Predicting Trends In Numismatic Premiums?

What do you mean by plunged? If the pre- case was silver = $1 numismatic = 99, then after it was 22/78, thats fairly normal. If you mean now the coin is only worth spot, that sounds really strange.

I don't recall using the term "plunged" anywhere?

Let's say - hypothetically - that our 5 Mark isn't a particularly-valuable coin from a numismatic standpoint. Call it "spot x 3" in grade. With the understanding that a coin's true value is only what someone is willing to pay for it (this is vital to my explanation, and the absolute truth about numismatics), maybe some guy somewhere needs only this single coin to complete his set. He's a bit motivated as a result, willing to bid over the other bidders - and that coin's "true" value - for it. When he does, that becomes a part of the permanent record for "what that coin is worth." The next guy looks at that auction result and says, "Gee, why'd that one go for so much? Is the winner a fool, or does he know more than I do about it?" He then gets sucked into bidding fever based on that result, and bids another one too high.

We now have not a single result, but a track record.

That's one vector for a coin to achieve inordinate value in the marketplace. An influx of new collectors wanting that series is another - a front-page article in a major publication about someone's "greatest set of all time" is sufficient to create such a spike. A third (probably the most common) is the investor, either taking a position in Precious Metals (PM) because he perceives the price to be artificially low, or (I'm looking at you, current reality) taking what he believes a more stable longer-term position in PM because the stuff he's accustomed to is too volatile for confident investing.

Either of those bouy prices, and a rising tide lifts all ships.

Here's an aside for you - did you know that for you, the investor, silver is rarer than gold? That's because of silver's manifold industrial uses. If it weren't for coinage, silver would be pretty tough to get, actually. The price of bullion silver is "artificially" lower because of those who are satisfied with their silver PM position being composed of "junk silver" coins, whereas the overall percentage of gold being used for recreational or investment purposes is far higher - even though it's quite as useful from an industrial standpoint as silver - for all the same metallurgical reasons of thermal and electrical conductivity - it's too darned expensive to be viable. Only 5% of mined gold goes to industry. Yet it's "worth" more.

Because we say it is.

So, the upshot of this little aside: Buy silver. Heck, buy gold. As long as you're OK with 10 years as a term, you will not lose. It might not skyrocket, but chances are the stock market will both skyrocket and crash half a dozen times in that ten years. Where it be when you're ready to sell? Your guess is as good as mine. Where will PM be at the end of that ten-year period? Either "better" or "a whole lot better." Industry doesn't need stock certificates or dollar bills - they need highly-conductive metals.

I've always contended that gold is historically overvalued by comparison to silver. Yes, silver is *slightly* more common in the Earth's crust than gold, but not by enough to count. Certainly not by 50:1 (the rough price ratio), especially when the ratio of mined silver to mined gold is only 8:1.

But that was only an aside, although it ties into my main point:

None of this stuff is objective. It involves people, the most subjective machines in the universe, and therefore nothing proceeds to "plan." Rules are useless; they're what we break for recreation. That is why I advise against ever believing there's a "system" to the discrepancies between bullion price, numismatic value and the cost of your next filling - somewhere in that process is at least one very much fallible human being deciding how much you'll pay for what you want to buy.

You aren't seeing it right. If the 5 mark with a silver value of $4 is bringing $100 that premium has NOTHING to do with the fact that it has silver in it. The value is from collector demand because it is scarce or rare, collectors want it for their collection, and it is not that available. Now if the silver value of such a coin rises to $28 does that increase the demand? If so from where? If I am chasing silver am I going to be willing to pay $100 for $28 worth of silver if I can buy $100 worth of silver for that $100? No. So the rise in silver value doesn't increase demand from the bullion investors. Does the rising in bullion value cause more people to decide to start collecting rare 5 mark coins? Probably not. So rising bullion doesn't create more collectors. What about the existing collectors? Well the rising bullion value makes the common 5 mark coins cost a lot more and this will price some of the collectors out of the market and they will stop collecting. This will REDUCE demand for the coin which creates a downward pressure on the price not up. So bullion is trying to push the price up while declining collector base is trying to push the price down. This is why low value coins react to a follow the spot market but high value coins do not or if they do respond it is at a slower rate.

So as a rule the price of common coins depends on the metal contact, but the scarcer the coin the more the value depends on the collector demand than on the spot market.

I have to say, I find it very hard to believe that coin prices trends are not predictable.

At some accuracy, even a much more efficient, much more liquid assets (such as the S&P500 for example) have been successfully predicted.

(If you wish to see some algo trading results for price predictions, I suggest starting here: http://arxiv.org/pdf/1107.0036.pdf ).


The problem I have when trying to apply these methods to the coin market are lack of data (I guess it is possible to purchase every greysheet issue and try this, but I haven't done that), lack of liquidity and (extremely) high spread between buyer and seller prices.

thanks supgog! real informative link, although I'm probably light years away from comprehending each and every single bit. just look at those formulae! haha funny you should mention that because I just bought a lot of key date coins that aren't yet very popular, but picking up there, and if I sell them at catalogue price, i'll be raking in a profit of almost 700%. or should I wait? that's not possible to predict now is it though? just gotta wait it out... I see myself scouting out for said coins everywhere now

As with ANY investment, economics teaches that the decision to remain invested is equivalent to the decision to invest at current prices (which assumes a liquid market with low buy/ask spreads).

So, would you buy a PR70 Jefferson Golden Dollar as an investment today? If not sell it.

What's it WORTH - that's a far harder question to answer... I suggest rubbing a lamp - about as accurate as "someone says it will rise/fall".


Since you quote a grade, you should check the price guide of the registry service, that will be one indication: PCGS or NGC where you will note that neither of those show any drop in price.

Next you can check the Greysheet or Bluesheet (subscription services, they sell single copies) (http://greysheet.com/) (the greysheet gives wholesale bid/ask pricing, i.e. what a dealer will pay you to buy your coin or sell it to another dealer for, the bluesheet is similar but covers sight unseen pricing for coins from the major grading services).

Another free service is NumisMedia FMV (Fair Market Value) here

FMV shows a small drop since issue, the TPG price guides are steady, I don't have access to current Greysheet/Bluesheet. All largely useless.


You can look for real sales in the worlds largest 1-off marketplace? For that most of us check eBay, which shows sales at anything from $3.50 to $140 in the last month or so. Seems to be highly dependent on the specific label and a lot on random chance...