Railroad Certificate Price Index
Discovering long-term price behavior
As most of you know, I record as many prices of North American railroad certificates as I can find. By 2004, I had been collecting prices for well over ten years. At that time I noticed prices for many certificates were softening. I decided I needed a method to measure whether my feelings were correct. I needed a method of measuring the overall market, not just single items. I decided to create some sort of market index which meant I needed to decide what to include in the index.
Getting a grip on certificate pricing is complicated by the small populations of collectible certificates. In most other hobbies, collectibles with populations under 1,000 are considered extremely scarce. In our hobby, varieties represented by one hundred certificates are not considered particularly scarce.
Highly-rare items appear for sale very infrequently and usually only in live auctions. Prices realized from those auctions may not represent widespread demand so much as the ability of auction houses to promote sales. For instance, one auction house might outsell another equally-qualified organization simply because it has more names in its customer list. Similarly, an auction house that delivers its catalogs three weeks in advance of sales probably outsells one with shorter lead times. Waiting until the last minute to contact buyers is probably not a good idea. All these variables contribute to the fact that prices of rarities are highly variable.
The prices of common items are also highly variable. Oftentimes, dealers can sell super-common items only by dropping prices to wholesale levels or below. In fact, some dealers could probably give away common items and net a profit merely by freeing up valuable storage space. The potential pitfalls of measuring common certificates is fraught with as many problems as rarities.
It ultimately made sense to me to look at medium scarce items. These are the kinds of items that intermediate and advanced collectors are always seeking to purchase. Taking the lead from typical stock market indexes, I tried to find a "market basket" of collectible certificates that met certain criteria. I reasoned that tracking a sufficient number of medium-scarce certificates over a sufficiently long time would allow us to discern trends in the collectors' market.
What certificates should go into the "market basket"?
I set out to find a collection of certificates that:
- had been known in the hobby for at least ten years,
- were collected by a sufficiently large number of collectors,
- had commonly sold for $25 to $200,
- appeared for sale two or more times per year,
- were in middle of the market (i.e., were neither too expensive nor too cheap)
- included a few sought-after autographs.
I conducted numerous searches of my database and initially compiled a list of 125 possible certificates for the "market basket." Almost two-thirds of the varieties in my initial selection failed one or more of the criteria listed above. Ultimately, my "market basket" of certificates shrunk to 43 primary varieties.
Modification to a 100-certificate "market basket."
In 2017, with records of over a million sale appearances, I examined populations again. With the perspective of an additional thirteen years, I decided that the ten-year familiarity requirement was unnecessary. In its place I substituted a more generous requirement that certificates needed five recorded sales prior to Dec 31, 1999.
I also relaxed the requirement that certificates needed to appear for sale twice a year. Rare and mid-rarity certificates tend to appear and disappear in cycles rather than come up for sale in predictable frequencies.
These modifications allowed me to add 57 more varieties to the original 43.
What certificates are included in my "market basket" index?
They are medium scarce certificates that tend to appear in formal live auctions. Unfortunately, as the number of formal auctions has dropped, many of those certificates are now appearing more frequently on eBay.
You may see images of all certificate varieties either as a collection of images and as individual listings. Both formats show the same information.
How to smooth the effect of "wacky" prices
It is extremely common to see medium-scarce certificates sell for $200 in one sale and $40 in the next. Large price ranges are common. Radical price behavior is the rule, not the exception.
This means that certificate prices are "spiky." If we plotted every sale of a "market basket" certificate, we would be unable to detect trends; there would be too many up and down jags. One of the ways to smooth spiky data is to display averages of prices through time instead of single prices. With a collection 100 certificates, prices, we could average the last two prices through time and display those averages on a chart. That is what is called a "two-sample moving average."
It stands to reason that the more sales we average, the smoother our price chart becomes. Oddly high and weirdly low prices become less and less noticeable as we add more and more samples to a moving average. In the case of collectible stock and bond prices, a 10-sample moving average would be nice. Unfortunately, we have an insufficient number of sales for such an average. After experimentation, it appears a 5-sample moving average removes the effects of extreme price behavior fairly well.
(It would be wonderful if we had records of dealer sales, but those transactions are hidden from view. Dealers do NOT send me any indications of their sales.)
Graphs of moving averages show trends very clearly, but they have a downside. They are "lagging" indicators. They do not show the value of the last sale, but the average value of the last five sales. In other words, it takes numerous sales to display a reversal of increasing or declining prices. In our case, where sales are few and far between, the chart of 5-sample moving averages will always trail the current market by six months to a year.
Measuring the value of the "market basket"
I calculate 5-sale moving averages for each of the 100 certificate varieties tracked. The total of all 100 moving averages gives an overall value of the market basket two times a year.
The total collection of the "market basket" of 100 certificates was worth approximately $12,470 at the end of 1999. Its value grew to $13,170 by the end of 2003. From that point forward, the market appears to have dropped 40% to $7,944 by the end of June, 2019. As of the time I write this toward the end of 2022, it appears the erosion in prices has slowed.
Analyzing the index
t is important to realize that this index DOES NOT and CAN NOT measure the entire breadth of the market. There is no market-wide price reporting system; dealers keep their sales secret and few collectors report their purchase prices to me. Moreover, I cannot possibly record all auctions and transactions that take place.
Acknowledging those limitations, I feel the index represents a reasonably consistent measurement across a fairly wide selection of RAILROAD certificates. My goal is not to establish a perfect index value; my goal is to spot trends and the trend has been down.
It is apparent that the hobby suffered dramatic price drops after 2003. Dealers tend to blame the decline on eBay. I FULLY AGREE! We could argue the reasons for low eBay prices ad nauseam, but agreement is unimportant. We cannot do anything about the price decline until we understand it. This last few price charts suggest several important conclusions to me.
- 1) Prices seemed tantalizingly stable for about a year from the last quarter of 2013 through the third quarter of 2014. Unfortunately, the index subsequently showed a steep drop through 2015 and 2016.
- 2) The high prices of 2003 are gone. They will not return until the hobby discovers how to attract significantly more participants. Price behavior in other markets clearly proves prices always decline much, much quicker than they recover. Therefore, even if the market turns tomorrow, it will require a long period of recovery. It is clear that the market decline will not reverse until a substantial number of new buyers appear.
- 3) I applaud high-quality, full-color auction catalogs, but I believe high commissions are negatively affecting the market. The percentages of items sold in live auctions were greatly higher in the 1990s than they are now. Prior to the year 2000, commissions ranged from 10% to 12%. Sellers routinely sold 50% to 70% of the lots they offered. Auction catalogs were minimally illustrated and were printed entirely in black and white. Today, commissions range from 18% to 25%. Commissions of that magnitude reflect enormous costs required to print and ship full-color catalogs. The alternative is to depend entirely on internet auction sites. Unfortunately, high commissions are occurring at a time of decreased collector interest. As a group, collectors are not bidding anywhere near the levels they once did. High commissions are NOT the cause of market decline, but they are certainly not helping. After all, collectors are well aware that they do not pay any commissions when buying on eBay.
- 4) Statistically, eBay sales are switching from auction formats to fixed-price formats. My analysis suggests that eBay fixed-price offerings have an unintended dark side. Many amateur sellers price their items higher than the eBay market will bear. Having tracked prices for so long, I can testify that amateur eBay sellers take an EXTREMELY long time to lower prices to the point where their items will sell. Some overpriced items have gone unsold over three years (!) simply because sellers will not relent on prices. Reluctance to drop prices forces buyers to sit on the sidelines and watch the same items go unsold week after week, month after month. Would-be buyers have no choice but to conclude that even eBay prices are too high! This is a very bad sign for professional dealers who compete not only against other professionals, but against thousands of eBay amateurs.
- 5) The sad truth is, we need MORE collectors. Prices will rise ONLY when the number of buyers increases. The first step is to get people excited about the hobby. Personally, I believe that is possible. I do not detect any shortage of collectors. They are buying on eBay every day and a lot of stuff they are buying is junk. But I question whether those collectors are fully aware that professional dealers and formal auction houses have tremendous numbers of certificates they will never see on eBay. The question is how to reach those kinds of buyers. Professional sellers cannot advertise or reach eBay buyers unless they sell them something. That means they need to sacrifice inventory at eBay pries. Who wants to do that? It is a classic "Catch 22" situation!
ONE FINAL COMMENT
This price index represents a market basket of 100 medium-scarce varieties from North American RAILROADS. I do NOT imply a correlation to certificates in any other specialty.