• Reduce text

    Reduce text
  • Restore text size

    Restore text size
  • Increase the text

    Increase the text
  • Print

    Print

Nutrition transition in Vietnam: new insights from recent statistical and econometric tools

In the past 30 years, Vietnam has experienced considerable economic growth. Economic development has also brought about a transition in the country’s demographics, epidemiology and nutrition. Researchers from the Joint Research Unit for Markets, Organisations, Institutions and Actor Strategies (UMR MOISA) and the Toulouse School of Economics — Research (TSE-R) studying Vietnam’s nutritional transition have developed new econometric methods to obtain detailed insights into the nutritional transition currently underway.

Street vendor in Hanoi, Vietnam © Galyna Andrushko
Updated on 06/07/2019
Published on 04/13/2019

Background

Since Doi Moi reforms in the mid-1980s, Vietnam has gone from being one of the poorest countries in the world to a lower middle-income country. The country’s considerable economic growth has been accompanied by demographic, epidemiological and nutritional transitions. Like other countries experiencing similar changes, Vietnam now faces the double burden of malnutrition: the number of overweight people — children particularly — is increasing, while under-nutrition continues to be a problem in rural areas.

In her doctoral thesis in applied mathematics, Ms Trinh used recently developed econometric and statistical tools to obtain quantifiable evidence on the nutritional transition in Vietnam, including:

  • non- and semi-parametric econometric estimation techniques;
  • model selection methods;
  • decomposition methods in economics;
  • compositional data analysis (CODA).

The empirical analysis in Ms Trinh’s thesis was based on six biennial waves of the Vietnam Household Standard Living Survey, conducted by the General Statistics Office of Vietnam (GSO) in partnership with the World Bank. The survey contains a very detailed module about household consumption that was used to calculate per capita total calorie intake and the macronutrient shares in per capita total calorie intake in Vietnamese households.

The thesis was supervised by a professor of statistics from the Toulouse School of Economics (TSE) and a research economist from INRA. The work led to the publication of three articles demonstrating that there is still room for income-based policies to fight under-nutrition in Vietnam. The articles draw attention to the fact that when analysing changes in total calorie intake or macronutrient shares in calorie intake, due regard must be given not only to changes caused by demographic shifts, such as increasing average income and urbanisation, but also to those caused by changing dietary preferences, such as the increased preferences for sweet foods and dairy products.

Relationship between calorie intake and household income

The article published by Trinh et al. (2018a) in World Development highlights the concave relationship between per capita calorie intake and per capita income, revealing that income-based policies remain pertinent to fight under-nutrition in Vietnam. There is effectively an income threshold up to which point calorie intake increases, and after which calorie intake stabilises.

The increase in average calorie intake between 2004 and 2014 is the result not only of socio-demographic changes in the country, such as increased food spending and decreasing household size, but also because of changes in food preferences, as seen in the changes to the relationship between calorie intake and income. The article underscores the fact that the latter type of change, mostly overlooked in the current literature, must be taken into account when studying nutritional transitions.

A macronutrient analysis of changes in calorie intake in Vietnam

The article published by Trinh et al. (2018c) in Economics and Human Biology serves as a complement to Ms Trinh’s article in World Development by focusing on the specific changes in macronutrient shares in total calorie intake over the study period. Calories from proteins and fats increased not only on average, but also specifically across various population quantiles for each macronutrient group. Calories from carbohydrates remained stable on average, although they decreased for the top 10% of the population and increased for the bottom 10%. Findings from the article demonstrated:

  • the importance of accounting for changing food preferences when analysing changes to macronutrient shares in total calorie intake, both in terms of average but also population quantiles;
  • the effects of demographic changes in the Vietnamese population, largely due to increased incomes and decreasing household size, which are positive and offset the negative effect caused by changing food preferences for proteins and fats. The opposite is true in the case of carbohydrates.

Relationship between socio-economics and diet

The article published by Trinh et al. (2018b) in Statistical Methods for Medical Research looks at diet quality, as measured by the macronutrient shares in total calorie intake. To study the impact of socio-economic factors on diet quality, the article made use of recently developed compositional data analysis (CODA) techniques. In this case, the object under study is a compositional vector whose total equals one — the type of data CODA was explicitly developed to deal with. The application of CODA to dietary analysis carried out for the article was the first such example in the literature. Among many determining factors, food expenditure is an important component. The article thus studied how food expenditure elasticities of macronutrient share could be calculated and used to analyse the substitution effects between macronutrients.

Outlook

Building on this work, researchers explored the link between body mass index and diet quality in China and Vietnam. The analysis drew on CODA techniques, and used total calorie intake and macronutrient shares in total calorie intake as variables to predict variability in body mass indices. The approach made it possible to distinguish the effects of the two variables, and allowed for more nuanced analysis of the impact of variation within the variables, whether calorie intake remained stable or not. After presenting her work, Ms Trinh was awarded a prize for best work presented by a junior researcher at the Vietnam Economists Annual Meeting in June 2018.
Research on the nutritional transition in Vietnam continues, and a project is underway to study the impact of climate change on the wellbeing of Vietnamese households, and their nutritional wellbeing in particular. The project is being carried out by Michel Simioni as a part of his work as a guest researcher at the Southeast Asia office of the International Centre for Tropical Agriculture (CIAT-Asia).

Application of research findings

Trinh, H., Simioni, M. et C. Thomas-Agnan (2018a). “Assessing the nonlinearity of the calorie-income relationship: an estimation strategy - With new insights on nutritional transition in Vietnam.” World Development 110: 192-204,  https://doi.org/10.1016/j.worlddev.2018.05.030.

Trinh, H., Morais, J. Thomas-Agnan, C. et M. Simioni (2018b). “Relations between socio-economic factors and nutritional diet in Vietnam from 2004 to 2014: new insights using compositional data analysis.” Statistical Methods in Medical Researchhttps://doi.org/10.1177/0962280218770223.

Trinh, H., Simioni, M. et C. Thomas-Agnan (2018c). “Decomposition of changes in the consumption of macronutrients in Vietnam between 2004 and 2014.” Economics and Human Biology 31:259-275, https://doi.org/10.1016/j.ehb.2018.09.002.

Doctoral thesis in applied mathematics, presented in July 2018 by Huong Trinh Thi, Maths, Information and Telecommunications Graduate School at the University of Toulouse III: “Adapting recent statistical techniques to the study of nutrition in Vietnam.” http://www.theses.fr/2018TOU10010