Determinants of Dietary Quality: An Empirical Study Using Data From Bangladesh

Abstract

This study investigates the determinants of dietary quality using data from a national household expenditure survey conducted in Bangladesh in 2000. Malnutrition is particularly severe in Bangladesh. Nearly one half of all mothers in rural areas are undernourished. The child malnutrition rate is among the highest in the world, with 55 percent of children under five years being stunted. This malnutrition is partially caused by a diet that has a high concentration of starchy staples and relatively little consumption of protein and micronutrient rich foods. Understanding how household income, food prices, education and household demographic characteristics determine food consumption patterns, and thus protein and micronutrient intakes, can provide crucial information for designing policies and intervention programs to improve food and nutrition security in Bangladesh. Reduced-form equations for household protein availability and demand for selected protein and micronutrient rich foods (meat, fish, pulses, vegetables, and fruits), as well as for dietary diversity, are estimated after testing and, if necessary, correcting for endogeneity of income. A reduced form equation for household energy availability is also estimated in order to asses whether the determinants of dietary quality differ from those of dietary quantity, the traditional focus of food security analyses. Using a variety of measures of dietary quality to investigate the role of its determinants, the study identifies the following principal determinants of dietary quality: income, education, gender of the household head, number of adult equivalents in the household, and prices of food commodities. These factors were found to have statistically significant and quantitatively strong impacts on the quality of diet. This study has found strong evidence that income and education are the most robust determinants of dietary quality. Women’s education was found to have particularly strong influence. While men’s education plays an important role in improving household dietary quality, its influence is not as strong as women’s. These results are consistent across the various measures of dietary quality employed as dependent variables. Female headed households are found to be at a disadvantage when it comes to both dietary quality and quantity. Limited mobility of women in Bangladesh may be one of the reasons that prevent them from going to market and getting access to a wide variety of foods that could improve the dietary quality. Those households whose main source of income is casual labor may get wages in food, presumably rice, which could constrain them from buying other types of foods. The disadvantage could also be due to the gender differences in knowledge and preferences and/or the ability of female headed households to put them into practice. Further research is needed to investigate this finding. As expected, the demand for protein, fish, meat, vegetables, fruits, and pulses are sensitive to price changes. The income elasticity computed at sample mean suggests that vegetables and pulses are necessities in Bangladesh while meat, fish and fruits are luxury goods. In answer to the second question asked “Do the determinants of dietary quality and quantity differ?,” this study concludes that income, gender of household head, number of adult equivalents in the household, and food prices are common determinants of both dietary quality and dietary quantity. While analysis reveals that men’s education has some limited influence (at the 10 percent level of significance) on energy availability (i.e., dietary quantity), education in general only has explanatory significance as a determinant of dietary quality. Given the crucial role of income and education in increasing access to sufficient and high quality food, this study recommends the design and implementation of well targeted poverty reduction and education programs. Education raises awareness about commonly consumed sources of micronutrients and protein, and thus has the potential for improving intake. Promoting women’s education in particular could be a significant policy tool for government and non-government organizations in addressing the issues related to dietary quality. This will require a departure from previous food security policies which tended to focus predominantly on dietary quantity. Vegetables, the most accessible non-staple food for the poor, have seasonal availability. As such, they are highly subject to price volatility, further disrupting dietary sufficiency and quality for the poor. In order to ensure consistent availability of nutrient rich vegetables throughout the year, crop diversification programs may be successful in promoting and helping to increase the cultivation of seasonally appropriate crops. Their increased supply in the market may translate into more stable and affordable prices. This study also advocates promoting plant breeding strategies for increased micronutrient content and bioavailability in staple foods. The potential of these technologies are enormous in improving dietary quality of populations relying mainly on cereal staples. Although not a direct conclusion from the analysis of the thesis data, improving access to resources, technology and information – especially for women – in support of homestead crop production has the potential to improve nutritional quality as well as general food security. Many rural households in Bangladesh are engaged in homestead gardening to some degree. It is the major source of vegetables for poor households. Therefore, targeting homestead production for technical and informational support, in conjunction with crop diversification programs, may be a direct route to enhancing dietary quality. Furthermore, since homestead production is often under the domain of women, increased support may help to address the disadvantages faced by female-headed households.