The Quest to Find Privacy Market Failure

by on March 19, 2011 · 0 comments

I guess the search for market failure in the privacy area is interesting to me. I wrote about it the other week too. It’s nice that those who prefer regulation feel obligated to justify that preference. It’s acknowledgment of the fact, increasingly well-accepted worldwide, that functioning free markets do a better job of discovering and satisfying consumers’ interests than any other method for organizing societies’ resources.

A recent market failure blog post called “Privacy and the Market for Lemons, or How Websites Are Like Used Cars,” seems to have piqued Adam’s interest. (See the comments.) In it, privacy and anonymity researcher Arvind Narayanan makes the case for privacy market failure. (Evidently, it’s an argument that others have made before.)

“In the realm of online privacy and data collection,” he says, “information asymmetry results from a serious lack of transparency around privacy policies. The website or service provider knows what happens to data that’s collected, but the user generally doesn’t.” Several economic, architectural, cognitive and regulatory limitations/flaws “have led to a well-documented market failure—there’s an arms race to use all means possible to entice users to give up more information, as well as to collect it passively through ever-more intrusive means.”

Alas, there’s no link at “well-documented.” I would like to see that documentation. But more importantly, what Narayanan appears to be speaking of as market failure—an arms race to get more information from Web users—is not one. That’s market action that Narayanan doesn’t like.

So where’s the market failure? Is it in the “impossibility” of navigating the Web aware of the privacy policies of each Web site? That’s certainly very hard to do. The premise is that markets need information symmetry for them to function. There’s a profoundly difficult question lurking here: Just how symmetrical does information have to be? 42?

Luckily, markets don’t require information symmetry. They work because both parties to nearly every transaction make themselves better off by their own reckoning. In the aggregate, market transactions make everyone better off without making anyone worse off. As with the information symmetry ideal, there’s no need for perfection in transacting, either. If some actors regularly make “bad” trades, but still end up better off on the whole, the market is working for them, too. (Their consistent failure to perceive their own self-interest may ultimately require them to seek the charity of those of us who do, but think carefully before depriving all consumers of the market’s opportunities just to protect some small number of incompetents.)

This logic is not defeated when a consumer gets less in a transaction than he or she thought, or bears a cost he or she didn’t know about. (“I thought I was getting a car that gets 35 mpg, but it only gets 31,” or “I thought the site would collect no information, but it associated my browser and IP address with an interest in golf.”) It is only defeated when the systematic operation of the market will reach sub-optimal results.

Will a series of transactions drive someone to their death of starvation or some other harm or ruin? Will the gears of commerce freeze? Market failure occurs when the rules, signals, and sanctions in and around a given marketplace would cause preference- and profit-maximizing actors to reach a sub-optimal outcome.

Narayanan believes there is a sub-optimal outcome—people are giving up too much privacy unaware of the consequences. I agree that people do not protect their privacy enough online. But correlating market behavior we rue to an “information asymmetry,” where every last detail of what happens to data isn’t clear, has not made the case for market failure.

The same “asymmetry” exists in the real world. I have no idea what my neighbors will do with the information they observe about my comings and goings or what my friends will say behind my back. Does this social market failure demand or justify a regulatory solution? Should I get a privacy notice detailing their plans? No. I’m on notice that stuff like that can happen with anyone and everyone I encounter. I go outside, subjecting myself to observation, because I am better off for doing so even though I pay an unknown privacy cost. Likely, Web surfers are similarly better off by their own reckoning despite paying a similar, unquantified privacy cost. The fact that they could pay less is imperfection, not failure.

As you can see, I love analogical thinking, but the analogy between the online environment and a used car market does not work for me. I also don’t see that the theory of the “lemons market” has borne itself out, either online or in the market for used cars. A lemons market should lead to stagnation—a no-trade equilibrium—but we have a robust online information environment and used car sales continue to flourish (in part because of market innovations).

In the end, Narayanan’s post is . . . a lemon. That is not a market failure—just a computer scientist who I think has hashed the economics a bit. It’s not my field either, so I welcome comments that educate and elucidate things for all of us.

Previous post:

Next post: