Unopposed but not Uncontested: Brokers and 'Vote Buying' in the 2017 Pati District Election.

• Author: Tawakkal, George Towar Ikral

This article examines the brokerage activity and vote-buying behaviour in the 2017 local government election in the district of Pati, Central Java. This election is an unusual example of an incumbent candidate running unopposed: a growing trend in Indonesian local elections (see article by Cornells Lay et al. in this special issue). Despite being unopposed, Indonesian election laws provide a "blank box" option that voters can choose instead of supporting the unopposed candidate on the ballot. This creates pressure on candidates to seek the support of both political parties and voters needed to win on election day. During this particular election, for example, there was a campaign led by two prominent political leaders opposed to the candidate encouraging citizens to vote for the "blank box" and, in so doing, deny the incumbent re-election.

Based on interviews with the incumbent candidate, brokers representing the incumbent and the opposition, campaign funders, gamblers, party leaders and voters in the Pati regency, we examine the brokerage activity and campaign activities during this election. Our findings point to three main conclusions. First, despite running unopposed, there was campaign activity such as building a tim sukses ("success team" or campaign organization) by the incumbent that involved vigorous efforts to mobilize and persuade voters months before election day. These efforts were undertaken despite the incumbent candidate and his campaign team expressing great confidence in the weeks prior to the election about their chances of victory. The efforts also involved robust discussion about how much money certain brokers would offer to voters.

Second, and relatedly, such discussions focused heavily on cultural considerations that involved demonstrating respect and humility to voters instead of more transactional considerations. The offering of cash to voters occurred well after the candidate had secured enough votes to win the election (as indicated by internal polling), suggesting that "vote buying" in this election was less about transactional exchange of votes for cash than about showing respect and humility to voters.

Finally, we found very little betrayal among brokers (e.g., keeping the money that was intended for voters) who were distributing money on behalf of the incumbent. Possible explanations for this unusual level of broker loyalty include the likelihood that the incumbent would win the election, the use of a "spy team" to monitor brokerage activity and clientelistic relationships between brokers working within the incumbent's tim sukses. While unopposed elections appear on the surface to lack strong competition, such campaigns can sometimes involve rigorous activities similar to those found in hotly contested elections where multiple candidates are competing.

Brokerage Activity and Vote Buying in Indonesia

In any complex organization, specialization of labour is required. Political campaigns are often large operations that include numerous individuals hired to help the campaign win elections. Having a separation of duties and responsibilities in the form of separate "teams" within the same organization should hardly be surprising. Most campaigns in Indonesia use tim sukses ("success team") representing complex and interconnected teams of campaign specialists. The tim sukses is often hierarchical and structured according to existing geographic political units, including separate "teams" for each level. (1) These teams include party leaders, prominent party members, business and religious leaders, and so forth.

A central facet of the campaign team involves the role of "brokers"--intermediaries between the campaign and voters themselves--who are tasked with persuading voters to support the candidate and often (but not always) distributing cash and other goods to the voters on behalf of the campaign. (2) There is often a hierarchy between higher-level brokers providing sums of money and lower-level brokers (whom they hire themselves) to distribute money to voters. While campaign specialists are common in virtually all democratic elections, brokers have received significant academic attention because of their role in patronage and clientelistic politics. (3) Susan Stokes and her colleagues have developed a "broker-mediated theory" of clientelism in Latin America where brokers are central actors in distributing resources from campaigns to voters. In their theoretical framework, brokers provide specialized knowledge about voters--including candidate preferences and their turnout likelihood--that candidates need to win elections. (4) They find that brokers know the voting tendencies of voters within their respective networks, tend to target the candidate's supporters for vote buying and, to a lesser degree, swing voters to the extent that the broker wants their candidate to win. Yet brokers also have a personal and financial interest in extracting rents by keeping the money intended for voters or otherwise not working diligently on behalf of their candidate. (5)

The ability of brokers to betray their candidates, and the candidate's concerns about possible betrayal, is the focus of a new and growing body of research. Edward Aspinall offered one of the first serious efforts to build a theory explaining the conditions in which such betrayal is likely. Aspinall contends that brokers consider the material resources available to a campaign as well as the likelihood of that candidate winning the election. Loyalty to the candidate is expected when the campaign has a strong chance of winning the election but does not have a wealth of resources to offer voters. "Predation" occurs when brokers keep some of the money for themselves and is most likely when the candidate has a wealth of resources but is likely to lose the election. "Defection", on the other hand, refers to brokers leaving their candidate to work for an opposing campaign and is most likely to occur when a campaign has few resources to offer and only a weak chance of winning the election. (6) Another example of brokerage betrayal occurs when brokers act as a "double agent" by working simultaneously for two opposing candidates, even offering money to voters on behalf of both candidates at the same time. (7)

Candidates have few ways to punish such betrayal by the brokers they hire, largely due to the illegal nature of vote buying and other patronage activities. There are no legal contracts between candidates and brokers that can be enforced by the court system, for example, leaving few options for candidates betrayed by the brokers they hire. Candidates have, however, designed several innovative mechanisms to limit the likelihood of betrayal. In Venezuela, party leaders created elaborate hierarchical structures to monitor the behaviour of brokers and ensure loyalty. (8) In Taiwan, party leaders would frequently select brokers who had personal ties to campaign officials and also closely oversaw voter lists used by brokers. (9) More recent research in Indonesia similarly shows that candidates have resorted to monitoring voter lists supplied by brokers, (10) hiring brokers with close personal connections to the candidate and his team (11) and even hiring women because they are believed to be more honest than men. (12) These tactics tend to be effective at limiting broker betrayal, and monitoring brokerage performance has been shown to result in higher "turnout buying" in other countries. (13)

For both Aspinall and Stokes et al., the likelihood of winning the election plays an important role in the decision by brokers to betray their candidate or remain loyal. Indeed, most of the extant literature on brokerage activity is based within a competitive electoral environment where two or more candidates are vying for the same office. The goal of campaigns providing gifts to voters is to win their support and thus gain more votes on election day than the rival candidate(s). These strategies vary considerably, ranging from "turnout buying", (14) attempting to persuade swing voters (15) and even "negative vote buying" which offers money for likely supporters of the opposing candidate to abstain on election day. (16) Despite the strategic variations, though, the purpose of "vote buying" is to win a competitive election between two or more rival candidates.

Yet what about unopposed elections where only one candidate--often an incumbent seeking re-election--is the only one on the ballot? In many countries, including the United States, unopposed candidates automatically win the election if they receive only a single vote. These situations thus represent the extreme version of uncompetitive elections where the unopposed candidate is guaranteed to win. In Indonesia, however, the election laws require candidates to win a majority of votes cast in the election. Thus, citizens choose between casting a vote for the incumbent or instead for a "blank box" (loosely representing "none of the above") in the polling booth and, consequently, the outcome of the race is not always guaranteed. Of the nine major elections with unopposed candidates in Indonesia in 2017, for example, the "blank box" received as low as 3.8 per cent of the vote in West Kalimantan and as high as 44.9 per cent in Southeast Sulawesi. (17) In Pati district, the "blank box" received just over 25 per cent of the vote. Thus, most unopposed candidates won with very large margins, although in some rare instances the unopposed election can be surprisingly competitive. (18)

The rather uncompetitive nature of such elections raises interesting theoretical questions that have received less scholarly attention. First, if a candidate is virtually assured of winning the election we would expect there would be little incentive to expend significant amounts of money and other resources to build a large campaign team, including hiring brokers and offering gifts to persuade voters. Absent a strong and...