Rarity of Ancient Coins.

I am wondering how other collectors judge the rarity of an ancient coin. Of course, the easiest way is to check a copy of Sears works, but like any other catalogs they have mistakes and of course the books become easily dated. I collect Byzantine so no update of Sear has occurred in decades.

In the artworld older works such as Rembrandt and Durer are judged by how many times pieces come to market, this of course makes sense since these were created before the concept of limited editions. This concept was done in the CLBC catalog for Byzantine coins, the authors used coin archives to track previous sales of the coins.

We have no idea how many of different types of a coin were made, I am very specialized so I find how many examples that are listed in primarily Dumbarton Oakes Catalog, that gives me some insight to rarity I also rely on Archeological data on stray coin finds and hoard finds.

I am curios how other collectors judge rarity, it does not matter if you collect Greek Roman or Other , I would enjoy to hear your insights.

 

Log in or Sign up to hide this ad.

When I get a coin, I create an email for myself of information about the coin and as many pictures of coins of exactly the same type that I can find (and sometimes of even the same dies if there's a lot of examples).

I search FORVM, acsearch, coinproject, various sellers, RPC, and all other specialized sites that I can find (I take RIC and other ratings into account but don't put too much stock in them. I have many R5 rated coins where you can find quite a few examples).

For things like FEL TEMPs, you could spend days collecting images. For others, I just find a handful of examples. Some really rare coins, I may find one or two and some no examples. This is how I usually judge rarity for myself. Then I just label it with a very general "common," "scarce," or "rare" (or "very rare" in those few or no examples cases).

 

For the coins I am focussing on at the moment...the Flavian dynasty, I use multiple approaches. First I check RIC. For this volume there are 5 general ratings C3 for extremely common, C2 for very common, C for common, R for rare, R2 (very few examples in the collections examined) and R3 unique in the collections examined. As the authors mention this should only be seen as a guide and not the last word on the number of surviving examples of each coin.

After I have checked RIC I check BMCRE. In general if a coin is not in the British Museum it is rare. However, it is not enough to check the print edition because the last revision was 1976 and so one could expect that the catalogue would be incomplete as hoards of coins have emerged in the last few decades. So, you need to check the online site for the British museum.

My next stop is OCRE. This catalogues coins of a number of collections but it does not include all known major collections. Still it is very useful for getting a ballpark idea of rarity.

Next is Acsearch. I pay for a subscription so I can see what an example has sold for at auction. The difficulty here is that it does not cover all auction records. However, it is still useful for getting an idea of rarity. Keep in mind of course that large auction houses do not deal in low value coins very often. The absence of these from the search records does not necessarily mean that they are rare but that they have not been sold as individual coins by major auction houses listed on Acsearch. Another issue is that sometimes you can see 5 examples of a coin on Acsearch. Does this mean it is common? Not necessarily, it might instead be a popular or highly sought after type that comes up for sale in the major auction houses.

I also search the member galleries at FAC. I have found this to be quite useful but again it only gives a general idea of rarity.

In short, I do not use one approach to gauge rarity I use a combination of approaches. If I have exhausted all of the above examples and have only found 2-3 examples I am reasonably comfortable in calling the coin rare. It is through this method that I have discovered that I own the second or third known example of some coins.

Of course another method is that I search the collection of @David Atherton. If it is a Flavian coin and he does not have it then it is rare indeed

 

I don't go out of my way to keep track of rarity. I collect what I like, but if the dealer or auction house notes its "rarity" then I also keep note of that.
What I consider "rare" is something in my collecting interests (Julia Domna, The Ladies or Stobi) that I have not seen or have rarely seen examples of. I figure if it is unusual for me after 20 years then it is a rarity.

 

@BenSi - I like the question, and I will be curious to read other perspectives. As I tend to buy relatively inexpensive coins (mostly Roman republican denarii and Parthian drachms), and prefer them to be in fairly good condition, most of the coins I have qualify as relatively "common", or are perhaps unique like this one only for it's unusual strike error:

This coin gets a (2) "very common" in Sydenham.

or unique as in this "plate coin" from Shore:

My first interest is in the connection to interesting stories and history, so rarity also not a major factor for me - which doesn't mean I am not pleased to find out that a coin I have is scarce or unusual.

Most often I think more about "rare" in terms of supply and demand and my ability to buy a coin in a condition that interests me rather than having any perspective on the number minted or number in the market.

How many results I get in acssearch (or other databases) is my other most used reference on "rarity" or "availability" - where I consider the number of search results gives me some relative feeling for how many are out there.

 

My approach is similar to @Orfew , but I don't put much stock into RIC's estimate of rarity. Often, the authors of RIC simply cite Cohen, not only as documentation of the coin, but they take Cohen's assessment of rarity and value uncritically. I can point to many coins in my own collection that RIC claims is common but they are close to unobtainable in the actual coin market. Here's an example of one and it also illustrates my method of determining rarity.

Rather, I have a two-step approach:

1) Look up examples in all of the online databases of which I'm aware, including records of previous sales and of museum holdings. These are: Wildwinds, OCRE, The ANS collection, The Coin Project, Coryssa, acsearchinfo, CoinArchives, Tantalus, the British Museum, CNG's archives, and V coins. Depending on the issue, I may also search ISEGRIM, MER/RIC, RPC online, or the Gallienus websites. Occasionally, I am unable to find another specimen of a coin apart from Cohen citing its presence in the French national collection. A coin is truly rare in that case.

2) If the coin appears rare (only half a dozen examples total among all the sites), I will do a die-study. If a coin appears to be struck with only a few different dies, then it's likely to have been in limited production or that only one or two hoards have ever been found containing it. For example, only nine examples of this coin with a right-facing bust appear in any reference, printed or online; they were produced by four different obverse dies and only a single reverse die. The denarius shown at the bottom of this post appears in only one other online source, the British Museum Collection, and it's a reverse die match, suggesting production by a single aberrant reverse die. There were only four other examples of this coin, struck by two obverse dies and two reverse dies.

The combination of very few examples online and produced by only a few different dies suggests true rarity.

 

I consider a coin to be rare if it appears for sale--anywhere--no more than once every two or three years on average.

Phil Davis

 

I look at rarity based on a few different factors. The most basic measure I look at is simply how many of the type have sold recently, i.e. in the past few years but I also try and look at historical data. Sometimes the past few years' data is seriously skewed by a recent hoard which may temporarily make a type look common when in reality it is quite rare. Sometimes a collection being disbursed over a short period of time can also make a type look common in the short term, for instance in my area of collecting the RBW collection sales meant that several rare varieties had multiple examples for sale over the course of only a few years. Before that collection sold, some of these types may have only had a single example for sale every 5-10 years. Even with the recent influx of bronzes on the market, it turns out that many of his coins wound up at the American Numismatic Society either through purchase at auction or donations so when I'm considering the possible number of examples on the market I try to keep this in mind since those probably won't ever be for sale again in my lifetime.

 

I only think about "market rarity", my chance for purchasing a coin on the Internet numismatic market in the 21st century. I am not interested in what was common when the great collections were being assembled by 18th century aristocrats.

A coin is "common", to me, if there are multiple examples for sale online. Market availability has a lot to do with dealer optimism. Some "scarce" types appear to be common because a few dealers have G or VG condition specimens on their web sites at 5x market value.

A coin is scarce if multiple examples are offered in a single year.

A coin is "extremely rare" if zero or one example appear for sale during the lifetime of an active collector. For example, this coin of Akraiphion in Boeotia which sold in a 2006 auction; no coins of the city had sold before that since 1879.

David Hendin and Arthur Houghton wrote an article on rarity for The Celator in 2008. They tried to relate absolute rarity to relative rarity. For example a ‘rare’ Seleucid bronze should have 10-30 known specimens, while a rare Jewish silver coin has 25-75 examples.

Valuable coins appear less rare than they actually are. A $15,000 coin with 10 known specimens seems to appear on the market more often than a $150 coin with 10 known specimens.

 

There are also levels of rarity depending on the relative scale of scope. Athens tetradrachms are very common but finding one from a specific pair of dies is extremely challenging.

In some spheres, US collectors delineate by die variety and by die state. At that level, almost every ancient coin would be unique.

 

It is ironic , a handful of factors should determine price, scarcity is one factor but that requires desirability as well. We also have to add condition including attractiveness and if lucky provenance as well.
That is the logical answer but I do find collecting to have an emotional pull as well , I have overpaid on a hard to get issue but I have gotten some deals on some too, in the end it should all wash out.

 

About 90 years ago a hoard of Roman silver numbering 81000 coins was found at Marcianopolis and named the Reka Devnia hoard. It was cataloged by N.A. Mouchmov giving a count of the number of examples found in the hoard. The coins ranged from Mark Antony to Herennius Etruscus but the ends are not as representative as the middle which includes my favorite Severans. Of the 7256 denarii of Septimius Severus, common types were reported with over 200 while rare ones were represented by one or no coins. You can not be too specific comparing the difference between a 1 and a 2 or a 3 but having 10 makes a coin likely to be scarce and 100 common.

There are problems. Before they were counted, the coins were divided between two museums and the most rare ones seem to have disappeared from one group. The assumption is that someone skimmed easy to recognize rarities. Worse, the report was published on the worst paper money could buy in 1934 and surviving originals are in terrible shape. The ANA library had one but said it was too fragile to lend. There are private photocopies out there but the chances of your finding one is not good. People who have them do not sell. I am not aware of it being online but it would seem that it might be somewhere.

Listings for this work are given as "RD" followed by the number of coins of that type. There are many different coins with the same RD numbers. They are organized by Cohen catalog numbers but all RDx means is that there were x examples in the 81000.

Rarity means nothing unless there are multiple people who are seeking the type and are willing to pay extra to get it. Many of us buy coins for other reasons (beauty, history etc.) and pay more for coins we like.

This popular coin of Septimius Severus is RD198:


The one below is RD1. Do you care? Should you?

 

AJ brings up another good sign of rarity: dies. If all known coins of a given type were struck from a single die set, the coin is probably very rare. If it is next to impossible to find two coins from the same die even when you have many coins, the type is probably more common. There are exceptions to this rule but it is usually pretty reliable. Frequently Martin and I post our examples of rare Septimius denarii. Note how often our two coins share the same dies. Below are my five examples of a rare Septimius Legionary (LEG XXII PRI). They share the obverse die and have two different reverses. I am aware of one other die for each side that I do not have. This is a good hint that it is not common. If you have five (or 50 or 500) coins of the same type but none share either die, the coin is not super rare.

 

Truthfully, I've burned out on "rarity" as a numismatic term - I own or have owned about a dozen "unique" coins, and I don't even know how many more that are only known from a few specimens. At the end of the day, who really cares about a line break or legend variant?

What does matter to me is historical importance. Just last week I submitted the highest bid I've ever placed for a coin minted by Saloninus as Augustus while under siege in Cologne by Postumus. Handily lost by about $350 because I need to have some common sense about this hobby - even if I will likely never see another one listed for sale.

 

Here's an interesting anecdote on rarity. I bought this victoriatus a few years ago and at the time it was missing from most major collections, even the British Museum didn't have an example until 1970 or so. When Crawford wrote his book there were only 3 known examples and there was only one on ACSearch. I was extremely excited to find my example and even more excited that it cost me almost nothing.

Then an example appeared at CNG. And then another and another. They've now sold five so far, two this year alone with another in the next CNG printed sale. Obviously a hoard is being disbursed. I probably could have made a nice profit on my example when I initially acquired it but now who knows? That said, I'm not too worried about it since I don't plan to sell anytime soon but if you're the type who feels your coins are an investment and you're willing to pay big bucks for rarities you should definitely keep this in mind. The finest known pre-hoard examples of this type pale in comparison to some of the new examples so even buying quality doesn't necessarily insulate you against a hoard being found and today's rarities may not be so rare tomorrow.

 

I think when people think of rarity they use as a basis point modern coins. This is not a good comparison. The U.S. mint produced in 2017 something close to 3 Billion dimes and just over 2 Billion quarters. I do not think that the entire population of Greek and Roman coins of any kind would come close to even one of these numbers. Just a very cursory examination using AC Search produced the following
2900 Hektes of Phokaia, 5900 Hektes of Mytilene, 4800 Athenian Tetradrachms 454-404 B.C. 1500 Tribute Pennies (denarii) of Tiberius and about 4100 denarii of Nero. Though I must admit that none of these figures are completely accurate. There were some Starr group V listed among the Athenian coins, Republican coins among the denarii of Nero and so on. Like I said cursory. The point I am making is that ancient coins even very common ones tend to have populations much smaller than modern coins. The authors of RIC VI felt that in order for a coin to be deemed rare there had to be less than 50 examples noted. The least common U.S. Morgan Dollar the 1893 San Francisco has a mintage of 100,000. I do not know how many still exist, however AC Search (again) lists 56 such coins.

 

how long an emperor has been in power also has rarity. There are also emperors ,we know nothing about it or that was killed immediately. The longer an emperor has been in power, several coins have also come out. or festivals special events , theme ec.

 

I don't really worry about rarity but that said I have a few "rare" coins (according to the dealer most of the time). This Marcus Aurelius as Caesar drachm is supposed to be quite rare, though under other rulers is a common type.

 

I think many here are touch on many aspects of rarity. It is truly a multi-layered onion.

First, how are we determining "type"? Is it a variant, like a dot on a column versus no dot, or is it a completely new design? I would label a mint the same as variant if the design the same.

Second, how well represented is that ruler? Rare types rare rulers are common, since most only want to pick up a coin of that ruler, not any particular reverse. It's the same as type versus date demand for modern collectors. I always chose rarer dates for a Type coin because I could get it with some effort for the same price.

Third would be "global demand", ie not only demand for the type or ruler, but of the city or civilization. A "common" type of a really rare city will always be judged as rare because of people just wanting one for they city, no matter the type.

Last is the most important. Demand. If you are the only one who knows about a coin, it may be rare but obtainable by you because no one else looks for them. I know of some really rare pieces I own multiples of because I have looked for them for 20 years now. Anyone looking at my collection will think they are pretty common, yet I challenge them to find more than 1 in any other collection. Many, many other coins that are scarce you will never be able to obtain because there is intense demand. It is not just price. Many other really well placed collectors have multiple dealers looking for this coin for them. They will always be offered it before you.

So it really is an interesting continuum of rarity intersecting with huge demand swings, with leads to multitudinous definitions of rarity, and all definitions are correct in their own way.

 

In the case of medieval coins things are often more extreme, with issues that can be dated to the exact day when the series began being minted to coins so unlikely that nobody even presumed might exist. There are also issues that we know existed from the royal ordonnances of the kings of France for instance, but are yet to be identified in the available material or further discoveries.

For the Crusader states statistics have been made by taking into account the known dies and the rate of die combinations and the results were that many million such coins had been struck during the reigns of the Kings of Jerusalem for instance, and a very likely probability exists that coins on the market were recalled to be restruck by the new administration.

All these things have a mark to leave on relative and general rarity but not necessarily on desirability. For a collector desirability is oftentime the driving factor, for a numismatist rare, ugly and unlikely is better.