Asian Women - The Research Institute of Asian Women

Asian Women - Vol. 34, No. 3

The Role of Social Capital in Shaping Policy Non-compliance for Chhaupadi Practice in Nepal

Miyang Jun : Yonsei University, South Korea
Ikhyun Jang : The Seoul Institute, South Korea
funding :

Journal Information
Journal ID (publisher-id): RIAW
Journal : Asian Women
ISSN: 1225-925X (Print)
ISSN: 2586-5714 (Online)
Publisher: Research Institute of Asian Women Sookmyung Women's University
Article Information
Print publication date: Day: 30 Month: 09 Year: 2018
Volume: 34 Issue: 3
First Page: 47 Last Page: 70


Chhaupadi is a Nepalese custom that prohibits women from entering the house and participating in family activities during their menstruation period. The Nepalese Supreme Court outlawed chhaupadi in 2005 and the parliament has recently passed a bill to criminalize the practice. However, the decision has not ensured compliance, as the practice is still widely upheld in many regions in Nepal. Previous studies have attempted to explain behaviors of policy non-compliance with the roles played by social capital. It has been argued that a group with a high level of bonding social capital, represented by the level of solidarity within the group, may build its own belief system and become an obstacle to national governance. This paper examines whether the behaviors of policy non-compliance are associated with the social capital the community possesses. It uses a survey targeting over 400 local residents in Tikapur, a local community of Southeastern Nepal where half of the residents follow the practice. This study employs ordered logit regression to analyze the effect of social capital on individuals’ perceptions of chhaupadi. The results suggest that bonding social capital is associated with positive perspectives toward the practice and therefore it is more likely to shape non-compliant behaviors to a policy that prohibits the practice of chhaupadi.


The Nepalese word chhaupadi refers to women in menstruation. There is a series of rules applied to chhaupadi, although they vary slightly depending on the individual family, community, and region. The core principles for the practice of chhaupadi exile women from their families and forbid them to participate in various daily activities during menstruation. Women who are menstruating are restricted from touching men, children, cattle, living plants, public taps, and fruit-bearing trees, and to enter temples and their houses (Kunwar, 2016). Dairy products, such as butter, milk, and yogurt, are excluded from their diet (Gautam, 2017). Most women spend their days and nights outside the house, in places such as cattle sheds. Even so, women are expected to engage in hard labor outside their homes, such as digging, collecting firewood and grasses, and farming (Khadka, 2014). Such rules are applied to women once they start their menstruation, up to 13 days for the first and the second period, seven days for the third period, and four days a month afterward (Bhandaree, Pandey, Rajak, & Pantha, 2013).

There has been slow progress in the institutional recognition of women’s rights and in addressing the negative impacts of the practice on the health and safety of women. Although Nepal’s Supreme Court outlawed the practice in 2005 (Bhandari, 2013), newspapers constantly share stories of women raped, killed by wild animals, bitten by snakes, or dying of cold during their stay in the sheds (Bhandaree et al., 2013). The practice became known as responsible for the high rate of uterine prolapse in women, menstruation disorders, and a high rate of neonatal mortality (32 deaths per 1,000) (Bhandaree et al., 2013). Both Save the Children and UNICEF criticized the practice along with Nepalese governmental organizations, including the Ministry of Water Supply and Sanitation and the Ministry of Women, Children and Social Welfares (Smith, 2018), and women’s fundamental rights are acknowledged by the Interim Constitution of Nepal of 2007 (Malla, 2007), which states that

- there shall be no discrimination because of gender;
- every woman shall have the right to reproductive health and reproduction; and
- no physical, mental, or other kind of act of violence shall be committed against any woman; and there shall be equal rights to parental property for sons and daughters.

Moreover, the parliament of Nepal passed a bill in 2017 to criminalize the practice of chhaupadi. The bill stipulates a three-month jail sentence or a fine of 3,000 Nepalese rupees (around 30 USD) and came into effect in August 2018 (Sedhai, 2017).

The practice is, however, still widespread, particularly in the Hindu area of the hill districts in the Far-West and some Mid-West regions. For example, in Achham District, more than 95% of women were practicing chhaupadi in 2010 (UN Resident and Humanitarian Coordinator’s Office, 2011). However, the practice has recently been brought into the plains belt of the region, as the people in the hill districts migrate (Central Bureau of Statistics, 2014; Khadka, 2014). For example, Tikapur Municipality in Kailali District had 30.1% of women following the practice in 2012 (Yonsei University & TOPEC Engineering, 2012). According to a recent survey conducted in 2017 based on 414 households in Pathariya1 and Tikapur,2 the number of women following the practice has increased by 49% (Institute for Poverty Alleviation and International Development, 2017).

Sedhai (2017) finds the root of this prevalent practice in the Hindu belief that menstruating women are impure. The menstruation taboo in Hinduism is well documented. Narayanan’s (2005) work on gender and priesthood in the Hindu tradition highlights the period of menstruation as one of the reasons that prevents women becoming priests. A separation of purity and pollution is said to exist and, from a male perspective, menstruation is a period of pollution. Women are not allowed to be ritual specialists, as they necessarily have a period where their activities are interrupted. Hembroff’s (2010, p. 19) research on the orthodox Hindu belief in the Dharmasastra literature3 notes that menstruation is associated with adharma, not with dharma, a multi-faceted model of right actions that upholds the “Hindu’s ideal religious world.” Such beliefs assign women to certain roles and shape their attitudes and behaviors, such as being separated from others during their menstruation period. Selvi and Ramachandran’s (2012) research examines a link between menstruation taboos and compliance with menstrual hygiene practices, such as not touching cooked food, plants, or followers and infants; not participating in puja;4 and using a separate vessel and mat. Basing the research on 600 women living in both rural and urban areas of Cuddalore district of India, the study makes the argument that the belief that “menstruation is religiously impure and ceremonially unclean” is responsible for compliance, especially for those women from higher socio- economic backgrounds.

While existing research finds the root of the menstruation taboo in Hinduism, it does not seem fair to argue that the religion as a whole is responsible for such a practice or the violations against women who subscribe to the tradition. According to the work of Guterman, Mehta, and Gibbs (2007) on menstruation taboos in major religions, Judaism, Christianity, Islam, Hinduism, and Buddhism all share taboos against menstruating women and associate them with impurity, spiritual and physical danger, and harm to society. While the level of severity varies according to the religion, they all require women to be isolated and excluded from a series of activities. All religions studied by Guterman et al. (2007) restrict menstruating women from attending religious services and engaging in sexual intercourse. In addition, not all individual believers in Hindu society comply with the practice of chhaupadi. While over 80% of Nepal’s population believe in Hinduism, the population that follows the practice varies widely depending on the location, and ranges from 30% to 95% (UN Resident and Humanitarian Coordinator’s Office, 2011; Yonsei University & TOPEC Engineering, 2012). In addition, many societies also possess social and cultural elaborations5 in their social structure, which maintain or reinforce control over women’s bodies and behaviors without a religious rationale. Such controls often appear to be socially and economically necessary, because they are directly connected to “family honor and the modesty code, community standing and reputation, the organisation of economy, age at marriage and marriage patterns” (Fox, 1977, p. 806). While a particular religion cannot solely be responsible for a menstruation taboo in a given society, religion is acknowledged as one of the most powerful forces that shape social norms, along with the family and social networks (Green, 2016). This study therefore understands the chhaupadi practice as a form of social control over a woman’s body and behavior that is interwoven in complex ways with religious beliefs.

This study focuses on the social aspects of the chhaupadi practice and seeks an explanation as to why it still persists despite a series of policy measures that prohibit it. In doing so, the paper uses a concept of social capital that refers to norms and social networks that stimulate collective action for mutual benefit in social relationships (Bourdieu, 1986). We believe that using the concept of social capital is useful in understanding the non-compliance behaviors in relation to the chhaupadi practice because, as Anderson (1979) argues, people tend to avoid unlawful behaviors because they wish to avoid the stigma of being a lawbreaker more than they fear the penalties. The policy measures that prohibit the practice and the fear of punishment would then not be very effective in persuading people to give up the practice, given that the idea of a menstruation taboo is deeply rooted in their social norms and therefore encouraged by collective actions (Sedhai, 2017), and when the stigma of being a deviant within the community by not following the practice is weighted more negatively than the stigma of lawbreaking. In addition, as the criminalization of the practice is largely driven by external advocates, its influence over the value placed upon the practice by the Nepalese is still in question.

To provide an understanding of the stickiness of chhaupadi, this paper focuses on the role of social capital in shaping the behaviors of policy non-compliance. In doing so, the paper explores the factors that interact with the behaviors of policy non-compliance in the next section, and discusses the role of social capital as an important contributing factor to the behaviors of policy non-compliance in the following section. The paper then presents the method and findings. Based on findings that suggest bonding social capital contributes to the persistence of the chhaupadi practice, the paper concludes by suggesting a need to reconceptualize the prevailing norms around chhaupadi by raising awareness of its unjustifiable violations against women.

Policy Non-compliance

Researchers have conceptualized policy non-compliance in relatively similar ways. Duncan (1981) defines the concept as a modification of behaviors that are accepted by government or society. Young (1979) and Anderson (1979) define policy non-compliance as the behavior of individuals that violates prescribed rules of government.

Existing knowledge explores factors that encourage policy non-compliance, which can be categorized in three ways. First, policy non-compliance is about the policy itself (Coombs, 1981). Non-compliance occurs when an instruction is considered to be vague, abstract, or inconsistent. It also occurs when the policy is considered to be undesirable for the community owing to political controversies. Second, policy non-compliance is about the implementers (Anderson, 1979; Young 1979). A policy is not complied with when implementers do not have an incentive to implement the policy, either because of a lack of sufficient resources or powers, or a belief that the policy is not in keeping with their own interests. Third, it is about the target groups (Dye, 1972; Young, 1979). Policy non-compliance occurs when it is not understood by the target group or when it is considered to be incompatible with their interests. Young (1979) highlights the position that target groups are unlikely to comply with government policy when it is perceived to be against their own traditions or customs. In particular, Anderson’s (1979) book Public Policymaking explains how a custom may cause policy non-compliance depending on one’s social relationships and group membership. When a law conflicts with prevailing beliefs and values or when one has a connection with people who rationalize and justify an unlawful behavior, one avoids becoming a deviant by complying with the accepted behavioral norms. The fear of punishment is therefore less effective when the stigma of being a deviant in one’s own community is given more weight than being a lawbreaker.

Given that chhaupadi practice is a custom that has been upheld as an unlawful behavior historically, this study aims to provide an understanding for such non-compliance, using the concept of social capital. The following section explores the role of social capital in shaping behaviors of policy (non)compliance.

Social Capital and Policy (Non)compliance

Social capital is used as a comprehensive term that includes norms and social networks that stimulate collective action for mutual benefit in a social relationship. Social capital has been conceptualized both at the individual and community level. Bourdieu (1986) sees social capital at the individual level and argues that it is the sum of resources individuals obtain from sustainable relationships in social networks. An individual’s social capital depends on the size of their network, and the amount of physical capital of each person in the network. Social capital is therefore closely related to the individual’s capacity to access resources in the society through social relationships. On the other hand, Putnam (1995) has conceptualized social capital in relation to the attributes of individuals or the community to which the individuals belong. Social capital in this respect refers to conditions that promote coordination and cooperation, and consists of cooperative networks, reciprocal norms, or social trust. These three components facilitate one another, as cooperative networks stimulate the growth of reciprocal norms and social trust.

Based on Putnam’s (2000) conceptualization, existing research has provided evidence on the influence of social capital over governance and policy outcomes. The main assumption is that a community with a higher level of social capital would have better governance and a safer and richer environment (Jones, Sophoulis, Iosifides, Botetzagias, & Evangelinos 2009; Woolcock, 1998). Putnam (1993) highlights the role of social capital in shaping the performance of local governments in Northern and Southern Italy. Social capital is measured by a number of local civil communities, the turnout rate in local elections, and the subscriber rates for local newspapers. The performance of the local government is measured based on the policy process, policy declaration, and policy implementation. The study suggests that a higher level of social capital in Northern Italy contributes to the better performance of their local government. Compared to Southern Italy, local residents in Northern Italy actively participate in and discuss local policy issues, and they trust the transparency of the policy process. As a consequence, local governance in Northern Italy gains more legitimacy and becomes more effective than that in Southern Italy. Putnam’s study conducted in 2000 resulted in a similar finding in the case of the U.S.A., where a higher level of social capital is associated with positive results in child welfare, safety, economic prosperity, health, and the happiness of local residents. Tavits’s (2006) study also highlights a positive relationship between levels of social capital and the performance of governments in Germany and the U.S.A. A higher level of social capital is associated with an increase in policy activism, and the extent to which governments allocate resources in the provision of public goods and services. Tavits (2006), however, indicates that social capital is more related to the policy activity of the government than the efficiency of administration.

However, not all studies of social capital indicate positive implications for governance and policy outcomes (Coleman, 1988). The associations are found to be varied according to the different types of social capital. Granovetter (1973) conceptualizes strong and weak ties in social capital, which are later named bonding and bridging, respectively (Gittell & Vidal, 1998, p.10). Bonding in social capital refers to social cohesion within a community. A community with strong bonding social capital would share the same culture, values, or traditions, which contribute to stimulating solidarity or unique reciprocal relationships within the community. In Putnam’s (2000) study an ethnic minority community in America offers a good example, in terms of providing reliable financial or labor supports between community members based on their strong social networks. Bridging in social capital refers to networks between groups from different cultures, values, or traditions. Bridging functions as an instrument to spread or obtain useful information from the wider community. While bonding and bridging in social capital indicate horizontal relationships between people, there is also a vertical dimension of social capital, “linkage” (Woolcock, 2002). Linkage in social capital forms alliances between individuals and others in positions of power that function as leverage for individuals to obtain resources, ideas, and information from formal institutions beyond their community.

How the three categories of social capital are responsible for policy (non)compliance has been the subject of various studies. Harris’s (2007) study argues that a high level of bonding social capital, represented by the level of solidarity within a group, is related to a high level of corruption in the society. This type of social capital prevents policy innovation or new ways of thinking as it encourages traditional shared values and norms within the group. Studies conducted by Aldrich (2012) and Aldrich and Crook (2008) also acknowledge that bonding in social capital creates obstacles to national governance in cases of disaster. Communities with a strong sense of bonding have internally strong solidarity and self-regulatory systems. This exclusiveness is often expressed as hostility against other communities, and creates distrust in government public policy on disaster management. Woolcock (2001), however, warns that bonding in social capital cannot be discussed without understanding the institutional context and the role of the state, especially in the case of developing countries. A high level of engagement between community members may be encouraged by an absence of formal institutions, or the presence of weak or hostile ones––a lack of linkage in social capital.

There is not much work on the role of bridging in social capital and policy compliance. Only a few studies highlight how bridging can play a role in increasing the integrity of governance, which may indirectly affect the level of policy compliance. Lee’s (2008) work on social capital and levels of corruption among OECD countries argues that some countries with higher bridging social capital show higher integrity than those with lower bridging social capital. Myeong and Seo (2016) argue that bridging social capital is positively associated with trust in government because it provides wider networks, horizontal associations, and opportunities to evaluate government services. Other studies have raised the role of bridging social capital in making people better off economically, and supporting them in times of disaster. The study of Zhang, Anderson, and Zhan (2011), based on the National Survey of Families and Households panel in the United States, highlights the theme that bridging social capital makes a positive contribution to economic well-being. Hawkins and Maurer’s (2009) study on the role of social capital during Hurricane Katrina in New Orleans, highlights bridging and linking in social capital as a means of positively contributing to long-term survival and wider neighborhood and community revitalization.

Despite the relative lack of studies on how linkage in social capital affects policy non-compliance, three ways of categorizing the relationship have been identified. First, policy non-compliance is often about how individuals perceive the credibility of the rules imposed. The research of Dietz, Ostrom, and Stern (2003) finds a higher level of rule compliance in the Maine lobster fishery, than in inshore fisheries. The difference between the two fisheries was found to be because of the different levels of credibility concerning the rules imposed. The Maine lobster fishery was governed by state-level rules that were created by formal and informal user institutions, while the inshore fishery was governed by national rules through a top-down regime. The research argues that the latter rules imposed on inshore fisheries created strong resistance in the community. Second, policy non-compliance occurs when individuals question the process of policy implementation. Anderson (1979) highlights the decision for the nuclear waste disposal site under President George W. Bush. Following enactment of the Nuclear Waste Policy Act in 1982, the Department of Energy designated three sites for waste disposal: Deaf Smith County, Texas; Yucca Mountain, Nevada; and Hanford Nuclear Reservation, Washington. While all of the communities fought against being selected, Congress summarily passed legislation and made Yucca Mountain in Nevada the final disposal site. This act faced strong opposition, including lawsuits, and various political and public relations tactics. Third, Boix and Posner (1998) argue that people are more likely to disobey when they are not convinced by the effectiveness of a policy’s implementation because they do not expect others, for example, to pay taxes or comply with regulations like themselves. Using the same example of a nuclear disposal site, Anderson (1979) also highlights how individual distrust concerning the effectiveness of a policy may address non-compliance behaviors. After approval of the disposal site was granted, Nevada exercised a veto that was quickly overridden by Congress. President Obama subsequently came into power in 2009 and cut most of the funding for the disposal site. Despite decades of struggle, nuclear waste disposal remains a problem without a permanent solution. The waste is temporarily disposed of in over a hundred sites around the nation and every site hopes to remove itself from the list.

Existing research shows us that bonding in social capital may result in non-compliance, especially when the policy is imposed, or is at odds with prevailing beliefs or customs. On the other hand, bridging and linkage in social capital may entail a good level of policy compliance when positive perspectives are created by individuals concerning the policy itself or its implementation. This study therefore aims to examine how different types of social capital shape individual perspectives on the chhaupadi practice in Nepal and, in turn, contribute to the retention of an unlawful tradition.


The study uses survey data created by the Institute for Poverty Alleviation and International Development, part of a research project on the capabilities of Tikapur, Nepal. The survey targeted 370 households, including 1,197 individuals in Tikapur and explored their perceptions across a wide range of life domains, as shown in Table 1. Here, we examine the role of social capital in shaping individuals’ perception of the chhaupadi practice that was expressed by 412 individuals in the data.

Table 1. 
Contents of Survey
Content Description
Respondent Details household, age, sex, marriage, education, jobs, total income, wage income, health, religion
Household Composition the same information as above for other household members
Child Education age, sex, marriage, education status, school type, child labor and income, health
Housing and Living
house ownership, construction materials, water supply, sanitation, garbage disposal, toilet, energy, assets
Access to Facility type of facility, availability, mode of transportation, time distance
Agriculture farm income, livestock income, cooperative, loans, inputs, training, irrigation, farm size, labor, etc.
Crop Production production, consumption, sales, price
Livestock Production type of livestock, number of livestock, price
Food Security food quality, quantity, choice
Social Capital Likert scale on cooperation, trust, concent, beliefs, safety, etc.
Governance Likert scale on government accountability, services, equity, leadership, community relationships, etc.
Networking &
friendly gathering, group meetings, religious festivals, etc.
Empowerment &
decision-making process, influences, responsibility, etc.
Satisfaction & Happiness perception of life satisfaction and happiness
Disease & Sanitation health and sanitation habits, vulnerability, sickness, etc.
Chhaupadi Practice perception of chhaupadi practice
Nutrition & Health diet & physical exercise
Health Status moving, vision, hearing, concentration, self-caring, etc.
Chronic Conditions pain or discomfort in joint, back, heart, asthma, stomach, and depression
Health System Response accessibility to health care
Health Outcomes general assessment of Healthcare information technology project in Tikapur
Health Literacy ability to read and understand written health information

The impacts of social capital on the individuals’ perception on chhaupadi were tested in relation to the three types of social capital discussed above: bonding, bridging, and linking. From the questionnaire, three questions were designed to measure each type of social capital. First, bonding social capital, which refers to solidarity within the group, was measured by the following statements: 1) I have a cohesive or unified neighborhood, 2) people in this neighborhood share the same values, and 3) people in this neighborhood can be trusted. Second, bridging in social capital, which reflects the solidarity between different groups, was examined through three statements: 1) I get along well with people from other backgrounds, 2) I am comfortable with people of other cultural backgrounds, and 3) I work with individuals and organizations outside the community to solve local problems. Third, linking social capital, which reflects the relationship between individuals and government, was measured by three statements reflecting the degree of linking: 1) My government is accountable for its policies, 2) my government takes responsibility for the needs of the community and development, and 3) my government delivers public services effectively. All questions above were designed according to a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree).

There are two ways to create a composite score (Warmbrod, 2014). First, an individual’s score on each type of social capital may be the sum of the individual’s responses to the three questions. Alternatively, a composite score can be calculated by a mean-item summed score, which is the summed score divided by the number of questions. Warmbrod (2014) showed no significant difference between the two methods unless there are missing values. As the data used in this study are not missing values, the sum of the questions on each type of social capital was used.

The dependent variable of this study is the individuals’ perceptions of chhaupadi. Existing research suggests that individuals value the practice of chhaupadi traditionally, culturally, and religiously, which contributes to the low levels of compliance with government policy that outlaws the practice. There is one question to measure individuals’ perception on the chhaupadi practice specifically: In your opinion, is it necessary for women to follow the practice of chhaupadi, where women change their lifestyle in terms of sleep, food, and things that can be touched? The question’s values range from 1 (strongly disagree) to 5 (strongly agree).

There are a number of controlled variables, including individual attributes, such as gender, age, and education level that are measured by years of schooling attained and social class (caste). Caste in this survey has four categories ranging from lowest to highest, with the lowest caste being a reference group. The residence of respondents (urban/rural) is also controlled. Regression analysis was employed to examine the relationship between dependent and independent variables. However, as the dependent variable (the perception on chhaupadi practice) is categorical, ordered logit regression analysis was used rather than simple ordinary least squares (OLS) regression to examine the impact of the independent variables on the perceptions of chhaupadi. The following equation shows the model used.

perception on chhaupatii = social ranki + genderi + educationi + urbani + social capital i+ϵi

The following hypotheses were tested:

Hypothesis A: When bonding social capital is stronger, people approve of the practice of chhaupadi more.

Hypothesis B: When bridging or linking social capital are stronger, people disapprove of the practice of chhaupadi.


Table 2 shows the demographic characteristics of individuals in the study. There were more females (61.3%) than males (38.6%), the average age of respondents was 33.76 years, and the average years of schooling were seven, which means that the average education level is between primary school graduation and secondary school. The average years of schooling for males were 8.48 years, which is slightly higher than for females. Some 40% of respondents belonged to the lowest social class, while half of the respondents were members of the upper social class (Class 3 and 4). Respondents are almost equally distributed between urban and rural areas.

Table 2. 
Demographic Characteristics of Respondents
Variable Male Female
Gender 159 (38.6%) 253 (61.4%)*
Caste  1 (Lowest) 65 (40.88%) 103 (40.71%)
 2 18 (11.32%) 21 (8.3%)
 3 40 (25.16%) 75 (29.64%)
 4 (Highest) 36 (22.64%) 53 (20.94%)
 Age 35.9 years old 32.3 years old
Average Years of Schooling 8.48 years 6.14 years
Note. *One female does not respond her Caste

Table 3 shows the level of social capital, which varies across gender and residence. Bonding social capital was relatively higher than other types of social capital, but there was no particular difference by gender or place of residence. Bridging social capital was higher in rural areas than in urban areas. Linking in social capital was higher for males than for females, but did show a difference by place of residence.

Table 3. 
The Level of Social Capital by Gender and Place of Residence
The Level of Social Capital* Gender Residence
Female Male Rural Urban
Bonding 10.56 10.88 10.67 10.82
Bridge 9.39 9.32 9.77 8.92
Linking 8.88 9.34 9.08 9.25
Note. *The level of social capital is ranged from 3 to 15, as this is the sum of three questions with 5-point Likert scale.

Table 4 presents the differences in the perception of chhaupadi practice across gender, caste, age, and place of residence. The perception of chhaupadi practice has values ranging from 1 to 5, where the highest score indicates the highest level of value placed on the practice. While gender and age group did not play an important role, perceptions of chhaupadi practice became more positive for urban respondents who were of a higher caste.

Table 4. 
Perceptions on Chhaupadi Practice
Caste Age Gender Residence
1 2 3 4 Age < 30 Age> 30 Female Male Rural Urban
3.1 3.5 3.62 4.04 3.50 3.502 3.51 3.49 3.34 3.65

Table 5. 
Regression Model
Model 1 Model 2 Model 3 Model 4 Model 5
Social class
2 .9644**
3 .9570**
4 (highest) 2.182**
(female) .7437
Age -.0068
Education -.0315
Urban .3752*
Bonding .2679***
Bridging .2167***
Linking -.1858***
Observations 412 412 412 411 411
Log Likelihood -493.47 -483.69 -486.91 -485.46 -462.71
Note. *coefficient is significant at 10% level; **coefficient is significant at 5% level; ***coefficient is significant at 1% level

The tables above show five models. Model 1 presents the relationship between the perceptions of chhaupadi practice and the controlled variables. Models 2–4 show the effects of bonding, bridging, and linking social capital on perceptions of the practice. Model 5 captures how the effect of social capital changes when all three forms of social capital are considered at the same time.

Model 1 shows that caste is the only variable that significantly affected the perceptions of respondents. The perception was more positive when caste was higher. Model 2 supports hypothesis A, indicating that a strong bonding social capital was related to a positive perception of chhaupadi practice. A stronger bonding social capital strengthened solidarity within the group, which led residents to believe more in their own traditions, values, and culture than in government policy. While most individual attributes, including education level, gender, and age did not have a significant impact, social class as measured according to the caste system played a role. This suggests that higher caste individuals are more likely to have a positive perception of chhaupadi practice than those of a lower caste. They are therefore more likely not to comply with the government policy bans on chhaupadi.

Model 3 shows the effect of bridging social capital on the perception of chhaupadi practice. While the existing literature indicates a positive relationship between bridging social capital and policy outcomes, bridging social capital here showed the reverse relationship by increasing positive perceptions of chhaupadi. While other individual attributes were non-significant, the caste system showed a significant impact. Model 4 presents the effects of linking social capital on the perception of chhaupadi practice. This is a negative coefficient, which means the respondents perceived the practice in more negative ways when they had stronger trust in the government and its policy. Model 5 considers the three types of social capital together. When bonding and bridging in social capital were treated together, a significant impact on the positive perception of chhaupadi practice was found. The other variables remained the same as in Models 1 and 2. Linking social capital still contributed to negative perceptions of chhaupadi practice.

To summarize, the empirical analysis shows that both bonding and bridging in social capital were related to an increase in positive perceptions for chhaupadi practice. It can be inferred that both bonding and bridging contributed to the low level of policy compliance by expressing strong ties among community members and thereby strengthened the retention of the practice. Individuals’ personal attributes consistently showed a non-significant impact apart from caste, which had a significant impact on participants’ perceptions. Individuals were likely to have positive perceptions of the practice when they belonged to a higher caste in all five of the models. Arguably, this adds to a further understanding of behaviors pertaining to policy non-compliance, as individuals in higher castes were likely to have higher social and political positions and possess power and influence over other community members. On the other hand, linking social capital was related to negative perceptions of chhaupadi practice. Respondents who had stronger trust in government accountability and its ability to undertake policy decisions and implementation had negative perceptions of the practice.


By examining the correlations, the findings suggest the role played by bonding social capital in the retention of chhaupadi practice in Tikapur. They raise a number of interesting discussion points, while possessing some empirical limitations and providing room for further studies.

First, while other individual attributes do not play a significant role, higher caste individuals tend to have more positive perceptions of chhaupadi. The relation is significant across all types of social capital. Important implications arise from this relationship, as the caste system itself is closely associated with other social and economic indicators in Nepal. Although Nepal adopted democracy in 1990, its caste system still prevails in everyday social relations in both the private and the public spheres (Bennet, 2005). The dominant elites are still confined to the men of upper castes (Brahmin and Chhetri), while those with less power, especially women and those in the lower castes, are marginalized. The caste system creates normative social differences in many aspects of Nepalese everyday life in terms of the foods people eat, the ritual places they enter, and the occupations they engage in (Nightingale, 2011). This normativity similarly shapes different educational opportunities. High-caste people, such as the Brahmin and Chhetri, are more able to access educational opportunities and can expect to obtain literacy skills to fulfill their social and religious roles. Thus, they tend to achieve better educational outcomes. As Jamison and Lockheed’s (1987) work on the Bara and Rautahat districts of Nepal shows, households in a higher caste are more likely to send their children to school compared to people in a lower caste. This difference in educational opportunities again creates an imbalanced distribution of power that favors those in higher castes. As a result, higher caste people in Nepal tend to be socioeconomically advantaged and occupy the majority of the ruling positions in the government, education, and the economy (Stash & Hannum, 2001). The fact that this ruling group in Nepalese society places great value on chhaupadi contributes to the retention of the custom, as they are in a better position to establish and spread normative behaviors for other Nepalese.

Stash and Hannum’s (2001) study on educational stratification in Nepal explains why people in a higher caste may have imposed stricter restrictions on women than on people in lower castes. Those in the ruling groups are mostly men in higher social and religious positions, as higher caste women are denied access to social positions outside their home, particularly in remote or socially conservative regions (Stash & Hannum, 2001). Privileged men possess better resources to implement stricter standards and govern women’s activities, for example, arranging an extra hut for the practice of chhaupadi. Similarly, in a study conducted by Boserup, Tan, and Toulmin (2013, p. 54) on women’s role in economic development, wives from wealthy households that cultivate their own land are encouraged to avoid all manual work outside the household and be secluded at home. It functions to distinguish women of wealthier households from “the despised” female laborers. Sticking strictly to the tradition would therefore be symbolized as a respectable behavior for the economically better off and would then be developed as a point of pride for those of a higher caste in Tikapur. The survey, however, fails to measure how strictly chhaupadi is performed, for example, the days women spend in the shed, the places they cannot enter, and the things they cannot touch. Therefore, we cannot provide any insight into whether the practice is more strictly enforced by the higher castes. Further research is necessary to address this limitation.

Second, both bonding and bridging social capital seem to contribute to the behaviors associated with policy non-compliance in regards to chhaupadi. Respondents with strong bonding and bridging social capital are more likely to value and uphold the practice, as they believe it is necessary; traditionally, culturally, or religiously encouraged; and they understand the practice as a caste-related norm. As disaster studies conducted by Aldrich (2012) and Aldrich and Crook (2008) explain, the solidarity within a group builds its own self-regulatory system that favors behaviors that maintain traditional values and norms over behaviors that comply with a government policy, particularly when the original system is first developed. Moreover, when policy compliance is more about avoiding the stigma of being a lawbreaker than the behavior that itself breaches the norm as Anderson (1979) suggested, the fear of being regarded as deviant within their community (by not following the practice) would have been a bigger challenge for the local residents of Tikapur than the stigma of lawbreaking. In addition, as women in Nepal are likely to be excluded from education, labor market opportunities, and decision-making processes over their property (Asian Development Bank, 1999), they tend to have less chance to form social capital. The impact of bonding social capital could be changed if bonding becomes stronger among women. The survey, however, fails to measure whether the respondents practiced chhaupadi, given that the Supreme Court decision outlawed it in 2005. This calls for further study, as there is a good chance that the respondents may not even be aware that the practice has been criminalized.

While some studies indicate that bridging social capital has positive impacts on the level of integrity and trust in governance (Lee, 2008; Myeong & Seo, 2016), the findings of this paper show that bridging social capital is associated with positive views on chhaupadi practice, which result in non-compliant behaviors. These results lead us to reconsider whether the survey questions used to represent bridging social capital appropriately reflected the functional element of bridging social capital for Tikapur residents. Bridging social capital is conceptualized as a network between groups from different cultures, values, or traditions that functions as an instrument to obtain or spread useful information from outside communities (Gittell and Vidal, 1998). However, the questions used to operationally define bridging social capital here largely focus on having “different backgrounds,” while leaving out the instrumental element of spreading or obtaining new information from outside the community. Further studies on communities that adopt this instrumental element of bridging social capital more fully would be useful.

Third, the role played by bonding social capital––the in-group solidarity for the behaviors of policy non-compliance––becomes clearer when taking account of the role played by linking social capital. The respondents who strongly perceive the accountability and effectiveness of policy implementation tend to have a negative perception of chhaupadi and are therefore less likely to uphold the traditional form of the practice. The survey supports existing research on the positive association between linking social capital and the behaviors of policy compliance. However, it fails to measure whether those people with high levels of trust in government have moved away from the practice, or modified it to be more flexible compared to those who have lower levels of trust in policy implementation. Further data on the rigor individuals employ in supporting chhaupadi would provide further knowledge of social capital’s role in shaping policy non-compliance.


This paper focuses on the social aspects underlying the persistence of chhaupadi in Nepal, despite a series of policy measures that prohibit the practice. Using the concept of social capital provides an understanding of how our social networks––formed by bonding, bridging, and/or linking––influence our perceptions and behaviors and thereby contribute to reinforcing, maintaining, or changing a practice complied with because of tradition, religion, and/or social norms. The findings suggest that one policy implication for further consideration is that a change requires more than the establishment of a law. As change pertains to the norms created, reinforced, or maintained by the types of social networks individuals engage in, government interventions should be geared toward reconceptualizing the prevailing norms around chhaupadi by raising awareness of its unjustifiable violations against women.


* This work was supported by the Ministry of Education of the Republic of Korea; the National Research Foundation of Korea [grant number NRF-2016S1A5B8925203].

1 Pathariya is a village development committee in Jhapa District in the Mechi Zone of South-Eastern Nepal.
2 Tikapur is a municipality in Kailali District in the Seti Zone of Western Nepal.
3 Dharmasastra refers to the treatises of Hinduism on dharma.
4 Puja refers to the act of worship.
5 See Fox (1977) for three types of control strategies over women’s body and behavior.

1. Aldrich, D. P., (2012), The power of people: Social capital’s role in recovery from the 1995 Kobe Earthquake, Natural Hazards, 56(3), p595-561. []
2. Aldrich, D. P., & Crook, K., (2008), Strong civil society as a double-edged sword: Siting trailers in post-Katrina New Orleans, Political Research Quarterly, 61(3), p379-389.
3. Anderson, J. E., (1979), Public policymaking, Boston, MA, Wadsworth Cengage Learning.
4. Asian Development Bank, (1999), Country briefing paper: Women in Nepal, Retrieved July 10, 2018, from
5. Bennet, L., (2005, December), Gender, caste and ethnic exclusion in Nepal: Following the policy process from analysis to action, Paper presented at the World Bank Conference on New Frontiers of Social Policy: Development in a Globalizing World, Washington, Retrieved from
6. Boix, C., & Posner, D. N., (1998), Social capital: Explaining its origins and effects on government performance, British Journal of Political Science, 28(4), p686-693. []
7. Boserup, E., Tan, S. F., & Toulmin, C., (2013), Woman’s role in economic development, London, Routledge.
8. Bourdieu, P., (1986), The forms of capital, In J. Richardson (Ed.), Handbook of theory and research for the sociology of education, p241-248, New York, Greenwood.
9. Bhandaree, R., Pandey, B., Rajak, M., & Pantha, P., (2013), Chhaupadi: Victimizing women of Nepal, Scientific Committee of Reviewers, p141-143.
10. Bhandari, S., (2013), Cultural conceptions and silent acceptance of stereotypes: A critical study on Nepalese women’s situation, (Unpublished Master’s thesis), University of Stavanger, Stavanger, Norway.
11. Central Bureau of Statistics, (2014), Nepal population and housing census 2011, Kathmandu, Nepal.
12. Coleman, J. S., (1988), Social capital in the creation of human capital, American Journal of Sociology, 94, pS95-S120.
13. Coombs, F. S., (1981), The bases of noncompliance with a policy, In J. G. Grumm, & S. L. Wasby (Eds.), The analysis of policy impact, Lexington, VA, Heath.
14. Dietz, T., Ostrom, E., & Stern, P. C., (2003), The struggle to govern the commons, Science, 302(5652), p1907-1912. []
15. Duncan, J. W., (1981), Organizational behavior, (2nd ed.), Boston, MA, Houghton Mifflin.
16. Dye, T. R., (1972), Understanding the public, Englewood Cliffs, New York, Prentice Hall.
17. Fox, G. L., (1977), ‘Nice girl’: Social control of women through a value construct, Signs: Journal of Women in Culture and Society, 2(4), p805-817. []
18. Gautam, Y., (2017), Chhaupadi: A menstrual taboo in far western Nepal, Journal of Nursing & Healthcare, 2(4), p1-4.
19. Gittell, R., & Vidal, A., (1998), Community organizing: Building social capital as a development strategy, Thousand Oaks, CA, Sage.
20. Granovetter, M. S., (1973), The strength of weak ties, American Journal of Sociology, 78(6), p1360-1380.
21. Green, D., (2016), How change happens, Retrieved from []
22. Guterman, M., Mehta, P., & Gibbs, M., (2007), Menstrual taboos among major religions. The Internet, Journal of World Health and Societal Politics, 5(2), p1-7.
23. Harris, D., (2007), Bonding social capital and corruption: A cross-national empirical analysis, (Environmental Economy and Policy Research Working Papers, No. 27), Retrieved from
24. Hawkins, R. L., & Maurer, K., (2009), Bonding, bridging and linking: How social capital operated in New Orleans following Hurricane Katrina, British Journal of Social Work, 40(6), p1777-1793. []
25. Hembroff, N., (2010), Orthodox Hindu attitudes to menstruation, (Unpublished doctoral dissertation), University of Lethbridge, Lethbridge, Alberta.
26. Institute for Poverty Alleviation and International Development, (2017), IPAID’s capability data series: Tikapur, Nepal, Unpublished raw data.
27. Jamison, T., & Lockheed, M. E., (1987), Participation in schooling: Determinants and learning outcomes in Nepal, Economic Development and Cultural Change, 35(2), p279-306. []
28. Jones, N., Sophoulis, C. M., Iosifides, T., Botetzagias, I., & Evangelinos, K., (2009), The influence of social capital on environmental policy instruments, Environmental Politics, 18(4), p595-611. []
29. Khadka, N., (2014), Chhaupadi pratha: Women’s condition and suffering, (Doctoral dissertation, Tribhuvan University, Kritipur, Nepal), Retrieved May 5, 2018, from
30. Kunwar, P., (2016), Coming out of the traditional trap, Asian Journal of Women’s Studies, 19(4), p164-172. []
31. Lee, J. J., (2008), A comparative study on the relationship between social capital and corruption among countries, Proceedings of the Korean Association for Public Administration (KAPA) Spring Conference, Jeon-ju, Korea.
32. Malla, S. P., (2007), Interim constitution of Nepal: Women’s rights and challenges of implementation, Renaissance Message, 2, p77-122.
33. Myeong, S., & Seo, H., (2016), Which type of social capital matters for building trust in government? Looking for a new type of social capital in the governance era, Sustainability, 8(4), p322-326. []
34. Narayanan, V., (2005), Gender and priesthood in the Hindu traditions, Journal of Hindu-Christian Studies, 18(8), p1-10. []
35. Nightingale, A. J., (2011), Bounding difference: Intersectionality and the material production of gender, caste, class and environment in Nepal, Geoforum, 42(2), p153-162. []
36. Putnam, R. D., (1993), Making democracy work: Civic traditions in modern Italy, Princeton, NJ, Princeton University Press. []
37. Putnam, R. D., (1995), Bowling alone: America’s declining social capital, Journal of Democracy, 61, p65-68.
38. Putnam, R. D., (2000), Bowling alone: The collapse and revival of American community, New York, Simon & Schuster.
39. Sedhai, R., (2017, August, 10), Nepal passes law to end practice of exiling women for menstruating, Independent Digital News and Media Limited, Retrieved June 9, 2018, from
40. Selvi, K. T., & Ramachandran, S., (2012), Socio-cultural taboos concerning menstruation: A micro level study in the Cuddalore district of Tamil Nadu, India, International Journal of Scientific Research Publications, 2(8), p1-7.
41. Smith, C., (2018, March, 18), Parliament in Nepal passes a bill to outlaw chhaupadi, Retrieved from Save the Children website:
42. Stash, S., & Hannum, E., (2001), Who goes to school? Educational stratification by gender, caste, and ethnicity in Nepal, Comparative Education Review, 45(3), p354-378. []
43. Tavits, M., (2006), Making democracy work more? Exploring the linkage between social capital and government performance, Political Research Quarterly, 59(2), p211-225. []
44. UN Resident and Humanitarian Coordinator’s Office, (2011, April), Field bulletin: Chhaupadi in the Far-West, (Transition Support Strategy RCHC Office, Nepal. Issue No. 01), Retrieved from
45. Warmbrod, J. R., (2014), Reporting and interpreting scores derived from Likert-type scales, Journal of Agricultural Education, 55(5), p30-47.
46. Woolcock, M., (1998), Social capital and economic development: Toward a theoretical synthesis and policy framework, Theory and Society, 27(2), p151-208.
47. Woolcock, M., (2001), The place of social capital in understanding social and economic outcomes, Canadian Journal of Policy Research, 2(1), p11-17.
48. Woolcock, M., (2002), Social capital in theory and practice: Where do we stand?, In J. Isham, T. Kelly, & S. Ramaswamy (Eds.), Social capital and economic development: Well-being in developing countries, p85-103, Cheltenham, UK, Edward Elgar Publishing. []
49. Yonsei University & TOPEC Engineering, (2012), Nepal Tikapur health services improvements project masterplan, (Unpublished report), MoHP, KOICA & Yonsei University, Wonju, Korea.
50. Young, O., (1979), Compliance and public authority: A theory with international applications, Baltimore, MD, Johns Hopkins University Press.
51. Zhang, S., Anderson, S. G., & Zhan, M., (2011), The differentiated impact of bridging and bonding social capital on economic well-being: An individual level perspective, Journal of Sociology & Social Welfare, 38, p119-142.

Biographical Note: Miyang Jun is an early career researcher with research interests in ‘family policy’, ‘welfare reform and conditionality’ and ‘social security benefits’. She has awarded a Ph.D. degree on lone mothers’ subjective well-being over the transitions from welfare to work from Department of Social Policy and Social Work, University of York. She now works for Institute for Poverty Alleviation and International Development, Yonsei University, South Korea. E-mail: (

Biographical Note: Ikhyun Jang (Corresponding Author) is an early career researcher in social policy, with research interests in ‘income distribution’, ‘welfare reform’, ‘comparative social policy’. He awarded a Ph.D. Degree on distributive outcome of new welfare policies from Department of Social Policy and Social Work, University of York. He now works as associate research fellow for the Seoul Insitute, South Korea. E-mail: (

Keywords: chhaupadi, menstruation, policy non-compliance, Nepal.