An Analysis of Worldwide Catalog Data

Back in this thread, Ron had broached the topic of an analysis of worldwide catalog data for the purpose of determining which collecting areas were the cheapest to collect. Never one to back down from a data-related challenge, I decided to attempt to replicate and update the analysis. So without further ado...

Background

Several decades ago, Michel Bégin, a collector from Québec, Canada, put together a study of worldwide collecting areas to determine which were the most cost-effective collecting areas. He did this by plowing through a 1997 Scott Classic Specialized Catalogue, compiling data on all mint issues from 1840-1940. He then quantified his finding through a simple formula that allowed him to capture which collecting areas had a low percentage of high-CV issues, and no issues with a really high CV.

Quantity with CV >$100
—————————–— x Highest CV
Total quantity issued

He published his findings on his website, where they sat until several years later, when the site lapsed and disappeared. It is currently only available through Internet archives.

About the Data

When Mr. Bégin undertook his study, he manually paged through a Scott catalog to assemble his data. I was not about to attempt that. I had the benefit of having access to data from the Michel catalog in electronic format, so that is the catalog I chose to use in assembling the current data.

Data in the Worldwide tables is from the 2015-2016 Michel catalog, with the exception of Germany data, which is from Michel’s 2015 Deutschland-Spezial.

The Table Types

Due to page-width limitations, each type (Mint & Used) is broken down into 2 tables, for a total of 4 tables.

Mint (#) — Mint issues by price range
Mint (%) — Mint percentage of total issues within the price ranges
Used (#) — Used issues by price range
Used (%) — Used percentage of total issues within the price ranges

For many issues, Michel lists catalog values for multiple mint conditions (MNH, MH, and/or MNG). The data used for this analysis follows an order of precedence — if there’s a MNH CV, that value is used. If no MNH CV is given, the analysis looks for a MH CV. Finally, if neither MNH nor MH is available, the analysis looks to MNG CV. A similar system is used for Used — Used CV first, but if none is available, CTO CV is used.

The Columns

IA — Issuing Authority
< €10 — Number of issues (#) or percentage of total issues (%) with a CV < €10.
€10-€50 — Number of issues (#) or percentage of total issues (%) with a CV of €10-€50.
€50-€100 — Number of issues (#) or percentage of total issues (%) with a CV of €50-€100.
> €100 — Number of issues (#) or percentage of total issues (%) with a CV > €100.
CV TBD — The Michel catalog has certain issues for which copies are known to exist, but there’s insufficient data for the catalog publisher to set a catalog value. This doesn’t mean the stamp is necessarily rare or valuable, but simply that there isn’t sufficient data to set a CV. This is denoted by “—,—” in the catalog, and is reflected here as CV TBD.
No CV — This is to denote those issues for which no catalog value is given (IOW, the space is blank in the catalog, not “—,—“).
Total — Total number of issues.
Max CV — Maximum CV for this IA and condition..
AQ — The Affordability Quotient, a variant of Mr. Bégin’s attempt to quantify which collecting areas are best on a budget. The lower the AQ, the more affordable the collecting area (however, see note below).

A Few Notes & Tips

• For many collecting areas, Michel’s data was subdivided into smaller parts (for example — USA regular issues, USA Postage Dues, USA Officials, etc.). I have not attempted to combine these into a single data point (for example — USA). If you wish to see all subparts of a collecting area, the easiest way is to use the IA search box (for example, entering “USA” will show 25 USA-related data points).
• There may be some translation errors in the IA names. The originals were in German, and I’m not fluent.
• I have not attempted to verify the accuracy of the data at an individual stamp level, as there were almost a million individual stamps in the data. I suspect there may be slight errors in places, as Michel sometimes lists set prices in addition to or in lieu of individual stamp CVs, which can lead to some duplication in the pricing data.
• I didn’t have access to unspecialized Michel data on the German areas, so the data from German areas is from the Specialized catalog, whereas the remaining Worldwide data is from unspecialized catalogs. Keep this in mind if comparing German collecting areas to non-German ones, as the German ones will contain far more listings than you’d find in an unspecialized catalog, including many rare varieties which can skew the pricing data.
• Finally, be aware that the AQ can be skewed by CV TBD and No CV. For mathematical purposes, CV TBD and No CV are treated as having a CV of zero in the database. This should not be mistaken for an actual CV of zero. For example, if an entire collecting area is only available Mint, there will be no Used values (No CV will be 100% of issues), and the Used AQ will therefore equal zero. I recommend you exercise caution with any areas with an AQ of zero.

An overview can be found here, with links to the tables contained within.