estimate that wrong or is there something misleading about the PCGS and NGC data?

What many people don't realize about the population reports is that the numbers are often greatly swollen by resubmissions. The "21" Mint State 1859-D gold dollars recorded by PCGS might actually be no more than ten or so separate coins. When you also realize the fact that many of the coins graded Mint State-60 or Mint State-61 by this service could just have easily been called About Uncirculated-58, then my original estimate of 10-12 truly Mint State 1859-D gold dollars seems a lot more accurate.

Another type of information that can easily be misinterpreted by neophytes is auction records. Many collectors make purchases based on the fact the a certain coin has or hasn't appeared a certain numbers of times at auction over the course of a few years.

I keep a meticulous database of Condition Census, rarity levels, and frequency of appearance for branch mint gold coins. I can cite numerous rare coins that have appeared in as many as five or six auctions over the course of one year. This is compounded by the fact that this exact coin may have appeared first as a PCGS AU-58, then as an NGC MS-61, then as a PCGS MS-61 and finally as a NGC MS-62. Same coin, four different holders. If you do not carefully track these appearances and study the plates in the auction catalogs (or view the coins in person), you can make some incorrect assumptions about the rarity of rare date U.S. gold coins.

VI. A New Type of Dealer Starts Buying Dated Gold

Until recently, the number of dealers who were serious buyers of rare date U.S. gold coins was fairly small. They tended to be knowledgeable purists who bought coins for collectors. They were stable in their business endeavors and didn't look at every coin as an upgrade or "breakout" opportunity.

This all began to change as price levels increased and value spreads widened. Suddenly, a date might be worth $1,500 in Extremely Fine-45 and $7,500 in About Uncirculated-50. The possibility of upgrading a coin from this proved to be extremely tempting; especially after grading standards became more loose.

The rare date gold market was then invaded by "crackout" dealers who would buy a wide range of coins in a wide range of grades. This type of dealer thinks nothing about submitting a coin five, ten or even fifteen times before he gets it into the "right" holder. If this dealer does not keep very careful track of the PCGS or NGC inserts from each of his submissions, the possibility of some--or all--of them becoming lost is great. I can cite numerous examples where the PCGS or NGC population for a truly rare coin is now totally out-of-kilter because one dealer submitted a coin ten times then either lost the inserts or is holding onto them for some reason.

VII. A Case Study: The 1840-C Half Eagle

To prove--or disprove--my thesis, I thought it would be interesting to look at a specific rare date gold issue and analyze its rarity. A good example is the 1840-C half eagle. This was a coin that I once believed was very rare in all grades and I would become excited whenever I had the opportunity to purchase an example. Today, I still think this is a scarce issue but I no longer seem to care about 1840-C half eagles when they are offered to me.

When my book "Charlotte Mint Gold Coins: 1838-1861" was published by Bowers and Merena in 1987, I estimated that 55-60 1840-C half eagles were known with just four or five of these in AU and none in Mint State. In 1999, I published a revised edition of this book and suggested that there were eight to ten known in AU and another one or two in Mint State. The May 2000 PCGS Population Report shows sixteen 1840-C half eagles having been graded in all AU grades and one coin in Mint State while the May 2000 NGC Census Report shows another sixteen in AU and three in Mint State. Has the 1840-C half eagle really become this much more available in the last decade and a half?

The answer is multi-faceted and somewhat complicated. I was clearly too conservative in my 1987 estimates; both in terms of this date's overall and high grade rarity. My 1999 estimates are more realistic but still probably on the low side. However, the PCGS and NGC population figures for this date are ridiculously inflated. The actual number of true About Uncirculated 1840-C half eagles is more likely in the area of a dozen pieces. There are three true Mint State coins that I know of and I was not aware of any of these in 1987 and two of them in 1999. The third is a coin that only recently "came out of the woodwork."

I believe that the following conclusions can be reached in regard to this date:

The PCGS and NGC population figures are inflated due to resubmissions. Many of the coins graded AU-50 and AU-53 by PCGS and NGC are coins that would have been graded Extremely Fine in the 1980's and early 1990's. A few very high grade pieces came on the market (the finest known Pittman coin, graded MS-64 by NGC and the PCGS MS-62 Milas coin which first surfaced in the 1994 James Stack sale). Both of these coins had been off the market since the 1940's and I had no idea that they existed until they reappeared. Price levels for this date have risen to the point that few collectors can afford to buy high grade 1840-C half eagles. Thus, this is now an "investor coin" in high grades and, as a result, the overgraded high grade pieces that exist tend to bounce between dealers and in and out of auctions.

Are rare date gold coins not as rare as we have been led to believe? In some cases, the answer is yes but in most cases I feel that a combination of factors have conspired to distort the true rarity of these issues. Hopefully, the next generation of collectors will be able to make sense of this increasingly confusing jumble of data.