Whose Signatures Are Worth Least
By Antonio Lopes · · 11 min read
A signature usually helps. When you compare the same author sold signed and sold unsigned, the typical premium across 165 well-represented names is 1.63× — a €60 book becomes about €98. For the biggest names it is far more, which is the subject of our main guide to what a signature is really worth.
This article is about the exceptions: the 31 authors whose signed copies sold for less than their unsigned ones.
That sounds like it should be impossible. Adding ink cannot subtract value. And it doesn't — what the data is actually showing is something more useful, and it is a lesson that applies far beyond signatures.
The authors whose signature moves nothing
Some detail behind those bars:
| Author | Signed | Unsigned | Signed lots | Unsigned lots |
|---|---|---|---|---|
| Herman de Vries | €20 | €60 | 11 | 40 |
| Dick Bruna | €58 | €110 | 12 | 78 |
| Claude Nori | €75 | €120 | 13 | 29 |
| Alec Soth | €55 | €86 | 131 | 80 |
| Stephan Vanfleteren | €71 | €106 | 69 | 82 |
| A.F.Th. van der Heijden | €80 | €117 | 73 | 77 |
| Dirk Alvermann | €32 | €44 | 24 | 26 |
| Arnon Grunberg | €56 | €69 | 41 | 110 |
| Cees Nooteboom | €58 | €70 | 26 | 33 |
| Antoine d'Agata | €64 | €65 | 98 | 35 |
First, the thing this does not mean
It would be easy to read that list as a ranking of who matters least. It is the opposite.
These authors' books are worth more than average, not less. The typical unsigned copy by one of the 31 sells for €70, against €59 for the 134 authors whose signatures do add value, and against €60 for the entire database. 23 of the 31 beat that €60 baseline, compared with 58 of 134 — that is 74% against 43%.
So these are not obscure or unwanted names. They are collected, they sell above the market, and several of them are among the most sought-after photographers and novelists in the database. The signature simply is not the scarce part of the transaction. That is a fact about supply, and supply has nothing to do with merit.
Reason one: the signature is not rare
The first explanation is the obvious one, and the data does support it — modestly.
Among the 31, signed copies make up a median 33% of everything that author sold, against 24% for authors whose signatures added value. Ten of the 31 have signed copies above 40% of their total lots; only 24 of 134 winners do.
At the extremes it is stark:
- Alec Soth: 131 signed lots against 80 unsigned. Nearly two out of three copies on the market carry his signature.
- Antoine d'Agata: 98 signed against 35 unsigned.
- Joan Fontcuberta: 80 signed against 28 unsigned.
- Bill Henson: 42 signed against 26 unsigned.
Photobook publishing does this deliberately. Signing sessions at Paris Photo, Offprint and Unseen are part of how the books are launched, and a run of 1,000 copies can leave the publisher with several hundred already signed. If most copies in circulation are signed, then "signed" is the ordinary state of the book and unsigned is the unusual one. Nobody pays a premium for the normal version.
This is the same scarcity logic that drives everything else in the trade, and it is worth reading alongside what collectable photobooks sell for, because photobooks are where this pattern is most concentrated.
Reason two, and the bigger one: the treasures are in the unsigned pile
The stronger effect is not about signatures at all. It is about what else is sitting in each pile.
For 24 of the 31 — 77% — the single most expensive lot that author ever sold was an unsigned copy. Among the other 134 authors that happens only 46% of the time. Compare the strongest 10% of each pile and the gap holds: for these 31, the top unsigned copies sell for 1.41 times the top signed ones, while for everyone else the top signed copies win by a wide margin.
Dick Bruna is the clearest case in the database. His signed lots typically make €58. His unsigned lots typically make €110 — and the reason sits right at the top of the list: three separate first-edition copies of Nijntje from 1955, which sold for €2,950, €1,698 and €1,300. Miffy's first appearance is a genuinely scarce children's book, and it was printed decades before the signing sessions that produced most of the signed material.
Bruna's signature is not worthless. It is simply that a 1955 first edition is worth enormously more than a signed 1985 reprint, and once a few of those land in the unsigned column they lift the whole side.
The same shape repeats. Herman de Vries's dearest unsigned lots are numbered artist's books in editions of 100 and 150 copies from the 1970s. Dirk Alvermann's are copies of Algerien from 1960. Ed Templeton's are Teenage Smokers from 1999. In every case the unsigned pile happens to contain the early, small-edition, hard-to-find material — and the signed pile contains the recent books people queued up to have signed.
So the honest summary is this: for most of these names, the unsigned pool contains the treasures. The comparison is not really signed versus unsigned. It is early-and-scarce versus recent-and-common, with the signature along for the ride.
The Alec Soth warning
One author shows why a single number can mislead badly in both directions.
Alec Soth's signed copies typically make €55 against €86 unsigned — a 36% discount, on 131 signed lots. But his most expensive lot of all is a signed copy of Broken Manual with a c-print, at €1,700, and a second signed Broken Manual at €1,100. His dearest unsigned copy made €451.
Both facts are true at once. There are so many ordinary signed Soth books in circulation that the middle of his signed range sits below the middle of his unsigned range — 72 of his 131 signed lots sold under €60. Meanwhile his signed special editions are the top of his market.
If you own a signed Soth, the typical figure is not your answer, and neither is the record. This is exactly the trap that reading an auction result properly is meant to help you avoid: what a group of copies typically makes tells you very little about the specific copy in your hands.
Two things I checked that did not hold up
I want to flag these, because the tidy version of this story is not quite what the data says.
"The unsigned copies are simply older." Plausible, but no. Across these 31 authors the typical signed lot dates from 2008 and the typical unsigned lot from 2005 — three years apart, which is nothing. Only 18 of the 31 have an older unsigned pool at all. The early scarce material matters because of its price, not because it is a large share of the pile.
"It's the bundles." Multi-book lots — "lot of 30 books", complete runs — usually sell for more and are almost never signed, so they could be doing the work. They are not. Bundles are actually more common in the winners' unsigned piles (6% of lots) than in these 31 (2%). Strip every bundle out and re-run the whole comparison, and 27 of the 28 names that still qualify remain below 1.0.
The finding survives both tests, which is why I am confident reporting it.
What this means if you own a signed book
Look up your author before you price anything. The typical premium is 1.63×, but the spread runs from about 0.3× to nearly 7×. An average is a poor guide to any individual name.
Ask whether the signature is the scarce part. If the author signed at festivals and book fairs, hundreds or thousands of identical signed copies exist, and yours is competing with all of them. If the author died before signing tours existed, the supply is fixed forever.
Check what the unsigned copies of your title are doing. If unsigned copies of a particular early edition sell for more than signed copies of the later one, that is telling you the edition matters more than the ink — the same conclusion our data reached about first-edition labelling, where the label alone turned out to be worth almost nothing.
Do not conclude your book is bad. A common signature means the author was generous and well-loved enough to be asked often. Two-thirds of the names on this list sell above the market average unsigned. The signature just is not where their value lives.
The quickest way to settle it for your own copy is to search for the title and compare the signed and unsigned results side by side. If you are starting from scratch, our guide to what an old book is worth walks through the features that move price in order.
How this was worked out
The method is the same one behind the main signature article, re-run from the raw data rather than carried over.
From 800,385 auction and marketplace results I removed non-book lots — the catalogue also carries jewellery, watches, model trains, football shirts, action figures and original artworks — leaving 576,341 book, map and print lots with a price. I then grouped lots by the author or maker named in each listing, and kept only names with at least 10 signed and 25 unsigned lots and an unsigned typical price of at least €20, so a ratio would mean something.
That leaves 165 names. The typical signature premium across them is 1.63×; 133 are above 1.0 and 31 are below. Those counts shift slightly depending on how strictly you strip out non-book lots — a looser filter that still includes autographed photographs and memorabilia gives 184 names and 42 below 1.0 — which is a reminder that the exact number matters less than the pattern. One photographer appears twice under two spellings of his own name, and both entries fall on the same side.
Two limitations, as always. Sellers describe their own lots, so this counts claimed signatures rather than authenticated ones. And every figure is a typical price rather than an average, because a single very large sale would otherwise rewrite an author's whole result — as the Alec Soth example above shows rather neatly.
