Election outcomes,legislative representation,and satisfaction with democracy

Abstract

This paper disentangles the relationship between election outcomes and satisfaction with democracy. As the first comparative study to employ a measure of satisfaction immediately before and after elections, we can be unusually confident that any changes we observe are attributable to election outcomes. Following previous work, we affirm that voting for parties that win more votes, more legislative seats, and more cabinet seats boosts satisfaction with democracy. In addition, we demonstrate for the first time that voters are sensitive to deficits in representation; satisfaction with democracy decreases when one’s party’s seat share falls short of its vote share.

Keywords

elections representation satisfaction with democracy voting behavior

In a legislative election, people vote for a party¹, which receives few or many votes and seats, and which does or does not enter government. We may assume that, everything else being constant, people hope that the party they support does well in the election; that is, it receives as many votes and seats as possible and, if it enters the legislature, it gets to be in government rather than in opposition.

This study is about how the electoral performance of the party one supports in an election affects satisfaction with the way democracy works in the country or region. A vast literature has shown that, while elections tend to have a positive overall effect on satisfaction with democracy in the public (e.g. Adam, 2014; Blais and Gélineau, 2007; Weitz-Shapiro and Winters, 2011; see also Ginsberg and Weissberg, 1978), winning an election increases the size of this boost and losing an election can not only decrease any potential boost in satisfaction but also lead to net decreases in satisfaction with democracy (cf. Anderson et al., 2005; Bowler and Donovan, 2002; Conroy-Krutz and Kerr (2015); Henderson, 2008; Listhaug et al., 2009; Singh, 2014; Singh et al., 2011).

We take these findings as our point of departure. Traditionally, the literature has focused on the government dimension: winning an election is construed to mean voting for a party that is part of the government which is formed in the aftermath of an election, and losing means voting for a party that does not enter the government (Anderson et al., 2005: 34). Our goal in this study is to determine if voters react to the vote distribution, to the seat distribution in the legislature, and/or to the seat distribution in the cabinet when they reassess their degree of satisfaction with the way democracy works after an election. As far as we know, this is the only comparative study with a measure of satisfaction with democracy immediately before and after elections, therefore allowing us to determine whether electoral outcomes across countries modify citizens’ evaluations of how democracy works for the first time.

After providing some background and our expectations, we discuss our results. We first reaffirm that voting for parties that win more votes, more seats, and more government representation boosts satisfaction with democracy. We also demonstrate, for the first time, that voters are sensitive to what we call “representation deficits.” Satisfaction with democracy decreases when one’s party’s seat share is not a fair reflection of its vote share, especially for those whose parties won relatively few seats and those with stronger connections to the party for which they voted.

Background and expectations

While satisfaction with democracy is a disputed concept (cf. Booth and Seligson, 2009; Canache et al., 2001; Fuchs, 1993; Linde and Ekman, 2003), it is generally thought of as an expression of approval of regime performance located between diffuse notions of support for democratic principles and specific attitudes toward political actors (e.g. Aarts and Thomassen, 2008; Anderson and Guillory, 1997; Blais and Gélineau, 2007). Weighing in on the debate about the meaning of “satisfaction with democracy,” Anderson (1998: 583) asserts that it “measures system support at a low level of abstraction.” Further, while satisfaction with democracy correlates with a variety of democratic attitudes (e.g. Klingemann, 1999; Kornberg and Clarke, 1994), it is generally thought to be conceptually and empirically distinctive (e.g. Anderson and Guillory, 1997; Clarke and Kornberg, 1992; Fuchs, 1993; Kornberg and Clarke, 1992).

Elections tend to increase satisfaction with democracy (e.g. Adam, 2014; Blais and Gélineau, 2007; Weitz-Shapiro and Winters, 2011; see also Ginsberg and Weissberg, 1978), and voting for a winning party can accentuate this increase (cf. Anderson et al., 2005; Bowler and Donovan, 2002; Henderson, 2008; Listhaug et al., 2009; Singh, 2014; Singh et al., 2011). There are two general mechanisms that are thought to underlie the relationship between winning and satisfaction; winning makes people happier by increasing the likelihood that their preferred policies will be implemented and it provides an emotional “payoff” associated with the experience of victory (see Anderson et al., 2005). Thus, voting for a party that forms the resulting government is especially important; one is more likely to see his or her policy preferences realized when his or her preferred party controls the executive, and the emotional reward associated with winning will be enhanced by such a large political victory.

Though individual-level factors predominate in empirical models of democratic attitudes (Anderson and Singer, 2008), scholars increasingly view such attitudes as held by “individuals who are embedded in political contexts” (Dalton and Anderson, 2011: 3). For example, recent research demonstrates that the relationship between winning and satisfaction is moderated by the characteristics and performance of the party that one supports and the makeup of the resulting government. For example, Curini et al. (2012) show that winning engenders satisfaction, but chiefly for those who are ideologically close to the resulting government, and Singh et al. (2012) find that voting for a party that enters the government leads to a bigger boost in satisfaction than voting for a party that only enters the legislature. Howell and Justwan (2013) find that satisfaction among winners is most elevated where the contest was close, while Singh (2014) finds that winning’s effect on satisfaction is especially pronounced for voters who selected winning parties to which they had strong psychological attachments. These latter two studies suggest that emotional investment in an election can indeed magnify the effect of victory on satisfaction. Meanwhile, losing an election can lead to a decrease in satisfaction, and the effects of losing are moderated by contextual factors (cf. Anderson et al., 2005; Singh, 2014).

Elections can also affect satisfaction by shaping real and perceived representational outcomes. Representative democracy, in theory, should provide both political voice and accountability (Dahl, 1971; Powell, 2000), and perceptions of the extent to which democracy performs its representational function are known to affect behavior and attitudes in the public (Almond and Verba, 1963; Miller, 1974; Thomassen, 2014). Satisfaction with democracy, in particular, has been linked to descriptive, substantive, and perceived representation (cf. Aarts and Thomassen, 2008; Brunell and Buchler, 2009; Ruiz-Rufino, 2013). Here we consider deficits in representation, which we argue condition the effect of winning an election on satisfaction, offsetting the emotional boost thought to underlie this link. If a voter’s party is “punished” by the mechanics of the electoral system, winning fewer seats than would be proportional to its vote share, he or she may become less satisfied with the way in which representational democracy works. Further, a deficit in representation, even for those whose parties performed well, can bring about a sense of defeat, which in turn can engender anger, sullenness, disillusionment (Anderson et al., 2005: 23–26; McCaul et al., 1992; Willson and Kerr, 1999), and lower levels of democratic satisfaction. Essentially, underrepresentation can harm satisfaction with democracy by making one frustrated with how the “system” treats his or her party.

Thus, we have two expectations. The first reflects previous work and anticipates that the degree of satisfaction is affected by one’s party’s performance in terms of votes, legislative seats, and cabinet seats, and that cabinet seats matter most because power is concentrated in government. The second is that voters are affected by the distribution of votes and seats and the extent to which these distributions create representational deficits; any potential boost in satisfaction with democracy associated with the election outcome will attenuate among those who voted for a party whose seat share under-reflects its vote share.

Data, models, and findings

We use the 13 panel election studies that were conducted for the Making Electoral Democracy Work (MEDW) project (see Blais, 2010) between 2011 and 2013 in 10 regions within five countries: Canada, France, Spain, Germany, and Switzerland.² Each of these surveys has two waves, usually with about 1000 persons responding to the pre-election questionnaire in the last two weeks of the campaign and about 800 of them responding to the post-election questionnaire immediately after the election. The surveys were conducted online among recruited panel members, with quotas that guarantee that the samples are representative of the population under study in regard to age, gender, education, and region. The two waves include a question asking respondents how satisfied they are with the way democracy works in their region or country (depending on the type of election) on a 0 to 10 scale. Because we have measures of satisfaction just before and after the election, we can be confident that the changes observed between the two waves are attributable to the election outcome.

The dependent variable is the change in satisfaction between the two waves, which can theoretically range from –10 to +10 (mean = 0.251; SD = 2.232; min = –10; max = 10). Figure 1 illustrates the mean difference in satisfaction with democracy before and after the election in each of our 13 election studies. The effect presented in the bottom of the figure presents the overall change when all observations are combined. Satisfaction with democracy is, on average, 0.23 points higher after the election.³ Nevertheless, as the figure makes obvious, there is a lot a variation across elections. In eight studies out of 13, the data indicate a significant increase in satisfaction with democracy after the election. The largest effect is observed in the Catalonian sample for the Spanish 2012 national election, with an increase of 0.95 points. However, four election studies show no significant change, and the Lucerne 2011 regional election led to an unexpected decrease in satisfaction. The absence of a uniform effect of elections on satisfaction with democracy makes it all the more important to investigate whether electoral outcomes play a role in stimulating or dampening the effect.

Figure 1.

The effect of elections on satisfaction with democracy. Brackets represent 95% confidence intervals.

As per our first expectation, we preliminarily focus on three indicators of a party’s electoral performance: vote support, legislative seats, and seats in government. These three indicators are related to one another in a mechanical manner: vote distribution determines the legislative seat distribution, which in turn influences the cabinet seat distribution. This causal stream guides our approach to our initial set of analyses.

We start with the first and most direct indicator of electoral performance, the proportion of votes obtained by the party supported by the Respondent. We regress change in satisfaction from wave 1 to wave 2 on the percentage the vote won by one’s party (mean = 21.025; SD = 12.425; min = 0.200; max = 44.630) and dummies for each election (with Lucerne’s 2011 regional election excluded as the baseline category).⁴ ^, ⁵ Abstainers are excluded from the analyses because, by definition, they did not support any party in the election. We control for the level of satisfaction in the first wave (mean = 5.948; SD = 2.571; min = 0; max = 10) in each analysis to neutralize floor and ceiling effects. As is standard in the satisfaction with democracy literature, we also control for gender (female = 1; male = 0, mean = 0.493, SD = 0.500), age (mean = 46.895; SD = 14.591; min = 18; max = 111), and education, which is measured on an ordinal scale ranging from less than high school to an advanced university degree⁶ (mean = 5.334; SD = 2.541; min = 0; max = 10).⁷

Model 1 of Table 1 indicates that, all else being equal, satisfaction with democracy after the election compared to immediately before is approximately 0.15 points higher on the –10 to 10 point scale when a party’s vote share increases by 10 percentage points. In line with our first expectation, better vote performance contributes to making people happier, though the impact is rather small. When we include a variable gauging the percentage of legislative seats (mean = 23.223; SD = 16.616; min = 0; max = 53.143) in place of the vote percentage, the results are very similar, as shown in Model 2 of Table 1. All else being equal, satisfaction with democracy after the election compared to immediately before is approximately 0.14 points higher when a party’s seat share increases by 10 percentage points. This also aligns with our first expectation.

Table 1.

Vote performance, seat performance, and satisfaction with democracy.

	Model 1	Model 2	Model 3	Model 4
Vote Percentage	0.015***		–0.034***	–0.026***
Vote Percentage	(0.002)		(0.006)	(0.006)
Seat Percentage		0.014***	0.037***	0.026***
Seat Percentage		(0.001)	(0.004)	(0.005)
Cabinet Seat Percentage				0.004***
Cabinet Seat Percentage				(0.001)
Satisfaction Before Election	–0.46***	–0.464***	–0.464***	–0.463***
Satisfaction Before Election	(0.009)	(0.009)	(0.009)	(0.009)
Female	–0.155***	–0.158**	–0.166***	–0.166***
Female	(0.043)	(0.043)	(0.043)	(0.043)
Age	0.001	0.0005	0.001	0.001
Age	(0.002)	(0.002)	(0.002)	(0.002)
Education	0.023*	0.021*	0.017^†	0.016^†
Education	(0.009)	(0.009)	(0.009)	(0.009)
Dummies for each election	✓	✓	✓	✓
constant	2.701***	2.761***	2.972***	2.968***
	(0.138)	(0.136)	(0.141)	(0.141)
n	8085	8085	8085	8085
R²	0.262	0.266	0.269	0.27
Pr > F	<0.001	<0.001	<0.001	<0.001

Note: Cells contain ordinary least squares regression estimates. Standard errors in parentheses. ^† p < 0.10; *p < 0.05; **p < 0.01; ***p < 0.001 (two-sided).

What happens if we include both votes and seats in the model? Model 3 presents the findings. We need to be cautious here, as there is strong collinearity between votes and seats. However, the results are quite striking, and they make a lot of sense. We see that the positive impact of the percentage of seats becomes much stronger when vote share is held constant. Controlling for the percentage of votes, a 10-point increase in the percentage of seats won by one’s party increases satisfaction with democracy by 0.37 units.

Model 3 also suggests, perhaps counterintuitively, that one becomes more dissatisfied when one’s party gets more votes. We need to keep in mind that this is after controlling for the proportion of seats obtained; a party that gets more votes than another with the same proportion of seats is a party that is unfairly treated by the electoral system. What these results tell us is that supporters of parties that get more votes in general tend to be slightly more satisfied (Model 1) but that they become dissatisfied if and when these votes do not translate into seats in the legislature (Model 3). Individuals become more negative toward democracy when their party’s seat share under-reflects its vote share. This suggests that representation deficits harm satisfaction, which aligns with our second expectation. To our knowledge, this result is the first piece of evidence that deficits in representation have an impact on satisfaction with democracy.

As per our first expectation, as power is concentrated in the executive we would predict that feelings of winning or losing are based first and foremost on whether one’s party is in government or not. Thus, legislative vote and seat shares may become unimportant to satisfaction with democracy if a voter’s party wins a high percentage of cabinet seats and thereby wields substantial political power. In Model 4 of Table 1 we add a variable capturing the percentage of seats in the resulting cabinet occupied by one’s party (mean = 25.516; SD = 34.731; min = 0; max = 100). The more seats one’s party has in government,⁸ the higher the increase in satisfaction. Further, even with the proportion of cabinet seats accounted for, the percentage of legislative seats continues to matter and, with seat shares held constant, more votes reduce satisfaction with democracy. Thus, governmental representation does not appear to wash out the importance of legislative representation, which goes against our expectation that one’s party’s presence in the cabinet matters most for satisfaction.

Another way of looking at the relationship between election outcomes and satisfaction is to distinguish four types of situations: when one’s party wins no seats in the legislature (four percent of voters in the sample), when it wins seats in the legislature but is not in government (47 percent of voters), when it is in government but is in a minority position within the cabinet (28 percent of voters), and when it has a majority of seats in cabinet (21 percent of voters). Model 5 of Table 2 presents the results, which are also illustrated in Figure 2. They show that, while voting for a party that wins no seats in parliament decreases satisfaction with democracy, satisfaction increases with the degree of power bestowed to the party one voted for. But the results also show that the gradation is nonlinear. There is no statistical difference in the predicted change in satisfaction between supporting a party represented in the legislature but not in government and supporting a party that is in government but not in control of the government. It seems that having one’s party in government is nice—provided that the party is in charge. This may be one reason why the winner/loser gap is smaller in proportional representation (PR) systems (Anderson and Guillory, 1997). Being a minority partner in a coalition government may be construed not to be a very positive outcome, especially if this is likely to be a handicap in the following election. The two things that really matter are, on the one hand, whether one’s party is in the legislature or not and, on the other hand, whether one’s party controls the government or not.

Table 2.

Parliamentary representation, government representation, and satisfaction with democracy.

	Model 5
In the Legislature but not in Government	0.654***
In the Legislature but not in Government	(0.115)
Minority of Seats in Government	0.675***
Minority of Seats in Government	(0.122)
Majority of Seats in Government	1.257***
Majority of Seats in Government	(0.12)
Satisfaction Before Election	–0.464***
Satisfaction Before Election	(0.009)
Female	–0.164***
Female	(0.043)
Age	0.0005
Age	(0.002)
Education	0.017*
Education	(0.009)
Dummies for each election	✓
constant	2.411***
constant	(0.168)
n	8085
R²	0.271
Pr > F	<0.001

Note: Cells contain ordinary least squares regression estimates. Standard errors in parentheses. ^† p < 0.10; *p < 0.05; **p < 0.01; ***p < 0.001 (two-sided).

Figure 2.

Parliamentary representation, government representation, and satisfaction with democracy. Brackets represent 95% confidence intervals.

Following from our second expectation—that representation deficits harm satisfaction with democracy—another way to illustrate the relationship between the electoral performance of one’s party and satisfaction, while also easing collinearity concerns, is to create a new variable that indicates the extent to which the party a voter supported is underrepresented as a result of the translation of votes into legislative seats. For each party, representation deficit indicates the percentage of votes it won in the election minus the percentage of seats it was awarded in the legislature (mean = –2.197; SD = 6.178; min = –19.177; max = 13.253). Positive values mean underrepresentation (a higher proportion of votes than seats) and negative values overrepresentation (a higher proportion of seats than votes). For example, a value of 10 on the representation deficit variable indicates that, for a given party, the proportion of seats in the legislative chamber is 10 percentage points lower than the proportion of votes.

As apparent from the results of Model 6 of Table 3, voting for a party that turns out to be underrepresented by 10 percentage points decreases satisfaction with democracy by 0.41 points. It is interesting to note that although 55 percent (53/96) of the parties we analyze are underrepresented (i.e. representation deficit < 0), the proportion of voters who voted for a party that is underrepresented is much lower, at 32 percent. In Model 7, we add in the variable capturing the percentage of seats occupied by one’s party in the cabinet that formed after the election. Again, the results indicate that the more seats one’s party has in government, the higher the increase in satisfaction. However, even with the proportion of cabinet seats accounted for, representational deficits continue to reduce satisfaction with democracy.

Table 3.

Representation deficits and satisfaction with democracy.

	Model 6	Model 7	Model 8	Model 9	Model 10	Model 11
Representation Deficit	–0.041***	–0.026***	–0.020***	–0.031***	–0.039***	–0.059***
Representation Deficit	(0.004)	(0.005)	(0.006)	(0.006)	(0.010)	(0.011)
Cabinet Seat Percentage	0.004***	0.003***	0.004***	0.00005	0.007*
Cabinet Seat Percentage	(0.001)	(0.001)	(0.001)	(0.001)	(0.001)
Satisfaction Before Election	–0.463***	–0.463***	–0.462***	–0.463***	–0.523***	–0.464***
Satisfaction Before Election	(0.009)	(0.009)	(0.009)	(0.011)	(0.009)	(0.009)
Female	–0.168***	–0.165***	–0.174***	–0.160**	–0.116**	–0.168***
Female	(0.043)	(0.043)	(0.043)	(0.057)	(0.042)	(0.043)
Age	0.001	0.001	0.0005	0.005	–0.003^†	0.0005
Age	(0.002)	(0.002)	(0.002)	(0.002)	(0.001)	(0.002)
Education	0.016^†	0.016^†	0.017*	0.021^†	0.013	0.016^†
Education	(0.009)	(0.009)	(0.009)	(0.012)	(0.009)	(0.009)
Campaign Activity			–0.048
Campaign Activity			(0.036)
Considered Other Party				0.05
Considered Other Party				(0.063)
Outcome Reflects Views					0.255***
Outcome Reflects Views					(0.010)
Seat Percentage						–0.001
Seat Percentage						(0.003)
Representation Deficit × Campaign Activity			–0.015**
Representation Deficit × Campaign Activity			(0.005)
Representation Deficit × Considered Other Party				0.018*
Representation Deficit × Considered Other Party				(0.008)
Representation Deficit × Outcome Reflects Views					0.002^†
Representation Deficit × Outcome Reflects Views					(0.001)
Representation Deficit × Seat Percentage						0.001***
Representation Deficit × Seat Percentage						(0.0004)
Dummies for each election	✓	✓	✓	✓	✓	✓
constant	3.025***	2.965***	2.989***	2.871***	1.941***	2.971***
	(0.109)	(0.136)	(0.137)	(0.171)	(0.139)	(0.140)
n	8085	8085	7929	5375	7881	8085
R²	0.269	0.27	0.272	0.269	0.327	0.271
Pr > F	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001

Note: Cells contain ordinary least squares regression estimates. Standard errors in parentheses. ^† p < 0.10; *p < 0.05; **p < 0.01; ***p < 0.001 (two-sided). Because the question about considering another party was not asked in Switzerland, the regional and national elections in Lucerne and Zurich are not included in Model 9. The baseline category for the election dummies in Model 9 is thus the 2012 Quebec regional election, rather than Lucerne’s 2011 regional election, as in the other models.

We plot the results of Model 7 in Figure 3. All else being equal, satisfaction with democracy rises after the election for people who are overrepresented—those with negative representation deficits. However, for individuals who are underrepresented—those with positive representation deficits—the post-election satisfaction boost is absent. In fact, among those with the largest representation deficits, satisfaction tends to decrease after the election. These results indicate that, irrespective of how much power the party they support exercises in government, voters remain concerned about how well their views are represented in the legislature.

Figure 3.

Deficits in representation and satisfaction with democracy. Shaded areas represent 95% confidence intervals.

We probe this further by assessing who is most affected by representation deficits. If underrepresentation harms satisfaction with democracy by making one frustrated by how the “system” treats one’s party, then individuals who are more invested in their party should be most affected by representational deficits, while those with a weaker attachment should be less affected. To test this, we interact representational deficit with the amount of campaigning one did for his or her party and whether or not one considered voting for a different party. The existing covariates from Model 7, including the percentage of cabinet seats, remain in the models.

A series of questions in the MEDW survey asked individuals whether they campaigned for their party by talking to others, attending meetings, or putting up a yard sign. Individuals who participated in each activity were assigned a 1, and others were assigned a 0. From these indicators we create an additive scale of campaign activity (mean = 0.325; SD = 0.616; min = 0; max = 3). As shown in Model 8 of Table 3, the coefficient on the interaction between campaign activity and representation deficit is negative and statistically significant, which indicates that individuals who are more devoted to their party are most affected by representational deficits. For those (the majority) who did no campaigning for their party, representation deficits still suppress satisfaction, but do so relatively weakly. This relationship is further illustrated in the upper-left quadrant of Figure 4, in which we plot the relationship between representation deficits and the predicted change in satisfaction among those who did not campaign for their party (“No Campaign Activity”) and those who tried to influence others, attended meetings, and displayed a yard sign (“High Campaign Activity”).

Figure 4.

Deficits in representation and satisfaction with democracy, conditional relationships. Shaded areas represent 95% confidence intervals.

To assess whether individuals contemplated choosing a different party, we use a question asking whether they “considered the possibility of voting for another party.” Individuals who reported doing so are coded 1, and others are coded 0 (mean = 0.407; SD = 0.491; min = 0; max = 1). As shown in Model 9 of Table 3, the coefficient on the interaction between this variable and the representation deficit variable is positive and statistically significant. This indicates that the negative effect of representation deficits on satisfaction with democracy is less pronounced among those who considered abandoning their party. Equivalently, representation deficits matter more for those who are more “loyal” to their party. This is further illustrated in the upper-right quadrant of Figure 4, in which it is clear that the negative relationship between representation deficits and the post-election change in satisfaction with democracy is steeper for those who did not consider voting for another party.

We can also assess the extent to which underrepresentation reduces satisfaction at the individual level by interacting the representational deficit variable with voters’ opinions about the election outcome. To do so, we use a question that inquires as to whether individuals believe that the outcome reflected the public’s views. The question wording is: “How accurately do you think the outcome of the election reflects voters’ views?” Individuals were asked to provide a number between 0 and 10, with 0 indicating “not at all accurately” and 10 indicating “very accurately” (mean = 6.441; SD = 2.392; min = 0; max = 10).⁹

The results of Model 10 of Table 3 indicate that, for those who believe that the election was fair, representational deficits do little to alter satisfaction; for these individuals, satisfaction tends to increase sharply after the election irrespective of their party’s representational outcome. Alternatively, for those who see the election outcome as unreflective of the electorate’s views, representational deficits amplify their tendency to become less satisfied with democracy after the election. This is apparent in the bottom-left panel of Figure 4; those who see the outcome as fair become more satisfied with democracy, irrespective of representational deficits, while those who see the outcome as unfair become less satisfied with democracy after the election, especially if their party has a steep representational deficit.

We showed above that the percentage of seats in the legislature is positively related to changes in voters’ satisfaction levels. Potentially, for individuals whose party performed well in terms of legislative seats, representational deficits will not matter much; satisfaction will be especially suppressed among voters whose party not only performed poorly but was also punished by the distribution of votes into seats.

To test this, we interact the variable gauging a voter’s party’s percentage of seats in the legislature with the representational deficit variable. As indicated by the positive and significant coefficient on the interaction term in Model 11 of Table 3, representation deficits harm satisfaction most sharply for individuals whose parties performed poorly in terms of legislative seats. This is clear in the bottom-right panel of Figure 4, which shows that, for example, for an individual who voted for a party that won 15 percent of the vote but only five percent of the seats in the legislature (a representation deficit of 10) satisfaction is expected to decrease by about 0.20 points. Alternatively, for individuals whose parties received a large proportion of seats in the legislature, representational deficits will have less of an impact on satisfaction.

The bottom-right panel of Figure 4 also indicates that, for individuals who supported a small party that was overrepresented, and thus had a negative representation deficit, satisfaction increases sharply after the election. Keeping with the scenario depicted in the figure, a voter whose party only received five percent of the vote but somehow managed to win 20 percent of the seats (a representational deficit of –15) would experience a marked increase in satisfaction. However, while such an outcome exists in theory it is unlikely to exist in reality and we thus cannot read too much into this result. In the elections covered here, for example, the lowest representation deficit among “small” parties receiving less than 10 percent of the vote is –1.62.

As shown in the online Appendix, we also discovered some weak and null results when estimating interactive models. First, macro-level characteristics, including whether the election was conducted in a federal or unitary system and whether the election was conducted with a majoritarian or proportional electoral system, do little to condition the relationship between representation deficits and satisfaction, while the impact of representation deficits is slightly more pronounced in national than in regional elections. Second, a voter’s opinion about the importance of the wellbeing of others does not condition the relationship between representation deficits and satisfaction; there is no evidence that some “virtuous” voters become less satisfied with democracy when their party is overrepresented. Third, political interest does not condition the relationship between representation deficits and satisfaction; voters’ satisfaction levels are equally responsive to representational outcomes, irrespective of their level of interest in politics.

Conclusion

The question that we address in this study is simple and straightforward: How do people react to the “performance” of the party they voted for in an election? More specifically we wished to determine whether people became more satisfied with the way democracy works if their party performed well and more dissatisfied if it performed badly, both in absolute terms and in relation to the way in which its votes were translated into seats. This study belongs to a stream of research that has documented the impact of “winning” or “losing” on satisfaction with democracy.

We have made a contribution on three fronts. First, and perhaps most importantly, this is the first study, we believe, to test and compare the effects of three aspects of electoral performance: how many votes a party gets, how many seats it obtains in the legislature, and how many seats it has in government. Our starting hypothesis was that each aspect would play a role, but that being in government would matter the most. We then put forth that individuals will become less satisfied with democracy when their party’s seat share is under-representative of its vote share. We find that people are more satisfied when their party has more votes and more seats in the legislature and in government, but that both types of representation matter. Further, we find that votes and seats do not matter in isolation; whether one’s party is fairly represented in the legislature, given its seat share, also counts.

Second, we have demonstrated that the extent to which election outcomes affect satisfaction with democracy is conditional on individual- and party-level characteristics. Those who are more invested in “their” party are more disheartened when their party’s seat share does not fairly reflect its vote share. Further, for individuals who perceive the election as fair, and for those whose parties garner a high proportion of legislative seats, electoral outcomes do relatively little to condition satisfaction with democracy.

Third, because we have used a panel survey with the first wave taking place immediately before the election and the second wave immediately after, we are able to tap the specific impact of the electoral outcome, as all the other individual level determinants of satisfaction with democracy are taken into account by controlling for pre-election satisfaction. This approach has been utilized in a few previous studies (e.g. Banducci and Karp, 2003; Blais and Gélineau, 2007; Singh et al., 2012) but, as far as we can tell, this is the first cross-national research with pre- and post-election measures of satisfaction with democracy. Applied to our 13 election studies, pre and post attitude comparisons revealed great variation in terms of how much the holding of an election impacts the mean level of satisfaction with democracy.

A number of interpretations and predictions can be drawn from our results. One of our findings is that satisfaction with democracy is sensitive to representation biases introduced by the electoral system. Indeed, satisfaction with democracy decreases when a voter supports a party that turns out to be underrepresented in the legislature, as compared to the proportion of votes obtained. This suggests that voters are more satisfied—or at least less dissatisfied—when seats are proportional to votes. Yet, another of our findings challenges this interpretation. Our analyses show that voters do not show the same dissatisfaction when representation biases lean in their favor—quite the contrary. Our models suggest that, if two parties were to obtain the same share of votes but a different share of seats, those voters who supported the advantaged party would report a greater increase in satisfaction with democracy. In short, voters’ reactions to representation biases depend on whether their party is advantaged or disadvantaged.

These results also help to understand why satisfaction with democracy is not higher, overall, in more proportional systems (Blais and Loewen, 2007: 51, Figure 3.8). Proportional representation typically leads to the formation of coalitions, and as a consequence there are usually more “winners’” in a PR election; that is, the governing parties have more supporters in proportional electoral systems than in majoritarian electoral systems. But it seems that the supporters of small partners in coalition governments are not particularly happy, which may account for the fact that satisfaction with democracy is somewhat lower after elections that produce coalition governments (Blais and Loewen, 2007: 53, Table 3.4).

To summarize, our results show that, taken individually, each of our three measures of electoral performance—vote shares, seat shares, and cabinet seat shares—is positively correlated with satisfaction with democracy. However, more importantly, our research reveals that the outcome of an election affects satisfaction in a more complex fashion. Our analyses present persuasive evidence that voters are responsive to biases in the translation of votes into seats. Future research might test whether these effects persist in the months following the election, as the effect of elections on satisfaction with democracy is known to taper rather quickly (e.g. Adam, 2014). In addition, future work may examine whether the outcomes of previous elections vis-à-vis a voter’s preferences shape the effect of the most recent election on his or her level of satisfaction with democracy—a dynamic that Chang et al. (2014) have studied looking across two elections. Finally, future work would do well to study elections for multiple levels of government to examine how satisfaction with democracy changes for those whose parties perform well at one level but poorly at the other(s).

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Supplemental material

The online appendices are available at .

Notes

References

Aarts

Thomassen

(2008) Satisfaction with democracy: Do institutions matter? Electoral Studies 27(1): 5–18.

Adam

(2014) Do elections bring optimism? Electoral Studies 33(1): 137–143.

Almond

Verba

(1963) The Civic Culture: Political Attitudes and Democracy in Five Nations. Newbury Park, CA: Sage Publications.

Anderson

(1998) Parties, party systems, and satisfaction with democratic performance in the new Europe. Political Studies 46(3): 572–588.

Anderson

Guillory

(1997) Political institutions and satisfaction with democracy: A cross-national analysis of consensus and majoritarian systems. American Political Science Review 91(1): 66–81.

Anderson

Singer

(2008) The sensitive left and the impervious right: Multilevel models and the politics of inequality, ideology, and legitimacy in Europe. Comparative Political Studies 41(4): 564–599.

Anderson

Blais

Bowler

. (2005) Losers’ Consent: Elections and Democratic Legitimacy. New York: Oxford University Press.

Arceneaux

Nickerson

(2009) Modeling certainty with clustered data: A comparison of methods. Political Analysis 17(2): 177–190.

Banducci

Karp

(2003) How elections change the way citizens view the political system: Campaigns, media effects and electoral outcomes in comparative perspective. British Journal of Political Science 33(3): 443–467.

10.

Blais

(2010) Making electoral democracy work. Electoral Studies 29(1): 169–170.

11.

Blais

Gélineau

(2007) Winning, losing, and satisfaction with democracy. Political Studies 55(2): 425–441.

12.

Blais

Loewen

(2007) Electoral systems and evaluations of democracy. In: Cross

(ed.) Democratic Reform in New Brunswick. Toronto: Canadian Scholars Press, pp. 39–57.

13.

Blais

Gidengil

Dobrzynska

. (2003) Does the local candidate matter? Candidate effects in the canadian election of 2000. Canadian Journal of Political Science 36(3): 657–664.

14.

Booth

Seligson

(2009) The Legitimacy Puzzle in Latin America: Public Support and Democracy in Eight Nations. New York: Cambridge University Press.

15.

Bowler

Donovan

(2002) Democracy, institutions and attitudes about citizen influence on government. British Journal of Political Science 32(2): 371–390.

16.

Brunell

Buchler

(2009) Ideological representation and competitive congressional elections. Electoral Studies 28(3): 448–457.

17.

Cameron

Gelbach

Miller

(2008) Bootstrap-based improvements for inference with clustered errors. The Review of Economics and Statistics 90(3): 414–427.

18.

Canache

Mondak

Seligson

(2001) Meaning and measurement in cross-national research on satisfaction with democracy. Public Opinion Quarterly 65(4): 506–528.

19.

Chang

Chu

Y-h

W-c

(2014) Consenting to lose or expecting to win? Inter-temporal changes in voters’ winner-loser status and satisfaction with democracy. In: Thomassen

(ed.) Elections and Representative Democracy: Representation and Accountability. Oxford: Oxford University Press, pp. 232–253.

20.

Clarke

Kornberg

(1992) Support for the Canadian federal Progressive Conservative Party since 1988: The impact of economic evaluations and economic issues. Canadian Journal of Political Science 25(1): 29–53.

21.

Conroy-Krutz

Jeffrey

Nicholas

Kerr

(2015) Dynamics of democratic satisfaction in transitional settings: Evidence from a panel study in Uganda. Political Research Quarterly. Epub online ahead of print June 16, 2015. DOI: 10.1177/1065912915589763.

22.

Curini

Jou

Memoli

(2012) Satisfaction with democracy and the winner-loser debate: The role of policy preferences and past experience. British Journal of Political Science 42(2): 241–261.

23.

Dahl

(1971) Polyarchy: Participation and Opposition. New Haven, CT: Yale University Press.

24.

Dalton

Anderson

(2011) Citizens, context, and choice. In: Dalton

Anderson

(eds) Citizens, Context, and Choice: How Context Shapes Citizens’ Electoral Choices. Oxford: Oxford University Press, pp. 3–30.

25.

Donner

Klar

(2000) Design and Analysis of Cluster Randomization Trials in Health Research. New York: Arnold Publishers.

26.

Fuchs

(1993) Trends of political support in the Federal Republic of Germany. In: Berg-Schlosser

Rytlewski

(eds) Political Culture in Germany. London: MacMillan, pp. 232–268.

27.

Ginsberg

Weissberg

(1978) Elections and the mobilization of popular support. American Journal of Political Science 22(1): 31–55.

28.

Green

Vavreck

(2008) Analysis of cluster-randomized experiments: A comparison of alternative estimation approaches. Political Analysis 16(2): 138–152.

29.

Henderson

(2008) Satisfaction with democracy: The impact of winning and losing in westminster systems. Journal of Elections, Public Opinion and Parties 18(1): 3–26.

30.

Howell

Justwan

(2013) Nail-biters and no-contests: The effect of electoral margins on satisfaction with democracy in winners and losers. Electoral Studies 32(2): 334–343.

31.

Klingemann

H-D

(1999) Mapping political support in the 1990s: A global analysis. In: Norris

(ed.) Critical Citizens: Global Support for Democratic Government. Oxford: Oxford University Press, pp. 31–56.

32.

Kornberg

Clarke

(1992) Citizens and Community: Political Support in a Representative Democracy. New York: Cambridge University Press.

33.

Kornberg

Clarke

(1994) Beliefs about democracy and satisfaction with democratic government: The Canadian case. Political Research Quarterly 47(3): 537–563.

34.

Linde

Ekman

(2003) Satisfaction with democracy: A note on a frequently used indicator in comparative politics. European Journal of Political Research 42(3): 391–408.

35.

Listhaug

Aardal

Ellis

(2009) Institutional variation and political support: An analysis of cses data from 29 countries. In: Klingemann

(ed.) The Comparative Study of Electoral Systems. Oxford: Oxford University Press, pp. 85–108.

36.

McCaul

Gladue

Joppa

(1992) Winning, losing, mood, and testosterone. Hormones and Behavior 26(4): 486–504.

37.

Marsh

(2007) Candidates or parties? Objects of electoral choice in Ireland. Party Politics 13(4): 500–527.

38.

Miller

(1974) Political issues and trust in government: 1964–1970. American Political Science Review 68(3): 951–972.

39.

Powell

(2000) Elections as Instruments of Democracy: Majoritarian and Proportional Visions. New Haven, CT: Yale University Press.

40.

Ruiz-Rufino

(2013) Satisfaction with democracy in multi-ethnic countries: The effect of representative political institutions on ethnic minorities. Political Studies 61(1): 101–118.

41.

Singh

(2014) Not all election winners are equal: Satisfaction with democracy and the nature of the vote. European Journal of Political Research 53(2): 308–327.

42.

Singh

Karakoç

Blais

(2012) Differentiating winners: How elections affect satisfaction with democracy. Electoral Studies 31(1): 201–211.

43.

Singh

Lago

Blais

(2011) Winning and competitiveness as determinants of political support. Social Science Quarterly 92(3): 695–709.

44.

Thomassen

(2014) Elections and Democracy: Representation and Accountability. Oxford: Oxford University Press.

45.

Weitz-Shapiro

Winters

(2011) The link between voting and life satisfaction in Latin America. Latin American Politics and Society 53(4): 101–126.

46.

Willson

Kerr

(1999) Affective responses to success and failure: A study of winning and losing in competitive rugby. Personality and Individual Differences 27(1): 85–99.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.42 MB