Economics of the book

Twenty or thirty years ago the single biggest cost item in the production of an academic book was typesetting.  Printing, broadly defined, came next, and paper last.  If you don’t print many copies, you don’t use much paper.  The picture for a trade publisher has always been different: probably paper has always been the biggest item.  When I started working for a trade publisher in the mid nineties it would make me wince that we would keep resetting books to make them fit an even working.  Luckily I quickly got used to it, because, of course, by doing so we avoided having that extra sig or two, and thus reduced our paper consumption over the 10,000 or 20,000 run by an amount that dwarfed the extra expenditure on typesetting.

The academic book that prints 10,000 copies is a very rare bird.  What used to be an automatic average run of 2,000-3,000 has now become more like 500.  Of course typesetting costs have totally caved in: we now expect to pay about a third or a quarter of what we used to.  We pay less for all prepress services, so the total plant cost is likely to be getting smaller over time, even though it may often still remain the largest item on the cost list.  Printing and binding costs have moderated too, partly because of equipment efficiencies, partly standardization, and partly because publishers are now willing to make compromises in specifications which they previously resisted — like unsewn binding and groundwood paper.

The elements in the cost picture for a book consist of

1. Paying the author (usually a royalty, probably with an advance against it, but occasionally a fee).

2. Plant cost (copyediting and perhaps rewrite, design, permissions, typesetting, proofing, indexing, origination of art etc.).  Some publishers include the cost of printer’s prep and plates here.

3. PPB — paper printing and binding + freight.  PPB costs vary with quantity whereas plant costs are invariable.

4.  Discount given to booksellers and wholesalers.

5. Overhead — the cost of running the business, which has in the end to be recovered from the sale of your books.

6. Net profit — what’s left over after all the other bits have been accounted for.

One of my bosses used to make me do net profit calculations in my head.  Receipts minus royalty, minus plant cost, minus PPB equals gross margin, then subtract overhead and you’ve got net profit.  Along the way an allowance had to be made for free/unsold copies.  Better than Sudoku as a brain work-out.

As academic books move towards digital, even on-demand, manufacture, the proportion of the total cost picture that is “printing” shrinks.  Paper is likely to be come the cost driver in all types of publishing: unless we all go e-book crazy before that happens.

The days when the publisher’s raison d’être was that he fronted the money to fund the inventory required to publish, are rapidly passing.  Sure most books are still printed in volume in the first instance, but it is now possible to be a publisher without ever holding any inventory.  Several such companies exist, doing all their printing by POD, only making the book in direct response to the receipt of an order.  Their cost picture is quite different from the conventional publisher’s.  Overhead is much lower — they have no warehouse — while PPB is higher (printing one copy will always give you a higher unit cost than printing 1,000).  The barriers to entry into our industry are now almost non-existent: if you only print after you’ve sold the book, your need for capital is obviously less.

See also Costing.

A philosophy of purchasing

Of course you want to get the best deal for your employer: that’s never in question.  But in what does the best deal consist?  It’s not obvious to me that the best deal must be the cheapest deal.  Nobody should overpay, but nailing suppliers to the floor jeopardizes their future ability to serve you.  If they are out of business they aren’t going to be much use to you.  There’s a cake there which the publisher and the suppliers need to share.

When I was starting out, the cake tin was no doubt tilted a bit too far towards the printers’ side.  Book prices were rising, and publishers were able to get by without gross, gross margins.  You could tell the suppliers were doing OK by the smart lunches they would take you to.  There’s obviously been a rebalancing now, and that’s probably ultimately to the good — just so long as it doesn’t go too far.

A publisher of a certain size should by and large deal with suppliers of a certain size.  If you are a gigantic publisher, using a small printer/binder is going to cause capacity and schedule troubles.  Even if you only send them a small part of your list, you just have too much work for them to handle comfortably.  And if you can’t send a large part of your list to any one of your suppliers, you are going to end up with far too many suppliers to be able to manage them all well.  A small publisher’s few titles will be totally trivial to a large supplier, whatever they may claim.  Stick with a supplier of whose workload you can be a decent proportion.  If you can noticeably affect the business by withdrawing your titles, the chances of your failing to get good service are greatly reduced.  Of course self-destructive behavior can manifest itself in the strangest places.

There’s no ideal number of suppliers.  You obviously need enough of them to be able to get your work done properly and promptly.  The nature of you list will affect this.  No point is saying to yourself “I can’t have more than three printers” if that means none of the three can handle that little 4-color cheap and cheerful children’s list you have to produce.  It’s all so obvious, but it does conflict with today’s urge to negotiate a scale deal with suppliers who all naturally want to get a volume guarantee in return for their discounted pricing.

The single most important thing a publisher can do to ensure good service and pricing from their suppliers is to pay all invoices on time.  The second, I maintain, is merely to answer the phone when a rep calls, and to see them from time to time.  A rep has a job to do too, and being able to put your name down on a call sheet is not nothing.  Failing to see reps for companies you do not do business with is ostrich behavior.  You need to be aware of what the competition is doing, and you never know when you may need help.

You should aim to estimate high and deliver low.  Things go wrong, and when they do they tend to cost money.  For a job to come in at less than your estimated cost takes a once-in-a-liftime miracle.  Your boss will probably not allow you to add something called a”contingency”, so round up on your estimates as much as you can get away with, so that you are protected against that almost inevitable cost overrun.  I always used to tell my people that if they made a mathematical error and estimated the total cost at $3,500 rather than the $35,000 it really should have been, the editor who got that estimate would say, “Isn’t it wonderful how they really got good pricing on this one.  Most of my books have to sell at between $45 and $50.  It’s great to be able to price this one at $12.95.”  They’d never call you on the error. Every dollar you save off the estimated costs goes straight to the bottom line, and that’s where the difference between a surviving publisher and a successful publisher is laid down.

One depressing reality is however that you can always buy anything cheaper than the cheapest price you have yet been given.  Just ask and you shall receive.  But be careful: too far down this road, and you may get to a place you wish you hadn’t.

However, in a longish career producing books I have never once been thanked for bringing a book in below budget.  I have occasionally been thanked for bringing it in on time or even early, but mostly I have been thanked for making the book beautiful.  This is kind of the opposite of the message that you’d want management to be sending to staff.

POD quality

Print on demand has gotten a bad rap.  This is mostly because in the early days of digital printing, in the early nineties, publishers had not yet managed to create an archive of their digital files — many hadn’t even started using digital files.  As a result, any book which was printed digitally had to be scanned.

Scanning is the problem.  Straight text will scan OK, and may be every bit as good as the offset printing.  But it can never be better than the original, so if your book was printed on groundwood with tiny flecks of bark in it, these will be scanned too and will show up as an annoying distraction.  It’s when we get to halftones that the trouble starts:  a good scan will lose a lot of detail.  Badly printed originals will tend to scan as shaddowy copies like you might get off the office copier when it was running low on toner.  And flat tints will be awful.  The scanner has two options: “I see it” or “I don’t see it”.  When it gets into a finely screened background tint it will miss many of the dots in the screened original.  You may get a strong moiré pattern or just a blotchy mess.  Thin rules will suffer from the same problem, and appear as randomly broken lines.

This all sounds awful, and often it was (is — because lots of old books still remain to be scanned).  But the proportion of digitally printed books which are made from scans is decreasing all the time, as publishers get more and more of their books created digitally and have an archived copy of the PDF.

With the PDF used for the offset edition, the digital version will be equally good.  I won’t say indistinguishable, because in an unexpected development, there’s a strong possiblity that the digitally printed halftones will actually be better than the offset versions.  A digital print engine has no dot gain, and the halftones will display detail in the shadow area which has plugged in the offset version.  It’s true the digital halftones will tend to “burn out” in the highlights, but mostly the result is superior.  I have several examples to prove this point.

Printed in . . .

For international trade we need to display the place of origin.  Conventionally we put this on the copyright page, but it would be fine on, say, the last page of the book.  Books used to get stopped at customs for failure to indicate place of origin.  This doesn’t seem to happen any longer.  Maybe we’ve become better at remembering to include it, or the law has changed, or maybe it is still happening and I’m just not hearing about it.

Books printed in Britain must by law carry the name of the printer and (I think) typesetter.  In America this is not a concern: our books can happily be imported into the UK without any such detail.  The reason for this is that the printer is jointly liable for any obscenity or libel in a book which they have printed.  So along with the publisher they can be sued in Britain by an aggrieved party.  It was well into my working life that the University Printing House in Cambridge first printed a book containing that most notorious of four-letter words, ****.

A CUP of coffee, 2

At our boss’s coffee mornings we used to jockey for the manuscripts from those editors known to be more sensible, less trouble.  Editors would also play the same game, doing what they could before the meeting to lobby for this or that production controller for their latest ms.  Being at that time a part-time editor, part-time production controller, I used to be able to secure all the manuscripts of that most cooperative of editors, myself.  I was able to keep a foot in both these camps because of the strange circumstances surrounding the firing of Tony, the best production controller CUP ever had.  This was indicated by his being chosen to work on The Cambridge History of Iran, a project who’s in-house editor was also the head of Production and thus Tony’s boss.  One Friday Tony took the maps for the latest volume of Iran home with him so he could work on them over the weekend.  He stopped at a colleagues’ house on the way as he was having a house warming party.  Convivial Tony left without the maps which were gnawed in the night by mice.  Our passionate boss fired Tony for this irresponsibility before he remembered the hiring freeze that had recently been declared.  The most dispensable editor was clearly me, so I was moved into production while retaining those parts of my editorial duties which nobody else wanted to take on.

Hard at work

Across from Sue was Cheryl, my assistant, and facing her across my table – for I had disdained a desk – I would sit reading my Times.  Here’s a portrait by Jack Bowles, one of our designers, and a fast friend, showing me in action, flowing locks, platform soles, bell bottoms – looking pretty cool.  Behind and across from me sat John and Arthur, my fellow production controllers and another assistant.  My recollection is that we talked and laughed all the livelong day.

The way a book was produced was governed by a book of rules generated by a Work-Flow Committee made up of representatives of all departments and interest groups.  What a copyeditor does to a manuscript is in part determined by what decisions the designer makes, and the decisions the designer makes are to some extent determined by what the copyeditor does.  The Work-Flow Committee solved this dilemma in true committee style by having the manuscript go through the design and copyediting departments twice: so each could regard themselves as getting it first at least once.  Once the manuscript had traipsed its way to our coffee morning, the production controller would send it out for an estimate to the printer indicated by the boss, notionally in discussion with us all.  This you’d do by writing a letter.  “Dear Mr Williamson” – we were still pretty formal in the early seventies.  The manuscript, along with the type specs written out by the designer, and a sheaf of notes from the copyeditor, would be mailed to the printer.  You’d then wait.  After about four weeks you might write another letter asking if they’d received the manuscript, and whether they felt they’d be able to cast-off and estimate anytime soon.  After another couple of weeks you’d probably telephone and be told: “Oh yes, Mr Hollick, don’t you worry. I have the manuscript in my in-box, and hope to get to it very soon”.  When eventually we did get the estimate for typesetting and printing we’d give it to Derek and he’d do a working, showing what the entire job would cost.  This would go to the editor who would use it to get a decision on price and print quantity.  The Production Controller supposedly had some responsibility for the numbers, but because we never did the calculations we never had to confront the answer, so our cost-control activities were notional rather than real.  We just took what we got; and why not – changing the printer would have meant going back to the coffee morning and in effect telling your boss he’d made a mistake first time round.  And it was well known throughout the industry that we’d take whatever we were given, and that you could thus get an extra couple of pounds a page out of us.  Diffusion of control, exemplified by the democratic efforts of the Work-Flow Committee, while admirable in many ways, did unfortunately have the effect of increasing our costs.  The whole system was back to front: you found out what the most audacious supplier dared say it would cost, and based on that, figured out what the price and print quantity of the book would be.  A financially sound system would decide how many copies to print at what retail price, and based on that come up with a budget for the Production Controller to spend.

A CUP of coffee, 1

We all sat in a huge ground-floor room; one where they’d obviously had to give up on the problem of how to fit in partitions.  Derek sat in the far left corner, bounded by Trumpington Street and Mill Lane.  His desk had the aura of being surrounded with stuff, but actually there really wasn’t anything there except a few chairs and some paper.  He sat in front of an enormous adding machine: he did all our calculations.

At the other corner on the Trumpington Street side of the room sat Sue, who did reprints.  A good-looking girl, she seemed to gleam in the sun which poured in the window all morning.  An editor once complained to our boss that he couldn’t concentrate on his work because Sue tended to wear low-cut blouses, and the sight of her tickled his libido so much that the joys of secondary school geography could no longer compete.  Typically of the time Sue was instructed to button up.  These were after all the days, the early seventies, when women were first being allowed to wear trousers to work. Makes the heart swell to have lived through such times of revolutionary social change.

Really revolutionary was the fact that we actually had to employ someone to do reprints.  Prior to the seventies the idea that you might be forced into a reprint was one of those things that would keep an academic publisher awake at night.  When books were printed by letterpress you needed to set the metal type up into pages, and then impose these pages into large, impressively heavy, formes which would be inked, and have a sheet of paper pressed against them.  After printing enough copies, the type would be dissed (distributed) and the metal reused for other jobs.  Nobody could afford to keep all their books standing in type: the cost of the metal and the acres and acres of warehouses that would have been required made that unthinkable.  So if you had to reprint, you had to reset the book, which made the whole operation about as cumbersome as printing the book in the first place.  Now, as at that time we used to feel proud of ourselves if we were able to bring out a new book within a year, clearly reprinting was a serious commitment both of money and time.  Therefore it was rarely done.  (The need to completely reset the book each time it was printed is thus, obviously, what explains that delight of bibliophiles: typographical errors that appear in one “edition” and not in another.)

The University Printing House, where we placed most of our work, was living off a glorious tradition.  Under the guidance of Stanley Morison they had become the country’s premier book printer.  Letterpress printing had risen at the Press to heights rarely dreamed of since Gutenberg, and now three or four decades later, they knew that they knew how to print by letterpress.  And they knew that they knew that offset lithography was an inferior printing technology, which nobody but a jobbing printer would contemplate using.  Offset, in a nice instance of technological determinism, is what made reprinting viable, and led us away from the world where your aim on first printing was to print as many copies as you thought you could ever sell, to the current situation where the first printing can almost become a test marketing shot.  As the UPH’s largest customer, we tended to turn our corporate backs on offset too.  The Printing House did run some tests for us, and were able to prove to everyone’s satisfaction that they were right: offset could be printed horribly.

So we were a letterpress operation, with a few odd exceptions.  I did get to go down to Gloucestershire to see a job on press which was being printed by collotype: not too many people can say that.  Collotype involved eggs and sunshine, believe it or not.  But most of what we did in those days involved the indentation caused by impression of ink onto paper.  Once, or was it twice, a week we’d all meet in our boss’ office, have a cup of coffee, share out the new manuscripts, and report on progress on the books in production.  I remember being asked “Have you ordered the blocks for McElhinney’s colour plates?”  I replied that they were being printed by “litho”.  “That’s not what I asked.  Have you ordered the blocks?”.  In the end I just accepted the rebuke, and promised to make sure blocks were available at the appropriate time.

For those who don’t get the “joke”, blocks, known as cuts in USA, are the plates of metal onto which art, line or halftone, was engraved so that pictures could be printed by letterpress.  One of offset lithography’s advantages is that the art is just photographed and exposed onto the photosensitive plate, along with the text, where it prints by virtue of the antipathy of grease and water.  Offset lithography, normally called offset in USA or occasionally litho (rhymes in America with “pith oh”, sort of) was commonly called litho in England, where it sounds like “Lie so” said with a lisp.

Singing the blues

Nothing beats the surge you get from coming on an old book folder that has a sig or two of blues in it.  That ammoniac smell takes you right back.  It is redolent of the successful completion of yet another job.

Blues, commonly called ozalids in UK, (Ozalid is actually a trademark in the USA) were the last proof you used to get before printing a book.  They were a cheap print made from the stripped-up negatives of the book (flats).  The flat was exposed to a sensitized paper, quickly “developed” and sent to the publisher as a proof that all was OK before plates were made from the same flat.  The paper was a creamy color and the type and illustrations appeared blue (the UK ozalid tended to be the opposite, a negative of the flat.  This was because of the right reading/wrong reading negative difference between the two sides of the Atlantic).

I always used to say that nobody with the word editor in their job title should ever be allowed to see a blue.  Editors regard a proof without any marks on it as a reproach — an indication that they were too sloppy to find the errors.  Of course by that stage in a book’s production, any correction would cost a fortune.  Thus it is probably only old production people who get weepy at the scent of a blue.

The Printed Picture

The Printed Picture by Richard Benson.  344pp, 8″ x 10-1/2″, 978-0-87070-721-6, The Museum of Modern Art, $60 (but Amazon is offering it at $37.80).

This is a brilliant book.  It’s one of those that made me say “I wouldn’t mind if I spent the rest of my life reading this book”.

Richard Benson is a born teacher, a photographer, a collector of printed pieces, a fan, especially of offset lithography, and former Dean of the Yale School of Art.  He writes beautifully and every page illustrates his immense enthusiasm for his subject.  The book works as a series of two-page spreads, each one discussing a particular printed piece.  We start with hand prints on cave walls and move through woodblocks, hand lettered pages, engravings, etchings, aquatints, mezzotints, stencils, silk screen prints, all manner of photographs, collotype, offset, etc, etc, to digital printing.  Centrally it is a study of the evolution of the printing of photographs.  The book contains an eight-page introduction to color theory, which any one of us can learn from.  Many of the pieces illustrated are beautiful in their own right, while others may be mundane — the sort of printed piece we’d all tend to throw away, but which nevertheless has its own range of technical problems magnificently overcome, which it takes someone with the right training and empathy to pick out and explain.

One of the many specially striking pieces shows a wood-engraved photograph from the last decade of the nineteenth century.  The photo was exposed onto a wood block which was then engraved by a craftsman/artist, and printed from a stereo by a steam-driven letterpress machine.  The result is beautiful: it’s like a hyperrealist painting in a way.  This one covers four pages, the second spread being given over to a detail of the eyes and nose of the subject, Frederick Law Olmsted.  The workmanship is amazing: all done by hand with a magnifying glass and a burin.

The book is beautifully printed by 4 color offset by GHP, West Haven, Connecticut on 100lb. Value Silk.  The challenges of printing images which were themselves originally printed (often excellently) by a wide range of different methods is triumphantly overcome.  It’s smythe sewn in a case made of matte laminated paper over boards, with a square back with board in spine.  Buy one, you won’t regret it.

Ink making

This You Tube video is from Canadian ink maker, The Printing Ink Company of Vaughan, Ontario

“This is my book”

1.  “This is my book” = I am the author.

2. “This is my book” = I own it, I bought a copy.

3. “This is my book” = I am the publisher — I paid to create it.

4.  “This is my book” = I work for the publisher/printer/typesetter/literary agent responsible for creating this book (perhaps “our” would be more usual than “my” here).

5.  “This is my book” = This e-book is on the Kindle/iPhone/computer/e-reader that I own.

Does this tell us anything about the “book”?

Nos. 1 & 2 point to the difference between book as a container of ideas/culture, and book as a physical manifestation of this culture carrier.  War and Peace may be your favorite book, but the ratty paperback you have at home may be a horrid object. Maybe Tolstoy preferred Anna Karenina.  Maybe the most handsome book you have is that Corpus of Roman Coins, but you have of course absolutely no intention of reading it.

No. 3 focusses on the economics of the business and is likely to mean the book as physical object (one that can be sold) rather than 1, though when it comes to handing out Pulitzers, Bookers etc. the publisher will inevitably want to connect to 1 more than usual.  It goes without saying, but of course I have to say it, that without a solid economic rationale none of these different “books” would exist (I’m willing to argue about No. 1 too).

Nos. 2 & 4 seem to be pointing to the same physical object, but if you were the editor or the agent, you may be thinking of No. 1.

Is No. 5 more like 1 or 2?  It sort of straddles them both.  I suppose we do actually “own” a free download from Project Gutenberg, but it’s not quite the same relationship as those bits of paper and board on the shelf.

Maybe philosophers and linguists have sorted all this out, and at the end of the day, we all know what we mean.  But there are instances, and will be more of them in future, when we need to be clear, at least in our own mind, which of these we really want to mean when we speak about a book.

(Based on a conversation with Mike Rosenhack)