The limits of autocratisation in Indonesia: power dispersal and elite competition in a compromised democracy

Abstract

How does democracy survive in a polity that has witnessed consistent autocratisation trends for an extended period of time? In Indonesia, new patterns of autocratisation and broader democratic decline emerged in the late 2000s, but unlike some of its Southeast Asian neighbours, the country has not crossed over into the territory of full-fledged autocracy. This article argues that two interrelated factors explain this outcome: first, power was so widely dispersed in the early democratic transition that its renewed monopolisation would be hard to achieve; and, second, this power dispersal has underpinned an intense inter-elite rivalry that rebels against any attempt at restoring personalist autocracy. At the same time, elites have cooperated in the weakening of democracy as long as a residual framework for competitive rotation of power remains in place.

Keywords:

• Democracy
• autocratisation
• Indonesia
• democratic decline
• elite competition

Introduction

In an age of democratic crises, scholars of comparative politics have extensively debated the meaning, drivers, and limits of autocratisation. Most fundamentally, there has been no consensus on its definition. While some authors have understood autocratisation as a process that leads to regime change from democracy to autocracy (Tomini, Gibril, and Bochev 2023, 121), others have defined it more broadly as covering ‘both sudden breakdowns of democracy […] and gradual processes within and outside of democratic regimes where democratic traits decline – resulting in less democratic, or more autocratic, situations’ (Lührmann and Lindberg 2019, 1099). Thus, in the latter concept, autocratisation can, but does not have to, produce regime change towards full autocracy. As an umbrella approach, it includes the possibility of autocratisation merely reducing the quality of democracy in a democratic regime without overturning it. Democratic decline, in this paradigm, is a sub-category of autocratisation, describing the weakening of democratic institutions and controls with an open outcome.¹ In is within these parameters that this article introduces the case of Indonesia, the world’s third-largest democracy. While experiencing significant autocratisation trends that led to democratic decline, Indonesia has not become an autocracy – raising questions about both the triggers for the decline and the country’s ability to withstand full-fledged autocratisation.

A growing body of literature has investigated the various dimensions of democratic decline in Indonesia in recent years. Reflecting a slow but noticeable decline of Indonesia’s democracy scores in international indexes, authors have highlighted trends of executive aggrandisement (Mujani and Liddle 2021), narrowing electoral space (Lay et al. 2017), intensifying pro-government media coverage (Tapsell 2017), repression of dissent (Mashabi 2020), increasing oligarchic power (Tadjoeddin 2019), and polarising religious conservatism (Bourchier 2019). President Joko Widodo’s decision to support the 2024 vice-presidential nomination of his son Gibran, and to mobilise state resources to secure his election, led to further concerns for the health of Indonesian democracy. To be sure, none of these patterns have been uniquely Indonesian. Indeed, the country joined a wave of global democratic decline that began in the mid-2000s and has affected old and new democracies alike (Diamond 2021). But within these worldwide developments, Indonesia’s democratic decline carries particular weight. This is because Indonesia was long depicted as a model Muslim democracy that successfully navigated the traps of reconciling democratic and religious orders (Anwar 2010). Thus, its democratic turbulence is of significant comparative interest, especially for the Asia-Pacific region and the group of struggling majority Muslim democracies.

However, while it is important to understand the drivers and dynamics of Indonesia’s democratic decline, it is equally crucial to explore the reasons for democracy’s survival. Unlike many of its Southeast Asian neighbours, Indonesia sustained a functional electoral democracy for the last quarter of a century. In the same period, Thailand experienced two coups; Myanmar first emerged from and then returned to military rule; the Philippines saw one president removed by mass protest and another initiating killings of petty criminals before the son of a former dictator took charge in 2022; Malaysia underwent an unstable transition after 2018; and the rest of the Southeast Asian region recorded enduring authoritarianism or semi-authoritarianism (Tudor and Slater 2021). Indonesia, by contrast, witnessed stable presidential rule after constitutional amendments became operational in 2004, with two presidents completing their terms without any attempts to impeach them. Indonesia’s regime is now best described as a compromised but functionally solid democracy. Given the democratic instability in the region, then, we need to ask not only why democracy declines in Indonesia – but why does it survive?

To find answers, this article explores the limits of autocratisation in Indonesia. It argues that two interrelated factors have prevented Indonesia from transitioning into authoritarianism: first, the power dispersal instituted by the early post-1998 reforms; and, second, the distinct lack of interest in a renewed project of personalist autocracy among the majority of elites (defined here broadly as actors with privileged access to economic and/or political resources). Conceptually, the article challenges the main notion in much of the autocratisation literature that views elites as the main actors of autocratisation (Lührmann and Lindberg 2019, 1096) and locates resistance to it mostly in judges, anti-corruption commissions, opposition parties, internal regime soft-liners, and civil society activists (Tomini, Gibril, and Bochev 2023, 122–124; Boese et al. 2021). Rather, the discussion here demonstrates that many elites, while working to limit the effects of democracy on their privileged position, have an interest in upholding it as the foundation of inter-elite power distribution. Following from that, and contrary to assertions by proponents of the cartel approach to Indonesian politics, the article highlights that competition between elite actors remains intense, with each of them trying to keep others from emerging as a new hegemon. I argue that this sense among Indonesian elites that a controlled democracy serves their interests better than authoritarianism is the main barrier to full autocratisation. Civil society activism and public opinion also play a role, but I contend that they are often secondary factors.

This article develops these arguments in four steps. The first segment discusses Indonesia’s democratic trajectory from its high point in the mid-2000s to the subsequent stagnation and decline. The second part identifies the limits of autocratisation and the corresponding pools of democratic survival. In the third part, I situate the post-1998 power dispersal as one of the main reasons why any autocratic challenger would face an uphill battle to restore authoritarianism, while the fourth analyses the continued inter-elite competition that keeps democracy alive. The conclusion, finally, contextualises the Indonesian case in the broader debate on democratic survival.

Indonesia’s democratic emergence and decline

Before exploring the limits to autocratisation in Indonesia’s contemporary polity, we need to briefly sketch its democratic development since the fall of long-time strongman Suharto in 1998. In that year, Indonesia entered a messy democratic transition, marked by institutional experiments and social conflict. There were four post-Suharto presidents in six years, reflecting the instability of that period. But the elite passed constitutional amendments in 2002 that became fully operational in 2004, and these changes stabilised the political system (Horowitz 2013). At the heart of the reforms were the direct election of the president and the setting of a high bar for presidential impeachments. At the same time, however, parliament received more powers, too, and new oversight agencies were created. In combination, these changes made Indonesia both more democratic and more stable – a fact that was recognised by international democracy indexes. Freedom House, for instance, upgraded Indonesia’s status in 2006 to ‘free’, and others followed suit. Based on these comparative measures (including V-Dem’s Liberal Democracy Index), Indonesia reached its democratic peak in about 2008. It was in this phase that Indonesia was often presented as a model for other young democracies (MacIntyre and Ramage 2008).

But by the late 2000s, it became clear that Indonesia’s democratic project had stalled (Tomsa 2010). The appetite for further reforms was fading in both the elite and the broader population, with overall stagnation as a result. President Susilo Bambang Yudhoyono (2004–2014), whose first term had witnessed wide-ranging political reforms and even the peaceful resolution of the separatist conflict in Aceh, merely administered the polity in his second. His successor, Joko Widodo (2014–2024), initially promised more reforms but quickly settled into the management of the status quo. Indeed, under his rule, the decline of democratic quality that had begun under Yudhoyono accelerated. In V-Dem’s Liberal Democracy Index, Indonesia scored 0.43 in 2020 – the lowest since 1999. Freedom House had already downgraded Indonesia to ‘partly free’ in 2013, and maintained that level throughout the Widodo presidency. In broad terms, Indonesian democracy declined from its 2008 peak in a gradual and moderate manner, diminishing its qualitative substance but leaving the formal democratic framework in place.

The democratic decline captured by the indexes manifested itself in numerous political and social arenas. To begin with, the Widodo government engaged in executive aggrandisement, defined here as the ‘incremental but systemic dismantling of checking mechanisms that liberal democratic constitutions typically put in place to ensure the accountability of the political executive’ (Khaitan 2019, 343). In 2019, for instance, Widodo cooperated with parliament to weaken the Corruption Eradication Agency (KPK) that had previously arrested many political leaders and government officials (Mujani and Liddle 2021). In parliament itself, Widodo manufactured a supermajority during his first term although he had begun his presidency only with minority support. In order to achieve this supermajority, Widodo used the regulatory powers that Indonesian presidents possess over political parties. Applying this authority, he helped remove the leadership of two opposition parties and had them replaced by pro-government politicians. This, in turn, undermined parliament’s oversight functions. In his second term, 82% of legislators were members of government-affiliated parties. In 2021, moreover, Widodo’s chief of staff tried to take over one of only two remaining opposition parties, and while this attempt ultimately failed, it paralysed the party and its oversight role.

There has also been a significant narrowing of the electoral space (Mietzner 2020a). Entry requirements for new electoral parties were tightened considerably between 1999 and 2019. In 1999, parties had to have branches in half of the provinces and half of the districts of those provinces in order to participate in elections. By the 2014 elections, this requirement had increased to 100% in provinces, 75% in districts, and 50% in sub-districts. At the same time, the parliamentary threshold increased from 2.5% in 2009 to 3.5% in 2014 and 4% in 2019. The main effect of these changes was that the cost of establishing a new party grew while the chance of entering parliament decreased. Predictably, only wealthy actors with large resources could take on this cost challenge and risk calculation. In presidential elections, the trend was similar. Presidential nominations were limited to parties or party coalitions that commanded at least 20% of the seats in parliament or obtained at least 25% of the votes in the preceding legislative elections. Combined with the increasing costs of presidential campaigns, this reduced the number of competitors over time. In the 2004 elections, five candidate pairs ran; in 2009, there were three; and in 2014 and 2019, the field shrunk to two nominated tickets.² In both the legislative and presidential election arenas, then, restrictive regulations and growing cost pressures limited the competition to affluent, and often oligarchic, actors.

Democratic decline was also reflected in the increasing integration of of large media empires into the government’s support base (Tapsell 2017). In the 2014 elections, the support of media companies towards presidential candidates was still divided between Widodo and his opponent, Prabowo Subianto. Subsequently, however, Widodo managed to pull most media enterprises into his orbit through both coercion and accommodation. Hary Tanoesoedibjo, for instance, owns a media conglomerate which in 2017 included ‘three free-to-air national television networks that command some 40% of prime-time audience share’ (Suzuki 2017). In 2014, he still supported Prabowo, but in 2017 declared his loyalty to Widodo (after having been told by the attorney-general’s office that a case had been opened against him). His daughter was later given a deputy ministry position in Widodo’s administration. Surya Paloh, another media oligarch and head of a party, had several proxies sitting in both Widodo cabinets. Erick Thohir, the owner of a media group that included a daily, was Widodo’s 2019 campaign manager and later became a minister. In this climate, coverage of the president became exceedingly friendly, delivering parts of the explanation for Widodo’s continuously high approval ratings – even during difficult times such as the COVID-19 pandemic.

Repression of dissent contributed to democratic decline, too. Government critics, both at the elite level and among ordinary social media users, were selectively pursued under Widodo to disincentivise the expression of dissenting views (author interview with Usman Hamid, director of Amnesty International Indonesia, Jakarta, 15 August 2023). The main instrument for this was a 2008 law that allowed authorities to criminalise supposedly slanderous or otherwise offensive online statements against officials. In Yudhoyono’s second term (2009–2014), there had been 74 such cases; in Widodo’s first (2014–2019), this number escalated to 233, with 82 of them directly related to the alleged insult of the president (Mashabi 2020). But other regulations were used as well: Islamist leader Rizieq Shihab, one of the most prominent Widodo critics, was first driven into exile by numerous charges and then imprisoned on his return in late 2020 because of ‘violations’ of COVID-19 rules. These repressive actions ultimately had a stifling effect on society: in a September 2020 opinion survey, 69.6% of respondents stated that citizens were ‘increasingly’ afraid of stating their opinion (Safitri 2020).

We already noted that oligarchs were major beneficiaries of electoral narrowing, and that media entrepreneurs increasingly became part of the government support structure. But these phenomena were part of a broader trend of the oligarchisation of politics (Fukuoka 2013). The number of oligarchs – defined here as actors whose primary political resource is the directly controlled, personal possession of wealth – in cabinet and other key political posts reached a new height in the second Widodo period. Many ministers in charge of major portfolios owned companies in business areas directly relevant to their portfolio. Among them were Luhut Pandjaitan, coordinating minister for maritime affairs and investment; Airlangga Hartarto, coordinating minister for the economy; Erick Thohir, minister for state enterprises; Sandiaga Uno, minister for tourism; and many others. Such a high concentration of oligarchs in cabinet was a departure even from the times of Suharto’s autocratic regime: under him, oligarchs were important financiers of politics, but were mostly kept away from direct political power. Beginning under Yudhoyono and intensifying under Widodo, oligarchs were put in control of leading ministries, handing policymaking to the top socio-economic elites. This, in turn, further consolidated the already entrenched exclusion of lower-class citizens from political influence.

As in other cases around the world, growing religio-political polarisation is also part of Indonesia’s democratic decline narrative (Afrimadona 2021). In Indonesia, the major cleavage in this regard is between supporters of religious pluralism on the one hand and advocates of a stronger role of Islam in state organisation on the other. The polarisation between the two sides was fed by growing Islamic piety since the democratic opening of the late 1990s but first became politically decisive in the 2014 elections. In that election, Widodo represented the pluralist camp and Prabowo the Islamist forces. Widodo prevailed in a fierce contest marked by disinformation and smears, and subsequent elections produced more polarisation episodes. The 2017 Jakarta gubernatorial election, which involved a Christian of ethnic Chinese descent, was one such episode, as was the 2019 re-match between Widodo and Prabowo. In the context of these polarisation contests, Widodo deployed repressive measures against radical Islamists (Fealy and White 2021). Two leading Islamist groups were banned, in 2017 and 2020, respectively. Polarisation levels declined somewhat after Prabowo joined the Widodo government in 2019, but the events of previous years had left Indonesia’s democratic fabric damaged.

In sum, there is ample evidence of democratic decline in Indonesia that confirms the evaluations of international democracy indexes. Nevertheless, the reduction in democratic quality in Indonesia has been less severe than in other countries, and unlike some of its regional neighbours, the country has not experienced democratic breakdown. On the contrary, Indonesia’s compromised democracy has been remarkably stable and remains functional in major aspects of democratic life. The task for us, therefore, is to explain what makes Indonesian democracy retain its stability and functionality despite gradually diminishing quality.

Limits of autocratisation

In order to understand why Indonesian democracy has not crossed over into autocratic territory despite its gradual weakening, let us revisit each of the arenas of democratic decline introduced above. In every case, the progressing democratic decline encountered strong limits that prevented descent into full authoritarianism. In the field of executive aggrandisement, for example, no president succeeded in personalising their rule. While both Yudhoyono and Widodo were powerful presidents thanks to the constitutional powers invested in them, the elites they aligned with did not allow them to establish non-institutional powerbases that could have enabled them to stay in office beyond their terms. Yudhoyono, when approaching the end of his second term, explored the idea of having his wife or brother-in-law succeed him, but that attempt was quickly aborted as it found virtually no support. Widodo, for his part, even launched a systematic and elaborate effort to gain a third term or to postpone the 2024 elections (Setijadi 2021). Despite Widodo’s high levels of popularity, however, the campaign failed because neither the elites in his ruling coalition nor the public backed it up. Indeed, Widodo’s own party rejected it.³ Thus, while elite actors sought to enter alliances with incumbent presidents to benefit from their authority to distribute patronage and other favours, this did not translate into the kind of personalist loyalty that executors of successful autocratisation projects such as Vladimir Putin could draw from. In Indonesia, presidential power remained bound to the institutional authority and limitations dictated by the existing order.

This also meant that elites watched carefully whether presidents stayed within their constitutional boundaries when dealing with other state institutions. Parliament, for example, forged strong bonds with both Yudhoyono and Widodo – and was happy to engage in democracy-damaging projects (such as the weakening of the KPK) if these served their joint interests. In doing so, however, the legislature always protected its autonomy and authority (Farhan 2018). It was this independent power that allowed parliament to negotiate with the presidency, and as such any attempt to reduce its leverage was fiercely rejected. Similarly, while incumbent presidents took some freedoms in interpreting court rulings, presidents generally respected the judiciary’s verdicts, even if they went against them. This was not because presidents enjoyed constitutionalism as a principle but because major violations of the rules of the game would have made them vulnerable to elite attacks.⁴ President Abdurrahman Wahid (1999–2001), for instance, was impeached partly because he appointed a police chief without seeking parliament’s approval, as was required by law. Hence, while executive aggrandisement in Indonesia has diminished democratic quality, it has mostly operated within legal-formal boundaries.

In the same vein, while electoral space has narrowed, elections are far from foregone conclusions, and having more resources than a competitor does not guarantee victory. Widodo’s rise from furniture entrepreneur in Central Java to president within a decade highlighted the opportunities that Indonesian elections still provide to those who manage to build meaningful connections with the electorate. While vote buying is rampant in many local legislative and executive elections (Muhtadi 2019), presidential elections involve too many voters for this strategy to be effective – and even in local contexts, there are limits to the practice. The 2024 presidential election provided further evidence of the continued competitiveness of electoral races and the chances of non-oligarchic actors. In the elections, the matriarch of the Indonesian Democracy Party of Struggle (PDI-P), former president Megawati Sukarnoputri, had to nominate a marginal party figure, Ganjar Pranowo, because his polling numbers were better than those of Megawati’s daughter, her preferred choice. Anies Baswedan, a former university rector, was also a candidate. Even the candidacy of Prabowo Subianto, Suharto’s former son-in-law, included some good news for Indonesian democracy: after failing twice in the 2014 and 2019 campaigns with proposals to return Indonesia to its pre-democratic constitution, he attracted voters in 2024 with a much softer message that no longer proposed to dismantle democracy. Indeed, he now denied that his previous campaigns aimed to do that (author interview with Prabowo Subianto, Jakarta, 14 September 2023).

The media landscape, while dominated by pro-government oligarchs, is also still producing competing streams of news to the citizenry (Tapsell 2017). This is because the various oligarchs owning the companies are often in strong rivalry with each other. Although the late 2010s saw exceedingly Widodo-friendly coverage, media conglomerates have an interest in releasing news that undermines their rivals and the political networks behind them. Therefore, news about corrupt behaviour of political actors continues to be available to the information market – it is just not delivered by the outlet whose owner might be discredited by it. In addition to this vested interest in sharing damaging information on competitors for power, there is also an economic incentive structure to provide interesting and non-streamlined information to customers. After all, media companies are businesses, and consumers are more interested in ­reliable – and sensational – news releases than in self-serving, monotonous coverage. Consequently, intense inter-company contestation and goals of profit-maximising have prevented the Indonesian media market from turning fully into a tool of government propaganda.

The repression of dissent has also been selective rather than universally implemented. While the police set up a special unit during the COVID-19 pandemic to monitor criticism of government officials and pursue perpetrators, the sheer amount of social media activity made persecution of all expressions of dissent impossible. In 2023, there were 197 million Indonesian Facebook users, 107 million users of Instagram, and 25 million Twitter users. To comprehensively screen this volume of posts and tweets, a much more sophisticated technological apparatus of supervision – such as the one deployed by China – would be necessary. Indonesia has not gone down this path, and has instead used selected cases to set precedents and create a deterrent effect. As noted above, this approach succeeded in motivating self-censorship, but it left large arenas of public discourse open. This, in turn, managed to convince the majority of Indonesians that democracy was being upheld, despite increasing monitoring of their opinions. On average, around two-thirds of Indonesians express satisfaction with the way democracy is practised, suggesting that tightened controls on dissent (which also more than two-thirds confirmed as occurring under Widodo) were not seen as overly problematic.

Likewise, the power of oligarchs – while immense – has not been all-encompassing. It is true that they have penetrated all layers of the political system, protecting their interests in the process. But oligarchs – who, it is worth recalling, are defined here through their wealth and not through a more classic concept of oligarchy as a small ruling elite – have not managed to monopolise the polity, and they have struggled to gain control of the highest political office available: the presidency. This has not been for lack of trying. Numerous oligarchs have attempted to win the presidency since 1998, but the post instead went to mainstream figures: Habibie was a technocrat, Wahid an Islamic cleric, Megawati the daughter of the country’s founder, Yudhoyono a reformist military officer, and Widodo a small-scale businessman from the regions. Oligarchs such as Paloh, Tanoesoedibjo, Thohir, or Aburizal Bakrie had to settle for second-tier posts as part of broader presidential coalitions led by non-oligarchic patrons. In these positions, oligarchs did much damage to democratic accountability processes, but their influence was checked by both presidents and other members of the presidential alliance. In other words, oligarchs became one collective actor among many rather than the hegemons that the oligarchy theory of Indonesian politics postulated (Hadiz and Robison 2004; Winters 2011; Fukuoka 2013).

Meanwhile, the threat of polarisation – although bringing Indonesia episodes of deep crisis – was counterbalanced by the need of all actors to seek compromise in order to access patronage resources. Despite Indonesia’s post-1998 reforms, patronage and clientelism remain the main principles of socio-political organisation (Aspinall and Berenschot 2019). As core elements of social life since pre-1998 regimes, they did not cause the democratic decline starting in the late 2000s, but they were obstacles to further democratisation. In regards to polarisation, however, they served as a barrier and neutraliser. As noted, Prabowo – after using Islamist forces as his vehicle in the 2014 and 2019 elections – joined Widodo’s government in 2019. Both Widodo and Prabowo mentioned their goal of overcoming polarisation as the reason for their alliance (author interviews with Joko Widodo, Jakarta, 14 October 2023; and Prabowo Subianto, Jakarta, 14 September 2023). Anies Baswedan, who won the 2017 Jakarta gubernatorial elections by aligning with Islamists, subsequently ran the capital as a catch-all politician. To be sure, polarisation interludes damaged Indonesia’s democratic quality, fuelling socio-religious intolerance and orthodoxy (Mietzner and Muhtadi 2018). But after elections, Indonesia’s key actors often prioritise collaboration over continued ideological contestation.

Hence, in each of the arenas of democratic decline, there have been formidable limits to further autocratisation. Indeed, as I will argue below, this common pattern in all fields is not coincidental. Rather, it shows a deliberate strategy of Indonesia’s leading socio-political forces to hollow out democratic substance but to let democracy survive as an organisational framework. This strategy, in turn, draws from two interrelated structural phenomena: first, the power dispersal instituted after 1998; and, second, the continued competition among elites for power that is, in their view, better regulated through a limited democracy than through autocracy.

Power dispersal

In discussing the reasons for Indonesian democracy’s survival, Slater (2023, 102) commented that this was mostly due to ‘Indonesian democracy’s founding mother: its vibrant, critical civil society. It is almost impossible to overstate just how important popular pressure and urban protests have been in keeping Indonesia’s cozy political elite in line’. There is no doubt that civil society remains a critical actor that Indonesian elites have to reckon with when making decisions. But it is also clear that civil society has become increasingly partisan, especially since the polarisation episodes of the second half of the 2010s (Mietzner 2020b). Many civil society groups have taken sides with either the pluralist or Islamist camp, leaving very few groups in the ideological centre that continue to fight for the defence of civil rights regardless of political orientation. The case of the KPK’s weakening was telling in this regard: some pluralist non-governmental organisations (NGOs) refused to join in its defence because they believed that the agency was infiltrated by Islamists. The reasons for Indonesian democracy’s survival, in spite of civil society’s increasing partisanship, must be located outside of the civil society arena, too.

One of the main barriers to full-fledged autocratisation in post-Suharto Indonesia has been the high level of power dispersal established in the early years of the transition. This wide-ranging distribution of power went well beyond that found in other post-authoritarian polities. In part, this was thanks to President B. J. Habibie’s desperate attempt to prove his democratic credentials by not only meeting but exceeding the demands made by the country’s leading reformers (Liddle 1999). As Suharto’s former deputy, Habibie’s only chance of staying in his post longer than just for a transitional period was to acquire a reputation as a radical reformer – and while he did not succeed in extending his presidency, the legacy of his reforms persists. Most importantly, he launched a decentralisation programme in 1999 that empowered the regions, and especially districts, over the centre. Originating from Sulawesi, Habibie was the first – and thus far last – president who was not from Java, Indonesia’s politico-economic power centre. It is widely accepted that had a Javanese succeeded Suharto, decentralisation would not have been pursued with such speed and intensity. Instead, Habibie left a political order behind in which fiscal resources – that in themselves are the basis for political power – shifted to the local level. By the 2020s, 43% of state revenue was spent by the regions.

Habibie’s decentralisation reforms were deepened in 2005, when direct elections for governors, mayors, and district heads began (Erb and Sulistiyanto 2009). These popular elections, which replaced ballots in local legislatures, gave local government heads a direct political mandate, boosting their position vis-à-vis the centre. Additionally supported by their increased fiscal powers, heads of local administrations turned into the primary political power holders in their regions – when they had simply been extensions of Jakarta’s apparatus prior to 1999. Anybody trying to overturn these powers of local leaders can expect significant resistance (author interview with Ganjar Pranowo, governor of Central Java, Semarang, 14 February 2023). In 2014, a coalition of parties supporting Prabowo – who had just lost the election against Widodo – passed legislation in the outgoing parliament that would have abolished direct elections of local government heads. But the latter mobilised against this move, informing the Minister of Home Affairs in no uncertain terms of their opposition. The public also rejected the new law. Outgoing President Yudhoyono, fearing a backlash from the regions and the public that could hurt his legacy, eventually issued a government regulation in lieu of law – that is, an emergency decree that allows the president to overturn laws by putting new stipulations in place until parliament can vote on them. When parliament reconvened, it did not challenge Yudhoyono’s decision, leaving the post-2005 reforms in place.

Another important decision made in the Habibie period that facilitated a wide dispersal of power related to the electoral and party system. In early 1999, it was decided to retain the system of proportional representation in the country’s legislative elections. Proportional representation had been used as the basis for elections both in the democratic experiment in the 1950s and during Suharto’s autocratic regime. In 1998, some actors had proposed to replace this system with a first-past-the-post, district-based system. Had this proposal prevailed, Indonesia’s polity would have taken a more majoritarian path than it eventually did. Instead, a wide range of parties representing diverse constituencies made it into parliament, and while we noted the gradual tightening of party creation and parliamentary threshold regulations, the party system remained without a clear power centre. Indeed, the party system levelled out over time: in the 1999 elections, no party achieved more than 35% of the votes; in the 2004 and 2009 elections, no party obtained more than 25%; and in the 2014 and 2019 elections, no party recorded more than 20%. At the same time, however, the party system was relatively stable, allowing competitors to consolidate: 10 parties were elected to the 2014 parliament, and five years later, nine of them re-entered the chamber, without a new entry. Accordingly, the institutional decisions made in the early democratic period prefigured a competitive party system without making it excessively fragmented.

The decentring of political power, as expressed in the party system, did not change significantly even after the introduction of direct presidential elections in 2004. In some contexts, such a shift to popular presidential ballots can produce majoritarian overreach, but it did not do so in Indonesia, at least not in a regime-changing manner. The reason for this is that despite having been equipped with strong popular mandates, Yudhoyono and Widodo opted for broad power-sharing rather than exclusive rule. Both included more parties into their coalitions than political mathematics would have required, and they handed cabinet posts to non-party actors as well. This inclusivist coalition building produced its own detrimental effects for democracy as the many vested interests of coalition members disincentivised reform (Mietzner 2023). But it also established constraints on potential autocratisation ambitions by presidents as any such attempt would have destabilised the coalition. Why, then, did presidents establish such broad coalitions in the first place? In addition to the usefulness of accommodating the many players generated by the electoral system of proportional representation, post-2004 presidents also drew lessons from Wahid’s impeachment in 2001. He was toppled because he had falsely believed that presidential power was absolute; Yudhoyono, Wahid’s minister at that time, concluded from this episode that stable government required inclusion rather than combative exclusion. Widodo, who initially promised to challenge Yudhoyono’s broad coalition approach, soon discovered its benefits and replicated it.

The 2002 constitutional amendments, which had established direct presidential elections, led to further dispersal of power in other fields, too. In the judiciary, for instance, a new Constitutional Court was created, breaking the judicial monopoly held by the Supreme Court thus far (Butt 2015). This meant that there were now multiple judicial checks on executive power, and the Constitutional Court in particular made extensive use of its powers. The constitutional amendments also gave new powers to parliament that, by implication, strengthened the multi-party landscape represented in it. From a rubber stamp under Suharto’s regime, parliament therefore developed into one of the most powerful actors of the post-1998 regime. Moreover, during the process of constitutional amendments, the police and the military were institutionally separated – the former had been part of the latter since the 1960s. While this separation was not enshrined in the constitution but based on a separate decree, the emancipation of the police was another cornerstone in the dismantling of the prior monopolar political system and the emergence of a post-Suharto landscape in which rival actors competed for power. In sum, the fast post-1998 reforms redistributed power among elite groups in a way that would make it hard for autocratic attempts at restoration to succeed. As the next section demonstrates, Indonesia’s key actors have been highly protective of this new competitiveness, despite their own interest in undermining major aspects of democracy.

Elite competition

In some analyses of post-Suharto Indonesia, the elite alliance ruling the country has been described as a cartel (Slater 2018). This model has been used to describe a constellation in which Indonesia’s parties bond together to jointly exploit Indonesia’s patronage resources and protect themselves from potentially damaging accountability mechanisms. There is much truth in this portrayal. Indeed, Indonesia’s key political forces – not only parties – have united in cases that affect their collective interests as a ruling political class. This means acting together on issues such as the weakening of the KPK, for example, as the agency had arrested many members of the elite across all sectors. The inclusiveness of presidential coalitions also seems to support the cartel theory, with the vast majority of parties and other major socio-political forces integrated into government in order to provide regime stability and perpetuate the status quo. In fact, much of Indonesia’s democratic decline, as discussed in the first section of this article, can be attributed to collective elite moves against democratic control institutions.

Yet the cartel model overlooks important elements of the inner workings of Indonesian elite alliances. Most importantly, it downplays the intensity of competition within such coalitions. Outside of collective interests, the fight for political advantage among the various actors is fierce, forming the very principle upon which alliances are built and operated. This competition is typically moderated by the president – another actor not sufficiently appreciated in the cartel model. Presidents run coalitions by constantly balancing the competing interests of their members – but also by playing them off against each other when necessary to sustain stability. But above all, members of coalitions jealously watch over each other so that no single actor among them gains an advantage that could allow the latter to emerge as a new political hegemon. This includes presidents, too – putting significant constraints on their ability to pursue autocratic ambitions. Thus, the cartel approach captures crucial dynamics of Indonesia’s democratic decline, but it is much less effective in explaining why elites have stopped short of fully abandoning democracy, and why they remain committed to competition as the organising logic of their interaction. To understand this phenomenon, we need to explore the alternative.

For many Indonesian elites, their continued support of democratic inter-elite competition is due to the unpalatability of the opposite: that is, living under restored autocratic rule (Aspinall et al. 2020). Under Suharto’s more than three decades in power, access to resources led through him, and him alone. This not only meant that the top post of politics was occupied and hence unavailable to any other individual or group for a generation, but also that privileges had to be purchased through unsavoury flattery. The president’s personal grudges could make or break an actor’s standing, potentially causing long-term damage to its institutional interest. In other words, while contemporary elites benefit from a diminished or compromised democracy – one that is controlled by the parameters of elite influence – they have little interest in pushing this weakened democracy into autocratic territory. Under autocracy, the prospect of specific elite actors gaining the presidency for one of their own in the short to medium term would be undermined; their real political weight might not be taken into account in the distribution of resources if the president chose to block certain players; and, as the Sukarno and Suharto autocracies demonstrated, the risk of eventual regime breakdown could undermine the position of authoritarian allies in the subsequent polity.

Under elite-controlled democracy, by contrast, the electoral and societal weight of an actor normally determines its representation in cabinet and its role in the distribution of resources. As this weight can change over time, each actor has a realistic expectation of having one of theirs elected to the presidency one day. Until that moment comes, participating in an elite alliance based on competitive allocation of positions is a satisfying outcome for most actors. Importantly, this outcome is the result of elections and subsequent negotiations, rather than of an autocrat’s whim. Another advantage flowing from this also figures prominently in elite calculations for sustaining a low-level democracy: that is, the benefit of democratic legitimacy as opposed to the inherent instability of regime authority based purely on repression. The upholding of an image of functioning democracy, as expressed in continuously high approval ratings for the way democracy is practised (about 70% in the 2010s and over 60% in the 2020s), is key to the elite’s interest in avoiding social upheaval and regime rejection.

The ways in which continued elite competition and ambition rebel against potential autocratisation campaigns were best manifested in an episode we already touched upon – that is, the failure of Widodo’s term limit evasion attempt (Mietzner and Honna 2023). Operating through his aides, the president had made it clear that he felt he needed more time in office to make up for the years lost through the pandemic. Either a third term – which required a constitutional amendment – or a delay of the 2024 elections were contemplated (author interview with Luhut Pandjaitan, coordinating minister, Denpasar, 16 September 2023). But, much to Widodo’s disappointment, the proposal found little elite support. Very few actors had an interest in Widodo staying on beyond his second term: his own party (PDI-P) was strongest in its rejection, believing that Widodo had not sufficiently represented its interests, and that a new president – possibly Ganjar – might be better for them (author interview with Hasto Kristyianto, PDI-P secretary-general, Jakarta, 22 June 2022). Prabowo, for his part, wanted to have a final shot at the presidency himself, making approval of a Widodo extension detrimental to his career planning. Even parties strongly pressured by presidential assistants into supporting the extension plan only paid lip-service to it. Nahdlatul Ulama (NU), the country’s largest Muslim organisation, did not come out in support either. While NU’s spiritual head was Widodo’s vice-president in his second term, it did not feel particularly privileged under his rule – in fact, at one point, Widodo had taken the organisation’s stronghold in the bureaucracy, namely the ministry of religion, away from it.

Accordingly, as Widodo had to realise, Indonesia’s elite actors were far too ambitious, and equipped by the post-1998 power dispersal with too many resources, to settle for another autocracy. The prospect of the re-monopolisation of power by a sitting president was unappealing to them. The presidential term limit, introduced in 1999 as one of the first post-authoritarian constitutional amendments, has been an insurance policy for elites to guard against autocratisation and to preserve their own chances at competing for the presidency in a reasonable timeframe. They were not willing to surrender this safeguard to Widodo, despite the fact that his presidency had brought benefits to them. For them, presidents leaving their post after two terms was the non-negotiable foundation upon which the Indonesian elite democracy rested. The agreement of some elites to the vice-presidential nomination of Widodo’s son Gibran was, in their long-term perspective on the matter, still reconcilable with that foundation. They believed, probably correctly, that power would rest with the new president, not Gibran – who many saw as only a vote-getter for Prabowo, the presidential candidate on the ticket.

The power dispersal instituted after Suharto’s fall, then, has sustained a level of inter-elite competition that obstructs incumbents’ possible attempts to navigate democracy into the territory of personalist autocracy. Any attack by an incumbent on existing elite privileges, including their own presidential ambitions, is certain to create a backlash – or just silent rejection, as in the case of Widodo’s attempt to outstay his welcome. As Slater (2023, 108) remarked, ‘Indonesia is living proof that a developing country’s political elites can embrace democracy for entirely pragmatic reasons, even when their democratic principles are paper thin’. Indonesian democracy survives, in other words, because its current minimalist format serves the interests of elites – and because the autocratic alternative wouldn’t, at least for now.

Conclusion

Our discussion above allows us to now step back and reflect on the drivers, pace, and limitations of democratic decline in Indonesia in a more conceptual manner. In the Indonesian case, elites with a range of vested interests have been the main actors behind both the country’s democratic decline and its continued survival as an electoral democracy. Elites have acted collectively, and in some cases individually, to undermine democracy’s oversight capacity and to keep the overall power of controlling the polity in the hands of the ruling class. This explains the instances of democratic decline beginning in the second half of the 2000s and continuing with more intensity throughout the 2010s and 2020s. The elite’s collective judgement at the outset of this process was that the reforms of early post-1998 democratisation had gone too far, and that elite controls needed to be restored. Increasing socio-religious conservatism in segments of the population provided a useful justification for the elite’s limiting of democracy’s liberal connotations. At the same time, however, elites also established limits to their own anti-democratic initiatives. There was no interest in a full re-establishment of autocratic rule, at least not among the majority of the elite. Whenever such reactionary initiatives were launched, as in Prabowo’s 2014 and 2019 campaigns or in Widodo’s attempt to extend his stay in office, there was significant opposition within the elite that thwarted them.

Hence, contrary to much of the autocratisation literature that has posited elites as the actors behind attacks on democracy but understudied their contribution to upholding it (Boese et al. 2021; Lührmann and Lindberg 2019; Tomini, Gibril, and Bochev 2023), this article has demonstrated that Indonesian elites are responsible for both the ‘minimalisation’ of democracy and for its continued endurance. Viewed through this lens, we can also better understand the pace of democratic decline in Indonesia and elsewhere. Slow and gradual, this pace differed considerably from the more rapid decline in other countries that often culminated in democratic reversal (such as in Thailand or Myanmar). The advantage of this slow-paced decline for the elite was that it made democratic setbacks almost unnoticeable for the societal mainstream, preventing the kind of communal conflict seen in its Southeast Asian counterparts. Democratic decline in Indonesia, then, was elite-controlled, regime-preserving, and socially tolerable.

In this dynamic, civil society and public opinion were secondary actors. In the early phase of democratic decline, they had still been the most important barriers to further erosion. But as NGOs, the media, and other non-state players became increasingly partisan, their effectiveness as democratic guardians diminished. As a result, elites themselves began to decide which parts of democracy were too important to them to be dismantled, and which ones could be attacked. This is not to say that societal pressure became irrelevant. But in most cases, it played a supporting rather than decisive role in protecting democratic rights. For instance, when pro-Prabowo parties tried to abolish direct local elections in 2014, public opinion was strongly opposed – but so was the incoming president, Widodo, his allies, and – importantly – the local chiefs themselves. With no elite unity on the issue, the campaign failed. Similarly, the public rejected Widodo’s quest for a third term – but, as we have seen, so did the majority of elites. Conversely, on matters in which there was collective elite unity but the public was opposed, elites generally prevailed – as in the case of the weakening of the KPK in 2019, the passing of a so-called Omnibus Bill in 2020 (on economic deregulation), and the legislation of a new criminal code in 2022. In short, elites turn against autocratisation attempts launched from within their midst, but are happy to cooperate on initiatives that strengthen the privileges of the ruling class while protecting the competitiveness of the overall system.

Finally, it is important to acknowledge that some readers might find the argument developed here circular or even tautological – democracy in Indonesia survived because elite competition remained high. But this notion is, in fact, far from being tautological. Democracy and elite competition are very different concepts; each can exist apart from the other. There are, for instance, autocracies in which elites compete for the attention of the dictator. The point this article made is that in the Indonesian case, most elites found the latter scenario unpalatable, and they thus defended a form of democracy that worked for them. They were helped in this regard by historical decisions made in the early period of the post-Suharto order that encouraged power dispersal. I have shown, then, that in some cases democracy is most likely to survive – albeit with reduced quality – if self-interested elites decide that its preservation is preferrable to autocracy. This means that clear boundaries are set for the progression of democratic decline, but it also means that a re-strengthening of democratic substance is equally impeded. While this constellation has frozen Indonesian democracy in a state of low-quality stasis, it has prevented it from joining the ranks of failed democracies – an outcome that might be a frustration to liberal activists, but should not be taken for granted either, given the trends elsewhere.

Ethics statement

Ethical approval for interviews in this research was provided by the Australian National University’s Human Research Ethics Committee (Approval No. 2022/083).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Marcus Mietzner

Marcus Mietzner is Associate Professor in the Department of Political and Social Change at the Coral Bell School of Asia Pacific Affairs, College of Asia and the Pacific, Australian National University, Canberra. He has published widely on Indonesian politics, and more specifically on the role of the military, political parties and the presidency. His most recent book is The Coalitions Presidents Make: Presidential Power and its Limits in Democratic Indonesia (Cornell University Press, 2023).

Notes

1 This article uses the term ‘democratic decline’ to denote a significant decrease in democratic quality. While multiple concepts have attempted to describe the same phenomenon – such as regression, backsliding, decay, or erosion – authors have pointed to weaknesses in all of these. For example, Lührmann and Lindberg (2019, 1099) have highlighted the problems inherent in the backsliding concept. ‘Decline’, on the other hand, is a simple yet effective way of capturing decreasing democratic quality.

2 Three candidate pairs ran in the 2024 elections.

3 Widodo, having failed to extend his term in office, then got his son Gibran nominated as running mate to Prabowo Subianto. But this was a much less attractive option for him than having his own term extended.

4 When Widodo advanced the vice-presidential nomination of his son Gibran, his brother-in-law, who was the chief justice of the Constitutional Court, engineered a verdict of the bench that allowed Gibran to run despite him not meeting existing age regulations. Other judges of the Court subsequently initiated ethics proceedings against the chief justice, and he was removed from office as a result. Several Widodo allies also broke with the president after the verdict.