A Novel Example

Yesterday’s post on the problem with novels was initially badly garbled due to an html coding error. It’s fixed now, but since the original version managed to confuse some people, let me offer an example (which you can also find in the comments on the original post).

Light in August, which has already been written, is available for about $8. It’s worth $9 to me. If I download a copy, $9 worth of social gain is created (a $1 gain for me, and an $8 gain for the estate of William Faulkner).

Now a contemporary novelist, say Sara Gruen (to pick a very good one) comes along and realizes that with $8.50 worth of effort, she can write a book for which I’m willing to pay $10. By offering it at $8.99, she induces me to read her book instead of Faulkner’s, and clears a 49 cent profit. Total gain: $1.01 for me, and $.49 for Sara, or $1.50.

(Of course Sara Gruen does not write books just for me, but she might well expend $8500 worth of effort to write a book for 1000 people like me, whereupon all the numbers scale up.)

In other words, by writing a very good novel, Ms. Gruen reduces the social gain from my reading habits from $8 down to $1.50.

That’s exactly the sort of thing that economists generally believe should be taxed. In fact, a perfectly analogous argument constitutes the entire case for taxing polluters.

So I repeat my question: If we want to tax polluters to discourage the production of, say, socially inefficient aerosols, why shouldn’t we tax Sara Gruen to discourage the production of socially inefficient novels?


55 Responses to “A Novel Example”

  1. 1 1 RJ

    When we tax the polluter, we’re declaring that his neighbors have a property right to carbon-free air. If we tax Sara Gruen, we’re declaring that the Faulkner estate has a property right to the attention of potential readers.

    Faulkner has a property right to my attention?

    My reading interests are rivalrous and excludable, it’s a private resource that producers must compete for; the author who writes the best novel gets my money and my fanbase. However, air is non-rival and non-excludable, it is a public good and thus a company that causes irreversible damage to the ozone/fresh air from excessive use.

  2. 2 2 RJ

    *sorry, html problems on my end too…

    …and thus a company that causes irreversible damage to the ozone/fresh air from excessive use is causing negative externalities from free-riding.

  3. 3 3 Jim W K

    When you compare the average reading time most of us afford ourselves in a lifetime against the time it would take to read all the excellent books ever written, it’s clear that there are many more excellent novels in the world than you’ll ever get a chance to read. So yes, there are enough books in the world already to keep every one of us replete with good reading material, which in market terms suggests the world doesn’t need any more books written.

    Let’s say on that basis we stopped producing works of fiction from now. There is a caveat that must be brouht to bear – namely, for how long would we keep this up? Because humankind would eventually reach a point at which all books were so old that there would be nothing in the fiction market to which those with a contemporary backdrop could relate.

  4. 4 4 suckmydictum

    When we choose to tax polluters, the goal of that legislation might run twofold:

    1. The tax is meant is decrease the amount polluters pollute (e.g. directly curb CO2 emissions).

    More subtly,

    2. The tax is meant to incentivize innovation so that polluters will avoid the tax by improving efficiency and pollute less by that means.

    Any tax on authorial externalities would be motivated exclusively by 1 since only by curbing new books being written will you mitigate the type of social loss SL is talking about. The analogous motive behind 2 does not hold for authors since authors can’t innovate to create books more efficiently. There are substitutes for dirty pollution which a carbon tax might encourage, but this is not the case for an author tax.

  5. 5 5 Steve White

    Your consumer surplus went up — like we expect since that is your choice. So the inefficiency is on the supply side. What is it?

    It’s that they both have some monopoly power. In a perfectly competitive market there wouldn’t be any profit so there couldn’t be a difference in the producers surplus.

    But every Econ 101 textbook points out this business-stealing externality and notes that in general situations where monopolies mark up prices (MC<P) have a deadweight loss you could (very) theoretically reduce with price controls or subsidies.

    So I still don't think there is anything new here. The only reason we don't subsidize the books is political economy and information constraints.

  6. 6 6 Lucas

    Are we sure Sara Gruen considers writing a novel an “effort”? Many artists claim making art is, if not enjoyable, necessary for them. Maybe Gruen thinks writing a novel is as much of an effort as breathing (preferably clean air).

  7. 7 7 Daniel


    Good call, Landsburg is ignoring that writers might find writing enjoyable and preferable to leisure time.


    What happens to your argument if it costs Sarah Gruen $0 to produce her novel? I think this is a legitimate explanation for why we have so many “also-rans”, that give their free-time without guarantee of a payoff.

  8. 8 8 Al V.

    I’m not sure I agree with the premise here. Your numbers apply to you, but I would guess that the market for contemporary novels is not mostly people like you.

    I read “Water for Elephants” after my sister-in-law recommended it. She mostly reads popular fiction, and bought “Water for Elephants” in hard cover, so I would estimate that she valued it at $30. Total gain: $3.50 to Sarah Gruen (assuming that she receives 50% of the purchase price), $12 less costs to the publisher, and $6 to my sister-in-law.

    My sister-in-law might value “Light in August” at $3, and therefore would not purchase it unless it was available as a free ebook.

  9. 9 9 Harold

    Jim WK. I(and others) mentioned this yesterday – I think there is some social value that is not captured only by entertainment of the reader. We could call this a positive externality perhaps? So even thought there are enough good books already in existence, we would not want to stop producing new ones.

    I liked David O’Cushman’s analogy with the water and orange juice. (Although there is a difference because orange groves would have to eventually turn into springs in the case of books.) We have an existing resource that costs almost nothing to reproduce, but lies in the hands of monopolists. Since a monopoly is an imperfect market, should we be intervening (with a tax on new books) to further protect the monopoly? Why should the Faulkner estate get all the social benefit? We can argue that the market failure is the high price of the existing book – I think this was comment #2 yesterday, perhaps we could have saved ourselves some time. Out of copyright books are quite cheap in e-format.

    To put it another way, the purpose if these proposed raxes is to correct a market failure, but since copyright causes a market failure to begin with, extra taxes are not going to correct it.

    If we drastically reduced the length of copyright to say 5 years, we would be approaching a better market. Anyone could produce and sell 5 year old books at very little cost. This would make 5 year old books very cheap, new books cheaper and would reduce the motivation to create more new books, which was the whole purpose anyway. Watch my lips – no new taxes!

    There is no similar situation with carbon emissions, so no comparison

  10. 10 10 John

    I guess the one thing that I didn’t quite understand is that this isn’t really about intellectual property; it’s about all new products. So the argument from yesterday could more generally be thought of as how do you justify taxing polluters without taxing all new products.

  11. 11 11 nobody.really

    1. Landsburg’s analysis seems oddly static, or oddly incomplete. The Faulker estate is selling a product with no marginal cost for $8, and is finding that people are switching to different products. Why doesn’t the Faulkner estate reduce its price? Presumably they have determined that their product is NOT a perfect substitute for other products (and that Landsburg is not representative of the buyers of their product), and that they can maximize their monopoly profits at a price of $8.

    As others have remarked, the arguable market failure is the failure to price the Faulkner book at its marginal cost. If we wanted to remedy this situation with taxation, arguably we should tax the profits of the Faulkner estate. If we set the taxes at 100%, the Faulkner estate might well treat his books as if they were in the public domain.

    (Of course, we face this problem will many aspects of intellectual property: The property can be expensive to create and almost costless to replicate. The producers of this property must find some way to recoup their investments, which inevitably involves pricing above marginal cost. Landsburg had proposed a kind of public financing for patents that might address this situation.)

    2. Given this analysis, Landsburg’s analogy breaks down. Landsburg analogizes the Faulkner estate to someone being harmed by Gruen’s book. To remedy the market imperfections here, I propose taxing the Faulkner estate, not Gruen. To remedy pollution, I propose taxing Gruen, not the Faulkner estate. Ergo, no analogy.

  12. 12 12 Jon Sealy

    As a novelist, this discussion is quite interesting to me. I won’t quibble that being a novelist might be socially useless, but I do think this reasoning could be broadened to quite a few other realms. You could make the argument that we’ve produced enough music, TV shows, movies, and introductory economics textbooks to last us a lifetime as well. Do we really need a remake of Robocop? Should we be taxing actors, filmmakers, musicians, textbook authors, et cetera?

    Here’s what you seem to be arguing: Social Gain = Price + Consumer Value – Production Cost

    In the case of Faulkner, Social Gain = $8 price + $1 consumer value – $0 production = $9.

    In the case of Gruen, Social Gain = $8.99 price + $1.01 consumer value – $8.50 production = $1.50.

    Maybe the production loss in the case of Gruen is an inefficiency, but I’m not sure it’s the same kind of externality as carbon emissions. You seem to be conflating two kinds of costs: the cost of Gruen’s labor (which is not an externality) and the cost of market share to Faulkner’s estate (which implies that market share is a kind of property, and that the right to market share can be infringed on).

    Regarding the social cost of labor to produce a novel, I think most novelists are working 40 hours a week at something else (teaching or copyediting or what-have-you). It’s not like Gruen would otherwise be curing cancer. Instead, she’d likely be screwing around on Facebook or watching TV. Maybe you can measure a cost of advertising dollars lost by her not watching so much TV, but the market does a pretty good job on its own of keeping too many novelists from making a fine living. In fact, there probably is an exterminator out there who is working on a novel right now, and if said exterminator publishes that novel, he’ll go on being an exterminator. The “cost” of his labor is not an externality, any more than the cost of watching TV in the evening.

    The externality you’re trying to tackle is the loss of Faulkner’s market share. In other words, why do I have a property-right to breathe carbon-free air whereas Faulkner’s estate does not have a property-right to market share? The property-right to breathe carbon-free air is the property-right to a healthy life, which came first in Jefferson’s list of innate rights. Regarding market share, it’s hard even to consider how that can be conceived of as “property,” because to do so would undercut everything about the capitalist system.

    For instance, why create a new social media site, when doing so would cut into Facebook’s market share? The obvious answer is that the new one would be better, or it would fail. Same goes, in theory, to a new novel. You’re free to go out and buy Faulkner; Gruen is just giving you the choice. Other commenters have already said this, but it really makes no sense to be arguing that Faulkner’s estate somehow has an innate property right to potential readers.

  13. 13 13 Jim W K

    Yes, to echo a point made above – one thing Steve’s analysis omits is that many writers enjoy writing to the extent that even if they remained unpublished the projects would still be intrinsically rewarding enough to confer net benefits on the writer.

  14. 14 14 Daniel


    Maybe it would help if you drew out some demand and supply curves to illustrate the point you’re trying to make. It seems to me that classic novels should have a perfectly elastic supply curve. Describe, via graph, how you can increase producer surplus by shifting the demand for classic novels to the right by increasing the the cost of its substitute.

  15. 15 15 Neil

    If faulkner’s book is worth $9 to you and costs $8, that presumably is true before gruen writes a new book. Why haven’t you downloaded it? It must be the case that before gruen writes her book, faulkner’s book gave you less than $8 value, say $7.99. so you left it unread. (Your whole point depends on the fact that there are unread old books–why is that?)

    Since the marginal cost of faulkner is zero, it is the faulkner estate (not gruen) who is depriving you of this $7.99 in value. If “old” books were priced at their marginal cost (zero) this problem would not arise. The value of the marginal unread old book would then be equal to the cost of your time, and the highest price gruen can fetch for writing a new one would equal its novelty value over the marginal old book. There is no externality in writing new books to be taxed. (In fact, I would argue, with this pricing situation there is too little incentive to write new books, but I won’t go into that unless someone wants to argue the point.)

  16. 16 16 Lawrence Kesteloot

    Doesn’t this assume that everyone who buys Gruen’s books would (in the absence of those books) buy Faulkner instead? That’s a pretty big assumption. Many people don’t like Faulkner.

  17. 17 17 Ben

    Therefore, any new companies that, by their entry into the marketplace, cause competitors to lose non-marginal-cost business should be taxed (if one believes in taxing polluters).

    For instance, company A spent X dollars on R&D, resulting in a product with $2 marginal cost and $8 amortization of R&D for a price of $10. It sells well because consumers value it at $11. Before the R&D cost has been amortized, company B just repurposes its highly-capable existing machinery and can manage to offer a slightly better product for $10.50 marginal cost, and then prices it at $10.99. Consumers substitute B’s product because they value it at $12. Company B reduces the social gain of consumers’ buying habits from $8 ($1 for consumers, $8 for company A) to $1.50 ($1.01 for consumers, $0.49 for company B). Therefore company B should be taxed at $6.50 per item.

    That’s clearly insane. So what’s wrong? This situation simply won’t happen from a game theory perspective. As soon as B makes its product available, A will drop its price by $0.02 and consumers will all switch back to A. This will go back and forth until A’s product is $0.50 cheaper than its original price and B will stop making its product. But B should foresee this before repurposing its machines, so B will never enter the market in the first place.

    Back in books, when an author releases a new book, other authors will simply drop their prices in response to the new entry. Knowing that will happen, if the new author still decides to write his book in the first place, it must be adding social value. The analog is not true with pollution — private and public incentives are aligned in the case of books, but they aren’t aligned in the case of pollution.

  18. 18 18 Daniel

    @ Neil,

    Right, Landsburg’s whole premise is fixed around the fact that books are a monopoly good, which we know produces deadweight loss, unless Landsburg wants to suggest some model by which the monopolist can have perfect price discrimination. That’s why I want him to draw out the graphs to explain what he means. I think if you’re shifting the demand curve to the left under monopoly conditions you’d be reducing the deadweight loss.

  19. 19 19 Neil


    Steve’s whole premise is fixed around the assumption that there is an old book that he would have bought had it not been for greun writing a new book. He is assuming some form of disequilibrium.

    Think of my desire to read old books over my lifetime as a demand curve, where the books are arranged from those which give me the highest value to those which give me the lowest. I have a marginal cost curve rising as I read more books. The equilibrium number of old books I will read over my lifetime is determined by the intersection of this demand curve and the mc curve.

    If old books are priced at zero, the rising marginal cost curve solely reflects my cost of time (so many books, so little time)and an efficient outcome is reached. I read old books up to the point where the value of reading is equal to the cost of time spent reading. The marginal old book has value to me exactly equal to my cost of time. If the old books have a positive price, I will read fewer old books and there is a deadweight loss. But it is the pricing of the old books that causes this deadweight loss, not gruen’s new book.

  20. 20 20 Kirk

    As a reader of books from all era’s (especially science fiction) I can state unequivocally that there are NOT enough books to satisfy someone for their lifetime. Anyone who has read novels from different eras can see extremely distinct writing styles and qualities that make some books great for some days, and older books better for others. Anyone who has read Twilight or Harry Potter (two books I HATE, for the record) can see that those novels fill a niche that no other novel from the past has. Examples abound of new styles, themes and ideas that fit far better for today’s reader.

    Sorry my argument is not economics related, but as an avid reader, I could not remain silent.

  21. 21 21 Daniel


    I agree with you, I just don’t know how landsburg could have made this simple error in basic micro and I want him to draw out the graphs in case he means something different.

  22. 22 22 Neil


    At this point I am not assuming Steve has made any error. In fact, I am expecting him to come back and devastate my argument. And if he does, I will learn something.

  23. 23 23 Doctor Memory

    While perhaps it would be a better world if it were so, nobody has ever had their lifespan significantly reduced, or required additional medical care as a result of reading a bad novel. (Nor have their crops failed nor any other value been extracted from them other than the time required to read the book.) Sometimes analogies obscure more than they reveal.

  24. 24 24 ThomasBayes

    For me, this is a fascinating post for the following reasons:

    1. Steven Landsburg is making a point that has never occurred to me and probably never would have had he not written this post.

    2. Although many of the commenters appear to believe so, I don’t think that the main point of this post is about books and the people who read and write them.

    3. I believe that Steven Landsburg HAS given considerable thought to this issue, and, in doing so, he HAS NOT overlooked many of the ‘obvious’ flaws in the argument.

    4. Many of the commenters appear to believe that Steven Landsburg HAS NOT given considerable thought to this issue, and, in doing so, he HAS overlooked many of the obvious flaws in the argument.

    Perhaps those commenters (item 4) are correct, but past evidence has taught me not to jump there too quickly.

  25. 25 25 Dirk

    I agree with many of the points already, but one point I didn’t see already was how to determine social gain.

    I can definitely agree that when you want to buy something costing $8 and are willing to $9, that there is a social gain of $1. You bought a product or service that gives you tangible benefits.

    On the producer side, Landsburg considers the profit the social gain. I can’t agree with this assumption. Money’s only benefit is the ability to buy things in the future. You can’t eat it, make cloths with it, or build a house out of it. The producer will only receive social gain when (s)he turns around and become a consumer.

    By contrast a polluter is producing both social gain and loss. The loss in this case is being excluding during the transaction, which is why governments try to regulate this cost back in. I see no such loss being produced by either book publisher.

  26. 26 26 Daniel

    @ThomasBayes #4 & Neil,

    Right, that’s why I think a graphical representation of his argument could further illuminate what he’s talking about.

  27. 27 27 Ben

    For an extensive discussion of almost exactly the same thing, see entries entitled “The Olympics, Bernie Madoff and Me” and “Arsenic and Gold Medals” from 2010.

  28. 28 28 Henri Hein

    Neil and ThomasBayes:

    I agree with you (#21, #23). That said, I think we have answered the question, as stated. Why should we tax polluters, but not artists? Because we do not want to risk a tax driving the next Shakespeare into becoming an exterminator (to use Jon Sealy’s example). I do not have that concern with CO2.

  29. 29 29 Daniel

    @ Ben,

    Good catch, but don’t you think this argument from Steve takes it a step further since racing medals have a fixed quantity while readers can be altered by the composition of books available. In other words, books should be closer to perfect competition since there are many buyers and many sellers?

  30. 30 30 Ken B

    @Henri Hein 27:
    The world might be a better place if Eminem had become an exterminator. So I don’t think your argument holds water.

  31. 31 31 Daniel

    @ Ken B,

    I think Eminem’s millions in record sales would beg to differ.

  32. 32 32 Ben

    @Daniel, I’m not sure racing medals have a fixed quantity; there is more than one bike race and certainly more than one snowboarding competition. But even if they did, I’d say people’s time dedicated to reading is relatively fixed also. To the extent it isn’t, the question just shifts to books versus video games rather than book A versus book B.

  33. 33 33 ThomasBayes

    #27: Surely you acknowledge, though, that there are some people who could be concerned that taxing pollution or CO2 might cause individuals or companies to avoid creating things that they believe are important.

    “Because I like novels and I dislike aerosols” doesn’t seem like a compelling answer to Steve’s question.

  34. 34 34 Daniel


    Makes sense, I guess. I’m just not seeing how the other racers could lower the amount that they’d charge society for their win in order to capture more market share like your explanation in #16 for the market for classic and contemporary books made clear.

  35. 35 35 Henri Hein

    @Ken B #29: Philistine.

  36. 36 36 Henri Hein

    @Thomas Bayes, #32:

    Sure, but that is an argument for not taxing anything. At the moment, I am interested in Landsburg’s question.

    I did not use the word like, but I do think the subjective nature of art makes it difficult to quantify. Both in terms of its social gain and its costs. Landsburg gave an example, but you cannot generalize this. See for instance Kirk #19, Al V.#7. I would have a hard time quantifying these values for myself. Quantifying it across a population is impossible. That is not true of the damage caused by a CO2 molecule. We may not be able to measure this with precision, but bracketing it is possible.

  37. 37 37 Steve Landsburg

    Steve White (#5):

    So I still don’t think there is anything new here.

    Nobody claimed there was.

    Everyone else: There are several comments I’m very much hoping to find time to respond to. I can’t promise I will, but I do hope to.

  38. 38 38 Mike H

    Whoa, whoa – I know that the costs of producing a novel already written are sunk costs, but this argument seems wrong to me.

    I shall think carefully about why, but…

    … wouldn’t it apply to any business with large fixed cost and small marginal cost?

    … if Apple spends $1B designing a smartphone, which costs $50 to produce, but they sell them at $400 each, does this imply one should tax Samsung?

    … if Walmart spends $100M on a megastore, does that imply one should impose a tax on new retailers?

    I think I have a nice counterexample now. I have to move my car out of a 2 hour parking zone, though, so I’ll only post it when I’ve dealt with that particular market inefficiency.

  39. 39 39 Fonzy Shazam

    This is a very interesting question, but I believe it is fatally flawed. Without some simplifying and unrealistic assumptions, the presumed market failure does not exist. Here is my argument:

    As others including Nobody.Really in comment #10 above point out, through the process of competition we should see the Faulkner estate, et al. pricing down novels to clear out any incentive for new novelists. We don’t see this in reality. Perhaps there is more going on here, and by more I do not mean the Faulkner estate is dumbly missing the boat or consumers are dumbly not realizing that everything marginally worth reading has been written. If there is a bit of an unseen benefit from new novelists, it may be enough to erase this presumed social cost. I believe there is a very good chance this is the case. Therefore, the question would be irrelevant as the premises of the argument are untrue.

    But let’s grant the argument as presented (assume the premises to be true) and answer the question. The problem with taxing the Sara Gruens of the world is that this tax could be more distorting than beneficial. Taxing away the competition from existing novel rights holders creates a rent. This will lead resources to flow into the chasing of the rent and out of the production of new resources. Rather than tax new novelists, set a price ceiling for all novels at marginal costs (presumably $.01 per copy).

  40. 40 40 Mike H

    So, lets rewind. Faulkner and Gruen are both considering writing a novel. To write a novel costs $8000. They sell for $10 each. A reader derives $12 worth of pleasure from the reading.

    Let’s suppose that if one novel is written, 1000 people will buy it. If two are written, 1200 people will buy one each.

    If neither Faulkner nor Gruen write, the net social benefit is $0.

    If only one novel were written, the net social benefit is $12000 – $8000, or $4000.

    If two were written, the net social benefit would be $14400 – $16000, a social cost of $1600.

    In this scenario, it is best if only one novel is written.

    Steve’s argument treats Faulkner’s efforts as sunk costs. Faulkner rolled up his sleeves earlier, hit the typewriter and got his novel out quicker than Gruen. Fair enough, he died 7 years before she was born.

    Now that his novel is already written, we can do the calculation again. Sara spends $8000 to write a novel, only 200 more people read novels, and only $2400 of social benefit is created at a cost of $8000, bringing the net social benefit down from $4000 to -$1600.

    If one wishes to consider this an externality, though, it doesn not automatically follow that Sara should be discouraged from writing. After all [puts on Coasian hat] Sara would have produced $4000 of social benefit if Faulkner had gone into the railroad business instead of taking up the pen. Perhaps it’s Faulkner who should have been discouraged, back between the two world wars.

    Now, this kind of conundrum presents itself in any situation where there are economies of scale. Economies of scale lead to barriers to entry, winner-takes-all markets, and so forth. Some of these effects could be interpreted (I suppose) as externalities. If we choose this interpretation, the question is – should we discourage later entrants, or impose the Pigovian taxes on the incumbents who can use their economies of scale to discourage competition?

    I suspect taxing the incumbents would increase competition, making the market more efficient overall, and that taxing new entrants would have the opposite effect. In the novel-writing industry, an obvious way to do this would be to shorten copyright terms. Alas, it is the incumbents who have the political power, and anyway, the new entrants are still busy with their day jobs.

    Do the welfare theorems still work if there are economies or diseconomies of scale?

  41. 41 41 Neil

    thomas bayes @24

    You are certainly right. This is not about what sort of books people like. It is about the process of creative destruction that has driven economic development for two hundred years.

    Although I am not convinced by Steve’s literature example, when I look at the gadget industry I do see his point. Is the reward to the latest marginal tweak in a smartphone really commensurate with its economic value to society?

    I don’t know. And I would not advise a tax on innovation. But it sure makes you think. Maybe creative destruction is like biological evolution. A wasteful, but creative, process.

  42. 42 42 Neil

    Clarification. I think creative destruction PLUS technology diffusion (copying other people’s costly innovations) drives economic development.

  43. 43 43 Jim W K

    Steve, one other potential criticism worth considering is this: the assumption that enough superb books have already been written to satiate us for a lifetime is possibly a precipitous and ill-conceived one. Who is to say that future writers won’t take literary writing to a whole new level beyond our current expectations?

    Just as some people in Roman times thought every invention that could ever be invented had been invented already, so too might the people brought up on Homer, Herodotus and Sophocles have hastily thought that this is the best literature can get. How wrong they’d have been.

    Just as the Greeks and Romans couldn’t forecast what was to come with the likes of Dante, Shakespeare and Flaubert, so too might we be ill-equipped to foresee what kind of quality might lie ahead in the future. Therefore we might be missing out if we assume everything that has already been written is enough.

  44. 44 44 Harold

    It is a red herring whether there are enough books already. It is pretty clear that if I read a new book I don’t read an existing book, or even if I am reading a new book by Krugman I am not reading a new book by Landsburg. If we were to arrange all the (fiction) books that exist in order of how much reading pleasure I would get from each one, then slot in any new ones that arrive as they are written, I do not think we would see any discontinuity.

    I could argue that perhaps a new book will have a transformational effect on me. I can also argue that an unread old book would have had such an effect.

    Clearly this is not the case for non-fiction.

  45. 45 45 Daniel

    @Landsburg 37,

    If you do respond, I hope you respond to Ben’s point in #17 first. His I think is the most straightforward criticism of your points.

  46. 46 46 Ken B

    @Daniel 31:
    You don’t know how good an exterminator Eminem might have been.

  47. 47 47 Daniel

    @ Ken B,

    Lol, true.

  48. 48 48 bigjeff5

    I think Dr. Landsburg is giving a false comparison for this example. Given his example, he is absolutely correct and I don’t think Water for Elephants would ever have been written, and there would then be no social cost.

    I base this on the actual number of sales for Water for Elephants thus far – 4.2 million, and the average household income – about $50k – the sales of this book are expected to replace to give a value for the author’s time that could have been spent doing something else.

    I’m assuming for this argument that it took a year to write Water for Elephants. Considering she was already a successful author, we probably can ignore the additional time/costs of getting the book published.

    So, at roughly $50,000/4.2 million, the effort per book is 1 cent. I wouldn’t expect more than double this number or less than half this number for the actual effort Gruen put into the book, so I feel it is a good approximation.

    Now we have real numbers for the comparison:

    Falkner value to Landsburg is $9
    Cost of book is $8
    Cost to produce is $0
    Social gain is $9 ($1 to Landsburg, $8 to Falkner estate)

    Gruen value to Landsburg is $10
    Cost of book is $8.50
    Cost to produce is $0.01
    Social gain is $9.99 ($1.50 to Landsburg, $8.49 to Gruen)

    Thus, best sellers provide a net social gain over previous works.

    This scales down for less successful works until the struggling author is better off doing something else, and does so.

  49. 49 49 Ben

    @bigjeff5 48,
    I don’t think you understand the hypothetical. First, what reason do you have to value Gruen’s time/effort/etc to write Water for Elephants at $25k-$100k? Landsburg has specifically set the cost of Gruen’s book at $8.50 per book — notice this is the cost (including labor, publishing, etc), not the price. You can argue that this is far too high ($8.50 >> $0.01), but that changes the hypothetical substantially and doesn’t answer the question. With the cost at $8.50, the price at $8.99, and the consumer value at $10, Gruen’s book displaces a $9 social gain with a $1.50 social gain ($10-$8.99 to Landsburg, $8.99-$8.50 to Gruen).

  50. 50 50 Harold

    I find Bigjeff5′s point interesting. Landsburg says “Now a contemporary novelist, say Sara Gruen… realizes that with $8.50 worth of effort, she can write a book for which I’m willing to pay $10″ He speculates that “she might well expend $8500 worth of effort to write a book for 1000 people like me” SL specifically ignores printing and publishing costs etc.

    Bigjeff5 says, well, why speculate! Instead of just saying it is $8500 for 1000 books, lets put in some realistic numbers, and he comes up with $0.01 per book. The book is still sold at $8.99. We now have gains of $1.50 for Steve and $8.49 for Sara. Steves argument is based on assuming that the effort for Sara to produce a new book is very much higher than the effort the Faulkner estate needs to expend to produce another copy. This may not be the case if most of the cost is due to publishing and publicity etc. I don’t know if Bigjeff5′s method is the best, but it seems that Steve’s is very arbitrary. Can we nail down the value of the effort required?

  51. 51 51 bigjeff5


    $50,000 is roughly the median income in the US. If Gruen were to something other than writing, the expectation is that she would generally earn about $50,000. She might earn a whole hell of a lot more, she might earn a lot less, but for a thousand Gruens out there the expectation is that they’ll average out to $50,000. That’s why I used that number.

    Producing the book is a fixed, one time cost, and I assumed about a years worth of effort. She’s a well published author, so I don’t think this is unrealistic. Her first book might have constituted several years worth of effort, but in the end it doesn’t change much.

    I ignored the cost of editing and advertising, but those are short term or fixed costs as well. You could probably raise the figure to $500,000 for this book, but as you can see that doesn’t change a whole lot. It’s a $0.12 cost per book instead of a $0.01, with the social gain of the Gruen still far out stripping the gain of the Falkner. She’s also going to have to sell at least half a million copies before it’s worth her time to write the book, so this should scale really well with the publisher’s confidence in the book. Better books are more likely to get big up-front expenditures, and are therefore more likely to make big sales.

    The cost of printing, once the costs of writing and editing and advertising are taken into account is so close to $0 that you might as well call it $0 – and if you don’t it’s the exact same cost for the Falkner.

  52. 52 52 Ben

    @Harold 50,
    Actually, Bigjeff5 is just speculating, and it’s wilder than Steve’s speculation. Steve isn’t trying to claim that it costs Gruen specifically $8.50 per book to write and publish a book, he’s posing a hypothetical where we take as given that it costs some particular author $8.50 per book to write something that some people will prefer over existing works whose marginal cost is essentially zero. Gruen is probably not the best example. But your critique is valid only if there could not possibly be any authors who capture readers who would otherwise purchase an existing book with very low marginal costs. Is that what you’re trying to argue?

  53. 53 53 Ben

    You’re still missing the point of the hypothetical, which is independent of whether Gruen specifically actually has a cost of $8.50 per book. If your hang up is about using Gruen specifically, consider my one of my favorite authors, the lesser-known Foo Bar. Mr. Bar would otherwise earn about $50,000 in the year it took him to create his book that is expected to sell just short of 5900 copies (his work is rather eclectic). Readers value his book at $10 and he prices it at $8.99. Now proceed with Steve’s hypothetical.

    But digressing from the topic at hand, let’s apply your logic to Steve Jobs writing a book. $50,000 is roughly the median income in the US. If Jobs were to do something other than writing, the expectation is that he would generally earn about $50,000. He might earn a whole hell of a lot more, he might earn a lot less, but for a thousand Jobs out there the expectation is that they’ll average out to $50,000. (Ok, well, actually that’s not true because I’ve described the mean yet quoted the median, but we’ll go with it) This probably seems crazy because we know Jobs’ time is very valuable — more valuable than the median American. Just like we know Gruen’s time is more valuable than the median American. Why? Because Gruen has written a book that millions of people want to read — the median American can’t do that.

  54. 54 54 Harold

    Ben, I feel that adding new books must reduce social welfare absent of some other externality that new books provide, since there are enough books with a huge range of types already out there. But Bigjeff’s argument reduces the social loss if the cost per book is much less. This means the externality needs to be much less to cancel out the social loss, and the production of new books may be desirable socially.

    Steve Jobs not working at computer / device design would be a social loss because nobody else produces anything as similar to the iphone as one book is to another. It comes down to what one is reading for. If it just pleasure whilst reading, then one book is much the same as another in terms of pure pleasure. If it is to “improve your mind” in some way, then it is difficult to say old books cannot do this as well as new books. If it is to improve one’s awareness of current situations, then current books are required. i think this needs more time, but I am pushed at the moment, so apologies if this is incoherent.

  55. 55 55 bigjeff5


    I don’t think I missed the point of the hypothetical at all. I just put some numbers in to test whether the hypothetical holds up to real world expectations. If these principles can’t hold up to the real world, what’s the point?

    In that vein, I’m arguing that Steve’s point holds until the author’s cost to produce the new book is less than the difference in value that Steven perceives for the books. For his Falkner/Gruen example, that is $1. With the simplified model where the author’s time and effort constitute the entirety of the cost (which is the one Steve used in his post) then the more copies an author sells the greater the social value.

    The $50,000 is just the average. If you were to poll all professional authors, I’d bet money that the average yearly earnings is remarkably close to $50,000. However, you can substitute real life earnings numbers to find out exactly when someone should bail on the whole writing thing.

    For example, suppose Gruen could earn $1 million a year doing something other than authoring. Assuming it takes a year to write the book, she needs to sell at least 1 million copies before the cost per book is $1 and there is as much social gain when Steve buys a Gruen book as when he buys a Falkner.

    Thus, if Gruen doesn’t sell a million copies of his book then Landsburg generates more social value when he buys the Falkner instead.

    Likewise for Steve Jobs, since we know he made many millions of dollars a year and since he probably isn’t going to sell nearly that many copies of his books, it definitely isn’t worth it for Jobs to write the book instead of doing what earns his millions. If he can do both at the same time, however, the cost per book plummets, possibly to nearly 0 regardless of the number of copies sold. In that case he should definitely write the book.

    Likewise, suppose Mr. Bar wrote his book as a hobby in his spare time, and still earned his regular wage. Most hobbies don’t result in an income, so his total cost is probably going to be very low – probably no more than a couple thousand dollars on the outside. In that case, selling 5900 copies at $8.99 means his per book cost was $0.34, for a $0.76 social gain per book over another book.

    It’s worth noting, I think, that this is a subjective per-reader, per-author, per-alternative valuation. For example, if Steven values the Gruen at less than the Falkner, there is never a case where buying the Gruen constitutes a greater social gain.

    As for your Foo Bar example, the Falkner is the greater social gain. Like I said, it’s not until the cost per book drops to less than the difference in perceived value.

    “Just like we know Gruen’s time is more valuable than the median American. Why? Because Gruen has written a book that millions of people want to read — the median American can’t do that.”

    Right, her time is more valuable because she sells more books. If she didn’t, her time would be less valuable. You don’t calculate the cost per book by what she earns as an author, you calculate it based on what she could be doing instead of writing. That’s the cost of writing instead of doing something else. What she earns as a writer is her contribution to the total social gain.

    Once she dips below the $50,000 point she is probably not pursuing the right career. However, half the authors out there are going to be below that $50,000 point (that’s just how statistics work), so they must be getting something else out of writing than just money. I imagine most authors stick with it because of their passion for the job, or they enjoy the lifestyle, or some other less tangible reason. Or, there could be a huge turnover rate below that point, where people try being an author for a year or two, fail spectacularly, then move on to something else.

  1. 1 #DailyBookQuote 23Oct13 : William Faulkner’s Light in August | Whatever It's Worth...
Comments are currently closed.