The quest for the elusive Giffen good has taken economists to the depths of the Irish potato famine, to the poorest parts of rural China and to the cages of lab rats at Texas A&M University. Now the Giffen good has been spotted at Disney theme parks. But what do Giffen goods really tell us about the way the world economy works?
Giffen goods were first described in Alfred Marshall’s ubiquitous textbook Principles of Economics. Marshall generously gave credit to Robert Giffen, an eminent Victorian who seemed to be on every economic committee imaginable, but of whom one biographer noted, his “not inconsiderable power and prestige appears to be disproportionate to [his] actual contribution to economic science”.
No matter. Thanks to Marshall, Giffen’s idea is now in every economics textbook: that idea is that, in some circumstances, price hikes can increase demand for a product that nobody really loves.
The canonical example is the potato, which was the cheapest source of calories for subsistence farmers in Ireland in the mid-19th century. During the appalling trauma of the great potato famine, the price of those potatoes rose, and the expense crowded out yet more pricey foods such as meat and milk. The Giffen good was a trap: the more expensive the potatoes became, the less ability you had to buy anything other than potatoes.
In the extreme, the Giffen behaviour itself cannot be sustained: things are truly desperate, people consume nothing but potatoes, and so if the price of potatoes continues to rise, they starve.
For a product that is unattractive and yet dominates the household budget, this is a tale that makes theoretical sense, but it is very much a curiosity.
“Only a very clever man would discover that exceptional case,” opined Marshall’s contemporary Frances Edgeworth, “only a very foolish man would take it as the basis of a rule for general practice.”
Economists no longer believe that potatoes were Giffen goods during the great famine, so the quest for the “exceptional case” has continued. In 1990, the economists Raymond Battalio, John Kagel and Carl Kogut persuaded lab rats to drink more bitter quinine water and less sugary root beer by raising the price of the disliked quinine water. (The “price” in this case was the number of times a rat had to push a button to get a drink of the quinine.)
This was classic Giffen good territory, but it is striking how difficult it has been to observe it in humans rather than lab rats. It was not until 2008 that Robert Jensen and Nolan Miller published persuasive evidence that in the poorest parts of Hunan, China, rice was a Giffen good. Jensen and Miller had conducted a randomised trial in which some households received vouchers that reduced the price of rice. (The subsidy varied but was about 10 to 25 per cent of the normal purchase price.) In response, households consumed less rice, not more — the vouchers had increased their purchasing power enough for them to want to buy tastier ingredients. When the experiment ended and the price of rice rose again, these poor households ended up buying more of that costlier rice.
The story, I am pleased to say, does not end there. Last year the economist Garth Heutel published evidence that theme park rides could be Giffen goods. Park visitors didn’t pay cash for each ride — instead, like the lab rats, they paid in terms of the effort required to enjoy each ride.
Heutel argued that within the constraints of a day at a theme park, there was a strictly limited time budget. A visitor might spend a couple of hours queueing for a popular rollercoaster, but just 15 minutes waiting to jump on a carousel. What, then, if the carousel queue time doubles to 30 minutes? In that case, says Heutel, people might actually decide to ride the carousel more often. Like the canonical potato, the carousel was consuming so much of their time budget that they barely had time to ride anything else. Using data from four Disney theme parks in California and Florida, Heutel finds that some theme park rides are indeed Giffen goods.
Gratifying as it is to note the inventiveness of economists, I would suggest that the real lesson of Giffen goods is that strange things can happen when people are backed into a corner. Two years ago, I noted that some of the cheapest foods, such as sliced white bread and no-frills pasta, had been rising in price most swiftly after the pandemic. The point is not that sliced bread is a Giffen good, but that people feel trapped by such price movements. If the price of fancy products soars, people who buy fancy products can always switch to something simpler. But for those who are already buying the most basic stuff, there is nowhere to trade down to.
The financial chaos of the past few weeks has thrown up another intriguing example. The US dollar and US Treasuries have some Giffen-ish qualities. Think of US Treasuries as being the potato of the financial world: while the potato is a no-frills source of calories, Treasuries are a no-frills source of stability.
Normally, when the US economy is thriving, the dollar is strong, and when the US economy is languishing, the dollar is weak — but when the US economy is in real trouble, the dollar often rises again. The reason? If the US is in trouble, everyone is in trouble — and if everyone is in trouble, it’s best to be in the safest place, which is the US dollar. This “dollar smile” pattern is not exactly a Giffen good, but it is reminiscent of the Giffen good’s counterintuitive movements.
One of the disconcerting market movements in the wake of President Trump’s tariff announcement on April 2 was that the dollar and US Treasuries did not rise — they fell. I’m not sure how much Giffen and Marshall can tell us about this — except that the vibes seem unsettling. When the blight truly takes hold, bad things happen.
Or perhaps the rollercoaster is a better analogy. The ride is rickety, the passengers are puking, and no — I’m afraid you can’t get off.
Find out about our latest stories first — follow FT Weekend Magazine on X and FT Weekend on Instagram
