Ever since the first edition of my catalog in early 1995, collectors have asked for rarity estimates. Starting with the first update of my online catalog in 2010, I finally rolled out my railroad stock and bond rarity system.

Why the wait? I have long been intrigued by rarity systems and have long tried to quantify relationships between rarity and prices. However, since my earliest days as a collector, I watched how collectors and sellers both misused, misrepresented and misinterpreted rarity estimates.

The main purposes of rarity systems. No matter the hobby or specialty, it would compelling to know how many items might exist on the whole planet. Sellers always love to say, "This item is one of only twenty known." But what does that knowledge really say?

By itself, not a lot.

Taken alone, knowledge of the populations of collectibles is like one part of two-part epoxy adhesives. In order to work, knowledge of populations can only work when combined with collector experience in the proper proportion. Novice collectors automatically think all low-population items are hard to find and are therefore quite valuable. Professionals know differently. In fact, they know that some low-population items are easy to find simply because they are not terribly desirable. Conversely, experienced collectors know many medium-rarity are hard to find because they are very deisrable for any number of reasons.

In short, collector experience translates the knowledge of populations into predictions of how frequently various collectible items will appear for sale.

With rare exception, sellers TRY to keep sellers ignorant of certificate populations. Our hobby is less than a half century old. Our knowledge of certificate populations is exceedingly meager. Knowledge is not publicly available. We almost never know how many certificates of any particular variety exist. We may never know. There is a very strong financial incentive toward maintaining ignorance, so most sellers do not even report serial numbers of the certificates they're selling. That is no accident.

I don't believe we will ever know certificate populations. We must find a stand-in.

Avoiding overstatement or understatement of rarity. In designing a rarity system for railroad stocks and bonds, I wanted to avoid making a system that would mislead collectors into thinking certificates were scarcer or more common than they really were. I am seriously concerned about creating misunderstandings because I feel that the overwhelming majority of collectors equate rarity with price. Most collectors actually believe that high-rarity certificates are automatically worth more than low-rarity certificates.

That is most emphatically not the case. Price is not a function of rarity; price represents DESIRABILITY.

Yes, there is a weak relationship between rarity and price, but only because collectors desire rarity to greater or lesser degrees. It should be obvious that rarity means different things to different collectors, thereby creating hugely different desires. It is the disparity in desire that causes wide and unpredictable prices ranges among items of similar rarity.

Discussions about designing a rarity system. About ten years ago, I began talking and corresponding with dealers, collectors and authors about designing a workable rarity system. The deeper the conversations went, the more we delved into the philosophy of collecting and what collectors really want rarity systems to tell them. We had numerous discussions about how finely we whould divide rarity scales and whether we should enumerate scales from common to rare or vice-versa.

What rarity systems need to tell. After years of consideration, I have concluded that collectors want rarity systems to answer two major questions: how valuable collectibles might be and how hard collectors will need to search.

Questions about value. Collectors always want to know whether certificates are vaulable or not. I cannot tell them that. I can merely give them price estimates. Price estimates represent actual offers and sales. Huge price ranges are normal, so I modify price estimates up and down based on experience, economics, time, attitudes, attractiveness, appearances, rarity, etc.

It is crucial to understand that I record and estimate prices. I do NOT estimate values.

Value is a personal consideration reserved for collectors, not catalogers. Value certainly takes price into account, but value is much, much more than a price estimate. Value considers how much money a collector has, how much he desires a certificate, his competing interests, his job, his relationship with his spouse, and many more factors.

To illustrate the concept, consider how people define good food, good whiskey, and good coffee. Value is an everyday consideration. People willingly pay more for items they consider to have good value even though their priorities shift constantly. Personal definitions of value make some people pay $100 for a meal while others of equal wealth settle for the senior special.

The compelling illusion that collectors can discover the ellusive relationship between value, price and rarity. Collectors generally believe that if they can find sufficient information about rarity and price, they can determine whether values are good or bad.

If value estimates depend on knowing exact rarity, collectors are in real trouble.

Rarity estimates usually depend on knowing populations, a statistic about which we know very little in our hobby. We only know exact populations for a pathetically small handful of varieties. (I offer immense gratitude to Scott Winslow and George LaBarre for supplying censuses of varieties they found in the Northern Pacific hoard!). Most of the time, we have only a vague understanding of the number of certificates that might exist. Our collective knowledge of existing populations of certificates is badly flawed.

On the other hand, the number of times that certificates have appeared for sale may be a sufficient substitute for population. I'll explain more in a moment.

Collectors also want to know about how much effort they will need to expend to acquire certificates. They want to know how hard they will need to search, how many sales catalogs they will need to buy, how many losing bids they will need to submit, how many dealers they will need to contact, and how long they will need to wait.

I suggest that if a rarity system can answer questions about value and effort, then collectors can approach their goals more effectively than they could if they knew precise populations.

As an example, imagine that a collector desires a certain certificate variety, but can learn only one fact. Which fact would give the most insight?

• The variety has 87 examples known.
• The variety has 12 serial numbers recorded between #7 and #221.
• The variety appears for sale about once every two years.

I agree that it would be wonderful to know the specific numbers of certificates that exist for every variety. We can't. I agree that we can estimate populations from known serial numbers. We cannot determine our accuracy.

We can, however, track how frequently certificates appear in collections, sales and price lists and thereby estimate potential rarity. Our tracking mechanism is most assuredly flawed, but it is one we can constantly improve.

The rarity system I am proposing is based on how many times certificate varieties have appeared since mid-1980. I have recorded earlier sales of certificates, but my best records start around the time that George LaBarre began publishing his 3-volume guidebook series, Collecting Stocks and Bonds (1980-81).

Although I have tried to track as many sales and appearances of railroad certificates as possible since that time, I cannot claim completeness. Instead, I argue that we have a workable estimate of how frequently certificates have appeared. The underlying assumption is that certificates should appear about as frequently in the future as they have in the past.

My rarity system is a seven-point system arranged in order from most common to most rare. This arrangement of common-to-rare seems to be the most popular scheme. (See my summation of popular rarity systems used for similar collectibles.)

Limitations. I cannot possibly record every certificate that appears for sale. There are too few hours in the day. Therefore, a large number of certificates escape my attention. The major limitations to the accuracy of my rarity system include:

• Focus on major sales. I record every certificate that appears for sale in major auctions (Spink, FHW, HSK, Gutowski, etc.) regardless of whether certificates sell or not. I am usually forced to ignore multi-item lots because descriptions rarely allow me to discover exact varieties included. (See the list of every major Auction Catalog sale recorded so far.)
• Missing major sales. I depend on major dealers and auction houses to send me copies of their catalogs. I cannot possibly afford to subscribe to all that appear. If there are recent catalogs missing, then those dealers chose not to donate. If there are older catalogs missing, then I have been unable to acquire them either through purchase or borrowing.
• eBay sales. Every day, I record eBay sales that completed the day before. Because of the number of inexpensive items sold on eBay, I restrict my recording to those certificates that sell for $24.50 or more (rounded to $25). I do not record cheaper certificates. The vast majority of unsold eBay items reappear, so I do not record unsold certificates.

Special notes

Overall rarity of ALL certificates. I have stressed time and time again that ALL our certificates are rare by comparison to other major collectibles. For instance, there are 15 examples of the 1804 silver dollar and the most recent sale realized $2.3 million in April, 2009. Yet, because those coins have appeared for sale many times, the 1804 silver dollar rates only R6 on my scale. There are currently almost 15,000 varieties of certificates of equal or greater rarity! Yet only a tiny number of certificates, mostly autographed, have ever fetched more than $1,000 at auction.

Let me put it another way. Common varieties of railroad certificates would be considered scarce in many hobbies.

Rarity designation BY VARIETY. My system quantifies the rarity of certificates by variety. I combine all types of issuance, cancellation, condition, and autographs into one rarity designation. A medium-rarity R4 variety might possibly be represented by unissued and issued certificates in both cancelled and uncancelled forms. If reported separately, each of the possible combinations might theoretically be classified as R6. In fact, a unique unissued, uncancelled remainder would rank as R7. If I assigned rarity estimates based on individual certificates instead of varieties, I would virtually guarantee mis-valuation. In this example, I can easily imagine inexperienced collectors believing that an unissued, uncancelled remainder certificate would be the most valuable certificate in the variety.

Unrecognized high-rarity certificates. My inability to "see inside" multi-item lots and my non-recording of sub-$25 eBay lots means that I create estimation errors among the cheapest certificates. Populations and appearances of pricier certificates are easier to track. Previously-unknown certificates routinely appear for sale on eBay and many sell for less than $50, if they even sell at all. Low-priced certificates with few appearances may be very scarce or very common. If sellers do not report serial numbers, we cannot really say. Collectors seldom contribute images and serial numbers for so-called "common" certificates, so low-priced certificates are seriously under-represented in my tracking system.

Dealers who use "stock images" in their sales. Some sellers (eBay and professionals alike) illustrate their offerings with images of certificates different than the ones they are selling. Whenever I encounter these situations and can read serial numbers, I record two certificates: the one being shown and the one being offered. While not necessarily true, I assume that sellers who use stock images have (or had) at least two similar certificates in stock. Theoretically, at least two certificates are for sale.

Multiple appearances in quick succession. It is not unusual to see certificates offered two, three, even several times in quick succession. When this happens with eBay sales, I assume that sellers are liquidating multiple certificates. The same thing, of course, happens with major dealers who frequently have several similar items in stock. When the same certificates appear multiple times in major auctions, items are probably going unsold at the minimum prices offered. Nonetheless, those multiple appearances are legitimate offers to sell and therefore must be counted as multiple appearances.

Appearances of "new" certificates. Collectors are often "jumpy" the first time previously-unknown certificates appear for sale. Since collectors are unsure of rarity, it is not uncommon for certificates to sell for higher prices at their second or third appearances than they did on their first appearance. Since my rarity system is based on appearances averaged over thirty years, it is possible that a single certificate that appeared seven times could shift a rarity estimate from R7 to R5. That is certainly possible if minimum asking prices were absurdly high. On the other hand, if the same serially-numbered certificate appeared for sale seven times with no others seen, collectors might feel more confident in assuming that such a certificate is genuinely scarce.

Appearances of mini-hoards. Several small collections of previously rare or unknown certificates have appeared for sale on eBay over the last few years. Near-identical certificates have often appeared for sale every week or two for several months and then they disappeared completely. Since my rarity system depends on 30-years worth of appearances, mini-hoards do not cause a serious under-estimation of rarity. Short-lived "blips" in supply might decrease rarity designations one or two steps, but over the long-haul, rarity estimates will probably not change much more. Unless, of course, additional "hoards" appear.

Ownership by "strong hands." It is inevitable that rare collectibles move from weak hands to strong hands. That means that non-wealthy collectors may have the pleasure of owning rarities for awhile, but sooner or later, rarities migrate to more wealthy owners. Wealthy collectors are not frivolous, but they can afford to pay "what rarity is worth." Once rarities enter collections of wealthy collectors, they may remain unseen for decades. Moreover, it is not uncommon for wealthy collectors to sell rarities by "private treaty." That means that reputable dealers serve as intermediaries to discreetly move collectibles from wealthy owners to others without any appearance in the open market. I point this out to stress that collectors should never over-estimate the likelihood of rarities re-appearing in the marketplace. For instance, I have been watching for certain certificates that sold in 1976. They were obviously purchased by a "strong-handed" collector because not a single certificate from three multi-item batches has appeared since.