Optimising romance

DATING is a treacherous business. There may be plenty of fish in the sea, yet many are unhygienic, self-absorbed, disconcertingly attached to ex-fish, or fans of Donald Trump. Digital dating sites, including a growing array of matchmaking apps, are meant to help. Their design owes more to hard-nosed economics than it does to the mysteries of the heart.

In a sense, searching for a mate is not so different from hunting for a job. Jobs, like prospective partners, have their strengths and weaknesses, which makes finding the right one a matter of complicated trade-offs. Such exchanges are different from other transactions, in that both parties must be enthusiastic about the match for it to happen. A supermarket, in contrast, does not particularly care whose wallet it is draining, nor does the power company agonise about whether a customer is worthy of its watts.

Alvin Roth, who won a Nobel prize in economics for his work on market design, made a career of studying such “matching markets”, where supply and demand are not balanced by price. Instead, people transact based on information. An apple-seller can nudge down his prices until the whole cart is sold. Yet if Apple were looking to hire two workers, it would not set a salary so puny that only two people applied. The quality of new hires often matters at least as much as their salaries.

Mr Roth, who won the prize jointly with Lloyd Shapley in 2012, found that the structure of matching markets made a significant difference in determining who wound up with whom. Systems designed to elicit people’s true preferences generated better matches between hospitals and doctors, for example. But the entire medical profession has an interest in improving matches, and so can set up a national clearing house to do just that. The lovelorn must instead rely on an array of digital matchmakers.

Good matches depend on good information. Even without digital help, people usually have some inkling of how much they have in common. Cosmopolitan strivers move to New York, say, rather than sleepier cities, in part because they will meet other ambitious types with similar interests. Within New York, the places people choose to spend their time—whether Yankee Stadium or a yoga studio—determine which sorts of people they come into contact with. Because it is expensive to live in New York, and to spend time sweating in a yoga studio or swearing in the stands, people in such settings can be reasonably confident those around them are in some sense like-minded.

But one critical bit of information is missing: whether there is mutual interest. The act of asking someone out is fraught. In the non-digital world, approaching a potential partner brings the risk of awkwardness or humiliation. Digital dating reduces this cost dramatically. Apps like Tinder and Happn, for example, reveal that a user likes another only when the feeling is mutual.

The best matching markets are those that are “thick”, with lots of participants. The more people there are seeking digital dates, the greater the chance of finding a good match. Odds improve that another person in the crowd also enjoys Wagner, Thai food, or discussions about the economics of matching markets.

The wealth of information many dating sites request may help to home in on the perfect match, but if the effort involved is enough to deter potential mates from joining in the first place, then it does more harm than good. When Tinder first launched, largely to facilitate casual sex, users assessed one another based only on looks, age and gender. Simplicity worked wonders; there are 26m matches made between Tinder users each day.

The advantages of thick markets are lost, however, if they become too “congested”, with users overwhelmed by the number of participants and unable to locate a good match among them. One response is to specialise. JSwipe, for instance, caters to Jewish singles while Bumble, an app where women must initiate contact, is meant to attract feminists.

But the most popular apps seek to help their users filter possible mates using clever technology. Tinder, for example, only provides users with profiles of fellow Tinderites who are nearby, to make it that much easier to meet in person. It has also introduced a “super like” feature, which can be deployed only once a day, to allow smitten users to signal heightened interest in someone. In addition, last year it started allowing people to list their jobs and education, to help users to sort through the crowds. Users get the benefits both of a big pool of potential partners and various tools to winnow them.

Sext and the city

The emergence of matching apps, for those seeking love or theatre tickets or a lift, has certainly made once-onerous tasks more convenient. They may also contribute to more profound economic change. Dating apps could strengthen the trend toward “assortative mating”, whereby people choose to couple with those of similar income and skills. By one estimate, the trend accounts for about 18% of the rise in income inequality in America between 1960 and 2005. A recent study of online dating in South Korea found that it boosted sorting among couples by education.

Better matching may also mean bigger cities. Metropolitan goliaths have long been melting-pots, within which those early on in their adult lives link up with jobs, friends and mates. Matching apps, romantic or not, make it easier to navigate the urban sprawl and sample all it has to offer. That, in turn, should make the biggest cities relatively more attractive to young people.

Apps cannot yet make break-ups less painful. And love remains mysterious enough that even the most refined algorithms cannot predict mutual attraction with confidence. But they clearly help, judging by their legions of users. After all, it is better to have super-liked and lost than never to have super-liked at all.

^{Visit our Free exchange economics blog}