Nevertheless, the topic requires discussion even within the current scenario. This is not because of the contested status, legitimacy or legality of the current parliament and government, but in fact because of a plan introduced in 2000, the Local Government Plan, which introduced an elected third tier of government in Pakistan through the “devolution of power”.

The discourse on democracy has expanded in recent decades to include more than just electoral processes. From being only about fair, free, competitive and regular elections, the definition of democracy now includes comments on responsive institutions and access to information. Civil and political freedom has also received a lot of attention within this expanded discourse, as has the notion of open political debates, the role of parties and the nature of political campaigns. These concepts have entered the discourse in the interest of allowing voters not just more choice but the ability to make that choice freely in an informed manner without undue political pressures. Because of this concern, a discussion on the role of elites, how they are chosen, and very importantly, social structures and pressures has also become a part of the discourse, based on the recognition that it is not always only political pressures and corruption that limit the ability of voters to choose freely but also the pressures that emanate from being part of a specific social milieu, and that electoral choices made within such contexts cannot always be considered free. Democracy is, therefore, not just a process. It encompasses within it a discussion of social forces and structures and how these interact with the process to create a certain group of political elites and allow often limited choices for voters, no matter how free the actual process of elections itself maybe.

Another addition to the discourse is the concept of “deliberative” or “consultative” democracy and a greater role for citizens in decision-making processes. This is closely connected to the idea that citizens may be better able to decide what they need most rather than the elites that represent them. Dalton, Scarrow and Cain describe this shift in the discourse in the following manner:

“The institutional impact of the reform wave of the late twentieth century can be understood in terms of three different modes of democratic politics. One aims at improving the process of representative democracy in which citizens elect elites. Much like the populism of the early twentieth century, reforms of this mode seek to improve electoral processes. Second, there are calls for new types of direct democracy that bypass (or complement) the processes of representative democracy. A third mode seeks to expand the means of political participation through a new style of advocacy democracy, in which citizens participate in policy deliberation and formation – either directly or through surrogates, such as public interest groups – although the final decisions are still made by elites.”

Despite this growing demand for more participatory forms of democracy, Diamond points out that the “unprecedented pressure (international and domestic) to adopt – or at least to mimic – the democratic form” has led not to the “deepening of democracy” but actually to the creation of “pseudodemocracies”, or “electoral authoritarianism”. He advances the view that “virtually all hybrid regimes in the world today are quite deliberately pseudodemocratic, in that the existence of formally democratic political institutions, such as multiparty electoral competition, masks (often, in part, to legitimate) the reality of authoritarian domination”

This paper approaches the study of democracy in Pakistan within the framework defined above. Pakistan’s recent decentralisation effort needs to be analysed within the discourse that has expanded to include more deliberative, substantive forms of democracy, and which takes into account the pressures emanating from social structures within which voters make their choices. At the same time, the paper also calls attention to the thesis that claims that many efforts within the recent wave of democratic reforms are more about the legitimisation of authoritarian regimes than an attempt to make democratic processes more deliberative or consultative.

Decentralisation is premised on the idea that by bringing governance, decision-making and provision of basic services closer to the people, government can be made more efficient and responsive based on more accurate information. The proximity between people and state can foster greater understanding and a better perception of the needs at the local level. At the same time, the closer contact promises greater transparency of decision-making processes and greater accountability of elected officials to the general populace. Most importantly, it has the potential to allow citizens to play a direct role in decision-making and implementation at the local level. Based on all of this, decentralisation has come to be characterised as a central element of the process of advancing democracy.

However, the relationship between decentralisation on one hand and participation, responsiveness and accountability on the other, or in other words, between decentralisation and the “deepening of democracy” is not that simple. It is, unfortunately, not a story of basic cause and effect. First, the phenomenon of capture of local government by local elites exists and has received much attention in recent literature on the subject. Second, social forces often work in ways that limit the impact of decentralisation and restrict people’s participation in decision-making and in civic bodies. In fact, they can also limit citizen’s access to better and greater provision of basic services. Furthermore, they can distort the theoretical perception that citizens demand as a homogenous whole in the interest of the greater, universal good. Third, leadership and patronage play a great role in determining the impact of decentralisation and in defining the relationship that exists between the state and its citizens, and can, therefore, limit or enhance the empowerment of citizens, as well as the extent of accountability and responsiveness that decentralisation brings about.

These factors combine to produce an impact of decentralisation that is very different from that envisaged by the traditional literature on decentralisation and its enhancement of participation, collective action, responsiveness and accountability, and it is this impact on democratic processes that this paper attempts to analyse.

Since LGP 2000 is only four years old, this paper bases many of its arguments on an analysis of the plan itself, rather than on its implementation. This methodology makes it possible to focus attention at this stage on the structural limitations of the plan, rather than on transitional problems. Nevertheless, this paper does draw some empirical evidence from the findings of an on-going field study, based at the Lahore University of Management Sciences (LUMS), on the impact of decentralisation in the province of Punjab. This material is used to support the analysis of social structures and their importance in electoral politics in Punjab (and by a very loose extension, in Pakistan), as well as the role they play in limiting the impact of decentralisation.

LGP 2000 has significantly extended representative democracy in Pakistan and has brought more marginalised groups into the electoral arena by introducing an elected third tier of government at the local level. However, this paper argues that this reform has not led to the deepening of substantive democracy in Pakistan because of three specific reasons. First, citizen role in decision-making is still extremely limited and that while power within the local government system resides at the district level, candidates at this level are elected not directly but indirectly by union councillors. The change, therefore, has not empowered citizens and is limited to an expansion of representative democracy. Second, and connected to the first, an analysis of the history of decentralisation in Pakistan leads to the conclusion that decentralisation has been implemented not to empower people or to make democracy more substantive, but actually to legitimise an authoritarian regime and to mask the growing centralisation of power. Finally, for decentralisation to lead to substantive democracy, it must be preceded by social reforms that aim to rectify existing social inequalities. This has not happened in Pakistan, with the result that elite capture and social determinants of electoral behaviour have significantly limited the ability of decentralisation to empower marginalized groups or to shift the old nexus of power.

Neither of these three reasons are problems of transition. The first is structural and inherent to the design of the plan. The second is inherent to the nature of the regime at the centre and the third, though social and not emanating from the plan itself, is completely ignored by LGP 2000 and thus goes unchallenged by the structure of the plan. Therefore, though it may be too soon to truly judge the impact of decentralisation itself in Pakistan, it is not too soon to examine the structure of the plan within this particular, unique socio-political context, and draw conclusions regarding its potential impact on democracy.

Local Government Plan 2000 (LGP 2000)

Local Government Plan 2000 (LGP 2000), later operationalised as the Local Government Ordinance 2001 in each province, is a dramatic and far-reaching political reform by any definition. Through a central ruling, the 200-years old colonial system of bureaucratic control over districts by the first two tiers of government was swiftly replaced by an elected third tier that connected 97 district governments through 307 tehsil (rural) and 30 town (urban) governments, to 6022 union governments in the four provinces of Pakistan. Within one year 114,418 new elected offices were created in the country at the union level, 3,264 at the tehsil level and 2,782 at the district level, of which 40,108 are reserved for women, another 36,734 are reserved for labourers and peasants, and 6,624 are reserved for religious minorities. The plan, therefore, significantly enlarged the electoral arena and allowed previously marginalized groups into the game.

The new three-tiered local government system is based on both direct and indirect elections. Union Administration (UA) represents the lowest tier, Tehsil/Town Municipal Administration (TMA) the intermediary level and District Government (DG) the highest tier of local government. A Nazim and a deputy Naib Nazim lead each tier. LGP 2000 decentralised service delivery and planning to the district, tehsil and union governments, instituted representative democracies at each level, placed the local bureaucracy under elected nazims at the district level, and created new mechanisms for citizen participation. The plan, thus, managed to empower elected officials vis-à-vis the bureaucracy, brought planning processes and service delivery mechanisms closer to the people, and increased the electoral empowerment of women, minorities and peasants. The plan has also been termed “participatory”, “bottom-up” and “people-centred” and promises to increase citizen voice in government decision-making processes.

The scope of the reform was thus extensive and it created great hype and excitement. The next three sections now analyse the design of LGP 2000, the intention behind it and the social context within which it was implemented to demonstrate why despite these revolutionary innovations, LGP 2000 cannot lead to a deepening of democracy in Pakistan.

A. LGP 2000: Advancing representative or substantive democracy?

The discourse on democracy has evolved in such a way that it is no longer enough to define democracy as electoral processes. Democracy is now defined as a concept that allows citizens a greater role in governance, specifically in decision-making processes. In Pakistan’s case, this direct citizen role in decision-making is still extremely limited.

LGP 2000 defines two specific roles for citizens. The first is as voters for offices at the union level. This has by far been LGP 2000’s most positive reform since it allowed marginalized groups to vote into office representatives from within their own communities. However, empirical evidence based on the findings of the on-going study suggests that minority, peasant, labour and women members of union councils are severely limited in their role within these councils. They are generally relegated to the backbenches and are discouraged from participating effectively by social norms based on a hierarchical caste-like biradari system, religious discrimination and patriarchy. This has progressively reduced their attendance in union council meetings, especially in the case of women, who send their husbands to participate in their place. There are even allegations that suggest that most members elected on reserved seats are proxies either for male family members or for more powerful groups within villages that have used the quotas to increase their hold over the councils.

The second role that LGP 2000 defines for citizens is in the delivery of certain village and community level services through Citizen Community Boards (CCBs). 25 per cent of the district development budget has been earmarked to fund proposals that are formulated by these voluntary, self-help citizen bodies for activities such as the improvement of service delivery within their unions and villages. CCBs function as distinct entities that are subject to accounting rules, monitoring and audits. They must take full responsibility for the implementation of their proposed projects and must also raise 20 per cent of the funds from within the community before applying for funding to the union or district. As such, their activities are separated from the planning activities of the local government and the interaction between state and citizens is limited to the acts of proposal formulation, submission, funding and monitoring. In other words, CCBs function like NGOs at the village level and require citizens to be extremely proactive with good implementational skills. CCBs are also susceptible to elite capture and to extensive misuse. In one village where a CCB has been working for over a year, most residents were unaware of the fact that it was registered as a CCB, and viewed it instead as a private venture of the person who had started it. Key respondents said that they thought the person who had initiated it had contributed 20 per cent of the funds personally in order to receive an 80 per cent contribution from the state to start a private venture.

There is also a provision for citizens to form Village or Neighbourhood Councils (VNCs), which are voluntary, self-help bodies through which citizens can become involved in service delivery and certain other municipal and welfare functions, such as the upkeep of parks and public streets and the care of the destitute. The plan also stipulates citizen involvement in various committees, such as School Management Committees (SMCs), Insaf Committees and Musalihat Anjumans. However, most of these have yet to be operationalised, and the few SMCs that were found in villages were usually made up of close friends, family and supporters of the union nazims.

Beyond this, citizens have no other active role in decision-making at the local level. LGP 2000 talks of having devolved responsibility for development planning to the local level to foster a participatory and bottom-up development planning process. However, what this means is that development planning is now done by officials at the level of the district and tehsil. It does not mean that more participatory spaces have been opened up for citizens at these or at lower levels. As far as planning and budgeting is concerned, the process is still entirely state-based and citizens only play a role in being able to apply for funds through CCBs to implement specific service-delivery projects.

In fact, even within the local government structure, the planning and budgeting process is based at the district level. The original plan provided for the planning process to occur close to the people at the level of the union, but this changed within a year of the plan’s announcement. According to LGP 2000 development plans were to be initiated at the union level where the UA would collect information through informal channels from citizen bodies at the village level. These plans would then be sent up to either the TMA, if the needs identified were municipal ones, or to the DG, if the needs were development ones. These two levels would then formulate development plans and budgets based on these recommendations, and send funds down to the UA accordingly. However, when LGP 2000 was operationalised in each province through the Local Government Ordinance 2001 (LGO 2001), this process was altered and responsibility for the preparation of the development plan and budget was given to the bureaucracy at the local level. This local administration system runs parallel to the local government system and is headed by a District Coordination Officer (DCO) who reports directly to the district nazim. Executive District Officers (EDOs) working under the DCO formulate plans, which are then sent to the nazim for presentation to the district councils for approval (NRB, 2001). Though these councils are made up largely of the direct representatives of people – the union nazims – the council is mandated to only discuss the plan and budget. It cannot vote on it.

This last point, in fact, leaves even the representative nature of the decentralisation effort open to questioning. Decisive power over development plans and finances is based within the extremely powerful office of the district nazim. Both the DCO (executive) and the district council (legislative) are significantly less powerful in what they can manage to do at the district level. Yet, the nazim is indirectly elected by union councillors while the district council is made up of union nazims, all of whom are directly elected. Citizens are, therefore, delinked from the office where power resides within the district, and they can access the nazim only indirectly through their union nazims. Field surveys support this argument in that there was no evidence in any of the three unions that citizens had ever had direct contact with the district nazim. In fact, many did not even know what his name was.

This unique structure of representation leads to two specific political trends. The first is the possibility (confirmed by some key respondent interviews) that Pakistan’s well-known and well-established tradition of patronage-based politics is being replicated within district councils, so that the ability of a union nazim to “win” extra funds for his/her union council (beyond the non-discretionary district-union fiscal transfers) is directly dependent on his/her ability to develop a strong, mutually beneficial relationship with the district nazim. The second impact is that the nazim’s accountability to his/her district constituents may be very limited. The district nazim’s election depends not on the votes of the majority of the constituents, but rather on a majority of a very limited pool of union councillors. At one level this means that district nazims may not command a majority of the public vote in that constituency and, therefore, may not represent majority interest, and therefore, may be unresponsive to the expressed needs of his/her constituents. At another level it means that it is very easy for candidates to influence the decisions of union councillors. Given the limited number of voters in this case (a rough average of about 1200 councillors per district) it is easier to promise favours, buy support and employ all sorts of other tactics to distort election results. It also heightens factional competition at the union level. In the case of one union studied, a councillor had managed to create a faction of 13 councillors who voted for the incumbent district nazim. The councillor was rewarded for this by the district directly sanctioning various development schemes for the village. This has weakened the relationship between this group of councillors and the union nazim, who had voted for the opposing candidate in the district elections, which had in turn led to stalemates over various union level decisions and development schemes, and had rendered the union government ineffective.

Therefore, while LGP 2000 has extended representative democracy in Pakistan, its design suggests that it will not be able to deepen it because there are few spaces for citizen participation, and because while the directly elected offices at the union level enjoy little substantial power, decision-making occurs within the indirectly elected, and therefore potentially unresponsive, office of the district nazim.

B. LGP 2000: Electoral Authoritarianism?

An analysis of the history of decentralisation in Pakistan suggests that the limitations on substantive democracy imposed by the structure of LGP 2000 may not be benign omissions, but in fact part of a process that seeks to legitimise a military regime and the growing centralisation of power at the centre. LGP 2000 is not Pakistan’s first encounter with decentralisation. Both Generals Ayub and Zia had implemented decentralisation during their military regimes in 1959 and 1979. Both these efforts failed to empower citizens or to establish substantive democracy at the local level, and the current plan bears enough resemblance to these efforts to lead to a contention that LGP 2000 may not have been designed to deepen democracy either.

Decentralisation in Pakistan has always been a top-down imposition. Under Ayub’s Basic Democracies system, it was instituted to create a loyal cadre of local level councillors who formed the electoral college for the election of the President. The 1959 decentralisation reform was complemented by the constitution of 1962, which centralised power in the office of the President and his nominated provincial governors. The same period also saw various interventions against political parties at the national and provincial levels. Most importantly, the bureaucracy was not reformed, and the most powerful office at the local level was that of the Deputy Commissioner (DC), who maintained significant control over locally elected officials. The bureaucracy could nominate some union council members and retained the office of the local government Chairman, as well as the power to “quash proceedings of the councils, suspend elected and official members of the councils, suspend resolutions and supersede the decisions and orders of the councils and even, in extreme cases, take over the functions of the bodies”. Therefore, despite the implementation of decentralisation, power remained, and in fact was consolidated, at the centre.

General Zia’s 1979 decentralisation reform continued this pattern of centralisation. While decentralisation was instituted through the Local Government Ordinance, the 8th constitutional amendment strengthened the office of the President at the same time against the elected federal and provincial tiers. The deliberate weakening of political parties continued and both the local government elections and the 1985 national and provincial elections were contested on a non-party basis, which led to the strengthening of personalised politics in Pakistan. Though Zia’s reform sought to reduce bureaucratic control over local government and stipulated the direct election of all local government offices, including that of the Chairman, local governments were restricted to the delivery of municipal services only. Therefore, it was far from being a substantive attempt at the devolution of power and the entire period was characterised instead by the strengthening of the power of the presidential office.

Both Ayub and Zia’s efforts had “resulted in the institutionalisation of clientelist, personalised politics, had disempowered the provincial elected tier, limited the power of elected local governments and placed ‘control’ in the hands of the bureaucracy, while maintaining the centralised nature of the state at the federal level”. Both regimes also did not accord the third tier of government constitutional recognition, and failed to define responsibilities clearly between the provincial and local levels of government, and between the bureaucracy and these levels. This lack of definition led to a system of governance that was characterised by competition and tension between each level and a continuous, hostile struggle at each level to retain some power. Not surprisingly, this tension contributed further to the consolidation of power at the centre.

It is interesting to analyse Musharraf’s LGP 2000 within this historical context because it allows attention to be focused on the innovative departures this plan made from the previous efforts, while at the same time allowing an analysis of the continuities that have been maintained. Looking first at the departures, it is easy to see that this is the first time that a decentralisation effort in Pakistan has been focused, at least in rhetoric, on the empowerment of citizens, and on bottom-up processes of participatory planning. The plan claims to be “people-centred” and to have placed power in the hands of people at the local level. This is borne out to an extent by the greater user role in service delivery through CCBs and by the increased reservation of seats for women, labour, peasants and minorities. LGP 2000 is also characterised by substantial decentralisation of service delivery functions from the provincial to local level and by the empowerment of elected local governments vis-à-vis the bureaucracy. For the first time in Pakistan’s history, the bureaucracy at the local level has been made subservient to an elected local office. Also, service delivery functions have been clearly separated between the province, district and tehsil for greater functional clarity and efficiency, leading potentially to better service delivery. Another departure is the vertical linkage created between the union, tehsil and district level of local government through the stipulation that union nazims constitute the district council while union naib nazims make up the tehsil council. The plan has also abolished the traditional rural-urban governance divide at the tehsil level in an attempt to allow urban revenue to flow more easily to rural areas through tehsil development plans and budgets.

However, at the same time, some important features of the plan have remained true to the decentralisation designs of the past. Most importantly, the third tier of government still has no formal constitutional recognition and is still dependent on the survival of the regime at the centre. Equally important is the fact that while many provincial functions have been devolved to the district and tehsil, no federal function has been devolved downwards. This means that the provincial tier has been significantly weakened vis-à-vis the centre. At the same time, interventions against political parties, such as the selective disqualification of politicians based on new educational criteria and the registration of new corruption cases, and the weakening of old political parties through political manipulation and re-engineering, have weakened the political sphere and concentrated power in the hands of the President and his newly created party. Elections at the local level have continued to be held on a non-party basis, which has strengthened personalised politics and weakened party structures at the local level. The weakening of the national and provincial elected tiers has led to heightened tension between these and the local tier, so that there is little political ownership of the new plan at these higher levels. At the same time, these tensions have contributed to strengthening the centre, not least because it has created the need for strengthened patronage links between the presidential office and each of the three elected tiers. Though the close link between local government and the presidential office is no longer as blatant as it was under Ayub when he designed it to be his electoral college, the perception within this tier that it owes its existence and power to the president is still very strong. These strong patronage links are exemplified by the 2003 mass resignation of all 24 district nazims in NWFP in protest against the MMA-led provincial government’s extensive interference in the workings of the local tier, and its reluctance to devolve certain functions downwards to the districts. The nazims appealed directly to the President, who intervened on their behalf and had all 24 restored to their offices.

These continuities reveal that once again, decentralisation has been used as a “complementary institutional change in a non-representative institution’s bid for centralization of political power”. One of the most important continuities is the weakening of party structures and the lack of a space within which political collectivities can form and develop. The significance of this becomes most evident when we take into account the unique social structures of rural Punjab and their impact on electoral politics, and by extension, on decentralisation itself.

C. Democracy and Primordialism

A central aim of democracy is to promote equality between citizens, especially in terms of their access to public goods and services. Based on this, decentralisation is supposed to advance democracy by bringing decision-making processes and service delivery mechanisms closer to people, under the assumption that this proximity will allow people greater and more equitable access to services. However, based on the findings of the on-going research study, this section argues that decentralisation has led to (a) elite capture at the local level, and (b) has not resulted in greater equity in service provision. The study contends that both these findings are a result of the fact that electoral behaviour in Punjab, and by extension in Pakistan, continues to be governed by primordial identities.

Elite capture has occurred at the local level because, like the previous efforts at decentralisation, LGP 2000 has not questioned or challenged the social context within which it was implemented. This means that there has been no attempt to address or change the deeply entrenched social inequalities that exist in Pakistan, which in rural areas are based to a large extent on unequal land ownership and on a biradari-based (clan-based) system of social organisation. Electoral behaviour in Pakistan is heavily influenced by biradari-ism and by the hold that large landowners have on village level politics. Even in Punjab, where it is believed that the social hold of large feudal-type landlords has been broken through land reforms and demographic changes over the decades, the biradari system remains strong and has a great influence on patterns of voting and leadership. The argument that this section advances, therefore, is that since severe social inequalities exist, no preceding social structural reform has been effective and the plan itself leaves these social structures unchallenged, decentralisation has merely led to elite capture at the local level.

Indeed, Pakistan’s four-year experience with decentralisation shows that the effort has increased the power of influential families by providing them with more seats to capture than they had the opportunity to previously. It is not uncommon to find that the MNA, MPA and nazim of a district are directly related, and that many seats below that level are also filled by relatives or loyal friends. With the decentralisation of many service delivery and planning functions to the local level, LGP 2000 has strengthened the old local power blocks and networks even further. The findings of various studies verify this contention. Keefer, et al, estimate that up to 70 per cent of seats at the local level have been won by the “rural gentry”. Bari and Khan (2001) point out that a majority of elected nazims own land in access of 25 acres and that “there is a hierarchy and clear pattern of land ownership. The higher you are in echelons of seats starting from reserved seats and going up to nazims, the more land you own” and that there is a positive correlation between the size of land ownership and the probability of success in the District nazim’s elections. Government-based reports of the local government election results claimed that a lot of new political blood had been thrown up by the local elections. The studies quoted above have found that this is true only at the level of the relatively powerless union councillors. As far as the powerful office of the district nazim is concerned, “30% of district nazims in Punjab were former MNAs or MPAs, and approximately 90% belonged to established political families”. The same results are evident at the union level, where most union nazims and naib nazims are old influentials, such as the numberdar, the old union chairman, or small local landlords. Even at this level LGP 2000 has not offered villagers a viable new channel for interest articulation and representation, but has simply offered the old channel a new name, that of the union nazim.

This impact of decentralisation can be blamed partly on the fact that people continue to choose leaders based on biradari alliances and loyalties, which dictate certain social hierarchies and thus perpetuate the social power of the local elite. Biradari-ism is also strengthened and encouraged by the fact that the new system of local governance is based on party-less political competition. This has serious implications because “political parties create spaces for participation by less influential individuals that are supported by collective ideologies and collective mobilisational efforts. In the absence of such parties it is usually powerful and resourceful local elite that are able to campaign for electoral seats”. Basically, in the absence of alternative forms of collectivisation, biradaris remain the most logical and coherent method of social organisation, and a central facet of the biradari system is the creation and maintenance of social hierarchies. Also, in the absence of parties, leaders and electoral candidates are identified by their biradaris, and votes are cast along biradari lines. In the villages surveyed, between 70 and 95 per cent of residents are a part of voting blocks that are formed on the basis of either a biradari or a biradari-based alliance, which then support candidates on the basis of their biradari or their social power (which also usually emanates from being a member of the dominant biradari). It is for this reason that this paper argues that LGP 2000 has made no attempt to challenge the social system or to make it more democratic. Democracy in Pakistan was always limited in its impact by clientelist, personalised politics and weak party structures. The fore-going analysis concludes that this has not changed through decentralisation. In fact, LGP 2000 has reinforced the old clientelist hierarchies.

The biradari system is also responsible for the continuing inequity in service provision. Gazdar (2003) finds that horizontal and vertical caste and class-based clientelist networks work to exclude the poor from access to key social assets and public services in Punjab and rural Sindh. The field study finds that this has not changed after decentralisation. Service provision remains inequitable and highly targeted in favour of the union nazim’s own village and biradari, and that opposition villages and biradaris are marginalized more after decentralisation than they were before. Though provision itself (especially of key services such as sanitation and street soling) has increased overall since decentralisation, it has targeted the more dominant biradaris and any groups or villages that aligned themselves with the nazim. In fact, an analysis of the findings revealed that those most in need of the services (i.e., groups that ranked the service as one of their top three needs) were least likely to be provided. The main explanation for this is that most dominant groups have had some level of access to these services in the past and therefore are not in dire need of them, while the previously marginalized “low-caste” communities that never had access are still being marginalized, and are, therefore, in desperate need of services like soling and sewerage systems. This shows that not only has service provision not become more equitable after decentralisation, but that this greater proximity between decision-makers and citizens has not increased demand responsiveness.

At the same time, however, there is evidence to support Wilder’s (1999) claim that electoral competition based on developmental needs exists in Pakistan and defines factional politics. Many of the groups that align with the nazim do so after careful consideration of whether this support will give them greater access to services. However, such considerations come into play only after the imperatives of biradari-based support have been satisfied. Factionalism exists but is a subset of biradari-based competition, in that it is defined by biradari-based alliances and is strongest in cases where two candidates are from the same biradari. This shows that biradari-based considerations are still the most important determinants of electoral behaviour, support and organisation.

This primacy of biradari-ism comes at the cost of universal collective action and creates factionalised interest articulation and extremely low levels of associational activity within the villages and communities studied. People are deeply divided along biradari lines, lead very separate lives and rarely cooperate over development issues even at the village level. Within such a scenario, to have eliminated parties and to have clumped a number of villages and biradaris within the same union (some areas of Punjab have as many as 15 villages within the same union), not only demonstrates LGP 2000’s disregard for the impact of these social structures, but also provides one of the most powerful explanations for its weak outcomes. It has not created new channels of interest articulation or new opportunities for civic engagement and community organisation. No village level meetings are mandated and other than the council at the union level, which is open only to elected councillors, villages have no opportunity or forum to discuss development needs and priorities. The result is that even after the implementation of LGP 2000, most villagers continue to name the pre-decentralisation village influential as their main channel of interest articulation.

The basic point that this section makes is that LGP 2000 has not resulted in more equitable service provision, which remains highly targeted and linked to electoral patronage. Combined with continuing social inequality and elite capture, such targeting has resulted in the further marginalisation of already marginalized groups. Social inequalities and the primary role of biradaris in determining voting behaviour has not allowed meaningful and effective democracy to take root or develop, and therefore, any discussion concerning a given reform’s impact on democracy must then concentrate on whether that reform challenged and affected these inequalities. LGP 2000 clearly chooses not to.

Conclusion and Recommendations

Decentralisation has had a positive impact on democracy in Pakistan in that it has replaced the old local administration with local governments, introduced thousands of new electoral seats and increased the provision of services at the local level. However, while it has significantly expanded representative democracy, it has not had much success in deepening substantive democracy in Pakistan. This is because it has not provided effective opportunities for participation by citizens in decisions that affect their lives, because the intention behind its implementation was the weakening of political collectivities and the greater centralisation of power at the centre, and because it has not challenged the social order that creates and maintains social hierarchies and thus marginalizes the poor because services are distributed not universally but in return for electoral support, which still follows the support patterns defined by biradari-ism..

A number of structural reforms can be instituted to improve the impact of decentralisation and allow it to truly empower citizens. First, and most obviously, elections at the local level should be based on political party competition, which will provide a system of social organisation capable of countering biradari-ism in rural areas. Second, elections should be based on the village rather than the union as the lowest level of local government, and villages should be single-ward constituencies, so that a union council would have the same number of councillors as villages in the union. This is recommended because there are many villages in Punjab that have been marginalized completely since 2001 because of the fact that they did not manage to elect even one councillor to the union council and this has severely impacted their developmental prospects. Also, from the perspective of a village, a union, especially one with 10-15 villages, is an extremely large unit and the district is a distant and abstract reality. Political competition is still beyond comprehension for most village residents, so that they readily fall back on the most obvious, coherent and intelligible unit of organisation – the biradari.

Third, tehsil and district nazims and naib nazim should be elected through direct elections. Fourth, there should be a reservation of seats for women, low-income groups and minorities not just at the councillor level but also at the level of the union nazim and naib nazim. Fifth, and very importantly, there should be a legal requirement for tehsil and district officials to organise mandatory, open and regular planning meetings at the level of the village in which villagers should be given an opportunity to prioritise their needs and plan the development process of their village. All of these are recommended because decentralisation will strengthen democracy in Pakistan only when it can provide an alternative to the current social inequalities and hierarchies and provide new channels for effective interest articulation and collective action.

Bibliography