Voting behaviour under doubts of ballot secrecy: reinforcing dominant party rule

ABSTRACT

Ballot secrecy is a cornerstone of electoral democracy, since its real or perceived absence can make voters reluctant to express their true preferences. Through survey data from Singapore, we show that doubts over ballot secrecy can alter voting behaviour even when the vote is secret and there are no individually-targeted punishments or incentives; specifically, they lead a small subset of Singaporean voters to support the dominant party, despite a preference for the opposition. We also examine individual-level correlates of doubting ballot secrecy: a tendency towards belief in conspiracies and distrust of the mass media are the strongest predictors. Finally, a counterfactual exercise demonstrates the sensitivity of election outcomes to marginal vote swings; it suggests that doubting ballot secrecy can secure the dominant party a small number of additional parliamentary seats, thereby buttressing dominant party rule without requiring any concerted action or overtly repressive measures.

Acknowledgments

Research carried out under IRB from the University of British Columbia. We would like to thank Isabel Chew, Eugene Choo, Gabriele Gratton, Allen Hicken, Alan Jacobs, Gillian Koh, Tommy Koh, Edmund Malesky, Steven Oliver, Thomas Pepinsky, Nico Ravanilla, Dan Slater, Netina Tan, Risa Toha, Michael Weaver, Meredith Weiss, and the audiences at IPERG at Universitat de Barcelona, Electoral Malpractice in East and Southeast Asia APSA mini-conference, National University of Singapore, Universitat Autònoma de Barcelona, University of New South Wales, University of British Columbia, and the AAS, AEE, EPSAC, and WPSA conferences, for insightful comments and suggestions. All errors are ours.

Disclosure statement

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

Notes

1 Ferree and Long, “Gifts, Threats”; Cruz, “Vote Secrecy and Democracy”

2 See Mares and Young “Buying, Expropriating” for an overview.

3 Gutierrez-Romero, “An Inquiry into”; Kiewiet de Jonge and Nickerson, “Beliefs”.

4 See Birch, Electoral Malpractice (chapter 5) for “soft” pressures in electoral contexts. We believe Gerber et al, “Is There a Secret Ballot”, is the only other study to analyze how doubts of ballot secrecy affect voting behaviour in a context free of electoral violence and vote buying.

5 Curato and Fossati, “Authoritarian Innovations”; Pepinsky, “Authoritarian Innovations”; Morgenbesser, Rise of Sophisticated Authoritarianism. See also Schedler, “Menu”.

6 See the Elections Department Singapore full explanation at: https://www.eld.gov.sg/voters_ ballotsecrecy.html

7 Following a 2017 parliamentary debate, for example, prominent opposition MP Leon Perera wrote on his Facebook page “Our votes in elections are secret and can never be traced - no ifs, no buts.” See the publicly accessible Facebook page of Leon Perera, post on March 2, 2017.

8 Gerber et al, “Do Perceptions”

9 Stokes, “Perverse Accountability”

10 Kiewiet de Jonge and Nickerson, “Beliefs”

11 We thank one of the anonymous reviewers for raising an interesting philosophical question: from an existentialist perspective (in which an individual is the sum of their actions independent of intentions), what does it mean to vote against ‘true preferences’ (in other words, does such an action reveal the ‘true’ preference to be insincere)? Engaging this, and related Foucauldian critiques, is unfortunately beyond the scope of this paper. From a practical perspective, in a system with mandatory voting where ballot invalidation is widely recognized as a form of protest, a voter who believes their vote is tracked and carries potential repercussions may see a PAP-vote as their only risk-free option.

12 Jolley and Douglas, “The Social Consequences”

13 To the best of our knowledge, only Gerber et al, “Is There a Secret Ballot” have examined this before, but they focus on the effect of social stigma rather than state-directed retribution for vote choice.

14 Based on publicly available data, it appears that just over 3 percent of otherwise eligible voters did not appear on the 2015 electoral rolls due to being previously struck off and not applying for reinstatement.

15 Oliver and Ostwald, “Singapore’s Pandemic Election”

16 Magaloni, “The Game of Electoral Fraud”

17 Tan, “Manipulating Electoral Laws”

18 Weiss et al, “The 2015 General Election”

19 Lee, The Media, Cultural Control, and Government

20 Slater, “Strong State Democratization”

21 Tan, “Manipulating Electoral Laws”

22 Gomez, “Restricting Free Speech”

23 Ong, “Online Repression”

24 Mutalib, Parties and Politics

25 Chua, Liberalism Disavowed

26 Weiss, Roots of Resilience

27 Oliver and Ostwald, “Explaining Elections”

28 US Department of State, “2015 Human Rights Report”

29 Tan, Singapore

30 Official statistics released by the Ministry of Manpower: http://stats.mom.gov.sg/Pages/ Employment-Tables2016.aspx, see Table 58, sheet “T58 cont”.

31 Yeo, “Electoral Politics”

32 http://www.independent.co.uk/news/uk/politics/generalelection/

general-election-2015-explained-voting-10227175.html

33 Elklit and Maley, “Why Ballot Secrecy”

34 Caselli and Falco, “Your Vote”

35 This is captured in a segment from a Singaporean cable network on April 12, 2006, where several citizens interviewed the country’s founding Prime Minister Lee Kuan Yew (link: here). At the 10 min 50 sec mark in the link, a young citizen states “You have a young generation of people (…) that are really fearful (…) You’re talking about the old people and the young people who are afraid to vote against the PAP.” The dialogue continues as follows: [LKY] “Are you fearful to vote against the PAP?” [ML] “Perhaps, yes, honestly, a little bit!” [LKY] “[Chuckles] How would we know that you voted against us?” [ML] “[Smiling] I think we all know that.”

36 These doubts are commonly raised in informal online discussions. See two platforms that engage these rumors: https://blog.seedly.sg/is-your-vote-secret-singapore/ and https://blackdotresearch.sg/is-your-vote-truly-secret/

37 Singapore statistics are official and taken from HDB (2014) and https://data.gov.sg/. These statistics refer to the resident population (full citizens and permanent residents) in public housing (i.e., 80% of the total population), the only publicly available data for Singapore. Results presented in our tables are restricted to Singaporean citizens only, since permanent residents do not have the right to vote.

38 See www.globalwebindex.com for additional details.

39 The Singapore-based Institute of Policy Studies IPS, for example, conducts widely-publicized Post-Election Surveys that ask a range of direct questions on political preferences and voting behaviour; findings typically comport closely with actual outcomes: IPS, “Report on Forum”

40 Riambau and Ostwald, “Placebo Statements”

41 Kuhn and Vivyan, “The Misreporting”

42 A randomly selected subset of respondents received modified response options that allowed for greater variance in belief, differentiating between (a) certainty that the ballot is secret and (b) confidence but uncertainty that the ballot is secret. The final option remained as well.

43 A randomly selected subset of respondents received a modified version of the question that referred to “a recent election”.

44 IPS, “IPS Post-Election 2011”; IPS, “IPS Post-Election 2015”;

45 Maruah, “Defending”

46 Kiewiet de Jonge and Nickerson, “Beliefs About Ballot Monitoring”

47 Stokes, “Perverse Accountability”

48 Gerber et al, “Is There a Secret Ballot”

49 Ferree and Long, “Gifts, Threats”

50 Cruz, “Vote Secrecy”

51 While the Straits Times is seen as government aligned, assessments of it are strongly (positively) correlated with those of the BBC News and ChannelNewsAsia (the dominant cable news provider in Southeast Asia), giving us confidence that it captures a general orientation towards the mainstream media.

52 We estimate two additional series of models. In the first, the DV becomes belief in individually-targeted penalties for opposition voters (‘1’ for respondents who express that belief, ‘0’ otherwise). In the second, the DV becomes voting against opposition preferences (‘1’ for respondents who have voted for the PAP despite a preference for the opposition, ‘0’ otherwise). The results of these additional models are largely consistent with the original findings; they can be found in the online appendix.

53 The results are weighted to address under-sampling of elderly and Indian respondents. Unweighted results are substantively similar. 

54 Sim et al, “Public Housing”

55 For example, some opposition voters prefer a PAP government, but with a larger parliamentary opposition. See Sudhir Thomas Vadaketh’s Musings from Singapore, specifically the post entitled “To help the PAP and Singapore improve, I’m voting opposition.” If these voters anticipated significant opposition gains, some may opt to support the PAP instead. 

56 Singapore’s GRC system was instituted in 1988 to address both normative and practical concerns. Some scholars have argued that it provides the PAP with electoral advantages (see Tan, “Manipulating Electoral Laws”, for example). Notably, the PAP won all GRCs until GE 2011, when it lost one. It lost a second in GE 2020.

57 One multi-member district inadvertently went uncontested in GE 2011 due to the disqualification of the opposition candidates who sought to contest it.

58 Tan and Lee, Voting in Change

59 James, “Struggle for Singapore’s Chinese heartland”

60 Tan and Lee, Change in Voting

61 Oliver and Ostwald, “Singapore’s Pandemic Election”

62 Weiss, Roots of Resilience; Oliver and Ostwald, “Explaining Elections”

Additional information

Notes on contributors

Kai Ostwald

Kai Ostwald is associate professor in the School of Public Policy & Global Affairs and the Dept. of Political Science at the University of British Columbia. He is also director of UBC's Institute of Asian Research and associate editor of Pacific Affairs.

Guillem Riambau

Guillem Riambau (PhD, Boston University 2012) is an assistant professor of economics at the Universitat de Barcelona, member of the Barcelona Institute of Economics (IEB), and the Institutions and Political Research Center (IPERC).