Week 1: Love as a Market

Orientation — why this course treats romance as a market, what that buys you, and what it costs

21 February 2028

Open the slides

A market is any arrangement where scarce things get allocated among people who want them, under rules nobody fully controls. Housing is a market. University admission is a market. This course’s claim is that courtship is one too, and that the tools economists built for the first two — search costs, signaling, screening, stable matching, mechanism design — say something true about the third.

The claim is not that romance is “just” economics, or that the vocabulary of markets improves anyone’s love life. It’s narrower and, we think, more interesting: wherever information is unevenly held and outcomes depend on what other people do, markets fail in specific, nameable ways. The semester is a tour of those failures, in order of how badly they hurt.

A market that actually failed#

Take a market that isn’t romance at all. Before 1950, US hospitals hired newly graduated doctors into residency positions through an informal process: interview, then wait for an offer. Competition for the strongest candidates pushed each hospital to make its offer earlier than the last one did — by 1945 offers were going out before students had finished the relevant year of study. Agreeing on a fixed date didn’t fix it; it just moved the deadline everyone raced to beat. By 1949 hospitals were making “exploding” offers — decide within a day or two, with no way to know what else might have come along and no time to compare. Nobody in this market was behaving irrationally. Every hospital jumping the gun was responding correctly to every other hospital already jumping the gun. The market unraveled anyway, and stayed unraveled until a redesign — a centralized clearinghouse taking ranked preferences from both sides at once — replaced it in the early 1950s (Roth 1984). Week 8 runs that redesign’s actual algorithm by hand.

Courtship has its own version of the same failure, and it doesn’t need a label to be recognizable: the pressure to “lock things down” before you’ve had a real chance to compare, because taking the time to compare reads, to the other person, as a lack of commitment rather than as due diligence. Nobody involved has to be acting badly for the whole arrangement to end up earlier and less informed than anyone actually wanted.

Outline#

• what makes something a market, and what makes a market fail
• why “love isn’t a market” is the position this course exists to test, not the one it assumes
• a market that unraveled for real, and what fixed it
• the shape of the semester: search, screening, matching, then living with what you matched into
• what to bring to Tutorial 1

One thing this framework won’t explain#

Redesigning the residency market fixed the timing — it didn’t ask whether earlier certainty was something residents actually wanted for its own sake, only whether the market allocating it was efficient. This course’s model is built the same way. It can tell you when a courtship market is unraveling and roughly what a better mechanism would look like, but it has no term for whether the anxiety of being rushed, or the relief of finally being chosen, is worth anything beyond the inefficiency it’s a symptom of. That gap — between what a market model prices and what it doesn’t even have a word for — isn’t a footnote. It’s Week 12’s actual subject, and this is the first of several places this semester where you’ll be asked to notice the framework going quiet.

Further reading#

Real economics, not romance, but the case study above is drawn straight from it:

• Roth, A. E. (1984). “The Evolution of the Labor Market for Medical Interns and Residents: A Case Study in Game Theory.” Journal of Political Economy, 92(6), 991–1016.
• Roth, A. E., & Xing, X. (1994). “Jumping the Gun: Imperfections and Institutions Related to the Timing of Market Transactions.” American Economic Review, 84(4), 992–1044.
• Stanford Magazine, “The Visible Hand” — a readable account of Roth’s work on unraveling markets, for anyone who’d rather start with the story than the journal article.