Agricultural production structure,market conditions and farmers’nutritional intake in rural China

WAN Yue ,HU Hao ,Wuyang HU

1 College of Economics and Management,Nanjing Agricultural University,Nanjing 210095,P.R.China

2 Department of Agricultural,Environmental,and Development Economics,The Ohio State University,Columbus 43210,USA

Abstract Nutrition poverty alleviation is an effective measure to improve the nutritional status of economically disadvantaged individuals,fundamentally improving their health and reducing poverty. Based on the Entitlement Theory and using the China Health and Nutrition Survey (CHNS) data,this paper examines the relationship between farmers’ nutritional intake,production structure and regional market conditions. Results show that farmers with diversified production have better nutritional intake than those who specialize. Furthermore,the correlation between regional market conditions and nutritional intake varies between economically disadvantaged and non-economically disadvantaged households. Market conditions significantly influence the carbohydrate and fat intake of economically disadvantaged households and are positively associated with the dietary structure and nutritional intake of non-economically disadvantaged ones. Moreover,income is positively correlated with the nutritional intake of non-economically disadvantaged households but not with economically disadvantaged ones.

Keywords:Chinese farmer nutrition,economically disadvantaged households,diversity of agricultural production,regional market conditions

1.Introduction

China has made remarkable achievements in poverty alleviation for more than 30 years. Based on the current poverty standard in China,by the end of 2019,there were 5.51 million rural residents in poverty,11.09 million less (or a reduction of 66.8%) than the end of 2018. As of 2019,the incidence of poverty rate was 0.6%,1.1 percentage points smaller than that in 2018 (NBSC 2019). The number of Chinese rural residents in poverty has been substantially reduced. However,the manifestation of poverty is also changing. The problem of poverty caused by illness is increasingly prominent. According to data of the Chinese Poverty Alleviation Office in 2015,44.1% of China’s rural poor were “impoverished due to illness”,with 2.4 million suffering from major diseases,and 9.6 million suffering from long-term chronic diseases. By 2018,of the 4.66 million poverty-stricken households,1.9 million(40.7%) (or 5.15 million individuals) were impoverished due to illness.

China has achieved comprehensive poverty alleviation(Fan and Cho 2021). To maintain and enhance the achievement,it is imperative to effectively prevent the risk of returning to poverty due to illness and realize the steady income growth of rural residents (Qinet al.2021).Amartya Sen,a leading economist and Nobel laureate,believes that the poor are poor because of a lack of human capital and a low level of health,and their ability or opportunity to create income is deprived. The medical community points out that unhealthy eating habits,such as excessive intake of sugar and salt or insufficient intake of protein and various vitamin elements,are often an important factor leading to various acute and chronic diseases,including diabetes,hypertension,dysplasia,and septicaemia (WHO and FAO 2003). Poorer households face higher levels of malnutrition,suboptimal health,and mortality (Wagstaff 2002). The economic and welfare loss caused by malnutrition is one of the primary reasons for individuals to fall into and stay in poverty. Economically disadvantaged households in rural areas generally lack a complete understanding of nutritional health and disease prevention,which in long term may lead to disease and poor health. From the perspective of nutrition,the reasons for returning to poverty among China’s rural residents are closely related to their nutritional status.

A number of challenges exist in dietary consumption and nutrition status of economically disadvantaged households in China (Liuet al.2013;Cuiet al.2014).First,these individuals have low disposable income and cannot meet their dietary needs. The quantity and composition of individuals’ food consumption are determined by many factors including the quality of food itself (Huet al.2013;Wuet al.2017;Zhonget al.2018).Economically disadvantaged households have limited ability to increase their food consumption as they have already spent a large proportion of income on it (Liet al.2021),and this situation may not be improved in the short term. Secondly,the composition of nutritional intake of economically disadvantaged households is not optimal.In addition,poverty caused by malnutrition can transmit over generations (Baiet al.2021).

Even a 1% increase in nutritional intake for economically disadvantaged households can have significant implications for society. Insufficient income,unstable food supply and poor dietary nutrition will have a profound impact on the physical wellbeing and health of economically disadvantaged households (Popkinet al.2006;Tiffin and Arnoult 2010). This may further affect human resources and economic development,widen the gap between the rich and the poor,curb social equity,and eventually have a negative impact on the stability and harmony of the country. Past literature has shown that major improvements may require significant shifts in current policy or practices. As a result,it may be more realistic to accomplish smaller and incremental improvements as a starting point. Policymakers should pay attention to the dietary and nutritional status of the poor,implement effective measures,even small steps at times,to ensure the basic food consumption and nutritional health of residents,promote the improvement of the country’s overall health,and achieve social equity. In view of this,it is necessary to analyse the main factors correlated with the nutritional intake of farmers and empirically test the possible paths of nutrition improvement for heterogeneous farmers,so as to promote the implementation and development of nutrition poverty alleviation.

Although the existing literature has analysed factors affecting individuals’ nutritional intake,there is still room for improvement. On the one hand,there are few studies on the nutritional intake of rural residents. In addition,few studies analysed the nutritional intake of individuals in poverty from the perspective of targeted poverty alleviation. As a result,the impact of nutritional intake on human-health capital and the role of nutrition poverty alleviation in targeted poverty alleviation policy remain understudied. On the other hand,there are few studies on the influence of food access on residents’ nutritional intake and the external factors that improve nutrition.Therefore,this study attempts to fill these voids.

This paper used data from the China Health and Nutrition Survey (CHNS) to study the relationship between farmers’ production structure,external market conditions,and their nutritional intake. Healthy nutritional intake means consumption in moderation as excessive nutritional intake may also lead to health problems such as obesity.This study calculated the gap between farmers’ nutritional intake and the standard recommended amount and analysed whether the improvement of farmers’ production and market conditions would improve the optimality of farmers’ nutritional intake. Rural residents were divided into economically disadvantaged and non-economically disadvantaged according to the poverty line. This study also captured differences in farm production structures and external market conditions in the analysis of farmers’nutritional intake.

2.Literature review and theoretical analysis

Generally speaking,factors related to residents’ food consumption and nutrition can be grouped into economic factors (Drewnowski and Popkin 1997;Duet al.2004;Tian and Yu 2013,2015;Zhouet al.2015;Zhuet al.2021),family characteristics (Behrman and Doelalikar 1987;Wrigleyet al.2003;Jensen and Miller 2010),regional policy,social factors,and food manufacturing technologies (Popkin and Bisgrove 1988;Solomons and Rainer 1995;Hu 2006).

Over the past decade,there has been increasing interest in research on how the agricultural sector can promote human nutritional outcomes (Turneret al.2013).In view of the importance of agriculture for the livelihood of the population in economically disadvantaged countries,Singhet al.(1986) studied whether the diversity of species grown on the farm contributed to farmers’ dietary diversity.In the case of functioning market conditions,farmers’production and consumption decisions are completely separable. Production diversity can affect dietary diversity if income is improved,thereby increasing the demand for a diversified diet. However,market failure is a common issue in economically disadvantaged countries. The high transaction cost of market access means that subsistence production is common,and farmers’ diversification of food and nutritional intake is restricted (Taylor and Adelman 2003;Wanet al.2021).

A series of papers have conducted empirical analysis on the relationship between production diversity and dietary diversity at family and individual levels. A general conclusion is that the increase of production diversity has a positive impact on dietary diversity (Joneset al.2014;Dillonet al.2015;Kumaret al.2015;Sibhatuet al.2015a;Hirvonen and Hoddinott 2017;Koppmairet al.2017). A study conducted by Joneset al.(2014) in Malawi found that,although production diversity is positively correlated with dietary diversity,families that pay more attention to survival have lower dietary diversity even after controlling for income. However,a study using data from four countries (Sibhatuet al.2015a) suggested that when production diversity is already high,further increases may reduce the income gains from specialization,thereby reducing dietary diversity.

The literature also recognizes the importance of the market for dietary diversity. Markets help generate income from agricultural production (Wenget al.2021),thereby increasing the family’s demand for diet quality and providing a variety of foods to meet this demand.The improvement of market conditions helps promote food diversity. Based on the size of the correlation coefficient between market conditions and production diversity,Sibhatuet al.(2015a) concluded that market conditions are more important than production diversity to improve dietary diversity. On this basis,they believe that increasing the diversity of farm food should not be overemphasized,and more consideration should be given to improving market conditions. This has triggered a debate with Remanset al.(2015),who believes that in some cases,market conditions may be difficult to strengthen,and thus it is still important to encourage regional production diversity. Sibhatuet al.(2015b)suggested that market conditions are also important to solve the problem of malnutrition and poverty in remote areas. Based on the above debate,the present study analysed the nutritional intake of Chinese farmers considering both their production diversification and possible access to markets.

Another contribution this study intends to make to the literature is to provide richer features of the markets related to farmers’ nutritional intake. In the related literature,the market plays roles through market positioning with producers (e.g.,the proportion of food consumed from production (Joneset al.2014)),production diversity in communities (Kumaret al.2015),and market conditions (e.g.,distance from the market(Sibhatuet al.2015a;Hirvonen and Hoddinott 2017)).However,in economically disadvantaged countries,the size and type of market vary dramatically,from small temporary facilities to large wet markets. This may lead to significant heterogeneity in the types of food available and in the choice sets from which households can build consumption mix and dietary diversity. In high-income countries,the literature linking the diversity of food supply to nutrition and diets is abundant (Caspiet al.2012).However,considering the food quality in economically disadvantaged countries and regions,the literature usually lacks detailed consideration of local marketization or transportations to markets. Therefore,this study attempts to close the gap.

The Entitlement Theory developed by Sen (1981) is widely used by scholars to study the relationship between nutritional health and poverty. The theory suggests that the deprivation of food access is one of the root causes of food insecurity. Individuals’ access to food depends on their ability and opportunity to access resources,technology,markets,social networks,and food aid programs. This paper followed the Entitlement Theory and explained food consumption and nutrition of farmers in poverty. The theory specifically points out that there are two reasons affecting residents’ nutritional intake:one is resource endowment,and the other is market access.

Nord (2007) regards household income,purchasing power,and the ability to trade as determinants of individual food consumption. Only farmers who are actively engaged in food production can obtain food through self-supply.For most other individuals in rural or poor areas,buying on the market is the main way to obtain food. As a result,how food can reach consumers in poor regions is affected by market factors. Chen and Wang (2010) argued that markets have become the dominant means for farmers to secure food. These studies all point to the conclusion that the food access ability of residents in poor areas is a key factor in their nutritional intake.

Based on the analysis of residents’ food acquisition paths,there are significant differences in food consumption and nutritional intake between urban residents and rural residents. The nutritional intake of urban residents is mainly obtained through market purchases,while that of rural residents comes from two important channels:household production and market purchase,which are determined by family income,the production structure of farmers,and the level of regional marketization. In terms of farmers’ self-production,due to the inseparability of production and consumption,farmers’ diversification of planting may lead to diversification of their food consumption. Furthermore,market purchase is not only restricted by family income,and the regional market development can also be an important reason affecting their nutritional intake. The higher the level of regional market development,the more opportunities the farmers have to meet their food consumption needs,thereby improving their nutritional intake. Therefore,the difference in individual income is not the only influencing factor of farmers’ nutritional intake;the marketization of the region where the farmers are located is also an important factor affecting their food consumption and nutritional intake.

This leads to the hypotheses of this study. First,based on the path analysis of farmers’ production and self-supply,a more diverse production structure leads to a more diversified food consumption,and the nutritional intake increases with the diversification of food consumption. Second,with a given income level,the higher the level of regional market development,the greater the chance of meeting farmers’ food consumption needs,thereby improving their nutritional intake. The specific analytical framework is shown in Fig.1.

Fig.1 Mechanism of production structure,market conditions and farmers’ nutritional intake.

This paper used China Health and Nutrition Survey(CHNS) data to test the impact of production structure and market conditions on farmers’ nutritional intake. Based on the household-level data,this study examined the nutritional intake of individuals in China’s less developed regions,with the intention to contribute to human capital enhancement and poverty alleviation.

3.Data and methods

The CHNS is jointly conducted by the University of North Carolina and the Chinese Center for Disease Control and Prevention. Samples in CHNS cover all areas in China and are divided into eastern,central,and western regions. The survey began in 1989 and there have been eight rounds of surveys so far. The CHNS data contain detailed records about family and personal characteristics,personal diet data during three consecutive days,and household food consumption,which can be used to analyse the nutritional intake of rural residents in China.The survey of family and individual food consumption was conducted by a consecutive 3-day 24-hour dietary review method using family food weighing method. The food consumption of each interviewee was collected,based on which the average daily nutritional intake of each interviewee was calculated.

Since the CHNS data of food consumption and nutrition were only updated to 2011 (the 2015 data have not been published),the literatures on food consumption and nutrition based on CHNS database are only up to 2011. The survey adopts the multi-stage stratified cluster random sampling method,and the specific steps are as follows:First,the simple random sampling method was used to select a total of nine provinces scattered in eastern,central and western regions of China. Second,counties were stratified in each province according to residents’ income level (low,medium,and high). Four counties (one high-income,two middle-income,and one low-income) from each of the nine provinces were randomly selected as rural samples according to a weighted sample table (a total of 36 counties were selected). Then,in each county,three villages were randomly selected respectively representing high,medium,and low schooling (a total of 108 villages were selected). The capital of each province and a low-income city were selected as urban samples.Third,220 community samples were selected from the urban and rural samples by the simple random sampling method,and then 20 family households were randomly chosen from each community for household survey. As a result,for each year,the sample contains about 4 400 households and 19 000 individuals.

This paper used five rounds of data,including 2000,2004,2006,2009,and 2011. After matching households in all five rounds of the survey,we removed missing values and observations of quantities exceeding five standard deviations above the sample mean (Tian and Yu 2015). Finally,our panel data include 6 446 households in rural areas,including 1 434 economically disadvantaged households and 5 012 non-economically disadvantaged households.

3.1.The choice of poverty standard

This study divided the samples to economically disadvantaged and non-economically disadvantaged households according to China’s poverty line standard in 2011 (2 300 CNY). The poverty line is defined as the minimum required to maintain the basic survival of a person and is the most widely accepted indicator for poverty in China.

3.2.Variable selection

Dependent variable of this study is nutritional intake of farmers. The intake of main nutrients1Nutrient intake is calculated based on the consumption of various foods. The tool used is the Food Composition Table(FCT). However,due to geographical differences,the FCT varies across countries. The “Chinese Food Composition Table” was developed by the China Center for Disease Control and Center for Nutrition and Food Safety and known as the Dietary Reference Intakes (DRIs).is a direct measure of nutritional status. The United Nations Food and Agriculture Organization (FAO) uses the minimum daily energy requirement (MDER) per person to estimate the proportion of undernourishment or chronic hunger in a population. The United Nations World Food Programme(WFP) also uses energy indicators to measure food security. Therefore,this study chose the nutritional intake index to measure farmers’ nutrition.

Key explanatory variables include production structure and regional market conditions. Referring to the existing literature (Huang and Tian 2019),this paper selected variable “the number of varieties of agriculture,horticulture,livestock and fishery production” to reflect the production structure of farmers. Specialized production is defined as farmers’ participation in only one type of agricultural production;diversified production is defined as farmers’participation in at least two types of agricultural production.The participation is a dummy variable,1 equals yes and 0 otherwise. Referring to the relevant literature (Sibhatuet al.2015a) measuring the level of marketization,this paper selected three variables in CHNS database,namely Urbanization Index2The methods and variables used to calculate these indexes were unclear in the original questionnaire. As a result,readers are advised to note this ambiguity.,Market Comprehensive Score Index,and Transportation Comprehensive Index,to measure regional marketization. In the robustness test,Urbanization Index was replaced by Economic Composite Index.

As mentioned above,farmers’ nutritional intake may also be affected by other factors such as family characteristics and regional differences. Therefore,this paper selected a wide range of control variables:the total household size,whether the household has fixed assets such as refrigerator to store food,gender,age,education level,self-assessed health level of the household head,total family income,and labour intensity.Studies using self-assessed demographics or perceptions as explanatory or control variables are common in both consumer and manufacturer research (e.g.,Bakeret al.2004;Huet al.2008).

3.3.Model selection

To estimate the influence of the relationship between production structure and marketization on nutritional intake of rural residents,this study constructed a consumption function in the following forms:

In these models,Yirefers to the average daily intake of different nutrients for residentiincluding the four main nutrients:carbohydrates (g),fat (g),calories (kcal),and protein (g).produceitis an index of residents’ production structure;urbanizationitrepresents the market conditions.produceit×urbanizationitis an interaction term between residents’ production structure and market conditions.Zitis a series of control variables,mainly including household characteristics and regional characteristics. This paper further analysed the correlation between production structure and market conditions on the standard deviation of nutritional intake of farmers. ΔYitrefers to the difference between the nutritional intake of the individuals in our sample and the dietary standard of Chinese residents,that is,the nutritional intake of households in the sample minus the recommended dietary standard intake of Chinese residents. This article set the reference dietary intake levels according to gender,age,and physical labour intensity based on the “Reference Intake of Dietary Nutrients of Chinese Residents” promulgated by the Chinese Nutrition Society3Based on the Chinese DRIs,the reference daily intake of calories,carbohydrates,fats and protein for men aged 18-49 years engaged in light physical activity is 2 250 kcal,120 g,90 g and 65 g,respectively;for 50-65-year-old men,the recommended intake is 2 100 kcal,120 g,84 g and 65 g,respectively;for 18-49-year-old women,the recommended intake is 1 800 kcal,120 g,72 g and 55 g,respectively;and for 50-65-year-old women,the recommended intake is 1 750 kcal,120 g,70 g and 55 g,respectively.. The specific explanation of relevant variables is shown in Table 1.

4.Results and discussion

The regression results of the four main nutrients (calories,carbohydrate,fat,and protein) intake of economically disadvantaged households are shown in Table 2 (1)-(4).The regression results show that production structure and market conditions have significant connection with the intake of calories,carbohydrate,fat,and protein in different degrees. Production structure is positively and significantly (at the 5% significance level),correlated with calorie and carbohydrate intake of the economically disadvantaged households. In addition,higher Urbanization Index is related to lower carbohydrate intake and higher fat and protein intake;higher Market Comprehensive Index is related to lower carbohydrate intake;and higher Transportation Comprehensive Index is related to lower intake of all nutrients.

The regression results of the influence of production structure and market conditions on the difference between the actual and recommended intakes of the four main nutrients for economically disadvantaged households are shown in Table 2 (5)-(8). The results showed that production structure and market conditions have no significant effect on the nutritional gap in calorie intake of the economically disadvantaged households. However,the fat intake of the households engaged in diversified production is closer to the recommended level than that of the specialized producers;households adopting less diversified production are more likely to have protein intake approaching the recommend level. In addition,higher Urbanization Index is related to carbohydrate intake approaching the recommended level,and lower Urbanization Index is related to protein intake approaching the recommended level. A lower comprehensive index of regional market and transportation is related to carbohydrate intake converging to the recommended level but is associated with fat intake being further away from the recommended level.

To sum up,the regression results show that the diversity of production structure is related to improved nutritional intake of the economically disadvantaged households. At the same time,regional marketization has a significant correlation with fat and protein intakes. To these households,the diversity of production structure has a significant relationship with reducing the nutrition gap between actual and recommended levels of fat and protein intake,and regional marketization has a significant relationship with decreasing the nutrition gap between actual and recommended levels of carbohydrate and protein intake. The results indicate that when market conditions permit,economically disadvantaged households may give priority to increasing the intake of fat and protein to improve their nutritional structure.

According to Table 2,other variables may also be related to the economically disadvantaged households’nutritional intake. First,higher education is associated with lower carbohydrate intake but higher fat and protein intake. Secondly,male-headed households have better nutritional intake than female-headed households. In addition,whether a household has a refrigerator has a significant correlation to their fat intake. For economically disadvantaged households,household income has no significant impact on their nutritional intake. Due to the limited income,their food consumption and nutritional intake may come from household agricultural production,and thus income may not play a significant role. A more diverse production structure may better contribute to satisfying their nutritional needs.

Table 1 Variable descriptive statistics

This study next analysed the impact of production structure and market conditions on the nutritional intake of non-economically disadvantaged households. The regression results on the four main nutritional intakes(calories,carbohydrate,fat,and protein) are shown in Table 3 columns (9)-(16). Production structure and market conditions have a significant (at the 5% level)connection with all the four types of nutritional intake. The regression shows that production diversification and the Market Comprehensive Index have significantly positive effects on all the four main nutritional intakes. Moreover,a higher Urbanization Index is related to a lower intake of carbohydrate but a higher intake of protein. At the same time,compared with production structure,market conditions can significantly affect whether the nutritional intake of non-economically disadvantaged households approaches the standard level.

According to Table 3,the effect of gender on the nutritional intake of non-economically disadvantaged households is significant. Male-headed households have better nutritional intake than female-headed households.Whether households have refrigerators has a significant correlation with their nutritional intake. In addition,household income has a significantly positive connection with the intake of calorie,fat and protein but a negative connection with the carbohydrate intake.

Comparative analysis of Tables 2 and 3 shows that the relationship between production structure,market conditions and nutritional intake is not consistent between economically disadvantaged and noneconomically disadvantaged households. Variable significance levels also differ between the two household groups. Diverse production structure is only significantly correlated with carbohydrate intake of the economically disadvantaged households,but it is significantly associated with all four main nutritional intakes of noneconomically disadvantaged households. Better market conditions are related to improved dietary structure of economically disadvantaged households with reduced carbohydrate intake and enhanced fat and protein intake;better market conditions are also associated with acarbohydrate intake more aligned with the recommended nutritional level.

Table 2 Regression results of the relationship between nutritional intake and production structure as well as marketization for economically disadvantaged households

For non-economically disadvantaged households,marketization is significantly related to the intake of the four main nutrients. With the improvement of market conditions,these households’ intake of calories and carbohydrate gradually decrease,while their intake of fat and protein gradually increase. However,for economically disadvantaged households,the improvement of market conditions may be associated with an increased gap between their actual protein intake and the recommended level,possibly resulting in excessive protein intake.

4.1.Endogeneity

The improvement of nutritional intake may promote labour output and labour income,causing endogeneity. The variable “vehicles” is used as the instrumental variable for farmers’ income. This variable is measured by the number of vehicles owned by rural households,including tricycles,bicycles,motorcycles,and cars. Vehicle number should be strongly correlated with income but not directly affect farmers’ nutritional intake. The estimated results are shown in Table 4.

The 2SLS DWH test rejects the hypothesisof exogeneity of the income variables at the 1%significance level (Table 4). This indicates that income is an endogenous variable. TheF-statistic of the weak instrumental variable test has an average value of 3.011,rejecting the null hypothesis of weak instrumental at the 1% significance level for all models.However,as Table 4 suggests,the regression results for the instrumental variable are qualitatively consistent with that in Table 2. Since this study does not intend to make causal inferences,Table 4 is not separately interpreted.

Table 3 Regression results of the relationship between nutritional intake and production structure as well as marketization for non-economically disadvantaged households

4.2.Robustness test

To test the robustness of the results,this study substituted several variables with alternative specifications.Considering that regional market conditions may influence the level of farmers’ nutritional intake by affecting their income,this paper also used the Economic Comprehensive Index and Community Comprehensive Score instead of the Urbanization Index,Market Comprehensive Score Index,and Traffic Comprehensive Index to represent market conditions. The estimated results are shown in Table 5.

Table 4 Regression results of the relationship between nutritional intake and production structure as well as marketization for economically disadvantaged households using 2SLS

Table 5 Robustness test of the relationship between nutritional intake and production structure as well as marketization for economically disadvantaged households

For all models,the regression results are robust. For instance,after replacing the Urbanization Index,the Economic Comprehensive Index is positively significant at the level of 1%,which indicates that higher regional economic conditions are associated with higher nutritional intake of economically disadvantaged households.

5.Conclusions and implications

Based on the data of 1 434 economically disadvantaged households and 5 012 non-economically disadvantaged households in rural China,this paper examined the relationship between production structure and market conditions to household nutritional intake from the perspective of the Entitlement Theory. It is found that production diversity can be positively related to enhanced nutritional intake. In addition,the correlation between regional market conditions and nutritional intake varies between economically disadvantaged and non-economically disadvantaged households. For economically disadvantaged households,urbanization has a significantly negative correlation with carbohydrate intake and a significant correlation with fat and protein intake. At the same time,urbanization can help economically disadvantaged households reach the recommended level of carbohydrate intake. Household income has no significant correlation with the nutritional intake of economically disadvantaged households.For non-economically disadvantaged households,Urbanization Index,Market Comprehensive Score Index,and Traffic Comprehensive Score Index all correlate significantly with better nutritional intake.

This study provides empirical evidence and support for decision-making to improve the human capital of rural poor and promote nutrition poverty alleviation in China. First,policymakers can encourage economically disadvantaged households to diversify their production.Due to the small and fragmented cultivated land and the loss of adult labour in rural areas,many rural residents,especially economically disadvantaged households,use private land for production and selfsufficiency. Therefore,the government needs to improve the development and utilization of land managed by economically disadvantaged households,encourage them to strategically engage in planting and breeding,and develop a variety of self-sufficient agricultural products. With inseparable agricultural production and farmers’ consumption,production diversity will promote the diversity of food consumption,improve diet structure,and increase farmers’ nutritional intake. This leads to improvement in farmers’ human-health capital and poverty alleviation effectiveness.

Second,infrastructure construction should be improved. In this paper,market and traffic indexes reflect regional infrastructure development. According to descriptive statistics,these indexes of our studied area are still far below those of urban areas. The lack of infrastructure restricts farmers’ access to markettransactions,seriously hindering their food consumption and nutritional intake diversity.

Third,nutrition knowledge communication and nutritional intervention need to be strengthened. Our results indicate that rural China faces over-consumption issues besides some deficiencies in essential nutritional intake. Specifically,the overall intake of calories and fat is insufficient,but carbohydrate and protein intakes exceed the recommended amount in some regions. While maintaining all other efforts combating malnutrition,relevant government agencies should also consider strengthening education on nutritional knowledge and promoting the recommended dietary guidelines in rural areas,especially among economically disadvantaged households.

Acknowledgements

The first and second authors make the following statement:This work was supported by the National Social Science Fund of China (19AGL020). This research uses data from China Health and Nutrition Survey (CHNS).We are grateful to research grant funding from the National Institute for Health (NIH),the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) for R01 HD30880,National Institute on Aging (NIA) for R01 AG065357,National Institute of Diabetes and Digestive and Kidney Diseases(NIDDK) for R01DK104371 and R01HL108427,the NIH Fogarty grant D43 TW009077 since 1989,and the China-Japan Friendship Hospital,Ministry of Health for support for CHNS 2009,Chinese National Human Genome Center at Shanghai since 2009,and Beijing Municipal Center for Disease Prevention and Control since 2011. We thank the National Institute for Nutrition and Health,China Center for Disease Control and Prevention,Beijing Municipal Center for Disease Control and Prevention,and the Chinese National Human Genome Center at Shanghai.

Declaration of competing interest

The authors declare that they have no conflict of interest.