CHAPTER 14: What generative mechanisms excluded indigenous people from social health protection? A study of RSBY in Karnataka Gayatri Ganesh, Tanya Seshadri, Anil MH, Mahesh Kadammanavar, Maya Elias, Philipa Mladovsky, Werner Soors Introduction The status of tribal communities in India Indigenous people, the world over, suffer violations of human rights and experience deprivations in basic human necessities, including health. According to the recent census, 8.6% of the Indian population (104,545,716 individuals) are tribal people (Census of India, 2011). Following independence, the Indian Constitution granted special provisions to the indigenous people classified as Scheduled Tribes (ST), of which reservation of seats in higher education, public employment and legal representation have since been a constant (Louis, 2003; Xaxa, 2001). Protective measures (Protection against Atrocities Act, 1989) and participation in planning for development (Provisions of the Panchayats (Extension to Scheduled Areas) Act 1997) – also mandated by the Constitution – are much less developed (Heredia, 2011; Louis, 2003; Xaxa, 2001). Studies have shown that the implementation of these measures has been weak and that the indigenous population have not been made aware of the rights accorded to them (Rout & Patnaik, 2013). Despite over six decades of affirmative policies, India’s ST communities still face hardship and inequities. For instance, the proportion of tribal people living Below the Poverty Line (BPL) is nearly twice as high as among the rest of the population (43.8% vs. 22.7%) (Planning Commission, 2007). Although some improvements have been made, literacy levels for the tribal population is abysmally low (47%) compared with the rest of the population (67%) (World Bank, 2011). These failures in substantial improvement for tribal communities has arisen out of low fund allocation for tribal development, violation of protective land measures and displacement from land/forests and livelihoods due to forest clearance policies, mining, construction of dams, fraudulent land transfers, forcible evictions from areas designated as national parks and other abuses of their land rights, and often by the federal government itself (Xaxa, 2012). Essentially, tribal people have been displaced from their lands in the name of development but have been denied the fruits of that development with inadequate compensation in land or restoring their livelihoods. For example, two states, Orissa and Jharkhand, have experienced rapid infrastructure development but have the largest percentage of tribal population living below the poverty line. Tribal people also face discrimination from larger society because of their societal structure, cultural practices, religious beliefs, and language (Xaxa, 2005). Overt poverty, displacement, difficulties adapting their lifestyle to a new environment, inability to enter forests to collect food and medical resources, discrimination, and deficient provision of health services all add to their predicament (Sundarajan et al., 2013; Ministry of Tribal Affairs, 2004). Health of tribal communities in India Few ST-targeted public health interventions have been introduced to complement the reservation policies (Mohindra & Labonté, 2010). ST health indicators are abysmal, with higher rates of mortality (for all ages), malnutrition, anaemia, malaria and tuberculosis than the rest of the population (Mohindra & Labonté, 2010; Subramanian, Smith & Subramanyam, 2006). The under-five child mortality of tribal children is particularly stark with nearly 96 deaths for every 1000 live births, well above the national average of 74 deaths per 1000 live births; around 53% of tribal children are stunted (lower height-for-age) (NFH3-3, 2006). However, the health and economic status of tribal communities vary considerably by where they live in India: tribal communities in rural areas of the states of Orissa, Chhattisgarh, Jharkhand, Maharashtra and Rajasthan have seen far lower declines in poverty relative to other groups; incidentally, the World Bank notes a high correlation between poverty and concentration of tribal populations (World Bank 2011). Tribal communities in Karnataka In the southern state of Karnataka, tribal people constitute about 7% of the population (4,248,987 people) (Census of India, 2011). The ST literacy rate is considerably lower than the average literacy rate in the state (53.9% vs. 75.6%) (Census of India, 2011). According to the 2005 Karnataka Human Development Report, “The human development status of the Scheduled Tribes is more than a decade behind the rest of the population of the state and they are the poorest and most deprived of all sub-populations in the state”. Decades of alarming health indicators have seen little or no improvement. While the proportion of institutional deliveries among ST women in Karnataka increased from 26.8% in 1993 to 41.5 % in 2005-06, the proportion of pregnant tribal women not receiving antenatal care has seen only slight improvement, from from 21.4 in 1993 to 17.8 in 2005 (Planning and Statistics Department, 2006; NFHS 3, 2006). In Mysore district – home to the state’s historic and touristic capital and a study district – ST women are still twice as unlikely to have an institutional delivery than non-ST women (Adamson et al., 2012). Considering the state programmes targeting the ST population (electricity, free housing, drinking water supply, midday school meals and free text books and school uniforms), the 2005 Karnataka Human Development Report concludes that while some programmes have been successful, most suffer from poor implementation and low effectiveness – particularly in the areas of poverty reduction, education and health (Planning and Statistics Department, 2006). The report also states that state functionaries have given insufficient attention to this vulnerable group and resources allocated for tribal development are under utilised. Skilled practitioners are unavailable to tribal people and a combination of distance from public services and neglect by the state in ensuring access to health services has contributed to the poor health indicators of tribal people. In the absence of a health programme specifically targeting tribal people, studying the implementation of social health protection (SHP) schemes in tribal areas provides an opportunity to understand if and how tribal people are excluded despite being eligible according to the programme design. In our study, we focused on a social health protection scheme, Rashtriya Swasthya Bima Yojana (RSBY), instituted by the Indian federal government in 2008, targeting the BPL population and described in detail in Chapter 6. While some studies have examined rates of awareness, enrolment and utilisation among different social categories including ST (Nandi et al., 2013; Devadasan et al, 2013) in RSBY, there is a paucity of studies exploring RSBY or any similar SHP scheme through the lens of social exclusion to explore who gets excluded from the scheme and how and why this occurs. We attempt at filling this gap by unfolding the process of social exclusion – as manifest in experiences and events – through its underlying generative mechanisms (Demetriou, 2009; Bunge, 1997; Tilly, 2001; Marchal et al., 2010). Given historical reasons for the apparent social exclusion of tribal people in India, we aim to examine if rural ST communities are also excluded from RSBY; and if so, to understand the mechanisms of their exclusion. Methods A full description of the study approach can be found in chapter 6. In this chapter, we focus on the experience of tribal households (n=553, 9.2% of total) across all four sites/districts in engaging with RSBY. A mixed-method approach was used for the overall study comprising of household surveys along with focus group discussions and in-depth interviews with community and different stakeholders. It is important to note that the household surveys were designed to provide estimates of the general population and not any specific group, like the tribal population. While the proportion of tribal households identified in the study is similar to the state proportion, the number of households captured was relatively small in size and not representative of the tribal population in the state. Hence, while the quantitative findings do provide valuable insights about the RSBY experience of tribal households in relation to the rest of the study households, we are cautious about generalising findings from these households. Furthermore, when surveyed on aspects such as ‘participating in local political activities’, what constituted those activities was not specified and only later explored in the qualitative approach. At every stage, we explore the experience of the tribal households in relation to the rest of the study households. We also chose multivariate analysis (logistic regression) to study the interaction between key variables both within and across SPEC dimensions and their influence on the odds of enrolling into the scheme (see chapter 8). In this chapter, we take the analysis forward by conducting multivariate analysis within the ST households to explore the influence of key SPEC variables on a household’s chances of enrolment within this community (described later). Eight of the focus group discussions were conducted specifically with tribal communities in two districts with a relatively higher proportion of tribal population, namely Mysore and Belgaum. These aimed at capturing information related to health seeking behaviour, access to RSBY and other welfare schemes and their experiences of exclusion. Findings from all sources were triangulated with each other to present a comprehensive picture of the tribal households’ experience in accessing RSBY in our study. Regression model The simple logistic model used for estimating the determinants of enrolment in RSBY among the ST households was defined by the following relationship (adapted from Parmar et al., 2014): Enrolledi = β0 + βi1Xi + βi2SEi + Єi Enrolledi is a binary variable that denotes the enrolment status of a household; Xi is a set of general variables; SEi is a set of key variables selected across social, political, economic and cultural (SPEC) dimensions; ЄI is the random error. The different variables included in regression analysis are shown in Table 2. The dependent variable, Enrolled, is a binary variable that indicates the enrolment status in RSBY of a tribal household in our study. A household is considered enrolled only if they enrolled in the enrolment camp conducted in their village/area. General variables (XI set): Since we looked for characteristics of the head of household that could determine its odds of being enrolled, we used gender, age, literacy status, and income dependence (coding explained in Table 2). The household size of the eligible household was also included, as only smaller households (5 or less members) can be fully covered under RSBY while larger households will always be partly covered given the design. SPEC variables: Following preliminary analysis, we identified key variables across social, political, econominc and cultural (SPEC) dimensions guided by the SPEC framework and literature review. A few variables that were found to be significant at the general population level were skewed or made irrelevant when focusing within the tribal households. For instance, all ST households were Hindu and nearly all of them were native-Kannada speakers (Table 1). The variables included in Model B (chapter 8) were adapted to be relevant to this community. Migration for work captured households where any member needed to migrate out for work. This is important due to the higher proportions of casual labourers in this community and, as such, households are explicitly targeted by the scheme. Food security reflects if a household reported having adequate food for at least 3 meals a day throughout the past year. Given the higher proportions of relatively economic poor households in this community (Table 2), this variable was included to supplement the wealth index. Forced to vote included any household where a member was reported to have been forced to vote in the recent elections. Results We present our results in three main sections. We first describe and compare the profile of tribal households in the study with non-tribal households. We then explore the tribal households’ experience in accessing RSBY in terms of receiving information and enrolling in the scheme in relation to other study households, and then amongst the tribal households. In the end, we attempt to describe possible exclusionary processes in society due to which tribal households are excluded from accessing the schemes. Profile of the tribal households Tribal households constitute 9% (n=553) of our study households. More than half of these households belong to Mysore district, constituting 20% of total households from the district. Tribal households were largely Hindu, native-Kannada speaking, nuclear families with a median household size of 5 (Table 1). The adult literacy rate was significantly lower than for the rest of the study population, with one in two adults never having gone to school. The proportion of adults with an education above primary school was one-third that of the non-tribal population with nearly half of all tribal adults engaged in casual labour. The socio-economic disadvantage is further reflected by the proportion of households with the ‘poorest of the poor’ BPL ration card (Antyodaya or poorest of the poor card, a card distributed by the government to households that received the lowest scores among the BPL census to avail food subsidies) among tribal households, which is twice that of the non-tribal households (Table 1). Table 1. General profile of tribal and non-tribal households Characteristics Total number of households Total population included District Bangalore rural Belgaum Mysore Shimoga Religion Hindu Muslim Christian Others Median household size Type of households Single Nuclear Joint/Extended Age group Less than 18 years 18 – 59 years 60 years & above Sex ratio (No. of females per 1000 males) Adult literacy rate* Education* Never Went to School (18 years & above) Occupation* Upto primary school Upto high school Above high school Not earning (18 years & above) Casual wage labourer Self employed Salaried Kannada speaking households Household has a ration card Type of ration card ‘Poorest of poor’ BPL Card BPL Card APL Card Tribal households Frequency Per cent (%) 553 2879 17 3.1 185 33.5 305 55.2 46 8.3 553 100.0 0 0 0 5 (1-23) 12 2.2 262 47.4 279 50.5 877 30.5 Non-tribal households Frequency Per cent (%) 5487 30238 482 8.8 2833 51.6 1199 21.9 973 17.7 4934 89.9 431 7.9 17 0.3 105 1.9 5 (1-30) 92 1.7 2379 43.4 3016 55.0 9,190 30.4 1,712 59.5 17,803 58.9 290 952 10.1 3,245 969 10.7 1,010 44.8 50.4 7,487 59.0 35.6 520 141 200 737 26.0 7.0 10.0 36.8 3,126 6,967 3,468 9,325 14.9 33.1 16.5 44.3 924 271 70 489 533 86 421 22 46.2 13.5 3.5 88.4 96.4 15.6 76.1 4.0 5,929 4,692 1,102 3923 5268 441 3974 810 28.2 22.3 5.2 71.5 96.0 8.0 72.4 14.8 * 18 years & above, n = 2,002 people among tribal households, n= 21,048 people among non-tribal households Experience with accessing RSBY Tribal households reported poor awareness with one in two households having never heard about the scheme or seen the RSBY card, similar to other households. However, the enrolment rate and the proportion that finally received the cards were significantly lower among tribal households than their counterparts (Figure 1). Similar to other households, the main sources of information about the scheme for the aware tribal households were Gram Panchayat (GP) members (36% v 41%), and local health workers (15% v 18%). Finding out about the scheme from friends/relatives was higher for the tribal households than others (34% v 20%). Figure 1. RSBY experience of tribal and non-tribal households (n = 553, 5487) While the proportions of enrolment and receiving the smart card among tribal households was significantly lower than the others, the experience at the camp was better in respect of having their thumbprints and photographs taken (98% v 91%), paying the correct amount of Rs. 30 at the camp (85% v 73%) and receiving the hospital information booklet (11% v 11%). Even the proportion of cardholding households that were fully covered (up to 5 members for large households) was higher among tribal households when compared to others (57% v 40%). While reviewing the interaction between RSBY and ST households, it becomes clear that the bulk of ST households are excluded at one particular step: enrolment (Figure 1). While other social categories including Scheduled Castes have an enrolment rate of 40%, only 30% per cent of ST households pass enrolment – the lowest score among all social categories. Approximately half (47%) of ST households that did not enrol reported that they were unaware of the enrolment camp venue and timings. In chapter 8, we identified key determinants for enrolment in to RSBY for the study households. Social category appears to be a key determinant to predict a household’s odds for enrolment. Among the different categories, the odds of an ST household being enrolled was 40% less than those of the other backward class (OBC) category while independent of other SPEC variables. We next explore various SPEC variables within the tribal households to try and understand why some households were able to be enrolled while some did not. A description of the different variables included in the analysis is given in Table 2. It is important to note here that 71% of the head of tribal households were illiterate, with 77% engaged in generating income (mostly casual labour) (Table 2). Around two-thirds of the tribal households reported to have members associated with some local social organisation, like women’s groups, youth groups, or self-help groups. It is also important to point out that 52% of the tribal households belonged to the lowest wealth index quintiles, reflecting the higher relative economic poverty. Table 2. Definition and description of variables used Variables Definition Enrolled General Head of household 1 if enrolled, 0 otherwise Tribal households Frequency Per cent 165 30% Female 1 if female, 0 otherwise 94 17% Elderly Illiterate Income dependent 1 if elderly (age 60 years and above), 0 otherwise 1 if illiterate, 0 otherwise 1 if not an earning member, 0 otherwise 139 393 125 25% 71% 23% 1 if the household has more than 5 members, 0 otherwise 180 33% 1 if no household member is part of any local social organisation, 0 otherwise 1 if no socio-political contacts, 0 otherwise 1 if no household member participates in local political activities, 0 otherwise 1 if distance from nearest PHC >5 km, 0 otherwise 223 40% 353 449 64% 81% 171 31% 62 59 11% 11% 22 4% 70 13% 81 103 124 155 15% 19% 23% 29% Large household SPEC Social participation Socio-political contacts Political participation Access to health services Migration for work Food security Forced to vote Wealth index 1 if migrant in household, 0 otherwise 1 if household did not have enough food for 3 meals a day in the past year, 0 if otherwise 1 if any household member was forced to vote in the elections for a given candidate, 0 if otherwise Q 1 to Q 5; Q 1 refers to the wealthiest 20% households Q1 while Q 5 refers to the poorest 20% households. Q2 Q3 Q4 Q5 Acronyms used: Q – Quintile, PHC – Primary Health Centre, SPEC – Social, Political, Economic and Cultural Within tribal households, having an elderly-headed household and belonging to a small household increased the odds of the household being enrolled, similar to a pattern seen overall in chapter 8. When studying why tribal households fare poorly when compared to non-tribal households, socio-political exclusionary processes were clearly identified and outlined in chapter 8. However, when looking within the community of tribal households, it is the relative economic poverty that appears to decrease the odds of getting into the scheme. The socio-political processes hence appear to mainly operate at the societal level between tribal and non-tribal households. Table 3. Determinants of enrolment in RSBY within tribal households (n=553) Key variables General Head of household Female Elderly Illiterate Income dependent Large household SPEC Social participation Socio-political contacts Political participation Access to health services Migration for work Food security Forced to vote Wealth index Q2 Q3 Q4 Q5 No of observations OR SE 0.616 0.513 0.694 1.634 0.496 (0.295) (0.339)** (0.234) (0.350) (0.228)*** 0.780 1.340 0.781 0.861 1.102 1.691 0.915 0.599 0.412 0.355 0.507 5827 (0.207) (0.226) (0.275) (0.219) (0.330) (0.354) (0.514) (0.352) (0.342)** (0.332)*** (0.316)** Acronyms used: Q – Quintile, SPEC – Social, Political, Economic and Cultural Robust SE in parenthesis, *p<0.1, **p<0.05, ***p<0.01 Mechanisms of exclusion Description of ST households’ characteristics and the incidence of their exclusion from RSBY are of little value without knowledge on how and why exclusion actually occurs in a particular context. Building upon a critical realist rationale (Demetriou, 2009; Bunge, 1997; Tilly, 2001; Marchal, 2010), in the present section we explore the process of social exclusion as the interplay between individuals and institutions, the interaction between social structure and agency. Concretely, we move from description to the identification of generative mechanisms – which we will then resume in the discussion section – applying the iterative abstraction known as retroduction (Astbury & Leeuw, 2010). In the FGDs, tribal respondents reported that information about welfare schemes does not reach them directly. A few people who are literate or semi-literate get to know about welfare schemes (sometimes through NGOs) and they spread the information among the rest of the settlement (usually through the tribal leader). For some elders, such information appears to be accessible through the younger generations due to their frequent travels outside the settlement for education or work: Directly they [local government functionaries] will not come tell us anything. They circulate the information among themselves only. Three or four of our literate people they will come to know, maybe through organisations like yours [NGOs] and they will inform the tribal leader...Our children will go mix with others in the village and they will tell us about the different schemes going on. (FGD, Tribal member, Mysore) Lack of political networks By far the most pronounced form of exclusion was exclusion from information about welfare schemes by Gram Panchayat members. In the case of RSBY, it is the insurance company’s responsibility to conduct information and education campaigns. They in turn rely on local Gram Panchayat members and health workers to identify eligible households, at times issue them a token for participating in the enrolment camp and to inform them about the scheme and camp timings. RSBY actors state that since these local players are better informed and able to take on this role compared to external organisers, they are given the responsibility at their village level. FGDs with tribal beneficiaries revealed repeated accounts of Gram Panchayat members providing information (on RSBY and other welfare schemes) only to their relatives, friends and others who favour them politically or contribute to their campaigns. ST respondents’ perception of being “useful” to a political leader ranged from being a patron of a local politician to being hired for political rallies and odd jobs. In addition, ST respondents also state that the requirements of daily subsistence gave them no time to “run after” politicians to do their bidding. They also believed that the lack of clout and inability to contribute to campaign funds by tribal communities lead to their overall neglect by politicians of any kind: The government will introduce some schemes and that will reach the panchayat, what he will do is only give the information to people whom he wants, but not for us poor people...If you are good to the leaders, running behind them, doing their work, then they will do your work [processing welfare applications or giving information about schemes]. But I am a daily wage [worker], I am of no use to him and I have to work everyday or my family will have no food. (FGD, Tribal member, Mysore) Gram Panchayat members admitted to giving information to their friends and others who frequently visited the GP office more often because it was convenient. Some reported that the time provided to them by the insurance companies and third-party administrators (TPA) was too short (a day to a week) to inform all those eligible, and to organise an enrolment camp. Faced with this short timeline, they then resorted to at least gathering some beneficiaries on the list. Political neglect Tribal respondents reported that Gram Panchayat members seldom visited their settlements, only doing so during election time. In times of natural disasters like forest fire or heavy rains, the local politicians are conspicuous by their absence. Therefore, they tend to believe that if a welfare scheme is targeted explicitly for tribal communities, then they will more likely see some benefits. ST respondents also expressed trust in NGOs working for the organisation of self-help groups and specifically for development of tribal communities, who they believe are genuinely concerned for the wellbeing of tribal communities. In contrast, they perceive local government officials to be “middlemen”, more focused on amassing profits and exploiting forest resources than helping tribal communities: Schemes do not directly reach us. Always the government people, panchayat people, forest officials are there to take their cut from us. If we sell an ox, they will take 70% of what we get for it. Only the NGOs… those who help us with selling forest-made items, they should be given the money to give to us directly. (FGD, Tribal member, Belgaum) Neglect of ST communities seems to impact also at a higher level of programme implementation. During the enrolment period prior to our research, the tribal settlements around HD Kote in Mysore district and Khanapur in Belgaum district, and sharing the worst social indicators of the state – were the last to be enrolled, and enrolment could not be completed before the policy period took off. When the enrolment camps were prematurely shut down, enrolment rates were a mere 9% and 15% in HD Kote and Khanapur respectively. The following year, it was decided to start enrolment not where it had left off but in the better performing districts, leaving the ST populations in wait. Lack of political voice Interviews with tribal representatives, like women self-help group leaders and former tribal leaders, revealed their helplessness in mobilising other leaders among the tribal community to come together for protests to the district administrative headquarters, even to demand for essential public services. They acknowledged that other vulnerable social categories, particularly SC, were much more able to voice their demands, due to better political mobilisation. In the opinion of the interviewed representatives, most tribal people were unable to see the world beyond their settlement and were ignorant of their entitlements. On the other hand, several ST respondents explained that their elected representatives to the Gram Panchayat were more often than not “weak” and “ineffective”, with a lack of focus on the legal ST entitlements, and unable able to question administrators and policymakers for the flawed implementation of existing welfare schemes. In general, they saw their representatives as excluded themselves from local political clout. They don’t tell him [representative] anything. They don’t give him any information. They don’t call him also for the meeting. What can he do if they don’t allow him to find out anything for his people? (Interview, ST woman’s self help group leader, Mysore) For those few beneficiaries who do make it to the village council and request information, the perception is that they are often ignored or treated without respect. Instances of verbal abuse by Gram Panchayat members referring to the beneficiaries’ “poor physical appearance” and “illiteracy” were reported. ST respondents’ state that they are often made to wait long hours, told to come back later, or that the concerned person is unavailable and are ultimately prevented from meeting a higher official. They point out that it can take all day to get information, and again they emphasise that for daily wageworkers this is problematic: They will not allow us to go into the office and ask [information]… They say, “You people stay outside”. If we want to see a higher official they will say things like “How can you see [him], look at your clothes!” or “He is not here, you have to wait”. We will be waiting whole day but they will not let us see anyone. (FGD, Tribal member, Belgaum) Low literacy and education Literacy rates and educational levels are low in tribal communities, and even lower among the women (see Table 1 and Table 2). As a series of communications on RSBY are in written form – benefits, eligibility, camp schedule, and list of empaneled hospitals–it is tempting to consider lack of literacy and education as a mechanism of (involuntary or passive) exclusion: Uneducated people, they do not know to ask anything. They just come when we tell them to come; they take the photo and go. They do not know why they came also. We also don’t tell the information because they are not asking. They don’t know they have to ask. (Interview, GP, Mysore) However, low literacy rates and educational levels can also be interpreted as outcomes of a long-lasting exclusionary process. In tribal tradition, oral culture has always been more important than written culture, as recognised by the Ministry of Tribal Affairs (2013) and evidenced in other literature (Bhukiya, 2010). While both government and tribal people express a need for improved education, the near absence of efforts to bring this about is an exclusionary process in itself. The fact that ST respondents frequently testify to being blamed for being illiterate reinforces the latter interpretation. Spatial and social isolation Mainstream literature describes spatial isolation, or distance, as an important explanation of involuntary exclusion of ST communities (The World Bank, 2011). While, historically, this is largely true and still the case in Central India’s tribal belt, it is far from always the case in South India. In Karnataka, most ST households’ cluster in tribal settlements often located at the edge of a larger village or in relocated villages outside their original forest environment. Isolation then becomes much more social than spatial, and distance is not the real issue (Thorat, et al. 2007). A critical incident during our data collection can illustrate this: while conducting FGDs in HD Kote, Mysore district, we were confronted with three villages within sight of each other and at equal distance from the main road to which they were equally well connected. In village one and three, the TPA had come by to inform them about RSBY and enrol them, but not so in village two. What distinguished village two was that is was entirely tribal. ST respondents confirm that camp organisers usually aim at the Gram Panchayat headquarters for conducting camps to ensure maximum participation in terms of numbers, making no extra efforts to reach out to tribal settlements. Gram Panchayat members, in turn, are perceived to ignore tribal settlements and villages for day-to-day matters and to only acknowledge their existence and issues for election campaigns. Discussion According to the World Health Organisation’s Social Exclusion Knowledge Network (SEKN, 2012), “(Social) exclusion consists of dynamic, multi-dimensional processes driven by unequal power relationships. These operate along and interact across four dimensions – cultural, economic, political and social – and at different levels including individuals, groups, households, communities, countries and global regions. Exclusionary processes contribute to health inequalities by creating a continuum of inclusion/exclusion. This continuum is characterised by an unjust distribution of resources and unequal capabilities and rights”. In this paper, we examined whether tribal households face social exclusion from RSBY, and if so, how and why. Our results show that tribal communities have lower rates of awareness and enrolment in RSBY when compared with non-tribal households in the areas studied. We identified that the process of exclusion includes the following – possibly overlapping – generative mechanisms: lack of political networks, political neglect, lack of a political voice, low literacy and education, and socio-spatial isolation. Using T.H Marshall’s (1977) classic work, these mechanisms can be grouped under one umbrella, i.e. be interpreted as a denial of citizenship. Marshall defines citizenship as, “...a status bestowed on those who are full members of a community. All who possess the status are equal with respect to the rights and duties with which the status is endowed”. As he traces the evolution of civil rights (right to justice, rights to property and free speech), political rights (voting freely in fair elections) and social rights (equal membership in a community), it is clear that citizenship is not a given fact but a slow and arduous journey (Betteille, 1999). While the ST community of India may have been conferred their legal or political rights as citizens by the state (and certain protections in the Constitution), their civil rights are often trampled on, and their social rights as equal members of a society are far from realised in everyday living. While citizenship demands equality before the law and equity in society, the relationships of tribal communities with mainstream society are characterised by inequity. Being denied citizenship is to have civil rights without the power to enforce them (as when tribal households are excluded from information about welfare), political rights without political effectiveness (when political representatives are absent or weak) and social rights without recognition from the rest of society of their equal worth. We have looked at social exclusion as a complex process for which we identified three sub-mechanisms, which we hypothesise are encompassed in a larger common mechanism, i.e. denial of full citizenship which, in this case prevents tribal households from benefiting from social health financing schemes. Nathan and Xaxa (2012), commenting on the exclusions of tribal communities in India, state that “Exclusions are of two forms. One is exclusion from access to or denial of rights to various services, such as health, education, housing, and water, with sanitation also being more recently included as an essential service. The other form of exclusion is that of deprivation of the right to express one’s views, of representation and voice in terms made famous by Hirschman. These two forms of exclusion often go together, with lack of representation and voice being manifested in inadequate provision of services”. It is precisely this ‘going together’ which we refer to using the term ‘denied citizenship’ and have evidenced in this study. We found that a vital conduit to gain access to information about welfare programmes, including RSBY, is the local political network. This echoes the wider sociological literature that finds that those with political power control information (as a resource) and divert it towards restricted groups that they prefer to be in the know (Murphy 1988). Exclusion occurs through ‘social closure’ (Weber, 1978), defined as monopolisation of resources by individuals and groups using rules of exclusion to dominate others in society, which in turn propels social inequalities. ST households lack an entry point into these political networks, as they are usually poor and are not considered to have much leverage in the wider community (status) that a local politician could gain from. The institutional bias against tribal communities manifests in the larger neglect of tribal welfare and the disrespect (such as verbal abuse, failure to answer queries, delaying applications for welfare schemes, criticism of physical appearance and shunned from meeting with public officers) reported at the local government structures. This denial of recognition of tribal people as equal members of society reduces their life chances – a term we borrow from Max Weber (1978) meaning the probability an individual has to improve his or her quality of life through access to important social resources, including healthcare. The chance of access to information about welfare schemes for tribal households is under the influence of the inequitable power relationships that dominate everyday interactions with authorities, one that has persisted for generations. Our results suggest that this bias extends to the tribal representative in local government, who is sidelined by the dominant members, making their representation ineffective and muting the tribal voice. While the concept of social closure has been used in the sociology of organisations (and professions), the political exclusion of tribal communities could be built in to understand how institutional bias is constructed and maintained through everyday interactions (Rosigno et al., 2007) between tribal people and local government. The world inhabited by the tribal households we studied in Karnataka was not the isolated forest dwelling abode that the life of tribal people is often portrayed to be in India. Many tribes have been forced out of the forest to resettle near or in villages with limited access to the forests. In our study, 92% of the tribal households had permanent houses (most constructed via government welfare schemes), 82% had electricity and 79% had a drinking water facility nearby (Table 1), figures similar to the rest of the rural study population. While the literature on tribal communities (Planning Commission, 2007; Stephans et al., 2005; Ministry of Tribal Affairs, 2004; Betteille, 1991) talks of the geographical isolation of tribal community as a determinant for their lack of access to resources, more and more tribal communities are being shifted out of forests and are relocated outside. We argue that for the latter: their lack of access arises not from geographical remoteness per se but from social isolation due to differences in social and cultural customs from larger society (Xaxa, 2001) and thus excluded even when at arms length. Regarding tribal settlements in remote areas or at the edges of larger villages, Nathan and Xaxa (2012) comment that the lack of infrastructure (be it roads, schools, or primary health centres) is not only due to higher cost but also due to persistent marginalisation: “remoteness (…) is not just a matter of geography”. Instead of tailoring welfare programmes to the particularities of tribal communities and targeting them specifically, the common strategy has been to consider ST communities as not having specific needs when entering modern society. According to Xaxa (2005), contact and exposure to modern society by some tribal communities has not necessarily benefitted them more than those with minimal exposure. For the implementers of RSBY, a social exclusion lens unveils certain challenges. The implementation of the scheme is bound up today with these existing mechanisms that exclude tribal communities. For instance delegating responsibility to create awareness and organise enrolment camps for RSBY in a village to Gram Panchayat members and health workers is one way of involving the community; however, lack of sensitisation of those involved in reaching out to the vulnerable allows existing perceptions and exclusionary processes to continue unchallenged. In terms of overall numbers, this may go largely undetected until the policy maker asks the question specifically in relation to tribal communities and then reflects on processes to offset the control of information by local political networks, like bringing in additional focused monitoring systems, incentives to include tribal settlements, etc. Our study points to an urgent need for implementers of RSBY to reflect on the exclusion faced by tribal households and develop such processes. In terms of overall numbers, their scattered settlements and increase in effort in terms of resources – financial, manpower and time – to reach out to them, tribal households are largely ‘unattractive’ in terms of the business model of implementation of RSBY (the government pays premiums per household enrolled to the insurance companies thereby making the companies responsible for creating awareness and enrolling households). However, it is of great significance given the social mandate of RSBY, and hence, the implementers need to bring in processes that negotiate between the mandate and the scheme’s model of implementation. As Kabeer (2000) points out: “The rationale for social policy lies in the recognition that neither individual need nor the collective good can be left solely to private initiative and that there is a case for purposive public action to be taken”. These misaligned priorities (between a business and social model) can pose a moral hazard. Frazer (1989) and Gore (1993) discuss exclusion occurring through ‘unruly practices’ where, despite institutional rules, there is a gap in implementing them as intended. According to Kabeer (2000), there are likely to be unofficial norms that shape the actual provision of goods and resources to which groups are officially entitled. She stipulates that ‘unruly practices’ are more likely to occur in the public sector because the public provisioning is meant to deliver/cater to social need and curtail exclusions in communities. As a result, ‘unruly practices’ are more likely to occur because the rules clearly instruct otherwise. As the private sector is concerned with the business end of the scheme, discrimination is likely to occur only if it interferes with this pursuit as it seems to do in the case of RSBY and tribal communities. With stricter regulation by the state, positive incentives to motivate or penalties for not enrolling tribal populations should be considered for insurance companies; and with a clearer focus on groups that face social exclusion, the state might begin to meet its goals of health protection for the most vulnerable. Currently, ST households (as well as others) have no forum to air their grievances about RSBY. The state needs to provide a transparent avenue for them to register their complaints and get information about their entitlements. One way forward would be to enlist trusted civil society or non-governmental organisations as independent regulators to ensure that information and welfare schemes reach the tribal people, building an effective partnership between the state and civil society without the state taking a back seat. Limitations: Our study is not free of limitations. Firstly, our study did not cover the Chamarajanagar area of Karnataka where the largest tribal population resides, mainly within hilly forest reserves, hence limiting the generalisability of our findings to the tribal population across the state of Karnataka. Secondly, tribal communities were not asked prior to the survey who they thought were important socio-political contacts or what they understood as political participation (Table 2). These perceptions were more clearly examined during the qualitative discussions which followed the survey. Thirdly, since the findings of this paper are part of a larger study on social exclusion in access to health services vis-à-vis health-financing scheme, the study as a whole did not focus on the particular exclusion of tribal households. Therefore, interviews with implementers are not specific to the tribal condition alone. Fourth, some processes of exclusion identified were also seen to affect other vulnerable groups like scheduled caste households, woman-headed households, casual wage labourers, etc. to different degrees, influenced by the particular relational exclusions those groups face. Further research is required to explore the validity of the discovered mechanism within these groups. Conclusion Our findings suggest that ST households face exclusion from awareness about, and enrolment into, RSBY when compared with the rest of the population. By using local political networks to spread information about RSBY, tribal households are affected by the unequal power relations that govern their interactions with the local authority. The mechanism of their exclusion is what we call ‘denied citizenship’: a combination of a lack of political networks, a lack of a political voice in the existing climate of political neglect, cultural discrimination and social-spatial isolation. This study is the first of its kind to document the experience of tribal households in accessing a health-financing scheme in India and to explore possible mechanisms of their exclusion. Our study has relevance for policymakers and implementers of RSBY and similar welfare schemes that need to recognise that social inequities deny tribal communities access to the schemes that they are entitled to. Also, the policy implications have relevance to implementers outside of India to consider a social exclusion lens, the particular exclusions and discriminations that sections of the population might face, anticipate practical impediments and the gaps between policy or scheme guidelines and the ways in which they are implemented in reality. References Adamson, P.C., Krupp, K., Niranjankumar, B., Freeman, H.A., Khan, M., & Madhivanan, P. (2012). Are marginalised women being left behind? A population-based study of institutional deliveries in Karnataka, India. BMC Public Health 12(3). doi:10.1186/1471-2458-12-30. Accessed from: http:// www.biomedcentral.com/1471-2458/12/30. Astbury, B., & Leeuw, F.W. (2010). Unpacking black boxes: Mechanisms and theory building in Evaluation. American Journal of Evaluation 31(3): 363-38. Accessed from: http://www.itg.be/internet/ds/tde/doc/ Astbury%20%26%20Leeuw%20.pdf. Beteille, A. (1999). Civilization, State and civil society. Economic and Political Weekly 34(36). Accessed from: http://www.epw.in. Béteille, A. (1991). Society and Politics in India: Essays in a Comparative Perspective. London School of Economics Monographs on Social Anthropology. New Delhi: Oxford University Press. Bhukiya, B. (2010). Adivasis and land assertion in Andhra agency. The IUP Journal of History and Culture 4(4): 60-72. Bunge, M. (1997). Mechanism and explanation. Philosophy of the Social Sciences 27(4): 410-465. Accessed from: http://goo.gl/IVHs37. Collins, K.M.T., Onwuegbuzie, A.J. & Sutton, I.L. (2006). A model incorporating the rational and purpose for conducting mixed-methods research in special education and beyond. Learning Disabilities: A Contemporary Journal 491: 67-100. Accessed from: http://www.mmr.sagepub.com/content/41/56. abstract. Creswell, J. W., Plano Clark, V. L., Gutmann, M. L., & Hanson, W. E. (2003). Advanced mixed methods research designs. In A.Tashakkori & C.Teddlie (Eds.), Handbook of mixed methods in social and behavioral research. Thousand Oaks, CA: SAGE. Demetriou, C. (2009). The realist approach to explanatory mechanisms in social science: more than a heuristic? Philosophy of the Social Sciences 39(3): 440-462. Accessed from: http://goo.gl/qtbT3p. Dasgupta, R., Nandi, S., Kanungo, K., Nundy, M., Murugan, G., & Noeg, R. (2013). What the good doctor said: a critical examination of design issues of the RSBY through provider perspectives in Chhattisgarh, India. Social Change 43(227). Das, J., & Leino, J. (2008). Evaluating the RSBY: Lessons from an experimental information campaign, 85–93. Economic and Political Weekly 41(32). Accessed from http://www.epw.in. Devadasan, N., Seshadri, T., Thrivedi, M., & Criel, B. (2013). Promoting universal financial protection: Evidence from the Rashtriya Swasthya Bima Yojana (RSBY) in Gujarat, India. Health Research Policy and Systems 11:29. Fraser, N. (1989) Unruly Practices: Power, Discourse and Gender in Contemporary Social Theory, Cambridge: Polity Press. Gore, C. (1993). Entitlement relations and ‘unruly’ social practices: a comment on the work of Amartya Sen. Journal of Development Studies 29(3): 429-460. Heredia, R. (2011). Quotas and minority rights: recapturing the constitutional vision. Economic and Political Weekly 46(30). Accessed from: http://www.epw.in. Hou, X., & Palacios, R. (2011). Hospitalization patterns in RSBY: preliminary evidence from MIS. In Robert Palacios, Jishnu Das & Changqing Sun (Eds), India’s health insurance scheme for the poor: Evidence from the early experience of the Rashtriya Swasthya Bima Yojana (Chapter 6). New Delhi: Centre for Policy Research. International Institute for Population Sciences. (2006). National Family Health Survey NFHS 3, Karnataka report (2005-2006). Mumbai: International Institute for Population Sciences. Kabeer, N. (2000). Social Exclusion, Poverty and Discrimination: Towards An Analytical Framework. IDS Bulletin 31(4). Louis, P. (2003). Scheduled Castes and Scheduled Tribes: the reservation debate. Economic and Political Weekly 38 (25). Accessed from: http://www.epw.in Marchal, B., Van Belle, S., & Kegels, G. (2010). Theory-driven inquiry for health systems research. Accessed from: http://www.itg.be/internet/ds/tde/index.html Marshall, T. H. (1950). Citizenship and Social Class. Vol 11 Cambridge. Miles, M.B., & Huberman, A.M. (1994). Qualitative data analysis: An extended sourcebook. Thousand Oaks, CA: SAGE publications. Ministry of Tribal Affairs (2004). Issues relating to Scheduled Tribes 37th Annual report. New Delhi. Retrieved from: http://www.tribal.nic.in Ministry of Tribal affairs (2013) National Tribal Policy (Draft). New Delhi. Accessed from: http://www. legalpundits.com/Content_folder/tribalpolicy.pdf. Mohindra, K.S., & Labonté, R. (2010). A systematic review of public health interventions and Scheduled Tribes. BMC Public Health 10(438). Accessed from: http://www.biomedcentral.com/1471-2458/10/438. Murphy, R. (1988). Social Closure: the Theory of Monopolization and Exclusion. Oxford: Clarendon Press. Nandi, S., Dasgupta, R., Kanungo, K, Nundy, M., & Murugan, G. (2012). Challenges in attaining universal health coverage: Empirical findings from Rashtriya Swasthya Bima Yojana in Chhattisgarh. BMC Proceedings, 6(5): p.12. Accessed from: http://www.biomedcentral.com/content/pdf/1753-6561-6SI-O5.pdf. Nandi, A., Ashok, A., & Laxminarayan, R. (2013). The Socioeconomic and Institutional Determinants of Participation in India’s Health Insurance Scheme for the Poor. PLoS One. Nathan, D., & Xaxa, V. (Eds). (2012). Introduction and overview. In Nathan, D., & Xaxa, V. (Eds). Social exclusion and adverse inclusion: development and deprivation of adivasis in India’. New Delhi: Oxford University Press. Office of the Registrar General & Census Commissioner. (2011). Census of India 2011: Provisional Population totals: Paper 1 of 2011. Accessed from: http://www.censusindia.gov.in/2011census/ population_enumeration.aspx. Planning Commission of India. (2007). Poverty estimates for 2004-2005. Press Bureau of India. Accessed from: http://pib.nic.in/newsite/erelease.aspx?relid=26316. Planning and Statistics Department, Government of Karnataka. (2005). Human Development Index Report 2005. Bangalore. Accessed from :http://www.isec.ac.in/shdr_kar05.pdf. Palacios, R. (2011). A new approach to providing health insurance to the poor in India: The early experiences of Rashtriya Swasthya Bima Yojana. In India’s Health Insurance Scheme for the Poor. In Robert Palacios, Jishnu Das & Changqing Sun (Eds), India’s health insurance scheme for the poor: Evidence from the early experience of the Rashtriya Swasthya Bima Yojana. New Delhi: Centre for Policy Research.1-37. Parmar, D., Williams, G., Dkhimi, F., Ndiaye, A., Asante, F.A., Ankomah, F., Kojo Arhinful, D., & Mladovsky, P. (2014). Enrolment of older people in social health protection programs in West Africa–Does social exclusion play a part? Social Science & Medicine 119: 36–44. Popay, J., Escorel, S., Hernandez, M., Johnston, H., Mathieson, J., & Rispel, L. (2008). Understanding and tackling social exclusion. Final Report to the WHO Commission on Social Determinants of Health. Lancaster: WHO Social Exclusion Knowledge Network. Retrievedfrom:http://www.who.int/social_ determinants/knowledge_networks/final_reports/sekn_final%20report_042008.pdf. Rajasekhar, D., Berg, E., Ghatak, M., Manjula, M. & Roy, S. (2011). Implementing health insurance: The rollout of Rashtriya Swasthya Bima Yojana in Karnataka. Economic and Political Weekly 46(20): 56-63. Accessed from: http://www.epw.in. Reddy, S.K., Selvaraj, S., Roa, K.D., Chokshi, M., Kumar, P., Arora, V., Bhokara, S., & Ganguly, I. (2011). A Critical Assessment of the Existing Health Insurance Models in India. Accessed from: http:// planningcommission.nic.in/reports/sereport/ser/ser_heal1305.pdf. Rosigno, V. J., Garcia, L. M., & Bobbit-Zeher, D. (2007). Social closure and processes of race/sex employment discrimination. Annals of the American Academy of Political and Social Science, 609(1). Rout, S., & Patnaik,P. (2014). Decentralized government in the context of conflicts over development in India. Journal of Developing Societies. 30(2). Stephens, C., Nettleton, C., Porter, J., Willis, R., & Clark, S. (2005). Indigenous peoples’ health–why are they behind everyone, everywhere? The Lancet 366(9479). Accessed from: http://www.thelancet.com. Subramanian, S.V., Smith, G.D., & Subramanyam, M. (2006). Indigenous health and socioeconomic status in India. PloS Med 3(10): e421. Sun, C. (2011). An analysis of RSBY enrolment patterns: Preliminary evidence and lessons from the early experience. In: Palacios R, Das J, Sun C, editors. India’s health insurance scheme for the poor: evidence from the early experience of the Rashtriya Swasthya Bima Yojana. Chapter 4 New Delhi: Centre for Policy Research. Sundararajan, R., Kalkonde, Y., Gokhale, C., Greenough, P.G., & Bang, A. (2013). Barriers to malaria control among marginalized tribal communities: A qualitative study. PLoS One 8(12): e81966. Tilly, C. (2001). Mechanisms in political processes. Annual Review of Political Science 4: 21-41. World Bank. (2011). Perspectives on poverty in India: stylized facts from survey data. Washington D.C.: World Bank. Thorat, S., Mahamallik, M., & Venkatesan, S. (2007). Human poverty and socially disadvantaged groups in India. UNDP India Discussion Paper series 18. New Delhi: Human Development Resource Centre. Weber, M. (1978). Economy and Society. Berkeley, CA: University of California Press. Xaxa, V. (2001). Positive discrimination: why do tribals lag behind Scheduled Castes. Economic and Political Weekly 21(27). Accessed from: http:// www.epw.in. Xaxa, V. (2005). Politics of Language, religion and identity: Tribes of India. Economic and Political Weekly 40(13). Accessed from: http://www.epw.in.