A study published in World Development investigates the effect of food price subsidies in India on nutrition among the poor. Such a study is particularly important because of the high incidence of undernourishment in developing countries and because of the high political and public support that food price subsidy programs receive in countries across the world despite widespread evidence that these programs suffer from poor management and targeting. There exists a longstanding debate among nutritionists and policy makers on the extent to which nutrition among the poor in developing countries improves through a food price subsidy and increased income. Previous studies have shown mixed results in both cases. This study seeks to fill this knowledge gap by investigating both the effect on nutrition of an increase in food grain subsidy in India and the effect of an increase in income resulting from the subsidy program.
The study focuses on an increase in food price subsidy to poor families resulting from the introduction in 1997 and expansion in 2002 of India’s Targeted Public Distribution System (TPDS). This subsidy is targeted at households with incomes below the official poverty threshold, which allowed them to purchase, at approximately two thirds of the market price, 10 kilograms of rice or wheat per household per month from 1997 onwards, this was expanded to 35 kilograms in 2002. The study investigates how the increase in food price subsidy after the TPDS was implemented and expanded affected per capita calorie, protein and fat intake, total food consumption and consumption of specific food items in poor families in rural India. The study is primarily based on data from three rounds of National Sample Surveys for the years 1993–94, 1999–2000 and 2004–05. These are nationally representative surveys covering between 120,000 and 125,000 households each year. The last two rounds were conducted about 2 years before and 2 years after the expansion of the TPDS. The surveys provide information on the quantities and values of wheat and rice supplied by ration shops as well as purchased in the open market. The study estimates underlying changes in nutrition by converting detailed data on food consumption into three nutrient intake categories: calories, protein, and fat. The amount of each food item consumed is converted to average daily nutrient intake.
Geographically, the study analyses consumption patterns across two groups of districts, allowing for comparisons between changes experienced in consumption patterns and overall nutrition. The study first conducts an empirical analysis, estimating the effect of price subsidies on incomes in 66 districts where wheat and rice are the main staple commodities. In these districts, the subsidy will purely have an income effect on households receiving the subsidy as it allows them to buy the same amounts of wheat and rice at discounted prices. This is followed by an analysis of the effect of price subsidies on nutrition on poor households in 15 districts where coarse grains (maize, jowar, ragi, and bajra) are the main staples and are generally cheaper than subsidized wheat and rice. In this analysis, the study specifically estimates the effect of the price subsidy on nutrition and consumption patterns due to the perceived income increases.
The study finds a number of important trends and changes related to per capita consumption and calorie intake in both groups of districts. In high wheat- and rice-consuming districts, the two staple grains accounted for 65 percent of the calorie intake in the pre-TPDS period and 67.5 percent in the post-TPDS period. Conversely, in high coarse grains consuming districts, households derived almost 50 percent of their calorie intake from coarse grains and 19 percent from wheat and rice. In the post-TPDS period, the share of wheat and rice increased to 26 percent and of coarse grains fell to 41 percent. Overall coarse grains were the cheapest source of calories in the pre-TPDS period and remained so in the moderate wheat- and rice-consuming districts after TPDS implementation. In high wheat- and rice-consuming districts subsidized wheat and rice became the cheapest sources of nutrition after TPDS implementation.
Furthermore, the study found that in the pre-TPDS period households in high wheat/rice-consuming districts had a calorie consumption that was 7 percent higher than the per capita calorie intake in moderate wheat/rice-consuming districts, this gap was reduced to 2 percent in the post-TPDS period. The study notes consistent increases of the proportion of calories derived from wheat and rice in both districts from about 5 percent prior to TPDS expansion to between 10 and 15 percent after TPDS expansion. Similarly, over the course of the study, households increased their total share of calories from the more expensive sources, namely pulses, milk products, edible oils and meat by about 1 percentage point.
The study estimates the effect of food price subsidy on nutrition through an empirical model where nutrition is measured through per capita daily calorie intake, per capita daily protein intake and per capita daily fat consumption. Based on these results, the study analyses whether these changes in consumption and nutrition patterns were caused by an increase in food price subsidy. This analysis estimates that increases in income resulting from the food price subsidy changed consumption patterns but had no effect on nutrition. More specifically, the analysis shows that a 100 percent increase in the subsidy amount, which is equivalent to a 0.54 percent average increase in income for poor households, increased calorie intake from wheat and rice by 2 to 4 percent; increased calorie intake from sugar and sugar products by 21 to 28 percent, lowered calorie intake from coarse grains by 40 to 75 percent, and lowered intake from other foods by 2 to 5 percent. Despite these consumption changes, the overall calorie intake remained virtually unchanged.
The analysis also shows that households allocated some of the increase in income from the food price subsidy to expenditures on non-food items. Overall, estimates of the effect of food price discount on calorie and protein intake are negligible and statistically insignificant; there is weak evidence of a modest increase in fat intake. The study indicates that the relative decline in the price of wheat and rice, due to the subsidy, changed consumption patterns towards increased consumption of wheat and rice and lowered consumption of coarse grains. There is also some evidence of an increase in consumption of sugar and edible oil which are more expensive sources of calories than wheat, rice, or coarse grains.
The paper concludes by highlighting two key implications that originate out of these findings. The results suggest that the poor generally have other constraints to more diverse nutrition and that households generally increase consumption of the subsidized grains with additional actual or perceived income changes from price subsidy programs. The second key implication is that while the food price subsidy program fails to improve nutrition among the poor, it changes consumption patterns that may in turn have a substantial impact on agriculture markets, this is likely to be an unintended and perhaps undesirable effect of the policy. The results from this study also imply that the further expansion of India’s food subsidy program is not likely to significantly impact nutrition rates in India.
The full paper can be accessed here
By Bas Paris, IFPRI