2 THE PARTIES

Party Powers

One of Indonesian parties’ most consequential powers over the president is the right to nominate presidential candidates. Parties equipped themselves with this power in the 2002 constitutional amendments to compensate for losing the authority to elect the president in the MPR (Crouch 2010). For the 2004 elections, a transitional arrangement was put in place by which parties that held 3 percent of the DPR seats or had gained 5 percent of the votes in the preceding legislative elections could make a presidential nomination. This increased to 20 percent of the seats or 25 percent of the votes in 2009, and stayed at that level for the 2014 and 2019 elections. For the latter, there was much discussion on revising the threshold because, for the first time, the legislative and presidential elections were held on the same day in that year (Gobel 2019). This meant it was difficult to uphold the logic of drawing a threshold for presidential nominations from the parties’ performance in the preceding legislative elections (which in 2004, 2009, and 2014 had taken place a few months before the presidential ballot). Nevertheless, the high threshold was left in place for 2019—it was now based on the result of the legislative elections held five years earlier. The high threshold for presidential hopefuls has the advantage for parties that every candidate typically needs to lobby several parties to get nominated, which reduces the possibility of nominees running on—and implementing—an anti-party agenda. For incumbent presidents seeking re-election, high thresholds limit the number of potential challengers. It was no surprise, then, that most parties and Widodo agreed to keep the threshold high.

In practice, the high nomination threshold allows parties outside the president’s own to join a pre-electoral alliance and participate in the distribution of spoils after victory (Slater 2018). As noted, parties of the pre-electoral coalition can expect more of these spoils than those that seek access to the presidential alliance at or after inauguration day. This power of parties to force their way into a nominating coalition is particularly relevant given the traditionally small size of the president’s nominating party and his or her sometimes-weak role in it. In 2004, Yudhoyono’s Democratic Party (PD) won a mere 7.5 percent of the votes, and could only make a nomination because of the transitional threshold regulations. For his re-election in 2009, Yudhoyono could have relied on a sole nomination by PD (it had won 26 percent of the seats), but many parties rushed to co-nominate the president as polls predicted his win. Widodo’s PDI-P was just below the nomination threshold in 2014 and 2019, requiring additional parties to file his candidacy. In Widodo’s case, nomination by non-PDI-P parties served his interests, as it compensated for his weakness in his party. In 2014, it was public knowledge that PDI-P chair Megawati had nominated Widodo—the leader in the polls—only because her own chances to win were slim and because no other member of the Sukarno family stood ready to replace her. To be sure, Widodo’s enthusiasm for PDI-P was equally limited.¹ In this situation, the existence of other parties in his nominating coalition was advantageous to the parties as well as the nominee and future president.

Beyond their formal nomination authority, parties typically offer their grassroots campaign networks to presidential candidates in exchange for later, post-electoral rewards. As Aspinall and Berenschot (2019) showed, local executive and legislative elections in post-Suharto Indonesia are won mainly by candidates employing an effective network of brokers and grassroots campaigners. These networks organize events, make material promises to specific communities, and frequently distribute cash. In presidential elections, however, this system is unworkable for a national campaign. With a population of 275 million, an archipelago stretching over 3,200 miles from west to east, and no effective campaign financing mechanisms, Indonesia makes it impossible for a single presidential campaign to set up an apparatus that could replicate what has proven successful at the local level. Hence, presidential candidates are forced to shift much of the campaigning to the about 250,000 legislative candidates that parties nominate in local legislative races. This army of nominees, it is hoped, will add promotion of the presidential candidate to the campaign for their seats. This hope was particularly high in 2019, as presidential and legislative campaigns took place concurrently. As could be observed on the ground, candidates did not necessarily implement the orders from their party headquarters to put the presidential candidate’s image on their campaign posters, especially in areas where the latter was unpopular (Mietzner 2019). Despite these frustrations, most presidential hopefuls and incumbents still believe that the support by party and local candidate machines remains crucial to their campaigns.

After elections, parties can leverage the legislative powers of their members in the DPR to gain concessions from the president (Sherlock 2012). These legislators can support or oppose bills and budgets, approve or reject the president’s nominations for key positions, and either attack or praise government representatives when appearing at parliament. Whether implicitly or explicitly, parties promise the support of their legislative caucuses for the president if given cabinet representation and other benefits in return, while presidents expect that parties’ membership in the government will convince them that it is inappropriate to vote against its policies in the legislature. In practice, the picture is less clear-cut. Parties do not exercise full control over their legislators, and at times, party leaders are divided among themselves over how to direct their caucuses on specific issues. Yudhoyono, for instance, tried unsuccessfully in 2009 and 2011 to codify the voting behavior of his coalition parties in the DPR through a written contract. However, this failure to enforce the loyalty of legislators by contractually binding their party chairpersons did not lead him to dismiss the importance of the parties altogether. Rather, he—and Widodo after him—developed a dual strategy by which party leaders and legislators had to be courted in parallel; this approach recognized that party chairs played an important role in filling senior leadership positions in the DPR but also acknowledged that the presidential apparatus had to work with individual legislators to secure specific deals. In his presidential memoirs (or, more precisely, a voluminous collection of thoughts on his presidency), Yudhoyono concluded that despite the unreliability of parties, including them “in a political coalition still makes our politics way more stable” (SBY 2014, 39).

Another asset of parties in their negotiations with presidents is their ability to bestow legitimacy during and after elections. This legitimacy capacity is particularly useful when presidents come under attack on issues of identity or faith (Aziz 2020). In such circumstances, Islamic parties can protect the president with their religious credibility and connections in the Muslim community. Except for Wahid, a Muslim cleric, all Indonesian presidents since 1945 have been mainstream nationalist figures—that is, Muslims who may practice some elements of their faith but otherwise lead what can be described as secular lifestyles. Thus, they have been vulnerable to questions about their devoutness. In the 2014 elections, Widodo had to confront false rumors spread by his opponents that he was a secret Christian of ethnic Chinese origin. As part of his pre-electoral coalition, the traditionalist Muslim party PKB took it upon itself to run a campaign in Islamic boarding schools to counter this smear. In one such school in Bogor, the party distributed pamphlets that showed pictures of Widodo performing the hajj (notes by the author, Bogor, June 7, 2014). Two years later, when the Widodo government came under siege from a mass mobilization of Islamists—which protested against alleged blasphemy on the part of the Christian and ethnic Chinese governor of Jakarta, Basuki Tjahaja Purnama, who was a Widodo ally—the Muslim parties in the president’s coalition were again called upon to act as a buffer against notions that Widodo was lacking in faith. Hence, some political parties play the same legitimacy-producing role as large Muslim organizations that, as separate entities, are also part of presidential coalitions.

Indonesian parties, then, hold important constitutional, institutional, and cultural powers that presidents have to manage. These powers can either act as stabilizers of presidential rule or be used against it. From our discussion above, it is clear that presidential management of parties goes well beyond the need to secure a legislative majority. For that, a simple 50-percent-plus-one formula would be sufficient. Instead, presidents have sought to include a much wider range of parties that not only deliver votes in the legislature but also offer other services. The explanation for this pattern partly lies in the broad party powers. Presidents are interested in benefitting from a variety of party functions, from electoral support to crisis protection and leverage in order to balance difficult relationships with their own parties. And above it all towers the permanent fear of presidents that parties not integrated into the coalition could create “chaos” in the legislature and society at large. As Yudhoyono expressed it, every party brought into the coalition is one party less to worry about in terms of its potential to trigger discontent and instability. Yudhoyono was well aware of the efficiency losses inherent in oversized coalitions: “convincing all these parties in the coalition is not easy, and in probably 30 percent of cases, I had to accept the reality [that I could not prevail with my policy]” (interview, Jakarta, December 2, 2014). But, he asserted, building such broadly inclusive party coalitions meant that 70 percent of his policies could be adopted, and—as remarked in his memoir—stability could be secured. As later chapters will demonstrate, many critics would view Yudhoyono’s assessment of his success rate as overly optimistic.

Presidential Assets

While we have noted the many assets that parties can use to extract concessions from the president, it is important to begin the discussion of presidential strengths vis-à-vis the parties by recognizing one of their key weaknesses. Remarkably, since the presidency of Megawati Sukarnoputri, long-established parties have been unable to produce competitive candidates for presidential elections from their leadership ranks. In every presidential ballot after 2004, competitive candidates have been political outsiders who quickly established personalist parties with the sole goal of running for president, or figures whose role in parties was marginal but who commanded high poll ratings. The first category of candidates includes Yudhoyono and Prabowo Subianto, both former military generals who gained name recognition through their military service under Suharto (indeed, Prabowo was married to Suharto’s daughter). Yudhoyono established PD before the 2004 elections (Honna 2012), while Prabowo—who unsuccessfully but competitively ran in the 2014 and 2019 elections—set up the Great Indonesia Movement Party (Gerindra) in 2008. Neither party has strong institutional or ideological roots, which was highlighted by the fact that Yudhoyono handed over the leadership of PD to his son Agus Harimurti in 2019. In the second category is Widodo, who rose to political prominence by having been mayor of the Central Java town of Solo from 2005 to 2012 and governor of Jakarta from 2012 to 2014. Mostly inactive in PDI-P, he was the only alternative for the party if it wanted to stand a chance in the 2014 elections. Post-2004 presidents, therefore, have benefitted from the incapacity of established parties to challenge them at the ballot box.

In addition to the inability of traditional parties to groom presidential hopefuls from their leaders, most of the personalist parties that sprung up between the late 2000s and late 2010s have not made their patrons competitive either. The National Democrats Party (Nasdem), founded by media tycoon Surya Paloh in 2011; the People’s Conscience Party (Hanura), established by former armed forces chief Wiranto in 2006; and the Indonesian Unity Party (Perindo), created by billionaire Hary Tanoesoedibjo in 2014, are all examples of parties headed by rich but unpopular figures whose ambitions to become president were strong but unfounded. After coming to terms with that reality, they discovered that their best opportunity to stay engaged in politics below the level of the presidency was to align with a popular presidential candidate who, in case of victory, could offer them ministries. Thus, the Indonesian party landscape after 2004 has been populated by two types of parties, as table 2.1 shows. The first type consists of more conventional, historically grown ones appealing to a specific socio-political community, and the second comprises personalist vehicles set up for presidential candidacies. Importantly, most parties in both categories have been reduced to endorsing the presidential campaigns of outsiders to gain access to post-election patronage. Having gained office with the very popularity that most political party leaders lacked, post-2004 presidents have paraded their political capital with great self-confidence. In his memoirs, Yudhoyono devoted an entire chapter to the argument that popularity cannot be bought—a not-so-subtle swipe at those party leaders who had tried exactly that (SBY 2014: 363–366).

This post-2004 separation between party politics featuring unpopular operators and presidential races dominated by outsiders is not unusual in presidential systems (Carreras 2012). However, it came as a disappointment to many Indonesian party leaders. They had hoped that the party-based nomination threshold would allow them to maintain their control over who could run in presidential elections and who would win (interview with former Golkar chairman Akbar Tandjung, Jakarta, February 11, 2008). Instead, the focus of presidential competition shifted from the party arena to figures whose main asset was popularity in the polls and who only needed party support to get nominated and gain a campaign network. One of the main reasons for this split between the party and presidential arenas has been the trend within parties to appoint chairpersons whose sole task is to secure the party’s funding (Mietzner 2015). This pattern has been obvious for personalist parties—in which the wealth of the “owner” is often the only reason the party exists—but it has also affected the socially rooted parties. In the absence of a functioning party funding system (with private donations low and state financing minuscule), party chairs are chosen primarily based on their ability to supply operational funds. As one senior party functionary conceded, “we would love to put people in the position of party chair who are popular and could win the presidential elections for us. But that is not how it works. The first question on our mind, when evaluating candidates for party chair, is: can this person pay our party’s bills?” (confidential interview, Jakarta, December 7, 2019). Hence, structural changes in party development have put faceless party financiers in charge, weakening the parties’ standing in presidential ballots.

Closely related to electoral power as a key element of presidential political capital is the anticipated ability of a popular candidate to increase the legislative vote of the parties that nominated him or her. This so-called coattail effect is both expected by parties and promised by presidential incumbents and nominees. In Indonesia, Golkar nominated Widodo for re-election in 2016, three years before the election and long before his own party did. Golkar’s secretary-general, Lodewijk Freidrich Paulus, was open about the reason for this move: “we hope that nominating [Widodo] so early will help our legislative candidates. That’s why we did it” (Mardiansyah 2019). As in the case of presidents’ hopes that their nominating parties’ networks will mobilize their resources for the presidential campaign, the coattail effect for the parties is often less pronounced than desired. In 2019, only PDI-P and Gerindra seemed to benefit from the candidacies of Widodo and Prabowo, respectively, while other nominating parties stayed largely flat (Golkar lost votes compared to the 2014 elections, when it did not nominate an incumbent president). Nevertheless, the theory of a coattail effect continues to attract interest and support within party elites, allowing presidents or candidates leading in the polls to use it as leverage. This was proven again in October 2022, when Nasdem nominated the popular Jakarta governor Anies Baswedan as its candidate for the 2024 presidential race, openly explaining this early choice within an expected coattail effect.

Another asset that Indonesian presidents can deploy in negotiations with parties is their authority to determine the outcome of intra-party conflicts. In Indonesia’s legal system, parties have to register the composition of their central boards with the president’s Ministry of Justice and Human Rights (Rahman 2016). Only if the latter certifies this registration is the leadership board deemed legitimate. This certification, in turn, allows the party to submit legislative candidates to the General Elections Commission (KPU), replace its MPs in the DPR, or seek access to state subsidies. Therefore, the president—through his or her minister—can decide which claimant in a party leadership dispute is the officially legitimated representative allowed to engage with the state. While Yudhoyono was reluctant to use this power and largely left intra-party conflicts to the courts to decide, Widodo was not so hesitant. In 2015 and 2016, his government’s meddling in the internal affairs of Golkar and PPP led to the removal of their respective party leadership boards that had nominated Widodo’s rival Prabowo Subianto in the 2014 elections (Mietzner 2016). In their stead, two rival leadership boards loyal to the president were installed and received government confirmation. This shift in Golkar and PPP allowed Widodo to broaden his party coalition in the DPR, taking it from minority to majority status. Consequently, parties have to consider the risk of executive interference in their affairs should they oppose the president, giving the latter a useful power instrument when dealing with parties inside and outside their coalitions.

Indonesian presidents also possess important patronage distribution capacities from which parties seek to benefit. Chaisty, Cheeseman, and Power describe three of these patronage powers contained in the presidential toolbox: cabinet appointment authority, budgetary powers, and the exchange of favors. To begin with, cabinet posts are essential for parties to channel money into their underfunded organizations. Ministers representing a party in a coalition are not only expected to defend the party’s policies but also to make their ministry’s resources available to the party. Under Yudhoyono, for instance, PKS officials used their hold over the agriculture ministry to demand kickbacks in return for granting beef export quotas; the party’s president, one recipient of the funds, was arrested in 2013 (Kramer 2014). In the Widodo period, the social affairs minister, who was also the PDI-P deputy treasurer, diverted money from the COVID-19 social assistance budget into the party’s coffers.² The president’s authority to draft the state budget is of interest to parties, too. Presidents can negotiate with parties over individual budget items that might benefit a party’s leaders or its main voting base. One example was a project to produce new identity cards for all citizens—most parties were believed to have pocketed funds siphoned off from the project budget, with the then Golkar chairman going to prison in 2018 for his role in the affair. Finally, presidents can hand out informal favors to parties; for instance, presidents often contribute, directly or through allies, to the funds necessary for holding a party congress, especially if they have an interest in potential decisions emerging from that congress.

The final two elements of the conventional presidential toolbox—namely, the president’s legislative and partisan powers—are also relevant for Indonesian presidents’ relations with parties, but in slightly different ways. Recall that in Indonesia, of the president and the legislature must jointly promulgate laws. This equips presidents with powers superior to standard package or partial vetoes but also forces them to seek consensus with party-affiliated legislators. However, most of these negotiations occur in the legislature’s caucuses or committees rather than at the party leadership level. Only key issues are brought into a rare forum between the president and coalition parties. Accordingly, discussing the process through which presidents negotiate with party-based MPs over legislation and vice versa must concentrate on parliament as its main arena and thus will be dealt with in the next chapter. As for partisan powers—which refer to the control of presidents over their party—the experiences of Yudhoyono and Widodo were decisively different. Yudhoyono was the de facto founder and sole patron of his party. While he faced occasional internal strive over personnel matters (Honna 2012), the party elite accepted his supremacy in the policy domain. By contrast, Widodo had no partisan powers, given his marginal role in PDI-P. Significantly, however, this lack of partisan power drove him to build stronger ties to other parties to leverage the latter against of the influence of PDI-P. Thus, while Yudhoyono’s “ownership” of his party was an advantage for him, Widodo turned his weakness in this field into a political strength by mastering the art of coalitional balancing.

We have now reviewed both the powers Indonesian political parties hold vis-à-vis the president and the assets that the latter can bring to bear in order to lobby, reward, and punish the former. The picture that emerged from this discussion is one of great leverage on both sides. On the one hand, parties recognize the powers of the presidency, and so they seek to capture or align with it to gain resources and policy influence. On the other hand, presidents need parties to receive support for nominations, campaigns, and post-electoral governance. This constellation provides an ideal breeding ground for coalitional presidentialism—an arrangement from which both sides benefit. The next section offers a portrait of how this arrangement typically takes shape in the Indonesian context and explains its endurance as the foundation of president-party relations in post-2004 Indonesia.

Balancing Party Coalitions

As noted, Indonesian party coalitions take shape before an election and are subsequently modified as post-election alliances. In much of the coalitional presidentialism literature, this distinction between pre-electoral and post-election coalitions is downplayed (Chaisty, Cheeseman, and Power 2017, 216). But the Indonesian case shows that the difference is significant. Pre-electoral coalition parties gain more of the share of the cabinet seats, and they are typically ideologically closer to the president. Indeed, it is possible to speak of a two-phase model of presidential coalition-building: in the first phase, parties often side with the candidate closer to their politico-ideological orientation, while in the second, some parties from the losing side are integrated, receiving a lower ministry share as a result. Thus, despite a growing body of literature on the irrelevance of ideology in Indonesian party politics (Ambardi 2009), important ideological traces remain in pre-electoral coalition-building. The main cleavage is between those who promote a stronger role for Islam in state organization and those who support the pluralist mandate of the constitution. In the 2004 and 2009 elections, Yudhoyono—whose party adopted the descriptor “nationalist-religious”—portrayed himself as more religious than his main rival Megawati, and the parties supporting them in the elections divided roughly along those lines. In the 2014 and 2019 ballots, Widodo took the mantle of the pluralists, while Prabowo was supported by conservative Muslim parties and groups (Power 2019). There were some shifts and exceptions over time, but in broad terms, this cleavage persisted.

Even after elections, these ideological tensions continue to influence the broader coalitions built to secure the president’s hold on government. PDI-P remained outside of Yudhoyono’s government for the entirety of his rule, and so did Gerindra, Hanura, and Nasdem. Conversely, PKS and PD did not join any of the Widodo coalitions. Personal animosities played a role in these decisions as well. For example, Megawati’s legendary grudge toward Yudhoyono (she felt that he had not properly informed her of his intention to run against her in 2004, although he was in her cabinet) led her to prevent her party from entering Yudhoyono’s coalition and PD from joining Widodo’s. It is true, of course, that ideological lines blurred considerably in the process of putting together post-election party coalitions. But as table 2.2 demonstrates, the support of a party toward a president in the election continued to matter greatly in determining if or how many cabinet seats were given to the said party. The table’s analysis of the composition of the second Widodo cabinet shows that the only non-nominating party to be included in the ministry in 2019 was Gerindra, and that the latter was the only cabinet party to receive a lower cabinet seat share than its seat share in the DPR would indicate.³ All other parties in cabinet obtained a ministry seat share significantly larger than their DPR seat share. Other non-nominating parties received no representation in cabinet. This suggests that the cost of being outside of a president’s nominating coalition remains high, despite the chance of being integrated into the post-electoral alliance.

While ideological default lines are visible in electoral coalitions, it is important to emphasize that the overall mechanics of coalitional presidentialism in Indonesia remain non-ideological. Ideological polarization tends to occur during elections (hence its influence on a president’s electoral coalition), but periods between elections are mostly marked by patronage deals that integrate former electoral opponents into the alliance (albeit with reduced privileges) and dispense some limited benefits to the remaining opposition. The inclusion of Gerindra into the second Widodo ministry (with the president’s rival in the 2014 and 2019 presidential elections Prabowo Subianto becoming minister of defense) is an example of integrating former adversaries. Oppositional figures staying outside of cabinet, for their part, are typically accommodated through arrangements made in the legislature (more about this in the next chapter). Ideology, then, should neither be fully dismissed nor overstated in its significance for presidential politics. It does play a role in the early formation of alliances that are subsequently formalized in a nominating coalition in presidential elections, but it is not the organizing principle of day-to-day coalitional presidentialism. Furthermore, it is common for some actors to change their ideological positions for practical reasons—Prabowo, for instance, eventually abandoned the Islamists who had supported him in 2014 and 2019, hoping to build a more inclusive platform for his anticipated bid in 2024. Therefore, the technical aspects of coalition management remain our analytical priority.

When managing their coalitions, presidents can use reshuffles (or the threat of reshuffles) to punish parties deemed disloyal. Indeed, threats are more common than actual wide-ranging changes to cabinet. During Yudhoyono’s and Widodo’s rule, only once was a party’s cabinet seat share reduced, and only once was a party expelled from cabinet. In 2011, Yudhoyono took away one of the four cabinet seats PKS had held, penalizing it for voting against the government in the DPR. Widodo, for his part, removed PAN from his cabinet in 2018 when the party nominated Prabowo Subianto for president in the 2019 elections.⁴ Despite the rarity of such punishing reshuffles, the mere mention of their possibility has a disciplining effect. According to Luhut Pandjaitan, one of the president’s closest aides, “cabinet reshuffles are the only thing that keeps parties in line. Even if the president just mentions the word ‘reshuffle,’ parties get nervous and hold emergency meetings” (interview, Jakarta, June 15, 2016). Although the prospect of reshuffles might prevent open dissent and coalition collapse, it often fails to produce the kind of full-time loyalty presidents seek. In the case of PKS, the party continued to take steps Yudhoyono and other coalition parties viewed as hostile. Despite being threatened with expulsion from the cabinet again in 2012 and 2013, PKS retained its remaining ministries until the end of the president’s term in 2014. Andi Mallarangeng, a former Yudhoyono minister, recalled that the president told him that he did not remove PKS from the cabinet because that would have increased the power of other parties, especially Golkar (interview, Jakarta, June 20, 2022). Hence, presidents judge the cost of removing parties from cabinet to be higher than that of tolerating minor acts of disloyalty.

The management of Indonesian party coalitions is poorly institutionalized but for different reasons in each presidency. Yudhoyono tried to create a coalition council to coordinate policies and enforce binding discipline rules but largely failed in that effort. Irked by what he saw as a lack of discipline among coalition parties in his first term, he created a Joint Secretariat (Setgab) for his alliance at the beginning of his second term in 2009. But this institution proved to be ineffective. In April 2012, a meeting of the Setgab—under the leadership of Yudhoyono—decided that PKS had (again) violated the code of ethics all parties had agreed to in 2009. But no sanctions were imposed. Widodo, by contrast, opted not to establish a council. During his presidency, meetings of the party chairpersons occurred rarely and irregularly. In fact, Widodo deliberately tried to reduce such meetings to a minimum and instead preferred dealing with party leaders individually. As one of his assistants explained, “it is normally during joint meetings when the party chairpersons put forward collective demands that the president then finds difficult to refuse. In one-on-one negotiations with parties, the president can balance the positions of the various parties better, and he can exploit differences between them” (confidential interview, Jakarta, February 8, 2017). Although neither president faced serious threats to their presidencies from their party coalitions, Widodo’s strategy was overall more successful in preventing parties from voting openly against government policies. Despite occasional friction, he pushed through some controversial government programs with a striking level of coalition unity (Setijadi 2021).

Widodo’s divide-and-rule approach to his party coalition was the key to improving his own standing vis-à-vis his own party. At the beginning of Widodo’s first term, his relationship with PDI-P was poor, with some of its leaders calling for the president’s impeachment (Rastika 2015). At a PDI-P congress in Bali in April 2015, Megawati lectured Widodo in front of the party crowd, telling him that a president is supposed to implement the directives given by his party and that thus far, he had failed to do so (notes by the author, Denpasar, April 9, 2015). In what was seen as the ultimate punishment, the president was not allowed to speak at the congress—a highly unusual incident. One local branch leader felt that Widodo “does not want to serve the party that nominated him, and if he continues this way, we will try to get rid of him” (interview, Denpasar, April 10, 2015). As noted, Widodo subsequently broadened his coalition. Supported by more parties, he was able to change the power balance with PDI-P. The most important strategy in this regard was asking other parties to take a public stance against an idea endorsed by PDI-P but rejected by him. This enabled Widodo to inform his party that there was resistance to the PDI-P policy, and that the government therefore could not adopt it. As a senior Golkar official reported, “Widodo is smart. He often asks us to take a stand against this or that because he feels he can’t reject a request by Megawati and PDI-P directly” (confidential interview, Jakarta, December 9, 2019). The changed power balance was on display at the next PDI-P congress in Bali in August 2019. This time, Megawati was much softer in her approach to the president, Widodo was invited to speak, and there was no talk of impeachment (notes by the author, Denpasar, August 9, 2019).

But PDI-P’s new politeness toward Widodo could barely mask its deep-seated disappointment over the president’s success in emancipating himself from his party. In the leadership’s internal accountability report circulated at the 2019 congress, the party lamented that there still was “no correlation between the elected president and vice president and the party that nominated him” (PDI-P 2019, 4). Excerpts from Megawati’s 2015 speech were re-printed and underlined in the report, suggesting that the party had not given up on its self-proclaimed right to control the PDI-P-nominated president—but also that it now grudgingly accepted that it could not exercise this control. Much to PDI-P’s dismay, Widodo had managed his presidential coalition in a way that served his interests more effectively than those of PDI-P. This again highlighted that partisan powers are not necessary for a president to run a party coalition, and that their absence can produce shifts in the alliance that disadvantage the president’s party. PDI-P’s secretary-general Hasto Kristiyanto conceded this in 2022 when he defined firmer control over the next president as one of the party’s main goals for the 2024 elections (interview, Jakarta, June 28, 2022).

Broad party coalitions, then, put presidents at the center of managing a multi-actor contest for resources and policy influence. Maintaining an alliance with so many parties, based on a sense of mutual dependence, carries benefits and risks for the president. The more parties are included, the higher the opportunities for the president to use one party against another and to fend off requests that he or she deems inappropriate. However, the danger for presidents lies in parties forming a collective stance against the head of state. Such issues could relate to the interests of the political class as a group or parties as organizations. Faced with a united front, presidents are often forced to give in—turning the inclusiveness of the coalition from an asset into a liability. Thus, party-based coalitional presidentialism—especially if it comes in the form of oversized coalitions, as in the Indonesian case—is both durable and prone to stagnation. It is durable because it profoundly satisfies both the parties (which can express their ideological preferences in elections while retaining the chance of joining government if defeated) and the president (who can benefit from the parties’ resources and neutralize their potential threats, as well as play individual parties off against each other for even greater advantage). Naturally, such juggling of multiple interests comes at the cost of bold reformist policy initiatives. Filtered through various layers of presidential or party interests, initiatives for major change are at risk of foundering or being watered down. In the next section, we explore a case study that showcases both the reasons for the durability of broad party coalitions and the stagnating impact it can unfold.

The 2018 Vice Presidential Nomination

As we have seen, coalitions between Indonesian presidents and political parties have settled into an equilibrium that is carefully balanced through a set of unwritten rules (such as that pre-election partners are privileged) and a mutual understanding of the costs of a possible failure (which leads presidents to rarely punish parties through reshuffles and parties to refrain from seriously pursuing the president’s impeachment). This balance, which allows both sides to survive without excluding or destroying the other, forms the basis for the party dimension of coalitional presidentialism in Indonesia. To illustrate how the broad outlines of this equilibrium manifest themselves in practice, the discussion below analyzes the negotiations through which Widodo and his pre-electoral coalition arrived at the choice of the vice presidential running mate for the incumbent in 2018. This process led to frustrations on both sides but still sustained the conviction that all involved actors benefitted. As such, the episode is an appropriate metaphor for Indonesian party coalitions per se.

Selecting a vice presidential running mate is an important moment in any presidency (Kamarck 2020). Especially for incumbents up for re-election, picking a running mate without outside interference is a test of their power. While first-term presidents often have to accept deputies recommended by their sponsors, incumbents are expected to have accrued enough political capital to choose the person they think is best suited to assist in the campaign and when in office. In 2014, Widodo’s coalition of nominating parties had imposed a running mate on him. Jusuf Kalla, who had also been Yudhoyono’s vice president in his first term, was the preferred choice of Megawati and other party leaders. They had looked for someone who was both useful and non-threatening, as they did not want to select a candidate with prospects of gaining the presidency on his or her own. Kalla fit the bill: at the time of the 2014 election, he was seventy-two years old, had lost the chairmanship of Golkar in 2009, and was an experienced administrator. As such, he was a politician well beyond his prime and without an independent power base but still a safe choice as far as his governmental expertise was concerned. Because of the existing term limit, Kalla could not be nominated again for the 2019 elections. Accordingly, a new vice presidential nominee had to be recruited. After having expanded his coalition and his presidential power in his first term, Widodo appeared to be in a better position this time to make his own choice.

The golden standard of a smooth and self-determined vice presidential pick, and thus the model for Widodo to follow, was Yudhoyono’s selection of Boediono in 2009. During his first term, Yudhoyono had grown tired of Kalla’s autonomous political maneuvers. Instead of the powerful and wealthy Kalla (who was still chairman of Golkar at the time of the 2009 election), Yudhoyono wanted someone more low-profile: that is, a loyal assistant administrator rather than a self-confident politician. Boediono was a perfect match. A rather bland technocrat and academic, Boediono had been in several ministerial roles before the 2009 elections but had no connections to party politics. At sixty-six years of age and with non-existent popularity ratings in the polls, he also appeared unlikely to be a candidate in the 2014 elections. Boediono’s profile suited the leaders of Yudhoyono’s coalition parties, too. They accepted that in a large multi-party coalition, it was implausible for one of them to become vice president, given that this would unfairly privilege one party over the others. In this context, they were satisfied with endorsing a neutral, non-political actor who could not threaten their interests, and who made sure that in the next elections—for which no incumbent would be able to draw from government resources—all parties could push their own candidates.

But Widodo failed to replicate Yudhoyono’s success, and he inadvertently turned the selection of the 2019 running mate into a major power contest between the incumbent president and his nominating parties (Fealy 2018a). To be sure, Widodo was in a slightly different position from the one Yudhoyono had found himself a decade earlier. Before the 2009 ballot, Yudhoyono had such a strong lead in the polls that he could choose a running mate without electoral considerations. Boediono, a Javanese like Yudhoyono, was not going to attract voters from outside the president’s regional and ethnic core support base. Widodo, by contrast, faced a significant challenge from Prabowo Subianto, and so he had to make the potential to increase his overall electability a major criterion of his vice presidential choice. Beyond this significant difference, Widodo made a series of mistakes that Yudhoyono had avoided. Most importantly, while Yudhoyono had early on closed the door to the notion of selecting a party chair as running mate, Widodo not only tolerated the ambitions of party leaders but actively fueled them. Throughout the selection process, he allowed at least three party chairmen to enthusiastically propagate their mistaken belief that they had a chance of being selected. Golkar chairman Airlangga Hartarto, PKB leader Muhaimin Iskandar, and PPP chair Muhammad “Romy” Romahurmuziy all credibly claimed that the president had made encouraging remarks to them about their chance of becoming the vice presidential nominee.⁵ Some palace sources insisted that Widodo deliberately egged them on so that he could use their competition to pick a neutral outsider.

If this was Widodo’s plan, it failed badly. Even in the last week before the nomination in August 2018, all three men still believed in their chances. In the meantime, Widodo was working with his pollsters and staff on making his real pick. His main concern was a candidate’s contribution to his electability and his feeling of personal compatibility. Finally, a few days before the nomination deadline, Widodo made his choice: it was Mahfud MD, the former chief justice of the Constitutional Court, who commanded high approval ratings and also had links to NU, the country’s largest Muslim group. The NU connection was important as Widodo tried to gain the organization’s support to challenge expected accusations from the Prabowo campaign that the president was anti-Islamic. Being presented with the latest survey numbers that showed Mahfud as the most promising running mate, Widodo looked at one of his pollsters and said, “Okay, let’s do it” (confidential interview, online, August 7, 2018). But in yet another grave error, Widodo did not inform his partners in the party coalition—and neither did he discuss with NU whether Mahfud was a choice it could support. Instead, the news about Mahfud’s imminent nomination leaked to the press, which is how the president’s partners found out. As Mahfud began to give interviews in the press on why he accepted the nomination, a rebellion was brewing within the party coalition and NU against Widodo’s decision.

The key summit between Widodo and the chairs of his coalition parties to discuss the vice presidential candidacy was scheduled for August 9, one day before the nomination deadline. On the evening before this meeting, NU leaders—including the PKB chairman—gathered at NU headquarters and threatened the president with desertion should Widodo nominate Mahfud (Detik 2018). NU made it known that it did not view Mahfud as a NU cadre, and that it had proposed to Widodo its list of four NU-endorsed figures. This list included the NU-linked party chairpersons of PKB and PPP, the NU chairman Said Agil Siraj, and its supreme spiritual leader, Ma’ruf Amin. The latter was a conservative cleric who had played a significant role in the 2016 mobilization of Islamists against the Christian and ethnic Chinese governor of Jakarta; it was his certification that the governor had committed blasphemy that justified the protests (Fealy 2018a). Following the NU gathering that threatened Widodo, assistants to the president began to inform party leaders individually that the president had picked Mahfud. In one-on-one communications with the palace, party leaders did not openly voice their opposition to Mahfud. Simultaneously, however, the elites consulted among each other. In these meetings, party leaders shared their dismay over how Widodo was about to nominate Mahfud over their heads, and that the parties needed to warn the president that he could not make such a momentous decision unilaterally. After some back and forth, the party chairpersons decided to use the August 9 meeting to reject Mahfud and nominate Ma’ruf instead (confidential interviews with party officials, Jakarta, March 7, 8, and 10, 2019).

The president arrived at the August 9 meeting unprepared. Mahfud had been told by palace assistants to wait in a restaurant nearby and expect a call to join the gathering for the public announcement. It is unclear whether Widodo was told before entering the room that the party leaders would reject Mahfud or whether that decision was only conveyed to the president in the meeting. In whatever way he was told, Widodo decided on the spot not to challenge his coalition and endorse the choice. To protect the president’s public integrity, it was decided to present Ma’ruf’s selection as Widodo’s choice. Awkwardly, as the president and his coalition parties announced the nominee to the press, Ma’ruf was not even present, and Mahfud had been told to go home quietly. Ace Hasan Syadzily, a senior Golkar official, described the deal as a “good consensus. The parties had made their voice heard. The president listened. Ma’ruf is a respected figure, and we can live with that choice. We would have preferred our own chairman, but that’s life. And there was one thing we could all agree on: the selection of Ma’ruf was the best way to leave all parties in the same position for the 2024 elections” (interview, Jakarta, March 15, 2019). As with Boediono in 2009, parties ultimately endorsed a vice president because he was too old (Ma’ruf was seventy-six and thus fifteen years older than Mahfud during the election) and not popular enough to launch his own presidential bid. Unlike in 2009, however, the parties had to force the president into this consensus, while Yudhoyono had proposed Boediono to them in a much more orderly manner.

The conflict surrounding the vice presidential nomination—which Yudhoyono and Widodo approached differently but with the same outcome—reflects the political centrality of the equilibrium between presidents and their coalition parties. This equilibrium is only rarely disturbed in a serious fashion, but the 2018 vice presidential nomination was one such rare case. On the one hand, the incumbent president was popular and on course for re-election, and the parties had no competitive alternative to nominate on their own. On the other hand, the parties held the power of formal nomination, which they felt Widodo did not sufficiently respect. In the conflict that ensued, both sides renegotiated their respective roles and eventually settled back into the power balance that benefited them both. While Ma’ruf was not Widodo’s preferred pick, he could accept the selection because it kept his party coalition together and secured NU’s support in an election partly fought on each candidate’s Islamic credentials. For the parties, the president had given in to their demands, reassuring them that they are crucial to the sustainability of presidential coalitions. Both sides became aware that although they had considered walking away from each other, neither could follow through with such drastic action. The conflict brought into sharp focus what both sides could lose if coalitional presidentialism were to collapse: the president could have potentially lost his nominators and electoral machine network, and the parties risked their access to the spoils of government had Widodo gone on to win without some of them. Continuing their arrangement offered the least risk for both.

Recall that one of the patterns emerging from our comparative analysis in the introduction was that presidents whose own party is small tend to build oversized coalitions. We speculated that that might have to do with their greater anxiety compared to other incumbents with larger partisan bases. While this chapter has further substantiated that hypothesis (Yudhoyono directly confirmed it), the above discussion has also pointed to additional motivations. Most importantly, the benefit of a broad party coalition offers presidents the chance to reduce their dependence on their own party. But there are risks, too: if parties form a united front against the president, the parties have a high chance of prevailing. This pattern was visible in the nomination of Widodo’s running mate in 2018, but it is also affecting the policy arena. In September 2019, an alliance of Widodo’s coalition parties pressured the president to endorse revisions to the law on Indonesia’s Corruption Eradication Agency (KPK). The KPK had been a major nuisance to the parties, with dozens of their senior officials going to prison for corruption since the agency’s creation in 2003 (Blakkarly 2015). Widodo, too, felt unhappy about the KPK’s wide-ranging powers, but similar to Yudhoyono before him, he had been hesitant to act against it out of fear of a public backlash. However, faced with a collective stance formed by his coalition parties, Widodo acquiesced (Warburton 2019). In such cases, the regressive effect of broad-based coalitional presidentialism is apparent: the KPK lost significant elements of its powers, undermining anti-corruption policies (Mujani and Liddle 2021). Similarly, through Ma’ruf’s selection, a conservative cleric with a reputation for limiting minority rights became vice president, instead of Mahfud, a popular former chief justice who had presided over numerous democracy-supporting rulings from 2008 to 2013.

The logic of achieving maximum inclusiveness of party coalitions in post-2004 Indonesia has become so entrenched that it is hard to ascertain what the polity would look like without it. Since Yudhoyono’s ascension to power, there were only brief windows into such a counter-factual scenario. Yudhoyono was confronted with a hostile majority of opposition parties between October and December 2004, and Widodo faced a similar scenario at the beginning of his term (in his case, as noted, he had proudly claimed in the preceding campaign that such a minority status would not bother him). Both Yudhoyono and Widodo were shocked by short demonstrations of the parties’ powers (which consisted mostly of monopolizing DPR committees and delaying presidential initiatives) without actually experiencing consequences that could have threatened their presidencies. To be sure, governing without broad party support would be more unstable, but neither Yudhoyono nor Widodo wanted to test just how unstable—or what the potential benefits of smaller coalitions could be. It is worth emphasizing again that no Indonesian president operating under the post-2002 constitutional arrangements has faced an impeachment process, not even in its earliest stages. This fact indicates a tendency toward risk aversion on the part of presidents—but also shows their knowledge that implementing the broad outlines of coalitional presidentialism offers significant protections from impeachment and instability. As a result, the equilibrium generated by coalitional presidentialism has protected Indonesian democracy from destabilizing conflict, while at the same time paralyzing the polity in a state of policy conservatism.

However, political parties are only one element in the coalitional presidentialism architecture. There are others that presidents can resort to for assistance in elections and government. Including a wide range of non-party elements into presidential coalitions allows presidents to build additional safeguards against disloyalty by political parties or other non-party allies. One of these additional coalitional presidentialism actors is the legislature (as well as individual legislators). Although much of the coalitional presidentialism literature has combined parties and legislatures into one main actor presidents have to deal with, a separate treatment may be more appropriate. From a perspective of party strengths, parties hold powers as organizations that partially lie outside of their representation in legislatures (such as mobilization capacity during and between campaigns). The next chapter will showcase the other side of this equation: that is, the political weight of the DPR and its members that is unrelated to—or only loosely associated with—the parties they affiliate with. In reality, individual legislators can exercise so much power that presidents need to target them individually to secure their cooperation.