Some things that were true about publishing for decades aren’t true anymore
Back when my father, Leonard Shatzkin, was active with significant publishers — the quarter century following World War II — he observed that very few books actually took in less cash than they required. That is not to say that publishers saw most books as “profitable”. Indeed, they didn’t. They placed an overhead charge of 25% or 30% or more on each book so most looked unprofitable. But that didn’t change the fact that the cash expended to publish just about every book was less than the cash it brought back in.
The exceptions were usually attributable to a large commercial error, most commonly paying too much of an advance to the author or printing far more copies than were needed. But, absent that kind of mistake, just about every book brought back somewhat more revenue than it required to publish it.
This led Len to the conclusion that the best strategy for a publisher was to issue as many titles as the organizational structure would allow. That was a lesson he passed along to the next generation of publishing leadership that came under his influence. And the leading proponent of that business philosophy was Tom McCormack, who worked for Len at Doubleday in the late 1950s, then went on to Harper & Row before he ascended to the presidency of then-tiny St. Martin’s Press in 1969. Tom often credited the insight that publishing more books was the path to commercial success as a key component of the enormous growth he piloted at St. Martin’s over three decades.
(I checked in with Tom, who is long-retired as a publishing executive but a very active playwright, about how many books didn’t claw back the cash expended. He told me that his “non-confirmable recollection” is that the percentage that did at least get their money back ranged from 85% to 92%. He recalls “incredulity” from his counterparts in other houses, whom he believes simply couldn’t “wrap their minds around the meaning of the statistic: revenues minus disbursements.” He went on to tell me that this number “seemed effectively irrelevant to them. They had an overriding and deeply flawed notion of something they called title-profitability. They thought they were analyzing the profitability of a title with their ‘p&l’.”)
Despite the apparent immutability of the fact at the time that most titles brought in incremental margin, many publishers who were losing money would come to the opposite conclusion. They would decide they should cut their lists, pay more attention to the titles they published, and create more profits that way. I remember discussing the futility of that approach in the 1980s with my friend and client, Dick McCullough, who was at that time the head of sales at Wiley. When I observed that the publishing graveyard was littered with the bones of publishers who pursued cutting their lists as the path to profits, Dick said of their efforts to cut “yes, and very successfully too”.
I got another lesson about this reality in the late 1980s when a company I consulted to (Proteus Books) sued its distributor (Cherry Lane Music) for a failure of “due skill and competence” in the sales efforts for Proteus Books. One of Proteus’s expert witnesses was Arthur Stiles, who had been Sales Director at several companies, including Doubleday, Lippincott, and Harper & Row. Stiles confirmed that big and competent publishers routinely put out thousands of copies of titles in advance of publication, with extremely few failures in terms of getting the initial placements. He was testifying in a time that was still like what my father experienced: the industry’s title counts were growing, but so were the the number of bookstores in which they could be placed.
Those days are over. And, coupled with the ebook revolution, the implications of that are profound.
A few things happened to change the environment so that it became no longer true that even big publishers could get all the distribution they needed on every title to assure a positive return of cash.
1. The title output of the industry has grown enormously. In the 1960s, the total output of the industry was in the neighborhood of 10,000 titles a year. Now it is something more than 30 times that number published traditionally, with a multiple of that number being self-published. Each new book is competing against more new titles every two weeks than a book fifty years ago would have competed against in a year!
2. Nothing published ever dies. Fifty years ago, stores were smaller and, while there’s no easy way for me to measure this, I’d guess that the active backlist across publishers was probably no more than 25,000 titles. Superstore growth in the 1980s, the efficiency of Ingram as a national wholesaler, and computer systems that helped stores track their inventory and sales fueled backlist expansion. Even in the early 1990s, the total of truly competitive titles was probably in the low six figures. But then came Amazon’s unlimited shelf space and Ingram’s Lightning Print to deliver one copy at a time, and, even before ebooks, the competitive set of available titles had probably jumped to seven figures.
3. Bookstore shelf space is declining. Nobody who has been reading this blog needs much elaboration on that point.
What that means is that a list-cutting therapy that McCullough and I saw in the 1980s as suicidal and which McCormack explained repeatedly was folly is no longer crazy. (Oh, how I wish my dear departed Dad was around to discuss this with!) And the new conjecture in this blogpost is that the day might come when a publisher with an extensive backlist might decide that the most profitable path would be to hardly publish any new titles at all!
The portfolio of any longstanding publisher today contains a lot of backlist which is pure profitable gold in the ebook era. Contracts often give publishers the rights to a book for the life of copyright if they continue to sell it. (I’ll confess here that there is a caveat to this point coming up in an italicized postscript below.) So a major publisher doing $600 million and up (of which there are six), almost certainly has triple-digit millions of sales in its backlist, which is increasingly shifting to digital. Even the most sober industry observers are seeing revenues exceeding 50% from ebooks in the next two or three years, which would mean that substantially more than half the units of these books are selling electronically.
So, let’s say you’ve got a company doing a billion dollars in annual revenue and barely eeking out a profit or perhaps even losing money. With a strategy of continuing to publish what you own as ebooks, you can see digital backlist revenue of $150 million, decaying by 10% a year, with gross margins giving you $100 million or more in cash flow. Offloading all the print operations for which you own rights to a distributor or competitor will provide incremental revenue as well. (You only need help for the offline print sales. Getting the online sales requires no operational capability.) You’d then need a minimal organization to do some marketing (not a lot), sign up and put out some additional titles that would be chosen for being risk-free (not a lot), and to handle the administration and royalty processing for your thousands of contracts. Five or ten million ought to cover those costs very handily.
Of course, the other thing you could do is sell your rights to that backlist. But I think it would require somebody to overpay in relation to your net discounted cash flow to make that attractive because the costs of keeping it all for yourself would be so minimal.
One hopes that today’s publishers are looking at the simple statistic Len and Tom authored: revenues minus disbursements by title. No doubt today’s biggest publishers are looking carefully at the performance of their copyrights in a way that sorts the new titles from the backlist. But doing so is only useful if they’re apportioning their costs properly across the title base. If they are, what is described in this post will be evident if and when it is true. In the meantime, careful focus on new title acquisitions and accepting that the healthiest way to manage for the future might be to reduce the commitment to new title development will have to replace the clear truths that guided smart publishing strategy for previous generations.
The history and analysis are all valid, but there is one big monkey wrench in this scenario I’ve sketched. There is a provision in the 1978 copyright law that allows authors to reclaim rights to their books after 35 years. Titles published in 1978 become eligible for reversion, called “recapture” apparently, starting in 2013. (With logic that is ironically typical of what Congress does when it touches copyright law, older titles are on a slower track for liberation.) Agents are planning for this; publishers will have to deal with it. I am given to understand that publishers can only retain these books for life of copyright by, in effect, reacquiring them. (Should be lots of fun!)
So, in fact, the backlist attrition might be faster than 10% (but it might not, because ebooks may create more readers for backlist than we had before as well.)
It is also true that many publishers have already been moving in the direction I suggest: pruning their new title counts and being particularly cautious with midlist. Of course, there was a conviction by many that list-pruning was a good strategy even before it actually was a good strategy, but the execution of it has been much more rigorous over the past decade.