Some Thoughts on the Maturing of the Rare Book Market at the Start of the 21st Century

[A talk given by Ken to the Fellowship of American Bibliophilic Societies]

Greetings. I'd like to thank you for having me here today. I'd like to talk about what I see as dramatic changes that have taken place in the rare book market in the last 10-15 years, and even more dramatic changes in the past 5 or 6 years. I want to talk about what kinds of changes we've seen, what has caused them, what do they mean for now, and what are their implications for the future.

I'll try to talk in terms that are applicable to the market in general and not limited to, or focused on, my own specialty of modern first editions. Probably most of my examples -- to the extent that I illustrate my theses with examples -- will come from modern firsts, since that's my specialty and where most of my experience lies. So this may sometimes result in a bias, or even in getting something plain wrong -- generalizing from a phenomenon in modern firsts when in fact that phenomenon is specific to modern firsts. If so, I'm hoping your questions and comments after the talk will show me the errors of my thinking.

For the most part, though, I feel confident in this approach for two reasons. One is that modern firsts is a legitimate and inextricable part of the current rare books market, especially in the US and Britain, but also around the world. And secondly, especially in the US, modern firsts is a more volatile part of the rare book market than many other segments and therefore trends in the overall market are often more readily visible in this segment than elsewhere.

As early as the early 1990s we began to see prices accelerating at an unprecedented rate, especially prices of "high spots," or those books most universally agreed to be valuable, collectible, etc. Several factors contributed to this: the beginning of the longest economic boom in American history played a part: more people were beginning to have more disposable income than before. Ironically, the recession that preceded the boom contributed to this by encouraging a small but relatively wealthy segment of the population to diversify their investment holdings, with rare books being added to the list that previously included stocks, bonds, real estate, fine art and other collectibles. In this context, books were rightly seen as having been undervalued for a long time: fine first editions of significant books were selling for anywhere from 1/10 to 1/100 of the prices of a similarly scarce Picasso print, say.

Also, in certain circles -- notably including, but not limited to, Hollywood -- there began to be a certain cachet, or sexiness, to book collecting that occasionally brought it to the general public consciousness in a way that hadn't happened before.

And finally, the building of, and sale of, certain notable collections -- the Garden collection, in particular, being one -- lent credence to the notion that there was still a challenge to assembling a fine book collection, even one built upon, as the Garden was, the "received knowledge" of previous collectors, bibliophiles, bibliographers, and scholars.

All told, the early '90s were an auspicious time for book collectors and the game was pursued with money, verve, and a discriminating -- and at times partly mercenary -- eye.

All this contributed to good times for book collectors and booksellers, and a general increase in prices -- which included a sometimes astounding increase in prices of high spots: as more people pursued the rarer books, they became even rarer and the ratio of demand to supply skyrocketed, dragging prices of the most sought-after books upward at a frenetic pace. The trend was exacerbated by the fact that as more money came into the market, new collectors were more likely than in previous times to start at the top -- buying the best and most expensive -- rather than beginning with modest collections and inexpensive books and gradually working their way up to more expensive rare books.

So, much of the pressure was focused on the top end of the market, and drove those prices upward disproportionately fast and disproportionately high.

A corollary effect of this was to bring a larger number of collectors' copies of these high spots into the market more rapidly than ever before, also creating an upward pressure on prices. You may wonder: how does an increase in supply force prices upward?

In the past when prices were moderately stable or increasing at a slow, even pace, collectors would have to wait until their books doubled in value, more or less, before they could even break even. For them to realize a substantial profit when they sold their books they'd have to triple in value, at least, and this could take not just years but even decades.

In the early '90s, however, we were seeing books -- high spots at least -- begin to sell at ten times what they had brought only a decade earlier. Even at the dramatic wholesale to retail differential in the rare book trade -- often a 100% markup is involved -- a collector who had bought a book in 1982 for $125 could sell it for 5 times that much in 1992, and the dealer could still double the price he paid for it.

So an unusually large number of copies of these high spots began to be recycled within the collector market, always predicated on the fact that the price increase has been so great. Since these were collectors' copies -- that is, already selected a decade or more earlier as being among the best copies available -- they typically came into the market at the high end of the current price range. Collectors wanting to maximize their profits on these books that they were disbursing from their own collection -- i.e., get as high as possible a percentage of current retail -- and dealers wanting to have these best possible copies in their stock combined to further fuel the upward pressure on prices of the rarest, or at least most sought-after and collected volumes.

The early '90s saw price records broken one after another in a continuous stream, at least for the most collected books.

This trend continued and increased through the mid-90s, and was fueled by what we now call the "dot-com bubble" but at the time was more generally called the "new economy." Thousands of overnight millionaires, maybe tens of thousands, contributed to this simply by virtue of the fact that in any demographic there will be a certain percentage of book collectors. The dot-com millionaires were no exception and they were both a wealthy and educated group. And, despite some efforts in the marketplace otherwise, they tended to be more inclined to collect a first edition of Moby Dick than a first edition of Lotus 1-2-3.

The dot-com bubble burst, as we all know, but the dot-com revolution continued to contribute to changes in the rare book market, most notably by harnessing the power of computers to match databases of books wanted with databases of books available. Very soon after this process began in earnest, the networking capabilities of the Internet began to influence not only the availability of books -- or viewed from the other side, the availability of customers -- but much more importantly the availability of information.

For generations, the rare book market had operated by means of a fairly rigidly defined hierarchy, or food chain. There were book scouts, who scoured the land for valuable books of all sorts -- and here I mean "valuable" in terms of "books that would have some value to someone," i.e., "saleable" books, not necessarily expensive books. There were a large number of general bookstores in most big cities, which would buy books in great volume and had sufficiently educated and experienced staff to recognize when a book was particularly rare and valuable -- and here I do mean, in general, expensive -- and knew moreover whether the store was likely to have a retail customer for such a book or whether the best way of selling it was to funnel it to a specialist bookseller, someone whose stock consisted exclusively, or at least primarily, of the rarest books in a given field or number of fields, and who was also therefore most likely to be in a position to have an actual retail customer for such a book.

This basic configuration of how books moved through the market went on for years, essentially unchanged, even into the early and mid-'90s. It was predicated on two things -- market position for the bookstore and specialist, that is being in a position to move those books, whether to another dealer or a retail customer, and, more importantly for all 3 -- especially the scout -- a body of information about what was valuable -- what people wanted and would pay for and wasn't commonly available. This body of information came slowly, over time, through much experience and trial-and-error (usually including a lot of errors), and eventually the market resembled a somewhat efficient vehicle for moving rare books out of attics and tag sales and into the hands of collectors and institutions.

The advent of the Internet changed all that profoundly, although it's taken some years for the change to fully permeate the marketplace.

I remember in 1997 we issued our third catalog of Native American literature -- books written by American Indian writers. It was actually our fourth such catalog, although one of them was never issued. We did our first in 1992 and it was received very well, especially by institutions. No one I know had ever done a catalog of that field before and we got a very favorable response to the catalog. In terms of the percentage of books sold, we did 50-75% percent better than our general literature catalogs, and those were already doing quite well by the general standards of the trade. Our second Native American catalog was in 1994 and did even better and with our third catalog we sent out a half dozen advance copies to the heaviest buyers from the first two, while the catalog was still at the printer. We sold so many books from those advance copies that we cancelled the print run: the catalog had mostly sold out and the remaining books, even if every one of them sold, wouldn't have brought in as much as the printing and mailing costs would have been.

Then in August 1997 we sent out our fourth, the third to be generally issued, and on the first day it hit I got a call from a librarian, effusively thanking me for my very informative catalog and telling me that he had found 2/3 of the books on the Internet for 1/3 the price.

I don't know if he thought he was being clever and catching me out, as though this were a game and he had scored a point, or if he was being sincere and was just oblivious. But I do know he didn't buy any books from that catalog -- despite the fact that he was deeply interested in, and committed to, that field. And I also know that his call was, for me, the clearest wake-up call I could have gotten. It was clear that the rules were changing, the playing field was changing, the market was changing. And what was changing most of all, and fastest of all, was the availability of information. All those books that he had found on the Internet had been available the day before. But he didn't have the information to recognize them. Now that he had the information -- which I had freely given out in a kind of transaction that had worked for both sides in the earlier market -- he did not need my books.

Native American literature is a field primarily of scarce, not rare, books. Most are scarce because they were printed by small publishers in small editions because the publishers had rightly recognized that there was only a small market for them. A lot of these were books an average browser might well not have stumbled across in a year's worth of browsing used bookstores, or might have missed even if they were there. When encountered, they could often be bought for $3 or so. And we could buy a book for $3, price it $45 -- or roughly double what a new, widely available book might cost -- and feel we were doing a good thing, both for ourselves and for our customers -- and for posterity, by ensuring these books were preserved. Given the cost of printing and mailing catalogs, not to mention some sort of compensation for the time and effort it took to find and catalog the books, $45 would be a little better than a break-even point for us. We'd make a few dollars on a book at that price, but that's all. For our customers, they had the benefit of our having done the work for them of finding these books and the benefit of our knowledge in recognizing them, and they would get a book that might be 10 to 100 times scarcer than a new book (or more) for roughly twice the price of a new book.

In its own way, this was a market that could be characterized as efficient -- that is, it worked. It got the books from wherever they were located -- a small used bookstore in Kansas or Idaho, for example, to where they belonged -- that is, where they would complement an existing collection, be seen in context, etc., often someplace like the Beinecke or Newberry libraries. Still, in terms of efficiency, it left a lot to be desired. It could take a year and a half to amass enough books for a catalog, a couple of months to prepare the catalog, and only then would the books begin to sell and the customers be able to add to their collections. There were plenty of other inefficiencies, too. We might print 2000 copies of a catalog and mail them, and in the end 80% of the buying was done by 20% of the customers -- a typical scenario in any kind of business. But if you don't know which 20% that might be, you have to print all those catalogs and send out all the ones that don't elicit any orders, just to be as sure as possible that you'll reach the buyers who will order.

Also, we did the calculations a number of times and concluded that every book we sold in such a catalog cost us about $30 to sell, not including the cost of the book. On the $500 and $1000 books, that's not so bad, but spending $30 to sell a $45 book can hardly be characterized as efficient. (One could, of course, calculate this differently, with those costs amortized over the total dollar value of sales as opposed to figured on a per book basis; this would make a more reasonable number, but in a field like Native American literature -- where the books tended to be scarce but not rare, it wouldn't be a huge difference: there tended to be a lot more books under $50 in those catalogs than over $500.) What made this approach work, however, or at least be viable, was that there was a scarcity of information as much as, or more than, an absolute scarcity of books. While many of these books were, and are, in fact truly scarce, customers for them are just as scarce, or more so. In absolute terms, the supply usually outstripped the demand for all but the scarcest of them. But the supply was not available to fill the demand except through the intermediary of the book catalog (and more generally, the entire "food chain" of the rare book market referred to earlier).

The advent of the Internet changed that, and booksellers were among the first large group of retailers to embrace the net en masse. As far back as 1997, there were already 6000 used booksellers on the Internet -- this at a time when a recent census of booksellers concluded that there were slightly over 7000 used booksellers in the country. (Now the number online is somewhere north of 10,000, but some significant portion of those would not be included in a current census as they are more along the lines of part-time hobbyists or individual collectors selling off small portions of their collections.) A recent New York Times article pointed out that 65% of the used booksellers on the Advanced Book Exchange -- the largest database of used books and used booksellers on the Internet -- have only an Internet presence: no bookstore, no catalogs, no book fairs. These do, however, include a significant number of full-time booksellers who either have closed their brick and mortar stores or, more commonly, began their bookselling careers at a point where having only an Internet presence was, and is, sufficient to support a legitimate full-time business. Whatever the actual numbers, the percentage of knowledgeable full-time booksellers who have their books listed on the net is extraordinarily high -- probably as high as or higher than any other comparable field.

So the Internet suddenly made, in effect, all books available to everyone, directly, and pretty much instantly laid waste to the traditional hierarchy, or food chain, within the book trade in years past -- scout, general bookseller, specialist.

The disparities that still existed were in the area of information: the bookseller listing a Maurice Kenny title, for example, might not recognize him as being a Native American writer. The librarian building a collection might not know that Carol Arnett was Native American, or George Plymell, the Beat writer and first publisher of Zap Comix.

In the first years of the Internet's prolific growth, it was not uncommon to find specialist dealers trying to figure out how to hold onto their specialized knowledge -- which they'd generally worked long and hard to acquire -- rather than giving it away for free, once and for all, by posting their listings on the net. Their concerns and fears were justified, too: in the early years of the Internet, there were those who were computer-literate enough to simply take the postings of all dealers, copy them and post them as their own, running an algorithm on the prices to mark them up, so that if a book sold they could buy it from the dealer who actually had it, sell it to their customer and make the markup as free and clear profit. This happened numerous times and one large database actually made this a feature of their offering: you could re-list all the books that other dealers had posted on the database, and they would give you software that allowed you to mark the prices up en bloc and make it look like you were the bookseller who had all the books. They even gave the resulting phenomenon a name -- a "super-site" they called it -- and provided it as an inducement to get dealers, or at least aspiring "dealers," to list with them. (From a marketing perspective, this was probably not so unwise: if any individual could overnight become the owner, or at least lister, of millions of books, then the potential number of subscribers to their site was not limited by the actual number of booksellers out there, who actually had books to sell, but could include anybody and everybody.)

There were, as you might guess, a few problems with this approach. For one, in the case of dealers who had written exhaustive catalog descriptions based on their own knowledge and research and putting the books in historical and literary context, there was the question of copyright violation. Those writings belonged to the dealers who wrote them, and appropriating them without permission, and without attribution, is not allowed under our copyright law. The ABAA took a strong stand against this on behalf of our members, because many of the most knowledgeable dealers, who wrote the most thoroughgoing catalog entries, were ABAA members.

But beyond that, there were more practical problems: if a dealer appropriated the listing for a first issue of, say, The Sun Also Rises, in a first issue dust jacket, what was he supposed to do when a customer asked "What's the issue point on the dust jacket?" and he didn't know? Or even worse -- and in some cases even more likely -- what if a customer asked, "What's an issue point?" or "What does 'first issue' mean?", and the "dealer" himself didn't know? This happened many times, when a lot of information was available to people who did not have the knowledge or experience to understand that information and know what it meant.

Still, for all its limitations, the Internet has meant that much more information is readily available to collectors, librarians and even booksellers than ever before. And information, in every market, is the real key to efficiency. Almost by definition, a market in which every participant has 100% of the available and relevant information will tend toward being the most efficient. And information-rich environments, or markets, tend to reward creativity, ingenuity, and innovation: if everyone knows exactly how much it costs to put a new roof on a house -- how much time it takes, how much the materials cost, etc. -- then the person who figures out how to make a higher quality roofing tile for the same price as existing ones, or an equivalent tile that costs less than current ones, will have the advantage in the marketplace. He'll be rewarded for his effort and knowledge; his customers will get the benefit of his knowledge, and the best deal possible; and the rest of the players in the market, his competitors, will be encouraged to be similarly creative or risk falling behind.

So what does all this mean for booksellers and collectors? That's not so clear right now, but we can take a few stabs at it based on what we know about markets in general, and also what we know about this market in particular.

For one thing, as I've told collectors -- and even other dealers -- numerous times in the past couple of years, we are probably going to look back on these few years before and after the turn of the century as a Golden Age for book buyers. More books have become more readily available to more people at one time than has ever been the case before or will ever be the case again. There simply aren't another 10,000 dealers out there to list their stock online, beyond the ones who have done it already. (Although here I should introduce qualifiers: Europe, South America and the Far East have been several years behind the US in this process and are just beginning to approach the kind of market saturation that has been the case here for several years. For those people whose collecting interests run to books more likely to be found in those countries than in the US, your Golden Age is just beginning.)

During this Golden Age, there has been tremendous competitive pressure driving prices downward for all but the most scarce or highly sought-after books. When there are 20 copies of a legitimately scarce book available online, the dealer who comes up with the 21st copy has a strong incentive to keep the price down, to compete with those already available: there aren't likely to be 22 customers for those 21 copies all at once, and each dealer wants to sell his own copy to the next customer who turns up. But I think we'll find that, over time, there will be more than 22 customers for that book, and that those first 21 copies, as they are sold, will not be replaced by 21 more. The number of available copies of these legitimately scarce books -- and if there are only 20 copies available of a book among the stock of 10,000 book dealers, representing the vast majority of books on the market in a country of 300 million people, we can feel safe in calling them "legitimately scarce" -- that number will tend to dwindle over time by attrition, and yet for most of the time that they'll have been available, there will have been very strong competitive pressures driving the prices down.

In a certain sense, the customer these days is in the position of the book searcher back in the days of AB Bookmans Weekly (Book searching being another niche in the book trade effectively obliterated by the Internet). In those days a searcher could advertise for a book wanted by a customer, get quotes for several, sometimes dozens, of copies and then pick the one to buy. The buyer had all the options, knew all of what was available, and could choose among them. These days each individual customer can do that. The difference today is that dealers can also see what prices books are being offered at, and they can and do condition their listings accordingly: by far the largest amount of bandwidth used by the major book database search engines involves booksellers pricing their books, not customers buying them. But the buyer has the advantage of seeing all the books available, and since there's no corresponding database of collectors on the Internet, the buyer is at a great advantage.

Of course this is only true of those books that are common enough that there isn't a sufficient demand, at a single moment, to outstrip the available supply. For those books in high demand, or truly rare, or both, the same situation works toward an opposite end: dealers who have copies of one of these high spots can see what's available and price their book competitively, but these copies will tend to cluster at the top end of the price range and, if a dealer has a particularly special or nice copy, it may be in his interest to price it higher than any other copy, so that the price itself signifies that this is a superior copy. In this way the same market forces that exert a downward pressure on prices for "medium rare" books exert an upward pressure on prices of rare books and high spots. And in a market that is, for the reason previously discussed, tending to emphasize high spots more than in the past, this upward acceleration of prices can be truly astounding.

About three years ago I wrote an article called "Trends in Modern Book Collecting" that discussed some of these phenomena and processes that I've mentioned today. One thing I said, however, bears noting because it was so wrong: The price of a fine first edition of The Catcher in the Rye had recently gone from about $1250 -- where it had been for quite some time -- to about $3000, fairly suddenly. And then, virtually overnight -- really somewhere between 6 months and a year -- it shot up to $6000. I said in my article that we could be fairly certain it would take longer to double from $6000 to $12,000 than it had taken to get from $3000 to $6000. Well, in April 2001, a copy sold at auction for about $33,000 and a month later another copy sold in the market for $35,000. These days, even fairly shabby copies of the book command over $12,000. My mistake, I believe, was in only seeing one side of how these pressures are working on the market, and in fact my guess is that as this Golden Age of book buying passes, and even medium rare books become scarcer by attrition, a similar upward pressure on prices will be felt on an ever-larger pool of books.

So far -- as promised, or threatened -- all of my examples have been from the realm of modern firsts, but I believe similar examples could be drawn from a wide range of collecting areas in the rare book world, although perhaps sometimes not quite as starkly obvious as the ones from modern firsts. In 1992 I bought a collection of Western Americana manuscripts with several other dealers, and we put out a joint catalog. Americana is a field I know very little about, and I had to do a great deal of research to do my share of the cataloging. But it was more than just learning about the material: I had to try to learn something about the Americana market, which seemed to me to work very differently from the market for modern firsts. A friend of mine, an Americana dealer, used to tell me that he would get a copy of a book, catalog it for $50, get 10 orders for it, and then when he got another copy of it -- maybe a year or two later -- he'd catalog it for $50 again. I asked him why, and he said, "Because it's a $50 book." In the field of modern firsts, where one assumes a certain volatility in the prices -- in part because the books are so modern and a few more years elapsing can make a significant difference in terms of how available copies are -- we would take 10 orders for a $50 book as a sign that our estimation of supply and demand was skewed and surmise that the price of the next copy -- especially if it were going to take a year or two to turn up -- should be much higher. I think a lot of areas of the book trade tend to work more like the Americana field than like modern firsts -- prices change more slowly because the books have a longer track record. Nonetheless, in the past decade or so we have seen books in a wide variety of fields break all previous price records.

The Frank T. Siebert collection of American Indian and American frontier books and manuscripts set a number of records that were astonishing: many of the items sold for several times what any previous copy had ever brought. Our 1992 Western Americana catalog, which was scoffed at for its high prices at the time -- three of the four dealers involved in purchasing the collection were primarily modern literature dealers, and we tended to apply our own rationales to pricing the material -- now looks almost like a bargain catalog, and prices in the Western Americana field -- especially for the truly rare material -- have escalated to unprecedented new levels, not just incrementally but often, like the Siebert collection, multiples of previous highest prices.

We've seen the same thing with manuscript archives. In 1992 shortly after she won the Nobel Prize for Literature, Nadine Gordimer's archive sold for $300,000. A few years later, the Allen Ginsberg archive sold for almost a million dollars, and last year the manuscript of one novel -- Jack Kerouac's On the Road -- sold for $2.6 million and helped put the value of the rest of his archive, which went to the New York Public Library, at an estimated $8 to $10 million. Susan Sontag's archive sold for $1.1 million a couple of months ago, and Norman Mailer's archive will reportedly be put on the market this year at an asking price of $2.5 million. So, in this field as well, the top end is reaching unprecedented levels at an ever-increasing rate. (I could here also illustrate the corollary that I discussed earlier with modern firsts: that "medium rare" material has had downward pressure on the prices, and is in many cases much harder to sell today than it was in the past; but with manuscript archives, even though that is approximately true, the forces affecting the prices are somewhat different than those I have identified for medium rare books.)

Audubon folios and quartos have broken price records repeatedly through the '90s, often by a large margin, as have the Edward Curtis North American Indian portfolios. And getting back to literature: a copy of Moby Dick sold to a dealer for $48,000 at the California book fair this year, breaking the previous price record for that book by about $12,000, and then sold to a collector for $76,000. Another is on the market at $85,000 now. And last year, at the New York Book Fair, a children's book dealer put out a copy of The Hobbit, which he had held onto for many years, at the world-record price of $50,000. It sold to a dealer instantly, who sold it a few minutes later to another dealer for $80,000, who then put it on the market at something like $125 or $150,000. Now, even so-so copies of The Hobbit, when they turn up, are priced in the $30,000 to $50,000 price range.

All of which is to illustrate that there have been changes in the market and the forces at work in it in recent years, and to try to emphasize that they are not random changes nor are they successful attempts to manipulate the market by a small group of booksellers: nobody in the field has enough clout to do that, it seems to me, or if they do then they don't need to do it. On the contrary, these are mostly changes that have come about because of a specific series of forces at work in the current marketplace that are different -- sometimes in degree, sometimes in kind -- from the forces that long governed the rare book market. Understanding those forces and the changes they have produced should help all of us -- collectors, librarians, and booksellers alike -- view this new market from a more informed perspective, and navigate it with a surer hand.

To summarize: there is far more information available today than ever before about what books are out there, what they are priced, etc. There are also far more books actually available to any given buyer than ever before, and as well there are more potential customers more readily available to dealers than has ever been the case before. All of these are combining to keep a lid on prices of scarce, but not rare, material and at the same time are helping force prices of the truly rare and/or most sought-after books to new highs.

And, if there is one last thought that can encapsulate what a lot of this boils down to, it is this: it's a Golden Age for book buyers, so buy now, buy lots; the prices are only going to go up in the future, both for the rare material and even for the medium rare material which, at the moment, is being kept somewhat artificially low. And if that sounds like what booksellers have been saying to collectors and librarians from time immemorial, I suppose that shouldn't be surprising: as a bookseller, I have to believe that's a good thing. And I hope, in elucidating how and why I believe these various factors have combined to make that true, I will have persuaded you to believe it as well.