***1882-S $20.00 PCGS MS62+ CAC, ex Fairmont***

a guest blog by Richard Radick

Introduction: Why it takes time (a lot of it!) to grade 400,000 gold coins.

In the first piece in this series, I distinguished three categories of coins in the Fairmont hoard: (1) the pedigreed Fairmonts, (2) the non-pedigreed Fairmonts, and (3) the ungraded Fairmonts. I defined the “ungraded Fairmonts” as coins from the Fairmont hoard that have either not been submitted to PCGS for grading, yet, or have been deemed not worth the time and expense of grading.

This three-part classification reflects a gradation in my knowledge about the three categories: information about the pedigreed Fairmonts is quantitative, detailed, and quite precise; information about the non-pedigreed Fairmonts is semi-quantitative, inferential, and depends heavily on indirect estimates from time series; and information about the ungraded Fairmonts is largely speculative. In other words, I don’t really know very much about the ungraded Fairmonts, and this seems likely to remain the case.

Individual coins can move between these three categories. For example, I know of several onetime non-pedigreed Fairmont coins have been “promoted” to pedigreed status, and there have also been a few demotions going the other way. However, by far the most common transition has been from the ungraded Fairmont category to either pedigreed (about 2%) or non-pedigreed (about 98%) status. Indeed, in early 2018, all the coins of the Fairmont hoard were presumably ungraded.

Preparing a hoard of 400,000 gold coins for market is not simple; for example, one cannot simply send the coins to PCGS for certification and grading, and expect to receive the encapsulated coins back in a month or two.

Consider some numbers. It appears to me that, with its present staff, PCGS can process about 10,000 coins per day, or perhaps 200,000 coins during a 20 working-day month. Thus, the only way PCGS could conceivably grade a hoard of 400,000 coins, even over two months, would be to dedicate its staff entirely to that one task, to the exclusion of everything else. This is a commitment that PCGS surely would not agree to make.

Accordingly, grading the coins of the Fairmont hoard has been a prolonged process that began in early 2018 and has continued up to the present time, an interval of almost six years, at an average rate of some 5,000 - 10,000 coins per month.

PCGS probably dictated the rate, but the marketing managers for the Fairmont hoard (presumably Stack’s Bowers, but perhaps some third party) determined the sequence. The double eagles and eagles were graded first, and the half eagles followed a couple of years later, with substantial overlap.

It now appears that the grading process for the Fairmont double eagles and eagles has largely run to completion, although smaller batches, typically numbering dozens (rather than hundreds or thousands) of coins, continue to appear. There is also some evidence that the grading process for the Fairmont half eagles is starting to wind down.

The dog that didn’t bark: What seems to be missing among the Fairmont coins.

In one of the Sherlock Holmes stories, the fact that a dog did not bark during the night when a crime was committed became an important clue to solving the case. Likewise, in studying the Fairmont coins, I have learned that it is useful to consider what seems to be missing, as well as what is present. In this case, what seems to be missing are the two tails of the grade distribution, the low-grade coins at the bottom and the high-grade coins at the top - not totally absent, to be sure, but noticeably so.

The low-end ungraded Fairmonts

At the low-grade end, the explanation is probably fairly simple: the coins are not missing, but rather they simply have never been graded. Low-grade Fairmonts exist, and have appeared in the Stack’s Bowers (SBG) auctions. There just have not been very many of them. However, those that have shown up also tend to be from scarcer issues.

The table below presents a list of the five lowest-grade half eagles, eagles, and double eagles from the auctions to date. The experts will instantly recognize several scarce issues; for the rest of us, I have included mintage figures as a proxy. Ignoring a couple of ringers, the lowest grade coins in the Fairmont sales have, indeed, been from scarce issues.

It seems very unlikely that most of the low-grade coins in the Fairmont hoard would be from scarce issues. Rather, SBG undoubtedly filled their sale rosters for these scarce issues with the best available, and in several instances, the only available coins were low-grade. For the more common issues, there is an abundance of Fairmont coins available, and so high-grade examples were chosen for sale.

To state it a bit differently, given that the Fairmont hoard contained a G-6 1870-CC eagle and two VF 1878-CC double eagles, there is every reason to believe that it also contained more numerous worn examples for other, more common, issues from the same era. There is, however, no evidence that large numbers of such worn, common-date coins from the Fairmont hoard have ever been graded. I have inspected the PCGS Certs for literally thousands of Fairmont coins, both pedigreed and non-pedigreed, and I have rarely encountered coins with grades below about VF-35 among the common dates.

There is, in fact, little incentive to have worn common-date gold coins graded, Fairmont or otherwise – their numismatic value exceeds their bullion value only slightly, and the difference is, oftentimes, comparable to the grading cost.

Indeed, I think it is likely that some ungraded Fairmont coins have been consigned directly to SBG’s Precious Metals (i.e., bullion) sales during the past several years. For example, in mid-2020, SBG sold several hundred ungraded common-date half eagles in their Precious Metals sales, including 115 moderately-worn 1895 $5 coins. About the same time in mid-2020, my time series show that PCGS graded almost 1,700 non-pedigreed Fairmont 1895 $5 coins. It is tempting to connect the two events: the better coins were sent to PCGS for grading, and the numismatically less attractive coins were consigned directly to the bullion sales. 115 / 1,700 = 6.8%, a number to keep in mind going forward.

If we accept the likelihood that low-end ungraded Fairmonts exist, the natural follow-up question is: How many of them are there?

An analysis of the Fairmont half eagles minted between 1834 and 1866 may offer a clue. As a reminder, I call these simply the no-motto half eagles (NM $5s), because I include the Classic Head coins in the category, along with the no-motto Liberty Heads (the Classic Head coins, of course, also do not bear the motto: “In God We Trust”). At this time, I believe my database for the Fairmont NM $5s is more-or-less complete.

As of late 2023, there are 3,254 known Fairmont NM $5s, 1,189 pedigreed and 2,065 non-pedigreed. These coins represent about 12% of the 27,188 NM $5s that PCGS has graded, and are distributed over the 121 varieties of the PCGS taxonomy, a taxonomy that ranges from the 1854-S (with two examples graded by PCGS) to the 1861 (with over 2,000).

I broke these 121 varieties into three groups: Group #1 - 78 varieties with 10 or fewer known examples from the Fairmont hoard (six of these, including the 1854-S, have zero), totaling 356 coins; Group #2 - 27 varieties with between 11 & 50 known examples, totaling 504 coins; and Group #3 - 16 varieties with over 50 known examples in each, totaling 2,394 coins and topped by the 1861 half eagle with 744. I then determined the grade distribution for the coins in each of these three categories, and normalized each distribution to total 1,000 coins for intercomparison.

Among these coins are 240 with PCGS-Details grades. In order to include them, I arbitrarily assigned each of them the lowest possible within-category numerical grade - thus, for example, an AU-Details coin became an AU-50.

The figures below show the results. Because XF & AU coins account for a large majority (over 80%) of coins in all three groups, I have included a second figure that shows only the lower left portion of the main plot - i.e., the lower tail.

The pattern is striking: Group #3 (the common dates, plotted as green columns) has only a small fraction of its coins below VF-30, relative to the other two groups. This can be quantified: for Group #1 (which includes all the scarcest varieties), 7.0% of the coins are VF-25 or below; for Group #2, 5.2%; for Group #3, 1.1%; and overall, 2.4%. In other words, the common-date Fairmont NM $5s are deficient by perhaps 5% in the lower grades, and this group (comprising the high-mintage coins) accounts for almost 75% of the total.

So, circumstantial evidence (discussed previously) suggests that almost 7% of the Fairmont 1895 half eagles (a common date) were sold, ungraded, in SBG’s Precious Metals sales, and an analysis of the NM $5s arrives at a number not much different - perhaps 5% of the common-date Fairmont NM $5s were evidently deemed not worth the time and expense of grading. Extrapolating (perhaps rather wildly), this may indicate that some 5% x 380,000 = 19,000 low-end coins from the Fairmont hoard remain ungraded, or, if round numbers are preferred, 20,000 coins.

Possible high-end ungraded Fairmonts

During the more than five years of Fairmont sales, a total of 194 coins grading MS-65 or above (i.e., gems) have been sold, from among a total of some 5,356 (pedigreed) Fairmont coins - in other words, less than 4% of the total. This seems rather low.

It seems even lower when we consider that the pedigreed Fairmonts comprise less that 2% of the coins in the Fairmont hoard. One thing we can probably safely assume is that all gem Fairmonts have been graded with the pedigree, so the 194 coins are likely the complete tally. Relative to the numbers in the hoard, the number of Fairmont gems is miniscule - about 0.05%. By comparison, much larger fractions of the total number of coins graded by PCGS are gems, as the table below indicates.

Even if we ignore the double eagles, the discrepancy is a factor of 30 to 50. Of course, Doug likes to remind us that the PCGS numbers for high-grade coins are inflated, perhaps by a factor of 2 or maybe 3, but 30? Or even 50? Something doesn’t add up.

One possibility, of course, is that gem Fairmonts are intrinsically scarce. Doug favors this explanation.

Perhaps the Fairmont hoard, when it was first assembled some 90 years ago, never contained many gems. Unlike many US gold coin hoards, the Fairmont hoard was not amassed from newly-minted coins and then exported; indeed, its oldest coins were minted almost a century before the hoard was formed, around 1932.

The Fairmont hoard does contain numerous mint-state coins that clearly did not circulate, but rather had resided for years or decades in bank vaults before their export, where they may not always have been handled very carefully.

After their export, the coins continued to reside in a vault for over eighty more years, in bulk, and again may not always have been handled very carefully. Gem grade gold coins are rather delicate, and it is certainly plausible that relatively few Fairmont coins survived as gems.

Or, perhaps the hoard was cherry-picked at some point in the intervening years, and its gems are already circulating among the stock.

Or, perhaps more Fairmont gems do exist, and are being withheld from grading and sale. After all, this happened once before - Fairmont coins were being graded and sold for almost four years before the number and condition rarities of the Hendricks Set and its five successors first appeared. What happened once could happen again - just saying.

Or, some combination of inflated numbers, relatively few gems to begin with, careless handling, a bit of cherry-picking, and withholding?

Time will tell - either the “missing” gem Fairmonts will appear, or they won’t.