Elsevier

Social Science & Medicine

Volume 232, July 2019, Pages 58-76
Social Science & Medicine

Family income and nutrition-related health: Evidence from food consumption in China

https://doi.org/10.1016/j.socscimed.2019.04.016Get rights and content

Highlights

  • Five important mechanisms of income impact on BMI and overweight are highlighted.

  • The endogeneity of income is addressed by using minimum wage as an instrument.

  • Rising income increases the adults' BMI and the propensity to be overweight.

  • Dietary diversity plays the most significant role in explaining the income impact.

Abstract

With increasing family income, the prevalence of overweight has risen and become a serious threat to individual health and a major public health challenge in China. This study attempts to shed light on the mechanism of income impact on the adult health outcomes of BMI and overweight through five potential channels: nutritional intakes, dietary diversity, dietary knowledge, food preference, and dining out. Using the panel data from the China Health and Nutrition Survey (CHNS), we investigate the causal relationship between income and health by considering the changes in the minimum wage as a valid instrument to address the endogeneity problem of income in health estimation. The results indicate that rising income increases the adults’ BMI and the propensity to be overweight; approximately 15.58% and 16.20% of income impact on BMI and overweight could be explained by the five channels considered, respectively. Among the five channels, dietary diversity plays the most significant role in explaining the income impact. We observe significant heterogeneity in income-BMI gradients across various income quantiles and subsamples. Specifically, income-BMI gradients tend to increase along with income percentiles, and income has a significantly positive impact on BMI and overweight for the male sample but it shows no significant impact for the female sample.

Introduction

With the rapid changes in urbanization, economic growth, technical change, and culture, remarkable changes in structures of diet and body composition have been indicated by extensive literature (Popkin and Ng, 2007), especially in low and middle income countries (Abdulai, 2010; Misra and Khurana, 2008; Popkin and Ng, 2007; Popkin, 2015; Popkin and Du, 2003). For instance, in the developing countries diets are shifting to more fats, more added caloric sweeteners, and more animal source foods (Popkin and Ng, 2007; Popkin and Du, 2003), as described by the process of nutritional transition (Popkin, 1993, 1999). Particularly, China has travelled along the path of economic transformation, and its remarkable progress has had important implications for income growth over the past four decades (Brandt et al., 2008). Rising income has bestowed many benefits on households in China and has facilitated poverty alleviation both regionally and nationally (Zhang and Donaldson, 2008). It has been indicated that consumers have experienced a remarkable nutrition improvement and a dramatic dietary change in China (Tian and Yu, 2015), and physical activity (Monda et al., 2007). This gives rise to the prevalence of nutrition related health issues in China, and it has been reported that approximately 39.2% of adults in China aged 18 and older are estimated to be overweight (Zhou et al., 2017). Such trends pose serious threats to individual health as they increase the risk of noncommunicable disease (Shimokawa, 2013; Tafreschi, 2015), and higher health costs not only for households but also for the entire nation. One available estimation by Popkin et al. (2006) shows the future health cost of the overweight epidemic (and the direct consequences thereof) will approach 9% of China's GDP by 2025.

It has been well documented that rising income is associated with certain negative health outcomes along the nutritional transition, such as higher rate of being overweight (Tafreschi, 2015; Bakkeli, 2016), while the mechanisms behind or the exact channels through which income affects individuals’ body mass index (BMI) (calculated by weight in kilograms divided by the square of height in meters), and being overweight have been much less studied. A clear understanding of these channels has profound consequences not only for individuals to improve their health but also for policymakers to improve public health in China, as reducing income is not a rational option to reduce the rate of being overweight. It is also important to identify which channel is the most important one in transforming income growth into being overweight so that policymakers can focus on it when making policy changes in an effort to enable people to enjoy both higher living standards and better health.

The international literature has identified several factors that may be associated with both rising income and health outcomes of increasing BMI and prevalence of being overweight. One of the most important effects of income growth is the increase in the quantity of foods consumed, which could be measured by nutritional intakes. Generally, nutritional intakes assess the consumption of the three macronutrients: carbohydrate, fat, and protein (Mendez et al., 2005). It is a conventional belief that low nutritional intakes are a consequence of low income. However, the literature has not achieved a conclusive agreement on the extent to which income drives nutrient consumption. For instance, Skoufias et al. (2009) estimate income elasticity for various macro- and micronutrients in rural Mexico and find mixed results. They obtain positive income elasticities for fat but negative income elasticities for protein for the poorest households, while another study finds that in China, higher income tends to raise nutritional intakes of protein and fat but decrease intakes of carbohydrate in past decades (Huang and Gale, 2009). Thus, this concept must be reconsidered in the analysis of the channels through which income growth leads to increasing BMI and the propensity to be overweight.

Nutritional intakes, however, reveal limited information about diet quality and the associated health consequences (Doan, 2014). Investigating the consumption of calories or individual nutrients can provide only a partial understanding of the structural changes in diet quality and diet-related issues that accompany the nutrition transition. As a qualitative measure of food consumption, some studies have shown that dietary diversity can be used to reflect individuals’ access to a wide variety of foods and is also a good proxy of the nutrient adequacy of the diet (Morseth et al., 2017; Torheim et al., 2004; Vandevijvere et al., 2010). The empirical literature on dietary diversity has been consistent in proving positive income effects on diet variety. For instance, Moon et al. (2002) find a positive linear income effect on diet diversity in Bulgaria, and another study finds similar results in Germany (Thiele and Weiss, 2003). A more recent study by Doan (2014) indicates a significant and positive income effect on dietary diversity in China from 2004 to 2009. Nevertheless, few studies have investigated whether dietary diversity serves as a channel through which income has an influence on adult health consequences.

In addition, dietary knowledge is another channel considered to transform rising income into increasing BMI and prevalence of overweight. We expect a strong link between family income and dietary knowledge because rising income gives individuals a greater possibility of obtaining more sources of information regarding nutrition and health, such as dietary knowledge (Clément and Bonnefond, 2015; Xie et al., 2003). Individuals with higher incomes are more likely to have access to the internet, podcasts, classes (Zhou et al., 2014), and mobile phones, and such access has been shown to significantly improve access to information and dietary quality (Sekabira and Qaim, 2017). Internationally, improving dietary knowledge has been shown to help people adjust their eating habits and exercise behavior in ways that keep them from becoming overweight (Bonaccio et al., 2013; Clément and Bonnefond 2015; Nayga, 2000; Wagner et al., 2016). As far as we can tell, no study has specifically investigated how income affects individuals' dietary knowledge, which in turn could affect individuals’ health.

Food preference and dining out could also transfer income effect to nutrition-related health. With the development of the economy and increasing incomes, especially, in China, the food preference has been shifting away from high-carbohydrate food towards dense high-energy food (Batis et al., 2014; Clément and Bonnefond 2015; Curtis et al., 2007; Du et al., 2004); these changes might lead to prevalence of overweight. As one of the lifestyle changes, dining out has been indicated to have significant association with increasing income (French et al., 2010; Liu et al., 2015). These changes may also contribute to remarkable increase in BMI and being overweight since people who are regularly dining out, normally, have unhealthy consumption habit and less quality of the food (Machado-Rodrigues et al., 2018; Watts et al., 2017).

The overall goal of this study is to understand the relationship between rising income in China and the health outcomes of increasing BMI and prevalence of overweight to help policymakers formulate policies to address this rising public health concern. To achieve this goal, we have three specific objectives. First, we examine the impact of income on health by using minimum wage as a potential instrument to address the endogeneity of income in BMI and overweight estimation. Second, we seek to understand the income effect on the various channels—nutritional intakes, dietary diversity, dietary knowledge, food preference, and dining out. In this article, we focus solely on food consumption as the research perspective to detect how family income influences health through various aspects of food consumption. Finally, we illustrate how and to what extent income affects individuals’ health through the potential channels considered by gradually decomposing the overall income effect on BMI and being overweight.

The existing literature on adult health relies mainly on the subjective measure of health by using binary or ordered categorical variables that fail to meet the requirements of heterogeneous income gradients. As BMI is continuous in nature, this paper also highlights the heterogeneous association between family income and adult health using unconditional quantile regression (Firpo et al., 2009) and a panel structure of the data. This type of econometric analysis helps identify which subgroups of adults are likely to improve or worsen their health when family income increases or decreases. Additionally, we also examine the possibility of heterogeneous income effect on male and female. The results of quantile regression show that in general, the income effect on health tends to increase from the lower quantile to the higher quantile; the results also demonstrate that family income contributes significantly to BMI and overweight for the male sample but is insignificant for the female samples.

In the next section, we introduce our econometric modelling approach. Section 3 briefly presents the data, and section 4 discusses the empirical results and accounts for the distribution of BMI. The last section concludes.

Section snippets

Benchmark model for the relationship between adult health and family income

To investigate the relationship between adult health and family income, we start with two benchmark models. As aforementioned, BMI is one of the most important indicators measuring an individual's health. It is estimated using the estimation strategy by Goode et al. (2014) as follows:BMIit=α0+β0logMit+γX+εitwhere Mit is the family income inflated to 2011, and β0 indicates the change in BMI when the income changes by 1%. X is a vector of control variables, including gender, Hukou, age, age

Data

The dataset used for this study is from the China Health and Nutrition Survey (CHNS), which is an international collaborative project between the National Institute of Nutrition and Food Safety at China Centers for Disease Control and Prevention and the Carolina Population Center, University of North Carolina at Chapel Hill. The CHNS is longitudinal and includes nine waves of 1989, 1991, 1993, 1997, 2000, 2004, 2006, 2009 and 2011, consissting of 9 provinces (Heilongjiang, Liaoning, Jiangsu,

Empirical results

All models are estimated with an RE panel and a pseudo-fixed-effects estimator. The RE estimator assumes that logMit is uncorrelated with any unobserved factors that may also influence the outcome variables. However, as individuals self-select into income variation activities, this assumption may be violated, which could lead to biased estimates. Therefore, in addition to the RE estimates, we also use a pseudo-fixed-effects estimator, as proposed by (Mundlak, 1978). The MK estimator includes

Conclusion

With the substantial increase in family income, the prevalence of overweight has risen and has become a serious threat to individual health and a major public health challenge in the transitional economy of China. After using minimum wage as a valid instrument to address the potential endogeneity of income in health estimation, this study attempts to shed light on the impact of family income on the adult health outcomes of BMI and overweight through the potential channels of nutritional

Funding

This research uses data from the China Health and Nutrition Survey (CHNS). We thank the National Institute for Nutrition and Health, China Center for Disease Control and Prevention, Carolina Population Center (P2C HD050924, T32 HD007168), the University of North Carolina at Chapel Hill, the NIH (R01HD30880, DK056350, R24 HD050924, and R01-HD38700) and the NIH Fogarty International Center (D43 TW009077, D43 TW007709) for financial support for the CHNS data collection and analysis files from 1989

References (78)

  • H. Liu et al.

    Household composition, income, and food-away-from-home expenditure in urban China

    Food Policy

    (2015)
  • J. Liu et al.

    Access to variety contributes to dietary diversity in China

    Food Policy

    (2014)
  • Z. Liu et al.

    Can land transfer through land cooperatives foster off-farm employment in China?

    China Econ. Rev.

    (2017)
  • M.A. Mendez et al.

    Overweight exceeds underweight among women in most developing countries

    Am. J. Clin. Nutr.

    (2005)
  • K.L. Monda et al.

    China's transition: the effect of rapid urbanization on adult occupational physical activity

    Soc. Sci. Med.

    (2007)
  • K. Ogundari et al.

    Examining the heterogeneity in calorie–income elasticities: a meta-analysis

    Food Policy

    (2013)
  • B.M. Popkin

    Urbanization, lifestyle changes and the nutrition transition

    World Dev.

    (1999)
  • Barry M. Popkin et al.

    Dynamics of the nutrition transition toward the animal foods sector in China and its implications: a worried perspective

    J. Nutr.

    (2003)
  • H. Sekabira et al.

    Can mobile phones improve gender equality and nutrition? Panel data evidence from farm households in Uganda

    Food Policy

    (2017)
  • S. Shimokawa

    When does dietary knowledge matter to obesity and overweight prevention?

    Food Policy

    (2013)
  • J. Strauss et al.

    Chapter 34 Human resources: empirical modeling of household and family decisions

  • D. Tafreschi

    The income body weight gradients in the developing economy of China

    Econ. Hum. Biol.

    (2015)
  • S. Thiele et al.

    Consumer demand for food diversity: evidence for Germany

    Food Policy

    (2003)
  • X. Tian et al.

    Using semiparametric models to study nutrition improvement and dietary change with different indices: the case of China

    Food Policy

    (2015)
  • M.G. Wagner et al.

    Nutrition education effective in increasing fruit and vegetable consumption among overweight and obese adults

    Appetite

    (2016)
  • X. Wang et al.

    The rise of migration and the fall of self employment in rural China's labor market

    China Econ. Rev.

    (2011)
  • B. Xie et al.

    Effects of ethnicity, family income, and education on dietary intake among adolescents

    Prev. Med.

    (2003)
  • A. Abdulai

    Socio-economic characteristics and obesity in underdeveloped economies: does income really matter?

    Appl. Econ.

    (2010)
  • D.W.K. Andrews et al.

    Inference With Weak Instruments (Working Paper No. 313)

    (2005)
  • J. Angrist et al.

    Instrumental Variables And the Search For Identification: from Supply and Demand To Natural Experiments (Working Paper No. 8456)

    (2001)
  • F. Asiseh et al.

    Family income and body mass index – what have we learned from China

    Health Econ. Rev.

    (2016)
  • C. Batis et al.

    Longitudinal analysis of dietary patterns in Chinese adults from 1991 to 2009

    Br. J. Nutr.

    (2014)
  • J. Binkley

    Low income and poor health choices: the example of smoking

    Am. J. Agric. Econ.

    (2010)
  • J. Blanden et al.

    Accounting for intergenerational income persistence: noncognitive skills, ability and education

    Econ. J.

    (2007)
  • L. Brandt et al.

    Large Shocks and Small Changes in the Marriage Market for Famine Born Cohorts in China

    (2008)
  • A.C. Cameron et al.

    Microeconometrics: Methods and Applications

    (2005)
  • J. Cawley

    The impact of obesity on wages

    J. Hum. Resour.

    (2004)
  • Y. Chen et al.

    Does raising family income cause better child health? Evidence from China

    Econ. Dev. Cult. Change

    (2017)
  • M. Clement et al.

    Does social class affect nutrition knowledge and food preferences among Chinese urban adults?

    Camb. J. China Stud.

    (2015)
  • Cited by (51)

    View all citing articles on Scopus
    View full text