Some Thoughts on Rare Coin Pricing

I have been working on a few coin-related projects lately and one component which has made me stop and collect my thoughts on many occasions involves pricing. And then it hits me: if issues about coin pricing confuse me, how confusing must they seem to new collectors? Here are a few random thoughts about coin pricing.

Try to see if you can answer this question without cheating: what do you think an 1893-S double eagle in PCGS MS65 is worth? Without knowing much about the series, I'm going to predict that your guess is in the mid-to-high four figures; maybe as high as $12,500-15,000. What if I told you this was a $44,063 coin and could prove it?

Which brings us to Thought #1 of this blog: can a market be made by one coin trade? Quick answer: "yes but..."

The 1893-S double eagle, it turns out, is the Poster Child for Numismatic Condition Rarity. It is common as dirt in the lower Uncirculated grades and only marginally scarce in MS63. It is very scarce in MS64 with an estimated value  of $9,000-10,000 (which in and of itself might prove surprising to you for a coin with a population of 40 pieces, just at PCGS...). But it turns out that in MS65, this date is a beast with a current population of just two at PCGS and none finer.

Which is all well and good but how do you price a coin like this when one has never sold? Well, you wait until one does sell which is exactly what happened in Heritage's 10/12 auction, when a PCGS MS65 example brought an impressive $44,063.

I'm not going to make a value judgement about this coin as it isn't my intention to state whether it was a "good deal" or a "bad deal." But  if another piece were to come up for sale, I would have to use the $44,063 figure as a baseline comparison. As long as the population figure (i.e. the supply)  stays low for this issue, we can assume that the demand will remain fairly consistent. For better or worse, this $44,063 is what we are left to work with, even if the underbidder at the Heritage 10/12 suddenly thinks the coin is now worth $20,000.

Here's the thing about coin pricing: in very thinly traded markets (like coins with populations of two and none better) the market price becomes whatever the last trade is. In the case of very rare coins, this makes sense. In the case of a condition rarity in a series which is not typically collected by date, especially in Gem (like our aforementioned 1893-S double eagle) it makes sense but it is harder to embrace.

What about what I call "outlier prices?" Can a pricing structure be based on pricing anomalies?

This is a harder question to answer. Let's use another 1893-S double eagle as an example; this time in MS64. While scarce, enough have appeared at auction over the last few years that we can make some good assumptions.

The last MS64 to sell at auction was a PCGS MS64 with CAC approval which brought $12,925 in Heritage's 1/13 auction. It was a nice coin and one of the few examples of this date in MS64 with CAC approval; even so, the price realized has to be considered an outlier given other comparable coins.

The two previous APR's for the 1893-S in MS64 are $9,975 for a PCGS/CAC coin in Heritage's 2/10 sale and $9,890 for a non-CAC PCGS example in Heritage's 2/09 sale. The non-CAC coin, interestingly, was nicer than the CAC (in my opinion).

Why did the one coin bring nearly $13,000 while the other two brought a shade under $10,000? It could be a number of factors. It is possible that two "crackout" dealers felt that it had a shot to upgrade to MS65 (not likely given its scratch on the face) and they bid the coin up. Or, Type Three double eagles may be a bit stronger now than they were a few years ago (possible but are they nearly 30% higher, in the case of this date?). We'll never know the answer for certain but I don't feel strongly enough about this price to make the bold statement that all other 1894-S double eagles in PCGS MS64 with CAC approval are now worth $13,000. To me, they are still $10,000 coins and this one "outlier" price realized doesn't necessarily mean a new price level has been established.

There are many other scenarios in which an outlier price can be attained.

Let's take a random example: an 1855-S  eagle in AU55. This is a coin which could have a potentially huge range in valuation. And there are many factors why this range could be so dramatic. If it was a "real" AU55 with original surfaces and good eye appeal it could be worth double the amount of a crappy processed example. If it was a nice AU55 with an SS Central America pedigree (in the original holder) it could be worth even more. My point is that we could see dramatic variations in prices if enough examples of this date were sold at auction.

I mentioned that the range on this date could be vast. While a junky, low end "buying it for the plastic" AU55 could be worth $7,000-9,000 in the current market, a choice PCGS AU55 with a CAC sticker could bring as much as $12,500-15,000. And a coin with all the bells and whistles (PCGS/CAC/pedigree/old holder) could bring close to $20,000.

Which brings me to another thought.

Some coins are easy to price. An Iowa half dollar in an MS65 holder is worth around $150. An 1882-S Morgan dollar is a $350 coin (unless it has spectacular multi-colored toning but that's another issue).  Other coins are hard to price and they require value ranges. As I pointed out above, the value range for an 1855-S eagle in AU55 could run from a low of $7,000 to a high of close to $20,000. Same date, same grade, same value, right? In the case of very rare, esoteric coins this is far from the case. There are a ton of factors which influence value and, unfortunately, this is often not reflected in pricing guides.

Would it be possible to create a pricing guide which reflected the fact that there are variations in value for many coins? In the case of 1793-1814 Large Cents such pricing already exists. This pricing assumes that there are at least three variations within each grade; let's call them "A" for nice coins, "B" for average coins and "C" for below average coins. These variations may not matter much for MS65 Iowa half dollars but they matter alot for coins like Chain Cents or 1855-S eagles or even common date Dahlonega half eagles in EF45. I would love to sell multi-tiered pricing for various series and believe that this could be done, albeit with a great deal of effort.

I've written this before but I believe that one thing that holds coins back from literally exploding as an asset class among sophisticated investors is a lack of high quality pricing. Someday, someone is going to realize this and they are going to create a proprietary pricing system (its not hard to do but it is extremely labor intensive and requires input from extremely knowledgable market players and specialists) which will revolutionize numismatics. Until then, many new collectors will have to gamble that the 1855-S eagle which they just paid $15,000 for is a $15,000 1855-S and not a $7,500 1855-S. CAC approval is a start, but dissemination of information is a necessary next step.

Pricing "Difficult" Coins: A Real World Model

I have written a number of blogs in the past few years about how I price rare coins. Despite this, I still get many questions from new and experienced collectors about pricing. I'd like to share a specific coin that I recently handled and explain how I came up with buy/sell prices. As I have written, I find many of the published price guides to be of little or no use when it comes to complex, infrequently traded coins. When I make decisions at shows, in my office, or in the auction room on what to pay for a coin, I tend to put a lot more credence in auction records. So, if you'd like to play at home, I suggest that you follow along with the PCGS auction archives on pcgs.com as this is a major source of information for me when I make pricing decisions. Here is a "real world" model and the thought process(es) that went along with my pricing decision.

1863 Half Eagle, Graded MS60 by NGC and CAC approved

This is a coin I handled earlier this year and it is one of the first pieces in a while that, as soon as I saw it, I said "I have to own this." Before I discuss my thoughts about how to price it, let me discuss a little about the issue and about the coin itself.

Only 2,442 business strikes of this year were made and my experience is that the 1863 half eagle is rare in all grades, especially in AU50 or better. I jogged my memory and couldn't recall having seen an example I thought was better than AU53 to AU55 in more than a decade. And, I remembered that this was an issue that typically comes with zero in the way of eye appeal. A quick look online showed me that the PCGS population was none for Uncirculated coins and five for AU58; NGC had graded two in Uncirculated (an MS60 and an MS61) and five in AU58. At the time, CAC hadn't approved a single 1863 in any grade; a good indication that the eye appeal of the typical example was not good.

(How can you, even without my experience, make the same conclusions? Look at the pictures of the 1863 half eagles sold at auction during the last ten years. Are the coin fresh and original or are they bright, abraded and processed? Then, look at the number of auction records. A quick scan of the PCGS archives showed a total of 30 records since 1941. What was immediately impressive to me about this figure was that the highly-regarded 1864-S half eagle had 32 auction records in that time period!)

Of course, all these statistics are just gobbledygook if the coin itself isn't "all there." As you can see from the photo above, this coin had really good eye appeal. In fact, my first question was "why is this only in a 60 holder?" (I recently overheard heard a wholesale dealer, who I regard as one of the top three graders in the world, refer to the MS60 grade as "dumb" and that he "hated it." I tend to agree with him but, in this case, I was smitten with the coin; even it was in the funkiest of all Mint State grades.)

So, at this point I was sold. What would I pay?

With no auction records for an Uncirculated coin, I looked at AU58's. The two most recent sales were $14,950 by Stacks Bowers in August 2012 and $14,375 by Heritage in May 2010. A quick look at images for both coins showed two pieces that were no better, in my opinion, than AU53 to AU55. So, after digesting this, I decided that I would pay at least $17,500-20,000 for a coin that was a real, CAC-quality AU58 (the last "real 58" I had seen was the Bass II coin which sold for $13,800 back in 1999...).

Having concluded that a "real" AU58 was worth as much as $20,000, I figured it would be OK to pay at least $30,000 for a really nice MS60. I wanted confirmation and then decided to see if there were comparable coins that had recent auction records in this grade. Back to the archives I went.

I didn't really find any good comparables for the 1864-P and 1865-P, two dates that I regard as somewhat similar to the 1863; at least in terms of overall desirability. I then looked at the 1863-S; an issue with 17,000 struck but a low survival rate. I believe that this date is about twice as available as its Philadelphia counterpart but, like the 1863-S, it is extremely rare in AU58 and above.

In their June 2011 auction, Stacks Bowers sold a nice NGC AU58+ 1863-S for a remarkable $25,875. This was the single best example of the date that I had seen in years and I thought the price realized would be strong but I was clearly not expecting a winning bid of over $25,000. But this was as good a comparable as I could find and it made me think that if a "gem slider" 1863-S half eagle was worth nearly $26,000 then a somewhat nicer example of a decidedly rarer date (the 1863-P) had to be worth at least $30,000-32,500.

After negotiations, I was able to purchase the 1863 half eagle in this price range. I sent it to CAC where it was approved, thus becoming the first and only stickered example of this date. I listed it for sale in the mid-30's and within a few hours I sold it to a specialist who had been looking for a high grade 1863 half eagle for many years.

And what exactly does this all prove? Here are a few thoughts that I gleamed:

1. With CDN Monthly Summary showing a "bid" of $20,000 for this date in MS60, I knew that I wasn't going to get any help from published price sheets. But that's not a surprise, given that no MS60 coin had ever traded.

2. A few things convinced me to stretch on this coin: its true rarity in all grades, its Civil War date of issue and its great eye appeal. But if I had been offered an 1863-P half eagle in MS60 that was ugly and processed, I might not have figured it for much more than the $20,000 or so that I decided a properly graded, attractive AU58 was worth; maybe even less, in fact.

3. When you are contemplating a purchase of a coin such as this 1863 half eagle, you have to be prepared to stretch. My quick analysis made me think it was a great deal at $25,000 and probably too much of a stretch at $40,000. So, at $30,000 I was still all in and at $35,000 I probably would have been as well but not without some complaining to the seller.

4. How effective is the comparable method I mentioned above for determining value? It can be very effective but it is fraught with potential landmines. Let's say there was just one comparable and it was from over a decade ago--would that be effective? Or what if there were three records and one was 100% higher for a comparable coin) than the other two--would you, as an informed buyer, know the circumstances behind this sale? Is it effective to compare a coin like an 1863 half eagle to, say, an 1863 eagle? Or is this too much of an "apples to oranges" scenario.

5. The bottom line is that no matter how pseudo-scientific we as dealers or collectors try to make pricing, a lot of the numbers that get placed on really rare coins are instinctual. If you are knowledgeable, you'll have a gut feeling that the price is "right" or its "wrong."

Would you like to read more about my thoughts on coin pricing? If so, feel free to email me at dwn@ont.com and fire away with some off your questions.