Institutional context and life satisfaction: does the rule of law moderate well-being inequalities? Open Access

Social theorists have suggested that if sociologists want to be relevant to modern society, they have to understand how the rule of law functions (Turner 2009). Similar to the inequality debate in subjective well-being (Kelley and Evans 2017a; Schneider 2019), we might consider the rule of law as a means to an end in its contribution to society's development and individuals’ well-being (Krygier 2016). However, despite its potentialities, the arguments have ranged from considering the rule of law as an ‘unqualified human good’ given its capacity to limit arbitrary power (Thompson 1975) to more skeptical views arguing that it can also serve to enhance the power of the ruling class (Fine 2002). Since life satisfaction has long been considered a dimension of mental health (Headey, Kelley, and Wearing 1993; Uecker 2012), we build on the medical and political sociology literature to argue that institutions do not only influence well-being but also the extent of its disparities (Bakhtiari, Olafsdottir, and Beckfield 2018; Beckfield 2018; Beckfield et al. 2015; Jutz 2020; Sosnaud and Beckfield 2017). Particularly, we examine the variations in the association between political power and subjective well-being by how the rule of law is instituted across societies.

The rule of law's contribution to economic and political development has been well-studied by social scientists (Acemoglu and Robinson 2012; Fukuyama 2012). This study aims to broaden the discussion of the rule of law's implication by shedding light on its role in people's everyday lives. Following discussions in social theory (Fine 1994; Krygier 2016; Thompson 1975) and Schneider's (2019) emphasis on the role of individual-level characteristics on the association between country-level exposures and subjective well-being, we argue that the rule of law moderates the association between power and subjective well-being.

We focus on the moderation of individual political power because its functioning is particularly amenable to the influence of macro-level institutional factors, especially the rule of law, which regulates how power at the personal level is exercised (Beckfield 2018). Moreover, political power presents many power phenomena features in sharper relief, making it ‘the power par excellence’ (Poggi 2001:30). In the Fundamental Causes Theory (Link and Phelan 1995; Phelan, Link, and Tehranifar 2010), power is a key resource that facilitates individuals to maintain health advantages. Nevertheless, universal, fair, and accessible laws might enable ordinary people to achieve many life goals through institutionalized means rather than relying on personal power. The rule of law may constrain the exercise – and abuse – of power by the powerful. The perversions and pathologies of power are the fundamental problems that the rule of law tries to solve (Krygier 2016). On the other hand, some sociologists (Calarco 2018; Fine 1994; Roscigno 2011) cast doubt on the effectiveness of power constraint rules. Despite well-established, fair, and transparent rules, powerful individuals could use their resources to manipulate and negotiate their advantages. Thus, the rule of law will serve the ruling class's interest by constraining only the powerless and proving opportunities to the powerful for enhancing their advantages. This study empirically investigates whether, in a fairer and more open system with transparent and universally applied rules and procedures, individual power may matter less or more to subjective well-being. We use a unique dataset of 30,491 individuals from 27 countries with diverse social and political characteristics to assess how the rule of law interacts with the personal power to influence subjective well-being.

The contributions of this article are threefold. First, we join recent efforts to theorize and systematically test how macro-level institutional factors shape individuals’ well-being (Kelley and Evans 2017a; Ólafsdóttir and Beckfield 2020; Schneider 2019). In particular, we incorporate the rule of law into this literature and examine whether it tempers or enhances the role of subjective power on individual well-being. Our findings illustrate how the rule of law may influence the ways in which social inequality is translated into inequality in well-being and provide insights into why disparities in well-being vary across different social contexts. Similar to the evidence in national income inequality, the rule of law itself does not affect subjective well-being. However, our evidence suggests that the rule of law reduces the effect of power on life satisfaction. Second, this study deepens our understanding of how access to power, an essential but understudied dimension of social stratification, is related to an important life outcome. By analyzing cross-national data, we show that this relationship is not fixed but depends on the contextual institutional configurations. Finally, because our analyses clearly identify institutional features that reduce disparities in well-being, this study may provide guidance for policy design and implementation.

In the next section, we discuss the theoretical bases of the rule of law role as moderator of individuals’ access to power and their life satisfaction. Moreover, we further present a brief survey of the recent literature that assesses the relationship between power and well-being. After describing our data, measurements, and analytical strategy, we provide the results and, lastly, discuss our findings.

In recent years, the literature on the role that macro-level exposures have on individuals’ well-being disparities has reached its momentum (Ólafsdóttir and Beckfield 2020). Country-level inequality has been the hallmark of this debate, with results that contradict the common assumption that inequality negatively affects the population's well-being (Kelley and Evans 2017a; Ngamaba, Panagioti, and Armitage 2018). In addition, medical and political sociologists have suggested that institutional arrangements are another important macro-exposure that plays a fundamental role in the production of well-being and its disparities (Beckfield 2018; Beckfield, Olafsdottir, and Bakhtiari 2013; Olafsdottir and Beckfield 2011).

The rule of law is an institutional ideal and, as such, contains schemes and resources that organize power (Beckfield 2018). O’Donnell (2004) distinguishes between the minimal assertion of the rule of law and the democratic rule of law. The former understands the rule of law as laws publicly promulgated by an appropriate authority, and fairly applied by relevant state institutions, including consistent application across cases; neutral to class, status, or relative power; and pre-established procedures. The latter affirms political equality and constraints against abuse of power by ensuring political rights, civil liberties, and mechanisms of accountability. Other authors (e.g. Dougherty, Gryskiewicz, and Ponce 2018) termed them as thin and thick definitions of the rule of law, respectively.

As Acemoglu and Robinson point out (2006), institutions have the capacity to influence political actions, political equilibria, and the distribution of power in the present and the future. The rule of law's effect on individuals’ power is a central micro-level implication of this institutional arrangement (Hadfield and Weingast 2014; Krygier 2016; Thompson 1975). Both thin and thick forms of the rule of law are required for institutions to function effectively (O’Donnell 2004). As schemes, institutions are an idea and a model of social relations that motivate policies, and they require employees and economic capital to operate. Schemes (i.e. a symbolic system that constitute categorization) and resources for enacting them (i.e. material capabilities) are coordinated by laws and state policies that define the rules of the game (Beckfield 2018).

Empirical evidence supports the effects of more efficient and impartial norms on individual well-being (e.g. Ciziceno and Travaglino 2019; Ferrara and Nisticò 2019; Graafland and Compen 2015; Helliwell and Huang 2008; Nikolova 2016; Ott 2010). Moreover, the central definition of the rule of law provides insights to understand how it could modify the effects of individuals’ access to power. Institutions define the rules that enable power to be deployed. The relative weight of coercion and persuasion, the two primary forms of exercising power for the power holder's interest, depends on the context (Castells 2013). Therefore, different institutional settings might explain variations in how specific power positions could ensure access to well-being. For example, scholars interested in the relationship between social context and health (Bakhtiari et al. 2018; Beckfield et al. 2015; Olafsdottir and Beckfield 2011) argue that complex sets of institutionalized civil rights could alter the causes of health, illness, healing, and well-being. The literature suggests two competing predictions on how the rule of law moderates the relationship between power and well-being.

First, the rule of law may mitigate the functioning of personal power. The literature has considered the power-tempering role as its primary function (Krygier 2016; Thompson 1975). The rule of law influences what has been termed the production function of an individuals’ well-being (e.g. Schneider 2019). A strong rule of law ensures all citizens’ political equality by constraining the abuse of power and promoting political rights, civil liberties, and mechanisms of accountability. It is considered an institutional feature that fosters social trust across countries, explaining the ‘Nordic exceptionalism’ in this regard (Delhey and Newton 2005; Sonderskov and Dinesen 2014). The notion of the rule of law also generates a virtuous cycle where laws apply equally to everybody; the rule of law may constrain the exercise of the power of the powerful, and no individual or group could rise above the law (Acemoglu and Robinson 2006, 2012; Krygier 2016; Thompson 1975). As a result, the rule of law establishes a pluralistic distribution of power and equality among individuals in access to well-being. Thus, in a society with a strong rule of law, ordinary people or people with no special access to political power can achieve many life goals through institutionalized means, rather than resorting to personal advantages. Therefore, personal power may be less relevant to well-being in societies with a strong rule of law.

Nevertheless, from a second perspective, skeptical views cast doubt on the effectiveness of institutional arrangements on the reduction of social inequalities (Calarco 2018; Kalev 2009; Powell and DiMaggio 1991; Roscigno 2011). According to Fine (1994), the emphasis on the inhibition of power oversimplifies the multiple functions of law and lost sight of the legal form's limitations. The rule of law cannot be isolated from the power relations that originated them. As Roscigno (2011) has pointed out, there is a potential decoupling between what institutions profess to do and how they operate. Therefore, although formalization mitigates inequalities (e.g. Anderson and Donald 1995; Majumdar and Marcus 2001), rules per se cannot constrain the advantages of the powerful and their privileged access to well-being. The appropriate coordination between schemes and resources is not ensured by the sole existence of a legal corpus. Moreover, socially advantaged individuals could use their agency to negotiate their advantages even in seemingly fair and well-functioning institutions (Calarco 2018). Thus, as an alternative hypothesis, the rule of law could enhance the advantages of the powerful or mirror them. This alternative hypothesis challenges the very same definition of the rule of law as neutral to relative power.

Not only do government officials from WEIRD (Western, educated, industrialized, rich, and democratic) countries advocate the rule of law, but also those from countries around the world with different political systems and in different stages of development: ‘No other single idea has ever achieved global endorsement’ (Tamanaha 2004:3). However, countries vary in the strength in which they apply the rule of law, and adherence to it could be only a declarative slogan. Estimations of the World Justice Project (2019) suggest that around 1.5 billion people cannot obtain justice for everyday problems; 4.5 billion people are excluded from the opportunities that the law provides; 253 million people live in extreme conditions of justice. North and colleagues (2012) estimate that only 15% of the world population could be considered as living in countries where the rule of law is respected, which they call ‘open access societies.’ In addition, not all countries experience the same flaws in their rule of law (O’Donnell 2004). Haggard and Tiede (2011) show that developing countries exhibit different rule of law ‘syndromes’ due to corruption, risk of expropriation, and violence, but not necessarily formal institutional arrangements. Therefore, cross-national variations in the rule of law could yield variations in how personal resources affect well-being. Our central aim is to assess the contingency of this relationship between life satisfaction and political power across countries based on the degree in which laws are defined, universally applicable, and fair.

Both theoretical expositions and empirical evidence suggest that at the individual level, access to power is consequential to well-being. Power is an important dimension of social stratification and power relations are crucial in the formation and reproduction of hierarchies and permeate all the dimensions of human activity (Castells 2013, 2016; van Kleef and Cheng 2020; Weber 1978[1922]). Following a long-standing relational tradition, we understand power as the capacity that enables particular actors (e.g. group, person, organization) to affect the decisions, will, and values of other actors asymmetrically for their own interests (Blau 1986; Castells 2016; Emerson 1962; Weber 1978[1922]). In the political arena, power is a measure of the influence that particular individuals have when there is conflict over which policies should be implemented (Acemoglu and Robinson 2006; Chaney 2013).

Social psychologists have long argued that the desire for power is a fundamental human motive across cultures, genders, age groups, and personalities (for a review, see Anderson, Hildreth, and Howland 2015). People engage in a wide range of goal-directed activities to manage their power because of the benefits that it affords in terms of autonomy, one's sense of control, reduction of stress and anxiety, boosting of self-esteem, among others (Kifer et al. 2013; Mitchell et al. 2020; Sherman et al. 2012; Stroope 2015). Hence, individuals are incentivized to seek power by the intrapersonal outcomes that it generates, and the lack of power may entail detrimental psychological consequences.

Empirical studies have recently provided evidence supporting the theoretical prediction of the consequences of access to power on psychological well-being (Anderson, Kraus, et al. 2012; Kifer et al. 2013; Yu and Blader 2020). Keltner and colleagues (Keltner, Gruenfeld, and Anderson 2003) have observed that individuals in positions of power (e.g. parent, priest, or political pundit) enjoyed increased positive affection, while reduced power is associated with negative affection. Moreover, a recent study (Jin, Tam, and Tao 2019) found that perceived political power is linked to individuals’ happiness and life satisfaction in China. Likewise, through political power, individuals could have leverage in policies and decisions that directly affect their living conditions (e.g. school admission, land use, security, and social assistance). Several studies have shown that political power affects factors that enable individuals’ access to well-being, such as wealth and social capital (Fisman, Schulz, and Vig 2014; Shen 2019).

Despite the recent progress in the research on the links between power and life outcomes (Jin et al. 2019; Yu and Blader 2020), we know little about how power functions in different institutional contexts. As argued in the preceding section, the access and functioning of power are highly dependent on the institutional context. This article seeks to contribute to the research on power and well-being by assessing the rule of law as an institutional determinant for the functioning of power. By taking a comparative framework, this article explains the variation in the relationship between power and subjective well-being.

This study uses a large and unique data set of individuals from 27 countries from the 2019 Gallup's World Poll (Gallup 2019), which includes measurements of life satisfaction and subjective power status. Countries are heterogeneous in terms of geographical location, political institutions, and levels of economic development. The countries were selected to incorporate the unique and proprietary measurement of subjective power of this study. The diversity of countries attempted to approximate our selection to a sample of countries that reflects the diversity of worlds’ country. Within countries, samples of citizens are probabilistic and representative of the population over the age of 15 (Table 2S in the supplementary material). The total sample comprises 36,456 individuals. After a listwise deletion, the analytical sample includes 30,491 respondents. No major differences in descriptive statistics between both samples are found. Following previous studies using Gallup's World Poll (Easterlin, Angelescu, and Zweig 2011; Helliwell et al. 2018), we use individual weights provided by Gallup that adjust for gender, age, and, where reliable comparative population data are available, education, or socioeconomic status. In addition, due to differences in sample sizes between countries, a robustness check of level-two weights is also reported in the supplementary material (Table 3S).

Life satisfaction. Individual life evaluations are measured by answers to the Cantril Ladder of Life (Cantril 1965), which uses an 11-point visual scale as a framing device and makes it possible to assess general life satisfaction. ‘Please imagine a ladder, with steps numbered from 0 at the bottom to 10 at the top. The top of the ladder represents the best possible life for you, and the bottom of the ladder represents the worst possible life for you. On which step of the ladder would you say you personally feel you stand at this time?’

Perceived Political Power. We assessed individuals’ perceptions of their positions in the power hierarchy using the visual analog method. The respondents were presented with a ladder with 10 rungs and asked: ‘Please imagine a ladder with steps numbered from zero at the bottom to ten at the top. Suppose we say that the top of the ladder represents people in this country who have THE MOST INFLUENCE over their local government, and the bottom of the ladder represents people in this country who have THE LEAST INFLUENCE. By ‘influence over the local government,’ I mean influence over the officials and behavior of the city/county government. On which step of the ladder do you think you, personally, stand?’ Previous studies have also used perceived positions in a power hierarchy as a proxy of power (e.g. Anderson, John, and Keltner 2012; Wang 2015). Moreover, sub-national politics account for a large proportion of the activities of the government across countries (Fisman and Gatti 2002; Trounstine 2009), including land use, schooling, security, etc. Therefore, it is a good proxy of the general construct of political power.

Rule of Law Index. The World Justice Project (WJP) developed the WJP Rule of Law Index (Botero and Ponce 2009) to serve as a quantitative tool for measuring the rule of law in practice. Scholars of the law and society suggest that it is the most ambitious and comprehensive effort to measure the rule of law globally (Versteeg and Ginsburg 2017). The WJP Rule of Law Index's methodology and definitions are the products of consultation and vetting with academics, practitioners, and community leaders from 128 countries and jurisdictions and 17 professional disciplines. Conceptually, the index is rooted in both thin and thick definitions of the rule of law used in the literature (Dougherty et al. 2018; O’Donnell 2004; Versteeg and Ginsburg 2017). The WJP Rule of Law Index's conceptual framework comprises eight factors that include constraints on government powers, the absence of corruption, open government, fundamental rights, order and security, regulatory enforcement, civil justice, and criminal justice. The original index was rescaled to a range from 0 to 10 and grand-mean centered.¹

Individual – and country-level indicators are included as control variables. Among individual-level variables, gender is included based on the literature on subjective well-being, indicating that men have a stronger power-based desire for rank than women (Hays 2013). In addition, we control for age, educational level, area of residence, health status, marital status, and having children because they are potential predictors of life satisfaction and factors of social stratification (Diener et al. 1999; Diener, Oishi, and Lucas 2009; Grusky and Ku 2008). Finally, recent evidence (Jin et al. 2019) in status inconsistency shows that subjective wealth compensates for the effect of power on subjective well-being², and economic conditions are important determinants of subjective well-being (Boyce, Brown, and Moore 2010). Therefore, we controlled for an 11-points indicator of subjective wealth and two binary indicators of economic hardship: (1) not enough money for food or (2) shelter.

Regarding country-level controls, the rule of law could be closely related to other macroscopic indicators of institutional quality. Therefore, to argue that our main institutional variable measures the rule of law and not a generalized quality of institutions (Bjornskov 2008), we included the mean-centered 2019 Democracy Index (The Economist Intelligence Unit 2019) and 2019 Freedom of Press Index (Reporters Without Borders 2019) as control variables. Moreover, we considered GDP per capita based on the literature that suggests that developed countries can spend more on their judicial system (Messick 1999), which could confound the effect of the rule of law on life satisfaction. Besides, we control for income inequality based on cross-national studies that suggest Gini's effect on individuals’ life satisfaction (Inaba et al. 2015), and that the income gap could affect the distribution of power in certain societies. The urban population as a percentage of the total population is another relevant control due to urbanization's effect on the distribution of power and quality of life (Dodoo and Tempenis 2002). Additionally, country's employment to population rate over the age of 15 addresses the effect of employment on power hierarchies and life satisfaction (Bless and Granato 2018). Finally, we control for the interview mode (telephone vs. face-to-face) (see Table 2S in the supplementary material).

Table 1 summarizes the weighted descriptive statistics of all the variables in the analysis. The average age of the sample is 42 years. 51% are women, and 52% have an intermediate educational level. Regarding family life, 56% of the respondents declared being married or partnered, and 45% have at least one child. One in four respondents live in rural areas (25%), one in three declared hardship in food (32%), and one in four declared hardship in access to any shelter (26%).

Table 2 reports results from models assessing the overall association between life satisfaction and political power. The distinction between Models 1 and 2 is that Model 2 includes country-level fixed effects, in addition to controls at the individual level. Consistent with the literature examining the association between these two constructs, the coefficient of power is positive and significant (B = 0.069, p < .001), indicating that in general, greater access to power is correlated with greater life satisfaction. This effect is similar in the models with (Model 2) and without (Model 1) country-level fixed-effects. In addition, further analysis shows that the magnitude of the effect of political power is large and comparable to other important socioeconomic indicators, such as the income percentile.

After testing the association between life satisfaction and political power at the individual level, we used random-effects models to assess the effect of the rule of law and the interaction between the rule of law and perceived power. Models include region fixed-effects. Thus, the estimations only rely on variation within regions. Model 1 in Table 3 shows that the centered Rule of Law Index has a significant effect on life satisfaction when country-level controls are not included (B = 0.165, p < .001). However, this effect is smaller and non-significant when macro-level indicators are controlled for (Model 2). Further analysis shows that the Rule of Law Index overlaps to a great extent with GDP per capita and the degree of urbanization. In addition, we cannot rule out the possibility that some of these country-level controls could be mediators of the rule of law effect. Therefore, the interpretation of the controlled models should be made with caution.

As a cautionary note, inference using p-values assumes probabilistic sampling. In our case, we have probabilistic and representative samples of individuals within countries and a non-probabilistic sample of countries. Thus, although level-1 coefficients are unbiased, the research design does not satisfies the fulfills the criteria of a fully multilevel probability sample (Lucas 2014). Following one of the common strategies for inference from non-probabilistic samples according to Berk (2004), we assumed that our selection of diverse countries in terms of region, political system, and level of development approximates a random sample of countries. In our case, it is unplausible to conceive a random sample of countries that fulfills the probabilistic criteria of level 2 units. As explained by Hox (1995:1), ‘in real research one may have a convenience sample at either level.’

Several analyses were conducted to check the robustness of our findings. All of them are reported in the supplementary material in Tables 3S.

First, scholars have warned about the effect of influential cases in multilevel modeling when conducting cross-national studies (Van der Meer, Te Grotenhuis, and Pelzer 2010). In our case, the analysis runs the risk that second-level slopes are unreliable due to the sample's influential countries. To probe the robustness against this potential bias, we estimated the random-effects models using a Jackknife estimation. This procedure systematically leaves out each country from the dataset and reports the average of the estimated effects. Model 1 in Table 3S shows these results. Both main effects and cross-level interactions are consistent with the main findings. Therefore, we can rule out the risk of influential countries.

Second, institutional quality indicators have been criticized because most of them are created based on ad hoc committees (Giannone and De Frutos 2016; Tasker 2016). The Rule of Law Index is one of the most transparent indexes with detailed information about its elaboration process (Botero and Ponce 2009; World Justice Project 2020), and it avoids most of the flaws of other measurements (Skaaning 2010). Nevertheless, we assess the robustness of our results using alternative measures of the rule of law. First, we used the sub-index of Human Rights and Rule of Law of the 2019 Fragile States Index (Fund For Peace 2019). As Acemoglu and Robinson (Acemoglu and Robinson 2012) acknowledge, a weak rule of law is one of the central causes of state failure. Therefore, we might expect similar results to our main findings. Second, we used the 2018 Bayesian Index of Corruption (Standaert 2015). Modern rule-of-law scholars and practitioners (North et al. 2012) define the independence of judiciaries from corruption and political manipulation as a key aspect of the rule of law. Hence, we also expect a similar effect when using this indicator. In the sensitivity analysis, these indicators are also mean-centered across countries and reverted for an interpretation in the same direction than the Rule of Law Index. Model 2.1 and Model 2.2 in Table 3S of the supplementary material shows the effects of the Human Rights and Rule of Law sub-index, and Models 3.1 and 3.2. consider the Bayesian Index of Corruption. The main effects and cross-level interactions are consistent with the findings using the Rule of Law Index. Overall, these sensitivity analyses suggest that our results are not an artifact of the specific measurement of the rule of law used in this study.

The use of these alternative country-level measurements is also related to what King, Keohane, and Verba (1994) termed observable implications of the theory. If our theory of the Rule of Law's role is correct, we might also expect the observed moderation for the Bayesian Index of Corruption. Besides, at the individual level, the rule of law may also moderate the effect of other individual resources such as wealth: people with no special access to economic resources can also achieve life goals in societies with a stronger rule of law. Model 4.2 shows a significant interaction between subjective wealth and the Rule of Law in the same direction than the interaction with subjective power. Although we do not aim to provide causal evidence, these implications are consistent with underlying expectations, strengthening our theory's explanatory power.

Nonetheless, revisions of Rule of Law indexes have been questioned for the extent to which they can be distinguished from neighboring concepts. Veersteeg and Ginsburg (2017) examined the convergence between the rule of law indexes, including the WJP index, and other neighboring concepts. They ruled out the claim that poor conceptualizations or experts’ halo effect could generate convergence with (1) democracy, (2) human rights, (3) constitutionalism, (4) judicial independence, and (5) GDP per capita. However, they cannot conclude that the rule of law indexes captures something different from corruption. To test the robustness of our estimation against this criticism, we estimated the interaction between subjective power and the Rule of Law Index, simultaneously controlling for the 2018 Bayesian Index of Corruption. The analysis shows a consistent negative interaction with a slightly larger coefficient (B = -.015, p < .05). Therefore, the power-tempering effect of the rule of law is distinguishable from the role that country-level corruption can play.

Finally, studies using Gallup data do not generally discuss the implications of different sample size across-countries (Table 2S, in the supplementary material). The original dataset only provides weights for within-country estimations. However, countries have different sample sizes that are not entirely proportional to their population, generating potential biases in the estimates. Therefore, we built a country-level weight, which assigns the same weight to each country in the analyses. Model 5 in Table 3S shows these results. The findings match the estimations using only individual-level weights.

Overall, the main effects and cross-level interactions are robust against influential countries, alternative country-level indicators, observable implications, and differences in sample sizes across countries.

This study aims to understand how an important macro institutional feature modifies the relationship between individuals’ subjective well-being and perceived political power. We find that in societies with a higher degree of the rule of law, the relationship between perceived power and life satisfaction is weaker, net of other individual and country-level indicators. Thus, we confirm that the rule of law plays a power-tempering role in modern societies (Acemoglu and Robinson 2012; Krygier 2016; Thompson 1975).

This study opens new avenues to the literature in political sociology and social determinants of well-being. Students of social determinants of well-being have called for considering alternative macro-level exposures to examine the production of individuals’ well-being (Kelley and Evans 2017a, 2017b; Schneider 2019). During recent years, an important body of research has been simultaneously accumulated to reveal how the institutional environment, in particular, the welfare regimes, are linked to the level and distribution of population health and well-being (Bakhtiari et al. 2018; Bjørnskov, Dreher, and Fischer 2010; Ferrara and Nisticò 2019; Hall and Lamont 2009; Jutz 2020; Olafsdottir and Beckfield 2011).

We put these different frameworks in dialogue to assess the well-being consequences of an important feature of the broad political-institutional context, the rule of law. We found that well-defined, universally applicable, and fair laws reduce the effect of political power on subjective well-being. This finding is substantial for well-being inequalities because power has been suggested as one of the resources used by people to gain well-being advantages and reproduce social gradients (Link and Phelan 1995; Phelan et al. 2010). This study extends the previous research by going beyond the welfare state and country-level inequality to examine how a feature of the upstream institutional arrangements of the political system interacts with personal resources to influence well-being outcomes. Similar to the evidence showing that inequality does not directly affect well-being or under specific conditions (Kelley and Evans 2017a; Kenworthy 2017; Ngamaba et al. 2018), we did not find a significant direct effect of the rule of law on subjective well-being. Nevertheless, its power-tempering role is confirmed, consistent with the discussion addressing the interaction between macro-level exposure and individual-level characteristics in the function production of well-being (Schneider 2019).

Moreover, following Pleasence and Balmer (2019b), this study may illustrate how the rule of law could expand people's legal capabilities. In their argument, an effective open legal system requires people to have the opportunity to decide whether and how to use legal frameworks. Thus, if disadvantaged individuals are more likely to encounter civil problems, the rule of law will provide the legal empowerment to solve them, weakening the negative effect of legal problems on well-being (e.g. Douglas 2018; Horn, Vahidy, and Charters 2011). Although the rule of law is a much broader concept than legal capabilities, the former can be understood as an institutional enabling condition of the latter. There are also certain commonalities. For instance, our definition and operationalization of the rule of law (Botero and Ponce 2009; O’Donnell 2004) incorporate whether people are able to voice concerns and bring specific complaints to government officials in carrying out their legal duties in practice, which is consistent with proposed measures of legal capability. Further studies may leverage recent methodological developments (Pleasence and Balmer 2019a) to examine legal capability at the individual level as one of the possible mechanisms tempering social inequalities in well-being.

Our findings also illustrate the importance of systematically theorizing and investigating how different types and aspects of the institutional context shape individual well-being through various pathways. Both theorization and empirical investigation have started to emerge but are far from adequate. Moreover, different institutional features may complement or compete with one another to influence health and well-being. Therefore, more comprehensive and composite measures of the institutional context are needed. In terms of the rule of law, this study takes the first step assessing its effect on well-being and inequalities. Future studies will need to assess other institutional features that interact and/or complement the rule of law. As Fine suggested about the modern centrality of the rule of law, ‘we cannot privilege politics or law or love in isolation’ (1994:211). A comprehensive understanding of well-being inequalities production is needed.

By focusing on the rule of law, the study emphasizes the importance of assessing the quality of institutional arrangements, above and beyond the arrangements per se. The institutional approach has long considered institutions shaping micro-level actions by imposing enforceable rules and moderating the political economy of political power (Acemoglu and Robinson 2006, 2012; Mahoney 2010). Sociological institutionalism (e.g. DiMaggio and Powell 1983) argues that once organizational models are institutionalized, they diffuse. Therefore, the successful ways a country constrains power are imitated by other countries in an isomorphic process. It could explain the widespread adherence to the rule of law principles across governments (Tamanaha 2004). Nevertheless, it could also be argued that states often make cultural and discursive claims to achieve institutional legitimacy (Roscigno 2011; Roscigno et al. 2015; Suddaby and Greenwood 2005). Thus, countries encounter a decoupling between declared isomorphic organizational forms and how they actually work. Our analyses have shown how institutional quality's heterogeneity plays an important role in moderating the access that powerful and powerless people have to well-being. Countries with fair, universal, and well-defined laws can off-set the capacity that enables individuals to affect the decisions and will of other actors for the sake of their interests. Therefore, policy design and implementation might consider the rule of law as an institutional feature that may reduce well-being disparities by regulating political power.

This article also contributes to the study of power as a determinant of health and well-being. Power is an important and unique dimension of social stratification, but empirical research on its consequences for individual well-being has only started to emerge recently. Our paper adds to this small body of literature by extending the study of the effects of political power on life satisfaction to societies with diverse political and institutional configurations. Moreover, we show that the consequence of power may not be uniform across different social contexts; rather, it depends on a specific feature of the institutional environment, i.e. the rule of law.

Future studies might expand our findings by considering multidimensional measurements of life satisfaction or other dimensions of mental health. In this study, we assessed the effect of power on overall satisfaction with life. This practical approach could veil what van Praag et al. (2003) have characterized as the ‘anatomy’ of subjective well-being, where general satisfaction relates to different domains of life (job, family, financial, housing, health, and leisure). Regarding the specific countries in the sample, they include a wide array of geographical locations, cultural traditions, and political systems. Nevertheless, we did not consider countries from the Arab world due to data availability. Qualitative evidence from the Middle East (Barber et al. 2014) has recently shown that individuals frame political functioning and political power as essential determinants of quality of life. Therefore, further case studies could focus on these countries as exemplary cases to study the rule of law's role in those contexts or include the subjective power measurement in the full sample of countries in Gallup World Poll.

Finally, our cross-sectional data does not enable us to make causal claims about our variables’ relationships. As shown by Kragten & Rözer (2017) for the relationship between income inequality and individuals’ health, cross-sectional and longitudinal models exhibits divergent results, which is explained by the omission of confounders in the former type of models. We rely on ample evidence in social psychology, suggesting a causal effect of power on subjective well-being. Moreover, in the case of the Rule of Law, we have included a set of theoretically motivated potential confounders aimed to address this omitted variable bias. For instance, Kragten & Rözer (2017) explained that cultural differences between countries potentially confound income inequality effects. In this study, we aimed to control these differences by regional fixed effects. In future data collections, the inclusion of the measurement of subjective power will enable control for additional time-invariant country-level characteristics. But this study certainly illuminates venues for understanding the role of the Rule of Law and institutions more broadly in reducing social inequalities.

The authors gratefully acknowledge the funding support of the Research Grant Council of Hong Kong (GRF-14618618). Dr. Olivos thanks the contribution of the RGC Postdoctoral Fellowship Scheme.

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

Lei Jin is associate professor at the Department of Sociology at Chinese University of Hong Kong. Her research interests lie in population health and health disparities.

Francisco Olivos is a Research Assistant Professor at Lingnan University, Hong Kong. His research interests are stratification, cultural sociology, and subjective well-being.