Interview with Gwangeun Choi on Economic and Political Inequality in Cross-national Perspective

Gwangeun Choi presented the paper, “The Link between Economic and Political Inequality in Cross-National Perspective” at the Politics and Inequality conference held in Warsaw, Poland in December 2018.

Dr. Choi recently received a PhD in Government at the University of Essex in the UK. His research interests are in the areas of democracy, quality of democracy, political inequality, economic inequality, perceived inequality, redistributive preferences, redistribution, and universal basic income. His latest article, “Revisiting the Redistribution Hypothesis with Perceived Inequality and Redistributive Preferences” appeared at the European Journal of Political Economy (2019).

We asked Gwangeun Choi for an extended abstract of his Politics and Inequality conference paper and, via email, some questions about his research. We are thankful for his positive and detailed response.

Extended Abstract by Gwangeun Choi

It is widely believed that there exists a debilitating feedback cycle linking economic and political inequality. However, there has been a lack of empirical evidence about this association, particularly, in cross-national comparative research. It is largely because cross-national measures of political inequality are underdeveloped. To fill this gap, this study introduces the Political Inequality Index (PII) and the Political Power Inequality Index (PPII). The PII is composed of the two dimensions: participation and representation, which are based on the reconceptualization of political inequality from the perspective of a middle-range conception. The PPII comes from the indicators that measure the distribution of political power across socioeconomic position, social group, and gender, which the Varieties of Democracy provides. This inquiry then investigates the two-way causal relationship between economic and political inequality. In the first causal direction, net income inequality is used as a proxy for economic inequality, while in the reverse causal linkage political inequality is supposed to influence market income inequality and redistribution separately, as income inequality is considered as an outcome of the two different distributive stages. In doing so, both causal directions between economic and political inequality are integrated into a unified framework. With respect to estimation techniques, a system GMM estimator for a dynamic panel data model, which is an increasingly popular estimation method, is mainly used to address the issue of endogeneity. The findings show that net income inequality does not significantly affect political inequality and that political inequality appears to have little impact on market income inequality, while political inequality seems to contribute to economic inequality by influencing redistribution in a negative direction.


The research you presented at the Politics and Inequality conference was on the relationship between political inequality and economic inequality in cross-national perspective. How did you get interested in this topic? And how is this topic connected to other research that you are doing?

I first became interested in a concept ‘political equality’ while I was doing research on the quality of democracy. In my framework designed to conceptualize and measure the level of democracy, political equality was one of the three core principles of democracy. Thus, it was easy for me to construct a new measure of political inequality, building on this framework. The next step was to investigate the reciprocal relationship between economic inequality and political inequality, as I realized that there is a lack of empirical evidence on this linkage although no one seems to doubt the widespread belief of the vicious cycle between economic and political inequality.

What is most challenging about measuring political inequality, and why?

I think that the most challenging part is to provide convincing theoretical arguments on the conceptualization of political inequality. Measuring the quality of democracy is also faced with the same issue. My study on democracy and political inequality and several other studies attempting to measure them with relatively thick concepts reached a consensus in excluding both minimalist and maximalist approaches. However, this does not guarantee that the majority of scholars agree with a specific middle-range concept of democracy or political inequality. This is therefore a more pressing issue than a range of measurement problems.

Continue reading “Interview with Gwangeun Choi on Economic and Political Inequality in Cross-national Perspective”


Five Problems with Measuring Political Inequality

How much political inequality is there? Is political inequality rising, falling, or staying the same? To answer these questions, we would have to measure the idea of “political inequality.”

Here are five main problems in measuring political inequality:  

1. Political influence is hard for scientists to observe.  Political influence is notoriously difficult to measure because it is an interaction between power wielders that is more inferred than directly observed.  We tend to “see” power after the decision is made, not during the decision process.

2. The range of potential political resources is extremely diverse and heavily dependent on context.  Political resources are anything one can use to influence a political decision: social or psychological factors – material, ideational, a personal attribute, a group level attribute, an authority position, a network connection – or an action, such as political participation. In international perspective, this is further complicated by seeking a measure that is functionally equivalent across nations.  

3. Political outcomes is also hard to measure. To answer the question, “does political inequality matter?”, we would have to empirically demonstrate that governmental decisions systematically favor some groups over others. Some recent work in the U.S. is exemplary (see the Gilens and Page argument below). Similar work outside the American context is rare.

4. Political equality never existed. Is political equality a real, empirically visible end of the continuum? If political equality is an ideal then does a theoretical endpoint belong in an empirical measure?

5. We need to specify the particular type of political inequality. Political inequality can be found anywhere within the political process. Let’s simplify the political process to two parts – voice and response.   Voice refers to how constituencies express their interests to decision-makers directly or through representatives.  Response refers to how decision-makers act and react to their constituencies and is expressed via policy and symbols.

The Gilens and Page Measure of Political Inequality

A recent article on inequality and policy outcomes by Gilens and Page highlights the promise and the difficulties in measuring political inequality. Their unique data consists of 1779 policies taken up by the U.S. Congress from the 1980s to the early 2000s. Each policy is matched to a public opinion survey research question asked during the time the policy was introduced (“this policy says X, to what extent do you agree with it?”) and to a set of interest groups who have taken a position on the policy. With these data, they gauged the extent to which the policy outcome reflects (a) the will of the median voter – identified within the surveys and (b) types of interest groups, such as economic elites, business interests, and mass public interest groups. They found that policy outcomes tend to favor the will of economic elites, not the median voter. Their study provides solid, further evidence of the paucity of pluralism in American democracy.

Gilens and Page recently published a spirited and convincing defense of their findings, but we should consider how the basis of their study — the measure of political inequality — has some fundamental problems.

A. They chose policies based on whether they were asked in public opinion surveys, and that means the many, not-so-famous policy debates that also shape key economic distribution policies were excluded.

B. Nor can it account for the policies that are off the Congressional agenda, the type of power that Bachrach and Baratz (1962) warned that is most pernicious: the power to compel voters to not even ask for the policy in the first place.

C. It is also specific to the American experience; though it can be replicated elsewhere, so far there is no cross-national equivalent to these data.

Gilens and Page conducted what is likely one of the best and most unique studies on American political inequality, and it’s just a start.

Josh (2)


Joshua Kjerulf Dubrow is an Associate Professor at the Polish Academy of Sciences. This post is adapted from the article, “Political Inequality is International, Interdisciplinary, and Intersectional,” published by Sociology Compass in 2015.


Cross-National Measures of Political inequality of Voice

Dubrow, Joshua Kjerulf.  2010.  “Cross-National Measures of Political Inequality of Voice.”  ASK: Research and Methods 19: 93-110.


Social scientists have long argued that political power is a key dimension of stratification, yet few empirically analyze political inequality or explicitly discuss the methodological implications of their measures of it. Political inequality is a distinct dimension of social stratification and a form of power inequality whose domain is all things related to political processes.  It is a multidimensional concept – comprised of voice, response, and policy – that occurs in all types of governance structures.  Conceptions of political inequality of voice reflect the well-established finding that position within the social and political structure impacts individual and group political influence. I argue that definitions and measures of political inequality of voice should focus on the extent of influence given its connection, but not reduction, to economic resources.  This article proposes and evaluates cross-national structural measures of political inequality of voice based on the relationship between socioeconomic status and political participation.  I explore the relationships between the measures and the rankings of European countries using data from the European Social Survey 2008 and the Economist Intelligence Unit Index of Democracy 2008’s “political participation” category.

Statement on the Study of Political Inequality

Social scientists have long argued that political power is a key dimension of stratification (Weber 1946; Lenski 1966; Dahl 2006), yet few empirically analyze political inequality (Winters and Page 2009).   Although attention to global inequality has increased in the social stratification literature, most examine income (Firebaugh 1999; Milanovic 2002; Neckerman and Torche 2007), some examine health (Goselin and Firebaugh 2004), and almost none examine political influence (Anderson and Beramendi 2008).  Most discussions of political inequality consist of philosophical debates over whether political equality is possible, or even necessary (Verba 2006; Bohman 1999; Dahl 2006; Ware 1981).  The few empirical discussions neither explicitly discuss the methodological implications of their measures of political inequality nor discuss how they can be applied cross-nationally (Winters and Page 2009; Anderson and Beramendi 2008).  This is a huge gap in our knowledge of how modern societies work.

This brief statement has three parts.  First, I present many definitions of political inequality, and argue each implies a distinct empirical measure.  Second, I suggest some empirical measures of political inequality.  Third, I offer a sketch of the field of political inequality.

Continue reading “Statement on the Study of Political Inequality”

Dahl on Measuring Political Equality

Here is what Robert Dahl had to say about measuring political equality:

According to Dahl, in On Political Equality


“Achieving truly well-grounded judgments about the future of political equality in the united states probably exceeds our capacities.  One reason is that, unlike income or wealth, or even health, longevity, and other possible ends, to estimate gains and losses in political equality we lack cardinal measures that would allow us to say, for example, that “political equality is twice as great in country X as in country Y.”  At best we must rely on ordinal measures based on judgments about ‘more,’ ‘less,’ ‘about the same,’ and the like.” 

He goes on to say that we might be able to develop ordinal measures by qualified observers to ascertain measures of more, less, or about the same.

More Notes on Piven and Cloward’s (2005) Discussion of Power Resources

Piven and Cloward (2005) discuss power resources.  “…it is perhaps the most important dispute in the discussion of power.  What are to be regarded as power resources?” (36)

Distributional approaches:
Weber: “all conceivable qualities of a person and all conceivable combinations of circumstances may put him in a position to impose his will in a given situation” (as cited in Wrong 1979:23).

Collins (1975):  “the resources for conflict are complex,” comprising just about everything

Dahl (1961): “anything that can be used to sway the specific choices or strategies of another individual”

Oberschall (1973): “anything from material resources… to nonmaterial resources”

Giddens (1985): allocative vs. authoritative resources

Etzioni (1968): utilitarian/material resources vs. normative/symbolic (see also Gamson 1968)

Tilly (1978): economists’ factors of production

Mills (1956): occupation of key positions/”command posts.”

One kind of resource can be used to gain another.

Piven and Cloward: “from this perspective, power resources are the attributes or things that one actor can use to coerce or induce another actor… almost everyone has something that can be used to influence somebody” (37).

Key resources are not widely distributed, but are concentrated at the top of the social hierarchy (37). 

Yet, those at the lower end do still influence social change.  How?

Interdependent relations, or interdependency approach: 

Piven and Cloward:  “the effective exercise of power in electoral representative institutions… does not result simply from a general currency of things or traits and the pattern of their distribution but rather depends on the specific relationships that make particular things or traits useful and important.”  (38).

“…in addition [to disembodied attributes, resources for power] are derived from the patterns of interdependence that characterize all social life” (39). 

“People have potential power, the ability to make others do what they want, when those others depend on them for the contributions they make to the interdependent relations that are social life” (39).

Power is embedded in interaction.  Power resources – or, rather, what are considered as power resources – are dependent on the interactional context.  “Control over capital is an effective resource for exercising power over others because those others are already entangled in a system of economic relations that makes them dependent on entrepreneurs for the means of production and subsistence” (39).

The exception is force – either the threat or actual physical force.  This is an important and possibly far-reaching exception, as force is embedded in many social interactions. 

The basic assumption is that actors are equal in the power relationship.

Giddens (1984): “much of day-to-day interaction is routine and is subject only to reflexive monitoring and rationalization” (34).  Piven and Cloward: “most of the time, people only try to make their everyday lives.  They do not try to make history” (41).

These interactional contexts in which power arises are in “the systems of interdependence that constitute societies” (40).  Systems are plural: “A web of complex networks of political, economic, and cultural interdependencies has to be analyzed if the actual potential for power by different participants in these networks is to be deciphered” (41).

The interdependency approach shares intellectual terrain with “exchange theorists.”  Power is “an attribute of relationships,” not of social actors.  They criticize Blau’s theory that power imbalances are the result of asymmetrical contributions to needed things:  employers have more power because they supply the jobs and those with less power have less to contribute to whatever is needed.  “Needed services” contributions… but landowners, do they contribute more than the workers who build the estate?  How is this measured?  P&C accuse Blau of justifying—even morally – power imbalance.

But if equality of contributions is inherent in mutual, interdependent relationships, why is the distribution of economic and status resources unequal?  Why don’t people use the potential power embedded in these interactions?

Value of contributions cannot be measured, and is therefore not the answer.  There are seven reasons why participants do not exercise their potential power:

1.  Consciousness: “people must recognize their potential power before they can act on it.”
2.  Coordination: in circumstances of collectivities, action requires coordination across similarly situated actors (as opposed to husband and wife relationships).
3.  Staying power: “power seekers must be able… to tolerate the costs” of acting.
4.  Control over supply of alternatives: “power seekers” must be able to prevent the other parties from going around them, circumnavigating them, to get what they want (via scabs, etc.).
5.  Exit: “contenders do not respond to challenges by simply exiting… or threatening to exit” (43) the relationship.
6.  Third-party leverage: “power seekers must be “free from constraints” from another, third party, that could intervene (e.g. courts intervening in labor disputes, e.g. Reagan firing the air-traffic controllers.
7.  Force:  power seekers must be free from threat or use of physical force.

“Social rules inhibit the activation of interdependencies and hence restrict the wide exercise of power” (43).

“Rules… [specify] the behaviors that are permissible by different parties in interdependent relations” (44).  Only certain rules are of interest: those that have direct bearing on power exercise.  The system of interdependencies, i.e. the context in which power interactions operate, include “rules” which legitimate “the actions available to some contenders while delimiting the actions available to others” (44).

Key to P&C’s theory is the legitimation of power exercise.  Legitimation is institutionalized, embedded in institutions that organize social life, such as law, policy.  Focus of power struggles is over rules—their legitimacy and their enforcement.  Once rules are legitimized, power interactions are characterized as being one-sided because the power seekers who attempt to break the rules no longer have legitimate means of challenging other parties. 

The role of the state is to formulate, impose and enforce the rules.

Agency in P&C’s approach:  They go back and forth on the constraints of social structure and the capacity of people to think and act beyond the constraints of social structure.  Because people are reflexive, they have the capacity to transcend constraints of social structure.  Reflexivity can therefore change the dynamic of power interactions, enabling rule breaking.


Piven, Frances Fox and Richard A. Cloward.  2005.  “Rule Making, Rule Breaking, and Power” pp. 33 – 53 in The Handbook of Political Sociology: States, Civil Societies, and Globalization edited by Thomas Janoski, Robert Alford, Alexander Hicks, and Mildred A. Schwartz.  Cambridge: Cambridge University Press.

Defining and Measuring Political Resources

The definition and measurement of political inequality, in some formulations of the concept, requires a definition of political resources.  Let’s start with a definition of political inequality.

— In the study of political inequality, political resources are viewed as a dimension of social stratification, including the ability to influence both governance processes and public policy.
—  Like economic goods and services, political resources are scarce, valued, and fought for.

What are political resources?  Are they different from power resources?  Let’s be a bit simple at first and say there are two perspectives (borowed from Piven and Cloward 2005):

(a) Distributional:  Resources are anything that can be used to influence an outcome.  Resources are used, but it is not power itself.  Resources are distributed unequally.  “Power resources” is used to describe any resources used in the exercise of power.  Political resources are resources used in political decision-making, or for all areas of social-life that are make claims toward a legislative/decision-making body (from school-boards to national government). Political inequality refers to structured differences in the distribution and acquisition of political resources. Power is an attribute of people.

The term “power resources” is misleading, as it suggests that power itself can be distributed.  Most distributional theorists argue that power is relational.  For example, one actor’s political resource is only a resource if it is perceived as a resource by the other actor.

(b)  Interdependency:  Resources are never strictly defined and can take the form of anything actors can do within an interaction.  Resources are actions available to the participants in the interaction.  These resources are valid because they are an integral part of the interdependent relationship.  The nature of the interdependent relationship reveals the types of actions (resources) available to each participant.  For example, in capitalist economies, ownership of land and wealth is a valid resource.  Employers have power over their employees because the employees are dependent on the employer for their economic livelihood.  Power is an attribute only of relationships, not people themselves.

The interdependency approach is different from the distributional approach because it assumes that each actor in the interaction has equal power resources.  Employers can only make employees work because employees agree to work.  If employees decided not to work, such as a work-strike, then the employees could be said “to have power over” the employers.  This approach does not adequately account for “force,” or physical coercion.

Resources are political when they enable claims-making toward a legislative/decision-making body.  For example, romantic relationships have elements of power, where each participant has a range of actions or range of resources at their disposal to get what they want despite the resistance of the other.  But this behavior is not political.

How to measure political inequality?

If political inequality is the unequal distribution of political resources, then the measurement of political inequality is dependent on the measurement of political resources.  But, how can we measure “anything?”  Dahl (1996) defines political resources as “almost anything “– including money, reputation, legal status, social capital and knowledge, to name a few — that has value and can be used to achieve political ends.  If we want to answer the question, “how much political inequality is there,” “anything” is too vague a measure of political resources and too context dependent. 

Sorokin (1959 [1927]) defines political stratification this way:  “If the social ranks within a group are hierarchically superimposed with respect to their authority and prestige, their honors and titles; if there are rulers and the ruled, then whatever are their names (monarchs, executives, masters, bosses), these things mean that the group is politically stratified, regardless of what is written in its constitution or proclaimed in its declarations” (11).  To Sorokin, authority, prestige, honors and titles are political resources.  Authority position seems to be the main determinant of who has power and who does not.

Some have measured political inequality in terms of political participation, specifically “voter turnout.”  There is political inequality if there are divisions in who votes and who doesn’t.  Some go broader and define political inequality in terms of the level of democratization.  Measuring political inequality with level of democracy assumes that the introduction of political rights and civil liberties leads directly to reduction of inequalities.  But, as Verba et al (1978) point out, for democracy to reduce inequality, rights and liberties are not enough; citizens must also be engaged in political participation (see also APSA 2004).  Thus, it is not democracy alone that matters, but what citizens do with the rights and liberties allowed by democracy.  Democracy cannot be a measure of political inequality or political resources. 

What would be a measure of political resources?  Is there a core set of “political resources” that can be used in every political situation?  One plausible measure of political resources is experience in political affairs, which is obtained through political participation. 

Democracy as a measure of political inequality does not shed much light on the link between economic and political inequality, i.e. the degree to which nations are internally-stratified in terms of political resources.  Democracy does have a relationship to economic outcomes, but it is not equivalent to political inequality.
As Verba et al (1978) point out, for democracy to reduce economic inequality, rights and liberties are not enough; citizens must also be engaged in political participation (see also APSA 2004).  Thus, it is not democracy alone that matters, but what citizens do with the rights and liberties allowed by democracy.  The relationship between participation and redistributive policies is further complicated by within-nation social stratification.  Political participation is stratified, such that the advantaged tend to participate more than the disadvantaged.  Economic distributive policy reflects the interests of the advantaged precisely because the advantaged are more politically active.  Political non-participation of the disadvantaged leads to an increase in economic inequality, or maintains its status quo. 


Piven, Frances Fox and Richard A. Cloward.  2005.  “Rule Making, Rule Breaking, and Power” pp. 33 – 53 in The Handbook of Political Sociology: States, Civil Societies, and Globalization edited by Thomas Janoski, Robert Alford, Alexander Hicks, and Mildred A. Schwartz.  Cambridge: Cambridge University Press.

Verba, Stanley, N.H. Nie and J. Kim. 1978. Participation and Political Equality: A Seven-Nation Comparison. Cambridge: Cambridge University Press.