Provincial-level agencies of the Chinese National Bureau of Statistics randomly selected 1,300 households in Hunan and Gansu provinces to participate in this study, and then randomly assigned them to either a comparison group or one of three treatment groups. In the three treatment groups, households received vouchers entitling them to a price reduction of 0.10 yuan, 0.20 yuan, or 0.30 yuan respectively off the price of each 500 grams of the staple good (rice in Hunan and wheat flour in Gansu) over the next five months. These subsidies represented substantial price reductions: the average pre-intervention price of rice in Hunan was 1.2 yuan per 500 grams, while the average price for wheat flour in Gansu was 1.04 yuan per 500 grams. The month’s supply of vouchers was distributed at the start of each month, with each household receiving vouchers for 750 grams per person per day—about twice the average per capita consumption of these staples prior to the intervention. All vouchers remained valid until the end of the intervention.
Data was collected through a standard income and expenditure survey, as well as a 24-hour food recall diary that was to be completed by each household member. Respondents were asked to report everything they ate and drank each day by listing the components of all things consumed in the diary. Diary entries were then compared to official data on the nutritional value of foods in order to more fully understand any changes in nutritional intake caused by this intervention.
Results and Policy Implications
There was no evidence that the consumer price subsidy improved nutrition, and results suggest that it may have actually reduced the intake of calories and several important vitamins and minerals in Hunan. In Hunan the price subsidy on rice caused households to exhibit Giffen behavior—they cut back on their consumption of rice, vegetables (primarily cabbage and spinach), pulses (bean curd and tofu), and fats (primarily cooking oil), and increased their consumption of seafood. A great deal of calories were lost from the reduced consumption of rice and pulses in particular, and only part of this loss was recovered through the increased consumption of seafood, leading to a net decline in calories consumed. Additionally, large declines in other nutrients such as Vitamin A resulted from the reduced consumption of vegetables such as spinach.
In Gansu, there was less evidence of Giffen behavior; but nutritional intake did not increase for any nutrient group. One interpretation of this result is that as in Hunan, subsidized households in Gansu sought to add taste to their diets, but since they still could not afford meat or seafood, they opted for the lower-cost option of increasing edible oils, which are often eaten with bread made from wheat flour, the staple good. Since increased consumption of oil did not correspond to decreased consumption of other foods, overall average caloric intake did not decline; however, because oil is largely devoid of nutrients, nutritional intake did not increase either.
While food-based welfare programs such as price subsidies enjoy certain advantages such as superior targeting, ease of administration, and political acceptability, the primary justification for choosing subsidies over other policy instruments is the presumption that they can be counted on to improve nutrition. If Giffen behavior, of the kind of observed in this study, is widespread, then policymakers may need to reconsider when food-based subsidies are appropriate.
2006 - 2007