How moderates and extremists find happiness: Ideological orientation,citizen–government proximity,and life satisfaction

Abstract

While the topic of life satisfaction and its determinants has drawn increasing attention among political scientists, most studies have focused mainly on macro-level variables, and often overlooked the role of individuals’ attitudes vis-à-vis their governments. The present article attempts to fill this gap by examining whether citizens’ left–right self-placement and ideological distance from their governments exert an independent effect on life satisfaction. Utilizing a dataset spanning a quarter century and containing nearly 70,000 respondents, we demonstrate a curvilinear relationship between ideological orientations and happiness, with self-identified radicals on both ends of the spectrum happier than moderate citizens. Moreover, we show that while propinquity between self-position and government position contributes to happiness, this effect is highly mediated by individual locations along the left–right spectrum: centrists report higher levels of happiness the closer they are to their government, while the opposite is true for radicals. The normative implication of our findings is that moderate governments may present a comparative advantage in enhancing the overall level of happiness of their citizens.

Keywords

life satisfaction political ideology ideological congruence happiness left right

While the topic of life satisfaction (often used interchangeably with ‘happiness’)¹ and its determinants has long been studied by economists (Bjørnskov et al., 2008; Blanchflower and Oswald, 2004; Di Tella et al., 2001; Easterlin, 1995, 2001; Frey and Stutzer, 2000, 2002; Helliwell, 2003; Ovaska and Takashima, 2006), it has also drawn increasing attention in the political science literature. For example, Inglehart (1990: 45) listed it as an intrinsic part of political culture that reduces the potential for revolutionary change. From the perspective of ordinary citizens, on the other hand, the most direct and tangible means of judging the effectiveness of a given policy is whether it improves their lives. This prompts Layard (2006: C32) to argue that ‘the prime purpose of social science should be to discover what helps and hinders happiness,’ while Tavits (2008: 1607) called on the political science profession to pay greater heed to ‘the most fundamental goal of every citizen – to be happy.’

Several recent studies have examined the effect of political factors on life satisfaction, focusing mainly on macro-level determinants such as institutional conditions (Bjørnskov et al., 2010), quality of governance (Helliwell and Huang, 2008; Ott, 2010), and policy outputs (Pacek and Radcliff, 2008; Whiteley et al., 2010). In this respect, Álvarez-Díaz et al. concluded that ‘politics emphatically does matter for … identifying the conditions that make human life rewarding’ (2010: 902).

However, concentrating on aggregate-level variables neglects the potential impact of individual attitudes vis-à-vis governments on life satisfaction. The few studies that have explicitly investigated this topic mostly focus on either the effect of citizens’ self-positioning, to show that right-leaning citizens report higher levels of happiness (Tavits, 2008; Taylor et al., 2006), or the impact of governments’ ideological stances (Álvarez-Díaz et al., 2010; Di Tella and MacCulloch, 2005).

In this article, we combine these two approaches by investigating the effect of both individual self-placements on the left–right spectrum and their proximity to the government’s location along the same scale. Furthermore, we depart from the existing literature by observing a contrast between moderate and radical citizens and its consequences for life satisfaction. As discussed below, this distinction turns out to be highly relevant both theoretically and empirically. By analyzing a large dataset encompassing 70,000 cases from 40 countries that span a quarter century, we find that (1) instead of a linear relationship often assumed in the literature, individual ideological orientations have a curvilinear effect on happiness, with higher life satisfaction scores at both extremes, and (2) congruence between self-position and government position on the left–right spectrum enhances happiness, but this effect only applies to moderates. These two findings suggest that centrist governments may present a comparative advantage in enhancing the overall level of happiness of their citizens.

This article is organized as follows. The next section reviews the literature on political factors affecting happiness, including a discussion on the impact of individual and government ideology on life satisfaction that leads to our main hypotheses. Section three introduces the dataset and describes each variable. Next, we present the empirical results, highlighting distinct proclivities between centrist and radical respondents. The last section summarizes our findings and discusses their implications.

Ideological orientation and happiness

Among studies that discuss how political factors affect life satisfaction, most works concentrate mainly on macro-level determinants. Several scholars have compared how regime and institutional differences across countries or subnational regions affect citizens’ level of happiness, while others have emphasized institutional quality as a key contributor (Helliwell, 2006; Ovaska and Takashima, 2006). For instance, Helliwell and Huang (2008) point out that honest and efficient delivery of public services enhances happiness in poor countries, while life satisfaction in rich countries is more influenced by the conduct of political and electoral institutions. Bok’s study of American states (2010) also stressed the positive role of institutional measures related to state capacity to boost happiness. In addition, Frey and Stutzer (2002) found that direct democracy raises life satisfaction due to citizens having greater ‘procedural utility,’ namely, a greater say in monitoring and controlling policy outputs.

Following an analogous macro-level approach, scholars have, in addition, investigated whether democracy itself influences life satisfaction. While democratic institutions may not constitute the main determinant of human happiness (Inglehart and Klingemann, 2000: 180), a number of studies have confirmed a link between democracy or democratic values and happiness (Graham and Pettinato, 2001). For example, Inglehart (1990: 41) observed ‘a remarkably consistent tendency for high levels of life satisfaction to go together with the persistence of democratic institutions over relatively long periods of time’, and Dorn et al. (2005) reported that a country’s Freedom House or Polity IV index score substantially affects its aggregate life satisfaction. In contrast, Bjørnskov et al. (2008) produced the opposite result using both Polity IV scores and indicators of good governance such as freedom of the press and low corruption.²

Most extant studies have overlooked the (possible) impact played by micro-level ideological factors, leaving an important vacuum. While positive appraisal of policy procedures can increase life satisfaction (Helliwell and Huang, 2008; Whiteley et al., 2010), it is worth asking whether the relationship between policies and happiness is based solely on how much citizens benefit from them, or also derives from their policy preferences. In other words, does ideological orientation exert an independent influence on life satisfaction? Napier and Jost (2008), for example, showed that right-leaning citizens in the USA are happier than their leftist compatriots (see also Taylor et al., 2006) and Tavits (2008) reported similar findings in a cross-national study.

Three main explanations are advanced in this regard. First, ideological position may capture the effect of religion. Right-leaning respondents tend to be more religious, and religiosity has been found to have a positive affect on happiness (Frey and Stutzer, 2002). Second, there may be an income effect, given positive correlations between ideology and income and between income and happiness. Finally, Napier and Jost (2008) propose a ‘system justification theory perspective’: someone who is happy with the existing social order would prefer an ideology that stresses its preservation (that is, conservatism). Conversely, left-wing ideology, often associated with progressivism, is likely to attract people seeking to change the status quo. Furthermore, left- and right-leaning citizens may react differently to the same set of objective circumstances. For example, Malahy et al. (2009) found that in the face of economic inequality, life satisfaction remained unchanged for conservatives, but declined for progressives, suggesting that the former may be happier because they are more ready to accept existing conditions.

Based on the findings reported in the aforementioned studies, we formulate our first hypothesis:

H1 (ideological position hypothesis): Citizens with right-leaning ideological orientations report higher levels of happiness than those who place themselves on the left.

Another strand of the literature highlights the contrast between extreme and centrist ideological positions rather than left and right. As long as half a century ago, several scholars noted similarities shared between right- and left-wing extremists, such as authoritarian attitudes, dogmatism, and radical methods of political engagement (Eysenck, 1954; Rokeach, 1960). McClosky and Chong pointed out that extremists on both ideological ends are characterized by resentment toward mainstream politicians and policies and an attraction to totalitarian measures (1985: 343). Similarly, Greenberg and Jonas (2003) concluded that, in addition to left versus right, there exists a separate ideological dimension pitting those who adhere rigidly to their views (radicals) against those who are more flexible (centrists), while other works have demonstrated that extremists are more cognitively sophisticated than centrists because they have greater need to justify their views (Kemmelmeier, 2008; Sidanius, 1985). It has also been noted how citizens with more extreme attitudes, having no doubts about the infallibility of their opinions, are more predisposed to emotion-driven expressions of their views (Claassen, 2007: 373). This is relevant given the (largely positive) influence of individuals’ emotional experiences on assessing life satisfaction (Suh et al., 1998).

Taken together, these findings suggest a relationship between ideology and happiness that goes beyond the simple linear linkage stated in H1, and lead to two opposite, albeit equally plausible, hypotheses as yet untested in the literature:

H2a (ideological extremism hypothesis): Citizens holding ideologically extreme positions report lower levels of happiness than moderates due to their feelings of persecution and alienation.

H2b (ideological extremism hypothesis): Citizens holding ideologically extreme positions report higher levels of happiness than moderates due to a stronger belief in the veracity of their views.

Insofar as ideology motivates governments’ choice of programs, where a government is located along the left–right spectrum, and consequently what types of policies it seeks to implement, can influence citizens’ happiness. For example, from a macro-level perspective, Álvarez-Díaz et al. (2010) showed that greater welfare spending and stricter regulatory policies lead to higher life satisfaction, and both Radcliff (2001) and Pacek and Radcliff (2008) concluded that government intervention in the economy enhances satisfaction by bolstering a social safety net. These results imply that having leftist governments makes citizens happier. However, other studies offer a contradictory view: Veenhoven (2000: 91) reported ‘no link between the size of the welfare state and the level of well-being within it,’ Tavits (2008) found no significant effect for social spending when controlling for other factors, and Bjørnskov et al. (2007) pointed out that excessive government consumption may decrease overall life satisfaction. Thus, whether and how government ideology affects happiness remains open to debate.

That said, one may surmise that what matters for life satisfaction is not simply the ideological profile of governments per se, but rather, from a micro-level perspective, the relationship between governments and citizens. Insomuch as a government’s economic and social policies are predicated on its ideological stance, outputs may depend on which party is in power, ceteris paribus, and citizens may be happier with a government pursuing a program closer to their own views. This would be coherent with the spatial theory of voting (Adams et al., 2005): if a voter derives greater utility from programs implemented by parties which are ideologically closer to her own ideal point, citizens may also make similar assessments when evaluating the impact of government programs on their life satisfaction.³ This leads to the following hypothesis:

H3 (ideological proximity hypothesis): The ideological proximity between citizens and their government is positively related to the former’s level of happiness.

Whether and how ideological proximity to the government affects citizens’ life satisfaction has been a largely neglected topic. Among the few exceptions, Di Tella and MacCulloch found that respondents are indeed substantially ‘happier when the party in power has a similar ideological position to themselves’ (2005: 378). However, the impact of ideological proximity on life satisfaction was only tested indirectly through an interaction between citizens’ self-placement and government positions. Moreover, respondents’ ideological positions were collapsed into two broad categories (left and right), which precludes more sophisticated analysis on distinctions within each group and similarities across groups. Dreher and Öhler (2011) followed a similar method, classifying governments as leftist, moderate, or rightist, but did not find the same significant relationship between citizens’ and governments’ ideological positions as Di Tella and MacCulloch (2005). Finally, the study by Taylor et al. (2006) yielded the surprising finding that citizens’ left–right orientation exerts greater influence on their life satisfaction when they live under a government from the opposite ideological camp.⁴

Part of the explanation for this unexpected result may lie in the distinction we made between citizens with moderate and radical ideological orientations. According to H2b, greater life satisfaction among radicals derives from a stronger belief in the correctness of their views. However, if their (hypothesized) higher level of happiness also derives from perceiving themselves as part of a minority (that is, an ideological ‘purist’ preference or attitude), then anything that threatens this status (for example, a government closer to their own radical position) should make them less satisfied.⁵ This line of reasoning assumes that moderates base their subjective life satisfaction on utilitarian considerations, with an emphasis on concrete gains or losses resulting from the implementation of certain policies, in contrast to extremists, who derive happiness more on expressive grounds (Brennan and Lomasky, 1993) and for whom abstract self-justifications matter more for life satisfaction.

Alternatively, it is possible that radicals could be the ones most easily disappointed by contradictions between the stated goals of a government theoretically close to their own views and its actual (policy) performance. This could happen given that a (relatively) radical cabinet, compared with a moderate one, faces considerably greater external and internal difficulties when trying to pursue its objective of altering the status quo, making citizens closer to this cabinet particularly frustrated. This scenario involving expectations of large-scale change and subsequent disillusionment (see Stimson, 1976) is less likely among centrist voters, precisely because they are usually located closer to the status quo. The fact that niche (that is, radical) parties on both sides of the ideological spectrum tend to lose votes after participating in government (Buelens and Hino, 2008; Deschouwer, 2008; McDonnell and Newell, 2011) offers important, albeit indirect, evidence for both of the arguments posited above.

Whether based on expressive rather than Downsian incentives or due to disappointed policy expectations, the preceding paragraphs raise considerations not covered by H2b and H3. Consequently, we introduce the following two hypotheses that account for a conditional relationship between ideological proximity and citizens’ self-placement, on one hand, and life satisfaction, on the other:

H4 (conditional ideological proximity hypothesis): The ideological proximity between citizens and their government is positively related to the former’s level of happiness as long as citizens are moderate.

H5 (conditional ideological extremism hypothesis): Citizens holding ideologically extreme positions report higher levels of happiness than moderates due to stronger belief in the veracity of their views. This gap increases as the ideological distance separating radical citizens from the cabinet increases.

As we will discuss below, H4 and H5 can be quite relevant for the overall impact on life satisfaction given certain government ideological positions.

Data and measurement

To test our hypotheses, we use individual-level measures of life satisfaction and ideological orientation obtained from the World Values Survey (WVS), which employs the same battery of questions across countries and time with respect to our main variables of interest. Five waves of the WVS are currently available (roughly, 1980, 1990, 1995, 2000, and 2005), containing about 1000–1500 respondents in each participating country. Our sample consists of countries rated ‘free’ by Freedom House at the time of the survey, since fair and competitive elections are a prerequisite for analyzing how the relationship between the ideological positions of individual citizens and their governments affects happiness.⁶ This leaves us with data from 40 countries, covering both established and new democracies (see Appendix Table A1), with an average of two surveys per country. The total number of observations is around 70,000, the largest number we could analyze without missing key variables.

Our dependent variable is respondents’ level of satisfaction with life (SWL). In each survey, respondents are asked the following question: ‘All things considered, how satisfied are you with your life as a whole these days?’ Response categories range from ‘dissatisfied’ (which is assigned a value of 1) to ‘satisfied’ (which is assigned a value of 10). The mean value in our sample is around 7.0, with a standard deviation of 2.1, which suggests considerable variation in the SWL.

Respondents’ ideological self-placement (labeled ‘SELF’) is measured on a 10-point scale, with lower values indicating more leftist orientations. Responses to this question also encompass a great deal of variation: the mean value is 5.40, with a standard deviation of 2.09. It is worth noting that around 19 percent of respondents placed themselves in extreme positions (defined as SELF < 3 and SELF > 8).

While recognizing that political competition in most countries does not revolve around a single set of issues, there are both theoretical and pragmatic reasons to base our analysis on a unidimensional ideological spectrum. The left–right schema constitutes ‘a universal solvent’ that takes in major political conflicts (Barnes, 1997: 131), which ‘in the long run tends to assimilate all important issues’ (Inglehart, 1990: 292). The flexibility in defining the spatial schema renders it ‘immediately understandable and easily translatable across cultures’ (Laponce, 1981: 27).

Empirical studies have shown that, even in countries with historical and economic backgrounds very different from the established Western European democracies where these spatial semantics originated, left–right semantics are meaningful to mass publics (Dalton, 2006). For example, Wiesehomeier and Doyle find that ‘the Latin American electorate has a clear and coherent understanding of the ideological tenets of left and right,’ and votes accordingly (2012: 26). Furthermore, citizens in new democracies are capable of learning to recognize the left–right schema as they become familiar with open political contestation (Freire, 2006). Thus, it is appropriate to measure citizen and government positions along the left–right scale.

To test the impact of voter-government spatial distance (labeled ‘PROXIMITY’) on SWL, it is necessary to measure governments’ ideological positions. Since the WVS surveys do not ask about individual perceptions of parties’ left–right locations, cabinet position in each country is derived from a pool of six expert surveys: Castles and Mair (1984), Huber and Inglehart (1995), Benoit and Laver (2006), Wiesehomeier and Benoit (2009), the Chapel-Hill expert surveys (Steenbergen and Marks, 2007), and the expert scores in the Comparative Study of Electoral Systems (CSES) dataset.

In all cases, the position of the government in country i at time j when the WVS survey was conducted $(\bar{P_{i j}})$ is estimated as the average position of the parties in government weighted by their respective vote share. This is based on previous studies concluding that government policy positions track the weighted mean position of those parties in cabinet better than a simple average (see Warwick, 2001). Also, we always selected parties’ scores from the expert survey temporally closest to the corresponding WVS wave to estimate government positions in each country. Finally, we normalized all the SELF and expert left–right scores on a 0–10 scale to allow for direct comparisons.

As a result, our PROXIMITY variable (that is, the spatial distance between voter i and government j) is estimated as the negative quadratic distance between SELF and $(\bar{P_{i j}})$ :

P R O X I M I T Y_{a i j} = - {(S E L F_{a i j} - \bar{P_{i j}})}^{2}

(1)

where SELF_aij is the ideal point of voter a in country i at time j along the left–right spectrum and $(\bar{P_{i j}})$ is the position of the government in country i at time j on the same spectrum. In our sample, the average value of PROXIMITY is −8.2, with a standard deviation of 10.9.⁷

Control variables

Of course, political variables are not the only determinants of happiness. Previous works have identified a number of socio-demographic and economic factors as important influences on life satisfaction. At the individual level, in addition to socio-demographic traits typically employed in the literature (for example, gender, marital status, age, and age squared, to account for the curvilinear relationship between age and life satisfaction), we include variables measuring respondents’ self-reported health, given the strong correlation between this indicator and subjective well-being (Frey and Stutzer, 2002), their level of generalized trust as a proxy for cognitive social capital (Helliwell, 2003; Helliwell and Huang, 2008), level of satisfaction with one’s household financial situation, a dummy for parenthood (coded as 1 for respondents with children) (Tavits, 2008), and the post-materialist index included in the WVS.

Furthermore, we include two additional variables that could mediate the relationship between SELF and life satisfaction. First, religion attendance is used as a proxy for religiosity. Bjørnskov et al. (2008) listed the frequency of religious attendance as a factor that significantly affects life satisfaction. Second, many works have affirmed a positive relationship between personal income and life satisfaction (for example, Diener and Diener, 1995; Frey and Stutzer, 2000). We have therefore created the following variable: by employing a subjective answer to a country-specific, 10-category, income-scale question, we measured the median income level in each country. Our dummy income variable assumes a value of 1 if the respondent reports an above-median income in her country.⁸

Given that our dataset covers 40 countries, it is necessary to control for the impact of several aggregate-level (that is, country-level) variables on individual life satisfaction. Concerning economic indicators other than personal income, we included each country’s average growth rate in the five years preceding each survey, assuming that the worse (or better) a country’s overall recent economic performance, the lower (or higher) satisfaction with life would be. Given debates over the effect of GDP per capita as a predictor of life satisfaction across countries (see Schyns, 2002), we used the logarithm of GDP per capita, estimated under purchasing power parity (PPP). Also, since Di Tella et al. (2001) and Frey and Stutzer (2000) underlined the significant negative impact of unemployment on life satisfaction, we controlled for a country’s average unemployment rate in the five years preceding the survey.⁹

In line with the discussion above on macro-level political variables, formal institutions that enhance the quality of resource allocation and public goods provision should increase life satisfaction (Helliwell and Huang, 2008). We therefore include the first dimension scores extracted from a principal component analysis of the widely used World Bank governance indicators relating to effectiveness, regulatory efficiency, the rule of law, lack of corruption, voice and accountability, and political stability (Kaufmann et al., 2002).¹⁰

We also included dummies for post-communist countries, Latin America, and Asia, which previous research has shown to be highly significant (Bjørnskov et al., 2008, 2010). Besides controlling for similar cultural backgrounds, these dummies also largely correspond to new democracies (with a few exceptions, such as Japan, among Asian countries).

Finally, we added period-fixed effects to the model (one dummy for each wave of the WVS) to account for joint macro trends over time, such as business cycles, and for the changing country composition of our sample across waves (see the Appendix for a full list of control variables).

Empirical analysis

Since our dependent variable (that is, life satisfaction) is a 10-point scale, we used an ordered logistic model. In addition, we corrected standard errors for intra-group correlation and heteroskedasticity by clustering individuals at the country-year level (for example, Spain 1981, Spain 1990, and so on). We also ran a sequential ordered logit, which relaxes the assumption of parallel regression in an ordered logit (see Boes and Winkelmann, 2004). All the qualitative results reported below hold intact. Moreover, the sequential ordered logit confirms that the categories of our dependent variable are monotonically related to an underlying latent variable, thus affirming that the ordered logistic model is appropriate.¹¹

In any ordered logistic model, an underlying score is estimated as a linear function of the independent variables and a set of cut points (or thresholds). The probability of observing a given outcome therefore corresponds to the probability that the estimated linear function is within the range of the cut points estimated for the outcome. In our case, there are 10 possible categories (from 1 to 10). We are interested in identifying at which point of the latent scale category ‘1’ changes to category ‘2’ (and similarly for the other categories). This is what we mean by a ‘cut point’. In particular, in all the subsequent analyses, our benchmark score will be the probability of moving above the cut point of 7, that is, the probability of being more satisfied than the average value in our sample.

Table 1 reports the six models we estimated. Model 1 directly tests the ideological position hypothesis (H1). SELF has a highly significant and positive coefficient, corroborating findings in previous works that right-leaning citizens seem significantly more satisfied than their leftist counterparts. Note that this is true after controlling for income and religion attendance, two of the main reasons advanced in the literature to explain this relationship. This suggests some deeper motivations behind the association between conservatism and happiness. To explore such linkage in more detail, we added a SELF-squared variable in Model 2 to check the possibility of a curvilinear relationship between ideology and happiness. The results show that the squared term of SELF is significant and that Model 2 clearly improves upon Model 1 (as can be seen by comparing the AIC information criterion). As SELF increases, the SWL decreases until it reaches a minimum of around four, after which it increases again. Therefore, between the rival hypotheses H2a and H2b, there is significantly stronger empirical support for the latter. In other words, respondents holding ideologically extreme positions appear to be happier than moderates.¹²

Table 1.

Explaining Happiness Across the World (Ordered Logit Regression).

	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Individual attributes
PROXIMITY	−	−	0.00144	−0.000965	0.00197	−0.00118
			(0.00166)	(0.00292)	(0.00383)	(0.00294)
SELF	0.0340^***	−0.126^***	−0.140^***	−0.101^***	−0.129^**	−0.0977^***
	(0.00538)	(0.0191)	(0.0273)	(0.0289)	(0.0403)	(0.0288)
SELF squared	−	0.0157^***	0.0172^***	0.0124^***	0.0149^***	0.0123^***
		(0.00198)	(0.00267)	(0.00290)	(0.00411)	(0.00287)
PROXIMITY*SELF	−	−	−	0.00374^*	0.00454^*	0.00387^**
				(0.00147)	(0.00199)	(0.00149)
PROXIMITY*SELF squared	−	−	−	−0.000419^**	−0.000513^*	−0.000427^**
				(0.000156)	(0.000205)	(0.000156)
Income	−0.101^***	−0.0961^**	−0.0957^**	−0.0928^**	−0.0161	−0.0924^**
	(0.0304)	(0.0306)	(0.0306)	(0.0303)	(0.0389)	(0.0302)
State of health	−0.530^***	−0.531^***	−0.531^***	−0.530^***	−0.518^***	−0.531^***
	(0.0221)	(0.0219)	(0.0219)	(0.0219)	(0.0317)	(0.0218)
Post-materialism	0.0410⁺	0.0368⁺	0.0371⁺	0.0382⁺	0.0693^*	0.0399⁺
	(0.0209)	(0.0209)	(0.0209)	(0.0208)	(0.0295)	(0.0211)
Age	−0.0385^***	−0.0386^***	−0.0386^***	−0.0386^***	−0.0360^***	−0.0385^***
	(0.00376)	(0.00375)	(0.00375)	(0.00375)	(0.00540)	(0.00378)
Age squared	0.000405^***	0.000402^***	0.000403^***	0.000403^***	0.000365^***	0.000403^***
	(0.0000384)	(0.0000382)	(0.0000382)	(0.0000382)	(0.0000533)	(0.0000384)
Gender	0.114^***	0.119^***	0.119^***	0.117^***	0.127^***	0.117^***
	(0.0178)	(0.0175)	(0.0175)	(0.0174)	(0.0208)	(0.0175)
Satisfaction with financial situation	0.402^***	0.401^***	0.401^***	0.401^***	0.436^***	0.401^***
	(0.0172)	(0.0172)	(0.0172)	(0.0172)	(0.0255)	(0.0172)
Religion attendance	−0.0340^***	−0.0365^***	−0.0364^***	−0.0359^***	−0.0449^***	−0.0361^***
	(0.00636)	(0.00644)	(0.00640)	(0.00634)	(0.00909)	(0.00623)
Trust in people	−0.159^***	−0.161^***	−0.161^***	−0.163^***	−0.116^***	−0.170^***
(Continued)
	(0.0261)	(0.0259)	(0.0258)	(0.0256)	(0.0338)	(0.0253)
Being married	0.380^***	0.384^***	0.384^***	0.384^***	0.359^***	0.384^***
	(0.0265)	(0.0267)	(0.0266)	(0.0266)	(0.0348)	(0.0265)
Children	0.0874^***	0.0876^**	0.0874^**	0.0869^**	0.0870^*	0.0889^***
	(0.0264)	(0.0267)	(0.0266)	(0.0267)	(0.0371)	(0.0269)
Country-year attributes
Log of GDP per capita	0.265⁺	0.245	0.248⁺	0.251⁺	0.209	0.272⁺
	(0.149)	(0.150)	(0.149)	(0.149)	(0.203)	(0.161)
Avg. growth (5 years)	0.0627^**	0.0615^**	0.0612^**	0.0587^**	0.0533^*	0.0552^**
	(0.0199)	(0.0197)	(0.0198)	(0.0199)	(0.0235)	(0.0205)
Avg. rate of unemployment (5 years)	−0.0140⁺	−0.0146⁺	−0.0146⁺	−0.0145⁺	−0.0104	−0.0167^*
	(0.00831)	(0.00841)	(0.00835)	(0.00821)	(0.0130)	(0.00819)
Quality of institutions	0.0762	0.108	0.107	0.106	0.142	1.285
	(0.135)	(0.134)	(0.133)	(0.132)	(0.166)	(1.201)
Quality of institutions*log of GDP per capita	−	−	−	−	−	−0.123
						(0.127)
Country attributes
Asia	−0.596^***	−0.578^***	−0.576^***	−0.571^***	−0.530^**	−0.628^***
	(0.157)	(0.160)	(0.160)	(0.161)	(0.181)	(0.181)
Eastern Europe	−0.0356	−0.0382	−0.0320	−0.0267	0.0334	−0.0688
	(0.147)	(0.148)	(0.150)	(0.148)	(0.145)	(0.154)
South America	0.649^*	0.631^*	0.630^*	0.629^*	0.615^*	0.617^*
	(0.292)	(0.292)	(0.292)	(0.288)	(0.305)	(0.288)
Temporal variables
Wave 2	−0.189	−0.169	−0.173	−0.182	−	−0.115
	(0.153)	(0.153)	(0.153)	(0.155)		(0.166)
Wave 3	−0.230	−0.205	−0.209	−0.225	0.00349	−0.123
	(0.175)	(0.175)	(0.175)	(0.176)	(0.146)	(0.203)
Wave 4	−0.314	−0.287	−0.291	−0.303	−0.00734	−0.173
	(0.236)	(0.235)	(0.234)	(0.233)	(0.133)	(0.262)
Wave 5	−0.334	−0.309	−0.313	−0.321	−	−0.188
	(0.219)	(0.220)	(0.220)	(0.220)		(0.252)
Observations	69705	69705	69705	69705	42601	69705
Pseudo R²	0.100	0.101	0.101	0.102	0.113	0.102
AIC	251869.1	251610.5	251609.5	251577.2	154229.9	251557.6
Log likelihood	−125902.6	−125772.2	−125770.8	−125752.6	−77080.9	−125741.8
Country*Year	80	80	80	80	46	80
Country	40	40	40	40	33	40

Notes: Clustered standard errors over Country*Years are shown in parentheses. Cut points have been suppressed to conserve space (available on request). + p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001.

Source: WVS (all five waves).

Figure 1 illustrates this point. In this figure, we plotted the probability of reporting a life satisfaction score above 7 as the value of SELF changes, holding all other variables fixed at their mean. We have also superimposed a histogram portraying the frequency distribution of SELF (the scale of the distribution is given by the vertical axis on the right-hand side of Figure 1). One can easily observe that Model 2 predicts a quadratic relationship. While it is true that conservatives are happier than progressives, as underlined in the literature, groups at both extremes are more satisfied than centrists.

Figure 1.

The Impact of SELF on the expected probability of SWL exceeding 7.

In Model 3, we introduced the PROXIMITY variable in order to test our ideological proximity hypothesis (H3). While this model is an improvement over Model 2, and the new variable has the expected positive sign, PROXIMITY fails to reach conventional statistical significance. However, this is not the last word on the issue. Indeed, as noted above, both H4 and H5 assume a conditional relationship between ideological proximity and citizens’ self-placements, on the one hand, and life satisfaction, on the other. To test these two hypotheses properly, two interaction terms between SELF and PROXIMITY are added to our analysis, while assuming ∂(SWL)/∂(PROXIMITY) to be substantially higher for moderate values of SELF and lower for extreme values of SELF. Conversely, we should expect that for citizens who place themselves on the extreme left or right, SELF would have a greater impact on the SWL when PROXIMITY decreases.

Testing this in Model 4, the results show that both interaction terms between PROXIMITY and SELF are highly significant, while the information criterion highlights that Model 4 improves on previous models. In order to understand the substantive magnitude of the effects found in Model 4, as well as the associated uncertainty, we simulate the marginal effect on our life satisfaction benchmark score by moving PROXIMITY from its mean (−8.2) to a rather low value (−0.5) (a change that corresponds to roughly one-half of the standard deviation decrease in PROXIMITY) as SELF changes.¹³ As predicted by H4 (see Figure 2), the results show that while PROXIMITY does not exert a significant marginal effect for respondents professing radical orientations (less than 2 and greater than 7), its greatest impact is found among moderates (self-placements of 5 and 6). This effect is not insubstantial: for a respondent who placed herself at 5, for example, the marginal effect of PROXIMITY increases our benchmark by 1.5 percent, comparable to the impact of gender, parenthood, or religious attendance.

Figure 2.

The Marginal Effect of PROXIMITY on the expected probability of SWL exceeding 7 as SELF Changes.

The first three panels in Figure 3 replicate Figure 1 (that is, the expected probability of reporting a life satisfaction score of more than 7 as the value of SELF changes) for three different values of PROXIMITY: at its minimum, average, and maximum values. As H5 underscores, the curvilinear relationship between SELF and happiness changes its shape according to our expectation (more convex for low values of PROXIMITY and flatter for high values of PROXIMITY). To generalize this conclusion, in the lower-right panel of Figure 3 we have reported the marginal effect of a one-unit increase in SELF on our benchmark life satisfaction score as both SELF and PROXIMITY change. The figure shows that this marginal impact is very large at extreme values of SELF when an individual is far away from the government’s position (that is, a low value of PROXIMITY). Indeed, at the negative extreme of PROXIMITY (−50), moving SELF from 0 to 1 (that is, toward a slightly less radical position) decreases this probability by 6 percent, while moving SELF from 9 with 10 (that is, becoming even more radical) increases the probability by 8 percent. In contrast, the same one-unit change when SELF is equal to 4 has a negligible effect on satisfaction. At the same time, as we keep increasing the value of PROXIMITY, the marginal impact of SELF at its extreme values also declines, exactly as predicted in H5. For example, when PROXIMITY equals 0, moving SELF from 0 to 1 decreases the probability by 2.0 percent (only one-third of the previous magnitude), while moving SELF from 9 to 10 increases it by 3.8 percent (less than half compared with the previous scenario).

Figure 3.

The Marginal Effect of SELF on the expected probability of SWL exceeding 7 as SELF as well as PROXIMITY Changes.

Given that no government in our dataset is located at 0 or 10 on the left–right scale, the aforementioned examples only represent hypothetical scenarios. Nevertheless, this illustration is quite relevant when one focuses on the linkage between voters’ and governments’ ideological positions.

According to the results in Model 4, an ideologically moderate government would make centrist voters, who constitute the vast majority of respondents in every country analyzed here, (slightly) happier (see Figure 2). Moreover, the same reaction would be found among extreme voters, given that they are by definition far away from a moderate government (as illustrated in Figure 3). In fact, the magnitude of this effect is larger than that for centrists. Conversely, an ideologically radical government would reduce satisfaction among centrist voters (since the cabinet is spatially distant from them), while it produces contrasting effects for extreme voters: a radical left government would make extreme right voters happier, while reducing satisfaction among extreme leftists (because the cabinet is located so close to themselves); the opposite occurs under a radical right government. This is illustrated by Table 2. This leads to the conclusion that a centrist government increases citizens’ average level of happiness through a combination of effects found for the SELF and PROXIMITY variables.

Table 2.

The Linkage Between the Cabinet’s and Voters’ Ideological Positions and its Expected Impact on the Average Level of Happiness Within a Country.

		Government’s ideological position
		Moderate	Extreme
Voters’ ideological position	Moderate	+	−
	Extreme	+	+ and –

According to the estimations of Model 4, Table 3 shows what happens to the overall (country-level) probability of life satisfaction being higher than 7 (that is, our benchmark score) when, counterfactually, the ideological position of the government changes by one-unit steps from 3 to 7 along the left–right scale, given the actual distribution of SELF in our sample. This is reasonable given that 70 percent of governments in our dataset fall into this ideological range.

Table 3.

A Counterfactual Scenario: What Happens to the Overall Probability of a Life Satisfaction that is Greater Than 7 if the Ideological Position of the Cabinet Changes.

Cabinet position on a 0–10 left–right scale	Probability	95% c.i.
3	0.439	0.438	0.440
4	0.482	0.481	0.483
5	0.503	0.502	0.504
6	0.501	0.500	0.502
7	0.477	0.474	0.480

Note: The probabilities are constructed using parameter estimates for Model 4 in Table 2.

As shown in Table 3, our benchmark probability changes in the expected direction: it increases as the government moves toward a more moderate position, and vice versa. How relevant is this finding? If we compare the impact of a government’s ideological shift with, for example, changes in a country’s rate of GDP growth, then moving a government’s position from 7 to 5 (that is, in a moderate direction)¹⁴ produces an impact on happiness roughly equivalent to an increase of around 2 percent in average GDP growth over five years. Thus, the impact is far from negligible.

Regarding the control variables, all the individual-level variables are significant and carry the expected sign, with one exception. The SWL appears to increase among the healthy, married respondents, parents, those who are satisfied with their financial situation, and those reporting higher generalized trust and more frequent religious attendance. In addition, happiness is higher among women and post-materialists (albeit only at the 90 percent confidence interval in the latter case) and has an (anticipated) curvilinear relationship with age (reaching its minimum value at around 48 years). On the other hand, we find a surprisingly negative and significant relationship between happiness and income: a person earning an above-median income appears less happy than someone below this median. To explain this, one may speculate that the relationship between (personal) income and the SWL is mediated by the context in which a person lives (see Clark et al., 2008). For example, Helliwell (2006) has shown that changes in income exert only marginal influence on happiness in wealthy countries. Moreover, within countries, income makes a greater difference among the poor (Graham and Pettinato, 2001). Thus, the sign of the income coefficient in Table 3 may be attributable to the fact that being wealthy matters more for happiness when a person lives in a relatively poor country rather than in a richer one.¹⁵

Concerning other macro-level control variables, one can see that satisfaction increases with improving economic trends (lower unemployment, higher GDP growth, and higher GDP per capita). Two of the three regional dummies are significant (positive for South America and negative for Asia), while temporal dummies are not.¹⁶ Interestingly, the quality of institutions variable is never significant, contrary to Helliwell (2006). However, we should note that the first observation for the World Bank governance indicators on which our quality of institutions variable is based is 1996. To arrive at measures for earlier periods in our dataset, we followed Helliwell and Huang (2008) by extrapolating the World Bank data from 1996 into the past, but there are questions about the validity of this method.

In Model 5 we therefore replicated Model 4, but only analyzing WVS surveys since 1996. As seen in the second to last column of Table 1, all the previous results remain intact, while quality of institutions is once again positive but insignificant. Helliwell and Huang (2008) also discussed the possibility that the impact of quality of institutions is mediated by cross-national differences in wealth. We therefore added an interaction term between quality of institutions and GDP per capita in Model 6. However, this new interaction also fails to reach significance. This does not mean that the quality of institutions in a country does not matter for happiness. Rather, our analysis suggests that the effect of this factor is probably absorbed by its impact on economic variables. Indeed, if we rerun Model 4 without the three macroeconomic variables, quality of institutions becomes highly significant.¹⁷

Conclusion

This article has investigated several hypotheses on the impact of ideology on individual life satisfaction. Do individual ideological orientations and voter–government congruence along the left–right spectrum really affect life satisfaction? In other words, does ideology buy happiness after all? The answer is affirmative, with some notable caveats.

To summarize, there is a curvilinear relationship between ideological orientations and happiness, with those who locate themselves toward both extremes on the left–right scale feeling more satisfied than centrists. This challenges previous work which found a linear association, with higher levels of happiness among right-leaning citizens. Furthermore, we demonstrated that propinquity between self-position and government position also contributes to happiness, but that this effect is heavily mediated by individual ideological orientation: centrists are more satisfied the closer they are to their government, while for citizens with radical views, proximity to government diminishes happiness.

This implies that ideologically moderate governments enjoy a comparative advantage in terms of their ability to enhance the overall level of happiness, since centrists far outnumber extremists and the former report higher levels of life satisfaction as their government’s left–right position approaches their own. On the other hand, citizens with more radical orientations may prefer the (self-)perception of isolated ideological purity and care little about where the cabinet stands, so being distant from a moderate government does not render them less happy. In contrast, an extremist government would not only alienate the majority of voters who are centrists, but also fail to make more radical voters happier.

Within the large literature on representation, policy congruence between voters and governments has often been used as an indicator of the quality of democratic representation (for example, Powell, 2000). Other scholars have noted a link between government policy representation and system support (for example, Curini et al., 2012; Ezrow and Xezonakis, 2011). Building on these works, the results of this article suggest that moderate governments may have the advantage of not only improving representational quality and system support, but also of boosting happiness among the citizenry.

The findings presented here are also relevant to academic and journalistic accounts regarding increasing political distrust among citizens in many advanced democracies, and the consequent rise in support for extreme parties (particularly on the right side of the ideological spectrum) in some countries. Some mainstream parties may seek to contain a radical competitor by constraining the latter with responsibilities of government. Whatever the electoral payoff of such a strategy, this study suggests that it carries the risk of alienating large segments of the population, not only those with moderate views, but also the core supporters of the said radical party, leading to a lose-lose situation with respect to citizens’ life satisfaction. Our results complement works that examine the failure of radical parties in government and offer a new perspective for future studies on the consequences of radical parties in office.

Footnotes

Appendix

Table A1.

Countries Covered in the Study (and the Corresponding WVS Wave).

Country	WVS 1	WVS 2	WVS 3	WVS 4	WVS 5
Argentina		1995	1999
Australia		1995			2005
Austria		1990
Belgium		1990
Brazil					2006
Bulgaria			1997
Canada	1982	1990		2000	2006
Cyprus					2006
Czech Republic			1998
Denmark	1981	1990
Estonia			1996
Finland		1990	1996		2005
France	1981	1990			2006
Germany		1990	1997		2006
Great Britain	1981	1990			2005
Hungary		1991
Iceland	1984	1990
Ireland	1981	1990
Italy	1981	1990			2005
Japan		1990		2000	2005
Lithuania			1997
Mexico				2000	2005
Netherlands	1981	1990			2006
New Zealand			1998	2004
Norway	1982	1990	1996		2007
Peru				2001	2006
Poland			1997		2005
Portugal		1990
Romania					2005
Slovakia			1998
Slovenia					2005
South Africa			1996
South Korea				2001	2005
Spain	1981	1990	1995	2000	2007
Sweden	1982		1996		2006
Switzerland	1989		1996		2007
Taiwan					2006
Thailand					2007
Uruguay			1996
USA		1990	1995	1999

Notes

Author biographies

Luigi Curini is Associate Professor of Political Science at Università degli Studi di Milano. His research centers on the spatial theory of voting and party competition. His articles have appeared in the European Journal of Political Research, Journal of Politics, British Journal of Political Science, West European Politics, Electoral Studies, and Party Politics.

Willy Jou is Assistant Professor at Waseda University in Japan. His research interests include public opinion, party systems, and ideology, with particular emphasis on new democracies. His articles have been published in Asian Survey, British Journal of Political Science, Communist and Post-Communist Studies, Europe-Asia Studies, International Political Science Review, and Party Politics.

Vincenzo Memoli is Assistant Professor at Università degli Studi di Catania. His main research interests include democracy, public attitudes, and public opinion. His articles have appeared in the British Journal of Political Science, Acta Politica, and Rivista Italiana di Scienza Politica.

References

Adams

James F.

Merrill

Samuel

Grofman

Bernard

(2005) A Unified Theory of Party Competition. Cambridge: Cambridge University Press.

Álvarez-Díaz

Ángel

González

Lucas

Radcliff

Benjamin

(2010) The politics of happiness: On the political determinants of quality of life in the American states. Journal of Politics 72(3): 894–905.

Barnes

Samuel H.

(1997) Electoral behavior and comparative politics. In: Lichbach

Mark Irving

Zuckerman

Alan S.

(eds) Comparative Politics: Rationality, Culture, and Structure. Cambridge: Cambridge University Press.

Beck

Nathaniel

Katz

Jonathan N.

(2011) Modeling dynamics in time-series-cross-section political economy data. Annual Review of Political Science 14: 331–52.

Benoit

Kenneth

Laver

Michael

(2006) Party Policy in Modern Democracies. London: Routledge.

Bjørnskov

Christian

Dreher

Axel

Fischer

Justina A.V.

(2007) The bigger the better? Evidence of the effect of government size on life satisfaction around the world. Public Choice 130(3–4): 267–92.

Bjørnskov

Christian

Dreher

Axel

Fischer

Justina A.V.

(2008) Cross-country determinants of life satisfaction: Exploring different determinants across groups in society. Society and Welfare 30(1): 119–73.

Bjørnskov

Christian

Dreher

Axel

Fischer

Justina A.V.

(2010) Formal institutions and subjective well-being: Revisiting the cross-country evidence. European Journal of Political Economy 26(4): 419–30.

Blanchflower

David G.

Oswald

Andrew J.

(2004) Well-being over time in Britain and the USA. Journal of Public Economics 88(7–8): 1359–86.

10.

Boes

Stefan

Winkelmann

Rainer

(2004) Income and happiness: New results from generalized threshold and sequential models. IZA Discussion Paper No. 1175.

11.

Bok

Derek

(2010) The Politics of Happiness. Princeton, NJ: Princeton University Press.

12.

Brennan

Geoffrey

Lomasky

Loren

(1993) Democracy and Decision: The Pure Theory of Electoral Preference. Cambridge: Cambridge University Press.

13.

Buelens

Hino

Airo

(2008) The electoral fate of new parties in government. In: Deschouwer

Kris

(ed.) New Parties in Government: In Power for the First Time. London. Routledge, pp. 157–74.

14.

Castles

Francis G.

Mair

Peter

(1984) Left–right political scales: Some ‘expert’ judgments. European Journal of Political Research 12(1): 73–88.

15.

Claassen

Ryan L.

(2007) Campaign activism and the spatial model: Getting beyond extremism to explain policy motivated participation. Political Behavior 29(3): 369–90.

16.

Clark

Andrew

Frijters

Paul

Shields

Michael A.

(2008) Relative income, happiness and utility: An explanation for the Easterlin paradox and other puzzles. Journal of Economics Literature 46(1): 95–144.

17.

Crooker

Karen J.

Near

Janet P.

(1998) Happiness and satisfaction: Measures of affect and cognition? Social Indicators Research 44(2): 195–224.

18.

Curini

Luigi

Jou

Willy

Memoli

Vincenzo

(2012) Satisfaction with democracy and the winner–loser debate: The role of policy preferences and past experience. British Journal of Political Science 42(2): 241–61.

19.

Dalton

Russell J.

(2006) Social modernization and the end of ideology debate: Patterns of ideological polarization. Japanese Journal of Political Science 7(1): 1–22.

20.

Deschouwer

Kris

(ed.) (2008) New Parties in Government: In Power for the First Time. London: Routledge.

21.

Diener

Marissa

(1995) Cross-cultural correlates of life satisfaction and self-esteem. Journal of Personality and Social Psychology 68: 653–63.

22.

Di Tella

Rafael

MacCulloch

Robert

(2005) Partisan social happiness. Review of Economic Studies 72(2): 367–93.

23.

Di Tella

Rafael

MacCulloch

Robert

Oswald

Andrew

(2001) Preference over inflation and unemployment: Evidence from surveys of happiness. American Economic Review 91(1): 335–41.

24.

Dorn

David

Fischer

Justina A.V.

Kirchgässner

Gebhard

. (2005) Democracy and happiness revisited. Paper presented at the European Public Choice Society Conference, Durham, 31 March–3 April.

25.

Dreher

Axel

Öhler

Hannes

(2011) Does government ideology affect personal happiness? A test. Economics Letters 111(2): 161–5.

26.

Easterlin

Richard A.

(1995) Will raising the incomes of all increase the happiness of all? Journal of Economic Behavior and Organization 27(1): 35–47.

27.

Easterlin

Richard A.

(2001) Income and happiness: Towards a unified theory. Economic Journal 111: 465–84.

28.

Eysenck

Hans J.

(1954) The Psychology of Politics. London: Routledge and Kegan Paul.

29.

Ezrow

Lawrence

Xezonakis

Georgios

(2011) Citizen satisfaction with democracy and parties’ policy offerings. Comparative Political Studies 44(9): 1152–78.

30.

Freire

Andre

(2006) Left–right ideological identities in new democracies: Greece, Portugal and Spain in the Western European context. Pôle Sud 25: 153–73.

31.

Frey

Bruno S.

Stutzer

Alois

(2000) Happiness, economy and institutions. Economic Journal 110: 918–38.

32.

Frey

Bruno S.

Stutzer

Alois

(2002) What can economists learn from happiness research? Journal of Economic Literature 40(2): 402–35.

33.

Graham

Carol

Pettinato

Stefano

(2001) Happiness, markets, and democracy: Latin America in comparative perspective. Journal of Happiness Studies 2(3): 237–68.

34.

Greenberg

Jeff

Jonas

Eva

(2003) Psychological motives and political orientation – the left, the right, and the rigid: Comments on Jost et al. (2003). Psychological Bulletin 129(3): 376–82.

35.

Helliwell

John F.

(2003) How’s life? Combining individual and national variables to explain subjective well-being. Economic Modelling 20(2): 331–60.

36.

Helliwell

John F.

(2006) Well-being, social capital and public policy: What’s new? Economic Journal 116: C34–C45.

37.

Helliwell

John F.

Huang

Haifang

(2008) How’s your government? International evidence linking good governance and well-being. British Journal of Political Science 38(4): 595–619.

38.

Huber

John D.

Inglehart

Ronald

(1995) Expert interpretations of party space and party locations in 42 societies. Party Politics 1(1): 73–111.

39.

Inglehart

Ronald

(1990) Cultural Shift in Advanced Industrial Society. Princeton, NJ: Princeton University Press.

40.

Inglehart

Ronald

Klingemann

Hans-Dieter

(2000) Genes, culture, democracy, and happiness. In: Diener

Suh

Eunkook M.

(eds) Culture and Subjective Well-Being. Cambridge, MA: MIT Press.

41.

Jou

Willy

(2010) The heuristic value of the left–right schema in East Asia. International Political Science Review 31(3): 366–94.

42.

Kaufmann

Daniel

Kraay

Aart

Zoido

Pablo

(2002) Governance Matters II: Updated Indicators for 2000/01. Washington, DC: World Bank.

43.

Kemmelmeier

Markus

(2008) Is there a relationship between political orientation and cognitive ability? A test of three hypotheses in two studies. Personality and Individual Differences 45(8): 767–72.

44.

Kim

In Chul

(2005) A sense of alienation towards government: An analytic framework. International Review of Public Administration 9(2): 55–64.

45.

Laponce

Jean A.

(1981) Left and Right: The Topography of Political Perceptions. Toronto: University of Toronto Press.

46.

Layard

Richard

(2006) Happiness and public policy: A challenge to the profession. Economic Journal 116: C24–C33.

47.

McClosky

Herbert

Chong

Dennis

(1985) Similarities and differences between left-wing and right-wing radicals. British Journal of Political Science 15(3): 329–63.

48.

McDonnell

Duncan

Newell

James L.

(2011) Outsider parties in government in Western Europe. Party Politics 17(4): 443–52.

49.

Malahy

Lori W.

Rubinlicht

Michelle A.

Kaiser

Cheryl R.

(2009) Justifying inequality: A cross-temporal investigation of U.S. income disparities and just-world beliefs from 1973 to 2006. Social Justice Research 22(4): 369–83.

50.

Napier

Jaime L.

Jost

John T.

(2008) Why are conservatives happier than liberals? Psychological Science 19(6): 565–72.

51.

Ott

Jan C.

(2010) Good governance and happiness in nations: Technical quality precedes democracy and quality beats size. Journal of Happiness Studies 11(3): 353–68.

52.

Ovaska

Tomi

Takashima

Ryo

(2006) Economic policy and the level of self-perceived well-being: An international comparison. Journal of Socio-Economics 35(2): 308–25.

53.

Pacek

Alexander

Radcliff

Benjamin

(2008) Assessing the welfare state: The politics of happiness. Perspectives on Politics 6(2): 267–77.

54.

Powell

G. Bingham

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

55.

Radcliff

Benjamin

(2001) Politics, markets, and life satisfaction: The political economy of human happiness. American Political Science Review 95(4): 939–52.

56.

Rokeach

Milton

(1960) The Open and Closed Mind: Investigations into the Nature of Belief Systems and Personality Systems. New York: Basic Books.

57.

Schyns

Peggy

(2002) Wealth of nations, individual income and life satisfaction in 42 countries: A multilevel approach. Social Indicators Research 60(1–3): 5–40.

58.

Sidanius

James

(1985) Cognitive functioning and sociopolitical ideology revisited. Political Psychology 6(4): 637–61.

59.

Steenbergen

Marco R.

Marks

Gary

(2007) Evaluating expert judgments. European Journal of Political Research 46(3): 347–66.

60.

Stimson

James A.

(1976) Public support for American presidents: A cyclical model. Public Opinion Quarterly 40(1): 1–21.

61.

Suh

Eunkook

Diener

Oishi

Shigehiro

. (1998) The shifting basis of life satisfaction judgments across cultures: Emotions versus norms. Journal of Personality and Social Psychology 74(2): 482–93.

62.

Tavits

Margit

(2008) Representation, corruption, and subjective well-being. Comparative Political Studies 41(12): 1607–30.

63.

Taylor

Paul

Funk

Cary

Craighill

Peyton

(2006) Are We Happy Yet? Washington, DC: Pew Research Center.

64.

Veenhoven

Ruut

(2000) Well-being in the welfare state: Level not higher, distribution not more equitable. Journal of Comparative Policy Analysis 2(1): 91–125.

65.

Warwick

Paul V.

(2001) Coalition policy in parliamentary democracies: Who gets how much and why. Comparative Political Studies 34(10): 1212–36.

66.

Wheeler

Robert J.

(1991) The theoretical and empirical structure of general well-being. Social Indicators Research 24(1): 71–9.

67.

Whiteley

Paul

Clarke

Harold D.

Sanders

David

. (2010) Government performance and life satisfaction in contemporary Britain. Journal of Politics 72(3): 733–46.

68.

Wiesehomeier

Nina

Benoit

Kenneth

(2009) Presidents, parties, and policy competition. Journal of Politics 71(4): 1435–47.

69.

Wiesehomeier

Nina

Doyle

David

(2012) Attitudes, ideological associations and the left–right divide in Latin America. Journal of Politics in Latin America 4(1): 3–33.

70.

Zechmeister

(2006) What’s left and who’s right? A Q-method study of individual and contextual influences on the meaning of ideological labels. Political Behavior 28: 151–73.