the dismal science

Matt Yglesias has a snarky column today about how the printed book is an obsolete technology whose only plus is “a nostalgia-soaked experience”.

He mentions some advantages of e-books (quick access, less bulky), but fails to think about any disadvantages.  Here’s a few:

  • Low resolution.  I read a lot on my computer, but it’s still a fraction of the resolution of print, and it’s bad for high-res graphics: comics, maps, diagrams, art books.
  • Screen size.  Not a problem for reading a novel, but a double-page spread can provide a lot of information, and it’s just not the same thing to thread that data through the small screen of an e-book reader.
  • Price.  I don’t have a Kindle because I can’t afford one.  At $199 for a color Kindle, it’s a significant loss if you lose it on a trip or drop it in the bathtub or whatever.  There’s much less sense of loss if you misplace a printed book.
  • Eyestrain.  Print is still the most comfortable way to read long texts.  (If you don’t think so, wait till you have middle-aged eyes.)
  • Reliance on dubious megacorporations and changing technology.  I’m still using books I bought 40 years ago; Random House can’t do anything to interfere with my enjoying them, plus the NSA is certainly not checking which parts of what physical books I’m reading.

Edit: Alert readers Carsten and JDHarris offered two more points:

  • It’s easy to put several books next to each other, hard (or at least very expensive) to do the same with e-readers
  • Books will never stop working mid-story because you forgot to plug them in last night
Searchability works differently for e-texts and physical books, and each has its advantages.  I’d hate to have to consult the OED in physical form: it’s way too big, and some types of searching (e.g. by language) would be impossible.  On the other hand, a book you know well, as a physical object, affords quick access you can hardly even define for an e-book.  You can flip immediately to a dog-eared page.  You know that a certain passage is this far into the book, top of the left-hand page.  You can make marks on the edges to point to often-consulted pages, and make notes in the margins which themselves become searchable by riffing the pages.

Plus, I’d say a bunch of people agree with me, because my print sales are pretty healthy.  For the LCK, this year, my sales are 57% print; for APAF, 42%.  Given that the print book is twice as expensive, I’d say that indicates that people still find the format valuable.  It’s probably significant that for the novel, the majority prefers the Kindle: genre books are a good match for e-books.

Yglesias is no doubt correct that print books are not likely to be a growth industry.  But print is far from disappearing, and people are going to continue to make money off it.  Especially that Jeff Bezos fellow.

I like almost anything by William Poundstone, so I pounced on his latest, Priceless.  It’s about prices.

It’s a bit miscellaneous… Poundstone wants to tell you about psychophysics and pricing consultants and even takes a foray into feminism.  But the overall theme is the irrationality of humans about money, something that’s come as a challenge to some modern overvaluations of The Market.

More precisely, it’s not that humans are irrational; it’s that they’re not rational in the way economists want them to be.

The main myth is that there is such a thing as a single price for an item… even for a single consumer.  It starts with the fact that all our senses and valuations are relative, not absolute.  That is, we have senses of how things compare, but very little good sense of the absolute value of things.

One neat bit: anchoring.  For instance, here’s a number: 56.  Now here’s a question: what’s a fair value for a dinner for two?

You are convinced, of course, that you answered the question without even thinking of the number I named.  Which indeed has nothing to do with dinners for two– I wrote it down before even coming up with the question!  But if we were doing this experiment for real, the results would be that people are influenced by the number; the answer they provide will be skewed in its direction.  It’s an anchor– somehow it affects the next calculation.  It works even if the subjects are told to ignore it.

These days, people happily use anchors in negotiations and in everyday pricing.  One of the simplest techniques is for your store to offer products at three price levels.  People tend to dismiss the lowest level as too cheap, probably shoddy, and the highest as overpriced.  They select the middle one.  But you can set the high and low ends almost arbitrarily.  The very same product, for instance, might be a medium choice at an ordinary supermarket and a low end choice at Whole Foods.

High end stores will have an item on display whose price is absurd.  But it serves as an anchor, making the real high-end items seem like a relative bargain.

Other pricing games show that people are not terribly good at quickly evaluating complex odds.  For instance, which of these offers is better?

  • a 10 in 12 chance of winning $9, otherwise you lose $3
  • a 3 in 12 chance of winning $91, otherwise you lose $21

If you calculated the average payoff, great.  Most people don’t apply the economist’s algorithm, though.  They evaluate whether something seems like a sure thing (and they are none too good at judging rare odds), and they are much more averse to losing what they have than to chances of winning new things.

Then there’s the ultimatum game, which drives economists crazy.  It’s simple: get a bunch of subjects, in pairs.  Each is offered $10.  Person A gets to divide the $10 any way they like, and person B can either accept or reject the division.  If they accept, the money is divided that way.  If they reject it, no one gets any money.

The economist’s reasoning: Person B should accept any offer, including $10 to $0.  After all, even if they’re offered only a pittance, it’s free money.

What actually happens: Most people offer a $6 to $4 split, which is usually accepted.  Person B may or may not accept a $7 to $3 offer.  They almost always reject anything more biased toward person A.

As noted, this experiment just bugs economists, and they’ve repeated it over and over.  (One of the forays into feminism: if person B is a woman, she’ll get slightly lower offers.  This is true even if person A is female.)

I don’t think it’s that shocking, myself.  People hate unfairness, and they don’t like it even within a game.  (The experimenters should probably just join a public gaming server for awhile, if they really think that the artificiality of games, the fact that no money is changing hands, suspends judgments of fairness.)  As social animals, we need to be extremely conscious of unfairness.  We didn’t evolve to be homo economicus, and if we had, we probably wouldn’t have developed an elaborate civilization.

Curiously, results from the ultimatum game differ across cultures.  Wherever there are markets and cooperative projects– which is to say, in most cultures– selfishness is punished.  Among the Machiguenga, who have neither, offers were consistently lowballed, and accepted.  The Ache hunter-gatherers, whose hunting superstars generously share their catches, would offer person B more than half.  So did the Au, of Papua New Guinea, but in that case, person B would reject the deal.  Among the Au, generous offers impose a debt, and people don’t always want to take on that obligation.

Anyway, fun book, lots more stuff like this.  As mentioned, though, it covers a lot of ground and not all of it may stick with you.

In your recent post, you noted that the (first) world is moving to a frivolity/art economy. What will that mean for employment? After all, robots are making our cars and, increasingly, our burgers— and, like it or not, those are jobs that in the past were filled by regular people whose natural talents were in menial work. Not everybody has the talent or desire to spend their time writing novels or direct avant-garde movies…and that’s OK; people shouldn’t be cut out of an economy because of their natural skill set. What will that mean for employment or the economy? It seems unrealistic to give everybody a $30k/year minimum income, for example, as I’ve seen suggested; that seems fiscally undoable.


There’s a grim meathook future answer and a nice answer, depending on whether we follow our current plutocratic path or not.

But first I’d just note that your question seems rather regressive.  It’s like Aldous Huxley assuming that the future must include a huge population of subhuman menials.  Dumb repetitive work is what machines do very well; those jobs just disappear.

The grim meathook answer is what’s happening today: lots of low-paid service work— call center employees, Wal-Mart greeters, nannies, waiters, nursing home attendants, home sales party presenters, bodyguards, flight attendants, SEO farm writers.  What humans do better than robots, for the indefinite future, is deal with other humans.  We need and value human contact, and anyway most of these jobs, even if they’re not exciting, require a generalist.  Humans deal well with the moderately unexpected.

Or to put it another way, automation targets expensive, repetitive jobs.  When you get rid of the $40/hour factory jobs, you have a large population that is forced to take $10/hour service jobs.  Even for Mr. Scrooge it’s not worth bothering to replace those jobs.

In the more optimistic future, we use the increased productivity that automation brings to improve everyone‘s life.  That’s what even the curmudgeonly old USA did in the liberal era, so it’s not unthinkable.  Poverty used to be universal; in mid-century America the vast majority were middle class; an even richer society could, if it chose, eliminate poverty entirely.  (That “we’ll always have poverty” is a myth to comfort the 1%.  We could end absolute poverty globally for a surprisingly small sum.)

I don’t think anyone has ambitions limited to factory work or bagging groceries.  Everyone has some dream that they’d love to be paid to do. In our economic system, maybe it’s too silly or specialized to pay well, but a world where the robots do all the heavy lifting is one where everyone can be a specialist or a frivolist.

But even in the more ideal world, it remains true that humans are better at making other humans happy.  When you’re 94, you probably don’t want to be surrounded only by robots.  So ‘elder care’ is still a human niche, but it’s seen as valuable rather than degrading and paid enough to make it attractive.

The SEO writer may not exist in the happier future, but only because he’ll be doing something far weirder.  In the Incatena, rather than a thousand different jobs with a million people in each— a situation that may be automatable— there’s a million different jobs with a a thousand people in each.

In general we’ve moved from an agricultural economy to manufacturing, then to a services industry.  What’s next?  Probably a frivolity economy.


I think this bothers some people.  A standard complaint about capitalism is that it devotes enormous energy to producing things we don’t need and shouldn’t want.  Many thinkers, left and right, seem to think we should just produce the bare necessities– or at least, avoid obvious excesses, like plastic surgery and bondage gear.

My contention is that what capitalism is good at is producing things people want, whether or not either anarchist professors or austerian central bankers approve of them– and that it should produce those things.  In fact, if it didn’t, the economy would collapse.  In the far future, I expect human activity to be little but frivolity.

Continued on Research Access

From Ha-Joon Chang:

If even the IMF doesn’t approve, why is the UK government persisting with a policy that is clearly not working? Or, for that matter, why is the same policy pushed through across Europe? A certain dead economist would have said it is because the government is “in reality instituted for the defence of the rich against the poor“. Dead right.

The dead economist is Adam Smith, and as Chang points out, there was no mystery about it in Smith’s day: only the wealthy could vote, and they ran the government for their own benefit.

When universal suffrage came, they were terrified of redistribution– or even a more equitable distribution of newly generated wealth.  They can’t directly attack democracy in the First World, so they attack “politics” instead.  When they can, they insulate wealth from politics entirely– not hard to do when ordinary voters don’t understand the stakes.  Thus we get reasonable-sounding independent central banks, balanced budget rules, IMF oversight, attacks on inflation that doesn’t exist, and ‘technocratic’ governments imposed on Italy and Greece– that is, governments that will do what the European wealthy think is best for Germany.

The irony at the moment is that there’s been a change of heart over at the IMF.  They’ve withdrawn support for austerity, and are suggesting to Britain that maybe sending the country back into recession isn’t that great an idea.  The Treasury promises to fight back.

Why do the Very Serious People love austerity?  After all, they’d themselves be richer if the economy was back on track.  I don’t think this a great mystery either.  Ideologically, it’s congenial to them– it sounds like the tough stuff leaders are supposed to say.  But best of all, the toughness is all faced by other people.  Austerity is an opportunity to beat back the claims of the middle class and the poor: cut social programs, fight attempts to reduce the dominance of  the 1%, and divert attention from how the financial industry caused the recession in the first place.

If you read Smith, it’s striking how the same preference existed two hundred years ago.  The conservatives preferred “cheap years”… i.e. recessions, when labor was pleasantly low-priced.  They’re going to do fine in bad times, and they sense that there will be less pressure on them if everyone else is feeling pinched.  As has long been noted, a system is in the most danger of revolution not when things are getting worse but when living standards are improving.

While I’m touting economics links, here’s an interesting essay from Brad DeLong on why corporations work at all.  As he points out, their structure is Soviet: they’re autocratic, huge-scale, centrally directed enterprises.  The USSR stagnated, so why don’t they?

He offers several possible reasons:

  • Soviet industries could be propped up indefinitely; unprofitable corporations do slowly decline, go bankrupt, or get taken over.  So there is a mechanism for replacement, however slow.
  • Huge enterprises may simply be the best organization for producing certain goods– planes, for instance: there’s a high cost if, say, you have the engine ready but not the wings.
  • Stockholders will discipline an incompetent CEO.  Pause for laughter.  OK, DeLong quickly admits that the theoretical oversight is mostly a failure, but he suggests that the punishment mechanism isn’t so much the stockholders as the stock market.  A crashing stock price talks very loudly, and creates a mechanism for a hostile takeover.
  • Finally, corporations are intensely useful to government, and thus are favored by taxation and economic policy.  Corporations do most of the work of tax collection (income and sales), and they’re encouraged to provide a good deal of health insurance.

(I’d also add that beating the Soviets is a pretty low bar.  They were more concerned with destroying opposition than in improving their methods.)

Undoubtedly there’s something to all of these ideas.  But I think DeLong misses some more Marxian possibilities:

  • Inertia.  We now think monarchy was a terrible system, but it persisted for a thousand years.  Even if the Next Big Thing was here, competition between entire systems can go on a long, long time.  (A Martian observer might have concluded from the history of the 18th century that UK was onto something, but it would take nearly two centuries before the majority of countries adopted democracy and capitalism.)
  • A better way of organizing production would be unlikely to benefit those currently in charge… who will therefore resist it tooth and claw.  Suppose the Next Big Thing is something as simple as Valve’s no-manager, vote-with-your-desk system.  Is EA going to adopt it?  Obviously not; the people who run EA would lose power and probably money, even if the employees of EA on the whole benefited.  And it’s the people who run EA who get to decide.  It’s the same problem Chang is pointing out: the entrenched interests are happy with things as they are.

Besides, what would happen if a new style of governance became available?  DeLong (the piece was written in 1995) reviews one such case:

Fifteen years ago it was fashionable to hold up the Japanese corporation as an example, to say that its managers regarded shareholders as only one stakeholder interest among many, and to say that the Japanese corporation was a superior organization and the wave of the future. Now it is fashionable to praise the American form of organization, with an active market for corporate control and with strong pressure on managers to do whatever they can to boost stock prices now.

In other words, a new system becomes a new fad, and a lot of people get rich writing bestsellers about it.  But the magic of Japan Inc. wore off abruptly when Japan plunged into recession in the early ’90s.

On the other hand, as I’ve mentioned before, the fad for highly paid CEOs was entirely successful.  In the 1960s it was accepted that American CEOs should get about 50 times the pay of their base workers.  Now it’s 500 times.  Corporations aren’t better run or more productive or more stable or more competitive; the effect has simply been to shift wealth from the 99% to the 1%.  Why did this fad succeed when the fad for Japanese-style management didn’t?  Pretty obviously, because it doesn’t require much convincing to tell executives that they should help themselves to ten times the salary.  It’s not that it was a good idea; it’s that it appealed to the people who make the decisions.

When change happens, it’s likely to come from either a new region or a new industry, and be very ignorable for years or decades or centuries.

I have a suspicion that top-down command works better for creating  vast new project– whether it’s an oil refinement industry or a spaceship or a computer or an online mega-bookstore– than it does for running an ongoing business.  Capitalism recognizes this to some extent, in that older businesses have more and more dilute ownership.  But this also means that there’s a kind of ongoing bias in favor of autarchy.  New firms, like the big web firms of the ’90s, are likely to be started by visionary hotshots, and that reinforces the elite’s notion that all corporations should be run by autarchs– even if, as happened recently at JC Penney, the hotshots come in and ruin the company.

So a more democratic corporate structure might have to wait for the next big thing that doesn’t happen to be a megaproject.  I can see it happen, for instance, if we move from the service economy to an art economy, for instance.  (Which I don’t see as unlikely: creative work is very hard to delegate to robots.)

My light reading for about a month has been An economic history of medieval Europe, by N.J.G. Pounds.  I recommend it half-heartedly.

What do we know about medieval economics?  Frustratingly little, it turns out.  Every few pages Pounds has to remind us that there just isn’t much data.  He goes over what we have, but it’s really impossible to build up the sort of overall statistics that we take for granted today.  It’s almost impossible to get estimates of production of goods, or even to definitively answer questions like when the moldboard plow was actually adopted, or whether the 1400s were a period of depression or a time of productive improvement.

The book is from 1974, so it’s possible that nearly 30 years have produced a slightly clearer picture.  E.g. I don’t quite buy his statements that medieval technology was stagnant, not after reading Jean Gimpel’s The Medieval Machine.  Pounds even mentions some of the same inventions, such as the blossoming of mill technology and the later focus on mining.

Perhaps the biggest surprise is in the first few chapters, on the late Roman Empire.  We have a picture of a flourishing, sophisticated, rich urban civilization, but in many ways this is an illusion.  Most of the Empire was at a barely subsistent level; the western half was “an uninhabited wilderness, broken by islands of cultivation”; trade was minimal; large-scale enterprises were undertaken only by the state; Rome itself basically produced nothing.  The East was of course richer and more urbanized.

The book also emphasizes that the lot of the peasant, from Roman times till well into the modern age, frankly sucked.  At the subsistence level, the peasant couldn’t afford much in the way of urban wares, so the cities remained small and trade was largely in luxury goods.  The empty spaces on the map filled in, and new towns appeared, but that just meant there were more and more peasants and on more marginal land.  The only real improvement in living conditions were a) in colonizing new lands, especially in Eastern Europe; and b) after the Black Death, when depopulation temporarily created a labor shortage.

And as Eastern Europe filled up, the feudal lords exerted more and more control and turned most of the peasants into serfs.  In Western Europe the tendency was for the king to rein in the nobles, which was a little better for the peasants.

There’s a discussion of the guilds; Pounds seems to think that they never amounted to much.  They always tried for monopoly power, which in theory could restrain the economy; but in most places it was the merchants, not the craftsmen, who really ran the economy.  (Often they supplied the raw materials and even the tools.)  In any case, when the urban cloth workers grew too expensive, the merchants simply outsourced the work to the rural areas.

One surprising assertion is that the cities had trouble feeding themselves– the northern Italian cities had to import grain from as far as Sicily.  This seems a bit odd when none of them exceeded 50,000 residents.  But perhaps the surplus of the European farmer really was that low– or perhaps the situation illustrates the price differential of wagons vs. ships.

There’s quite a bit about how the early fairs developed into permanent commercial markets, and how the early traders operated.  Currency was scarce, so things were arranged such that little money had to actually change hands.  You’d bring your alum to Bruges, take home a shipment of woolens, and payments were mostly handled by moving numbers within the bankers’ ledgers.  (The last few days of a fair were devoted to the settling of accounts.)  At first the big merchants actually traveled across the continent; later on they simply employed local agents.

Kings and other lords were constantly interfering in the market.  Landowners, including the king, were generally able stewards of their own estates.  Their powers of enforcement over the rest of society were limited, which led to interesting compromises.  Close control was impossible, but on the other hand focusses of wealth could be seized.  Thus cities were given a large measure of autonomy, but were also easily taxed.   In many countries the king had the right to all mineral resources– but as he could hardly mine everywhere, what this came down to was that anyone could mine, but owed a tax to the king.  Most states were highly in debt to the banks, but didn’t scruple at confiscating a local bank or defaulting to a foreign one.

Another lesson is just how miscellaneous Europe was.  Generalizations at the national level are almost useless; you have to look at each region and even each town.  The Elder Scrolls continent of Tamriel, with its ragged multiplicity of races and regions, is actually not a bad model, certainly much better than the usual fantasy expedient of one uniform country with perhaps one exotic neighbor.

I recently got into a conversation with a work colleague regarding piracy. Specifically, the basic assumption that I came to is that internet piracy of media is that we do not value the artist’s effort in production, nor do we comprehend the inherent monetary value of art (I positioned it as a symptom of post-modernism, that art is for art’s sake, and art-for-monetary’s sake is blasphemous.) He stated that the cost of media is just too high, and he also opposes (much like a lot of things) certain studios so refuses to give them his money, but still wants to enjoy the media produced under their umbrella. My question is, knowing that you have written on this before: What are your initial thoughts on the MegaUpload extradition? And what are your thoughts pertaining to the idea of media piracy— why do we do it? Ultimately, I’m not trying to understand why the government is wrong in trying to push SOPA and PIPA through, nor am I trying to figure out which one of us was right, if any, just your opinions on the subject.


As a content creator myself, I think people should pay for their art.  (Except sometimes; see below.)

But in general, I think ‘piracy’ is a sign of market failure, and disappears when providers make it easy to get content in whatever form is desired.  My best argument for this is the enormous success of Steam.  Steam is at root a DRM system, but it’s so packed with features for the gamer that it’s actually a pleasure to use, and I don’t buy games any other way now.  You can buy your games there, see what games your friends are playing, chat with them, use games on multiple computers and platforms.  There are frequent sales, so those $60 games will soon be available for half that— or wait a year and get them for under $10.  You can review games or get links to other reviews before you buy.  You automatically get updates and patches.  I recall an interview with Gabe Newell where he said people thought Steam was crazy to get into Russia, which is notorious for piracy, but they’ve done very well there.

The music industry earned a lot of hatred for attempting to continue its lucrative old model— selling physical CDs for $20 a pop— well into the electronic age.  Their business model had very little to do with helping artists; they wanted the profits of the manufacturer / distributor… precisely those costs that go away with electronic distribution.  I don’t know the figures, but I’d guess that most people are pretty happy buying single songs from iTunes for a buck.

The next battleground is movies.  Why haven’t the studios come up with a Steam-like service where all movies are available on all platforms, at reasonable prices?  Instead of tracking down pirates and alienating customers, they should be figuring out a feature-packed, cheap distribution system of their own.

During the SOPA fight many people pointed out the absurdity of the jobs claims made by the industry, which were based on the idea that any pirated viewing represents a full theater admission lost.  Of course it doesn’t.  If someone wasn’t able to pirate the movie, he would probably have skipped it, or waited till it was free on TV or at the library.  Or maybe he’d pay $1 for a used DVD, or $2.50 for an iPad version.

Many businesses have discovered the joys of variable pricing— Tim Harford’s The Undercover Economist goes into detail on this.  Econ 101 tells you that there’s a single price— let’s say $17.95— where the supply curve and demand curve intersect, and which maximizes sales and profits.  Econ 101 is wrong!  All those consumers who’d happily pay $10, or $5, or $1, don’t get what they want, so their sales are lost.  And those consumers who’d pay $50 or $100 aren’t being milked for enough simoleons!

It’s not easy to work up a pricing scheme where everyone pays the maximum they’re willing to… but plenty of markets do their damnedest to try.  Airlines are very good at this— it’s almost the case that no two flyers get the same price.  Books are a good example: you can get a hardcover for $25, a trade paperback for $15, a mass market paperback for $9, a Kindle for $5, a used book for $3, a library book for free.  Steam approximates this for video games, the chief variables being willingness to wait, and assiduousness in checking sales.  (Game companies have even figured out how to get the really motivated customers to spend more, largely with premium editions and DLC.)

And note, there’s a role in this model for free.  I don’t mind if you lend your copy of the LCK to a friend, or donate it to the local library!  A certain amount of free distribution builds recognition and good will, and in the long run increases sales.  (Neil Gaiman has experimented with giving books away for a limited time; it always increases actual sales as well.)

Plus, art needs art as a source of ideas and inspirations— it would be a bad thing if the heirs of Shakespeare controlled performance of his plays.  It’s right there in the Constitution: copyright is a balance between the interest of creators and consumers.  The sign that the balance has gone too far toward the consumers would be that artists are starving and not producing any art— and they’re not!  We’re awash in art!

If I read anti-industry people too much, I start to sympathize with the conglomerates.  People make all sorts of rationalization for what sounds like entitlement and miserliness.  But really such things are a reaction to the obvious greed and stupidity of the distributors (who aren’t even the creators).  If your friend doesn’t want to pay $17.95, that doesn’t justify him in paying $0, but if they were smart the distributors would find a way to get him to pay $5, or whatever.

I also think we’re going through a slow transition from a system dominated by middlemen, to one where artists handle their own production and distribution, and of course the middlemen are squawking.  But it’s their own fault for not adapting.  Books and music, and even indie movies, could easily be produced by their creators.  Maybe not blockbuster movies, but somehow I don’t think we’ll ever have too few of those.



« Previous PageNext Page »