Explaining municipal governance in Kosovo: local agency, credibility and party patronage

ABSTRACT

What can explain the varied effectiveness of internationally led attempts at statebuilding? This article seeks to answer this question by comparing the contrasting trajectories of governance in two municipalities in Kosovo: Hani i Elezit/Elez Han and Kamenica. In Hani i Elezit, evidence suggests that effective and accountable governance is embedded. However, in Kamenica informal and clientelist practices persist and residents are less satisfied with the municipality’s performance. As the nature and extent of internationally led statebuilding has been similar in both municipalities, explaining variation requires an analytical shift to how local leadership interacts with, receives and ultimately shapes statebuilding processes. The data are based on fieldwork from 2012 to 2015. The article focuses on two critical dimensions of statebuilding: capacity building and social accountability. It argues that the impact of externally led statebuilding strategies depends on the orientation of the political leadership of the municipalities. The article identifies features of the political environment, namely credibility and the organization of political parties, which constrain the kind of public-oriented leadership necessary for effective and accountable governance.

KEYWORDS:

• Statebuilding
• clientelism
• patronage
• credibility
• local governance, Kosovo

Introduction

Internationally led statebuilding has become a policy marked by vast scope and complexity. Defined as a set of actions to establish or reconstruct effective and autonomous state institutions where these have been seriously eroded or are missing (Caplan 2006, 3), attempts at internationally led statebuilding has meant that domestic institutions have been increasingly shaped and re-cast by international and regional organizations, individual donors, or transnational NGOs. As Chandler and Sisk note, ‘statebuilding historically was for most part quite endogenous’ but now ‘in engaging to build states through military deployments, civilian capacity building, and development aid flows, international actors have changed the nature of contemporary statebuilding’ (2013, xxii). Still, international actors rarely operate alone but rather depend upon a complex chain of delegation and co-option while being assisted by numerous local partners (Mac Ginty 2011).

Much of the existing literature tends to emphasize how this kind of statebuilding rarely produces the expected results. Belloni points out that, although progress is often projected through international reports and evaluations, emerging governance structures tend to be ‘only superficially democratic, effective and accountable and are perceived as illegitimate, constraining and unsuccessful by those experiencing them’ (Belloni 2012, 1). Accordingly, the dominant outcome is that state institutions ‘tend to adopt a ‘western’ form under the influence of internationalized norms, but keep functioning according to other social logics’ (Bliesemann de Guevara 2010, 115). So ‘what evolves often has the appearance of democracy because local elites are able to easily hoodwink gullible international observers by applying the nomenclature and physical facades of the new. But behind such facades persist the very ways such reformation is designed to eliminate’ (Roberts 2013, 95).

Such conclusions, however, tend to conceal variations in the outcomes of externally led statebuilding processes. Effective statebuilding can manifest across sectors, geographical regions or levels of government, whether this be police reform in Sierra Leone (Albrecht 2010), customs management in Kosovo (Skendaj 2011) or democratic governance in the municipality of Brčko in Bosnia-Herzegovina (Bieber 2005). These cases of relative success demonstrate that we need to accumulate empirical evidence to explain when statebuilding is effective, when it is not, and why. This article investigates the evolution of effective statebuilding at the municipal level in post-war Kosovo, a site of massive and intense attempts at internationally led statebuilding. I focus on two important dimensions of the efforts to build effective and accountable governance: building up social accountability and building up municipal capacity to deliver public services.

Empirically, I compare the contrasting trajectories of two municipalities: Hani i Elezit/ Elez Han (referred to here as Hani i Elezit) and Kamenica. All the evidence suggests that effective and accountable governance has been embedded in Hani i Elezit, whereas in Kamenica, informal and clientelist practices persist and residents are far less satisfied with the performance of their municipality. Based on this comparison, I argue that, as the nature and extent of statebuilding has been similar in both municipalities, other factors explain why statebuilding has played out differently in both cases. Analytically, I shift away from perspectives that prioritize international agency to assess how local leadership, mainly embodied in the municipal Mayors, interacts with and shapes external efforts of statebuilding. I demonstrate that debates about sequencing (Paris and Sisk 2009) – what should come first: building up the capacity of the state or building bottom-up accountability – are somewhat secondary because these statebuilding strategies are shown to have no independent effect. Rather, they only have impact if, as in Hani i Elezit, these measures are given impetus by local leadership on the ground. When international strategies diverge with local political agency, the effectiveness of statebuilding strategies is likely to be constrained (Beysoylu 2018; Selenica 2018; Tadic and Elbasani 2018; Triantafyllou 2018; Troncota 2018).

This research contributes to the now prominent literature on the importance of local factors in determining outcomes in statebuilding, which has helpfully turned attention to how local actors react to and influence statebuilding (Beysoylu 2018; Elbasani 2018; Kursani 2018; Mac Ginty 2010, 2011; Phillipps 2018; Selenica 2018; Triantafyllou 2018; Troncota 2018). The research demonstrates the importance of one particular type of local agency: political leadership. I also explain why different forms of local agency have emerged in Kosovo’s municipalities. The article is organized in three sections. Section I elaborates on the concept of effective statebuilding and outlines the research design. Section II traces the impact of the orientations of the local political leadership on capacity building and social accountability at the municipality level. Section III further interrogates local leadership and identifies features of the political environment which may act as constraints to statebuilding: credibility and party structure.

Contrasting Hani i Elezit and Kamenica

Statebuilding in Kosovo’s municipalities

In many respects, Kosovo represents a crucial case for the efficacy of attempts at internationally led statebuilding, as the level of aid directed towards various aspects of governance has been unprecedented. Since 2009, the amount of aid allocated to governance as a proportion of all aid flows into Kosovo has mounted to around 60–80 per cent; in the other countries of the region it has not reached above 20 per cent. Yet, it is the scale of resources that has been most striking: in 2009, for instance, the international community dedicated $345 per person to governance sectors. This amount towers over the aid spent on statebuilding activities in other high-profile efforts, for example, in Afghanistan ($62 per capita) and Iraq ($41 per capita), and far surpasses that allocated to other countries in the Southeast European region (after Kosovo, Bosnia-Herzegovina receives the second highest at $44) (OECD 2014).

Establishing effective and accountable local governance has been a core element of this massive effort at statebuilding. From the very start, reforming municipal government was an important assignment for the UNMIK administration, which sought to effect an early transitioning away from the Yugoslav system of self-management towards a more liberal-democratic system (Cocozzelli 2010, 141). In April 2004, the UN Security Council issued a statement calling for ‘more effective local government through the devolution of central non-reserved responsibilities to local authorities and communities in Kosovo’ (OSCE 2010). Later, the 2007 ‘Ahtisaari proposal’ intensified the development of municipal governance by instituting a far-reaching de-centralization of administrative competence to the municipal level, including on sectors such as education, health care, economic development, and urban or rural planning.

This shift of administrative competencies to the municipal level has been accompanied by a clear idea of the form of governance that should emerge: democratic, rule-based and grounded in citizen participation. Such normative standards have been inspired by the European Charter on Local Self-Government, a multinational legal instrument that defines the principles of local autonomy and aims to protect citizens’ rights to effectively participate in the making of decisions which affect their everyday environment (OSCE 2010). To this end, the Law on Local Self-Government has accorded significant weight to transparency and accountability procedures, as well as mandating municipalities to adopt participatory structures in decision-making, such as petitions, citizen committees or representation by assembly members (Republic of Kosovo 2008).

In 2009, the largest projects within the public sector policy and administrative management category were directed at the municipal level, as is the case with much of the aid dedicated to governance in Kosovo (OECD 2014). Those 15 years of internationally led statebuilding has generally been directed towards the two strands of activity that are the focus of this article: (1) capacity building, which entails the provision of technical assistance to the municipality and (2) social accountability directed at fostering accountability within the society.

Despite these efforts, evidence suggests that effective and accountable governance remains elusive in Kamenica municipality, which is located in the east of Kosovo. Over 80 per cent of those surveyed in Kamenica believe that only people with strong political connections have their voices heard, suggesting that informal connections are important for accessing public goods. In addition, only 32 per cent of those surveyed in Kamenica agreed that people are, in general, treated equally by their political leaders, implying integrity-undermining practices, like string pulling, are common (Jackson 2014, 2018).

If Kamenica municipality is broadly representative of a general trend of citizens’ experience of governance in Kosovo, then Hani i Elezit municipality, located by the south-eastern border with Macedonia, represents an outlier case, where the majority of people feel they have equal access and are treated equally by the local authorities. These differences are confirmed by a 2012 UNDP survey (summarized below in Table 1), where Kamenica proved to be mainly below the national average for key indicators of municipal governance, while Hani i Elezit was way above average, again confirming its outlier status.

Table 1. Effective and accountable governance compared

Municipal capacity indicators	Kamenica	Hani i Elezit	Kosovo average
Citizen trust in competence of municipality to solve local problems (%)	22	83	65
General satisfaction with the work of the municipality (%)	54	93	69
Citizen considerations taken into account a lot or to some extent by municipal officials (%)	52	80	55
Respondents very or somewhat informed on the work of the municipality (%)	42	84	49

Source: Elezaj, Duri, and Haskuka (2012).

Controlling for alternative explanations

Two key potentially confounding conditions are controlled for. First, the key variables of statebuilding – the type and extent of the resources, personal, expertise, training, programmes and projects dedicated to statebuilding – are controlled for as both municipalities have been subject to similar the capacity building and social accountability measures. Since 2006, both states have been beneficiaries of major interventions at the municipal level undertaken by the largest donor organizations that distributed aid according to the population size of the municipalities. As an illustration, the European Commission’s two-year project titled ‘Support to Local Government’ has worked in both municipalities in improving the functionality and transparency of the municipal departments and the financial management of the municipalities. The OSCE’s ‘Local Governance Programme’ has provided training and capacity building in both municipalities and the ‘Effective Municipalities Initiative’ Programme supported by USAID has directed significant inputs into both municipalities. Projects undertaken by international donors aimed at increasing citizen participation and raising awareness on democratic issues have also been common, such as the Swiss-funded ‘LOGOS’ project.

Another alternative explanation is difference in socio-economic characteristics, which scholars have identified as shaping ‘bottom-up’ pressures for patronage and clientelism (Kitschelt and Wilkinson 2007). Some economic features do vary. For example, it could be that one of Kosovo’s first modern industrial plants, a cement factory from the 1930s, has provided more economic security to Hani i Elezit, in contrast to Kamenica, which has less large-scale industry. However, there is no evidence that this source of employment has led to more economic security in the municipality. In fact, the citizens of Hani i Elezit report marginally more economic insecurity, with the percentage of households that cannot afford basic goods and services (e.g., public utilities, medicine, a meal with meat, clothes) higher than Kosovo’s average, whereas in Kamenica it is generally lower than the national average (Elezaj, Duri, and Haskuka 2012).

It could be that Kamenica’s small-scale, rural– and thus less collectivized – economy makes it more difficult to mobilize demands for public goods. However, while Kamenica certainly did have more of an agrarian economy, since 1999 there has been considerable ‘urbanisation’ in Kamenica. It is worth noting that Kamenica has had an industrial sector, mostly revolving around pottery. Employment patterns are also comparable: though more people are looking for work in Kamenica, the employment rate is slightly higher than in Hani i Elezit.

It has also been argued that ethnic cleavages will increase patronage by increasing the salience of targeted goods (Kitschelt and Wilkinson 2007). While it is the case that Kamenica has a small non-Albanian minority, at under 10 per cent of the population the relative size of this has not been considered significant enough to represent a confounding factor. As the data in Table 2 summarize, socio-economic patterns do not vary a great deal across the municipalities, implying that socio-economic characteristics do not represent an alternative locus of explanation.

Table 2. Background socio-economic characteristics

Key characteristics	Hani i Elezit	Kamenica
Population	10, 000	36,000
Size	83 km^2	423 km^2
Population ethnic Albanian (%)	99	90
Estimated unemployed (%)	23	30
Estimated employed (%)	26	28
Economic well-being: cannot afford to pay for public utilities (%)	58	38
Primary education only (%)	50	44

Sources: (Elezaj, Duri, and Haskuka 2012; OSCE 2014, 2015).

Since the socio-economic structure and the nature and extent of statebuilding have been similar in the two municipalities, we should look for alternative actors and processes shaping the outcome of international statebuilding in both cases. Such an assumption requires a shift away from statebuilding perspectives that prioritize international agency – the kind of strategies that are used (Paris and Sisk 2009); how many resources are deployed (Dobbins 2007; Zuercher 2006); what kind of mandate statebuilders may have (Caplan 2006); or how they are organized (Holohan 2005). Specifically, I focus on local agency and leadership to assess what happens on the ground: how local actors interact with and shape the effects of internationally led attempts at statebuilding. While the ‘local turn’ has been a prominent feature of the statebuilding literature in recent years, helpfully turning attention to how local actors react to and influence statebuilding (Mac Ginty 2010, 10), more research is required on how and why specific forms of local agency affect statebuilding on the ground.

I trace how capacity building and social accountability has been facilitated or limited by local leadership. As such, the article draws on Migdal’s model of policy implementation articulated nearly three decades ago (Migdal 1988). Migdal emphasized that in many countries, state policies are not just fluently applied on the ground but rather are channelled through ‘implementers’, people at the local level responsible for the implementation of state policies, who have leeway in deciding which aspects of policy are realized (Migdal 1988, 200). Implementers respond not to the overarching authority of the state but to the incentives and pressures they face at the local level. The differing responses to local incentives and pressures explain the fragmentation and inconsistences in outcomes at the local level. The research investigates both how municipal leadership ‘implements’, but also examines those local pressures and incentives that determine this response.

Data collection and analysis

Empirical research into Kosovo’s municipalities is scarce, therefore this research relied mostly on primary data. During 2012 and 2013, I made repeated visits and stays to become acquainted with the social and political circumstances in both municipalities. Additional research was also conducted in 2015. I interviewed different actors: past and present Mayors, local officials, key informants and civil society groups. Additionally, I conducted long, semi-structured interviews with 15 randomly selected citizens in each municipality. I conducted an additional shorter survey of the relevance of clientelist practices with over 80 participants in each municipality. Finally, I consulted political analysts, international diplomats and NGO employees working on the ground.

Understanding why international agency has played out differently in Kamenica and Hani i Elezit requires analysing how the mechanisms of capacity building and social accountability have related to the different outcomes. The analysis utilizes the process tracing method to unpack and assess the different parts and steps of this relationship (van Evera 1997). At each analytical stage, the evidence was examined for patterns, triangulated with secondary data, and assessed in light of countervailing explanations, to ensure reliability. The findings of the research were also triangulated through research into other municipalities.

Assessing different dimensions of international attempts of statebuilding

Capacity building

Capacity building has been a ‘generalised tonic prescribed for municipalities in poor health’ (Grindle 2007, 106) and is expected to work through developing certain capabilities – human, technological, infrastructural – that facilitate the implementation of new processes. Yet when tracing how those capabilities operate on the ground, one can more often than not notice disruption and deviation of projects, trainings, technology and infrastructure from their intended aim. Therefore, capacity building often depends on the underlying orientation of political leaders, how it is received on the ground and to what effect. When a clientelist agenda interferes in the process, as is the case in Kamenica, the effect of capacity building is neutralized and deviated to other ends. In Hani i Elezit, however, capacity building initiatives have had a stronger effect, precisely because they have dovetailed with the political leader’s general orientation and commitment to the public good.

Consider attempts to develop capacity to implement participatory budgeting mechanisms in Kamenica undertaken by a leading international donor in 2011. After diagnosing that the municipality lacked the capacity to develop an engaging mechanism for citizens to discuss the budget, the donor invested in communication tools and training for staff, as well as organizational resources. A report by the donor lauded the programme a success based on the increase in the number of people who attended these sessions, as well as the size of the budget allocated to this participatory mechanism. Insiders working on the project, however, revealed that while participation increased, this was because the public hearings were subject to ‘client packing’: ‘At first, we thought it was a success then after the meeting we realized that everyone there was placed there by the political party. The Mayor used his party contacts to fill the room just so they could tick the boxes and then everyone was happy’ (NGO director Pristina 2015). Mediated through the orientation of the Mayor, the effect of this capacity building project was mitigated because the communication tools and training were used to mobilize the Mayor’s political supporters rather than ensuring that the new capacity is deployed to create a genuinely open and public meeting.

The same undermining of capacity building efforts, especially with regard to human capital and technological development, has occurred because of the type of organizational culture cultivated in Kamenica by political leaders. Having worked closely with successive Mayors in Kamenica, a leading expert and director of a municipal reform programme in Kosovo complained that information sharing has been very poor, there has been a deferential and controlling culture, and motivation amongst some staff has been extremely low (USAID project director 2012). Due to this poor sense of public mission, new technological inputs, such as computers, have not been used to their fullest potential and have not been helpful in terms of improving how the municipality operates. Many staff members in Kamenica have also undergone trainings via seminars implemented by international organizations or specialist CSOs. Yet because for most staff there is no incentive to work efficiently and towards targets, as rewards and promotion are not configured according to clear and objective standards, attending training is considered to be more of a ‘break’ from work rather than a chance to develop personal capacity (USAID project director 2012).

In general, capacity building inputs have not been channelled effectively in Kamenica. In contrast, in Hani i Elezit new technology and training have had an important impact because it has converged with the political agenda of the leading authorities, especially the Mayor, Rufki Suma, in office since 2009, has strived to improve the efficiency and integrity of how the municipality operates. The cultivation of a ‘public service’ culture, which stands in stark contrast to the clientelist-induced dysfunction and inertia of the other municipalities has been crucial to these shifts. In a series of interviews with the Mayor, he cited a number of actions which have used inputs by the international community to advance statebuilding in ways that often departed from the normal statebuilding script.

Using technology has been an important aspect of his management style. The Mayor is able to monitor late arrivals to work through an electronic check in and check out system, paid for by donor funds. Every municipal staff worker who arrives to work after 8 am has been sent home without pay; as a result, there are now generally no late arrivals. Cameras installed throughout the municipality have not only allowed the Mayor to castigate those he deems to have spent too long on a coffee break, but also to assess whether public work programmes have been running efficiently. According to the standards of international statebuilders, this may be an uncomfortable level of surveillance. However for the Mayor, this is merely a way of checking on the fidelity of public officials (Mayor Hani i Elezit 2015).

While municipal staff have attended many trainings provided by the international community, the Mayor has harnessed these developments in human capital to a script of organizational standards that are rather different from those advocated by the international community. This has shaped the organizational culture of the municipality in the spirit of the TMK, the Kosovan Defence Force (successor to the Kosovo Liberation Army (KLA): ‘it is about getting the job done for the community’ he tells me (Mayor Hani i Elezit 2015). As the director of public and emergency services described: ‘the work culture is very strong a good ethic, it is very open and we all want to do a good job for the municipality’ (Municipal director Hani i Elezit 2015). Dismantling the ‘privilege culture’ is another strong signal which has shifted expectations about how public officials behave. Citizens in other municipalities cited the culture of privileges, especially expenses, as an indicator of the lack of public service. Mayor Suma has struck a blow to the privilege culture in Hani i Elezit by taking resources for expenses from municipal officials and assembly members and transferring it to public good. The money has gone to funding mobile phone credit for teachers so that they can communicate better with each other, as well as with parents of students – again, this builds on the initial purchase of the phones through donor funds. This kind of policy was not enacted in any of the other municipalities I studied.

Additional policies nurturing trust in the municipality have centred on entrenching universal entitlements. Hani i Elezit is the only municipality in Kosovo in which the mother of every new born child receives 100 euros, and the only municipality in which the funeral expenses of citizens are covered by the municipality. For a country with the scant resources of Kosovo, this has been a radical set of ‘cradle to grave’ policies – yet this has been only made possible via the winning of additional funds for infrastructure projects from the international community (Mayor Hani i Elezit 2015). So, while the international community focus on technical processes of transparency and accountability, the Mayor has, rather uniquely, used international funding to push forward in convincing citizens that the formal state institutions care for all citizens. In a similar vein, on entering the municipality on his first day, the walls were covered with flags, posters and pictures of the two dominant parties, all of which had conveyed a sense that when residents entered the municipality building they weren’t entering the public domain, but a highly politicized realm carved up by political parties. The first task for the Mayor’s staff was to remove all these symbols from their offices. As the Suma explained ‘we needed to show that the municipality is dedicated to citizens and not to political parties’ (Mayor Hani i Elezit 2013a). Interestingly, the activities underpinning this transformational leadership are not in the repertoire of international statebuilding, this raises the importance of ‘local statebuilding’ and of local ingenuity in the process.

Social accountability

Drawing on research into social capital and associational life, social accountability strategies are based on the idea that engaged citizens, municipal assemblies and civil society organizations (CSOs) will exert a strong pressure on political leaders to implement the statebuilding model (Fukuyama 2014; Mungiu-Pippidi 2015; Peruzzotti and Smulovitz 2006). Tracing the mechanisms of social accountability, however, one can see that this strategy cannot be assumed to be a straightforward or automatic boon for statebuilding. The research shows that effectiveness depends on the right organizational, institutional and social environment, which cannot itself be shaped by citizens or CSOs, but is ultimately reliant on the political leaders of the municipality. This section will argue that the effect of social accountability interventions is dependent on the degree to which political leaders at the municipal level are willing to provide a favourable environment for citizen participation, civil society and municipal assemblies. The impact of social accountability interventions in Kamenica have been blunted because political leaders have exploited the organizational dependence of CSOs on the municipality, have not tried to address feelings of powerlessness of citizens, nor provided effective institutional settings for accountability. This is in stark contrast to Hani i Elezit, where these initiatives have been animated by the Mayor.

Citizen engagement and civil society

Examining CSO behaviour in Kamenica has revealed certain limitations, which suggest CSO promotion is not a straightforward conduit for effective statebuilding. The first limitation relates to the freedom of CSOs to act as the theory assumes. One CSO representative expressed the logic quite clearly: ‘if you are a CSO and need support, you have to support the political leaders somehow to get financing’ (Youth centre representative 2013). While some CSOs receive international funding in Kamenica, the research found that most are actually dependent on municipal funding or cooperation for their continued existence (Journalist Kamenica 2013). CSO representatives across Kamenica complained that this dependency has made them wary of challenging political leaders. Not only has this dependence blunted the effectiveness of CSOs, it has cultivated an unfavourable environment for CSOs who genuinely want to hold the municipality to account. According to one source, because a Women’s Rights CSO opposed the Mayor, it cannot receive funds from the municipality and is likely to close down (Journalist Kamenica 2013).

A related limitation is the sense of deep inequality in power between political leaders and citizens. A CSO representative in Kamenica said citizens have felt powerless because one day they, or a member of their family, may need a job from the municipality, which means that there is a degree of self-censorship and citizens’ discontent has been often muted. One CSO director described the contrast between independently organized debates, where citizens have expressed anger and contempt for the municipality, and those where the political leaders have been present, wherein there has been a general subservience and deference (Youth centre representative 2013). A local radio journalist described how ‘people have had no faith that they can change things. They feel like they experience a kind of dictatorship. Whenever someone starts to speak out, they have been challenged by a bigger network of people’ (Journalist Kamenica 2013). This ‘challenge’ has been especially pertinent for opposition activists, one of whom, a history teacher in the municipality, told me how he has been ‘sent away’ to a village (to which it costs 60 euros more a month to travel) as a punishment for campaigning against the Mayor. He told me, ‘the political leaders are basically forcing him to leave his job to make space for ‘one of their people’ (LVV representative Kamenica 2013). Such a power imbalance in Kamenica has been reflected in data collected by the author (presented below in Table 3), where nearly half of all respondents strongly agreed that the political leaders have much more power.

Table 3. Power imbalance compared

Responses to statement: Political leaders generally have much more power than citizens in the municipality	Kamenica (n = 101)	Hani i Elezit (n = 96)
Citizens that strongly agreed (%)	45	2
Citizens that agreed (%)	34	53
Total strongly agreed/ agreed (%)	79	55

Source: (Jackson 2014).

Contrast the feeling of powerlessness in Kamenica with the results in Hani i Elezit, where only two per cent of respondents strongly agreed. While a majority in Hani i Elezit still believed that some power imbalance exists, this is perhaps due to the natural distinctions between citizens and political office holders. Indeed, what the research found there is a deliberate effort by the Mayor to close the gap between citizens and politicians in the municipality. The Mayor told me that close consultation with citizens has been an essential part of his modus operandi. ‘For every investment project, the citizens are involved 100%. In other municipalities, Mayors are only interested in cutting the red tape. I am always personally involved. If there are any problems during the development, I am engaged and I am constantly consulting with citizens’. Citizen engagement has not been political window-dressing but, according to the Mayor, has been an essential ingredient for successful governance: ‘There is no way we can implement project without 100% support of the citizens’ (Mayor Hani i Elezit 2013a).

Opposition politicians are even willing to concede that the Mayor has done a lot to close the power imbalance. The local Vetëvendosje (LVV) party secretary said that the main strength of the Mayor is that ‘he has been very close to the citizens. He has become close because he identifies problems together with the citizens and implements those solutions together (LVV representative Kamenica 2013).' This view is also echoed by citizens. A 45-year-old chemistry teacher in the municipality suggested ‘[In other municipalities] there is a big [power distance] difference between the leaders and citizens. However, I saw [Mayor] Suma this one time while he was clearing away the snow at midnight – nobody else does that. The Mayor shows he cares with his actions’ (Resident Hani i Elezit 2013).

The Mayor has conveyed a sense that he is predisposed to act in an open and indiscriminate way. One neat illustration of this instinctive openness is the mobile phone he carries everywhere with him and which rings every few minutes. During one interview, it became clear that whenever it rang, the Mayor had no idea who was calling. I asked his assistant why he never stored names on his phone: ‘He never records who is calling, he just answers and whoever it is, he will speak with them. He tries to be open and available for citizens the whole day’. When asked who tends to call him, the assistant replied: ‘Oh everyone in the municipality has his number’ (Municipal official Hani i Elezit 2013).

Municipal and public assemblies

Social accountability works through structures, notably municipal and public assemblies, and though these have been enshrined in the Law on Local Self Government, another shortcoming of the social accountability logic in Kamenica is that they have not been given ‘life’ by political leaders. Instead, among CSOs and the public in Kamenica, there is a widespread (but not total) lack of faith in formal institutions, especially the municipal assembly and public meetings, to provide sufficient means of holding the municipality to account. Many felt that these formal fora are just not credible. A 33-year-old architect in Kamenica described his experiences to me:

I do not go to public meetings so often because...well, last time, the municipal government manipulated the meeting. I had a proposal for a business park to be built in the municipality, but they decided for a residential building to be constructed in that area … it will only be Mayor’s close circle that could benefit from this building. And anyway, the Mayor gets his people to go to these meetings so he can dominate. (Resident Kamenica 2013c)

A Kosovan employee of an international organization, whose job is to keep a close eye on developments Kamenica’s municipal assembly, concluded that the current Mayor ‘has neutralised the municipal assembly’, and that it has been generally a common tactic for all Mayors to undermine accountability mechanisms of the public institutions. He gave an example:

If the Mayor has opponents, the Mayor will get ‘revenge’ on them because he controls the education and health sectors. So if the wife of the opponent works in education, next year she will be re-allocated into another village. Then once the opponent falls back into line in the assembly, the wife is placed back into her original position. Everyone knows what is going on but people do not want to clash with the Mayors. (OSCE Officer 2013)

While in Kamenica the municipal assembly has been described as ‘just a theatre with scenes’ (NGO director Kamenica 2013), the Mayor of Hani i Elezit has deliberately endeavoured to make public meetings an effective forum where concrete problems can be solved. For public meetings, he and colleagues went from door to door to make personal invitations to citizens and stimulate attendance through the offer of a free lunch and coffee (Mayor Hani i Elezit 2013b). Citizens think he has also made major steps to increase the credibility and effectiveness of public meetings. One resident, having witnessed a public meeting held in a village, said he was ‘really surprised by the meeting’ as ‘there were a lot of constructive discussions’. He added, ‘I know for a fact that the requests from citizens have been acted upon and implemented. We are not used to that in Kosovo’ (Resident Hani i Elezit 2013).

Traits of leadership

The research demonstrates that local politicians in Kosovo have used their privileged access to information and resources to shape the outcomes of various attempts at statebuilding. In Kamenica, as elsewhere in Kosovo’s municipalities, the emphasis has been on clientelist leadership – that is, prioritizing the interests of those who vote or may vote for you and putting the consolidation of power at the top of the agenda. Contrast this with Hani i Elezit, where public leadership has defined the exercise of political agency.

Leadership is understood as the ‘mobilisation and organization of people and resources in pursuit of particular ends’ (Leftwich 2009, 14). What explains the predominance of clientelist leadership in Kosovo’s municipalities? Why is public leadership so rare? Differences in leadership orientations are often discussed with reference to rather vacuous concepts such as ‘political will’ and there have been few attempts to understand the incentives, opportunities and constraints that explain difference in local agency in statebuilding settings (Leftwich 2009). This section contributes to filling this gap. The aim is not to create a theory about how political leaders behave in all situations, but rather, to identify certain features of the political environment which may act as constraints on public leadership or encourage clientelist leadership. I hypothesize that certain political constraints on leadership have existed in Kamenica (and other municipalities) but not in Hani i Elezit. I thus examine the behaviour of the dominant political figures, and the conditions that facilitate or constrain their behaviour, focusing on two conditions: credibility and the nature of political organization.

The constraining effect of credibility

Political scientists examine the essential elements of political order, especially qualities such as legitimacy and authority. However, my research revealed another element very important for politics in Kosovo’s municipalities: the political credibility of local actors, that is, the belief they are believable or trustworthy (Gourevitch, Lake, and Stein 2012). Credibility ‘entails a judgement of a receiver’ and is therefore a relational concept: it is not something that leaders possess but it is up to citizens to attribute credibility to political leaders, and so it is a quality that cannot be acquired in one moment but needs to be constantly nurtured (van Zuydam 2014). Political credibility, therefore, is a specific form of trust, generated within a particular context of interaction between voters and politicians.

The argument made here is that the level of political credibility enjoyed by political leaders shapes the opportunities for the kind of leadership they are likely to exercise. For most leaders, political credibility is in short supply in Kosovo’s municipalities due to a structural feature of Kosovo – that it is a ‘young democracy’. Keefer and Vlaicu (2007) have advanced an important insight that as credibility is mostly developed through an evaluation of past actions (i.e., reputation of past performance), an intrinsic structural constraint of young democracies is that politicians have no ‘track record’ upon which to demonstrate trustworthiness (Keefer and Vlaicu 2007).

During interviews with citizens in Kamenica, a recurring theme that emerged was a lack of trust that politicians will keep to promises. One resident, for instance, explained that ‘I don’t even want a politician to help me … I don’t expect that’ (Resident Kamenica 2013b), while a local lawyer and keen observer of the political scene in the municipality concluded that ‘it is naive to believe that politicians will deliver on what they promise to do’ (Resident Kamenica 2013a). Interviews with politicians revealed that they are aware that they have little credibility. The Mayor of Kamenica explained how ‘[Kosovan] Albanians had a lot of hope after the war but now they are disappointed by the politicians’ (Former Mayor Kamenica 2013). The local Democratic League of Kosovo (LDK) secretary in Peja municipality explained how ‘not a high score is given to honest or intelligent politics … people do not believe in the honesty [of politicians]’ (LDK representative Peja 2013).

National quantitative data illustrate how little credibility politicians tend to have as political parties have been rated the least-trusted organizations in Kosovan society. In 2008, over a 1000 residents were presented with a list of 10 organizations, like the Mosque, the government, civil society and political parties, and were asked by Gallup how much they trusted each organization. At the bottom of the pile were political parties with only eight per cent of respondents having a lot of trust in political parties compared to, for instance, 63 per cent for NATO and 43 per cent for EU institutions (Gallup Balkan Monitor 2008).

How can politicians convince voters that they will deliver on their promises? There are generally two different strategies to decide upon: to make your appeal on the provision of public goods or through the provision of personalized and targeted promises to individuals (Keefer and Vlaicu 2007). When states lack technical capacity, it can be challenging to deliver public goods, therefore a candidate may assess the capacity of the municipal administration and foresee that seeking credibility through the provision of public goods is riskier compared to a more targeted approach. Providing a public waste management system may take time to deliver and be translated into tangible gains for the electorate – or the municipality may just lack capacity to deliver this public good. Moreover, for any electoral gain to be realized, voters need to know about the development of this system but it is quite costly to constantly provide information about public good initiatives through, for instance, leaflets, adverts or media appearances.

In contrast, clientelistic, personalized and targeted appeals – such as employment, small infrastructure and minor resources – are more efficient in generating political credibility. Offering a job to someone, for instance, can be delivered immediately after the election and the fulfilment of this promise can be more easily observed by the beneficiary taking up the role. In fulfilling the electoral offer, credibility is then generated and the politician could be quite confident that he or she has secured votes for future elections too.

Due to the generally low capacity to deliver public goods in the nascent municipal structures of Kosovo, the most efficient way of demonstrating that you can deliver goods, and thus start to build credibility, has been to shun appeals based on public goods and make personalized and individualized clientelistic promises about private goods to the electorate, a tendency which does not favour the development of effective and accountable governance. Consider a politician who has tried to build credibility through promising jobs to various families: openness about the allocation of these jobs may cause a backlash by aggrieved citizens, and distributing jobs in this way may also violate the formal laws and processes. Instead, prioritizing the client network is helped if a leader consolidates his or her ability to act with discretion and behind closed doors rather than through distributing authority to other actors, such as civil society or the municipal assembly, or through ensuring that the administration works transparently.

Without a ‘track record’ to draw upon, leaders have found it difficult to develop credibility in Kosovo’s municipalities. Combined with the low capacity to deliver public goods this leads to politicians to seek a track record through personalized and targeted electoral offers. This is not an ‘iron-clad’ constraint but is still a relevant factor in shaping opportunities for different kinds of leadership. The case of the Mayor of Hani i Elezit, however, demonstrates that this can be overcome in a way which largely depends on what the Mayor had done prior to entering into politics.

In other words, there has been a context outside of the municipal administration in which the Mayor nurtured the sense that he is a trustworthy citizen who acts in accordance to the interests of the community. As the Mayor explained to me, the reasons people trust him ‘is not about one or two years but thirty years of hard work’ (Mayor Hani i Elezit 2013a). This work began when he was a leader within the youth division of the LDK, at that point a Kosovan-Albanian resistance movement. In the late 1990s, his local reputation grew when he became Commander of the local unit of the Kosovan Liberation Army. Yet it was really in the post-war period that Suma proved himself as dedicated to serving the community. Having remained in Hani i Elezit throughout the NATO bombing, he led a core team in rebuilding the damaged municipality. Schools, homes and roads were repaired under his direction and readied for returnees. Suma set up a system of taxation to generate resources to clean the town and organized informal directorates relating to various aspects of life. (Mayor Hani i Elezit 2013a).

Credibility is difficult to measure with any precision, but interviews with citizens and opposition politicians affirmed that the Mayor has enjoyed a great deal of personal credibility. Due to this prior nurturing of a public track record, Suma had credibility in abundance prior to entering politics. He never had political ambitions, but such was his standing across Hani i Elezit that a group of citizens pleaded to him to run in the municipality’s first direct election for Mayor in 2009. He reluctantly agreed, but only on the condition that he wouldn’t campaign. Remarkably, through his reputation alone, Suma managed to defeat the political machines of the three largest parties. His sustained efforts have been striking. When in 2015 pictures emerged of the Mayor in the middle of a freezing January night driving a snow plough through the blocked streets so that by the morning stranded residents resume their normal lives, citizens were simply not surprised as this was considered just another deed in a series of good deeds stretching back three decades (Municipal director Hani i Elezit 2015). This ability to nurture trust could be due to particular conditions in Hani i Elezit, especially its size. Suma himself admits that the relatively small size of Hani i Elezit helped in the nurturing of his credibility: it is easier to cultivate a reputation amongst 10,000 people than 100,000 people.

Unlike nearly all other local political leaders, Rufki Suma’s high credibility, mostly nurtured prior to him holding office in Hani i Elezit, has enabled him to exercise public leadership. When political credibility is high, it can allow politicians to pursue public leadership because, despite challenges with technical capacity, it makes it easier to get buy in for public plans from the electorate, and more likely that the electorate will be more patient with the reform processes and attuned to the future benefits from building up effective and accountable governance.

The nature of party organization

The nature of political organization is where the second constraint on public leadership is found. Mayors find themselves part of political parties and the way political parties are organized, which structures the incentive shaping the behaviour of political leaders at the municipal level. The core distinguishing feature of the organization of political parties in Kosovo is that they have been extremely hierarchical in nature: nearly all decision-making and command over resources has been concentrated in the hands of the leaders, or in the leaders’ close inner circle. Those at the top can be understood as ‘super patrons’: undisputed leaders who sit at the apex of the organization, charting its course and ultimately managing the internal flow of resources (Jackson 2014). This core quality of political parties has long been recognized by analysts in Kosovo. A report written in 2006 suggested that ‘elected representatives perceive few compelling reasons to be thankful to voters for getting elected. They certainly feel as they owe more towards respective party heads’ (KIPRED 2006, 20). In my own interview with the Mayor of Decan municipality in 2013, he admitted that the obligation he has for his party is just as important as the one he has to citizens (Mayor Decan 2013). One development expert with two decades of experience in Kosovo summed up the organizational structure of parties in Kosovo: ‘You are either the big man at the top or you take orders, there is no middle management. It takes strong personalities to break the mould and those who do try to do so are normally sidelined’ (International Development Professional 2013).

The absence of internal mechanisms that could allow party members, especially municipal leaders, to challenge the orders of the leader at the top has reinforced the acute hierarchy. Rules favouring open and democratic debate have been generally cast aside in favour of informal and flexible organization that serves the writ of the party leader. In the words of Philip Keefer, this means there has been an absence of ‘organisational arrangements that provide for collective action’, such as subjecting party leaders to oversight by members (Keefer 2015, 229).

The acute organizational hierarchies of political organization in Kosovo have meant the political survival of local politicians is utterly dependent on support of the super patron, who has control over money for electoral campaigns. The Mayor of Hani i Elezit explained how virtually no candidates running at the municipal level in Kosovo have access to their own resources but must rely on financing from the party leadership (Mayor Hani i Elezit 2013b). Leaders of political parties are also important for according authority and credibility to the candidates at the local level when fighting election campaigns. The importance of this ‘reflected glow’ of the party leader is indicated in the significance municipal candidates accord to party leaders visiting their municipalities to rally support for them (KIPRED 2006).

The main constraint hierarchy has exerted on public leadership and the implementation of the statebuilding model is that local leaders must, first and foremost, ensure that the super patron’s interests and patronage requests are well managed at the local level. To illustrate, a local employee of an international organization, and something of an insider in Kamenica municipality, described a situation in 2012, when the Mayor received a request from the party centre to employ someone as a security guard in one of the schools. The Director of Education actually refused at first until he received a phone call from a minister in Pristina, who demanded the person be hired. As the insider explained ‘it’s about power in the end and he had to give in’ (OSCE Officer 2013).

The pressure on Mayors to fulfil requests sent from ‘the centre’ has not been incidental but constant, meaning politicians, who may ordinarily want to implement the effective and accountable municipal governance, have been sidetracked by the compelling logic of serving the patronage interests of the super patron. Analysts elsewhere have also established that due to the hierarchical structure of the political parties in Kosovo, leaders of parties sometimes place pressure on local leaders to conduct policies that may not necessarily be in their best interests, but which are demanded by the party centre (KIPRED 2006). This has certainly been the case in Kamenica, where one political analyst explained that successive leaders have neglected to really take on the important challenges that confront the municipality, such as improving living conditions and making sure that the administration runs according to the rule of law, precisely because they have been under ‘intense pressure from higher levels’ to direct most of their efforts to consolidate power for the party (Political Analyst 2015). Consequently, according to an evaluation undertaken by the IMF, while the municipality only 60 sixty workers, there are currently over 200 employed, due to ‘all the jobs the Mayor has created because of requests from all the party leadership’ (Political Analyst 2015).

Party organization has therefore mattered for the type of leadership that has emerged in Kosovo’s municipalities. Exercising public leadership has been difficult because of the pressure to fulfil requests from the centre. It has made more sense from the point of view of political survival to refrain from strengthening the municipal assembly, to abstain from increasing the integrity of the administration and to avoid making all decisions public because these activities would make it more difficult to serve the patronage requests and interests of the super patron.

If party organization has constrained public leadership, we would expect to see in Hani i Elezit some other form of party organization. Significantly, Mayor Rufki Suma has never been a part of a political party – he is the only Mayor in Kosovo independent of a party structure. The Mayor himself admitted to me that this has given him significant freedom to pursue policies oriented towards the public good. In a revealing exchange, the former PDK Mayor of Hani i Elezit admitted that the incumbent Mayor Rufki Suma has a distinct advantage when it comes to statebuilding, because he is an independent MP. The former Mayor spoke in very positive terms about personal qualities of the incumbent Mayor but also identified how important it was that the Mayor was not part of a political party (Former Mayor Hani i Elezit 2013).

Former Mayor:

Yes he has an advantage! If you come to power because of the party, you have to listen to the party. Because he is independent he does not have to follow a line that someone else has set.

Author:

What kind obligations did the party put on you?

Former Mayor:

Nothing directly, but indirectly you have to give something back to them, for example employment to people. You are dependent on them for your political base so it is important to give something back.

Author:

Is this common in Kosovo?

Former Mayor:

It is not just here. Under the party flag there are always obligations.

Though political parties receive very little attention in debates about statebuilding and institutional development, the research here demonstrates that they are central to the type of political agency that is exercised. The research supports Philip Keefer’s work, which in recent years has persuasively argued that the provision of certain government models is heavily influenced by the nature and organization of political parties because political parties are crucial in structuring political incentives and choices (Keefer 2015).

Conclusion

Internationally led statebuilding is not going away from international politics any time soon. It is still high on the agendas of powerful states and is central to the various strategies that aim to address sources of global risk. This research has used a comparative research design to demonstrate how statebuilding policies were not implemented in a straightforward manner, but rather channelled through local agency – in this case, municipal Mayors who have facilitated or limited the impact of these endeavours.

The influence of local agency cannot be assumed in every case. At the municipal level, geographical distance means that statebuilding relies on local implementers, a structural factor that provides latitude for local agency. The extent to which the analysis – that effective statebuilding may depend, as in Hani i Elezit, on impetus provided by local leaders – can be applied to other cases depends somewhat on whether this ‘scope condition’ is met. In general, more evidence is also needed on how and the extent to which local agency may determine statebuilding outcomes. Nevertheless, this research has demonstrated the value of seeking explanations for the varied effectiveness of statebuilding in local agency, a realm of explanation that goes beyond those accounts privileging what statebuilders ‘do’ rather than ‘what occurs’.

Subsequent to identifying the importance of local leadership, this article explored why different types of leadership have emerged by examining local pressures and incentives. In Hani i Elezit credibility nurtured prior to his involvement in politics, allied to his independence from party machines, allowed the Mayor to gain the ‘buy in’ for his more public-oriented governance. In contrast, the absence of credibility amongst Kamenica’s leadership seems to have encouraged clientelist politics, a dynamic amplified by the need to fulfil patronage requests orchestrated by the upper echelons of a party hierarchy. These factors may not be just germane to the particular setting here, since the absence of credibility and trust in politicians and the hierarchical patronage-based organization of political parties resonates with many other settings. While the process-tracing method has sought to identify these factors as playing a distinct role in this case, they are likely to be mutually reinforcing with the nature of leadership in turn affecting the extent of these constraints. Exercising particularistic leadership may exacerbate credibility constraints, for instance. Further research should consider how different types of leadership may undo or bring about additional constraints.

The constraints identified in this case are also not exhaustive; in other settings additional sources of motivations may be relevant. More comparative research is needed to explain why local actors act as they do by examining locally determined incentives and pressures, which could be rooted in economic structure, social norms or political imperatives. These can only be assessed initially through in-depth research, but patterns from case studies may emerge which may make it possible to build up mid-range theories that explain local agency, explanations that would seem important too for practitioner perspectives.

Such research should appreciate the differentiated nature of local agency (Beysoylu 2018; Kursani 2018; Phillipps 2018; Selenica 2018; Tadic and Elbasani 2018; Triantafyllou 2018; Troncota 2018). While the ‘local turn’ in the statebuilding literature has provided new insights into statebuilding dynamics, the literature has been less disposed to exploring the different facets of local agency, to understand their relevance and how they may contradict each other. Heathershaw, for example, points out how many influential accounts treat local agency as though it is a unitary force rather than differentiated and complex (Heathershaw 2013). The research here demonstrates, not only that political leadership is a critical and specific form of local agency, it can also shape the impact of other forms of local agency, such as civil society. Mayors were identified here as crucial forms of local agency, but future research should be sensitive to other forms of agency, such as economic or non-state actors, that may also be important in facilitating or constraining statebuilding processes.

Notes on contributor

David Jackson is a researcher and adviser at U4 Anticorruption Resource Centre based at the Chr. Michelsen Institute in Bergen, Norway. David is specialized in governance and anticorruption, with a particular focus on the role of informal institutions and social norms. He holds degrees from Oxford University, the Hertie School of Governance and the Freie Universität Berlin.

Funding

This work was supported by the Fritz Thyssen Stiftung and the Chr. Michelsen Institute.

Disclosure statement

No potential conflict of interest was reported by the author.