Skip to main content
NCBI home page
Public Opinion Quarterly logoLink to Public Opinion Quarterly
. 2024 Jul 11;88(SI):472–494. doi: 10.1093/poq/nfae022

The Dynamics of Electoral Manipulation and Institutional Trust in Democracies: Election Timing, Blatant Fraud, and the Legitimacy of Governance

Masaaki Higashijima 1, Hisashi Kadoya 2,✉, Yuki Yanai 3
PMCID: PMC11300039  PMID: 39109089

Abstract

This paper explores the dynamic relationship between electoral manipulation and popular trust in political institutions. Governments often manipulate election results by resorting to electoral fraud. They also tilt the electoral field by opportunistically deciding when to hold elections, in other words, election timing maneuvering. How do these two different types of electoral manipulation affect citizens’ trust in the government, legislature, and election management bodies (EMBs)? We suggest that although the short-term effects of election timing manipulation are unclear due to its ambiguous nature as an electioneering strategy, substantial electoral margins created by timing maneuvering facilitate smooth decision-making, leading to boosting trust in the government and legislature over the long run. In contrast, as electoral fraud is an unambiguous form of manipulation, it may undermine trust in the government and parliament, although such effects may not last. By combining an original dataset of election timing with existing survey data comprising 335,000 citizens from fifty-eight democratic countries, we find evidence in support of our theoretical expectations.

Introduction

This paper explores the dynamic relationships between electoral manipulation and institutional trust in democracies. Political leaders often utilize various techniques of electoral manipulation to stay in power. Even among democracies, levels of electoral integrity differ (Norris 2017). Electoral manipulation refers to the means by which political leaders tamper with election results to deviate from public vote preferences in favor of the regime (Higashijima 2022). By resorting to electoral fraud, incumbents bias election results with illegal measures such as violent and nonviolent intimidation of voters and opposition candidates, vote buying, ballot stuffing, and multiple voting (Kelley 2012; Simpser 2013; Norris 2014). Manipulating electoral rules, governments can also bias election results (Boix 1999; Lust-Okar and Jamal 2002; Boone and Wahman 2015; Washida 2018; Chang and Higashijima 2023). These distinct types of electoral manipulation affect popular perceptions of and trust in governments and electoral institutions in dissimilar ways (Reuter and Szakonyi 2021; Szakonyi 2022).

To explore the relationships between electoral manipulation and popular perception of political institutions, this paper centers on the maneuvering of election timing, a common form of electoral manipulation in democratic states. Although the burgeoning literature has begun to illuminate how various types of electoral manipulation influence the popular perception of government and elections, we know little about the impact of election timing maneuvering. Presuming that calling off-schedule elections is the typical technique of electoral manipulation for democratic leaders, the election timing literature has examined the causes and consequences of such elections, mostly focusing on macro-level variables such as whether economic conditions and constitutional arrangements affect the government’s decision to hold early elections (e.g., Ito and Park 1988; Kayser 2005) as well as whether early elections favor the seats and votes of ruling parties (Schleiter and Tavits 2016; McClean 2021). Shedding light on micro-level consequences of election timing manipulation, recent studies examine how citizens evaluate governments’ manipulation of election timing (Blais et al. 2004; Schleiter and Tavits 2018; Morgan-Jones and Loveless 2023; Turnbull-Dugarte 2023). These studies, however, have primarily focused on the immediate effects of election timing manipulation without considering what long-term implications election maneuvering may have on citizens’ views on politics. Furthermore, the general literature on electoral fraud fails to adequately consider the long-term consequences of election fraud, save for a few studies focusing on satisfaction with democracy (Higashijima and Kerr 2023; Morgan-Jones and Loveless 2023) and confidence in an election authority (Lundmark, Oscarsson, and Weissenbilder 2020). Although our primary focus is on election timing manipulation, we also analyze the consequences of blatant electoral fraud, thereby underscoring important features of timing maneuvering.

Given these gaps in mind, this paper illuminates how citizens’ confidence in political institutions, specifically trust in the government, legislature, and election management bodies (EMBs), is affected by election timing manipulation and electoral fraud. Political trust in institutions is defined as the popular belief in the righteousness of these institutions, which is an important indicator of capturing political legitimacy (Turper and Aarts 2017, p. 417). Introducing a dynamic perspective, we suggest that the effect of election timing manipulation on citizens’ institutional trust may change according to the electoral cycle in the short run, namely immediately after an election; election timing manipulation does not have clear effects on institutional trust due to its ambiguous nature regarding the government’s intentions to call early elections. However, over the long run, namely during nonelectoral periods between the last and next elections, election timing maneuvering helps governments gain confidence from citizens: greater majorities manufactured by opportunistic election timing facilitate smooth decision-making during nonelectoral periods, leading to a boost in voters’ confidence in government and legislature.

Conversely, electoral fraud is an unambiguous type of electoral manipulation in that citizens can relatively easily notice a situation in which the incumbents used it to bias election results in their favor. It is therefore likely to exacerbate public confidence in political institutions (Bratton 2008; Kerr 2013; Reuter and Szakonyi 2021). Since election fraud coercively distorts election results, citizens are more likely to lose institutional trust immediately after fraudulent elections. However, public trust in the government and parliament might not be reduced further because it is possible for legislative bodies to recoup public trust through the phase of policymaking during nonelectoral periods, which is buttressed by parliamentary majorities made by election fraud.

We test these theoretical expectations by combining an original dataset of election timing with existing survey data, which comprises approximately 335,000 respondents from 258 surveys taken in 58 democratic countries (1979–2019) across the globe. We focus on democracies because the arguments and assumptions in this paper build primarily upon existing studies on election timing, which are based on the experiences of democratic countries. Taking advantage of the time differences between the dates of surveys and the last elections, we estimate the dynamic effects of election timing manipulation on popular confidence in the government, legislature, and EMBs. We find that citizens who experienced early, unscheduled elections have greater levels of trust in the government and legislature as these elections become distant in time, whereas such elections do not necessarily impact popular trust in EMBs. We also find partial evidence that blatant fraud is negatively associated with institutional trust, but it does not necessarily have long-term effects.

This paper makes two important contributions. First, we contribute to the literature on election timing. By exploring the dynamic relationship between election timing maneuvering and institutional trust, we show that, paradoxically, the manipulation of election timing gradually improves political trust in the government and legislature. Second, we also contribute to the general literature on the consequences of electoral manipulation. Measuring electoral proximity, we compare the long-term effects of two typical types of electoral manipulation—overt electoral fraud and election timing. In so doing, we suggest that different means of electoral manipulation influence the prospect of democratic consolidation in distinct manners.

Electoral Manipulation and Popular Perceptions of Political Institutions

Scholarship on electoral integrity has explored the manner in which various types of electoral fraud affect popular perception of electoral processes and political institutions. Although the effects of vote buying are ambiguous in this regard (Bratton 2008; Kerr 2013; Weschle 2016), studies have demonstrated that other methods of overt electoral fraud, including exposure to ballot stuffing (Reuter and Szakonyi 2021), voter intimidation (Bratton 2008; Kerr 2013; Reuter and Szakonyi 2021), and electoral violence (Kerr 2013; Gutiérrez-Romero and LeBas 2020), induce a negative evaluation. Using survey experiments in Denmark and Mexico, Aarslew (2023) reports that respondents who are assigned vignettes that reveal electoral malpractices (e.g., vote buying, voter pressure, or ballot stuffing) tend to change their views on the election and government in a negative direction. Other studies examine the effects of overt electoral fraud in general. Focusing on the case of the 2015 Nigerian elections, Kerr (2018) constructed an additive scale that reflects voters’ exposure to the five types of blatant fraud (i.e, double voting, underage voting, noncitizens voting, tampering or stealing ballot boxes, and voter intimidation at polling stations), finding that Nigerian voters’ election-day experiences are highly associated with their assessment of electoral integrity.

These studies robustly show that blatant measures of electoral fraud are easily recognized by citizens and identified by third parties, leading to declines in the popular evaluation of electoral integrity and democratic institutions. In contrast, other types of electoral manipulation such as electoral system reforms, malapportionment, and gerrymandering are more ambiguous and therefore more difficult for voters to detect (Ong 2018; Higashijima 2021). Szakonyi (2022) focuses on one form of institutional manipulation, the de-registering of opposition candidates. His survey experiments in Russia demonstrate that voters respond less negatively to this method of candidate filtering than to blatant electoral fraud.

Among others, another ambiguous form of electoral manipulation is the opportunistic calling of elections—election timing maneuvering. The vast literature on election timing in democracies has illuminated when and how calling early parliamentary elections contributes to manufacturing election results in favor of incumbents.1 In most countries under parliamentary systems, prime ministers are entitled to dissolve parliament and call early elections. By timing elections to coincide with favorable situations such as good economic conditions, incumbents can signal their competence to voters and hence win elections by larger margins (e.g., Ito and Park 1988; Chowdhury 1993; Kayser 2005; Roy and Alcantara 2012). By calling such surprise elections, incumbents can also generate a strong incumbency advantage by catching opposition parties off guard and unprepared for elections (McClean 2021). Indeed, Schleiter and Tavits (2016) show that calling early parliamentary elections enables incumbents to obtain higher proportions of votes and seats in parliament.

Although election timing manipulation may help incumbents bias election results, scholars also report nuanced effects of timing maneuvering on popular perception of government legitimacy. On one hand, research suggests that opportunistic election timing undermines the incumbents’ credibility, electoral advantages, and democratic legitimacy. For instance, Smith (2004) argues that opportunistic decisions to call early elections work as a credible signal of incumbent weaknesses. Similarly, analyzing a snap election in Canada, Blais et al. (2004) found that citizens who support opposition parties tend to hold stronger resentment about opportunistic elections. Adopting a regression-discontinuity design to analyze US mayoral elections, de Benedictis-Kessner (2017) found that in off-cycle elections, incumbents are more likely to lose incumbency advantages. Consistent with these observations, Schleiter and Tavits’s (2018) survey experiment demonstrates that incumbents’ opportunistic decision to call early elections negatively affects voters’ support for the ruling party. According to Morgan-Jones and Loveless’s (2023) study using survey data from twenty-six European countries (2002–2016), calling early elections significantly lowers citizens’ satisfaction with democracy. On the other hand, recent research also documents that early elections may recover political trust in the government by providing voters with a new chance to endorse or reject the incumbent. Focusing on the case of the 2017 UK elections with a quasi-experimental design, Turnbull-Dugarte (2023) found that the announcement of the snap election increased political confidence in the incumbent.

The extant work has advanced our understanding of how governments manipulate elections and the consequences of manipulation on popular perception. However, the existing studies fail to address at least two important issues. First, the literature on electoral manipulation mostly focuses on the relationship between blatant measures of manipulation and public perception. Although several important studies started investigating the impacts of early election calling on government trust and satisfaction with democracy (Morgan-Jones and Loveless 2023; Turnbull-Dugarte 2023), we still know less about whether election timing maneuvering influences other aspects of public institutional trust, including confidence in the legislature and EMBs in addition to the government. Moreover, as these recent studies exclusively focus on European countries where blatant electoral fraud is highly uncommon, more research is needed to evaluate the impacts of election timing manipulation while comparing it with overt forms of electoral manipulation beyond the context of Europe.

Second, most of the extant research focuses on the short-term effects of electoral manipulation on popular perception by using surveys taken around election time or vignette experiments in which respondents are asked how much they support the ruling party when receiving randomly assigned information on blatant electoral fraud or election timing. The adoption of these research designs, however, does not allow us to investigate whether the effects of electoral manipulation persist over time. To fill this lacuna, we estimate the short- and long-run effects of electoral manipulation on public trust in the government, legislature, and EMBs. Taking advantage of temporal distances between survey dates and those of previous elections (e.g., Eifert, Miguel, and Posner 2010; Higashijima and Nakai 2016; Michelitch and Utych 2018; Higashijima and Kerr 2023), this study measures the degree of electoral proximity. This research design enables us to estimate how respondents change their confidence in those political institutions depending upon how close they are to the last election with varying degrees of electoral fraud and dissimilar election timing.

Dynamic Effects of Electoral Manipulation on Institutional Trust

Short-Term Effects of Electoral Manipulation

How electoral manipulation affects popular trust in institutions differs according to the type of manipulation technique. Specifically, how clearly a particular electoral manipulation technique conveys the government’s crude intentions of maintaining its power significantly influences the manner in which citizens perceive electoral integrity, including their confidence in relevant actors and political institutions surrounding elections.

In the case of overt electoral fraud such as election violence and intimidation, vote buying, and other forms of election irregularities, the government’s intention to use those methods is unambiguous: they attempt to use these techniques of electoral fraud to bias election results to stay in power. In democracies, free media often reports governments’ use of those coercive measures such as incidences of election violence and episodes of ballot-stuffing and fraudulent voting. Citizens directly hear about experiences of electoral fraud from their neighbors, friends, and relatives because events related to electoral violence and cheating stand out. Consequently, many citizens easily gain access to information regarding the government’s practices of overt electoral fraud. This unambiguous nature of blatant electoral fraud is likely to result in negatively affecting popular evaluations of the government (i.e., the manipulator), the legislature (i.e., the decision-making body where ruling parties occupy seats through manipulated elections), and EMBs (i.e., third parties guarding against electoral fraud) (Bratton 2008; Kerr 2013; Gutiérrez-Romero and LeBas 2020; Reuter and Szakonyi 2021).

In contrast to blatant electoral fraud, manipulating election timing is a more legitimate and thus more ambiguous manipulation technique with regard to government intentions in the eyes of voters. Consequently, voters react to election timing manipulation in dissimilar manners. Holding early elections, even abrupt ones, is exercised within the framework of the law and thus is seen as a legitimate act. This enables political leaders to obscure their intentions for calling early elections. For instance, when calling early elections, Japanese prime ministers have often publicly announced that they did so to seek public confidence via snap elections, even though such unscheduled elections have actually targeted the best timing in biasing election results in favor of the Liberal Democratic Party while inducing the opposition’s unpreparedness (McClean 2021) and opportunistically capitalizing on good economic conditions (Ito and Park 1988).

Because governments can obscure their political ambitions in the case of election timing maneuvering, some voters may not even recognize the calling of early elections as a method of electoral manipulation (Blais et al. 2004). As opposed to overt electoral fraud, election timing maneuvering is legal and thus more difficult to be denounced. If citizens do not recognize that calling early elections is a technique of electoral manipulation, they are less likely to change their trust in the government, legislature, and EMBs after experiencing early elections.

Furthermore, even if citizens recognize the manipulation of election timing as an intentional electioneering strategy by the government, citizens’ reactions may be diverse, as suggested in the previous section. On one hand, in off-cycle elections where ruling parties’ organized interests are strongly mobilized (Anzia 2014), some voters may be more exposed to such political mobilization. Consequently, they may not necessarily decrease trust in political institutions (cf. Blais et al. 2004). Other voters may positively evaluate the early calling of elections as the government’s honest action to recover political accountability (Turnbull-Dugarte 2023). In contrast, other portions of voters may perceive the government’s decision in calling early elections as a credible signal of incumbent weaknesses and thus decline their trust in government and legislature (Smith 2004; Schleiter and Tavits 2018; Morgan-Jones and Loveless 2023). As early elections tilt the electoral field in favor of ruling parties, those voters may also lose confidence in EMBs, the third parties administering free and fair elections.

Given the ambiguous nature of election timing manipulation, we expect that calling a parliamentary election earlier than originally scheduled may not have a clear impact on trust in the government, parliament, and EMBs immediately after elections. Conversely, in line with existing studies, we also expect that blatant electoral fraud is more likely to have a negative impact on trust in the government, legislature, and EMBs.

Hypothesis 1a: Election timing manipulation does not affect trust in the government, legislature, and EMBs immediately after elections.

Hypothesis 1b: Blatant electoral fraud is more likely to undermine trust in the government, legislature, and EMBs immediately after elections.

Long-Term Effects of Electoral Manipulation

Electoral manipulation may also impact public trust in institutions beyond immediate postelectoral periods. However, the effect of electoral manipulation is likely to change according to two parameters: (1) type of electoral manipulation (i.e., blatant electoral fraud vs. election timing maneuvering) and (2) type of institutions (i.e., government/parliament vs. EMBs).

Both electoral fraud and election timing maneuvering enable political leaders to win elections with larger margins than they could achieve without resorting to those election manipulation techniques. Leveraging stable majorities achieved by electoral manipulation, political leaders can engage in smooth policymaking throughout nonelectoral periods. Greater seat premiums generated by these manipulation techniques make it easier for ruling parties to form majorities in parliament. In both presidential and parliamentary systems, legislative majorities are pertinent in facilitating smooth communication and collaboration between the legislative and executive bodies while averting policy stalemates (Tsebelis 2002). Against this backdrop, prime ministers can achieve a stable fusion of power between the two bodies, which enables them to hold strong political leadership and thus smoothly implement the policy promises that they made during election campaigning. Also under presidential systems, avoiding a divided government and hence reducing the number of veto players is important for presidents to demonstrate their leadership while avoiding political gridlock (Mainwaring 1993).

However, as already discussed, blatant fraud and election timing are different in how they contribute to securing stable majorities in parliament. If a large number of voters recognize that winning elections was made possible through electoral fraud, the parliamentary majority does not have legitimate grounds. The public distrust created by overt electoral fraud is not easily dispelled and is even more likely to be amplified unless the reputation is restored in some way. Specifically, although the government and parliament may be able to recover public trust over the long run during nonelectoral periods by implementing smooth policymaking facilitated by the majority, public trust in EMBs is likely to be further undermined because EMBs do not engage in policymaking during nonelectoral periods.

In contrast, election timing manipulation is likely to increase trust in the government and legislature over the long run. First, similar to blatant electoral fraud, election timing manipulation contributes to winning elections. Previous studies have robustly shown that early elections allow incumbents to gain higher proportions of seats and votes than on-time elections in democracies (Schleiter and Tavits 2016). In addition to making it possible to coast on good economic conditions (Ito 1990; Palmer and Whitten 2000; Kayser 2005), early elections contribute to winning elections by promoting the opposition’s unpreparedness at election campaigning, biasing election results in favor of incumbents (McClean 2021).

Importantly, election timing maneuvering does not involve illegal violence and cheating. Due to its ambiguous nature as a method of electoral manipulation, the manipulation of election timing does not have significant repercussions on public trust also in the long run. Rather than undermining institutional trust, election timing maneuvering contributes to boosting trust in the government and legislature by facilitating parliamentary majorities. The subsequent stable decision-making buttressed by parliamentary majorities enables the ruling parties to demonstrate their policy competence and competent political leadership. The government and legislature have a responsibility to make decisions and fulfill public needs. If these institutions can provide public services appropriately and meet public needs, it is then undoubtedly beneficial for citizens. Even if they do not engage in public goods provisions, strong leadership anchored by parliamentary majorities impresses voters that the incumbents are competent without facing strong opposition in the executive-legislative relationship. Indeed, many studies suggest that trust in these institutions increases if the incumbent can demonstrate competent policy initiatives through stable parliamentary majorities (Miller and Listhaug 1999; Mishler and Rose 2001; van Erkel and van der Meer 2016). On the other hand, since EMBs cannot enjoy such trust premiums during nonelectoral periods, levels of trust in EMBs do not change much throughout the election cycle.

Hypothesis 2a: Election timing manipulation is more likely to increase trust in the government and legislature but not in EMBs in the long run.

Hypothesis 2b: Blatant electoral fraud is more likely to decrease trust in EMBs but not trust in the government and parliament in the long run.

Data Analysis

Data and Models

We estimate the effect of election timing manipulation or electoral fraud on people’s trust in the government, parliament, and EMBs. In addition to the main explanatory variable, we follow the previous literature (Eifert, Miguel, and Posner 2010) and add the number of days that have elapsed since the last election before the survey as an explanatory variable. As written above, we expect to find the effect of timing manipulation only in the long run. To estimate the change depending on the time difference between a survey and an election, we include an interaction term of an early election dummy and the days elapsed since the last election.

As measures of our outcome variables—people’s trust in the government, parliament, and EMBs—we use survey data collected by international survey projects: Afrobarometer, the Asian Barometer, Latinobarómetro, and the World Values Survey.2 These surveys ask about people’s trust in the government, parliament, and EMBs, and provide 4-point scale measures for the responses.3 Each survey asks respondents how much they trust the government/parliament/EMBs: (1) not at all, (2) just a little, (3) somewhat, and (4) a lot.4 We analyze 258 surveys from 58 democratic countries5 fielded by these surveys, and the number of responses to the questions of interest is about 335,000. We distinguish democracies from autocracies by referring to Boix, Miller, and Rosato’s (2013) binary measure of political regimes. We limit our analysis only to democracies because some of the important assumptions to validate our arguments on election timing hold only in democracies. We assume that early election calling is positively associated with increases in the vote and seat shares of ruling parties (e.g., Schleiter and Tavits 2016). However, this assumption has not yet been substantiated in autocracies. Furthermore, election timing manipulation occurs only in legislative elections in democracies. However, even the timing of presidential elections is frequently manipulated in autocracies. If we extend our analysis to authoritarian regimes, we need to also include presidential elections, which complicates mechanisms and theoretical predictions.

Figure 1 presents the distribution of the responses pooled across all surveys analyzed in this paper. As shown in the figure, most people do not trust the government, parliament, or EMBs much. The mean levels of trust are 2.32, 2.23, and 2.32, and their standard deviations are 0.99, 0.97, and 1.00 for trust in the government, the parliament, and EMBs, respectively. The average trust in the government is highest in Burundi (3.40) and lowest in Colombia (1.73). Similarly, trust in the parliament is highest in Burundi (3.12) and lowest in Ecuador (1.66); trust in EMBs is highest in Uruguay (2.88) and lowest in Ecuador (1.74).6

Figure 1.

Figure 1.

Open in a new tab

Popular trust in the government, parliament, and EMBs. Source: Afrobarometer, Latinobarómetro, and the World Values Survey.

To measure the timing of parliamentary elections, we use our original dataset. Our dataset records the dates when each parliamentary election was initially planned and when they were actually held. To determine when the election was initially planned, we referred to either fixed schedules (e.g., the US congressional elections) or the conclusion of parliamentary terms. An election was classified as early or delayed if a temporal gap of more than two weeks existed between the two aforementioned dates.7 Using this information, we can tell which elections were called earlier than scheduled. To estimate the difference between early elections and on-schedule ones, we omitted cases where the election took place later than planned; we used a dummy variable indicating an early election. Among the 122 legislative elections included in this analysis, 25 elections were early elections, and 97 were held as initially scheduled.8

We constructed a measure of electoral fraud similar to the clean elections index provided by V-Dem (Pemstein et al. 2020; Coppedge et al. 2021).9 V-Dem’s clean elections index10 includes EMB capacity and autonomy, which we need to exclude because one of our outcome variables is trust in EMBs. Moreover, V-Dem’s index captures some covert forms of electoral fraud. We created a new measure of blatant electoral fraud by excluding the variables related to EMBs or ambiguous electoral fraud. Using the V-Dem’s index related to blatant electoral fraud,11 we conducted a factor analysis and extracted a single factor following the V-Dem’s clean elections index. We scaled it to a real value between 0 (fraudulent) and 1 (clean), reversed the scale, and obtained a variable representing the level of electoral fraud. The variable ranges from 0.016 to 0.889 within democracies.12 Countries such as Albania (1997, 2001), Burundi (2010), Colombia (2002), Kenya (2007), Malawi (2004), and Mozambique (1999) are democracies exhibiting high levels of blatant electoral fraud in our sample.13

Next, we calculated the number of days elapsed between the last election and the survey using the variable provided by the surveys14 and the election dates we collected. Figure 2 displays the distribution of the number of days elapsed. The mean of this variable is 717 days, and the standard deviation is 467 days.

Figure 2.

Figure 2.

Open in a new tab

The number of days elapsed since the last legislative election before the survey. Source: Afrobarometer, Asian Barometer, Latinobarómetro, the World Values Surveys, and an original dataset of election timing.

In addition to the explanatory variable of our concern, we controlled for some potential confounders. First, we controlled for the status of the country’s economy. As previous studies show, a good economy could trigger an early election because the government would like to hold an election when it expects to get more votes; furthermore, the country’s economic performance could affect people’s trust in politics because most people in any country prefer a good economy to a bad one. We used four variables to control for this confounding factor: two variables from NELDA (Hyde and Marinov 2012, 2019)—NELDA17, which records whether people think the economic performance of the country is good, and NELDA18, which records whether people think the country is in economic crisis—GDP growth rate at the time of the election, and the change in GDP per capita from the election year to the survey year.

Second, we controlled for the electoral margin in the election previous to the election whose timing is our primary interest. A government that won big in the last election might not care much about election timing. At the same time, a wider electoral margin could strengthen people’s trust in the government because it suggests that many other people support that government. We calculated the difference between the vote shares of the first and second parties using the data provided by V-Party (Lührmann et al. 2020; Pemstein et al. 2020).

Third, we controlled for the level of electoral fraud when we regressed trust in early elections. If the incumbents wish to stay in office, they may resort to whatever measures they can take (Simpser 2013). If this is the case, we then suspect that electoral fraud is associated with election timing. If people detect that an election is fraudulent, they will trust the government and parliament less. As this variable could be affected by election timing rather than vice versa, we compared the model with and without this variable. It turns out that the inclusion of this variable makes no difference in the effect in which we are interested. Thus, we report the results with this variable controlled.15

Fourth, we controlled for the level of democracy. While our dataset exclusively consists of democratic systems, it is crucial to acknowledge the inherent variability in the quality and extent of democratic practices across these nations. Notably, it is plausible that governments with lower levels of democratic governance exhibit a higher propensity to resort to early elections compared to their more democratic counterparts. Simultaneously, the trust of individuals in the government and parliament may be substantially eroded within nations that have yet to fully embrace democratic principles. To address the issue, we introduced the “polyarchy index” as a controlling variable, which is crafted from V-Dem’s “Electoral democracy index” (Pemstein et al. 2020; Coppedge et al. 2021). Given that V-Dem’s index incorporates the notion of electoral fraud (clean elections index) in its composition, we excluded this element and devised a new measure accordingly.16

Furthermore, we added four individual-level predictors to make our estimation more precise. The variables included are the respondents’ age, gender (female dummy), employment status (unemployment dummy), and area of residence (urban residence dummy)17 provided by each survey.

With these explanatory variables, we estimated parameters of linear regression models where the outcome variable is people’s trust in the government, their trust in parliament, or their trust in EMBs. We included country- and year-fixed effects and calculated standard errors by treating countries as clusters.

Results

Figure 3 18 and table 1 show the results of the regressions whose main explanatory variable is the indicator of an early election.19 The left panel displays the result for people’s trust in the government as the outcome variable, the middle presents that for trust in parliament, and the right shows that for trust in EMBs. In each panel, the horizontal axis is the number of days that have elapsed since the last election. The vertical axis shows the effect of early elections on trust. The solid line is the point estimate, and the shaded area around the line displays the 95 percent confidence interval. Because we include the interaction term of timing and the number of days elapsed, the marginal effect of election timing changes depending upon the days.

Figure 3.

Figure 3.

Open in a new tab

The marginal effect of the early elections on popular trust in the government (left panel), parliament (middle), and EMBs (right), conditional on the number of days elapsed since the last election. Estimation results are presented in table 1.

Table 1.

Regression results: effect of early elections on trust.

Outcome
Trust in government Trust in parliament Trust in EMB
Early election 0.003 −0.008 0.000
(0.967) (0.960) (0.999)
Days from last election −0.003 −0.001 −0.001
 (10 days) (0.000) (0.404) (0.348)
Early x Days 0.002 0.003 −0.002
(0.081) (0.178) (0.363)
Electoral fraud −0.660 −0.829 −0.661
(0.435) (0.028) (0.022)
NELDA17 0.046 −0.007 −0.055
 (Good economy) (0.577) (0.873) (0.549)
NELDA18 −0.144 −0.156 −0.118
 (Economic crisis) (0.083) (0.007) (0.229)
GDP growth 0.006 −0.006 −0.011
(0.446) (0.434) (0.324)
Change in GDP per capita 1.074 0.612 0.032
 (from election to survey) (0.017) (0.369) (0.953)
Electoral margin −0.003 −0.002 −0.004
(0.458) (0.423) (0.028)
Polyarchy −0.113 −0.250 0.713
(0.840) (0.415) (0.180)
Age 0.003 0.001 −0.001
(0.000) (0.140) (0.252)
Gender (female) −0.029 −0.021 −0.021
(0.000) (0.001) (0.195)
Unemployment 0.008 0.016 0.003
(0.223) (0.004) (0.842)
Urban −0.010 0.004 −0.116
(0.615) (0.801) (0.000)
Adj. R 0.133 0.113 0.048
Num. obs. 335,453 325,289 105,382
N Clusters 58 58 32
Open in a new tab

Note: p-values calculated with robust standard errors clustered by country are in parentheses. p-values refer to two-tailed tests. The first two models include year-fixed effects and all models include country-fixed effects.

As expected by our first hypothesis (H1a), an early election does not affect people’s trust in the government, parliament, or EMBs right after the election. In figure 3, the confidence interval in each panel contains zero when the survey date is close to the date when the country had the last national legislative election. The 95 percent confidence intervals of the expected effect on election day are [−0.16, 0.17] for trust in the government, [−0.31, 0.29] for trust in parliament, and [−0.32, 0.32] for trust in EMBs.20 These results show that the government cannot instantly enhance people’s trust in the government, parliament, or EMBs by calling an early election.

Similarly, figure 4 and table 2 present the results of the regressions whose main explanatory variable is election fraud. Again, because we include the interaction term of electoral fraud and the number of days elapsed, the marginal effect of fraud changes over time. Partially supporting the second part of our first hypothesis (H1b), electoral fraud is negatively associated with public trust in parliament in a statistically significant way after a few months from the election. Although the coefficients for trust in the government and EMBs are negative, they are not statistically different from zero. In figure 4, the confidence interval in each panel contains zero when the survey date is close to the date when the country had the last national legislative election. The 95 percent confidence intervals of the expected effect on election day are [−2.33, 1.05] for trust in the government, [−1.88, 0.04] for trust in parliament, and [−1.43, 0.66] for trust in EMBs.

Figure 4.

Figure 4.

Open in a new tab

The marginal effect of electoral fraud on popular trust in the government (left panel), parliament (middle), and EMBs (right), conditional on the number of days elapsed since the last election. Estimation results are presented in table 2.

Table 2.

Regression results: effect of electoral fraud on trust.

Outcome
Trust in government Trust in parliament Trust in EMB
Electoral fraud −0.644 −0.923 −0.385
(0.449) (0.059) (0.460)
Days from last election −0.002 0.000 0.000
 (10 days) (0.205) (0.996) (0.881)
Fraud x Days −0.002 −0.001 −0.003
(0.476) (0.745) (0.542)
NELDA17 0.029 −0.021 −0.038
 (Good economy) (0.717) (0.654) (0.689)
NELDA18 −0.163 −0.192 −0.105
 (Economic crisis) (0.063) (0.002) (0.282)
GDP growth 0.003 −0.014 −0.006
(0.732) (0.064) (0.560)
Change in GDP per capita 1.054 0.794 −0.069
 (from election to survey) (0.037) (0.249) (0.891)
Electoral margin −0.002 −0.001 −0.004
(0.590) (0.695) (0.046)
Polyarchy −0.422 −0.683 0.702
(0.354) (0.040) (0.221)
Age 0.003 0.001 −0.001
(0.000) (0.135) (0.187)
Gender (female) −0.029 −0.021 −0.021
(0.000) (0.001) (0.199)
Unemployment 0.009 0.016 0.004
(0.217) (0.004) (0.768)
Urban −0.013 0.000 −0.110
(0.517) (0.998) (0.000)
Adj. R2 0.132 0.112 0.047
Num. obs. 335,453 325,289 105,382
N Clusters 58 58 32
Open in a new tab

Note: p-values calculated with robust standard errors clustered by country are in parentheses. p-values refer to two-tailed tests. The first two models include year-fixed effects and all models include country-fixed effects.

Considering that our measure indicates clear instances of fraud, it is perplexing that such fraud does not erode public trust in the government. One possible reason for this unexpected finding is the relatively small variation in fraud within our sample, leading to a large standard error of the parameter. In fact, the point estimate is negative regardless of how close to the election the survey was conducted, as depicted in figure 4. We focus exclusively on democratic countries, where levels of blatant fraud tend to be lower compared to autocracies. Consequently, the observed fraud index in our sample predominantly consists of lower values and does not span the entire variable range (0–1).21 This limitation might be responsible for the large standard error and the lack of a significant effect.

However, as shown in figure 3, the effect of early elections becomes larger as time elapses. After a lapse of 500 days since the last election, the effect is statistically significant at the 5-percent significance level in the middle panel in figure 3. Similarly, the left panel shows a statistically significant effect after about 1,000 days since the election. In the long run, an early election improves people’s trust in the government and parliament. Five hundred days after the election, the estimated effect is approximately 0.2, about a fifth of the standard deviation. Given that our outcome variable is based on a 4-point scale measure, this effect is not small. These results support our second hypothesis (H2a). Interestingly, unlike these two outcomes, people’s trust in EMBs does not depend on election timing. Its 95-percent confidence interval always contains zero regardless of the days elapsed since the previous election. This may suggest that an efficient policymaking basis produced by stable parliamentary majorities only increases trust in the government and legislature but not trust in EMBs because EMBs are not directly related to the distribution of seats in parliament and thus policymaking processes. The overall results indicate that calling an early election could contribute to regime stability through people’s trust in the government and parliament after certain periods of time.22

In contrast, as can be seen in figure 4, electoral fraud does not change people’s trust in the government or parliament in the long run. Long-term effects on trust in the government and parliament tend to be flat. Trust in parliament becomes statistically insignificant as the last election becomes distant in time. These results are in line with H2b suggesting that both the government and parliament are able to recover public trust in nonelectoral periods where they can engage in smooth policymaking supported by stable parliamentary majorities manufactured through electoral fraud. In contrast, electoral fraud tends to become more negatively associated with trust in EMBs over the long run, although the effect is not statistically significant. These results offer partial support for our last hypothesis (H2b).

Conclusion

We have investigated how election timing affects people’s trust in political institutions. Timing change is an obscure electoral tool, and thus it makes no difference in popular attitudes soon after the election. However, the effect gradually increases over time. As time lapses since the election, people who experienced an early election have higher levels of trust in the government and parliament than those who had an on-schedule election. We suggested that this is because the government could secure a more stable majority by calling an election at an arbitrary time. A larger majority itself might make people trust the government more. Furthermore, with a firmer grasp of the parliament, the government might achieve its policy goals more efficiently, enhancing people’s trust in the government and parliament.

Our study leads to a few important directions for future research. First, although we focused on election timing manipulation in democracies, it may also be possible to apply our framework to the context of autocracies. To do so, more research needs to be done regarding the causes and consequences of election timing maneuvering in dictatorships. Second, our analysis conceptualized election fraud as unambiguous and election timing maneuvering as ambiguous techniques of manipulation. However, the ambiguity of electoral fraud may also differ according to the techniques utilized (e.g., Szakonyi 2022). Using subcomponents of the clean elections index in the V-Dem dataset, scholars can untangle dissimilar effects of electoral fraud on institutional trust through cross-national quantitative analysis. Finally, voter partisanship may influence the manners in which voters associate election timing maneuvering with their trust in institutions. Although we did not adequately touch on this issue due to numerous missing values of voter partisanship variables, future studies should consider the role of voter partisanship moderating between electoral manipulation and institutional trust.

Supplementary Material

nfae022_Supplementary_Data
nfae022_supplementary_data.pdf (864.3KB, pdf)

Acknowledgements

An earlier version of this paper was presented at the 2021 World Congress of the International Political Science Association. We appreciate the participants of the panel as well as the Public Opinion Quarterly editors and the three anonimous reviewers for their helpful comments.

Footnotes

1

Governments may also be forced to call early elections due to the breakdown of coalitions and a vote of no confidence. Building upon extant studies, our theory also explores how opportunistic election timing impacts institutional trust. Empirically, we consider this important issue by considering whether a government consists of a single party or coalition.

2

We analyze the data from Waves 1 through 6 of Afrobarometer, Wave 3 of the Asian Barometer, waves covering 2001 through 2017 of Latinobarómetro, and Waves 2 through 7 of the World Values Survey. Total number of survey waves is twenty-eight. For more detailed information on the methodology of each survey, see the following URLs: Afrobarometer (https://www.afrobarometer.org/surveys-and-methods/), Asian Barometer (https://www.asianbarometer.org/survey.html?page=s40), Latinobarómetro (https://www.latinobarometro.org/latContents.jsp), and World Values Survey (https://www.worldvaluessurvey.org/WVSContents.jsp?CMSID=FieldworkSampling&CMSID=FieldworkSampling).

3

In exact wording, not all of the questions we use ask for trust in the government, parliament, or EMBs. See Supplementary Material section B for details. For exact wording of the questions about our outcome variables, see Supplementary Material section C.

4

The datasets provided by the surveys assign the smallest value to “a lot” and the largest to “not at all.” In order to measure trust instead of distrust, we assign the value of 1 to “not at all” and 4 to “a lot.” In fact, the options for response vary slightly from survey to survey. However, in all survey waves, respondents are asked to rate their level of trust on a 4-point scale.

5

We excluded countries for which the other variables necessary for the analysis are unavailable. The countries we analyze in this study are listed in Supplementary Material table B2.

6

Trust in EMBs is measured in only thirty-two out of fifty-eight countries in our sample.

7

For a more detailed explanation, see Supplementary Material section B.

8

Supplementary Material figure B1 shows when an early election was held relative to the scheduled date. As can be seen in the figure, most early elections in our sample were truly early elections rather than “normal” elections that occurred just a few weeks earlier than the term expiration.

9

V-Dem is a dataset that provides multidimensional and disaggregated indicators of democracy based on expert-coded data. V-Dem aggregates expert-coded data by means of a measurement model to provide valid and reliable estimates of concepts. For more information on its methodology, see Pemstein et al. (2020).

10

It is labeled as v2xel_frefair in the V-Dem dataset. Clean elections index is formed by taking the point estimates from a Bayesian factor analysis model of the indicators for EMB autonomy (v2elembaut), EMB capacity (v2elembcap), election voter registry (v2elrgstry), election vote buying (v2elvotbuy), election other voting irregularities (v2elirreg), election government intimidation (v2elintim), nonstate electoral violence (v2elpeace), and election free and fair (v2elfrfair) (Coppedge et al. 2020, p. 47).

11

That is, we constructed the fraud index from vote buying (v2elvotbuy), other voting irregularities (v2elirreg), and government intimidation (v2elintim) included in V-Dem (Coppedge et al. 2020).

12

Supplementary Material figure B2 shows the distribution of the blatant fraud index for both democracies and autocracies.

13

Supplementary Material figure B3 presents the bivariate relationship between the fraud variable and the level of polyarchy that we constructed based on V-Dem (see Supplementary Material section B). We can see that the two variables are negatively correlated. Some countries with a high level of fraud are labeled in the figure.

14

The exact date (day) is not available in some surveys, and we input the first day of the month in those cases.

15

Results without this variable are presented in figure A1 and tables A1–A3 in Supplementary Material section A.

16

A more detailed explanation is available in Supplementary Material section B. We also created the liberal democracy index by excluding the fraud element from V-Dem’s “Liberal democracy index.” We present the results using the polyarchy in the main text and provide the results using liberal democracy in Supplementary Material section B.

17

Afrobarometer and the Asian Barometer provide the two-category variable of urban vs. rural residence. The other surveys contain the variable recording the size of the town. We treat the respondents whose town size is larger than 50,000 as urban residents.

18

We show the results of the model that controls for blatant electoral fraud here. The similar figure without controlling for electoral fraud is presented in Supplementary Material figure A1.

19

Supplementary Material tables A1–A3 present the regression results with other model specifications.

20

It is unrealistic that a survey is conducted on election day, however.

22

The results where we take into account the effect of a coalition government are presented in Supplementary Material section A.

Contributor Information

Masaaki Higashijima, Associate Professor, Institute of Social Science, University of Tokyo, Tokyo, Japan.

Hisashi Kadoya, Assistant Professor, Faculty of Political Science and Economics, Waseda University, Tokyo, Japan.

Yuki Yanai, Associate Professor, School of Economics & Management, Kochi University of Technology, Kochi, Japan; and Research Fellow, Graduate School of Law, Kobe University, Kobe, Japan.

Supplementary Material

Supplementary Material may be found in the online version of this article: https://doi.org/10.1093/poq/nfae022.

Funding

This work was financially supported by JSPS KAKENHI [grant numbers JP19H01447, PI: Y.Y. and JP23H00779, PI: Y.Y.].

Data Availability

Replication data and documentation are available at https://doi.org/10.7910/DVN/JT6Q4K.

References

  1. Aarslew Laurits F.  2023. “Why Don’t Partisans Sanction Electoral Malpractice?” British Journal of Political Science  53:407–23. [Google Scholar]
  2. Anzia Sarah.  2014. Timing and Turnout: How Off-Cycle Elections Favor Organized Groups. Princeton: Princeton University Press. [Google Scholar]
  3. Blais André, Gidengil Elisabeth, Nevitte Neil, Nadeau Richard. 2004. “Do (Some) Canadian Voters Punish a Prime Minister for Calling a Snap Election?” Political Studies  52:307–23. [Google Scholar]
  4. Boix Carles.  1999. “Setting the Rules of the Game: The Choice of Electoral Systems in Advanced Democracies.” American Political Science Review  93:609–24. [Google Scholar]
  5. Boix Carles, Miller Michael, Rosato Sebastian. 2013. “A Complete Data Set of Political Regimes, 1800–2007.” Comparative Political Studies  46:1523–54. [Google Scholar]
  6. Boone Catherine, Wahman Michael. 2015. “Rural bias in African electoral systems: Legacies of unequal representation in African democracies.” Electoral Studies  40:335–46. [Google Scholar]
  7. Bratton Michael.  2008. “Vote Buying and Violence in Nigerian Election Campaigns.” Electoral Studies  27:621–32. [Google Scholar]
  8. Chang Eric C. C., Higashijima Masaaki. 2023. “The Choice of Electoral Systems in Electoral Autocracies.” Government and Opposition  58:106–28. [Google Scholar]
  9. Chowdhury Abdur R.  1993. “Political Surfing over Economic Waves: Parliamentary Election Timing in India.” American Journal of Political Science  37:1100–118. [Google Scholar]
  10. Coppedge Michael, Gerring John, Knutsen Carl Henrik, Lindberg Staffan I., Teorell Jan, Alizada Nazifa, Altman David, Bernhard Michael, Cornell Agnes, Fish M. Steven, Gastaldi Lisa, Gjerløw Haakon, Glynn Adam, Hicken Allen, Hindle Garry, Ilchenko Nina, Krusell Joshua, Luhrmann Anna, Maerz Seraphine F., Marquardt Kyle L., McMann Kelly, Mechkova Valeriya, Medzihorsky Juraj, Paxton Pamela, Pemstein Daniel, Pernes Josefine, von Römer Johannes, Seim Brigitte, Sigman Rachel, Skaaning Svend-Erik, Staton Jeffrey, Sundtröm Aksel, Tzelgov Eitan, Wang Yi-ting, Wig Tore, Wilson Steven, Ziblatt Daniel. 2021. “V-Dem [Country–Year/Country–Date] Dataset v11.1.” Varieties of Democracy (V-Dem) Project. https://www.v-dem.net/en/data/archive/previous-data/v-dem-dataset/. Date accessed 9 May 2024.
  11. Coppedge Michael, Gerring John, Knutsen Carl Henrik, Lindberg Staffan I., Teorell Jan, Altman David, Bernhard Michael, Fish M. Steven, Glynn Adam, Hicken Allen, Marquardt Kyle L., McMann Kelly, Paxton Pamela, Pemstein Daniel, Seim Brigitte, Sigman Rachel, Skaaning Svend-Erik, Staton Jeffrey, Cornell Agnes, Gastaldi Lisa, Gjerløw Haakon, Mechkova Valeriya, von RöMer Johannes, Sundtröm Aksel, Tzelgov Eitan, Uberti Luca. 2020. “V-Dem Codebook v10.” Varieties of Democracy (V-Dem) Project. https://curate.nd.edu/articles/dataset/Varieties_of_Democracy_v_10_2020_/24858180?file=43736778. Date accessed 9 May 2024.
  12. de Benedictis-Kessner Justin.  2017. “Off-Cycle and Out of Office: Election Timing and the Incumbency Advantage.” The Journal of Politics  80:119–32. [Google Scholar]
  13. Eifert Ben, Miguel Edward, Posner Daniel. 2010. “Political Competition and Ethnic Identification in Africa.” American Journal of Political Science  54:494–510. [Google Scholar]
  14. Gutiérrez-Romero Roxana, LeBas Adrienne. 2020. “Does Electoral Violence Affect Vote Choice and Willingness to Vote? Conjoint Analysis of a Vignette Experiment.” Journal of Peace Research  57:77–92. [Google Scholar]
  15. Higashijima Masaaki.  2021. “Blatant Electoral Fraud and the Value of a Vote.”Japanese Journal of Political Science  22:1–14. [Google Scholar]
  16. Higashijima Masaaki.  2022. The Dictator’s Dilemma at the Ballot Box: Electoral Manipulation, Economic Maneuvering, and Political Order in Autocracies. Ann Arbor: University of Michigan Press. [Google Scholar]
  17. Higashijima Masaaki, Kerr Nicholas. 2023. “When Does the Honeymoon End? Electoral Cycles of Satisfaction with Democracy in Africa.” Political Psychology  44:709–28. [Google Scholar]
  18. Higashijima Masaaki, Nakai Ryo. 2016. “Elections, Ethnic Parties, and Ethnic Identification in New Democracies.” Studies in Comparative International Development  51:124–46. [Google Scholar]
  19. Hyde Susan D., Marinov Nikolay. 2012. “Which Elections Can Be Lost?”Political Analysis  20:191–210. [Google Scholar]
  20. Hyde Susan D., Marinov Nikolay. 2019. “Codebook for National Elections across Democracy and Autocracy Dataset, 5.0.” https://www.dropbox.com/s/amvyfi6cnpy19ew/NELDA_Codebook_V5.pdf?e=1&st=b33o4k3n&dl=0. Date accessed 9 May 2024.
  21. Ito Takatoshi.  1990. “The Timing of Elections and Political Business Cycles in Japan.” Journal of Asian Economics  1:135–56. [Google Scholar]
  22. Ito Takatoshi, Park Jin Hyuk. 1988. “Political Business Cycles in the Parliamentary System.” Economics Letters  27:233–38. [Google Scholar]
  23. Kayser Mark Andreas.  2005. “Who Surfs, Who Manipulates? The Determinants of Opportunistic Election Timing and Electorally Motivated Economic Intervention.” American Political Science Review  99:17–27. [Google Scholar]
  24. Kelley Judith G.  2012. Monitoring Democracy: When International Election Observation Works, and Why It Often Fails. Princeton: Princeton University Press. [Google Scholar]
  25. Kerr Nicholas.  2013. “Popular Evaluations of Election Quality in Africa: Evidence from Nigeria.” Electoral Studies  32:819–37. [Google Scholar]
  26. Kerr Nicholas.  2018. “Election-Day Experiences and Evaluations of Electoral Integrity in Unconsolidated Democracies: Evidence from Nigeria.” Political Studies  66:667–86. [Google Scholar]
  27. Lührmann Anna, Duüpon Nils, Higashijima Masaaki, Kavasoglu Yaman Berker, Marquardt Kyle L., Bernhard Michael, Döring Holger, Hicken Allen, Laebens Melis, Lindberg Staffan I., Medzihorsky Juraj, Neundorf Anja, Reuter Ora John, Ruth-Lovell Saskia, Weghorst Keith R., Wiesehomeier Nina, Wright Joseph, Alizada Nazifa, Bederke Paul, Gastaldi Lisa, Grahn Sandra, Hindle Garry, Ilchenko Nina, von Römer Johannes, Wilson Steven, Pemstein Daniel, Seim Brigitte. 2020. “Varieties of Party Identity and Organization (V-Party) Dataset V1” Varieties of Democracy (V-Dem) Project. https://www.v-dem.net/en/data/archive/previous-data/v-dem-dataset/. Date accessed 23 October 2020.
  28. Lundmark Sebastian, Oscarsson Henrik, Weissenbilder Marcus. 2020. “Confidence in an Election Authority and Satisfaction with Democracy: Evidence from a Quasi-Natural Experiment of a Failed Election in Sweden.” Electoral Studies  67:102216. [Google Scholar]
  29. Lust-Okar Ellen, Jamal Amaney Ahmad. 2002. “Rulers and Rules: Reassessing the Influence of Regime Type on Electoral Law Formation.” Comparative Political Studies  35:337–66. [Google Scholar]
  30. Mainwaring Scott.  1993. “Presidentialism, Multipartism, and Democracy: The Difficult Combination.” Comparative Political Studies  26:198–228. [Google Scholar]
  31. McClean Charles.  2021. “The Element of Surprise: Election Timing and Opposition Preparedness.” Comparative Political Studies  54:1984–2018. [Google Scholar]
  32. Michelitch Kristin, Utych Stephen. 2018. “Electoral Cycle Fluctuations in Partisan- ship: Global Evidence from Eighty-Six Countries.” The Journal of Politics  80:412–27. [Google Scholar]
  33. Miller Arthur, Listhaug Ola. 1999. “Political Performance and Institutional Trust.” In Critical Citizens: Global Support for Democratic Government, edited by Norris Pippa, 204–16. Oxford: Oxford University Press. [Google Scholar]
  34. Mishler William, Rose Richard. 2001. “What are the Origins of Political Trust? Testing Institutional and Cultural Theories in Post-Communist Societies.” Comparative Political Studies  34:30–62. [Google Scholar]
  35. Morgan-Jones Edward, Loveless Mathew. 2023. “Early Election Calling and Satisfaction with Democracy.” Government and Opposition  58:598–622. [Google Scholar]
  36. Norris Pippa.  2014. Why Electoral Integrity Matters. New York: Cambridge University Press. [Google Scholar]
  37. Norris Pippa.  2017. “US Elections Rank Last among All Western Democracies.” Electoral Integrity Project. https://www.electoralintegrityproject.com/eip-blog/2017/1/7/its-even-worse-than-the-news-about-north-carolina-american-elections-rank-last-among-all-western-democracies. Date accessed 9 May 2024.
  38. Ong Elvin.  2018. “Electoral manipulation, opposition power, and institutional change: contesting for electoral reform in Singapore, Malaysia, and Cambodia.” Electoral Studies  54:159–71. [Google Scholar]
  39. Palmer Harvey D., Whitten Guy D.  2000. “Government Competence, Economic Performance and Endogenous Election Dates.” Electoral Studies  19:413–26. [Google Scholar]
  40. Pemstein Daniel, Marquardt Kyle L., Tzelgov Eitan, Wang Yi-Ting, Medzihorsky Juraj, Krusell Joshua, Miri Farhad, von RöMer Johannes. 2020. “The V-Dem Measurement Model: Latent Variable Analysis for Cross-National and Cross-Temporal Expert-Coded Data.” V-Dem Working Paper No. 21. 5th Edition: Varieties of Democracy (V-Dem) Project. http://hdl.handle.net/2077/64305. Date accessed 9 May 2024.
  41. Reuter Ora John, Szakonyi David. 2021. “Electoral Manipulation and Regime Support: Survey Evidence from Russia.” World Politics  73:275–314. [Google Scholar]
  42. Roy Jason, Alcantara Christopher. 2012. “The Election Timing Advantage: Empirical Fact or Fiction?” Electoral Studies  31:774–81. [Google Scholar]
  43. Schleiter Petra, Tavits Margit. 2016. “The Electoral Benefits of Opportunistic Election Timing.” The Journal of Politics  78:836–50. [Google Scholar]
  44. Schleiter Petra, Tavits Margit. 2018. “Voter Reactions to Incumbent Opportunism.” The Journal of Politics  80:1183–96. [Google Scholar]
  45. Simpser Alberto.  2013. Why Governments and Parties Manipulate Elections: Theory, Practice, and Implications. New York: Cambridge University Press. [Google Scholar]
  46. Smith Alastair.  2004. Election Timing. Cambridge: Cambridge University Press. [Google Scholar]
  47. Szakonyi David.  2022. “Candidate Filtering: The Strategic Use of Electoral Manipulations in Russia.” British Journal of Political Science  52:649–70. [Google Scholar]
  48. Tsebelis George.  2002. Veto Players: How Political Institutions Work. Princeton: Princeton University Press. [Google Scholar]
  49. Turnbull-Dugarte Stuart J.  2023. “Do opportunistic Snap Elections Affect Political Trust? Evidence from a Natural Experiment.” European Journal of Political Research  62:308–25. [Google Scholar]
  50. Turper Sedef, Aarts Kees. 2017. “Political Trust and Sophistication: Taking Measurement Seriously.” Social Indicators Research  130:415–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. van Erkel Patrick F. A., van der Meer Tom W. G.  2016. “Macroeconomic Performance, Political Trust and the Great Recession: A Multilevel Analysis of the Effects of Within-Country Fluctuations in Macroeconomic Performance on Political Trust in 15 EU Countries, 1999–2011.” European Journal of Political Research  55:177–97. [Google Scholar]
  52. Washida Hidekuni.  2018. Distributive Politics in Malaysia: Maintaining Authoritarian Party Dominance. Oxon and New York: Routledge. [Google Scholar]
  53. Weschle Simon.  2016. “Punishing Personal and Electoral Corruption: Experimental Evidence from India.” Research & Politics  3:2053168016645136. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

nfae022_Supplementary_Data
nfae022_supplementary_data.pdf (864.3KB, pdf)

Data Availability Statement

Replication data and documentation are available at https://doi.org/10.7910/DVN/JT6Q4K.

Articles from Public Opinion Quarterly are provided here courtesy of Oxford University Press

RESOURCES