A Fresh Look at the Nutrition Transition in Vietnam using Semiparametric Modeling

Abstract

Policies aimed at reducing starvation and redressing nutritional deficiencies remain among the most widely accepted policies in the world. These policies can take many different forms, from subsidized prices of basic foodstuffs to cash transfers, and their effectiveness depends on the existence of a sensitivity of food demand to income variation and its magnitude. This paper revisits the issue of estimating the relationship between calorie intake and income. We present and compare estimates of this relationship for Vietnam which has undergone profound economic changes over the last 30 years. After estimating semiparametric generalized additive models are estimated, we compare their performances are compared to the performance of the classical double log model using the recently proposed revealed performance test. This methodology is implemented using successive waves of the Vietnam Household Living Standard Survey. The application delivers some new and interesting insights on nutritional transition in Vietnam between 2004 and today. The analysis focuses not only on the comparison of the general shape of the estimated curves but also on the decomposition of the evolution of average calorie intake in terms of changes in the structure of the surveyed sample (greater urbanization, for example) and changes in preferences as reflected by changes in the estimated curves over time.

Keywords

Replaced by

Michel Simioni, Christine Thomas-Agnan, and Thi-Huong Trinh, “Assessing the nonlinearity of the calorie-income relationship:an estimation strategy - With new insights on nutritional transition in Vietnam”, World Development, vol. 110, 2018, pp. 192–204.

Reference

Michel Simioni, Christine Thomas-Agnan, and Thi-Huong Trinh, “A Fresh Look at the Nutrition Transition in Vietnam using Semiparametric Modeling”, TSE Working Paper, n. 17-842, September 2017.