I happened to be around-in fact in close proximity, when coin dealers first began using numerical grading. There was nothing numismatic about the approach. The idea was to create a program to assist in the marketing of coins. At the time, dealers were using a variety of adjectives to describe the condition of their wares. Any serious attempts to standardize this verbiage had failed. The person responsible for first using numerical grading to describe coins other than United States Large Cents rightly predicted that customers would think a numerical grade provided a greater level of precision and consistency. They were also correct in assuming that it would seem more credible to borrow an established, thought rather obscure, numbering system than to invent a new one.
The first modern users of numerical grading were attracted to the system as a way to market uncirculated coins. The grading of circulated coins was already quite standardized because diagnostic features on the coin were used to determine the degree of wear. For example, it was generally agreed that for a United States Indian cent to deserve the grade of Very Good at least three letters in the word on the Indian's headband must be visible. There might be some question as to what visible meant, but collectors who were familiar with these diagnostics were seldom talked into paying a Very Good price for a coin that was no better than Good. Nor did they expect to receive a Fine grade coin at a Very Good price.
The situation was entirely different when grading uncirculated coins. Even the complete novice could see that some uncirculated coins were considerably more attractive than others, but there were no specific guidelines to rate these differences. One numismatist might decide a Morgan dollar with any noticeable marks on Liberty's cheek could not be rated as Choice Uncirculated. The next person could disagree and suggest that two or three light marks might be allowed for coins deserving this grade. The first dealers to use numbers to grade coins were looking for a way to segregate uncirculated coins into apparently consistent categories. Their plan worked far better than they could have ever anticipated.
From the outset, when grading Morgan dollars it was obvious that there was no sense in using all the available numbers within the uncirculated grade range. Numerical grading was primarily adopted as a way to massmarket fairly common coins. At the time, many of these coins were available to dealers for $4 each or less. If the dealer was offering a Gem Uncirculated example of one of these coins for $8 and a Choice Uncirculated example for $6 only two numbers would be needed to identify these coins. Actually three numbers were provided. A MS-60 was the lowest uncirculated grade. A MS-63, which would be considerably nicer, took the place of the Choice coin and a MS-65 represented the Gem quality. The dealers were prompted to follow this scheme primarily because previous business practices had to be considered. The dealer who first began using numerical grading had already sold many thousands of Morgan dollars. The dealer had to be certain the new numerical grades followed previous grading practices. The MS-63 grade couldn't be used to identify coins that were substantially better or worse than what had been sold as Choice BU. The person who had been content to purchase run of the mill Brilliant Uncirculated coins would find themselves owning what would now be called MS-60. Certainly the previous "Gem" buyer needed to be pleased with the new MS-65 grade coins they would receive and convinced that the quality of these coins was comparable to previous purchases.
Back then the value of the commodity being graded determined the practicality of using certain grades. It must also be remembered that the original reason for developing a numerical grading system was not to entice serious collectors into paying more for Morgan dollars, but rather to convince entry level customers to buy more coins. The new users of the numerical grading system understood that the last thing they wanted to do was confuse the customer. Even though it might have increased revenue, pricing a MS-61 at thirty-five cents more than a MS-60 and asking fifty-five cents over the MS-61 price for a MS-62 wouldn't work. To suggest to collectors that there might be coins worth more than the "Gem" prices they had already been paying to receive the best of the best would be a risky ploy. What these first "by the numbers" dealers couldn't have anticipated is how the numbers would alter the consumer's perception of what they were buying.
Complicating grading by creating too many categories would have also created havoc in the Grading Room where sorters where given thousands of silver dollars to process. Prior to using numerical grading, employees were instructed to sort the Morgan dollars into three groups. It was expected that in each lot of dollars there would be some average coins, better coins and a few really nice ones. Coins with heavy marks or rim nicks were put aside to be wholesaled. Toned coins were placed in a dipping box and then sorted after the color was removed. Once a particular project has been completed staff numismatists would quickly review the work and perhaps move some coins from one pile to another. If any truly outstanding coins were found these were set aside to be offered to favored customers or sold to one of the few dealers who would buy them at a premium price. When it was decided to experiment with numerical grading the plan wasn't even mentioned to the sorters because the new numbering system had no impact on their work.
Why didn't the first coin marketers to use numerical grading call their finest "Gem" coins MS-67, 68, or even 69? The marketing strategy was to sell more of what was readily available rather than to coax a few collectors into paying substantial premiums for superior quality coins. When numerical grading was first applied to Morgan dollars the vast majority of collectors were not prepared to pay significant premiums for slightly better quality. To suggest that a certain common date coin should be worth $8 in MS-63 and $25 in MS-67 would have been breaking new ground and unearthing few if any takers. That someone would willingly pay $8 for a "MS-63" was miracle enough. This is not to suggest that discriminating customers didn't recognize superior quality. Rather, the marketplace had not become sophisticated enough to appreciate how genuinely elusive a few very special coins might be. The purpose behind first using numerical grading was to add a new marketing wrinkle and increase the sale of coins that were readily available to the dealer.
Soon after the "by the numbers" marketing plan was launched a new trend in customer response was noted. Sales increased slightly, but hardly met management's always optimistic expectations. Those in charge understood that the overall coin market was slowing down somewhat and so from a timing standpoint the introduction of the new program was less than idea. After three months, an analysis of sales data exposed a curious trend. In the past, new customers would often begin making purchases by selecting Choice BU coins. One of the most popular offerings was a "Starter Collection" of 10 different dates complete with an album. After the introduction of numerical grading sales of this product dropped off. Many new customers were now selecting single coins from the Gem BU MS-65 listing. Because these coins has cost the dealer no more that the Choice BU pieces, per coin profits rose dramatically, but the amount of money the new customer was spending actually declined slightly. In short order, inventory control problems began to result. The numismatists were told to be a little less selective and move more coins from the sorter's "better" piles to the "best" group. Coins from the wholesale suppliers kept coming and the inventory of "MS-63" grade coins began to pile up.
The buyers were sent to coin shows with instructions to pick and choose instead of buying bulk lots as they had done in the past. The new strategy immediately created confusion and plenty of questions from the dealers the company often did business with. At the time, common date Morgan dollars usually traded at wholesale levels in roll lots or by the $1000 face value bag. Even if a slight premium per coin was being offered by the buyer, the selling dealer was reluctant to let them select a few coins from each roll. What would they do with the 15 or 16 pieces that were left over? For many years these coins were bought and sold as "original rolls" or bags with the understanding that the quality would vary someone and the buyer would take the coins as they came. To let someone choose for a fifty cent a coin premium meant the selling dealer would be left with picked over goods that would be difficult to sell to the next guy.
Back home, the dealer who first began using numerical grading to mass-market silver dollars was faced with another unexpected problem. Significant numbers of previous customers were now asking if they could trade in their Choice BU grade coins for the new MS-65 ones. Very few of those wishing to trade declined to do so after learning what it would cost. Because the dealer had published selling prices for both grades, he could hardly increase the asking price of the now highly desirable MS-65 grade coins or make obviously disappointing offers for the Choice BU coins he had previously sold. Staff members who had been doing productive work, were now forced to spend time dealing with these annoying exchanges. Only skilled graders could be entrusted with this work because it was necessary to make sure that the trading customer always received coins that were actually of better quality than they were trading in.
At the dealership, plenty of serious conversation ensued and on several occasions the company came close to scrapping numerical grading entirely. The staff member who had first suggested the plan and then developed it came under considerable fire. One adamant detractor pointed out that they weren't at all surprised. A letter from a customer pointed out that on a scale of 1 to 10 something called a 5 would be average while a 3 would be less than average. Who wanted to buy a less than average coin? If the company had called their Choice BU coins MS-65 and the Gems MS-68 then none of this would have happened!
During one gloomy meeting a person from marketing suggested the company send out a special price list to a few hundred of the company's newest customers. In this test mailing the MS-65 grade coins would be priced at a whopping 35% more than the MS-63 grade pieces. Nobody in the room believed such an aggressive pricing strategy would sell more MS-65 dollars, but it might convince customers to buy the MS-63 grade coins. Even new customers who weren't familiar with the traditional pricing structure couldn't be expected to pay $3.50 more for a coin that might be only slightly better than one they could own for $10.
The staff member who suggested the test received an extra nice bonus that year. Sales of MS-65 dollars went from good to sensational. After the test, the company's entire pricing structure was revised with all prices for MS-65 grade coins being raised significantly. A mailing to the entire customer list resulted in surprisingly few questions from customers. In fact, several of them called to say they were delighted to see that their investment packages was doing so well!
This new " price spread" between the grades made life much easier for the buyers. The primary buyer, who hadn't gained this status by accident, determined the team would go back to their previous practice of buying the coins as they came. The "garbage", that being anything less than what could be sold as MS-65 would be disposed of at a tempting price to a group of wholesale buyers. Along with these coins would go the many boxes of MS-63 and lower grade pieces that were now gathering dust in the company's inventory. Any minor losses incurred in disposing of this material were more than compensated for by the great profits generated from the sale of MS-65 grade coins. As the primary buyer expected, dumping thousands of Morgan dollars on the market immediately depressed prices. Once that was accomplished, the company's buying team had much less difficulty convincing dealers that they should be allowed to pick and choose coins from rolls and bags. The buyers were also aggressively trying to purchase better quality single coins. Dealers at smaller shows were amazed when one of Dealer X's buyers came along and bought piles of Morgan dollars at only a bit less than the dealer's "retail" price. What these sellers didn't know was that the coins they were able to move for $5 each were finding new homes with collectors at around $15 a pop.
In the coin business it's hard to keep a secret for long. That people were willing to pay considerably more for better quality common date Morgan dollars was important information to have and use. To the modern collector it may seem remarkable that these dealers and collectors didn't "get it". Now we understand that a MS-62 anything is worth more than a MS-61 of the same thing and that a MS-69 example of this coin is special. We didn't know that back then. I'm confident that if this initial effort to use numerical grading to market Morgan dollars had failed other dealers would not have attempted to make it work. When applied to coins other than Large cents the numbering system seemed illogical and arbitrary. I'm convinced that it did work because the company who tried it was willing to invest huge amounts of money to attract entry level customers to the hobby and they selected the right product to market.