The Health Cost of Rural Banquet Culture: The Mediating Role of Labor Time and Health Decision-Making—Evidence from Jiangsu, China

Abstract

Against the backdrop of health inequality among rural residents, focusing on the impact of cultural institutions during the transitional period on villagers’ health contributes to ensuring healthy lives for rural populations (SDG-3). Drawing on institutional theory, this study explores how rural banquet culture shapes residents’ health outcomes and the mechanisms through which this influence operates. Utilizing data from the 2021 China Land Economy Survey (CLES), we employ both mediation effect models and machine learning techniques. The findings indicate that a stronger presence of banquet culture within villages is significantly associated with poorer health outcomes among rural residents. Further analysis reveals that banquet culture is also correlated with increased adult and infant mortality rates and reduced life expectancy. Mechanism analysis shows that time crowding caused by social obligations and suboptimal health decision-making serve as important mediating pathways through which banquet culture influences health. Moreover, heterogeneity analysis suggests that higher levels of village autonomy and greater provision of public health goods can mitigate the negative health impacts of banquet culture. By uncovering the micro-level behavioral mechanisms through which cultural norms influence individual health, this research advances the understanding of informal institutions. It innovatively links ritualized social practices to measurable health outcomes, addresses a critical gap in public health research on rural populations, and provides policy-relevant insights for the development of targeted rural health interventions.

1. Introduction

Promoting sustainable health is a central objective of the United Nations Sustainable Development Goal 3 (SDG-3). Achieving this goal requires not only strengthening formal health systems, such as service provision, protection mechanisms, and resource allocation but also recognizing the often-overlooked influence of informal cultural practices on health outcomes [1,2,3,4]. In rural settings, traditional rituals and social customs continue to shape daily behaviors and health-related decisions in profound ways. However, the implications of these informal practices for long-term health sustainability remain insufficiently explored.

In China, one of the defining features of banquet culture is the hosting of feasts and the exchange of “gifts” (monetary contributions) during life-cycle events such as weddings, funerals, the birth of a child, and birthdays. The scope of these events often extends to the entire village, making it a typical example of how informal cultural norms can influence the health of rural residents [5]. While such gatherings serve to reinforce social cohesion and cultural identity, they can also impose substantial economic burdens and generate social pressure to overconsume food and alcohol [6,7,8,9]. Consequently, banquet practices may contribute to a range of adverse health outcomes, potentially threatening the long-term sustainability of rural public health.

Despite growing recognition of global health disparities between urban and rural populations [10,11,12,13,14,15,16,17], existing studies have predominantly focused on ecological [6,7,8], economic [18,19], and structural social determinants [20] while often overlooking the potential influence of deeply embedded cultural practices on the sustainability of rural health. In fact, culture can enhance individuals’ ability to perceive and understand health, while good health can facilitate human capital accumulation and increased economic productivity [21], thereby helping in the alleviation of poverty (SDG-1). As such, uncovering the mechanisms through which cultural norms interact with health outcomes has become increasingly essential for promoting the long-term sustainability of rural communities.

This study draws on data from the 2021 China Land Economy Survey (CLES) to examine the effects of banquet culture on the health of rural residents and the mechanisms through which these effects occur. Specifically, we explore how normative, coercive, and mimetic pressures related to banquet participation influence labor income and health-related decision-making, thereby shaping individual health outcomes. By integrating informal institutional theory with empirical analysis of health indicators, this research addresses a critical gap at the intersection of cultural sociology, public health, and sustainable development. The study contributes to the literature by deepening our understanding of how cultural practices affect health behaviors in rural contexts and offers evidence-based insights to inform rural health policy and intervention design.

The paper is organized into seven sections. Section 2 reviews the relevant literature, highlighting existing findings and pointing out unresolved issues. Section 3 develops the conceptual framework and introduces hypotheses concerning the health implications of banquet culture. Section 4 details the dataset and analytical strategies employed. Section 5 presents the core empirical results, followed by supplementary analyses in Section 6. The final section concludes with a summary of key insights and broader implications.

2. Literature Review

Anthropological research emphasizes that vernacular culture and rituals—such as weddings, funerals, and birthday banquets—constitute a vital part of cultural life in rural villages. These ceremonial gatherings play a critical role in reinforcing social cohesion, expressing collective identity, and facilitating reciprocal exchanges [22,23]. Participation in and organization of such events are typically governed by unwritten social norms. These rituals are often highly normative in nature, shaping individual behavior through shared expectations and perceived social obligations [24]. In developing countries, especially in rural areas with strong kinship networks, ceremonial banquets typically involve competitive consumption, symbolic social status, and the fulfillment of moral obligations [5,25]. These practices can impose significant economic and social burdens, contributing to tensions between traditional customs and modern lifestyles. Despite the evident influence of such cultural institutions on behavior and consumption patterns, limited research has examined their potential implications for public health.

Health outcomes among rural populations have primarily been examined through the lens of structural, behavioral, and environmental determinants. A substantial body of literature identifies factors such as access to healthcare services, educational attainment, income levels, working conditions, and lifestyle behaviors as critical contributors to rural health disparities [24,26]. Specifically, rural residents are often at greater health risk due to limited access to medical resources, lower levels of education, and inadequate health-related knowledge [4]. In addition, increasing attention has been given to social determinants of health—such as social capital, community engagement, and informal support systems—in explaining health disparities among rural populations [27,28]. Existing evidence suggests that social participation can promote positive health outcomes; however, in certain contexts, excessive social obligations may become burdensome, resulting in psychological stress or other adverse health effects [29,30]. It is also important to recognize that health-related behaviors are not purely individual choices but are shaped by cultural norms and institutional environments. For example, studies have shown that collective eating practices may encourage overeating, excessive alcohol consumption, and sedentary lifestyles [7,31].

While existing scholarship has extensively explored folk culture and the social determinants of health, limited attention has been devoted to understanding how cultural norms shape health-related behaviors and outcomes. Few empirical studies have examined the ways in which ritualized cultural practices—like hosting and attending banquets—impact the health of rural populations and the pathways involved.

Most current studies either treat culture as a contextual background variable or focus on macro-level institutional factors without offering a detailed examination of the micro-level behavioral pathways through which culture influences health.

3. Institutional Background and Theoretical Hypothesis

3.1. Banquet Culture and Its Governance in China

Banquets, as ritualized events involving the communal sharing of food and drink, represent cultural phenomena deeply embedded within social structures. Banquet culture is prevalent across major cultural systems worldwide and plays significant social, economic, and political roles in people’s lives [32]. For instance, in France, banquets reflect the fusion of gastronomy and social interaction [33], while in ancient Greece and Rome, they served as central arenas for social, political, and religious engagement [34]. Throughout the world, banquets with life-cycle rituals, such as wedding banquets in the West and traditional wedding banquets in India [35], carry multiple functions, such as social integration, identity, and economic exchange.

China’s banquet culture is an informal institutional, cultural practice rooted in social structures and traditional systems and constitutes a part of its long-standing historical heritage. Specifically, it is embodied in social events such as weddings, funerals, and festive celebrations, where banquets serve as a means to maintain interpersonal relationships, reinforce social identity, facilitate reciprocal obligations, and express emotional bonds [36]. Consequently, this cultural practice is characterized by a high degree of ritualization and normative social regulation. Distinct from general communal dining, banquets in China possess several defining features. First, as ritualized events, they typically follow standardized procedures and sequences. Second, banquets are embedded with ceremonial and moral significance [37]; guests are usually expected to offer monetary gifts, a practice commonly referred to as renqing (reciprocal obligation) in Chinese culture. Third, alcohol is a central element of banquets, which are typically characterized by the combination of elaborate dishes and baijiu (Chinese white liquor). As a result, banquets are colloquially referred to as “eating spirits” or “eating wine” (shijiu) [38]. This study specifically focuses on non-festival ritual banquets centered around major life events, such as weddings, funerals, and housewarmings, all of which involve the exchange of gifts and the communal consumption of food and alcohol.

Villages in China are communities traditionally organized around kinship ties, and the hosting of banquets in rural areas often involves mutual assistance from fellow villagers [38]. According to long-standing customs, life-cycle events such as weddings and funerals in rural communities involve complex and labor-intensive procedures that a single household is typically unable to manage alone. On such occasions, close relatives and neighbors commonly offer support to ensure the banquet is successfully carried out [39]. Over time, a system of mutual assistance for banquets has been established, and this model continues to be reflected in rural banquets today. However, under the pressures of social transformation, they have become instruments of enrichment or wealth display [40]. For the sake of face, villagers have continuously expanded the scale and cost of banquets, often hosting events beyond their financial capacity. This is typically reflected in the increasing number of guests invited, as well as the growing variety and quantity of dishes served [41]. For example, in some rural banquets in northern Henan Province, there are as many as 44 dishes per banquet; in Hubei Province, weddings and funerals are accompanied by banquets that last for at least three days. These extravagant and comparative rural banquets exacerbate the living burdens of villagers and hinder the modernization of villages. Additionally, this culture fosters unhealthy eating habits that may adversely affect health.

Since 2016, Chinese government documents have repeatedly emphasized the necessity of rectifying extravagance and wastefulness associated with rural banquets while continuing to promote the transformation of traditional village habits and customs. For instance, the No. 1 Central Document of China for 2024 highlights the importance of leveraging villagers’ self-governance and enhancing the incentive and restraint functions of village regulations to address the issue of excessive banquets. It is evident that villagers’ self-governance represents a pivotal measure in facilitating this rectification. Consequently, numerous villages have established wedding and funeral councils to regulate and standardize the frequency of banquets, the number of attendees, the value of gifts, and the expenditure on banquets. These efforts aim to reshape the widely prevalent banquet culture in rural China, mitigate its negative impacts, and promote an affordable and sustainable path of development.

3.2. Theoretical Hypothesis

Institutionalism theory suggests that institutions govern people’s activities and that individual behavioral choices are constrained and influenced by the institutional factors (both formal and informal) within which they are embedded [42]. Thus, informal institutions developed in traditional cultures exert a significant impact on individual choices, regardless of whether these impacts are in the long or short term [43]. As a typical form of informal institution, banquet culture in China is deeply rooted in the traditional system of rituals and the ethics of interpersonal obligations. It is a cultural practice that has gradually evolved through sustained interpersonal exchanges and kinship interactions. Banquet culture not only fulfills functions such as symbolic expression, social integration, and identity formation but also constitutes a routinized behavioral mechanism. It obliges individuals to adhere to specific norms and participation logic when encountering major life events such as weddings, funerals, or residential relocation. Particularly in rural China, banquet culture manifests as a deeply embedded informal institutional system within the social structure and emotional networks, exerting a profound influence on individual behavior and interpersonal interactions.

Neo-institutionalism frequently employs institutional isomorphism to explain such behavior, whereby individuals or organizations seeking legitimacy must adhere to specific institutional rules [44,45]. DiMaggio and Powell propose that this behavior arises through three primary mechanisms of pressure: coercive, mimetic, and normative [46].

The banquet behavior of villagers is influenced by these three pressure mechanisms. Specifically, coercive pressure arises from human interaction and face-saving culture within the social network, where the core rule is to monitor and constrain individual behavior through moral evaluations from others. As a key arena for the exchange of social favors, banquets serve important functions in maintaining social relationships and fulfilling symbolic obligations, and missing a banquet may be perceived as “arrogant” and “out of touch”. To maintain their reputation and social connections, villagers are compelled to attend banquets. Imitative stress stems from individuals’ perception, imitation, and adherence to others’ behaviors at banquets. At banquets, individuals observe and mimic how others use cigarettes and alcohol in their interactions [36] and tend to organize banquets by following the standards, scale, and procedures established by others. Normative pressure originates from the common norms of banquet culture that regulate and guide individual behavior. The mutual assistance system embedded within banquet culture encourages active participation in banquet activities, and Chinese villages are distinctly characterized by prominent kinship ties [47,48]. Despite working outside the village, Chinese individuals, influenced by a strong sense of “family”, return to participate in banquets.

Within the rural social context, interpersonal obligations operate in tandem with public village norms, forming an intertwined structure of normative, coercive, and mimetic institutional pressures. These combined forces shape villagers’ behavioral choices and ultimately influence their health outcomes. On the one hand, rural banquets present food safety and hygiene issues, including poor hygiene in food processing areas, cross-contamination between raw and cooked food, and the lack of health certificates among cooks [49]. These factors elevate the risk of contracting infectious diseases [50]. On the other hand, the host typically prepares dishes, predominantly meat-based, and beverages of a superior quality to the usual diet in order to entertain the guests [9]. Individual food choices and eating habits have been shown to be significantly influenced by peer dynamics [51,52]. At these banquets, villagers are influenced by their dining companions to increase their food intake [53], consume excessive amounts of unhealthy foods and meats, and alter their dietary structure, ultimately augmenting their health risks [54]. Furthermore, alcohol consumption constitutes a primary form of interaction between hosts and guests, often leading to excessive intake [55]. Medical research has shown that alcohol adversely affects health and contributes to the development of chronic diseases [56,57,58]. However, from the perspective of medical anthropology, illness is not merely a random or isolated event but rather a phenomenon that is socially and culturally constructed [59]. Based on the aforementioned analysis and findings, the following hypothesis is proposed:

Hypothesis H1:

The stronger the banquet culture, the lower the health status of villagers.

While banquet participation directly influences rural residents’ health, it may also affect their health through two important mechanisms: labor time crowding out and health decision-making. For rural residents in China, labor hours are typically divided between agricultural labor hours and non-farm working hours (NFWH). During peak agricultural seasons, villagers engage in farming activities at home, whereas during non-farming periods, they seek employment outside their villages. According to Social Identity Theory, individuals who are strongly identified with a group are more inclined to adopt behavioral norms that align with the group’s standards [60]. Under the influence of banquet culture, villagers adhere to mutual assistance whose boundaries extend beyond biologically defined kinship—namely, ties based on procreation and descent. As Carsten argues [61], kinship is not solely a biological given but is shaped through everyday practices and culturally mediated understandings of “relatedness”. The practice of returning to the village to attend banquets is influenced by both this mutual aid system and local conceptions of kinship, thereby reinforcing connections between individuals and the collective village community. However, traveling between workplaces and banquet sites may result in the crowding out of non-farm labor time, thereby reducing their income. Given that low or reduced income adversely affects health status [62,63], the following hypothesis is proposed:

Hypothesis H2:

The banquet culture impacts the health of rural residents through the crowding out of their labor time.

Embeddedness theory posits that individual economic behavior is influenced by social networks, which are invariably embedded within non-economic contexts such as culture and customs [64]. In China, where a traditional relational society prevails, gift-giving serves as a pivotal means of maintaining and fostering social networks, termed gift-giving expenditure. According to data sourced from the China Family Panel Studies (CFPS), the average proportion of gift-giving expenditure to household expenditure among rural households was 16.11% in 2014, 18.26% in 2016, and 18.66% in 2018, respectively. Elevated levels of gift-giving expenditure diminish current disposable income, thereby compressing short-term normal consumption when such expenditure constitutes an excessive proportion [65]. In essence, heightened gift-giving expenditure curtails household disposable income, prompting individuals to prioritize short-term consumption for basic needs over long-term health investments [66]. These investments encompass durable health products such as healthcare services [67], air purifiers [68], and water purifiers [69]. Given the preceding analysis, we propose the following hypothesis:

Hypothesis H3:

Banquet culture may impact villagers’ health decision-making, particularly in relation to long-term health investments, ultimately resulting in a decline in their overall health status.

While banquet culture may pose certain health-related risks, it is also essential to recognize its role in traditional rural society as a venue for social interaction, reciprocity, and emotional expression. Banquets function as informal mechanisms for mutual assistance and relationship integration. Therefore, any critique of banquet culture must acknowledge its dual nature: it serves both as a vehicle for social cohesion and as a potential source of health-related externalities. The framework depicted in Figure 1 is constructed on the basis of the above theoretical considerations and assumptions.

4. Data and Methodology

4.1. Data Collection and Sample Representativeness

The questionnaire for The China Land Economy Survey (CLES) was organized into two components: a village questionnaire and a rural household questionnaire. The survey employed the Probability Proportionate to Size (PPS) method, selecting 26 districts and counties across the 13 prefecture-level cities in Jiangsu Province. In each district, two townships were chosen, followed by one administrative village per township, from which 50 farm households were randomly selected. This multi-stage sampling ultimately covered 52 administrative villages and 2628 rural households, encompassing approximately 8000 individuals. As of the end of 2021, Jiangsu’s rural population stood at 22.18 million, meaning the survey covered roughly 0.036% of the rural population. This sampling ratio aligns with standard practices used in large-scale national social surveys—for example, the China General Social Survey (CGSS) employs a sampling ratio of 0.011%, the China Family Panel Studies (CFPS) uses 0.004%, and the United States General Social Survey (GSS) uses 0.0013%.

Regarding the representativeness of the sample, first, in terms of the typicality of banquet culture, Jiangsu—particularly its northern and central regions—has preserved rich traditional ritual customs. Jiangsu Province is renowned nationwide for the dishes, grandeur, and atmosphere of its banquets [70]. Furthermore, in terms of food-related aspects, according to data from the China Foodborne Disease Outbreak Surveillance System spanning from 2010 to 2020, Jiangsu Province ranked second in China in terms of the number of foodborne disease outbreaks associated with rural banquets [71]. More importantly, the structure and function of banquet culture in Jiangsu closely resemble those found in most rural areas of China, characterized by widely shared kinship networks and normative social pressures. At the same time, it is also comparable to rural banquets in other developing countries, such as the large-scale communal dining typical of Indian weddings [72] and the display of wedding scale as a marker of social status [73]. Second, in terms of regional heterogeneity, Jiangsu exhibits a marked development divide between its southern and northern regions: the south is more economically advanced and characterized by stronger governance capacity, whereas the north more closely reflects the developmental conditions of rural areas in China’s central and western regions [74]. Jiangsu is geographically divided into three regions: Northern, Southern, and Central, and the economic conditions of these areas closely resemble those of the eastern, central, and western regions of China. This regional diversity offers a natural “control laboratory” for identifying the differential impacts of banquet culture across varying socio-economic contexts. Accordingly, selecting Jiangsu as the study area not only enhances the representativeness of the findings within the Chinese context but also provides a solid foundation for future cross-national comparative research.

4.2. Defining Variables

4.2.1. Dependent Variable

In accordance with related research [75], this study employed the self-reported health status of rural residents as the dependent variable. To better capture individual differences in perceived health status, a five-point Likert-type scale was employed during the variable assignment process [76]. Respondents rated their health status on a five-point scale: 1 for “incapacitated” (indicating very poor health), 2 for “poor”, 3 for “fair”, 4 for “good”, and 5 for “very good” (optimal health). This scale captures a continuum from complete loss of function to optimal well-being, with higher values indicating better self-perceived health. This assignment logic not only aligns with the continuous characteristics of subjective health assessment but also facilitates the identification of nonlinear relationships between variables in subsequent statistical analyses, demonstrating strong explanatory power and operational consistency.

4.2.2. Independent Variables

The complexity of traditional wedding and funeral customs serves as a key indicator of the institutionalization of banquet culture [5], reflecting the ceremonial pressure and normative constraints individuals experience during major life rituals [77]. Therefore, this study uses the complexity level of traditional village wedding and funeral practices as a proxy variable for banquet culture. The questionnaire asked respondents whether traditional wedding and funeral practices in their village were complicated. The responses were divided into four levels, “complex”, “moderate”, “not complex”, and “indifferent”, reflecting respondents’ subjective evaluations of the complexity of wedding and funeral customs in their village. A higher score indicates a greater perceived simplification of these traditional practices. For ease of understanding, the index was reversed and then aggregated to the village level to obtain the average complexity degree of traditional wedding and funeral practices in that village. A larger value indicates a stronger banquet culture in the village, characterized by greater ritual formality, procedural complexity, and social pressure for participation in customary practices.

4.2.3. Mechanism Variables

This research identifies two key mechanisms: labor time crowding out and health decision-making.

Specifically, labor time crowding out was assessed primarily in terms of non-farm working hours and income. Labor time was measured by the number of days respondents worked in non-farm jobs during the year, whereas labor income was operationalized using household per capita non-farm wage income. To address heteroskedasticity, the income variable was transformed using the natural logarithm.

Health decision-making was examined in terms of Time Preference (TP) and Health Investment (HI). Given that health is a long-term investment [66,78], the time preferences of respondents were utilized to assess their inclinations in health-related decision-making. Time preference is a core concept in behavioral economics, referring to an individual’s tendency to trade off present versus future benefits [79,80,81]. Under conditions of poverty and social pressure, individuals are more likely to exhibit a “present-oriented” mindset, which affects their choices related to long-term outcomes such as education and health [82]. The questionnaire solicited responses from participants regarding their time preferences, with the results being categorized into three groups: those prioritizing only current benefits, those valuing both current and future benefits, and those emphasizing only future benefits. A higher value indicates a stronger preference for long-term benefits. Additionally, health investments were quantified based on the number of water purifiers and air purifiers within households. The quality of drinking water and air significantly influences individual health outcomes [83,84]. Studies have shown that among environmental risk factors, water and air pollution are among the leading contributors to morbidity and mortality in China [85]. Approximately 490 million rural residents in China still rely on solid fuels for cooking [86], resulting in severe indoor air pollution and adverse health effects [86,87,88,89,90]. Numerous studies have provided evidence for the positive health outcomes associated with water programs [91,92,93,94], with higher water quality shown to enhance cognitive function and energy levels, thereby improving health status [87,95]. Household water purifiers and air purifiers, which are marketed as health-related consumer goods, play an important role in improving drinking water quality [96,97] and mitigating indoor air pollution [98]. Although such devices may not produce immediate health improvements, their presence can serve as an indicator of a household’s long-term investment in health.

4.2.4. Control Variables

Drawing upon previous studies [99,100,101], this study controlled for variables that have been demonstrated to be closely associated with health outcomes. The analysis encompassed three levels: individual, household, and village. The measurement of these variables is detailed in

Table A1. Key variable descriptions.

Variables	Description
Dependent Variable
Health	Self-identified Health Conditions: 1 = Incapacity to work; 2 = Poor; 3 = Medium; 4 = Good; 5 = Excellent.
Independent Variable
Banquet Culture	Are the traditional wedding and funeral customs in your village complicated? 1 = Complicated; 2 = Moderate; 3 = Not Complicated; 4 = Indifferent.
Mechanism Variables
Non-Farm Working Hours	Total volume of non-farm work within a year (days).
Household Income from Non-Farm Work	Total income from non-farm work of a year (CNY).
Time Preference	What are your time preferences? 1 = I only focus on the current income, regardless of the future; 2 = I value both current and future benefits; 3 = I only value future earnings, not the present.
Health Investment	Air purifier (piece); Water purifier (set).
Control Variables
Individual Level
Gender	Gender (1 = male; 0 = female).
Age	Age (In full years).
Years of Education	Educational level (Number of years in schooling).
Emotional Conditions	In the past week, I have been in low moods. 1 = Almost Not (less than one day); 2 = Sometimes (1–2 days); 3 = Frequent (3–4 days); 4 = Most of the time (5–7 days).
Social Relations	The number of people who you can borrow CNY 50,000 from when you are in trouble.
Recent Medical Visit Situation	Have you gone to the hospital for your last illness or injury? 1 = Yes; 0 = No.
Household Level
Family Size	How many people are there in your permanent population (Living for 6 months or more throughout a year)?
Family Income	Total household income (yuan); Natural logarithm processing.
Type of Housing	What is your current housing type? 1 = Reinforced concrete structure; 2 = Masonry-concrete structure; 3 = Brick, stone and wood houses; 4 = Other, please specify.
Living Space	What is the living space of your house (m2)?
Medical Insurance	Do all members of your family have medical insurance? 1 = Yes; 0 = No.
Drinking Water Conditions	How do you get your drinking water? 1 = Indoor tap water; 2 = Tap water in the yard; 3 = Well water in the yard; 4 = Mineral water; 5 = Other, please specify.
Living Conditions	Water Heaters (Number); Air conditioning (Number); Washing Machines (Number).
Village Level
Village Elevation	Altitude of the village committee location.
Population Density	Population density (person/acre).
Economic Level	Per capita net income of the whole village.
Drinking Water Safety	Is there a drinking water safety project in this village? 1 = Yes; 0 = No.
Sanitation Facilities	Number of trash cans in the village.
Medical Quality	Number of practicing (assistant) physicians.
Tool Variables
Number of Traditional Villages in the County	The number of traditional villages in the county to which the village belongs.
Others
Village Level	Annual number of civil disputes. Annual number of public security violations.

of Appendix A.

Table 1. Descriptive statistics of key variables.

Variable Type	Variable	N	Mean	p50	SD	Min	Max
Dependent variable	Health	2420	4.049	4.000	1.055	1.000	5.000
Independent variable	Banquet culture	2420	2.881	2.857	0.192	2.417	3.313
Mechanism variables	Non-farm working hours	2352	73.580	0.000	118.600	0.000	365.000
	Household income from non-farm work	2420	7.954	10.400	4.629	0.000	14.670
	Time preference	2340	1.701	2.000	0.685	1.000	3.000
	Health investments	2413	0.370	0.000	0.765	0.000	13.000
Individual level control variable	Genders	2417	0.727	1.000	0.446	0.000	1.000
	Age	2420	62.160	64.000	11.430	18.000	92.000
	Years of schooling	2420	7.184	8.000	3.971	0.000	19.000
	Emotional state	2419	1.210	1.000	0.521	1.000	4.000
	Social relation	2357	5.604	2.000	18.570	0.000	500.000
	Recent medical treatment	2404	0.775	1.000	0.417	0.000	1.000
Family level control variable	Household size	2420	3.044	2.000	1.601	0.000	11.000
	Log of household income	2420	7.537	8.896	3.707	0.000	14.930
	Housing type	2418	1.798	2.000	0.765	1.000	12.000
	Housing area	2392	205.200	200.000	118.800	15.000	1500.000
	Medical insurance	2387	0.941	1.000	0.235	0.000	1.000
	Drinking water conditions	2417	1.877	1.000	25.340	1.000	1234.000
	Living conditions	2415	1.114	1.000	0.770	0.000	22.000
		2416	2.342	2.000	1.824	0.000	30.000
		2413	1.184	1.000	0.821	0.000	30.000
Village level control variable	Village elevation	2038	11.420	10.000	11.050	0.000	50.000
	Population density	2420	0.524	0.497	0.406	0.000	2.918
	Level of village economy	2384	9.958	10.080	0.582	7.765	10.820
	Drinking water security	2343	0.696	1.000	0.460	0.000	1.000
	Sanitation facilities	2343	509.800	146.000	870.200	2.650	4636.000
	Medical level	2420	1.042	0.000	1.435	0.000	7.000

presents summary statistics for the main variables.

4.3. Identification Strategies

First, an ordered probit model with village fixed effects (Oprobit) was used to assess the impact of rural banquet culture on villagers’ health. The model was set up as follows:

$${\mathit{Health}}_i^{*}={\beta }_0+{\beta }_1{\mathit{Banqcul}}_i+\eta X_i+{\mathit{Village}}_i+{\epsilon }_i$$

(1)

$${\mathit{Health}}_i=\left\{\begin{array}{l}1,{\mathit{Health}}_i^{*}\le {\mu }_1\\ 2,{\mu }_1<{\mathit{Health}}_i^{*}\le {\mu }_2\\ 3,{\mu }_2<{\mathit{Health}}_i^{*}\le {\mu }_3\\ 4,{\mu }_3<{\mathit{Health}}_i^{*}\le {\mu }_4\\ 5,{\mu }_4<{\mathit{Health}}_i^{*}\end{array}\right\}$$

(2)

where $i$ means rural residents $i$; ${\mathit{Health}}_i$ indicates the health status of the rural residents; ${\mathit{Banqcul}}_i$ indicates the strength of the banquet culture; $X_i$ refers to a series of control variables; ${\mathit{Village}}_i$ indicates village fixed effects; ${\epsilon }_i$ represents the random error term; and ${\beta }_1$ is the coefficient to be estimated, indicating the effect of banquet culture on health.

Second, Baron and Kenny’s study was regarded as reference [102], and then the following models were constructed for mechanism testing:

$$M_i={\alpha }_0+{\alpha }_1{\mathit{Banqcul}}_i+\eta X_i+{\mathit{Village}}_i+{\epsilon }_i$$

(3)

$${\mathit{Health}}_i={\theta }_0+{\theta }_1{\mathit{Banqcul}}_i+{\theta }_2{M}_i+\eta X_i+{\mathit{Village}}_i+{\epsilon }_i$$

(4)

where $M_i$ is the mediating variable. The rest of the parameters are consistent with the above. Since the mediating variables are all continuous variables, the OLS model was used in the mechanism analysis. ${\alpha }_1$, ${\theta }_1$, and ${\theta }_2$ are the coefficients to be estimated from which the underlying mechanism can be inferred.

5. Empirical Analysis

5.1. Baseline Results

Individual, household, and village-level control variables were sequentially introduced into the model, with the estimation results presented in

Table 2. Baseline regression results.

Variables	Health	Health	Health	Health
	Model 1	Model 2	Model 3	Model 4
Banqcul	−5.1186 *** (0.0638)	−4.9387 *** (0.1593)	−5.6797 *** (0.2314)	−0.3556 *** (0.0534)
Individual Controls		Yes	Yes	Yes
Household Controls			Yes	Yes
Village Controls				Yes
Village FE	Yes	Yes	Yes	Yes
Log-likelihood	−3017.1907	−2771.8203	−2647.5977	−2259.4591
Pseudo R2	0.0385	0.0925	0.1053	0.1015
Obs	2420	2354	2277	1913

The symbol *** indicated 1% significance levels, respectively. Cluster-corrected standard errors (village level) are in parentheses.

. The results of models (1) to (4) revealed that the estimated coefficients for Banqcul were all significantly negative at the 1% significance level. This finding indicated that a stronger banquet culture in the village was associated with lower health status among rural residents. Consequently, it was demonstrated that banquet culture indeed exerted a negative impact on the health of rural residents, thereby preliminarily supporting Hypothesis H1.

5.2. Mechanism Analysis

In which channel does banquet culture impact health? This question was addressed through the testing of Hypotheses H2 and H3. Initially, Model 1 in

Table 3. Mechanism test results.

Variables	Health	NFWH	Health	INFW	Health	TP	Health	HI	Health
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8	Model 9
Banqcul	−0.4179 *** (0.0436)	−89.8524 *** (5.1054)	−0.3432 *** (0.0548)	−3.1743 *** (0.2450)	−0.3651 *** (0.0515)	−0.6100 *** (0.0334)	−0.3808 *** (0.0529)	−0.1191 *** (0.0310)	−0.3871 *** (0.0427)
NFWH			0.0008 *** (0.0002)
INFW					0.0166 *** (0.0067)
TP							0.0876 *** (0.312)
HI									0.0643 *** (0.0225)
All Controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Village FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Pseudo R2	0.2254	0.2586	0.2313	0.1925	0.2294	0.1146	0.2279	0.2056	0.2280
Obs	1913	1860	1860	1913	1913	1855	1855	1911	1911

The symbol *** indicated 1% significance levels, respectively. Cluster-corrected standard errors (village level) are in parentheses.

demonstrated that banquet culture also adversely affected the health of rural residents at the 1% significance level in the OLS-estimated model.

Model 2 revealed that banquet culture crowded out the non-farm working hours of villagers. Model 3 further showed that non-farm working hours had a positive influence on their health, yet this positive effect was diminished upon the introduction of banquet culture. This finding suggested that banquet culture weakened its beneficial impact on health by reducing non-farm working hours, thereby conforming to the labor time crowding-out mechanism. These results were in line with the predictions of Hypothesis H2.

The outcomes of Model 4 indicated that banquet culture had a detrimental effect on non-farm work income; specifically, it reduced the non-farm work income of rural residents. Model 5 subsequently demonstrated that banquet culture further influenced health by decreasing non-farm income, thus validating Hypothesis H2.

In Models 6 and 8, the coefficients for banquet culture with respect to time preference and health investment in health decision-making among rural residents were both significantly negative at the 1% level. This indicated that banquet culture weakened the willingness and behavior of rural residents to engage in long-term health investments. These findings were consistent with the predictions of Hypothesis H3.

Models 7 and 9 revealed that time preference and health investments had a significant negative impact on the health of rural residents, as evidenced by passing the 1% significance test. Consequently, it could be concluded that time preference and health investments mediated the relationship between banquet culture and the health status of rural residents. This suggested that banquet culture reduced the health level of villagers by influencing their health decision-making. Therefore, Hypothesis H3 was confirmed.

5.3. Robustness Testing

5.3.1. Addressing Endogeneity

Compared to traditional regression models, machine learning algorithms exhibit a superior ability to address endogeneity problems and, consequently, have gained increasing attention in the field of causal inference [103,104]. Among these advanced methods, the dual machine learning (DML) model stands out by integrating semiparametric estimation models with machine learning techniques. This integration offers significant advantages in mitigating the curse of dimensionality and reducing the risk of model misspecification [105]. DML has also been applied to studies of rural population health, where it has demonstrated robust estimation performance in addressing endogeneity issues [75]. In this context, inspired by the work of Chen Y and Ye Q [75], we constructed a partially linear dual machine learning model and a dual machine learning instrumental variable model. These models were designed to tackle the endogeneity problem in a rigorous and effective manner. The underlying assumptions of the DML model, parameter configurations, and strategies for mitigating overfitting have been thoroughly elaborated in the study by Chernozhukov [105]. Therefore, this paper does not reiterate those details but instead focuses on the implementation process of the model.

First, with reference to the study by Chernozhukov et al. [105], the following dual machine learning partial linear model was constructed:

$${\mathit{Health}}_i={\theta }_3{\mathit{Banqcul}}_i+g(X_i)+U_i,E(U_i|X_i,{\mathit{Banqcul}}_i)=0$$

(5)

$${\mathit{Banqcul}}_i=m\left(X_i\right)+V_i,E\left(V_i|X_i\right)=0$$

(6)

where $U_i$ is the error term in the first stage, which satisfies the zero-mean assumption. $V_i$ is the error term in the second stage. The meanings of the remaining parameters are consistent with the previous section. Since Equation (5) may converge slowly with small samples, it needs to be accelerated with Equation (6) to obtain the unbiased estimator. Combining Equations (5) and (6) and using a machine learning algorithm, the unbiased estimator ${\theta }_3$ can be obtained.

Since dual machine learning solves the curse of dimensionality, additional quadratic terms were added for some of the economic and social variables at the individual and village levels to the control variables, thus reducing the omitted variables and nonlinear effects. Regarding algorithm selection, the linsvm algorithm (Linear Support Vector Machine) was used in conjunction with the data characteristics of the dependent variable in this study. The linsvm algorithm is well suited for linearly categorizable data [106] and showed the best predictive performance in many scenarios [107]. For parameter tuning, the dataset was divided using a 1:4 split ratio, and the corresponding estimation output is reported in Model (1) of

Table 4. Estimation after dealing with endogeneity.

Variables	DML	DML-IV		DML-IV (Changing Algorithms)
	Semi-Parametric	First Stage	Second Stage	First Stage	Second Stage
	Model 1	Model 2	Model 3	Model 4	Model 5
Banqcul	−0.4093 *** (0.0109)		−6.0045 *** (1.4025)		−6.2962 ** (2.7583)
IV_Tradvill		0.0089 *** (0.0010)		0.0565 *** (0.0192)
All Controls	Yes	Yes	Yes	Yes	Yes
Village FE	Yes	Yes	Yes	Yes	Yes
Kleibergen–Paap Wald rk F statistic		72.6100
Anderson–Rubin Wald test		51.0600 ***		72.2100
Obs	1913	1913	1913	1913	1913

The symbols *** and ** indicated 1% and 5% significance levels, respectively. Cluster-corrected standard errors (village level) are in parentheses.

. The results showed that the estimated coefficient of Banqcul was −0.4093, which was significant at the 1% level. This indicated that after mitigating the endogeneity, the banquet culture still had a negative impact on the health of rural residents.

Second, referring to Chernozhukov et al.’s study [105], the following dual machine learning instrumental variable model was constructed:

$${\mathit{Health}}_i={\theta }_4{\mathit{Banqcul}}_i+g\left(X_{\mathit{it}}\right)+U_{\mathit{it}}$$

(7)

$$IV\text{\_}Tradvil{l}_i=m\left(X_i\right)+V_i$$

(8)

where $IV\text{\_}Tradvil{l}_i$ is the instrumental variable, and the remaining parameters are consistent with the above.

The instrumental variables employed in this study were quantified by the number of traditional villages at the county level, selected according to the following criteria: Firstly, banquet culture constitutes a significant component of folk culture and serves as an emblematic representation of traditional rural culture [108]. The Measures for the Protection of Traditional Villages in Jiangsu Province stipulate that one of the prerequisites for identifying a village as traditional is the preservation of intangible cultural heritage, including folklore activities and traditional skills, which exhibit a distinctive regional vernacular culture. Consequently, the county-level count of traditional villages serves as an apt indicator of the robustness of local rural traditional culture, which is intrinsically linked to banquet culture. Thus far, the instrumental variable fulfills the criterion of correlation. Secondly, the number of traditional villages at the county level is exogenous with respect to economic and social conditions at the village level. Furthermore, it bears no direct relation to the health status of individual villagers. Therefore, this instrumental variable also satisfies the exogeneity condition. Data pertaining to traditional villages were sourced from the List of Traditional Villages in Jiangsu Province, which is acknowledged and published by the Department of Housing and Urban–Rural Development of Jiangsu Province, among other authorities, and is noted for its high degree of accuracy and authority. To date, a total of 554 traditional villages, spanning seven batches, have been announced.

After matching the county-level data with the CLES data, the estimation was conducted, and the results are presented as Models 2 and 3 in Table 4. The results indicate that the Cragg–Donald Wald F statistic is 116.8500, significantly exceeding the conventional threshold of 10. This suggests that the concern of weak instruments is effectively mitigated, thereby confirming the validity of the instrumental variable. The Anderson–Rubin Wald test also rejected the null hypothesis of weak instrumental variables at the 1% significance level, further confirming the appropriateness of the instrumental variables. In the first stage of the analysis, IV_Tradvill showed a strong and statistically significant positive association with banquet culture (p < 0.01), indicating that the county-level count of traditional villages can effectively predict the intensity of banquet culture.

In the second stage, after employing instrumental variables to mitigate endogeneity concerns, the estimated coefficients for banquet culture decreased substantially, indicating that banquet culture exerts a detrimental effect on the health outcomes of villagers.

In summary, the estimation results of this study remained robust after accounting for endogeneity. To prevent the issue of algorithmic overfitting, we re-estimated the instrumental variable model using a neural network algorithm. The results from Models (4) and (5) in Table 4 indicate that banquet culture significantly reduces the health status of rural residents. Hypothesis H1 was thus reconfirmed.

5.3.2. Variable Tests Replacement

First, the measurement of the dependent variable was changed. Specifically, the dependent variable was re-assessed using the average health level of household members. Second, the measurement of the independent variable underwent a transformation; banquet culture was recoded as a dichotomous variable. Specifically, values exceeding the mean were coded as “1”, whereas values below or equal to the mean were coded as “0”. The results in

Table 5. Robustness test.

Variables	Healthfam	Health	Health	Health
	Model 1	Model 2	Model 3	Model 4
Banqcul	−0.4056 *** (0.4328)		−0.2905 *** (0.0758)	−0.4930 *** (0.0923)
Banqcul_binary		−0.1621 *** (0.0243)
Tobacco			Yes
Alcohol			Yes
Environmental Pollution			Yes
All Controls	Yes	Yes	Yes	Yes
Village FE	Yes	Yes	Yes	Yes
Pseudo R2	0.2293	0.1015	0.1037	0.1044
Obs	1913	1913	1892	1913

The symbol *** indicated 1% significance levels, respectively. Cluster-corrected standard errors (village level) were in parentheses.

indicated that banquet culture continued to exert a negative influence on the health of rural residents.

5.3.3. Exclusion of Other Factors

According to related studies, smoking [109], alcohol consumption [110], and environmental pollution [111] also exhibit potential health effects. These factors were further controlled, and the regression analysis was conducted again. The results are presented in Model (3) of Table 5. The findings indicated that, although the coefficients had decreased, the adverse effect on villagers’ health continued to be statistically significant at the 1% level. In addition, the survey collected village-level data on the annual number of civil disputes and public security violations, which serve as proxy indicators for the local social security environment. Previous research has indicated that alcohol consumption during public gatherings often leads to behavioral disorders, thereby increasing the risk of violence and injury. Areas with more frequent community gatherings tend to exhibit significantly higher crime rates [112]. Therefore, banquets may contribute to incidents of violence and harm, potentially exerting adverse effects on health. To account for such confounding factors, we included the local social security condition as the dependent variable in the regression model. The results are reported in

Table 6. Regression results of banquet culture and village-level public security.

Variables	Annual Civil Disputes	Public Security Violations (Annual)
	Model 1	Model 2
Banqcul	6.3960 (38.2419)	0.4017 (2.4222)
All Controls	Yes	Yes
Pseudo R2	0.0735	0.0581
Obs	1874	1823

, Models (1) and (2). The findings suggest that banquet culture is not significantly associated with variations in local social security conditions. Although this finding diverges from existing studies, it can be reasonably interpreted within the context of rural China. First, rural banquets are typically embedded within kinship networks and ritual norms, characterized by strong organizational structures and a high degree of predictability. The involvement of elders and the presence of moral constraints serve as forms of supervision and social stabilization. This aligns with the theories of Human Territorial Functioning [113] and the Broken Windows Theory [114], which suggest that crime tends to flourish in environments lacking guardianship and social oversight. Moreover, in the context of China’s collectivist culture, which emphasizes social harmony, the negative effects of banquet culture are less likely to manifest as overt violence. In sum, our findings confirm that banquet culture does not impact health through pathways related to crime or violence.

5.3.4. Measurement Models Replacement

To mitigate the potential influence of the model specification, the regression analysis was re-executed using the ordered logit model, and the outcomes are presented in Model (4) of Table 5. The results remained consistent with the benchmark regression, suggesting their robustness.

6. Extended Analysis

6.1. Heterogeneity Analysis

Furthermore, considering that banquet culture exhibits variations across different regions and age groups, subgroup regression analysis was employed to examine potential heterogeneity. The results are presented in Figure 2, Figure 3 and Figure 4.

6.1.1. Different Age Groups

Differences in banquet culture may exert influences on various age groups. Consequently, the WHO age classification standard was referred to for grouping: 18–44 years old constitutes the young group, 45–59 years old represents the middle-aged group, and those over 60 years old comprise the elderly group. The results depicted in Figure 2 indicate that banquet culture significantly impacted the health status of the youth group, whereas no significant relationship of influence was observed in the elderly group. This suggests that rural youth are the primary demographic affected by banquet culture. A plausible explanation could be that rural youth have a heightened need to cultivate social network relationships. They must actively utilize banquets to forge connections and accumulate social capital within their acquaintance society [115]. Consequently, they are more susceptible to the health risks associated with banquet culture.

6.1.2. Impact of Self-Governance

Wedding and funeral councils are self-governing organizations established by villagers to transform traditional customs within the village. In the village-level questionnaire, respondents were asked, “Does this village have weddings and funerals councils responsible for streamlining wedding and funeral affairs?” and “Do Party members and public officials in this village take the lead in altering customs and traditions and promoting new social norms when managing wedding and funeral affairs?” Based on the responses, the survey results were divided into two subsamples: those answering “Yes” and those answering “No”. Heterogeneity in self-governance was then tested by regressing the subsamples separately. The results presented in Figure 3 indicate that banquet culture had a significant negative impact on villagers’ health. However, in areas with stronger self-governance, the negative impact of banquet culture on health was mitigated. This may be because, through self-governance, the frequency of banquets and adverse health behaviors associated with them could be reduced. Furthermore, the findings suggest that self-governance can, to some extent, diminish the negative influence of banquet culture on villagers’ health.

6.1.3. Health Public Facilities Supply

The supply of public health services also influences banquet culture. The sums of “health care expenses” and “rural medical expenses” obtained from village questionnaires were categorized as health funding (HF) and subsequently classified into high and low groups according to whether they were above or below the average level. Data regarding the number of annual health lectures (HLs) were also utilized. Villages were categorized into high and low groups based on whether the number of lectures exceeded three. As depicted in Figure 4, villages with lower health expenditures and fewer health lectures exhibited a negative impact of banquet culture exhibited a negative impact of banquet culture on rural population health. This indicates that the negative effects of banquet culture have not been effectively mitigated. In contrast, the effect of banquet culture on villagers’ health outcomes becomes positive in villages with high health expenditures and a high number of health lectures. This suggests that when the supply of health public goods is relatively adequate, banquet culture, as an informal institutional embedded in social relations, functions as a bond of social capital [116]. Banquet culture influences health outcomes by shaping social behaviors and strengthening collective action capacities [117]. However, for banquet culture to be effectively transformed into public health capital, it must be regulated and guided through appropriate governance mechanisms, particularly institutional arrangements for the provision of public health goods. In other words, the transformation of social capital into public health capital depends not only on cultural practices themselves but also on their proper institutional embedding within supportive governance structures [118].

6.2. Further Tests of Health

The further impact of banquet culture on various health aspects, including village population mortality, infant mortality, longevity, and mental health, is analyzed in depth. Village population mortality is measured as the number of deaths per 1000 people, termed Population Mortality (Popmor). Infant mortality is similarly calculated as the number of infant deaths per 1000 population, designated as Infant Mortality (Infantmor). Longevity is defined as the average age at death of the population. Mental health is assessed using the Depression Scale, where lower values signify better mental health (Mhealth). As demonstrated in

Table 7. Further tests of health.

Variables	Popmor	Infantmor	Longevity	Mhealth
	Model 1	Model 2	Model 3	Model 4
Banqcul	3.2215 *** (0.5807)	1.3162 *** (0.2168)	−1.8989 ** (0.7976)	0.0239 (0.0309)
All Controls	Yes	Yes	Yes	Yes
Pseudo R2	0.1888	0.1792	0.1441	0.3647
Obs	1864	1735	1816	1913

The symbols *** and ** indicated 1% and 5% significance levels, respectively. Cluster-corrected standard errors (village level) are in parentheses.

, the regression coefficients of banquet culture with respect to village population mortality and infant mortality in Models 1 and 2 are significantly positive at the 1% significance level. This suggests that banquet culture is associated with increased mortality. The regression coefficients between banquet culture and the life expectancy of villagers in Model 3 are significantly negative at the 5% significance level, implying that banquet culture is associated with a shortened life expectancy. In Model 4, there is no statistically significant effect of banquet culture on mental health. This may seem counterintuitive, as previous research has demonstrated that social relationships and support networks often buffer individuals against psychological stress. However, banquet culture in rural China differs from public gatherings such as family dinners or holiday celebrations in that it is more normative and coercive pressures. These features may undermine the potential emotional benefits typically associated with social interaction. Moreover, given the limited empirical evidence linking banquet culture to psychological depression in existing literature, this study adopts this non-significant result.

7. Conclusions and Discussion

7.1. Conclusions

Institutionalist theory has been extensively applied across fields such as political science, economics, organizational studies, and public governance and has demonstrated clear theoretical applications. However, relatively few studies have explored the impact of cultural institutions on population health. This study, grounded in institutional theory, empirically examines the impact of banquet culture on villagers’ health and its underlying mechanisms by employing mediation effect models and machine learning techniques using data from the China Land Economy Survey (CLES). The results indicate that a stronger banquet culture is associated with lower health levels among villagers. Further analyses reveal that banquet culture is also found to increase population and infant mortality rates, as well as to shorten the life expectancy of rural residents. Mechanistic analyses suggest that labor time crowding and health decision-making are important mechanisms by which banquet culture affects villagers’ health levels. Heterogeneity analysis reveals that the adverse effects of banquet culture are more pronounced among youth groups in areas with weaker self-governance and where there is an inadequate supply of public health facilities. Self-governance among villagers has the potential to mitigate the negative health impacts of banquet culture. Furthermore, the availability of public health facilities can also counteract these negative effects. In other words, the provision of additional healthy public benefits to villages may transform banquet culture into a form of social capital that fosters popular support for the provision of public goods.

7.2. Discussion

This study enriches the understanding of how deeply rooted cultural institutions influence health outcomes in rural contexts, especially in developing countries. Existing public health research on rural populations has primarily emphasized structural determinants of health—such as income, education, and access to medical services—while often overlooking the influence of informal rural institutions [24,26]. This study provides empirical evidence that banquet culture exerts a detrimental impact on the health of rural residents. This finding aligns with sociological and anthropological perspectives, which suggest that ritual obligations and normative pressures can heighten individual vulnerability, particularly in resource-constrained settings [5,25].

Nevertheless, the social functions of banquet culture—such as fostering village cohesion, sustaining interpersonal networks, and facilitating mutual assistance—should not be dismissed. From a policy standpoint, institutional embeddedness plays a pivotal role [117,118]. Specifically, the health-promoting potential of informal institutions can be activated when they are effectively supported by formal governance mechanisms. Therefore, policy interventions should aim to preserve and leverage these social functions while mitigating the associated negative health impacts. Based on these findings, we propose the following policy recommendations.

First, it is essential to strengthen rural cultural governance and guide banquet culture toward a more affordable and sustainable trajectory. Consequently, it is recommended that policymaking concerning rural customs and culture be intensified, with a focus on reinforcing the beneficial aspects of the informal system as social capital to enhance the provision of public goods in rural areas. Second, it is recommended to encourage the active involvement of village self-governance organizations in regulating banquet-related behaviors, thereby enhancing the overall quality of rural health governance. Government departments should proactively foster a supportive institutional environment for grassroots autonomy. Through village self-governance structures, collective behaviors can be more effectively guided and regulated, including the establishment of standards and procedures for banquets, in order to promote self-restraint among residents.

7.3. Limitations

Although this study is based on rural areas in Jiangsu, China, similar cultural practices involving communal dining, social obligations, and conspicuous consumption can be observed across diverse cultural contexts—such as the lavish wedding banquets in South Asia and the ostentatious displays of wealth in Indian wedding celebrations. These phenomena highlight the profound influence of cultural traditions on health-related choices and lifestyle behaviors among rural populations. This study provides new empirical experience in cultural governance, but it must be acknowledged that due to limited data, the analysis is confined to a representative region and does not explore regional variations in banquet culture across the country. Meanwhile, cross-sectional data have certain limitations in causal inference. Future research could employ panel data or conduct field experiments to obtain more accurate causal relationships, thereby further deepening our understanding of how cultural norms affect health outcomes.