Q. David Bowers
Rarity vs. Certification Reports
Some modern researchers prefer to base their opinions primarily upon certification service population reports and auction data, whereas I tend to use these but to take them with a grain of salt, sometimes with a whole shaker full. (More about auction data later.)
With regard to population reports published by NGC and PCGS, there are many issues of which just a fraction of those known have been certified. In fact, in 1992, PCGS founder David Hall stated that in his opinion, just 5% to 10% of the gems of certain Morgan dollar issues had been certified. As gems are more likely to be submitted for certification than are lower-grade coins, in his opinion even tinier percentages have been graded of lesser pieces.
Further, population report data are often skewed by the resubmission of the same coins. I recently purchased a rare Mint State Liberty Seated dollar that was submitted four times to a grading service before I got it. Thus, the population reports for that service show four specimens instead of one. Bruce Amspacher told me of an 1854 silver dollar that had been submitted to PCGS six times. Andrew P. Harris, M.D., advised me of a single rare Uncirculated 1878-CC trade dollar that appears in a grading population report as four coins.
As further examples of this, I mention a recent edition of the PCGS Population Report, which informed its readers that no specimens of the 1801 Proof "restrike" (novodel) dollar had been certified, that five Proof-64 specimens of the 1802 Proof "restrike" (again, novodel) had been slabbed, and that one Proof 65 1803 "restrike" (again, novodel) dollar had been checked. It is hard to make much out of these data, in my opinion.
A West Coast dealer told me of resubmitting a 1916-D dime to a leading grading service over two dozen times until finally it came back graded MS-65. A Massachusetts dealer told of a 1796 half dollar which was graded AU-53 when he owned it, but a year or two later, after multiple resubmissions to a grading service, it "improved" to MS-62. (Weimar W. White suggests that over a large number of coins submitted over a long period of time, differences caused by resubmissions of a single coin tend to average out. He cites an article, "A Comparison of PCGS and NGC Population Reports," by David J. Locker, The Numismatist, September 1991, as a concurring opinion on the same subject. Letter to the author, May 3, 1992)
Notwithstanding such resubmissions and other inconsistencies, it is probably the case that when a large number of coins are studied over a spread of different dates or varieties, the emerging patterns are relevant. In the following study, use has been made of comparisons based upon the numbers of Liberty Seated arid Morgan dollars certified in higher grades, for example. In series in which most pieces were originally sold to collectors, and few to the general public, NGC and PCGS data are the most valuable. Examples include Proof Liberty Seated, Morgan, and trade dollars.
Grading was, is, and probably always will be, subjective. However, in a relative sense the population reports are a very useful guide, as long as the population is not too small and also if the coins have sufficient market value, especially if they are worth $100 or more (lower value coins in many instances are not submitted, as certification can cost $10 to $20 or more per item). When thousands of coins are involved-such as with many Morgan silver dollar issues-some excellent conclusions can be drawn. If, for example, 10,000 specimens of Coin A have been certified in MS-63 grade, but just 2,500 of Coin B in like grade, it is a sure bet that Coin B is the scarcer of the two, and it is probably also safe to say that Coin B is about four times more elusive.
Continuing the above example, it is dangerous, however, to conclude that 10,000 (or even 20,000) represents the entire population of Coin A. There may be dozens or hundreds of bags hidden away. Further and as noted, if a coin is of low value, certification service reports have little meaning. Relatively few EF, AU, or low-grade Mint State 1921 Morgan dollars have been sent in for certification, as certification costs nearly as much-as the coins are worth.
Still further, it has been my observation that higher value Mint State and Proof coins held in private collections are often (in old-time holdings, usually) not certified, whereas those appearing in modern auction sales are likely to be certified by a grading service. As an example, as of September 1992, none of the five known 1885 trade dollars had been certified and, as noted earlier, no Proof restrike 1801 dollars had entered the portals of NGC or PCGS.
In the instance of great rarities-examples include the 1804 and 1870-S dollars and the 1884 and 1885 trade dollars-many if not most have changed hands publicly through the auction route. In contrast, expensive but not super-rare coins-the 1879-CC and 1889-CC Morgan dollars are examples- change hands mainly through private transactions. I estimate that of a given 100 Mint State 1879-CC dollars which change hands, only 5 or 10 appear in auction catalogues.
The farther we go down the rarity scale and the more common a coin is, the smaller is the percentage changing hands across the auction-block. The 1881-S dollar is very common in Mint State, and each year tens of thousands change hands. Relatively few are sold via the auction route.