Abstract
This article examines how social movements based in poor communities make electoral alliances with politicians in contemporary Indonesia. Drawing on case studies of the urban poor in two elections in Jakarta, we point to a pattern of adversarial linkages by which movements present candidates with demands – in this case about housing and livelihood security – which are then distilled in formal ‘political contracts’. Unlike institutionalised relationships between parties and social constituencies in many democracies, these linkages are ad hoc, pragmatic and characterised by mistrust. In Jakarta, they involved disaggregation rather than aggregation of interests, with movement actors in the second election in 2017 seeking concrete gains relating to land and livelihoods in particular neighbourhoods, rather than a broad programme of urban reform, as had been their goal in 2012. We suggest that such adversarial linkages are a feature of contemporary Indonesian politics. They allow marginal groups to make contingent political gains but are compatible with prevailing clientelistic patterns, which limit their potential to promote systemic change.
Introduction
How do social movements rooted among poor and marginalised citizens bargain with politicians, and win concessions from them, when electoral competition is personalised and clientelistic? This article 1 addresses this question by examining the 2017 gubernatorial election in the Indonesian capital of Jakarta, focusing on an alliance between a movement representing members of the urban poor dislocated by urban redevelopment and the winning candidate, Anies Baswedan. Taking our cue both from literature on voter-politician linkages, and from recent debates about the nature of political representation in Indonesia, we point to an emerging pattern of adversarial linkages in Indonesian local politics. In this pattern, interest groups – especially, but not limited to, groups representing lower-class constituencies – seek mutually beneficial alliances with political candidates, exchanging electoral support for policy or particularistic concessions. These linkages are adversarial in that they are characterised by a significant degree of mistrust, especially suspicion on the part of the movement actors that politicians will renege on their promises to citizens. Accordingly, a defining feature of these coalitions is that they are embodied in formal “political contracts” (kontrak politik) by which the interest group tries to bind the candidate to an agreed set of concessions.
The authors would like to thank the Indonesia Project at the Australian National University, which funded the project (ANU Indonesia Project Research Grants 2016/2017) through Amalinda Savirani. We also thank Ekamara Ananami, Amita, and Resandy for their research assistance. Aspinall's fieldwork was funded by the Australian Research Council (FT120100742). Savirani used a small portion of fieldwork data from the “Citizenship sub Project”, part of the “Power, Welfare and Democracy (PWD)” research cooperation between Universitas Gadjah Mada, Indonesia, and University of Oslo, Norway.
Literature on relations between politicians and citizens in democracies has identified three archetypal linkage types: programmatic, in which parties promise packages of broadly applicable policies; clientelistic, in which they target material goods and other benefits in a particularistic manner at individual voters or small groups; and charismatic, or populist, founded on the personal authority or appeal of a leader (Kitschelt 2000; Stokes 2009). The alliance formation studied in the present article was a hybrid of the first two types. The social movement upon which we focus ultimately eschewed the broad policy promises typical of programmatic politics, having felt betrayed by the failure of the incumbent governor to honour earlier commitments to respect the rights of the urban poor. Instead, the political contract drawn up with Anies Baswedan in 2017 was highly particularistic, detailing land status changes and other desired outcomes in named precincts and neighbourhoods. But this was not a purely clientelistic exchange: the group did not accept the material rewards that are commonly distributed in exchange for political support in Indonesian elections.
In making this analysis, we seek to contribute to recent debates about the nature of political representation and power in contemporary Indonesia. Over the last decade or so, analyses of Indonesian politics, especially at the subnational level, have shown that the organs of local democracy have been captured by elites, many of whom derive their influence from the fusion of state and economic power that characterised the authoritarian “New Order” regime (1966–1998). Facilitating this elite capture has been so-called “money politics” – a monetisation of political relationships that includes such phenomena as ticket-purchasing by candidates, vote-buying during elections, and lavish funding of privately-organised campaign teams, all sustained by rent-seeking behaviours in office (Hidayat 2007; Hadiz 2010; Aspinall and Sukmajati 2016; Mietzner 2007). This elite capture is widespread and obvious, and has not been seriously debated by scholars. However, one forceful interpretation – the “oligarchy thesis” promoted in slightly different forms by Richard Robison and Vedi Hadiz (2004) and Jeffrey Winters (2011) – has been challenged by analysts who view it as overly totalising. Some critics have noted various ways in which non-oligarchic forces exercise political influence, especially in policy-making (Mietzner 2013; Aspinall 2014). Others have drawn attention to electoral interventions by groups representing lower-class interests, such as labour unions, farmers’ groups or associations of informal traders (Aspinall 2013b; Ford 2014; Ford and Pepinsky 2014; Gibbings, Lazuardi and Prawirosusanto 2017; Mahsun 2017; Savirani 2016). One large action research project, in which one of the authors of the current article (Savirani) participated, aimed to encourage “democratic political blocs” through which popular movements would engage in electoral politics. The idea was that activists needed to save local democracy from elite capture by building electoral coalitions of civil society organisations and popular movements of labour, farmers and the urban poor (Priyono et al. 2007).
The present article builds on this strand in the literature but also step backs from it to make a more general characterisation of the emerging pattern of linkages between social movements and elected politicians in contemporary Indonesia. The coalition we observe in Jakarta was almost entirely uninstitutionalised, being based on an ad hoc deal reached between the candidate and representatives of the urban poor movement, with the help of several mediators who had feet in both camps. The coalition was not embedded in any institutionalised connection between social movement organisations and a party, as is often the case in party systems based on well-defined cleavage structures. Social movements representing lower-class interests in Indonesia have little choice but to engage in one-off transactions of the sort described in this article, rather than in the systematic and structured interest aggregation that is the sine qua non of programmatic politics. Rather than representing a fundamental challenge to Indonesia's clientelistic political order, such lower-class electoral engagement is thus readily accommodated within it, and represents just one element within the kaleidoscopically fragmented political order of post-Soeharto Indonesia (Aspinall 2013a). However, the case also shows that popular movements can gain access to the policy process, both by relying on the skills of movement entrepreneurs and allied intellectuals, and by improving their own technical knowledge and capacity. Finally, by making this argument we also seek to shed light on the politics of the urban poor in Indonesia. While this is an important group, it has been largely neglected in literature on Indonesian politics (Wilson 2017).
We develop our argument through several sections. First, we introduce the Indonesian case and our framework on politician–citizen linkages. Second, we introduce the urban poor and its politics in Jakarta. Third, we discuss initial attempts to forge an electoral alliance during the 2012 gubernatorial election, during which urban poor activists supported populist candidate Joko Widodo (Jokowi). Although movement activists were generally satisfied with Jokowi's performance in office, policies of evictions without negotiation resumed in 2014 once he was elected president and was succeeded by his deputy, Basuki Tjahaja Purnama (popularly known as Ahok) as governor, fuelling disillusionment. Fourth, we explore the formation of a new electoral alliance during the 2017 gubernatorial election, when activists replaced the broad programmatic sweep of the 2012 political contract with disaggregated, particularistic demands. Fifth, we examine how members of the urban poor movement tried to ensure that the commitments in the political contract became embodied in policy after the election. Finally, we conclude by considering other examples of such social movement electoral engagement and the light they shed on Indonesian politics more broadly.
Politician–Citizen Linkages
Much of the literature on linkages between politicians or political parties and social movements discusses situations in which parties are deeply embedded in and organically connected to particular social milieus. Parties themselves are often products of social movements. Obvious examples are the social democratic parties of Europe and elsewhere that were formed by labour unions, or the Islamist parties that arose from Islamic educational, charitable and similar organisations in many countries. In other cases, existing political parties develop linkages with new social movements by incorporating their issues into their existing platforms “through techniques of programmatic unity building” (Kitschelt 2000: 848). This, in short, is the model of the programmatic party, by which political parties incorporate and express the interests of particular social constituencies, allowing them to “map their issue positions on a simple conceptual alternative (Left and Right) based on underlying programmatic principles” (Kitschelt 2000: 848). Programmatic politics involves a process of interest aggregation in which the particular interests of small groups are abstracted and absorbed into over-arching policy positions that are then offered as bundles to the electorate.
In Indonesia, some parties possess organic connections to civil society groups. Notably, several parties have deep connections with Islamic social networks. Examples include the National Awakening Party (Partai Kebangkitan Bangsa, PKB) and the mass-based traditionalist organisation, Nahdlatul Ulama; the National Mandate Party (Partai Amanat Nasional, PAN) and the modernist organisation Muhammadiyah, and the Prosperous Justice Party (Partai Keadilan Sejahtera, PKS) and the Muslim-Brotherhood-influenced tarbiyah movement (Machmudi 2008). Such connections are partly ideological, in that these parties express the worldview of the parent movement, but also pragmatic in that elected politicians from these parties also direct state resources toward the parent movement when feasible.
In Indonesia's electoral democracy, however, two factors have impeded development of a generalised pattern of institutionalised connection between social movements and parties. The first factor is particular to poor communities: the historical elimination of the Left. The Left was wiped out in 1965–1966 and has never been able to resurrect itself as a mass force. Partly as a result, the typical left–right cleavage structure that defines party systems in many democracies is absent in Indonesia. The absence of a significant social-democratic or left-wing party based on the lower classes is frequently noted as one major source of the limited nature of the post-Soeharto democratisation (e.g., Robison and Hadiz 2004: 134–135; Winters 2011: 181).
The second factor is the emergence of a generalised pattern of clientelistic politics in post-Soeharto Indonesia. In clientelistic systems, politicians “create bonds with their following through direct, personal, and typically material side payments” (Kitschelt 2000: 849). Although space limitations prevent a thorough examination of the emergence of this pattern here, one obvious source is the adoption in Indonesia of candidate-centred rather than party–centred electoral systems (see, for example, Erb and Sulistiyanto 2009; Aspinall and Sukmajati 2016). The result that is that candidates for elective office have incentives to promote their individual candidacies and appeal to constituents by offering them particularistic benefits, rather than to pool their efforts and cooperate with copartisans in constructing broadly appealing programmes. A strongly personalised electoral pattern has emerged in which candidates for office construct personal success teams and woo voters by delivering them material benefits – whether in the form of pork-barrel projects or club goods to communities, or cash payments and other gifts to individuals.
One feature of this clientelistic landscape is a pattern of transactional deal-making, in which candidates bargain with local community representatives, offering them specified benefits in exchange for the votes of the community they lead. Such representatives are typically village heads, religious leaders, clan chiefs or similar informal leaders, and the benefits are usually club goods such as repairs to a road, houses of worship or other community infrastructure, or gifts of sports equipment to a local youth group, livestock to the farmers’ cooperative, cooking equipment to a women's welfare group and so on. Such agreements are sometimes spelled out in a written “political contract” (kontrak politik) that gives the appearance of being legally binding. Such political contracts are a “new phenomenon in Indonesian politics” (Gibbings, Lazuardi and Prawirosusanto 2017: 252). Both authors have observed such political contracts in widely varying electoral contexts (see also Sulaiman 2016: 63–64; Rohi 2016: 380). In form, political contracts borrow from the language of business agreements; they typically spell out a number of commitments that the candidate promises to deliver should he or she be elected.
Such political contracts are also products of the climate of generalised mistrust of politicians and their promises in contemporary Indonesian society. Anti-party or anti-politician sentiment has been observed since early in the democratisation process, but has become entrenched as ‘money politics’ has spread, along with widely held assumptions among voters that candidates seek office simply to benefit themselves, their family or their group. Therefore, as Rudi Rohi (2016: 380) has explained, some politicians try to overcome this trust deficit by offering formal contracts in order to borrow “the symbols of legal procedure and financial transactions—contracts—to convince voters that they would fulfil their promises and, most importantly, get them to deliver their votes.” As Gibbings, Lazuardi and Prawirosusanto (2017: 252) put it, “The term ‘contract’ implies a degree of accountability […] As opposed to money politics, political contracts tend to be seen in a more positive light.”
While political contracts are most frequent in the context of clientelistic deals offering small-scale pork barrel projects and similar benefits to villages or urban neighbourhoods, this model is also applied in other circumstances. In particular, in the absence of programmatic parties representing lower-class interests, it is available for use by social movements representing the poor. For example, a recent article by Gibbings, Lazuardi and Prawirosusanto (2017) drew attention to efforts by the Asosiasi Pedagang Kaki Lima Indonesia (APKLI, Association of Indonesian Street Vendors) and other groups representing informal traders to mobilise support for political candidates at various levels on the basis of formal political contracts. Notably, many labour unions have signed political contracts with candidates in local and even national elections (Aspinall 2013; Caraway and Ford 2014: 148; Ford 2004: 347).
For example, in a district-head election in Bekasi in 2012, a local labour union branch made a deal with the candidate from the Golkar Party, Neneng Hasanah Yassin, on three issues: a moratorium on outsourcing, tightening up labour law enforcement, and building a permanent office for a workers’ groups. Nanang was elected, but she only respected the deal during her first two years in office; for the next three she ignored it. In Karawang, another industrial areas to the east of Bekasi, a member of the Metal Workers Union (FSPMI) made a deal with Cellica Nurachadiana, a candidate in the 2015 election (Interview, FSPMI members 5 January 2017). In subsequent years, the regional minimum wage in Karawang was the highest in Indonesia, which the activists attributed to the contract. Therefore, the effectiveness of such contracts varies: much depends on the movement's capacity for mobilisation, and the candidate's future dependence on its support. Ultimately, given the ad hoc nature of such deals, they are vulnerable to abrogation. The fact that winning candidates are limited to two terms in office also limits their applicability, even in the best of circumstances.
The Urban Poor in Jakarta
In order to understand the political contracts created by urban poor activists and gubernatorial candidates in 2012 and 2014, it is helpful to briefly review the context of the urban poor in Jakarta. As with Indonesia as a whole, poverty is a major problem in the capital city. Official statistics from March 2016 show that 384,000 people – 3.75 per cent of the total population of Jakarta – were living in poverty. This is low compared to the national poverty rate of 10.86 per cent, but likely underestimates the problem considerably. In Jakarta, there is a significant unregistered population, many of whom are poor. A rough estimate is that around 3 million people live in settlements in the Greater Jakarta Region that lack basic infrastructure services (Padawangi and Douglass 2015: 533).
Housing for Jakarta's poor has been a problem in Jakarta since the colonial era (Abeyasekere 1989: 81). Poor people moving into the city have always found ways to shelter themselves, creating urban kampung: settlements characterised by informality, irregularity, illegality, flexibility and resilience (Jellinek 1991; Murray 1991). Many poor people occupy vacant state land, such as that found along railway tracks, rivers and roads and under bridges.
Since the era of Governor Ali Sadikin (1966–1977), housing for the poor has been a persistent political problem. Sadikin famously dealt with the influx of migrants at the outset of the New Order, and the slum areas they created, through a series of policies that mixed developmentalism with coercion. He declared Jakarta a “closed city” in 1970 (Abeyasekere 1989: 222), banning immigrants unless they had a valid ID card and a job. He initiated eviction programmes to make way for urban redevelopment, and street clearance to remove street vendors and becak (tricycle rickshaw) drivers. Starting in 1969, Sadikin also began conducting on-site residential renovations and providing basic infrastructure in Jakarta's kampung through a “Kampung Improvement Programme” supported by the World Bank. Throughout the New Order period, this coercive city improvement approach led to repeated waves of sometimes violent land clearances and the replacement of many urban kampung by roads, apartment buildings, shopping malls and other developments, but also by the gradual upgrading of many settlements (for a critique, see Werlin 1999).
The advent of democracy layered new political complexities on this context. Though post-Soeharto Jakarta governors have generally pursued similar urban development goals to their predecessors, they have also had to deal with conflicting constituency pressures, especially following the introduction of direct gubernatorial elections in 2005. On the one hand, elected officials have faced demands from middle-class voters who often view poor Jakartans and their informal livelihoods as the source of the traffic jams, flooding and other problems of congestion experienced by the city. On the other hand, the poor residents of Jakarta's urban kampung themselves constitute a major vote bank whose support can be the difference between electoral success and failure.
Democracy also widened the political space available to NGOs that advocated on behalf of the urban poor. Since the time of Ali Sadikin, when the Legal Aid Institute (Lembaga Bantuan Hukum, LBH) provided courtroom assistance to displaced squatters, middle-class social activists have tried to assist poor Jakartans pressured by government policies. During the late New Order, the rise of a new urban poor movement in Jakarta was facilitated by broader trends in the NGO scene, with a new sensibility of popular empowerment prompting students and other activists to attempt to organise residents of poor kampung and mobilise against evictions.
Several organisations emerged. One of the most important was the Urban Poor Consortium (UPC), which was established in September 1997. Founded on a belief that it was the people, not elites, who could initiate political change, UPC has consistently focused on organising the urban poor and on defending their social, economic and political rights. Since 1997, the UPC has carried out organising work in 58 kampung in Jakarta and established grassroots organisations such as SEBAJA (Serikat Becak Jakarta, Jakarta Becak Union). The UPC has also expanded its organisation nationally by establishing UPLINK or Urban Poor Link, which became JRMK (Jaringan Rakyat Miskin Kota, Urban Poor Network) in 2002. UPLINK was initially active in 14 regions, but it has maintained a presence in eight cities: Jakarta, Banda Aceh, Bandar Lampung, Yogyakarta, Surabaya, Sidoarjo, Makassar, and Kendari. In 2009, UPC decided that in order to maintain its independence, it would no longer take funds from donors. This decision was highly unusual in the world of Indonesian NGOs, many of which are acutely aware of their problems with donor dependency, and has had a large impact on the organisation, driving away some members, but also nurturing an ethos of self-reliance among its urban poor supporters.
This social movement represents the formal tip of a subterranean world of informal politics. The urban poor in most developing countries, as documented by authors such as Partha Chaterjee (2004), writing on India, and Javier Auyero (2001), on Argentina, frequently develop a mode of everyday politics in which “informal problem-solving networks […] ensure material survival” (Auyero 2001: 14). The precarious nature of existence of the urban poor, not only in terms of poverty, uncertainty of income and poor access to government services, but also frequently in terms of illegality of residential status and housing, makes them look to their social relations to ensure their survival. For example, Auyero explained how brokers (in his case, from the Peronist party) play a key role in connecting to poor people to the sources of state patronage and support that can make life bearable. Similarly, Chaterjee drew attention to how the poor in India, in order to attain access to recognition and state benefits, “must succeed in applying the right pressure at the right places in the governmental machinery”; such a process involves mobilising against existing structural inequalities but also requires local community leaders who “mediate between those who govern and those who are governed” (Chaterjee 2004: 66).
Fluid and complex patterns of brokerage are also typical of the everyday politics of the urban poor in Indonesia. For example, in her work on Indonesian informal traders, Sheri Gibbings has pointed to the mediation role played both by brokers who “are not street traders themselves, but have experience organising groups and connections to political parties or government officials”, and by operatives who are themselves experienced and respected street traders (Gibbings, Lazuardi and Prawirosusanto 2017: 269; see also Gibbings 2013). A similar dynamic is visible with the urban poor movement in Jakarta. Most of the key leaders and community organisers of the UPC have backgrounds as middle-class activists; the founder, Wardah Hafidz, is a nationally known figure, as well as being a graduate of Ball State University in the United States, while most of the group's community organisers have university degrees. While some UPC activists, including Wardah Hafidz, remain leading figures, they now claim they primarily play roles as facilitators, meaning they assist activists with urban-poor backgrounds, organised through JRMK, to take major leadership roles. For example, if the UPC receives an invitation to take part in a meeting organised by an urban poor network abroad, the group will send urban poor representatives, rather than the activists, to attend. Therefore, the backbone of the organisation is community leaders from poor kampung, who are groomed systematically by UPC to organise the urban poor in each kampung.
These community leaders themselves have varied backgrounds, but are typically individuals who have come to prominence by playing a leading role in organising protests or other activities for the movement itself; sometimes they were the initial contact point connecting their kampung to UPC, being the person who first sought out the organisation when eviction was initially threatened. Although these grassroots activists typically lack the party connections of urban brokers in India and Argentina, many are themselves respected as tokoh masyarakat (informal community leaders) by their neighbours, having the capacity to mobilise them for protests or influence their voting choices, and being well connected to other kampung-level notables and problem-solvers. Some of them have almost 20 years’ experience in organising citizens; such people include Edi Saedi and Guntoro (Gugun), a Kampung Tongkol (North Jakarta) resident, whose area was threatened for eviction under Ahok. Gugun has been working as a motorcycle taxi driver with the online platform Grab for the past two years and has become a prominent movement activist with a significant media profile in his own right, while still living in the kampung and seeking his income from a typical kampung pattern of work.
In Indonesia's cities, as in India and Argentina, the rise of electoral competition and new social movements has expanded opportunities for transactional bargaining between representatives of the urban poor and politicians, frequently involving mediation by NGO activists, lawyers, political operatives and other brokers. But such deal-making has taken time to evolve in Jakarta. Governor Sutiyoso (1997–2007), a former general, bridged the transition between the authoritarian and democratic eras, being appointed as the Soeharto era was drawing to a close but securing reappointment in 2002 with the patronage of then-President Megawati Soekarnoputri and extensive vote buying in the provincial parliament (Steijlen 2002). At this time, the UPC made its first attempt to influence the policy process through electoral politics, by nominating one of its own members to run as a candidate for governor. This campaign was largely symbolic, given that governors were still elected by the local parliament rather than the population at large. Moreover, the nominee, Rasdullah, a becak driver, did not meet the basic administrative requirements of candidacy, having failed to attain a high school education (Steijlen 2002: 517). During his decade in power, Sutiyoso pursued hard-line policies to urban problems, emphasising public order and intensifying efforts to remove street vendors and becak drivers (Nurbianto 2004). Urban poor groups that protested against his policies sometimes faced violent attack by preman, or street thugs (Kompas 2002).
Next came governor Fauzi Bowo (2007–2012), who had been Sutiyoso's deputy and before that a career bureaucrat and Golkar politician. In the 2007 election – the first direct gubernatorial election in the capital – Adang Daradjatun of PKS was his rival. With gubernatorial elections now using a direct vote, and requiring expensive campaigns, UPC shifted its strategy. It did not nominate its own cadre, lacking the resources to organise a campaign with a realistic chance of success, but instead joined a wider coalition. UPC/JRMK made a deal with Adang in which he promised to halt forced evictions for the duration of his five-year term should he be elected (Interview Edi Suadi, 25 May 2017). It was a vague deal, limited to this single item and without a detailed plan. But Adang lost, and Fauzi continued Sutiyoso's harsh urban improvement policies, with his term remembered as a period when the urban poor movement had few opportunities to advance its agenda (Widhiarto 2010).
A dramatic change came with the 2012 gubernatorial election. Former Solo Mayor Joko Widodo (Jokowi) was elected, gaining popularity both for his promises to tackle the city's numerous transport, flooding, housing and other problems, and for his ability to connect with ordinary voters, including the poor, especially through his trademark blusukan meet-the-people style encounters in which he would elicit citizens’ views about problems they encountered in daily life (Mietzner 2015). During the campaign, Jokowi and his running mate, Basuki Tjahaja Purnama (Ahok), made various promises to and agreements with members of the urban poor movement. In fact, Jokowi was not the first choice of these activists. In the first round of the election that year, UPC/JRMK supported Faisal Basri, a well-known economist from the University of Indonesia who was running as an independent and was well connected in civil society networks, including among some of the middle-class intellectuals and urban planners who were linked to UPC. When Faisal was knocked out, leaving Fauzi Bowo and Jokowi facing off in the second round, UPC/JRMK threw its weight behind Jokowi, doing so on the basis of an explicit political contract. The contract was facilitated by figures in PDI-P, including the well-known female parliamentarian and activist Eva Sundari.
The contract consisted of three main points. First, Jokowi and Ahok promised to involve members of the urban poor in planning, executing, and monitoring of three key governance issues: spatial planning, formulation of the local budget, and the development programme. Second, they promised “fulfilment and protection of citizens’ rights”. Specifically, they promised to legalise illegal kampung and to provide ownership certificates to people who had been residing on land for more than 20 years, if the status of that land was not disputed. They also said they would pursue kampung renovation rather than eviction. Third, they promised to protect actors in the informal economy such as street vendors, fishers, and traditional market traders. The contract was a brief one-page document, expressed in sweeping terms, but it was widely publicised in poor kampung.
Jokowi–Ahok, 2012–2017: Promises Kept, Promises Broken
Once Jokowi became governor and Ahok deputy governor, the pair had to perform a delicate balancing act if they wanted to keep their promises of overcoming Jakarta's problems, while also respecting their commitments to the urban poor. In particular, floods were worsening, and reducing their severity meant fixing the city's reservoirs and clearing rivers and canals. These steps required evictions of communities living along waterways. In contrast to the coercive approach of the past, Jokowi personally approached urban poor communities living in target areas, discussing these problems with them and listening to their wishes. 2
Recordings of such meetings are available online, for example: www.youtube.com/watch?v=ONn-QfoZMGc; and www.youtube.com/watch?v=thuIop2pb0s (22 February 2018).
This approach was evident in the evictions caused by upgrading of the Pluit reservoir in North Jakarta in 2013. This was a long overdue programme of the Jakarta government but was hard to execute because it required rehousing 35,000 people living around the reservoir. UPC/JRMK had long worked in the area and many residents were members. After the election, UPC/JRMK conceded that forced evictions could proceed but insisted that those affected first be provided with shelter at nearby locations to minimise disruptions to their daily lives. Jokowi agreed, and the Muara Baru multi-storied social housing complex was constructed for affected people prior to relocation. Mawar (pseudonym), a resident, said that she and her sister received keys to their low-cost apartment before they were evicted (Interview, 27 March 2015). UPC/JRMK members were also happy with the outcome (Interview, Melati (pseudonym), 25 March 2015).
Jokowi and Ahok also tried to fulfil their promises of better treatment of the urban poor by establishing a programme called “Kampung Deret”, which involved renovating row-houses in slum areas. They planned to renovate 74 slum locations, with the first and best known, in Petogogan, opened in April 2014. A total of 123 homes in Petogogan were renovated with funding from the Jakarta budget. The project was considered a success, creating a healthier environment with public facilities such as water installations and open space for a children's playing ground (Dewi 2014). However, the programme was halted in 2014 when Jokowi became president, in part because the Supreme Audit Agency found that the houses in Petogogan were illegal because they were built on state land and the owners lacked valid title.
When Jokowi became Indonesia's president in 2014, Ahok took his place. As Jokowi's deputy, and during his first year as governor, Ahok generally stuck to Jokowi's approach of using participatory methods to find housing solutions for the poor. For example, he was directly involved in the Pluit Reservoir clearances in 2013, and maintained good communications with JRMK at that time. As he consolidated his position, and in order to address broader housing needs, Ahok prioritised the construction of social-low-cost rental apartments (Rusunawa, or Rumah Susun Sederhana Sewa; literally, simple rental apartment buildings). The goal was to avoid the legal problems that had impeded the kampung deret programme by using land that belonged to the Jakarta government. The programme involved moving people rather than renovating their existing housing, and thus required evictions.
Meanwhile, to make good on promises to deal with Jakarta's flooding, in 2014 Ahok began a large-scale project to improve water flow along 13 rivers that had been narrowed by unauthorised settlements and become shallow due to sedimentation. Ahok aimed to clear settlements from up to five metres from the water's edge along riverbanks, to dredge sedimentation from more than 60 kilometres of riverbeds, to construct 42 kilometres of embankments and to clear green areas in order to improve water absorption. The estimated total cost was USD 190 million, USD 140 million of which was to be secured from a World Bank loan. In fact, the project had been initiated during Fauzi Bowo's governorship in 2010, but it was only under Ahok that major project execution took place.
One consequence of this project was accelerated evictions of people living along riverbanks and in designated green areas. According to a report prepared by LBH, during 2015 there were 113 separate instances of eviction, with almost 6300 families and over 8,100 small vendors affected (LBH 2017: 3). In 2016 there were 193 cases with more than 5700 families and more than 5300 small-scale enterprises forced to move (LBH Jakarta 2017: ii). In pushing through these evictions, Ahok applied a different leadership style than Jokowi, emphasising administrative efficiency and strong leadership over participation. According to the LBH report, most of the evictions were arbitrary, being carried out without negotiations with evictees and most without adequate compensation (LBH Jakarta 2017: 34, 48).
Evictees were promised alternative housing, in the form of rusunawa, but Ahok insisted on going ahead with evictions despite a serious back-log in rusunawa construction. The government planned to evict more than 12,000 families, but units were available for only 1000 (Wardhani 2015). In fact, the total number of such units owned by the city in 2015 was 14,187, 3 almost all of which were already occupied and subject to long waiting lists. The result was that the government asked most evictees to survive on their own while waiting for the rusunawa to be ready. Another problem was the location of the rusunawa units that were available. People evicted from the Kampung Pulo area were lucky; they were provided with units in Jatinegara Barat, which was close by. But evictees from other sites were less fortunate: those from Kampung Akuarium, Pasar Ikan in North Jakarta, as well as from Bukit Duri in South Jakarta, were moved to units in Rawa Bebek at the eastern edge of Jakarta, 30 kilometres away. Moreover, Ahok rejected alternatives proposed by UPC/JRMK and other urban NGOs like Ciliwung Merdeka and Arsitek Komunitas or “Arkom” (Community Architect), such as on-site upgrading. 4
PowerPoint presentation provided by Office of Housing and Buildings, Jakarta provincial government. Online: http://ciptakarya.pu.go.id/bangkim/perdakumuh/upload/89-6.%20Paparan%20Penanganan%20Kumuh%20DKI%20Jakarta%202016.ppt (22 February 2018).
On-site upgrading is an alternative to slum clearance. There have been several examples in the post-Soeharto period. In Solo, where Jokowi was mayor between 2005 and 2012, Arkom facilitated a participatory programme called “Kampung Renteng”; in Surabaya, following advocacy by UPC activists, the city government cancelled plans to evict a poor community living along the banks of the Stren Kali river in 2007, and instead carried out on-site upgrading with a requirement that houses be moved two metres from the river. On urban planning in these cities, see Bunnell et al. (2013).
Overall, then, Ahok, broke with Jokowi's pledge to relocate rather than evict (gusur bukan geser). The reason was political: Ahok was building a reputation as a can-do leader who could bulldoze through the objections of corrupt politicians, bureaucrats and other recalcitrants in order to fix Jakarta's many pressing problems. This approach was evident in his handling of budgetary, planning and other issues, and it earned him support, especially among the urban middle class. As resistance from poor communities affected by evictions mounted, Ahok became increasingly impatient with them, often denigrating them as stubborn and ungrateful, and depicting them as just another vested interest that stood in the way of his plans to improve the city (see, for example, Salim 2015).
A New Political Deal: The 2017 Election
The 2017 Jakarta election attracted much attention among analysts for the religious polarisation it engendered (e.g., Lim 2017). In September 2016, in the lead-up to the official campaign period, Ahok, who is a Chinese Christian, spoke at an event in which he warned his audience not to be “fooled” by people who use a particular Quranic verse to argue that Muslims should not be governed by non-Muslims. Ahok's opponents seized upon the comments, alleging that he had insulted the Islamic religion; this led to a series of demonstrations – the largest in post-Soeharto Indonesia – calling for Ahok's arrest for blasphemy. Ahok was eventually charged, and his trial proceeded throughout the campaign period. His two opponents took advantage of this issue to mobilise a sectarian campaign against Ahok, which contributed greatly to his eventual defeat in the second round held on 19 April 2017, in which Ahok won 42.05 per cent of the vote, compared to 57.95 per cent won by Anies Baswedan, a Muslim intellectual and former minister of education and culture.
While religious mobilisation was the determining issue in the campaign, indicated by how voting patterns broke down along religious lines (Warburton and Gammon 2017), much of the sectarianism occurred offstage, in private religious meetings and on social media. Despite large rallies opposing Ahok and calling for his incarceration, Ahok's rivals mostly avoided raising the religious issue directly in public election campaigning, such as at campaign rallies, in advertising and in televised debates. Instead, issues of urban planning and management, as well as social welfare, featured centrally. For example, the candidates presented differing ideas on housing. While Ahok made much of his plans to transform Jakarta into a modern city like Singapore, he also promoted construction of more rusunawa for poor Jakartans. Anies and his running mate Sandiaga Uno (commonly known as Sandi) proposed a programme of subsidised housing whereby Jakartans would be able to purchase homes without making a down-payment. This programme targeted persons with a monthly income of at least 7 million rupiah, putting it out of the reach of the poor.
The evictions that had occurred under Ahok also became a campaign issue. Though opinion polls indicated that the issue was not a major concern for Jakartans as a whole (only 10 per cent identified evictions as a policy of Ahok's that they disliked in one poll conducted two weeks before the final round: Charta Politika 2017: 50), both Anies and the third candidate, Agus Harimurti Yudhoyono, the son of former president Susilo Bambang Yudhoyono, criticised the harshness of Ahok's approach. Agus promised that he would “build without evicting” (membangun tanpa gusur) and, when challenged at a debate, said he would achieve this goal by “on-site upgrading” (Kahfi 2017). Much of Agus's campaign was targeted at poor Jakartans, so it is telling that his slogans matched those of the UPC/JRMK and other urban poor activist groups. Anies promised that he would not evict citizens but only temporarily “relocate” them (Astiana 2017).
Meanwhile, Ahok now had little support among activists. In the 2012 election, many civil society activists had rallied around Jokowi and Ahok, giving rise to a new mode of volunteer-based election campaigning (Suaedy 2014). By 2016, much of this support had evaporated. Urban poor activists, in particular, felt betrayed by the evictions that had occurred under Ahok, and were worried about what his programme of a “New Jakarta” would mean for the poor (Budiari 2017). However, in the lead-up to the first round of the election, held on 15 February 2017, it was not yet clear which of Ahok's opponents would advance to the second round. Accordingly, UPC/JRMK did not throw their weight behind any candidate, although some individual activists supported Anies or Agus. Instead, the groups campaigned on the theme “Gusur Ahok di putaran pertama“ – “Evict Ahok in the first round”.
Once Ahok and Anies won in the first round, knocking Agus out of competition, UPC/JRMK started to develop a detailed strategy. Ahok's record of evictions meant they wanted to make a deal with Anies. As already noted, Anies had spoken out against evictions, even though this was not a main plank of his campaign, which instead emphasised such matters as the zero-deposit housing programme, a small-scale entrepreneurship programme, and education and health care. Even so, the JRMK activists were assisted by the fact that they had an entry point into the Anies campaign: Marco Kusumawijaya, an urban architect, was the founder and former director of RCUS (Rujak Centre of Urban Studies). Kusumawijaya had long-standing connections to the urban poor activists of UPC and he was personally close to Sandiaga Uno, Anies's running mate. Early on, Marco had planned to run alongside Sandi himself when the latter was thinking of standing as gubernatorial candidate. Kusumawijaya become the main advisor to the Anies-Sandi team on urban issues.
In seeking cooperation with the Anies campaign, the UPC/JRMK activists, learning from their past experience, wanted to come up with a binding agreement that they would be able to use to hold Anies to his promises, should he be elected. In addition, they wanted the agreement to be detailed, unlike that of 2012, which had been couched in general terms. Finally, they wanted an agreement that would embody, as directly as possible, the concrete demands of the urban communities where they conducted advocacy work.
In order to bring such an agreement into being, the activists worked at three levels: a series of “kampung teams” compiled demands in the communities, a UPC/JRMK team consisting a mixture of urban-poor community leaders and middle-class organisers coordinated the enterprise, and an expert team drafted the agreement and negotiated with representatives of the Anies campaign. The kampung teams worked in sites spread across eight urban precincts (kelurahan) and 31 kampung. Some of these locations were long-term UPC/JRMK base areas, and others were places where the group had not previously been active. All were locations where residents had experienced evictions, or were expecting them within two years. Coordinators in these kampung – who were either longstanding JRMK community leaders or representatives chosen by community members – held small-scale consultative meetings (musyawarah) to determine residents’ aspirations. These aspirations were then compiled by JRMK. Coordinating this effort were mostly members of the urban poor, notably the JRMK community leaders, such as Gugun, as mentioned above.
Once they were compiled, the UPC/JRMK activists funnelled these aspirations up to the expert team, which consisted of academics, architects, urban experts and legal specialists, including some from organisations involved in urban issues, such as Marco Kusumawijaya's and Elisa Sutanudjaja's RCUS. Many of them were members of “Forum Kampung Kota” or Forum for City Kampung, a WhatsApp communication group established in early 2014 by Sandyawan Sumardi, a famous Catholic social activist and the head of Ciliwung Merdeka, an NGO concerned with people who live along the banks of the Ciliwung River. There were also scholars from the architecture department at Universitas Indonesia who had been engaged in UPC activities for a long time, including Herlily (a lecturer who had for years been bringing her students to UPC kampung and engaging them in various pro bono improvement projects). After receiving inputs from the kampung, via the JRMK coordinators, the expert team then drafted the agreement, translating the demands from the kampung into technical language, and negotiating with representatives from the Anies campaign. When they felt that the demands raised at the kampung level were vague, the UPC/JRMK coordinators contacted the kampung coordinators to seek more precise data, including about the legal and zoning status of disputed land.
It should be emphasised that this was a self-directed and self-financed effort. As mentioned earlier, UPC had not accepted funds from outside donors since 2009, and it applied this principle to electoral engagement, avoiding the taint of money politics. The kampung teams were separate from the Anies-Sandi campaign team, and if kampung coordinators were discovered to be already involved in the Anies campaign, they were asked to step aside from that role. UPC/JRMK also did not accept money from the Anies campaign to fund the various consultative meetings and coordination activities. Instead, kampung residents collected small donations to cover coordinators’ transportation costs and compensate them for income they lost by attending meetings. One of the UPC/JRMK coordinators, Guntoro, explained that they did not integrate themselves into Anies's success team because they wanted to maintain their independence:
Our target was to get a policy on halting evictions and providing housing rights from the new governor. That was our focus. If we were part of his team, our focus could be blurred because they have other programmes as well […]. We did not want our own team to be absorbed and used for things that were not our main focus. (Interview, 17 July 2017)
The product of this process was a political contract eventually agreed between the group and the Anies camp. This was called a “Perjanjian Tanah Penggarap” (which might roughly be translated as contract on land use) and was signed by Anies, Sandi and representatives of 31 kampung communities on 8 April 2017, 11 days before the second-round vote. The contract was an 18-page document that used the language of a legally binding instrument, purportedly obliging the parties to abide by the commitments it contained and, in the event of any violation, giving them the right to “sue the party in breach of contract in accordance with prevailing laws and regulations” (article 5). 5 The agreement obliged the 31 kampung representatives who were signatories to secure first-place victories for Anies in 125 voting stations in 31 locations, spread across eight precincts (kelurahan) and six subdistricts (kecamatan). The agreement listed all 125 of these voting stations, the vast majority of which were in areas where JRMK had long been active and which were sites of long-running land disputes (and where, it might be added, Anies was unlikely to lose to Ahok).
Legal experts have pointed out that the agreement is almost certainly not legally enforceable, not least because Anies signed it as a candidate rather than as governor (interview, Aldo Felix Januardy, 3 August 2017; see also Hukumonline 2009).
In return, Anies and Sandi promised that, if elected, they would fulfil the 46 listed points, each of which had been devised through the iterative process described above. The points were detailed and explicit. For example, point number 10 obliged the candidates to change the “Detailed Spatial Plan (Rencana Detail Tata Ruang) of Jakarta” so as to switch the legal status of land “in the Elektro Kampung, RW (Neighbourhood) 17, Penjaringan Precinct, Penjaringan Subdistrict, North Jakarta), from an industrial zone to a zone of very small houses (R2)”. Most of the 46 points listed very specific land use and planning changes of this kind, detailing a particular location – sometimes a specific street – and outlining the exact change in zoning and legal title that was required. In most of these cases, the purpose was to provide secure legal status to residents in settlements that had already been affected by, or were slated for, evictions. Several items obliged Anies and Sandi to legalise the status of street vendors in particular locations, or to mediate on their behalf with other authorities (such as the authority running the Sunda Kelapa port). A couple of points promised financial and employment assistance to becak drivers, who had been unable to work in most of Jakarta since the Sutiyoso era. The contract acknowledged that if Anies and Sandi lost the election, the agreement would not be valid. It also committed the pair, if they won, to establish a General Local Public Service Unit (BLUD, Badan Layanan Umum Daerah) to work on land security for the poor and to enforce the points in the agreement (article 46). The BLUD was part of the movement's long-term strategy: through it, the activists wanted to institutionalise an anti-evictions policy and spread it to other kampung in Jakarta over the long term.
The agreement also explicitly stated that Anies and Sandi would implement the points demanded by UPC/JRMK “insofar that they do not violate the law and existing rules and regulations” (article 3). This item was inserted on the request of Anies's team and reflected the “adversarial” nature of the relationship, indicating that Anies-Sandi's side would not simply agree to every demand from the urban poor. In order to anticipate this proviso, the expert team, using information supplied from the kampung, had been able to specify the legal status of much of the land under dispute and formulate demands in terms that accorded with spatial planning rules, especially with regard to zoning. This accounts for why the agreement contained so much technical detail. The expert team had been careful to include concrete demands that could be acted upon by the Jakarta government, avoiding general claims, sweeping rhetoric, and demands that required action by national rather than provincial agencies.
The spirit of what UPC/JRMK were trying to achieve through this agreement was not land reform, in which recipients received full legal title over land, but maximisation of land use for the poor instead of commercial interests. Most of the items included granted use rights (hak pakai), mostly to residents but in some cases to community cooperatives or associations. The goal was to ensure the land would not be converted into a capital asset and sold by recipients, but would instead perform a social function. The model, including the establishment of a land agency to monitor and control use of the assets, was based on an approach pioneered in Bangkok, where an organisation called CODI (Community Organisation Development Institute) ran a programme providing housing to the poor in 2003–2007 (Yap and Wandeler 2010). UPC activists had forged connections with their Thai counterparts and learned about this approach through the Asian Coalition for Housing Rights (ACHR).
Once the contract was agreed upon, it was the job of the kampung coordinators to campaign for Anies in their communities and deliver the promised ballot-station victories. Again, they did not accept funds from the Anies–Sandi success team, nor did they engage in much expensive door-to-door campaigning. Instead, they spread the word through their existing networks that the agreement had been reached and that it was now the community's job to deliver on its side of the bargain. This approach was effective, and Anies–Sandi won by significant margins in all designated voting stations, with a vote total at these sites that increased by 49.48 per cent, from 26,212 in the first round to 39,189 in the second round. However, it should be noted that this increase was not significantly higher than that achieved Jakarta-wide, where the Anies–Sandi vote rose by 47.72 per cent, from 2,193,530 to 3,240,332.
After the Victory: From Adversarial to Cooperative Linkage?
Immediately after Anies's victory, the challenge for the activists was to ensure that their political contract would get translated into policy. This was an important phase. After the election in 2012, activists had failed to build institutionalised links with the new government, instead placing trust in Jokowi's personal goodwill. In 2017, activists tried to build direct links with the incoming city administration.
They did so via a new body that Anies and Sandi established, a “synchronisation team” (tim sinkronisasi). The main task of this team was to devise a strategy to ensure that the promises the governor-elect had made during the campaign would be translated into concrete policies and, as far as possible, funded through the provincial budget which Ahok had drawn up earlier in 2017. The team was coordinated by Sudirman Said, a former minister of energy and mineral resources, and had 27 members, including respected figures such as Bambang Widjajanto, a former commissioner of the national Corruption Eradication Commission and Edriana Noerdin a gender specialist. Critically, it also included Marco Kusumawijaya, the urban affairs expert who was the major point of contact between the JRMK activists and the Anies camp, as well as NGO activists Irvan Pulungan (an expert on climate change) and Sukma Widyati (an expert on women's issues) who were also knowledgeable on urban issues (Rudi 2017). All had been members of the Anies–Sandi campaign team.
The role of the NGO activists in the synchronisation team was crucial. They can be regarded as “policy brokers” (Sabatier 1998), providing the critical linkage between the urban poor activists and the incoming government, and translating articles of the political contract into concrete policies. On topics like land use and spatial planning, Indonesian public administration is a forest of complex rules and regulations. Experts familiar with this domain are needed in order to steer movement demands through the bewildering formal language of the bureaucracy, and to avoid pitfalls such as inter-agency rivalry or overlap. The urban NGO activists involved in the synchronisation team had the task of trying to negotiate the items from the political contract through the bureaucracy, inserting them as policies into the incoming government's programme and ensuring that they had a budget allocation. 6
It is not unique for civil society organisations to become policy brokers in this way: Lay (2015, 2017) has explored policy lobbying in the national parliament on a bill on Aceh Government.
It is too early to tell how successful they were in this task, with Anies having only recently been sworn in as governor at the time of writing. However, active interaction between the kampung-level organisers and these NGO activists are continuing, at least in some of the kampung that were at the centre of the deal. After becoming governor, Aniesformalised a programme of building new housing for the urban poor in 16 kampung, calling their programme one of “land consolidation” (Sari 2017). One of the priorities for redevelopment is Kampung Akuarium, with residents being targeted for rehousing after they were evicted and their homes demolished in April 2016 during the Ahok era (Nugroho 2017). The plan to rebuild the 16 kampung has been included in the 2018 Jakarta local budget (APBD), and several of the target kampung are base areas of UPC that lent their support to the political contract. In short, in the immediate post-electoral context the adversarial linkages of the election period seemed to be giving way to a more structured policy linkage, but the ultimate outcome will depend on government follow-through in the years to come.
Conclusion: Elections, Popular Agency, and Political Linkages in the Context of Weak Institutions
Involvement by urban-poor activists in electoral politics in Jakarta between 2012 and 2017 involved ‘learning by doing’ – a phrase applied by Ford (2014) to labour-movement experimentation with electioneering in the Riau Islands. The Jakarta activists learned from the broken promises of 2012 and approached a challenger to create a new political contract that was, in their eyes, more concrete and more enforceable than what they had produced in 2012. Rather than being couched in broad aspirational terms, as with the 2012 agreement, the new contract contained very specific demands regarding land status and related issues in specific locations.
We must emphasise the positive aspects of this evolution for the urban poor activists and community members involved. Electoral engagement in 2017 involved a far deeper collective process than in 2012. It provided an opportunity for this movement to expand political participation at the kampung level among members of Jakarta's poor. Critically, it allowed members of these communities to learn about the urban planning regime that shapes their lives. For example, many kampung dwellers had the opportunity, for the first time, to access spatial planning documents and observe the locations and status of their homes and communities in those documents, and to think strategically about what sort of goals and achievements such documents allowed, in the long term. The importance of this process of learning should not be underestimated for movement participants.
Just as importantly, the 2017 engagement was a process that, in critical respects, led participants away from the patterns of vote buying and clientelistic exchange that are so often seen in poor communities during Indonesian elections. When poor Indonesians sell their votes in one-off material exchanges, they are underlining the paucity of their hopes for delivery of ongoing benefits from their elected representatives. Political contracts of the type discussed here are an attempt to bind the politician they support to future delivery, and to ensnare that politician in relations of accountability. Also, as we have seen, the urban-poor activists deliberately avoided being drawn into the web of clientelistic relations that arise around elections, refusing to join the Anies–Sandi campaign team or to accept cash payments from it.
In one critical respect, however, this evolution did not involve the type of learning that might be anticipated from the literature on political linkages: we do not see evidence of gradual evolution toward a model of programmatic politics in which particularistic demands are combined, generalised and expressed in terms of a platform with broad appeal. On the contrary, electoral engagement in 2017 involved a process of disaggregation rather than aggregation of interests. To be sure, the activists did not set aside their long-term goals, and they maintained a broad vision of wishing to achieve social justice for the poor. But they realised they had to start from very concrete goals. Learning from the experience of 2012, the urban-poor network in Jakarta moved from the general to the specific, trying to hold their candidate to a series of extremely precise demands, which they researched and moulded to ensure they would be achievable within the existing regulatory framework.
In part, this pattern resulted from dynamics associated with the urban poor as a social group. Negotiating complex social terrain is integral to the lives of the urban poor. As Ian Wilson has observed with regard to Jakarta:
Poor people operate politically in everyday life on multiple levels, which requires the managing of complex relationships, including with the powerful, in order to hedge risk, keep options open, encroach on spaces and opportunities formally denied them and defend any gains made. (Wilson 2017: 4)
According to Wilson (2017: 5), the result is politics that is “deeply pragmatic, instrumental and largely non-ideological”. As we considered briefly above, such observations accord with studies of the politics of the urban poor in other parts of the world. In order to survive, poor city dwellers forge instrumental links with numerous sources of authority, protection and assistance. Their struggle for existence involves not just cooperation but also competition for scarce resources, opportunities and patronage, and they often seek assistance from overlapping and competing networks, mediators and patrons. In short, these are patterns of life that can readily give rise to atomisation rather than to collective action. It is surely possible to view the disaggregation of demands in the 2017 Jakarta election as reflecting these underlying patterns. Indeed, other analyses of electoral engagement by groups claiming to represent the interests of the urban poor point toward similar conclusions. For example, Gibbings, Lazuardi and Prawirosusanto (2017) documented how the attempt by APKLI to rally informal traders’ support for Prabowo as a presidential candidate in 2014 stumbled badly when some of the group's affiliates felt compelled to support Prabowo's rival, Jokowi, because of their ties of obligation to local political patrons, or simply because they felt the election was too remote from their own affairs.
A similar dynamic can be seen among rural and farmers’ movements in contemporary Indonesia. The vast majority of such movements are built around struggles over particular land disputes, which typically pit village communities against outside developers, such as plantation companies or the state forestry agency. These communities themselves are often divided, as the developers seek local agents through which to gain access to land or negotiate on their behalf, and they frequently result in compromise solutions involving compensation or partial recognition of land rights. Accordingly, it has sometimes been observed in contemporary Indonesia that once their immediate demands with regard to rights to land have been redressed, such movements tend to sink back into passivity (Bachriadi 2010). This is in contrast to industrial workers, for example, many of whose demands (such as for improved wages and working conditions) are more readily generalisable across entire population groups, and whose conditions of daily labour on the factory floor facilitate rather than inhibit collective action. Therefore, it is perhaps not surprising that it is among industrial workers and trader unions that we see the most sustained and effective attempts to intervene in electoral politics.
What broader lessons can we derive from the Jakarta case about the nature and evolution of lower-class political agency in contemporary Indonesia? At a general level, and as already emphasised, we can point toward the critical importance of complex patterns of brokerage and mediation in this form of politics, reinforcing the conclusions drawn by Chaterjee, Auyero and others about the politics of the urban poor elsewhere. However, focus on mediation and coalitional politics should not blind us to the scepticism that underpinned the political contract forged during the 2017 Jakarta election. On the contrary, such a focus highlights the perseverance and inventiveness with which Jakarta's urban poor, like disenfranchised groups elsewhere, “struggle to make their claims to governmental care” (Chaterjee 2001: 75), actively seeking out political alliances and seeking new methods to make them as advantageous as possible.
However, attention to the modes of lower-class agency should not lead us to exaggerate their effects. Scholars have been observing electoral experimentation by groups representing lower-class interests in Indonesian politics for the better part of a decade. We have seen a steady accumulation of the number of case studies, and a commensurate accumulation of examples of political contracts agreed by social movements and candidates. Nevertheless, it is difficult to conclude that there is similarly steady accumulation and coordination of such activities, or even an accumulation of greater and greater gains achieved by poor people from such engagement. To be sure, the urban poor activists in Jakarta sought to achieve such goals. However, at the micro level, our study of the Jakarta election suggests that while learning is taking place, and that this model has the potential to challenge the clientelistic patterns that dominate Indonesian political life, the overwhelming impression is still of continued atomisation and fragmentation of social movement electoral engagement. There is significant scope for highly focused and one-off transactions of the sort discussed in this article, which includes a constant pull towards a clientelistic model, but the generalisation of gains achieved to more than just isolated pockets of social movement activism will ultimately require more sustained efforts to build programmatic politics.