Dynamics of Designing and Implementing Social Safety Net Programmes in Developing Countries

Social Safety Net as a part of social protection programmes is now very popular in many developing countries. Political discourses have a role in prioritizing and implementing such pro-poor policy. The objective of this paper is to understand the dynamics of political discourses that influences in adopting and sustaining Social Safety Net Programmes in these countries utilizing the common assumptions proposed by Hickey (2006) and Barrientos and Pellissery (2012). The analysis of this paper identifies that the assumptions developed by Hickey, Barrientos and Pellissery are not adequate enough to capture political discourses around SSNPs in the developing countries. These assumptions mainly help us to understand such discourses from the macro/structural perspective and fail to capture the politics at the micro level. The paper, therefore, argues for more rigorous empirical based research from cross-country perspective focusing both macro and micro level politics. Introduction Social Safety Net (SSN) as a part of social protection programmes is now very popular in many developing countries. These countries have introduced several safety net programmes, including cash or in-kind transfers, public works and income generation programs (Subbarao et al. 1997). There is a growing tendency to adopt safety net policies among the policy makers in these countries in order to serve purposes like reducing poverty, inequality and to enhance livelihood capabilities of vulnerable people also in the context of climate change shocks. For example, the success of conditional cash transfer programme in Latin America for reducing poverty has led to the adoption of such programmes in many African countries (Hurrell and MacAuslan 2012). Barrientos and Hulme (2008) have termed this rapid expansion of social protection programmes in developing countries as a ‘quiet revolution’. An analysis of 87 developing and transition countries during 1996-2006 shows that the mean spending on safety nets of these countries is 1.9 percent of GDP (Weigand and Grosh 2008). In thirty OECD countries, on the other hand, the share of social security is 13 percent of GDP (Townsend 2009 in Hurrell and MacAuslan 2012: 258). Though the percentage is trivial in developing countries compared to the developed countries, the share for this type of social protection is significant in terms of available resources and budget constraints faced by these countries. However, there are different schools of thoughts regarding its popularity in 9 Assistant Professor, Department of Public Administration, Jahangirnagar University, Savar, Dhaka-1342, Bangladesh,Email: harekrisnaju@gmail.com,


Introduction
Social Safety Net (SSN) as a part of social protection programmes is now very popular in many developing countries. These countries have introduced several safety net programmes, including cash or in-kind transfers, public works and income generation programs (Subbarao et al. 1997). There is a growing tendency to adopt safety net policies among the policy makers in these countries in order to serve purposes like reducing poverty, inequality and to enhance livelihood capabilities of vulnerable people also in the context of climate change shocks. For example, the success of conditional cash transfer programme in Latin America for reducing poverty has led to the adoption of such programmes in many African countries (Hurrell and MacAuslan 2012). Barrientos and Hulme (2008) have termed this rapid expansion of social protection programmes in developing countries as a 'quiet revolution'. An analysis of 87 developing and transition countries during [1996][1997][1998][1999][2000][2001][2002][2003][2004][2005][2006] shows that the mean spending on safety nets of these countries is 1.9 percent of GDP (Weigand and Grosh 2008). In thirty OECD countries, on the other hand, the share of social security is 13 percent of GDP (Townsend 2009 in Hurrell andMacAuslan 2012: 258). Though the percentage is trivial in developing countries compared to the developed countries, the share for this type of social protection is significant in terms of available resources and budget constraints faced by these countries. However, there are different schools of thoughts regarding its popularity in developing countries. One school claims that the Social Safety Net Programmes (SSNPs) are adopted by the incumbent governments for reducing poverty and vulnerability (Handa and Davis 2006;Rawlings and Rubio 2005;Santos et al. 2011;Soares et al. 2010). And in analyzing contribution of these programmes, most of the studies have mainly concentrated on evaluating the impact on poverty, vulnerability and targeting effectiveness. However, outcome of a development intervention is not a linear one; rather it is mediated by a variety of social, political, legal, economic and cultural issues. Among these, political dynamics might play the crucial role in prioritizing and implementing pro-poor policies. "Anti-poverty interventions are a political process in their own right" (Bastiaensen et al. 2005: 979). Therefore, the other school describes this popularity from political point of view. According to this school, there is a double-edged relationship between politics and SSNPs. On the one hand, political considerations influences in designing and implementing SSNPs and on the other hand, SSNPs also play a role in shaping national or local politics (Bebbington 2006;De La O 2013;Giovagnoli 2005;Hall 2008;Hickey 2007Hickey , 2009Hickey , 2012Hickey & 2013Hossain 2007;Hurrell and MacAuslan 2012;Jones 2012;Maiorano 2014;Pelham 2007;Pellissery 2005Pellissery , 2008Zucco 2008). The incumbent government might allocate resources in SSNPs in order to strengthen their political capital. Similarly, the right based relationship between the state and citizens promoted by the SSNPs may change the political landscape of a country.
However, there are very few studies that deal explicitly with political issues in Social Protection (SP) and particularly in SSN fields (Bebbington 2006;Hickey 2007Hickey & 2012Hossain 2007;Hurrell and MacAuslan 2012). Moreover, existing studies are context or programme specific and the analysis from cross-country perspective is really scarce. This paper is an attempt to study political aspects of SSNPs in comparative perspective by combining multi-country experiences based on analyzing relevant published secondary sources though they differ in many contextual cases. The objective is not to develop a common framework rather to understand the dynamics of political discourses that influences in adopting and sustaining SSNPs in developing countries. The political discourses are constructed differently in different cultures. Likewise, the patterns of governments (e.g. multi-party democracy, dictatorship); centralized/decentralized governance system, degree of bureaucratic autonomy, capacity of civil societies differ country to country. That is why it is difficult to generalize the political dynamics of designing and implementing SSNPs for all countries. This is also a limitation of this paper. However, we have adopted social constructionist perspective to capture different political discourses under some common assumptions developed by Hickey (2006) and Barrientos and Pellissery (2012). This paper mainly focuses on low income and developing countries and ignores developed countries. This is due to the fact that there is already a rich tradition of political analysis of social policies in developed countries (Hurrell and MacAuslan 2012). The analysis of this paper identifies that the assumptions developed by Hickey, Barrientos and Pellissery are not adequate enough to capture political discourses around SSNPs in the developing countries. These assumptions mainly help us to understand politics from the macro/structural perspective and fail to capture politics at the micro level. The paper, therefore, argues for more rigorous empirical based research from cross-country perspective focusing both macro and micro level politics.
To serve the purposes of this study, this paper uses the most frequently cited definition of safety nets: "as non-contributory transfers targeted in some manner to the poor or vulnerable" (Weigand and Grosh 2008: 1). For defining politics, we want to use the definition given by Leftwich. "Politics consists of all activities of conflict, negotiation and cooperation in decisions about the use, production and distribution of resources" (Leftwich 2006: 3).
The rest of this paper analyzes and compare findings based on the assumptions of the relationship between the politics and SSNPs advanced by Hickey (2006) and Barrientos and Pellissery (2012) in their landmark studies. In the later part, the paper identifies several micro level political discourses that influences in designing and implementing safety net programmes before drawing conclusion.

Putting Political Dynamics into Social Safety Net Analysis: General Assumptions?
As stated above, political dynamics in SSN is still an understudied topic and there is a lack of proven analytical framework. However, there is a growing tendency among academicians to uncover this important relationship. A considerable number of literatures have already established that political dynamics has an important role in designing and implementing SSNPs. In this context, Devereux has rightly maintained that "[a] safety-net programme is more than a transfer of resources from 'haves' to 'have-nots'; it is a relationship of power" (Devereux 2002: 673).
To uncover the underlying political dynamics of designing and implementing SSNPs, Hickey (2006) has developed some general assumptions about the politics of social protection programmes on the basis of his study on eight different countries namely Bangladesh, Mozambique, India, Lesotho, South Africa, Namibia, Uganda and Zambia. The assumptions are as follows:  To understand the pro-poor politics, we need to focus on particular trends of capitalist development, state formation and citizenship development from long term perspective.  The 'politics of crisis' might play more active role for emerging pro-poor policies than 'politics as usual'.  The key agents for promoting pro-poor policies are mainly located in political societies. Civil societies have very limited role in promoting such policies.  Patronage and clientelism act as a driving force for emerging social protection policies.  Targeted programmes are politically sustainable.  In case of design and implementation, administrative capacity, central-local relations, political culture are important.  Donors have significant influence in policy making process of the highly indebted countries. Barrientos and Pellissery (2012) have also offered an incremental framework based on their study in India, Brazil and South Africa which gives us several analytical hypotheses to understand the role of political dynamics in designing and implementing SSNPs in the context of developing countries. These hypotheses include (Hickey 2006: 6-11):  Imperfect political competition leads to weaken policies addressing poverty. The policies that emerge from this process tend to serve powerful interests and elites.  Direct democracy (policy making is based on voters' preferences) is favorable for emerging social assistance institutions.  Identity politics is sometimes beneficial for sustaining and enlarging social assistance programmes.
 Civil society play very nominal role in promoting SSN programmes.  The governments who are not dependent on their citizens for generating revenues from taxation enjoy higher degree of autonomy in designing and implementing propoor policies.
However, these assumptions as developed by Hickey (2006) and Barrientos and Pellissery (2012) about the relationships between political variables and pro-poor policies in general and SSNP in particular are not yet empirically tested and therefore it is too early to go too quickly to more specific universal hypothesis. It requires more rigorous comparative research. Hickey (2006) has also cautioned us to use these assumptions very carefully, because these may vary according to contexts and require further empirical studies.

Rights based versus patronage based politics:
Many authors (including Giovagnoli 2005;Hickey 2009;Hossain 2007;Pelham 2007) have tried to explain the relationship between politics and SSNPs from social contract perspective. From this perspective, SSNPs are gradually being viewed as a form of social/political contract between the state and the poorest people to engage in welfare activities (Hossain 2007). It refers to non-legal contracts between the state and its citizens and emphasizes political accountability of states towards their citizens (Hickey 2009). The basic assumption of this approach is that a social contract "denotes not only more legitimate, peaceful and consensual forms of political authority, but also a stronger commitment to social justice that can help relocate public policy within the realm of rights rather than patronage" (Hickey 2012: 427). In many developing countries, poor people constitute majority of their population. These people maintain very little give and take relationship with their states and thus states suffer from legitimacy (Hickey, 2012). However, SSNPs help to establish relationship between the state and its citizens and increase legitimacy (Barrientos and Pellissery2012). In this way, this approach helps us to determine the forms of politics that is required to support SSNPs. For example, the erosion of apartheid in South Africa has broadened the scope for more citizens to be included in the social pension programme and the absence of such contract in northern Uganda tends the ruling regime to deal with chronic poverty through piecemeal patronage based social funds (Hickey 2007(Hickey & 2012. The examples in these countries indicate the nature of contract between the state and citizens, which in turn determines the possibility of inclusion or exclusion based on rights or patronage in accessing to public services. Here, rights are similar to social rights, which can be defined from normative perspective as "a stronger basis for citizens to make claims on their states and for holding states to account for their duties to enhance the access of their citizens to the realization of their rights" (Cornwall and Nyamu-Musembi 2004). This approach also stresses on duties, which may include observance of law, military service, etc. (Ulriksen and Plagerson 2014). Analysis of several SSNPs in the context of developing countries reveals that most of the programmes have been sustained for longer period due to their patron-client relationship. Ruling party politicians at the national and local levels and their relatives are the major beneficiary of these programmes.
Moreover, the analysis of several SSNPs in cross-country perspective reveals that the programmes are generally targeting in nature (which serves mainly small segments of the total population) rather than universal. For example, total population coverage in Nicaragua (3 percent), Chile (6 percent), Colombia (5 percent), Mexico (20 percent), Honduras (15 percent), and Brazil (24 percent) (Bastagli 2009). The perceived benefits of these targeted policies restrict governments of these countries to transform minimal safety nets to inclusive social policy.

'Politics as Crisis' versus 'Politics as Usual':
According to Grindle and Thomas (1991), there are mainly two political contexts for propoor decision making: (a) 'politics as usual'-here policy makers focus on problems based on their preferences and values; and (b) 'politics as crisis'policy reactions in response to outside actions and events (Grindle and Thomas in Hickey 2009: 476). Pelham (2007) has termed these political contexts differently, i.e. 'politics as usual' as ex-ante or supply driven and 'politics of crisis' as ex-post decisions or demand driven, both of which define the relationship between the state and citizens. Though it is presumed that the pro-poor interventions are more likely to take place in 'politics as usual' context through channeling the popular interest in the political system, it is now broadly agreed that politics of crisis also provides opportunity for the development of pro-poor policies (Hickey 2009 (Bryant 2009;Chu & Gupta 1998;Giovagnoli 2005;Hickey 2009;Hossain 2007). Old age pensions in South Africa, on the other hand, are considered to be largely driven in the interests of the government and state (Pelham, 2007). Similarly, old age pension scheme in Bangladesh supports the political manifesto of the present government. Therefore, SSNPs can emerge both in the context of 'politics as crisis' and 'politics as usual'. However, as the developing countries suffer from adequate resources to finance development expenditures, 'politics as crisis' dominates in case of allocating resources for SSNPs.

Politics of Poverty Reduction:
In many of the developing countries, widespread poverty, unemployment, inequality have turned into one of the major political agendas in recent years. During their election campaigns, major political parties make their commitments on reducing poverty and income inequality. The best example in this case is the declaration of Fome Zero (zero hunger) policy by the Lula administration in Brazil immediately after his take-over into the power (Hall 2008). At the same time, it has been also observed that ruling political parties undertake several SSNPs to convince general people to cast vote for the incumbents. For example, the introduction of conditional cash transfer programme named Progresa (now Oportunidades) in Mexico "led to a 7 percent increase in voter turnout and a 9 percent increase in incumbent vote share in the 2000 presidential election" (De La O 2013: 2). Similar effects on electoral behaviour have been also observed in other countries. Bolsa Familia (launched in October 2003), a conditional cash transfer programme in Brazil has been designed to capture votes (Hall, 2008) and the popularity of this programme among the poor has helped to establish a patronage culture in the distribution of benefits, which has helped Lula to be reelected in 2006 election (Zucco 2008). In India, some of the SSNPs have been named after with the name of politicians in order to gain political benefits. For example, central government has introduced some social assistance programmes by prefixing the name of former prime minister 'Indira Gandhi' just before the elections in 1995 (Barrientos and Pellissery 2012). This type of identity politics creates favourable conditions for emerging such pro-poor policies, which in turn helps the incumbent government to stay in office for several terms (Feng and Gizelis 2002).
Forms of Government: There does not seem to be particular forms of government which are more or less favorable for emerging and expanding of such pro-poor policies (Hickey 2006;Hossain 2007). These programmes may emerge under dictatorship or democracy or in any other forms of government. In this case, the best example is China. Social protection programmes targeting the poor households have increased substantially in recent years. In 2010, it has passed the China Social Security Law and this part accounts for 10.4 percent of total government expenditure, which is significantly larger than other developing countries (Needham 2013). In Vietnam, budget allocation for non-contributory programmes was more than 2 percent of GDP in 2009 (World Bank, 2010). In Bangladesh, major SSNPs like VGD and FFW had emerged during the military regime. On the contrary, authoritarian form of government in Senegal appeared to be a constraining factor in reaching out to the poorest segment of the population (Subbarao et al. 1997). Therefore, the claim 'direct democracy is favorable for emerging social assistance institutions' made by Barrientos and Pellissery (2012) may not be true in majority of the cases. Rather, in multiparty democracy, there is a tendency of the incumbent government to distribute resources to the regions where it has strong political support (Aytaç 2014).

History of State Formation and State-Citizens Relationship:
In terms of choosing such pro-poor policies, history of state formation and state-citizens relationships also play an important role (Hickey 2006). Many of the countries in the South were under the colonial rule and most of them are still being influenced by the existing neoimperialist structure. The political elites of these countries are highly dependent on powerful countries and often have little option to consider the interest of their own citizens (Moore 2001). Moreover, policy processes of these countries are highly influenced by the donors. For example, the SSNPs in Uganda, Zambia, Colombia, Honduras, and Nicaragua are heavily dependent on donors funding and the governments have limited ownership in these programmes (Bastagli 2009;Hickey 2013;Ulriksen 2013). On the contrary, NGOs and other civil society members can play very nominal role in compelling governments to adopt propoor policies as it is expected rhetorically (Barrientos and Pellissery 2012;Hickey 2006). One of the reasons might be that many of the civil societies in developing countries are not that strong and dependent on donor funding. Existing social hierarchies in terms of education, gender roles and power relations hinder these organizations to play their roles effectively (Patterson 1998).
State-citizens relationship has a role in mounting SSNPs. Historically, European countries have been constructed based on taxing their citizens and this has directed the governments of these countries to engage in continuous dialogue with their citizens (Moore2001; Ulriksen 2013; Williams et al. 2011). We observe a positive correlation between this relationship and the expansion of social protection expenditure as it has been indicated in the introductory section of the paper. On the contrary, there is a limited relationship between the state and citizens in developing countries. The powerful political elites of these unearned states are highly dependent on aid or natural resources and have less enticement to serve their citizens (Moore 2001;Williams et al. 2011). Therefore, there is a dearth of wider social contract in many of these countries. However, we observe a growing tendency of introducing several pro-poor policies in general and SSNPs in particular among these political elites in order to enhance their legitimacy and to restrict protest against economic liberalization (Hickey 2006).

Policy Discourse of Development Partners
Most of the developing countries have adopted this pro-poor policy as a result of policy diffusion by the international agencies, like, World Bank, International Monetary Fund, World Food Programme, UNESCO, and UNICEF in the name of human capital development, a new paradigm of development following the failure of a neoliberal discourse. Successful implementation of such programmes in neighboring or influential countries also influences domestic policy makers of a country to adopt and sustain such policy options (Fenwick 2013). For example, following the implementation of conditional cash transfer programme in Brazil in the year 1995, many countries in Latin America like Mexico (1997), Nicaragua (2000), Argentina (2002) Uruguay (2005) have implemented such programme to fight against poverty (Fenwick 2013).
We can explain the dynamics of such policy diffusion from inclusive neoliberal perspective. According to Devereux (2002), critics of neo-liberal reforms view SSNs as politically palatable and ways of bypassing to take real initiatives to address the structural causes of poverty and vulnerability. They claim that neoliberal proponents promote structural reform in many of the developing countries under the disguise of pro-poor policies like SSNPs as a residual approach in order to deflect resistance from poor people and to sustain the existing resource redistribution process (Hickey 2007;Ruckert 2006). This mechanism also acts as a restraining factor to turn the poor into an agency and to bring their interests into political discourse. By assisting in the Food-For-Work programme in Bangladesh, the multi-lateral donors like World Food Programme (WFP) and Cooperative for Assistance and Relief Everywhere (CARE), which played the key role in designing and implementing the programme, have tried to realize their own agenda. Donors have used food aid as tool to maintain regional power balance (Brundin1979; McHenry and Bird 1977).
Now the question remains as to why do the policy makers of developing countries accept such policies which are imposed by the external agents? There are several reasons for accepting policy suggestions from development partners. First reason might be the symmetric relationship between the transnational policy ideas and the interest of domestic political actors. According to Fenwick (2013: 144), "… if transnational policy ideas and models can draw on or build upon already established ideational and symbolic beliefs, they can actually be used to further the motivations of key political actors and societal interests Many of the developing countries in the South can be characterized as neopatrimonial states, where "real power and real decision-making lie outside formal institutions" (Cammack 2007: 600). A form of kleptocracy from national to local levels exists in these states. Different forms of aids forwarded to the political and bureaucratic elites of these states by their multi-lateral and bilateral donor agencies not only increases the acceptability of these governments to their people by adopting popular programmes, but also help them to stay in power (Cammack 2007). Similarly, the development partners find it interesting to perpetuate their hegemony in all aspects of state affairs through their interaction with political and administrative elites of these nations. Secondly, these countries suffer from availability of inadequate resources or capacity to design better strategies for utilizing existing resources. Therefore, we observe a growing tendency of policy dependency on donors among these states.
The above analyses based on the two sets of assumptions mainly help us to understand power relations from structural perspective, which fails to capture the local level politics of local micro level actors. These local actors play an important role in the design and especially implementation of SSNPs. Following discussion will corroborate this argument.

Influence of Vested Interests at Local/ Micro-level
Existing literature whimpers for scant studies concerning micro level political discourses surrounding safety net programmes (Maiorano 2014). In reality, the design and implementation of SSNPs (in terms of identification of projects, beneficiary selection, formation of project implementation committees, grievance handling procedure, etc.) are highly influenced by rural power structures.
There are huge debates about the use of beneficiary targeting techniques because the effectiveness of a technique is largely determined by its impact on poverty reduction. It is a complex process which needs to deal with who should be included or excluded. There are mainly three types of targeting methods: (a) means-tested targeting by individual/household assessment; (b) demographic or geographic targeting; and (c) self-targeting (Devereux 1999;Coady et al. 2004). The targeting process under these methods has been expatiated in details by Coady et al (2004) in their study.
Local/micro level interests/politics become visible in interpreting the selection provisions and inclusion/exclusion of poor people in SSNPs because of the discretionary action exercised by the local political elites and street level bureaucrats. Local political elites play the central role in identifying and designing the projects whereas the local bureaucrats approve the projects, monitor their progress and allocate resources accordingly. For some observers, there exists a patron-client relationship between the politicians and bureaucrats which help them to bypass the formal institutional requirements. Consequently, these projects turn into money making machine both for politicians and bureaucrats at the rural levels due to this elite capture (Pritchett 2005). Pellissery (2008) has observed that the poor people had to pay Rs. 10 as registration fee to be included in the public works programme and 5 percent of their daily earnings to the work gang leader to get job for the next day. He has also maintained that corruption takes place through creation of low quality assets. Lack of accountability might be one of the reasons for this debilitated situation. "The problem in many developing countries is that they are still at fairly early stages of state building where institutionalized relationships and checks and balances are often weak" (Williams et al. 2011: S37). Absence of strong institutions is one of the reasons for poor accountability situation in these countries. For example, in India, there are almost 300 programmes being implemented under 13 different ministries and there is no single authority to coordinate these activities (Barrientos and Pellissery 2012). This mismanagement situation has created room for local elites to capture the programmes. On the contrary, in Brazil and South Africa, the implementations of SP programmes are being coordinated by strong autonomous institutions and by a single ministry respectively, which have been able to reduce the influence of local politics (Barrientos and Pellissery 2012). Accountability also depends on appropriate grievance mechanism. In Kenya, it has been observed that the village leaders who are responsible for managing the programmes are also responsible for hearing and redressing the grievances (Hurrell and MacAuslan 2012). Therefore, in this situation of elite captures, local people are unable to fully ensure their accountability.
Rhetorically, identification and design of the projects under SSNPs emphasizes to include multiple stakeholders including poor people. However, a plethora of empirical studies has observed that this type of participation could not empower poor people of the rural areas. In participatory development approach, the critiques argue that the discourse of participation is strongly influenced by power structures within a community. According to Uma Kothari (2001), "participatory development programmes emphasizing social inclusion draw previously marginalized individuals and groups in to the development process, but do so in ways that bind them more tightly to structures of power that they are not then able to question" (Uma Kathori cited in Williams 2004: 563). Participation at the local levels is highly influenced by several structural issues like gender, class, caste and ethnicity and it can thus be linked with more structural power struggles and transformation (Mohan and Stokke 2000;Williams 2004). For example, in one study in rural Kenya, it has been observed that though this approach has increased citizen participation through mobilization, it has failed to circumvent existing power structures (Sheely 2015). Similarly, other studies in Bangladesh, India, Malawi and Ghana do not show any evidence that the participation of poor people in project selection and implementation has enabled them to achieve political capabilities to hold political and administrative elites accountable (Chinsinga 2007;Hossain 2007;Pellissery 2008;Oduro 2015).
Local level politics of vested groups become more visible due to the lack of agency power by the poor people. In case of public works programme in Maharashtra and West Bengal, the inclusion into the programme depends on access to and connection with the local elites, political and casts identity (Das 2015;Pellissery 2005Pellissery & 2008. In Mozambique, relatives of local community leaders were selected in SSN programme (Hickey, 2006). In Argentina, access to income support programme called Jefes depends on political connections, rather than satisfying technical criteria (Giovagnoli 2005). However, most of the poorest people lack this agency power and find themselves excluded from the access to these programmes. For example, 65 percent of the poor are excluded in Bangladesh, 70 percent in Mexico, 59 percent in Brazil from Bolsa Familia (Planning Commission, 2015). In some cases, the allocations rather reaching to the poor people end up with the rich segment of the population. In Brazil, 15 percent of transfers went to the richest quintile and in Mexico, it is 4 percent (Lomelí 2008). Here "Access is the ability to benefit from things" (Ribot and Peluso 2003: 153). Most of the vulnerable people lack this capability to derive benefit from things. This is due to the fact that most assets and capabilities derive from economic, organizational and political processes and poor people have little control over these issues (Tilly 2006). In this sense, their poverty is relational and their rights to get access appropriate mechanisms become a non-issue in political discourse as they lack agency power (Mosse 2010). This is because poor people pursue local leaders individually, rather than collectively to get benefits, which actually undermines their agency power (Hurrell and MacAuslan 2012). Moreover, those who manage to get access into the programmes through their relationship with local elites, they in turn perpetuate this system of elite capture.

Conclusion
From the above discussions, it is clear that intergroup politics play an important role in designing and implementing SSNPs. It influences from either sides. Many agents like donors, political parties, and local elites have influence in adopting several pro-poor policies in general and SSNPs in particular. There is scarcity of robust analytical frameworks for doing comparative studies in this field. However, from the above analysis, we can draw some macro or general trends about the dynamics of designing and implementing SSNPs. We find that adoption of these programmes may take place in the both contexts of 'politics as usual' and 'politics as crisis' for achieving several political capital; any particular form of government may not be favorable for emerging such programmes; donors' influence is higher in those countries which are dependent on foreign aid and similarly social contract is also weaker than those which are dependent on tax. From micro perspective: most of the programmes are captured by local elites; and inclusion by target population into the programmes depends on political connections, ethnicity, gender or caste. Most of the poorest people lack this agency power and find themselves excluded from the access to these programmes. In this context, establishing strong institutions for monitoring and coordinating implementation activities, appropriate grievance redressing mechanism might help in improving accountability and reducing corruption. However, we need further rigorous comparative studies for generalizing these trends.