The most commonly cited example of a Giffen good is that of the Irish potato famine in the 19th century. During the famine, as the price of potatoes rose, impoverished consumers had little money left for more nutritious but expensive food items like meat (the income effect). So even though they would have preferred to buy more meat and fewer potatoes (the substitution effect), the lack of money led them to buy more potatoes and less meat. In this case, the income effect dominated the substitution effect, a characteristic of a Giffen good.
However, critics of this textbook example point to the fact that Ireland was in the grip of a severe famine at the time, and because of the shortage of potatoes – which led to higher prices – it was unlikely that people could have consumed more of them.
A more recent – and perhaps better – example of a Giffen good is offered by a 2007 study by Harvard economists Robert Jensen and Nolan Miller. Jensen and Miller conducted a field experiment in two provinces in China – Hunan, where rice is a dietary staple, and Gansu, where wheat is the staple. Randomly selected households in both provinces were given vouchers that subsidized their purchase of the staple food.
The economists found strong evidence of Giffen behavior exhibited by Hunan households with respect to rice. Lowering the price of rice through the subsidy caused reduced demand by households for rice, while increasing the price (by removing the subsidy) had the opposite effect. The evidence with regard to wheat in Gansu was weaker because two of the basic conditions for Giffen behavior were not fully observed, i.e. that the staple good should have limited substitution, and households should be so poor that they consume only staple foods.