Local Political Consolidation in Bangladesh: Power, Informality and Patronage

During the past decade Bangladesh has shifted from a competitively clien-telistic two-party system towards a dominant-party democracy. This article analyses how the ruling party has consolidated partisan political control at the local level. Using qualitative ﬁeld data from 2004 and 2016, and drawing on a post-structural analysis of the state, it shows how this extension of power has been achieved locally through interaction between formal and informal political initiatives. Four main types of informal activity are documented, through which the extent of local political competition has been reduced: circumvention, capture, brokerage and the creation of new organizations. These insights into the changing nature of Bangladesh’s long-standing partisan politics highlight how the state’s capacity to deliver local services, allocate resources and maintain stability has been enhanced through the ruling party’s control of local government structures, its elimination of political opposition, and its reshaping of local patronage arrangements.


INTRODUCTION
In 2015, Bangladesh met the World Bank's criteria for classification as a 'lower middle-income country', a striking turnaround for a nation given limited prospects when it seceded from Pakistan in 1971.This transition has been based on a relatively inclusive path of export-led development that, combined with effective public and private investments in human and social capital, has produced a pattern of pro-poor growth (Hossain, 2017;Lewis, 2011).Yet Bangladesh has also come to be regarded as a case of governance failure, characterized by an antagonistic form of 'winner-takesall' politics fought out between two main political parties, and with high levels of corruption (Islam, 2013).Some observers have therefore referred to the Bangladesh 'paradox', pondering the relationship between economic growth and social progress on the one hand and political dysfunction and lack of democratic consolidation on the other (e.g.Khondker, 2018).
Anxieties around governance have further increased with a recent shift from a merely dysfunctional form of politics towards a more specifically authoritarian type of single-party democracy.The ruling Awami League (AL) has been in power since 2008, and has sought to consolidate its position as the 'natural' party of government through its management of the economy, by undermining its political opposition, and by extending patronage roots further into local communities.In 2014, the party was returned to power when the main opposition Bangladesh Nationalist Party (BNP) refused to participate in the election.This was the first time that an incumbent government had been re-elected since the restoration of electoral democracy in 1991, signalling a shift from a competitively clientelist two-party system to a dominant-party democracy.This new period of political continuity has not made Bangladesh's politics any less harsh.The AL government has come under increased local and international criticism for its deteriorating human rights record, with reports of forced disappearances, extra-judicial killings, arbitrary arrests of journalists, and the imposition of new legal restrictions on civil society organizations (see, e.g., The Guardian, 2017;HRW, 2016).Nevertheless, the political consolidation project has continued and intensified, and the AL secured a third term of government during the national elections that took place in December 2018.
In making its decisive break with the recent past, the AL has disrupted an established elite political settlement that has pertained for the past three decades.This had produced conditions of competitive clientelism between two opposing party blocs driven not primarily by ideological difference but by short-term accumulation goals and intense political antagonism, organized through a patron-client pyramidal structure with the two main political parties at the top (Islam, 2013;Khan, 2013).The effort to move beyond this long-standing pattern of rotating plunder has required a national-level strategy with multiple elements: the successful management of a growing export-led economy (with an average of 6 per cent GDP growth achieved during the past decade); careful and pragmatic alliance building with key religious and military interest groups in order to minimize dissent; the abolition of the distinctive three-month neutral 'caretaker government' system that since 1991 had overseen each election; and the politicized use of the International Crimes Tribunal (set up in 2009 ostensibly to secure justice for crimes committed during the 1971 war) to weaken political opposition.
While the party's strategies have received some analysis at the national level (Riaz, 2016;Zafarullah, 2013), comparatively less attention has been paid to the micro-politics of how this consolidation has been achieved locally in rural areas.By consolidation, we refer primarily to the AL's success in obtaining and maintaining control of state resources.This has taken place against a backdrop of continuing efforts to build a more accountable and inclusive local government system through reforms to sub-district and community-level administrative institutions (Chowdhury and Panday, 2018).In what follows we identify and analyse changing political practices through which the AL has extended its power, including increased politicization of local administrative roles, the reduction of independent civil society space, the reshaping of patron-client relations, and the emergence of new forms of party-affiliated brokerage.We examine the ways that such strategies have been enacted locally, with efforts to work both through and around organizations and institutions, as the AL has attempted to strengthen its influence within community-level processes.In particular, the analysis highlights the changing interplay between formal and informal political initiatives.
The article draws on two sets of qualitative data from three local communities of Greater Faridpur district in south central Bangladesh, collected during 2004 and 2016 as part of a study of community-level power and politics.1That the two fieldwork periods were just over a decade apart makes it possible for us to analyse trajectories of change in these communities.The studies were based on interviews with community leaders, local officials and civil society activists, and structured focus group discussions with local people disaggregated by poor and non-poor villagers, and by gender.Three contrasting fieldwork locations in Greater Faridpur were chosen for the original study to reflect differences in infrastructure and connectivity, with one well-connected, one peri-urban, and one remote village community (Lewis and Hossain, 2008).For the follow-up study we returned to the three locations, using the same data collection approach.Fieldwork was mainly carried out during the months prior to the 2016 Union Parishad (UP) elections. 2This provided an opportunity to examine how the three UP Chairs in our study sites (local leaders nominally affiliated to the opposition BNP) were faring in the face of AL efforts to consolidate local power.The earlier study had reported a form of local politics primarily rooted in competition between patrons and only weakly linked to the interests of national political parties.By the time of our return, these local leaders had become increasingly marginalized by the extending power of the ruling party.Each was now struggling to maintain his authority in the face of efforts by local AL party activists, in conjunction with the local Member of Parliament (MP), to establish greater control of local institutions, connect with local constituents, and ensure that the party would be successful in the local elections.
We analyse these data using a post-structural approach to state formation (Lund, 2006(Lund, , 2016)).Changing forms of local political practice have enabled the AL and its supporters to consolidate local power by reshaping interactions between local government and civil society.We find that during the period under study the AL has gradually concentrated its control over state resources in these areas as a result of its growing use of informal connections between party and state networks.The result is a reduced level of local political competition and the achievement of electoral dominance by the AL.The effects of this strategy have not only disempowered the political opposition, but also strengthened the capacities of local patrons -in both formal and informal positions of power -to better meet the needs of their clients and thereby secure political support.In developing our analysis, we therefore draw upon, and reframe, recent conceptualizations of informality and patronage.In describing and analysing these trends, the article aims to make a contribution to the literature on the informal dimensions of political processes.A key strength of the approach is that the re-study offers a rare opportunity to trace how informal political practices have changed over time.
The argument in the article is developed as follows.First, we discuss changing local power relations in Bangladesh in the context of local government institutions and ongoing processes of decentralization.Second, we analyse changing local political practice in terms of the shifting interaction of formal and informal initiatives, with a selective critical discussion of literature on informality and patronage.The third section draws on this to present and analyse our empirical data.We identify four main ways through which this shift is being enacted locally through strategies of circumvention, capture, brokerage and the creation of new organizations.Each of these four dimensions of political consolidation is then briefly illustrated using empirical vignettes.

LOCAL POWER, INFORMALITY AND PATRONAGE IN BANGLADESH
The formal administration of local government in rural areas is organized around a system of elected Union Parishads or councils, that form the lower tier of the country's administrative system, with each one normally serving a total of nine villages.These are part of a long history of efforts by central authorities to establish control in rural areas, going back both to the British colonial period, and to three military governments that took control of the country during the 1970s and 1980s.The UP was originally established by the colonial authorities in 1870 with appointed rather than elected membership for the purpose of local control, tax collection and improving local roads.From 1983 onwards, UPs were constituted by an elected chair, nine ordinary members and three women members.The UPs continued to organize local economic activities and, while state authority came to be exercised mainly through them, political parties played little role.An existing informal power structure based primarily on clan, landholding and patronage became incorporated into these local administrative bodies through a 'loose coupling' arrangement, with the UPs remaining closer to rural administration than to genuine local government (Jahangir, 1982: 24).
The UPs serve as distribution points for central development resources that are allocated by elected, locally accountable members who chair various standing committees, convene and facilitate local dispute settlement, and organize the collection of local revenue from the licensing of marketplaces and awarding of fishing rights.Project Implementation Committees (PICs) oversee local development projects, chaired by elected UP members.Although many regimes have attempted to politicize local government, such as General Ziaur Rahman's attempt to build gram sarkar (village government) arrangements during the 1970s, or General H.M. Ershad's creation of the new upazila (sub-district unit) in the 1980s, these structures have until recently been only weakly articulated with national political parties, and have remained more firmly tied to local landowning elites.This pattern continued into the post-1991 period of parliamentary democracy, where local politics became the struggle to capture and control the lower part of the pyramid, where state resources were distributed to community-level clients, and local political leaders shifted their allegiance between the two political party blocs on a pragmatic basis.
Local administrative bodies such as the UPs and the upazilas have for many years also been subject to a range of 'capacity-building' initiatives supported by international donors such as the World Bank, as part of developmental efforts ostensibly to strengthen local governance and improve accountability to citizens.The 2009 Local Government (Union Parishad) Act further widened the role of the UP by introducing new arrangements at ward (village) level intended to promote 'social accountability' through the introduction of open budget meetings, improved arrangements for people to access local information and services, including issuing documents and certificates, and new formal arrangements for representatives of civil society groups and non-governmental organizations (NGOs) to play roles in local planning and budgeting (Ahmed et al., 2010).The Act also introduced a Citizen Charter to provide more transparent information about the UPs' responsibilities in relation to service provision and rights of citizens, aligned with a Right to Information Act that was passed in 2009.
When local elections took place in 2016, the AL gained formal control of most of the country's UPs.Nationally, a total of 2,661 AL chairpersons were elected, and only 367 from the BNP.Compared with 2011, the BNP's share of seats fell from 38 per cent to only 9 per cent, while the AL's increased from 49 per cent to 70 per cent.In our study area two of the previously BNP-affiliated UP chairs were replaced with AL candidates.In the third site there was no UP election since, following the encroachment of the town, it had decided to apply to become a municipality.
Our data show how the AL has used informal connections to develop tighter control over state resources and has successfully translated this into electoral advantage.This has been achieved through extending the brokerage roles and activities of local party workers, capturing local bodies (or bypassing them in cases where AL does not have political control) and, in a few cases, through the creation of entirely new local organizations.We do not suggest that local political consolidation should only be understood with reference to local factors.Rather, it has been substantially underpinned by a much wider set of national-level strategies such as: (1) extending the role and influence of MPs who determine the distribution of key central government resources that flow into local areas; (2) maintaining the continuing economic dynamism that extends household livelihood options through small enterprise and migration opportunities; and (3) increased control of civil society space so that conditions for service provision and economic support by civil society organizations are prioritized over rights-based work and critical advocacy.The AL's electoral success around the country has also been facilitated by a combination of straightforward electoral popularity, vote rigging and the use of organized violence. 3

THE SHIFTING INTERACTION OF FORMAL AND INFORMAL INITIATIVES
In order to understand further how this consolidation has taken place, it is necessary to focus first on the changing interaction between formal and informal politics, and second on the patronage relations that inform the dynamics of local politics.Bangladesh's political settlement (the balance of power between social classes) has enabled a relatively high level of economic growth and secured positive changes in relation to social indicators, particularly in health and education.However, this has not generally been understood as producing the conditions that can support either a stable or an inclusive democracy.For example, to meet the criteria of North et al.'s (2007) concept of an 'open access' political system would require the existence of formal property rights, enforceable contracts, and an efficient and impersonal bureaucracy, none of which are sufficiently present.The dominant form of political organization is one of competitive clientelism that requires that most forms of organized political action (and economic activity) will normally involve the purchase of support on a personalized basis (Khan, 2013).Without a state that can enforce the rule of law in a systematic way, forms of rent seeking are commonplace.In North et al.'s terms, Bangladesh can therefore be regarded as a 'limited access order' that meets few of the conditions for a functioning democracy.
Yet state and public authority, as Lund (2016Lund ( : 1200) ) has argued, needs to be understood not just in terms of how it matches or falls short of an idealized template, but also in terms of how it actually works, and how in this sense it is always 'in the making'.A post-structuralist approach emphasizes the ruling party's efforts to inhabit the state in ways that redefine the conditions and 3.While there were media reports of violence and vote rigging in many areas, we did not hear of this in our three study locations.
organizing logics through which people encounter and experience it.This requires an analysis of how the ruling party attempts to manage local-level political competition, rights and resources in ways that ensure the conditions not only for exercising governance, but also for consolidating its political power in local areas.This is achieved not only through the use of formal channels, such as by changing administrative rules or laws, but also through the deployment of a set of interlinked informal initiatives and activities.
The analysis of informality enables insight into the workings of power structures beyond the surface level and makes visible 'interactions and negotiations far away from the transparent and accountable systems of modern government' (Mitlin, 2014: 3).Yet as a mode of organizing, informality operates within and not just outside formal institutions.Formal and informal activities and relationships cannot easily be disentangled (Meagher, 1995;Wood, 2000), and each help mediate relationships between state and society.The informal dimensions of public authority are central to the working of local power structures.People's daily experience is one of interacting not only with the formal rules associated with governance, but also with an array of informal arrangements.Focusing on the extent and type of informality offers insights into how power is being exercised, by whom, and for whose benefit.Resisting a conceptualization of a fixed boundary between formal state institutions and informal relations, we favour a post-structural view in which the state can be seen to promote informality as part of efforts to secure its authority.As Blundo (2006: 799) shows, this approach can help reveal changes in the 'operational logics' of local government and the ways in which the informal may become institutionalized as part of the state's everyday mode of operation.Also useful here is Roy's (2009: 82) concept of 'calculated informality'.Roy characterizes informality as a 'mode of practice', and India's planning regime as 'an informalized entity, one that is a state of deregulation, ambiguity, and exception' (ibid.: 76).As a result, informality needs to be understood politically as integral to how the state operates rather than simply as a casual or spontaneous aspect of individual relationships.
Bangladesh's society is strongly hierarchical, and widely structured by patronage relations.Breman (1974: 18) defined patronage as 'a pattern of relationships in which members of hierarchically arranged groups possess mutually recognized, not explicitly stipulated rights and obligations involving mutual aid and preferential treatment'.Vertical social arrangements of mutual obligation and loyalty are constructed between unequal individuals through which negotiations around access to resources, protection, social mobility and political support take place.For Jahangir (1982), patronage is a personal transaction between two interested individuals as opposed to one directed at addressing the collective good, and he distinguishes both economic and political forms.This personalized quality is a key reason as to why it remains difficult for poorer people in Bangladesh to build 'horizontal' relationships of solidarity among equals, such as through joining coopera-tives or the community groups that both government and NGOs have tried to form at various times (Lewis, 2017).From the point of view of clients, such vertical relationships are not only asymmetrical and personal but also uncertain, since they remain discretionary rather than formalized or rightsbased.Patronage is central to the organization of informality within political practice.
Patron-client relations are rarely straightforward or fixed, but are complex, multi-stranded and always evolving.As Leonard et al. (2010: 482) point out, 'patronage is not disappearing so much as changing its shape'.In Indonesia, for example, Simandjuntak (2013: 96) describes the changing patterns of elite competition that allowed ordinary people to increase their capacity to choose 'which patrons were suitable for them'.For Piliavsky (2014) emerging forms of democratic statehood in South Asia have not dampened the practice of patronage which instead thrives 'alongside and often through it', persisting not as 'a feudal residue, but a current political form in its own right' (ibid.: 4).It should not only be seen in functional economic or political terms but also as a 'living moral idiom' that helps to structure people's political aspirations, engagements, practices and relationships (ibid.).Patronage should not be seen merely as a mode of exploitation, as a barrier to democracy, or a vestige of tradition, but also as a productive relationship.It can be understood productively as a linking mechanism between urban and rural spheres, or between bureaucracies and local traditional institutions.Patronage relationships can also be viewed as the means through which those who are poor or excluded undertake their civic engagement, seek to improve their security and practice politics.
Those who emphasize the importance of a vibrant civil society as a necessary condition for democratic consolidation tend not to view patronage favourably.It is usually regarded as antithetical to the working of civil society since it is enacted through unequal vertical relationships rather than ones based on horizontal forms of solidarity.4Such a distinction is challenged, but in another sense also supported, by Chatterjee's (2004: 4) conceptualization of governance and democracy in India.He has suggested that civil society, with its formal rules and laws, may serve powerful interests so that it is merely a 'closed association of elite groups', set aside from the broader population of ordinary poor people.The poor are instead forced to negotiate and contest their claims in relation to government within the wider and messier uncivil space of 'political society' where there are fewer formal rules and norms.Patronage here has a positive character, since it provides a means through which poor people can advance their interests in ways that challenge lawful but exclusionary 'formal' civil society frameworks.Martin's (2014) analysis of patronage from Pakistan questions the optimism of Chatterjee's perspective.He finds that patronage arrangements tend to benefit elites more than they do ordinary people, by actively undermining the possibility of universal service provision, limiting poor people's rights, contributing to the de facto privatization of public services, and reducing the potential for impartiality in dispute settlement.The assertion that ordinary people benefit from 'political society' as a sphere of patronage that bypasses 'bourgeois law' and bureaucratic procedure (because it allows them to extend their freedoms using means other than the legal rights supposedly available to them within 'civil society') ignores what Martin terms the 'dark side' of patronage.Our analysis also suggests an oversimplification implicit in Chatterjee's dualism, since there are intertwined elements of both civil and political societies apparent in local forms of organizing and action in Faridpur.But at the same time, Chatterjee's description of politicians who operate in political society by 'circumventing and breaking the law to appease their constituents' (Martin, 2014: 327) is consistent with an emphasis on the interaction of formal and informal initiatives.For example, while this produces arrangements that are -from the perspective of many ordinary poor people -both insecure and non-permanent, it may bring other benefits.A recent shrinking of civil society space observed in 2016 in Faridpur can also be set against the positive economic climate and the expansion of public services and social safety nets. 5

UNDERSTANDING LOCAL POLITICAL PRACTICE
The re-study provided an opportunity to analyse how informality and patronage have changed in rural areas.Influential studies such as 'The Net: Power Structure in Ten Villages' (BRAC, 1982) have tended to represent villagelevel power structures as closed, predatory and dominated by traditional elites.By the turn of the millennium, researchers from rural Bangladesh were reporting that the power structure was becoming less rigid, with a wider range of patrons with which poorer people could negotiate alongside traditional landowners, including non-landholding elites and civil society actors.Our 2004 data confirmed this picture, with forms of political negotiation and local problem solving that were consistent with what we (and others) had seen as an 'opening up' of local village-based rural power structures (Lewis and Hossain, 2008).This wider selection of potential patrons from which people could choose -including local elected officials, party politi- cians, NGOs and local business leaders -could also be accessed through a growing range of brokers. 6 This trend is consistent with Leonard et al.'s (2010) observation that there has been a general shift in the character of patronage relationships in many societies away from local land-based monopolies of power towards a situation where clients can benefit from growing competition between different potential patrons seeking influence, since it becomes possible for clients to exert more leverage within patronage relationships.The result has been that a more diverse and less unitary 'multifocal power structure' has taken shape in rural Bangladesh, bringing greater room for manoeuvre for people as they go about negotiating with local elites (Mannan, 2015).The earlier metaphor of the 'net' had been appropriate because it implied relationships that had mainly constrained, but now there was increased scope for poorer people to build more connections potentially linking them with productive resources and opportunities (see Gardner, 2012).
In Faridpur, our first set of data had suggested that these changing conditions were creating new opportunities to participate in, challenge and sometimes transform local institutions in more democratic ways (Lewis and Hossain, 2008).For example, there were signs of emerging pro-poor synergies between local government officials and civil society actors.Despite a national and local political system that had become highly adversarial and polarized between two confrontational political party blocs, we found some cases of positive, negotiated change taking place locally.In some cases, these were enabling ordinary people to have more participation in local decision making (such as in the choice of local infrastructure projects), or were strengthening the rights of the landless.For example, one elected UP leader was found to be cooperating with NGO-formed landless groups to strengthen land rights and increase his local popularity.Another social mobilization NGO working in the area had built a productive relationship with a UP Chair that enabled a group of landless to enforce stronger claims and establish rights in relation to an area of khas land 7 that was being encroached upon by local elites.
Local informal institutions were also changing.We found more inclusive shalish arrangements (village-level dispute settlement bodies convened by elders and local officials), supported by government and NGO projects, with outcomes that improved access to justice at least for some local women.Village elites seemed to be more ready to accept more gender-inclusive 6.Brokers are intermediaries who use their knowledge and networks to connect people to opportunities and resources.The concept of brokerage is useful because it captures a key aspect of agency which, in keeping with the idea of informality, is that local political practice may not follow orderly or predictable scripts but instead may take improvised or even disorderly forms (Mosse and Lewis, 2006).7. Khas land is newly available unowned land that emerges from rivers when they shift course.
It is earmarked by law for redistribution to low-income households under land reform legislation, but is frequently illegally occupied by local landowners.
shalish arrangements, particularly if this increased the reputation of local leaders, and helped to attract outside resources.At this time there was a fashion for donors and NGOs to support this type of modernization of the shalish.The findings from the earlier study also indicated a high degree of local diversity: these issues were playing out differently across the three settings that were studied, suggesting that context and locality -such as the state of the economy, the types and extent of civil society organizations that were operating, and the individual strategies of local political leaders -mattered considerably.
Returning to the same communities for the follow-up study, we encountered significant changes.A more dynamic economy meant that most people were now significantly better off than they had been in the mid-2000s, and in two of the three locations there had been major improvements in both the extent and quality of local infrastructure.At the same time, we found fewer NGOs or other civil society organizations doing rights-based work than we had observed before.With the local elections about to take place in UPs that were controlled by opposition BNP-affiliated popular local leaders, the ruling AL was now working hard to strengthen its influence and control of local government institutions.Where there was an elected official in place who was inconveniently from the opposition party, we observed various efforts to try to work around the problem.
A distinctive set of local informal political practices is occurring that takes four interrelated forms: (1) circumvention, where for example the bypassing of the UP provided a mechanism for the AL to distribute resources directly to its support base; (2) capture, such as the increased ability of AL activists to determine shalish outcomes, or to influence local School Committees; (3) brokerage, carried out by informal party 'helpers', that extends the AL's reach into local communities, such as a minority community lawyer who helped broker stronger ties with outside AL activists to ensure safeguarding of the rights and vulnerabilities of local community members; and (4) creation of new organizations, such as a newly-established auto-rickshaw driver association that was a work-around to avoid problems over rickshaw licensing.
While these are not new phenomena in rural Bangladesh, they contain new dimensions.Such practices extend and adapt existing types of informality and indicate important changes taking place in the conduct of local governance and politics.The need to analyse these changes is further supported by data from other studies, although to date work on this has been mainly drawn from urban contexts.For example, in a study of space and 'encroachment' in a Dhaka slum, Hackenbroch and Hossain (2012) draw attention to the transformations taking place in urban space through types of 'informal formalization'.They find that poorer residents' access to water and other services takes the form of protected business opportunities that may provide short-term benefits but ultimately restrict people's rights and choices.More recently, Jackman (2017) has analysed key aspects of the micro-politics of AL consolidation efforts through the changing use of different forms of intermediation.Jackman's data suggest that the previously well-documented significance of the mastaan (hoodlum or 'strong man' figure) was fading in the Dhaka slum areas where he conducted his ethnographic research, being replaced instead by a set of informal arrangements that were being more directly enforced through the AL party and its associates.
In our earlier study we had found that local political leaders, while nominally affiliated to national political parties, drew most of their legitimacy from their local reputations as elite patrons and from clan affiliation (Lewis and Hossain, 2008).Our 2016 data indicated growing political control from the centre, with a larger proportion of resources flowing into local communities (including those for development projects, an increased range of social welfare payments and the greater use of discretionary MP funds).There is a tighter grip on the institutions that are supposed to make local decentralized decisions about how these resources are to be used, and in the case of welfare benefits, decisions about who should get what.Local administrative arrangements have therefore been more strongly politicized in political party terms.Overall, the two main trends we observe in the contemporary picture drawn from our recent fieldwork are that while politics has always been partisan, it has become more so; and that the AL has become more adept at consolidating its power at the local level.
In this way, the informal political practices we have described are underpinned by, and intermingled with, changes made to formal arrangements.As Wood (2000) reminds us, there is permeability between formal and informal institutions.For example, in a formal change to regulations made since our earlier study, the government no longer requires that the UP PICs must be chaired by an elected UP member.This makes it possible for local political activists to achieve more influence in the allocation of resources in UPs even where the opposition party is formally in control.At the sub-district level, the Upazila Development Coordination Committees (UDCCs) that we had previously observed playing significant roles in the oversight of local development processes, have also now become sidelined.
Another example of a formal change that underpins these informal practices is the recent party politicization of the UP election process.In 2015, the government decided that, for the first time, voting in the elections was to be organized along strict political party lines.This was a significant difference compared with the previous election in 2011, since all candidates were now to be nominated by political parties, and would use party symbols on their campaigning materials and ballot papers to contest the election.Previously, political parties had given only informal backing to their candidates, and aspiring candidates could therefore file their nominations without requiring party approval.This is a formal change that reflects the AL's intention to create a greater level of centrally controlled party politicization of local processes.
In the following sub-sections, we move on to illustrate each of the four interrelated dimensions of informal political practice in more detail, using brief vignettes drawn from the data.

Circumvention
The bypassing of local institutions is a key element of political practice.For example, in the well-connected village, when it comes to local public services, ordinary people's access is not only heavily personalized (as it was before) but is now also increasingly party-linked.The elected members of the UP draw up lists of eligible people, as before, but the UP Chair -under pressure from the AL-affiliated Upazila Chairman -now also puts aside a proportion (believed to be around 50 per cent) for specific distribution to local party leaders, activists and supporters.This was also the case with the allocation and management of road repair and renovation projects, many of which are now under the control of party leaders and activists, allowing them to provide resources to clients, reward existing supporters and enlist new ones.In the new scenario, the UP Chairman and members are able to implement a smaller proportion of projects through the formal UP channels than we had observed during the earlier study.The UP's standing committees, which are supposed to widen local decision making and provide oversight of local development activities, were therefore less active (and in some cases totally inactive).
The UP is also engaged in local resource mobilization.However, when it comes to the leasing of local water bodies and allocation of marketplace trading licences, we found that people affiliated to the political opposition now considered it futile to participate in the tendering process, since it was fully controlled by the ruling party and not open.In the peri-urban village the local AL president, and not the BNP-affiliated UP Chair, was now the most powerful source of authority, securing a high degree of influence over the UP.For example, he was found to conduct the most important shalishes, such as those concerned with local land disputes, rather than the UP Chairman who has formal responsibility for this.We also heard some reports of shalishes that were now being convened by leaders of the local Chhatra League, the AL's student wing.
In both cases, the relationship between the MP and the Upazila Chairman is crucial in helping to cement these forms of bypass arrangement.As a result, decisions about who is entitled to social welfare benefits are no longer the sole preserve of the elected UP authority but are now shared with the local party.In the remote area, around 70 per cent of the vulnerable group development programme (VGD) cards are still distributed as normal through the UP, but the remaining 30 per cent are allocated by local AL leaders and activists.The Upazila Chair simply informs the UP staff of the MP's instructions.This is also the case with Test Relief and Food for Work (FFW) programme entitlements.Local AL leaders draw up lists of recipients and submit them to the MP, who then sends his approved instructions down to the UP officials.In the case of the FFW programmes, which require decisions about local infrastructure building and repairs, the Upazila Project Implementation Officer makes an assessment and issues instructions according to the MP's recommendations, thereby circumventing UP-level PICs.

Capture
Our earlier data made it clear that many of the local institutions that are regarded as forming part of local civil society, such as School Committees, Business Associations, Bazaar Committees and transport organizations, were sites of contestation between political parties.From our 2016 data, it was striking how comprehensively the AL had now secured effective control of civil society groups such as School Committees and Business Associations, as well as local administrative committees such as those responsible for leasing markets and water bodies.This finding is reminiscent of Lund's (2006) concept of 'twilight institutions' in which he suggests that alongside the more familiar arenas of politics, political authority can be wielded and negotiated within organizations that do not at first sight seem political.
For example, an important water body in one Union is controlled by the Bangladesh Water Development Board (BWDB) and leased out each year for valuable fishing rights.It used to be the case that the UP would arrange for an annual advertisement for leasing to be distributed, and any interested fishing group could bid for rights to fish, with the highest bidder granted access for the season.However, the ruling party, through the local leadership of the Jubo League (the youth wing of the AL), has in recent years ignored these rules, and explicitly forbidden other groups from participating in the bidding process.With the support of the local MP, who is also a minister, they suspended normal arrangements and effectively captured the tender process.As a result, political pressure was successfully exerted to ensure that the BWDB ignores all other applications in favour of this particular one, such that the rules that were previously adhered to no longer apply.
In the case of local religious associations such as Mosque or Temple Committees, we found that rather than being explicitly politicized, compared to the first study, they had mostly withdrawn into a narrower religious function.The wider community roles that we had observed in the earlier study -such as becoming more involved in shalish and in some cases making such bodies more diverse and inclusive -had diminished.An exception was when the Mosque Committee agreed to cooperate with a community group that was briefly formed to prevent a local gang from extorting money from people at an embankment crossing point, by broadcasting a warning message.Some attempts at capture were nevertheless resisted.In the peri-urban village we encountered a case in which the main local AL leader had become chair of the High School Committee, but had failed to secure re-election.This was because a controversy had emerged after he had refused some students the chance to re-sit exams.This decision had been perceived as illegitimate and made only for political reasons, and was resisted as an abuse of power.In the well-connected village, the MP and his associates (with the support of the AL chair of the local Bazaar Committee) attempted to profit from extending local shops onto public land that had long been used as a football field.The encroachment was protested by local people and their claims taken up by the UP Chairman.He successfully resisted the illegal activity by enlisting support from a group of journalists from the local Press Club, who campaigned against it using the media and attracting help from a nationallevel advocacy NGO.A human chain was organized during a visit to the area by the Prime Minister, and the unauthorized shops were demolished shortly afterwards.Such cases of resistance also illustrate the fragile nature of such state making highlighted within the post-structural approach.

Brokerage
Brokers actively constitute their social and political roles in ways that reflect local conditions, opportunities and constraints.The character of brokerage activities has changed from the earlier fieldwork so that today in Faridpur we find increasing numbers and an enhanced significance of what Berenschot (2019) calls 'party-affiliated brokers'.Two examples can be used to illustrate this.
In the remote village, a former leftist activist, who had gained the respect of villagers for his principled stands on various issues in the past, was elected in 2011 as an AL UP member, despite the UP's overall control by the BNP.He had built a good relationship with the local MP and assisted the AL extend its power further into the village, enabling the MP to bypass the UP Chair.His election as a local council member for the AL came at a time when he had been charged along with several other activists for a violent robbery and he was elected while in custody.When he and his colleagues were later released, the others were killed by military police in a so-called 'cross fire' incident (often a pretext for extrajudicial killings) but he had been saved through his relationship with the MP.This strategy has smoothed a path for the AL to incorporate the village more fully into its patronage network, despite the strength of local support for the BNP-affiliated UP Chair.By making a popular local political leader into a client of the AL he has become a valuable resource, offering a better way for the party to secure influence than simply trying to impose power from outside.The ruling party has been successful in this area in constructing alliances with locally rooted and trusted leftist activists.
In the peri-urban village, we found another intermediary from the local bedia ('river gypsy') minority community, many of whom have lived for many years as squatters in an informal settlement on railway embankment land.A community lawyer and AL activist since his student days, he has through his contacts so far been able to prevent their eviction.Members of this community have been joined in recent years by other locals seeking to escape the effects of nearby river erosion, and this influx has exacerbated pressures.The result has been periodic attempts by the police and other local authorities to remove bedia from illegally occupied land.The lawyer had recently used his contacts to bring the local MP to a community meeting and had managed to secure an agreement to protect bedia rights.With the support of the AL Upazila Chair he has also come to serve as an informal 'coordinator' for his UP, where he also oversees many local development projects.He explained to us that he has been able to access both MP and UP funds and these have enabled him to undertake madrasa, temple, mosque and school projects along with local road maintenance work.During the BNP regime, he told us, there had been fewer local roads and less regular repairs.He has also lobbied for and secured construction of a new bridge over the river nearby and is involved in plans to build a new local college.

Creation of New Organizations
The capacity both to break rules and to 'get around' them by creating new structures is another important dimension of informal political action.For example, we noticed on returning to Greater Faridpur that there was a new type of electric auto-rickshaw that had become available locally since our earlier study.This improved vehicle had quickly become popular with local drivers, since it saved labour and improved journey times, and it was affordable with the help of business loans that are easily obtained by NGOs operating in the local area.However, the new rickshaw was technically illegal because it had not yet been officially licensed for use by the government.In response to unhappy rickshaw operators, this bottleneck had recently been overcome by local activists from the ruling party who had managed to use their contacts to establish a new local rickshaw driver association that provided drivers with protection against prosecution.This made it possible to circumvent the rules and ensure that drivers could pursue their business without interference from the police.

CONCLUSION
There has been a shift during the past decade towards a more partisan form of local politics in rural Bangladesh, where most local organizations are now firmly controlled by the ruling political party.This is in contrast to the picture encountered earlier, when local leaders and office holders were only loosely associated with the national political parties.Local political consolidation by the AL has been achieved through the activities of partyaffiliated power holders playing both formal and informal political roles, in some cases gaining control of local decision-making bodies, and in others working around them using informal strategies.Formal changes to the rules, such as bringing party symbols into local elections, removing the need for PICs to be chaired by an elected UP member, and increasing the discretionary funds available to MPs for local distribution, have also helped underpin this consolidation of local power.
The cases discussed in this article add to our understanding of how postcolonial states take their shape through everyday micro-politics.As the ruling party attempts to re-engineer the country's political and administrative systems towards a one-party democracy, we see how in the post-structural sense Bangladesh's state continues to be remade (Lund, 2016).Here informality is not only something that occurs around the edges of 'formal' institutions, but forms an intrinsic part of the way the ruling party plans and governs.What Sen, in his discussion of hybrid political settlements, describes as the 'interplay of both formal and informal institutions in the growth maintenance process' (Sen, 2015: 51) has played an important role in the AL's preparations for and subsequent successes in the 2016 UP elections and more recently in the 2018 national elections that returned the party to power for an unprecedented third term.
These processes cannot be understood without paying close attention to how the benefits and gains are distributed among those not only with power, but also those without.While models of 'encroachment' by the powerful on the powerless (such as Hackenbroch and Hossain, 2012) are instructive, they do not fully explain the trajectories of local political consolidation observed in Faridpur, where outcomes are more locally varied and sometimes more positive for ordinary citizens than the macro-level templates of political economists might suggest.This mix of formal and informal practices implies neither a purely instrumental relationship that narrows people's political options to impose the AL's agenda, nor a simple opening up of patronage choices that necessarily benefits poor clients as they seek to advance their livelihoods.
Understanding this unfolding micro-politics on its own terms requires us to pay attention to the moral as well as to the instrumental dimensions of political patronage, as Piliavsky (2014) has argued.The ruling party deploys a moral component within patronage relationships when it projects its commitment to solving key local problems of inadequate services, weak infrastructure and poor law and order, and allocates resources for these purposes.It also does so by attempting to reframe the past with promises of a better future.Earlier periods of instability and disorder are today represented by the AL as now having been superseded by an improved governance regime that listens more closely to people and gets things done.For example, many local residents in our re-study perceived that law and order in Faridpur had improved compared with the earlier era of competitive clientelism, which was associated with ineffective policing and widespread mastaan activity.This picture fits within a powerful national narrative in which the AL is recovering its role as the 'natural' party of government that helped create the nation in 1971.
Within an overall context of export-led economic growth, the AL has delivered increased prosperity and relative stability, reshaping patronage arrangements alongside new forms of party-affiliated brokerage in ways that respond to local needs.In recognizing and consolidating informal and formal power at the local level, the party has become more adept at dominating the local power structure.The longer-term implications of this combination of consolidated AL power, a strong economy, weakened political opposition, and reduction of civil society diversity are not easy to predict.While the new and changing forms of patronage arrangements that are currently in place are based on long-standing social institutions, the effects of the ruling party's increasing dominance within the social networks that connect people, power and resources may damage the future stability and inclusivity of the political system.