Everyday Economics

Planning for Obsolescence

Not everything should last forever.

When I wrote my first textbook, the publisher wanted to combat the second-semester used book market by sewing a $100 bill into the binding of every 100th book. Every student would tear his book apart to see if he’d won. Printing with disappearing ink that lasts exactly one semester would also discourage the used book market. But instead of running lotteries or using disappearing ink, most publishers make used textbooks obsolete by periodically releasing revised editions. Did I mention that the fifth edition of my textbook is forthcoming next year?

The naive answer to why publishers might want to discourage the used book market is that they prefer to get paid every time a student buys a book. But by that logic, you should never sell your house when you can rent it: Why get paid only once when you could get paid every month? The logic is wrong because the sale price is likely to be far higher than the monthly rent. And the logic is still wrong when applied to textbooks, because the sale price for a book that can be resold is likely to be far higher than the price of a book designed to lose its value.

If a student is willing to pay, say, $30 for a textbook, that same student will be willing to pay $60 for the use of a textbook that can be resold for $30 at the end of the semester. (For the sake of simplicity, let’s ignore the fact that the $30 delivered a few months from now is worth a little less than $30 delivered immediately.) And for a book that can be resold twice, the second owner’s willingness to pay $60 means that the first owner is willing to pay $90.

Now, if you’re a publisher, is it more attractive to collect $30 apiece for three books printed with disappearing ink or to collect $90 for one book that’s designed to last? As long as it’s cheaper to produce one book than three, the publisher should opt for permanence.

That’s why economists are generally skeptical about allegations of “planned obsolescence.” Every few years, someone claims that General Electric knows how to make a light bulb last 1,000 years but suppresses the technology to keep us coming back for more light bulbs. Likewise, Ann Landers is forever publishing columns about how pantyhose are intentionally designed to run so women will need a new pair every two weeks.

But a woman who buys a $2 pair of pantyhose every two weeks has demonstrated her willingness to spend $52 a year on pantyhose. If Hanes could make a pair that’s guaranteed to last a year, she’d buy it for $52. That would be a better proposition for Hanes than having to make 26 pairs to collect the same revenue. So the Ann Landers theory makes no sense.

D oes that mean there’s no such thing as planned obsolescence? No, it means that planned obsolescence occurs only under special conditions. Mistrust, for example, is a special condition. If a publisher says, “Buy this book for $90, and you’ll be able to resell it next year for $60,” a student might well respond, “How do I know you won’t bring out a new edition next year and undercut my resale market?” Unless the publisher can quell such doubts, students won’t pay premium prices for books with lasting value, so publishers won’t provide them.

(Students, of course, are in some sense a captive market, forced to buy the books their professors assign. But there must, nevertheless, be some upper limit on their willingness to pay; otherwise, textbooks would sell for an infinite price. And whatever the upper limit is, it will always be higher for a book that can be resold than for a book that can’t.)

In most instances of planned obsolescence, customers have demanded it, and firms have provided it as a service. Maybe you’d rather not spend $52 on a pair of pantyhose that might get lost at the laundry. Or maybe you don’t have $52 in your pocket right now. By letting you buy 26 shoddy pairs at $2 apiece, the manufacturer provides you with the equivalent of either insurance (against the prospect of losing your entire year’s supply of hose at once) or a loan (by allowing you to spread out your payments over an entire year).

What brings all this to mind is the recent controversy over Monsanto Co.’s development of infertile seeds–seeds that yield crops that don’t reproduce so that farmers have to buy new seeds each year. From the farmer’s point of view, the opportunity to buy infertile seeds can be a great boon. Instead of paying $100 for seed that should last 10 years, you pay $10 for new seed each year, which insures you against the possibility of a disastrous and expensive crop loss.

(Are you worried that Monsanto would charge just as much for the infertile seeds as the fertile ones? Don’t be. Surely farmers are willing to pay much more for fertile seeds than for infertile, and you can be sure that Monsanto fully exploits that willingness.)

Many high-yield hybrid seeds are infertile, though not by design. Like mules, they’re naturally infertile. Taking its lead from the software industry, Monsanto had planned to convert this bug into a feature. But in the face of considerable public pressure, Monsanto has agreed to stop developing infertile seeds.

Much of the opposition had nothing to do with planned obsolescence and everything to do with concern that the Monsanto’s infertility gene might “leap” from its seeds to fertile seeds in adjoining farms and eventually render those fertile strains infertile. Such hypothesized contamination is a legitimate concern and quite plausibly a sufficient reason to applaud Monsanto’s decision. But an unfortunate side effect is the lost opportunity to provide some socially desirable planned obsolescence.