The Undercover Economist

How To Sell a Tuna

What fish markets teach about the economy.

Mmm. Fish. Click image to expand.

Economists tend to study “the market” in the abstract, but some markets are anything but abstract. The Aalsmeer auctions in the Netherlands burst with the color of more than 20 million flowers a day, while Tokyo’s early-morning Tsukiji fish market is enthralling enough to compete for tourist attention with the nightclubs of Roppongi. (The stink of alcohol in the tourist pens of Tsukiji suggests that many visitors have combined both experiences to make a night of it.)

These celebrated markets deserve attention from economists as well as tourists. A lot of economics assumes away the details of how market transactions actually work. That is a shame; the late John McMillan, an economist at Stanford, argued that markets will only work well if they are well-designed. The whole idea of a market is to allow gains from trade to take place; a badly thought-out market will often fail.

There is more than meets the eye even to a simple transaction such as buying fish.

Economist Kathryn Graddy once spent a month shadowing a trader at the Fulton Fish Market in New York, rousing herself in the early hours and paying protection money to park her car in a safe spot near the market. She discovered that market traders appeared to charge different prices to different ethnic groups. In particular, Asian buyers tended to win keener prices than white buyers. The likely explanation was that the Asian buyers had more price-sensitive customers in Chinatown.

Still, this was a surprising result.

A competitive market should eliminate the traders’ ability to charge white customers more, and the Fulton Fish Market should be competitive: It is the world’s second largest. (Tsukiji, more than five times larger, is in a league of its own.) Graddy discovered that even at Fulton, perfect competition was elusive.

Nevertheless, the better markets work, the more transactions they make possible. Some transactions are so tempting as to be irrepressible, much as we might want to repress them: Schoolchildren will be supplied with chocolate bars whether or not the school shop stocks them. Other transactions are inspiring in their resilience. In the chaos of Mogadishu, Somalia, private merchants have worked out ways to get the market for electricity to work without metering; customers pay according to the number of bulbs in their home or workshop.

Yet many transactions require a lot more support than that. Modern transaction infrastructure includes the legal system, credit registries, Visa, and corporate accounts. Buying shares may not seem very sophisticated, for instance, but when you buy a share you are lending money in exchange for a claim on the future profitability of a firm in, say, China.

This everyday financial transaction requires layers of accountants, regulators, brokers, and lawyers to make it possible.

Much economic progress consists of reducing transaction costs to allow more sophisticated transactions to take place. Although transaction costs have fallen, modern economies are ever more consumed by them. John Wallis and Nobel laureate economist Douglass North once tried to tot up transaction costs by looking at the size of sectors such as accountancy, law, retail, management, policing, and so on. They found that the transaction-cost sector doubled its share of the U.S. economy, from just over a quarter in 1870 to just over half in 1970. Transactions don’t come cheap, but since we were much richer in 1970 than in 1870, all this spending seems to have been worth it.

That is worth remembering. Transaction costs often seem like waste, chaos, or bureaucracy, but without all those lawyers, accountants, secretaries, and sales staff, a modern economy would be impossible. If only all transaction costs could be as thrilling a spectacle as Aalsmeer or Tsukiji.