What Kinds of Economic Inequality Really Matter?

Abstract

The chapter discusses the major reasons why economic inequality should be a source of concern and the forms of inequality that are principally implicated. It considers ten different arguments as to why inequality matters—two of them moral, two political, three economic, and three social. In each case it discusses the economic variable(s) whose unequal distribution is at issue, whether economic class inequality or ethnic group inequality is most salient, and what part(s) of the unequal distribution are the most problematic—i.e., is the problem primarily poverty at the lower end, privilege at the upper end, bipolarization, or the entire distribution?

1 Introduction

For a long time economists concerned about the distribution of economic well-being in real-world societies have focused on poverty—generally defined as insufficient economic resources to attain a minimal standard of living—as the key problem to be documented, understood, and addressed. Since the end of the colonial era, national governments and international institutions like the World Bank have viewed the alleviation and ultimate eradication of poverty as an important policy objective. Even in affluent countries such as the United States, poverty lines are defined, the extent of poverty below those lines is measured, and campaigns to reduce poverty are launched—e.g., the “War Against Poverty” in the U.S. by the Lyndon Johnson Administration in the 1960s. It surely makes sense to accord high priority to reduction of the immense suffering associated with poverty, especially when it is as widespread as in many developing countries.

For a long time too, most economists and policy-makers have seen sustained economic growth as providing the surest means to reducing poverty. History certainly shows that sustained economic growth has in the past done a great deal to reduce poverty—initially in the West, then in non-Western countries such as Japan and South Korea, and lately in “emerging nations” like China and India. In recent decades, it is true, concern has grown over the extent to which economic growth actually alleviates poverty; and the term “inclusive growth” has arisen to focus attention on policies to assure that growth does in fact do so. Up until recently, however, concern about poverty has not been matched by concern about economic inequality more generally. In other words, the focus has been on absolute economic deprivation (whether one falls short of a subsistence standard) rather than relative economic deprivation (where one stands in relation to others in one’s society). If economic growth is accompanied by a widening economic gap between rich and poor, this has generally not been seen as a problem so long as the growth also reduces the proportion and the numbers of the very poor.¹

In recent years, however, attitudes toward economic inequality seem to be changing. In late 2010, the then Managing Director of the IMF—a bastion of mainstream economic thinking—declared that “Lurking behind [globalization is] a large and growing chasm between rich and poor—especially within countries. An inequitable distribution of wealth can wear down the social fabric. More unequal countries have worse social indicators, a poorer human development record, and higher degrees of economic insecurity and anxiety” (Strauss-Kahn 2010). Concern about the “large and growing chasm between rich and poor” has been stimulated in part by an accumulation of evidence that in recent decades economic inequality has increased substantially in a great many countries around the world.² It is not only that the poorer strata—e.g., the bottom 10 or 20%—are receiving smaller shares of total income and wealth. There is also much evidence of a burgeoning share for the very richest stratum (see Piketty and Saez 2011), which has contributed to a growing sense that inequality at the upper end of the distribution confers excessive power on the very rich, with problematic consequences for the well-being of a society. Although this point of view has been expressed and discussed more fully in relatively affluent countries, where evidence on the growth of a super-rich class at the top of the income distribution is more extensive, it would seem to be just as applicable to less affluent developing countries, where a class of super-rich is also often to be found.

In a recent paper (Weisskopf 2011) I reviewed a variety of arguments that have been advanced to suggest that people would be better off if the distribution of key economic resources in a society were less unequal. In this paper I would like to develop this line of analysis further by addressing an issue to which too little attention has heretofore been devoted, namely: what kinds of economic inequality really matter? I begin in Sect. 4.2 by distinguishing various forms that economic inequality can take. In Sect. 4.3 through 6 I discuss a series of moral, political, economic, and social arguments that have been made for reducing inequality; in each case I seek to identify the particular form of economic inequality that is implicated in the argument. In Sect. 4.7, I conclude with a summary of the results and a brief discussion of their implications.

2 Forms of Economic Inequality

The various arguments for reducing economic inequality turn out to reflect concerns about distinct kinds of inequality that can differ in several dimensions.

The first dimension is the economic variable or variables whose unequal distribution is at issue. The potential variables of interest include (a) income, (b) consumption, (c) wealth, and (d) access to goods and services provided by governmental or non-governmental organizations(hereafter “public services”)—in particular, those that provide capability building services such as health care, sanitation, and education. In most cases it is the amount of the variable accruing to an individual that is most critical, but in some cases the endowment of the individual’s household or family is more important.

The second dimension involves the nature of the distributional entity whose unequal possession of an economic variable is the source of concern. Most often this is the individual person or household. In this case differences among them in possession of the relevant economic variable can be described as differences of economic class,³ and the population may be divided into a hierarchy of classes, each of which is defined by a pre-specified range of values for the relevant economic variable. (The range can be defined in absolute or in relative terms, i.e., as fractiles) Alternatively, the distributional entity of interest may be a group of people who share a pre-defined characteristic, independently of their economic status. Such groups may be defined ethnically (e.g., by race, caste, tribe, religion, native language) or geographically (by politico-administrative or topographical region). In this paper I will consider only ethnically defined groups, because concern about inter-ethnic inequality is generally much weightier than concern about inter-regional inequality⁴—if only because ethnic identity is difficult or impossible to alter, while regional identity can be altered through migration. Inter-group inequalities—as well as inequalities across separate hierarchical classes of individuals—are most readily measured by assigning to each group the group median⁵ for the variable at issue,⁶ which sets up a frequency distribution with a number of observations equal to the number of groups.

The third dimension addresses the part (or parts) of an unequal distribution on which concern is focused. I think one can usefully distinguish four distinct configurations of inequality, as follows:

• (a) Accentuated inequality at the lower end of the distribution. This configuration involves a predominant concern with the extent of poverty conceived of in relative terms—i.e., in relation not to a pre-specified poverty line, but to the societal median. It is motivated by Sen’s distinction between income and capability: “Relative deprivation in the space of incomes can lead to absolute deprivation in the space of capabilities” (Sen 1992, p. 115).⁷ In the case of class distributions for any given economic variable x, it can be measured analogously to a head count measure of poverty by the number (or proportion) of people falling below y*x _m, where x _m is the societal median value of x, and y is a pre-specified fraction no higher than—say—50%. Or it can be measured analogously to a gap measure of poverty as the total deficit in x under y*x _m of those in the head count, taken as a share of societal total x. The distributive share of the bottom 5 or 10% of the population provides a very rough indicator of the latter measure. In the case of group distributions one would want to focus on the number of ethnic groups whose median falls below the overall median, as well as the proportionate extent to which each group median falls short.

• (b) Accentuated inequality at the upper end of the distribution. This configuration involves a predominant concern with what one might best characterize as “privilege,” in opposition to poverty, also conceived of in relative terms. Just as absolute deprivation with respect to capabilities is linked to relative deprivation with respect to economic resources, so absolute advantage with respect to power, influence and autonomy in a society is linked to relative advantage in terms of economic resources. For class distributions a head count of people with more than some very high level of x would not be very informative, since it is their aggregate economic power that is the major source of concern. Thus it would be best to use a gap-like measure—i.e., the total surplus in x above z*x _m of those who have at least that amount of x, taken as a share of societal total x, where z is a pre-specified multiple of at least—say—10. Indicators such as the distributive share of the top 1% of the population provide a very rough approximation of this measure. For group distributions accentuated inequality can be measured by the proportionate extent to which the medians of the highest-placed ethnic groups exceed the overall median.

• (c) Inequality in the form of a weak middle of the distribution. This configuration reflects concern about what recent literature has labelled “polarization,” or more specifically “bipolarization,”⁸ which means that the size of the “middle class” is small in comparison with the sizes of the upper and lower classes in the distribution. In the case of class distributions one would need to pre-specify a middle range of values of the variable x at issue, from (1 − v)*x _m to (1 + v)*x _m, where v takes on a value—say—between 25 and 50 %. The extent of bipolarization could then be measured by the ratio of the head count of those outside that middle range to the head count of those within it, or—probably less informatively, because it would be dominated by the upper class—the share of total x accruing to those outside the middle range. The distributive share of the middle quintile or the share of the middle four deciles of the population provides rough approximations of the latter measure. In the case of group distributions, one would compare the number of group medians relatively distant from the overall median to the number of group medians relatively close to it.

• (b) Inequality spread over the full distribution. This configuration of “entirety” represents distributions that do not show, to any significant extent, the particular attributes encompassed by the three configurations listed above. It can be measured—if imperfectly—by any of the traditional measures of overall inequality, such as the Gini coefficient. Such measures can be applied either to economic class distributions or to ethnic group distributions, even though the number of different pre-defined ethnic groups is bound to be far, far smaller than the number of individuals (or families) in the relevant population.

In the following four sections I will discuss in turn the major moral, political, economic, and social arguments that have been made for limiting economic inequality, ⁹ characterizing each argument in terms of the societal objective to which the reduction of inequality is expected to contribute. In each case my aim is to determine what form(s) of inequality are at issue.

3 Moral Arguments

Moral arguments about economic inequality involve value judgments about what constitutes fairness in the distribution of economic resources or well-being. People differ greatly with respect to what they consider a fair distribution; there is widespread agreement, however, on the importance of the following two objectives for a good society:

3.1 Ensure that All Members of the Society Are Treated as Equally Worthy of Respect

At first glance this objective might seem to be largely a matter of law and jurisprudence. But no matter how fair and comprehensive the law may be, how one is treated depends a great deal on one’s economic resources; and it may well depend also on one’s ethnic group membership. People with far fewer economic resources than the societal average are likely to be disrespected and disfavoured in a variety of ways, whereas people with far more resources than the societal average will tend to be treated with undue deference and granted undue favours. And people belonging to an ethnic group stigmatized by entrenched antecedent disdain or disparagement are likely to be viewed by many members of other groups as inherently less worthy and less deserving.¹⁰

The economic variables whose distribution is at issue here are primarily wealth, and secondarily income and access to public services, all of which largely determine the economic resources that a person can bring to the table. Inter-individual (or inter-family) inequality is obviously very important for the way in which people are treated. But membership in an ethnic group at or near the bottom, or the top, of the wealth distribution of group medians can also significantly affect people’s treatment, since the social status of a particular group—which surely affects the respect accorded to its members, even independently of their economic status—is likely to be highly correlated with the median wealth of that group. The configuration of distributional inequality that is most damaging to the assurance of respect for all citizens is surely accentuated poverty at the lower end, for individual poverty constitutes the biggest obstacle to equal treatment. But accentuated privilege at the upper end is also relevant, since privilege can in many ways be translated into unequally favourable treatment.

3.2 Promote Equality of Opportunity for All Citizens

The opportunities available to young men and women, as they grow up and become adult citizens, depend both on their family background and on the nature of their residential community. This is because the acquisition of productive characteristics and skills by an individual child depends significantly on the richness of upbringing that parents can offer her/him as well as on the quality of the quasi-public resources—such as neighbours, peers and schools—that local communities can offer to children. Highly unequal family economic resource endowments result in corresponding inequalities of opportunity for young men and women, both directly via the effect of family endowments on the quality of parent upbringing and indirectly via the effect of family endowments on the choice of local community in which their children grow up. Unequal opportunity resulting from unequal family resources limits in turn the degree of intergenerational social and economic mobility. Widespread diffusion of good-quality public services in areas such as education and medical care can help to reduce the inequalities of opportunity faced by children in families of different resource endowments, but the extent of private family economic resources will still play a big role in determining the extent of the opportunities to which a child has access. It is therefore no surprise that an accumulation of empirical evidence suggests that the degree of social and economic mobility in a society is inversely correlated with the degree of economic inequality.¹¹

Over and above the inequality of opportunity attributable to inequality in economic resources, unequal opportunity may well also result from overt or covert discrimination against members of an ethnic group that has been historically marginalized by group-based negative discrimination. Even the ending of such negative discrimination—through various laws designed to eliminate current discriminatory practices—may well prove insufficient to ensure equal opportunity, because decisions made in a non-discriminatory market context are unable to overcome past negative discrimination when there is a tendency toward clustering and social segregation in associational behaviour, whereby members of a particular ethnic group prefer to intermarry, to live in the same residential neighbourhoods, and to join the same community institutions (see Loury 1987). These parental and community influences convey advantages or disadvantages that cannot be equalized by market forces; so full equality of opportunity may well require that compensatory steps be taken to reduce economic disparities between groups and thereby provide more equal access to important non-market resources and social networks.

The economic variable whose distribution most strongly affects inequality of opportunity is family wealth, since it is accumulated wealth that is most critical in determining what kind of upbringing, what kind of residential community, and what kind of education parents can offer to their children. Family wealth is surely the most important determinant of opportunity, but ethnic group membership can also play a role—especially in the case of ethnic groups at or near the bottom of the of the distribution of group median wealth, who may well suffer from past or present discrimination, and ethnic groups at or near the top of the of the distribution, who may well benefit from historically generated advantages independent of their family wealth. The distribution of access to good-quality public services can also be significant, but only to the modest (for reasons noted above) extent that it is not highly correlated with the distribution of wealth. The configuration of inequality that is most critical to the promotion of equal opportunity for all citizens is the entire distribution, for unequal opportunity is generated by privilege as well as by poverty—and indeed by inequality in any part of the distribution.

4 Political Arguments

Most people would agree that it is desirable for a society to be fairly cohesive, such that people have a real sense of community and common purpose with one another as fellow members of the larger society. Likewise, few would disagree that it is desirable that a society be fairly democratic, so that each citizen has the opportunity to influence governmental decision-making. Only if there is such cohesion and democracy will a political system—and the power that it vests in the government—be broadly respected as legitimate. Consider therefore each of the following societal objectives:

4.1 Promote Social Cohesion

As the historian Tony Judt has written, “Inequality is corrosive…it rots societies from within…it illustrates and exacerbates the loss of social cohesion” (quoted at the conclusion of Strauss-Kahn 2010). Many people will not have a sense of community with their fellow citizens if there are significant economic and social differences among individuals that are not readily attributable to differential effort or desert. Likewise, social cohesion will be difficult to sustain if there are substantial economic and social differences among ethnic groups—especially disparities in the extent to which different groups are represented in powerful and prestigious decision-making positions. The achievement of a high degree of social cohesion thus depends upon the limitation of economic and social inequalities across both individuals and ethnic groups.

The distributions of all of the economic variables under consideration—income, consumption, wealth, and access to public services—affect social cohesion, because they all affect social as well as economic status. The configuration of inequality in these variables that is most critical to the promotion of social cohesion is arguably the degree of bipolarization, because the prospects for community and common purpose are undermined if the economic status of a large proportion of the population is far above the median and that of another large proportion is far below it. By the same token, if there are significantly-sized ethnic groups whose economic status is far above the societal median and/or far below it, social cohesion becomes much more difficult to sustain.¹²

4.2 Promote Democratic Vitality

The integrity of a political system is undermined if many citizens are unable to participate meaningfully in a democratic process of decision-making. The first prerequisite for democratic decision-making is surely to have a functioning system of free elections for public office and a set of democratic institutions that allow for fair competition among candidates with different views on the policy issues of the day. Once the basic elements of a democracy are established in a society, the biggest potential threat to democratic vitality is arguably the extent to which political decisions can be influenced by people who are capable of deploying enormous economic resources in the political arena—whether to finance election campaigns, lobbying efforts, or outright bribery. The experience of the United States over the past three decades testifies to the political consequences of a growing and now vastly disproportionate share of wealth held by a very small fraction of the population at the top of the distribution: it has become increasingly evident that a tiny proportion of the population thereby wields tremendously disproportionate political power (see Mann and Ornstein 2012). Great wealth passed on across generations not only limits social mobility; it also can lead to a hereditary aristocracy antithetical to democracy. Although laws regulating campaign finance, lobbying and bribery obviously play a big role in determining the extent to which wealth can be used to influence public policy, those laws are themselves subject to the influence of politically powerful individuals and organizations that they finance.

The economic variable whose distribution most strongly affects the influence of the very rich on politics is obviously wealth (and the income it directly generates), for even high salaries do not provide a comparable surplus of funds that can be deployed in the political arena. Economic class differences among individuals are much more salient than economic differences across ethnic groups, since the capacity to use wealth to influence politics seems unlikely to depend on the ethnic identity of a wealthy person. The configuration of wealth inequality that is most critical to the strength of democracy is clearly the extent to which wealth is concentrated at the upper end of the distribution. If the lack of a strong economic middle class actually inhibits democracy, bipolarization would also make a difference.¹³

5 Economic Arguments

It is often believed that greater economic inequality is associated with faster economic growth. In fact, the relevant empirical evidence suggests that greater income inequality is not correlated with higher economic growth rates, either across countries during the same time period or within countries over time (see Alesina and Rodrik 1994 and Bruno et al. 1999). This is not really surprising, for there are a number of ways in which economic inequality can impede economic efficiency and growth. Reducing inequality can promote greater efficiency and growth by contributing to the achievement of the following objectives:

5.1 Improve the Development and Allocation of Human Resources

Innate talent—as distinct from acquired skill—can reasonably be assumed to be equally distributed across different socioeconomic classes as well as across different ethnic groups. But class and group differences have a significant influence on an individual’s prospects for developing talents and acquiring skills.

The greater the degree of economic inequality among individuals, the less likely it is that full advantage can be taken of people’s innate capacities to make productive contributions. Many poor people with considerable innate talent and ability will be consigned to a poor education and to jobs of little responsibility, and the barriers to advancement that they face are likely to reduce their motivation to work hard to develop their talents and apply their skills. Moreover, lack of adequate economic resources may well adversely affect the nutrition and health as well as the education of the poor. At the other end of the socioeconomic spectrum, many rich people with little innate talent and ability will nonetheless be able to get a good education and gain access to positions of responsibility in society. Furthermore, if some groups are far better represented than others in the upper echelons of a society, then many members of the poorly represented groups may lack sufficient incentive (due to doubt about their ability to succeed) or sufficient opportunity (due to lack of access to useful connections) to develop and apply their capacity to make productive contributions. Such constraints on truly meritocratic human resource allocation are likely to result in a significant loss of economic potential.

The economic variable whose distribution most strongly affects the ability of people to develop their talents and apply their skills is family wealth, since it is accumulated wealth that is most critical in determining what kind of upbringing and what kind of education parents can offer to their children. The distribution of access to good-quality public services is also independently significant, to the extent that it is not highly correlated with wealth distribution. The configuration of distributional inequality that is most damaging to the development and allocation of human resources is surely accentuated poverty at the lower end. But accentuated privilege at the upper end is also relevant, since it provides undeserved opportunities for advancement to the well-off. In this context economic inequality among individuals would appear to be more salient than such inequality across ethnic groups. However, to the extent that ethnic group membership—independently of wealth—is a source of discriminatory processes of selection for education, employment or promotion, then group inequality is also a source of concern with respect to human resource development and allocation.

5.2 Reduce Economically Costly Tensions and Conflict

Social and political tensions linked to economic inequality can have an adverse effect on economic efficiency and economic growth in several ways. People who are economically deprived are at times driven to challenge the established order by a variety of disturbing and sometimes violent means—such as strikes, protests, sabotage, and crime. The resultant instability can render property rights less secure and thereby depress investment and productivity growth; it may also give rise to costly efforts to combat disturbances and pay for security systems, prisons, etc. Furthermore, a high degree of inequality is likely to generate suspicions that growth-oriented reforms will only benefit the rich at the expense of the poor, thereby intensifying popular opposition to much-needed reforms.¹⁴ This is especially true of reforms that increase the scope of market forces and enlarge the private sector, which may well increase allocational efficiency and/or economic dynamism, but which also tends to distribute the resultant gains in a disequalizing fashion—absent systematic efforts to limit economic inequality.

Social and political tensions, and the problems they can raise, are especially likely to be generated by economic inequalities across different ethnic groups. That is because such inequalities are sometimes attributable to—and more often perceived as explained by—discrimination against members of less-well-off groups. As a consequence, intergroup inequalities in a society are considerably more likely to evoke strong feelings about the unfairness of the social order, and they are considerably more likely than inter-class inequalities to lead to social and political conflict. Moreover, a high degree of inequality across ethnic groups fortifies suspicions on the part of members of relatively deprived groups that growth-oriented reforms will benefit more advantaged and powerful groups at their expense, thereby generating opposition to such reforms.

The distributions of all of the economic variables under consideration—income, consumption, wealth, access to public services—affect social tensions and conflict, because they all affect relative economic and social status. Because of the strong feelings associated with inequalities across ethnic groups, such inequalities are likely to be even more critical with respect to social tensions and conflict than inequalities among individuals. The configuration of inequality in these variables that is most critical to the avoidance of tensions and conflict is probably accentuated poverty at the lower end, since the poor are likely to be more concerned by their own economic deprivation than by the economic prosperity if the rich. Accentuated privilege at the upper end, however, may well also be of some significance, since it tends to fortify the belief that the economic abundance enjoyed by the rich is aggravating the economic deprivation of the poor.

5.3 Promote Cooperative Solutions to Coordination Failures

In many situations it is not possible to write contracts governing all aspects of the behaviour of parties involved in production and distribution, and under these circumstances markets alone cannot be counted upon to generate the most economically efficient outcomes. Markets fail to coordinate activities efficiently (a) when there are inherently common resources to be exploited (e.g., fish populations on open waters), (b) when there are important strategic complementarities in investment (e.g., basic research and development, or infrastructural services that can benefit many different enterprises), (c) when residual income rights are not aligned with control rights over an activity (e.g., when land control rights are such that actual cultivators do not receive much of the gains from improvements in agricultural productivity), and (d) when contracts are incomplete or unenforceable, as is often the case in labour and credit markets.

Institutional and policy changes that address such coordination problems are often difficult to achieve, because they are bound to involve winners and losers, and compensatory side payments to the losers are difficult to guarantee ex ante and to effectuate ex post. Productivity-enhancing cooperative institutional and policy changes are especially difficult to bring about in contexts of asymmetrical bargaining power linked to an unequal distribution of wealth—and even more so if the richer and poorer parties are also divided largely along lines of ethnic identity. Richly endowed private parties, such as large landholders, can gain from a cooperative solution to the production and distribution of capital-complementing inputs—e.g., large-scale irrigation works to supply water—but such parties also often have private exit options that are at least as attractive—e.g., alternative water sources such as tube wells. In contexts of high inequality, therefore, well-endowed elites will be tempted to forego cooperative community-wide or governmental solutions to coordination in favour of more profitable private solutions.¹⁵

Over and above the specific kinds of incentive problems just discussed, a high degree of inter-individual and/or inter-group inequality may well impede economic performance more generally by obstructing the development of productivity-enhancing norms and institutions that attenuate problems of coordination failure. Ordinary market transactions, as well as complex multi-party economic projects, work much more smoothly the more that the individuals and groups involved can count on one another’s honesty, trustworthiness, and cooperative behaviour. In the absence of widespread norms of trust and cooperation, substantial resources must be devoted to monitoring, supervision, and contract enforcement in order to assure that the terms of a market transaction are respected or that inter-related economic activities are well coordinated. But it is difficult to develop and maintain norms of trust and cooperation in a society characterized by (a) large economic disparities between the rich and the poor and/or (b) multiple ethnic groups whose differences in economic and social status lead members of particular groups to distrust members of other groups.

The economic variable primarily at issue here is wealth, and in particular often land ownership. The configuration of distributional inequality that is most likely to impede cooperative solutions to coordination problems is accentuated inequality among individuals at the upper end of the distribution, which is the prime source of asymmetrical bargaining power. However, inequality across the entire distribution can contribute to the suspicion and distrust that undermine cooperative norms, and this applies to inequality both among individuals and among groups.

6 Social Arguments

In several ways economic inequalities tend to generate significant social costs that are not reflected in conventional measures of economic well-being. Greater economic equality can reduce such social costs by contributing to the achievement of the following objectives:

6.1 Improve the Health Status of Much of the Population

Better health—e.g., lower infant mortality and greater longevity—is for obvious reasons correlated with higher income and wealth, so the worst health outcomes in any society are to be found among the lower economic classes. But there is a plausible theoretical argument, as well as some contested statistical evidence, to the effect that the average quality of health in a population varies negatively with the overall degree of economic inequality—independently of the extent of poverty (measured in absolute terms).¹⁶

The argument focuses on stress as a key intermediate variable (see Wilkinson and Pickett 2009, Chaps. 3 and 4). Chronic stress compromises the immune and cardiovascular systems and thereby increases vulnerability to many diseases. Stress is increased by low social status as well as by poor quality of early childhood experience. There is evidence that the kinds of stress which have the greatest effect on a person’s stress level are “social evaluative threats”, such as threats to self-esteem or social status, in which others can negatively judge one’s performance.¹⁷ Differentials in individual or family income and wealth have a significant impact on whether people feel valued and appreciated, rather than disrespected or stigmatized. Differentials in the social status of one’s ethnic group have a similar significant impact. Increased economic inequality and social hierarchy thus serve to raise competitive stakes and personal anxieties about one’s worth, exacerbating stress levels and associated pathologies for people in all except perhaps the very highest strata of society.

It also stands to reason that the ability of a society to limit the spread of disease and other public health problems is impaired to the extent that a relatively poor segment of the population lacks access to good nutrition and health facilities. However, the severity of this problem cannot readily be attributed to the extent of (relative) inequality; it can be alleviated by reducing absolute poverty and/or by improving public health services for the poor.

The distributions of all of the economic variables under consideration—income, consumption, wealth, and access to public services—affect social as well as economic status and therefore also health outcomes. Because status insecurity is especially likely to result from differential treatment based on ethnicity, inequalities across ethnic groups are likely to be at least as critical—and perhaps more critical—than inequalities among individuals in affecting health outcomes. The configuration of inequality that is most relevant to health outcomes is the entire distribution, since differentials in any part of it imply differentials in status that can generate stress.

6.2 Promote a Better Quality of Life by Reducing the Over-Valuation of Purchasing Power

First highlighted by Hirsch (1977), “positional goods” are products or services that are inherently limited in supply, so that their acquisition by any given individual depends not simply on that individual’s own resources, but on how those resources compare with the resources of other individuals interested in acquiring them. Examples include an apartment in a desirable building, a house in a desirable neighbourhood, and an education at a top-tier school.¹⁸ Any product whose value to the consumer depends at least in part on distinguishing him/her from other consumers not able to acquire it—for example, a house that is larger, or a car that is flashier, or jewellery that is rarer than most others—also has some positional character, because the ability to acquire it depends on how one’s ability to pay for it ranks among all those who desire it. Such examples make it clear that positional goods are a matter of concern mainly for the higher income classes of a society, and for those who aspire to join their ranks.

Positional goods lead people to over-value purchasing power, because more of it not only increases one’s ability to acquire ordinary (non-positional) goods, but improves one’s rank in the competition for positional goods. Yet the putative improvement in rank is most likely to prove illusory, because the desire to improve one’s relative position tends to drive all prospective consumers of a positional good into a competition in which the generation of more and more purchasing power ends up just maintaining one’s relative position—a kind of “arms race” that increases nobody’s chances of acquiring the good (see Frank 2011, Chap. 5).

That those who can afford to compete in arms races for positional goods simply end up paying more for such goods hardly constitutes a significant social problem. The valuation of extra income or wealth for adding to relative as well as to absolute purchasing power, however, can indeed generate significant social costs, because it systematically favours individual private solutions over more efficient collective responses to social needs that often pose significant coordination problems. For example, it leads the rich to overvalue the loss of purchasing power they suffer when paying taxes and thus to undervalue public facilities that can be financed from taxes, which can thwart socially optimal choice with respect to the supply of public services in areas such as health, education, transportation and security. If all of the rich had to pay somewhat more taxes, each individual rich person would not suffer any loss of access to positional goods; but they would all gain from better public facilities. For another example, to the extent that people are interested in positional goods, they will over-value their wage or salary income relative to better working conditions, fewer work hours per week, or more vacation time; but the higher income they earn will bring little advantage in the competition for positional goods, and they will forfeit the possibility of trading some income for better working conditions or more leisure.

In general, the over-valuation of purchasing power for positional reasons imparts a bias against non-material amenities whose provision depends to a significant extent on collective decision-making and public regulation or provision—e.g., clean air, clean water, healthy recreational opportunities, safe working environments, as well as leisure time. A reduction in inequality can help to mitigate the social costs of undersupply of such amenities by reducing the intensity of competition for positional goods among those who can afford it. Probably the most efficient way to achieve this is to establish or expand progressive taxation of consumption expenditures, since that would most directly affect incentives to compete for positional goods among those most able to do so.

The economic variables at issue in the over-valuation of purchasing power include wealth and income, but consumption is most fully implicated. Since it is consumption among those who are relatively well-off that most affects the competition for positional goods, it is inequalities at the upper end of the distribution that are of primary concern. And it is economic class inequality, not ethnic group inequality, which matters, because the ability to compete for positional goods and the likelihood of over-valuing purchasing power is hardly likely to depend on a person’s ethnicity.

6.3 Promote Greater Ecological Sustainability

It is now widely understood that economic growth around the world is slowly but surely warming the earth, with potentially dire long-run consequences, because of rapidly growing emissions of greenhouse gases associated with predominant use of fossil fuels as a source of energy. In the coming decades it will be increasingly necessary to reduce emissions of such gases, which will have to be achieved by some combination of reduced growth in the demand for energy and reduced dependence on fossil fuels as a primary source of energy. These requisites of ecological sustainability put a premium on shifts of production and consumption away from energy-intensive material goods and toward less energy-intensive goods and non-material amenities, including in particular leisure time.

Economic inequality can make it more difficult to develop an ecologically sustainable level and pattern of economic activity in several ways. First, as noted in the previous section, high-end inequality is likely to lead to a bias against non-material amenities whose provision depends largely on collective decision-making and public regulation or provision; and ecological sustainability is precisely such an amenity. Controlling emissions of greenhouse gases poses a classic collective action problem—at the international as well as the national level; and government regulation (whether by taxes and subsidies or by quantitative controls) will clearly be required to address the challenge.

A second way in which inequality poses an obstacle to ecological sustainability is that it tends to encourage ever higher levels of consumption via a “demonstration effect” (a term first used by Duesenberry 1949; see also Nurkse 1953). When the consumption levels and patterns of the rich are widely publicized to the general public, they exert a considerable influence on those who can afford in some degree to emulate them. (Those who are poor, in absolute or relative terms, will of course have good reason and justification to try to consume more, without any outside encouragement.) The kinds of consumption that are most likely to be well-publicized are those that are advertised by profit-making firms, so the non-material amenities enjoyed by the rich will have much less of a demonstration effect than the marketed goods and services they buy. Thus the demonstration effect will operate mainly on the middle or (more significantly) the upper middle economic classes, stimulating them to consume at higher levels and with less regard to non-material amenities that are comparatively energy-non-intensive.

Finally, the need to promote ecological sustainability speaks to the debate between those who advocate economic growth as the best way to reduce poverty and those who argue that some kind of redistribution from rich to poor is the best way to do so. If we rely on economic growth to reduce poverty by raising living standards across the board, with little or no change in the degree of inequality, then there will be more production and consumption—and more energy use and environmental destruction—than if poverty is reduced by policies that have the effect of shifting some consumption from the rich to the poor in the context of a lower overall growth rate. Distributional considerations are likely in any event to loom large in the formulation of an ecologically sustainable energy policy. That is because people are unlikely to cooperate in the effort to reduce energy and fossil fuel use if the economic burden of their lesser availability is not seen to be widely and fairly shared. It will be hard to generate public support for the substantial changes needed to cope with global warming if the rich are able to continue enjoying an environmentally costly lifestyle while the poor bear a disproportionate burden of higher energy costs.

The economic variable of primary concern with respect to ecological sustainability is clearly consumption; insofar as greater wealth and income does not get translated into consumption, it is not of ecological concern. What matters is inequality of consumption among individuals, not across groups, since the likelihood of over-valuing purchasing power and the strength of the demonstration effect does not depend on a person’s ethnicity. And the configuration of distributional inequality that is most at issue is accentuated privilege at the upper end, which is most likely to stand in the way of the kinds of regulatory policies needed to reduce emissions of greenhouse gases.

7 Conclusion

In this paper I have considered both the reasons why economic inequality can be a matter for concern and the forms of inequality that are principally at issue. The major reasons for concern about economic inequality are reflected in the ten different arguments—two of them moral, two political, three economic, and three social—that I have addressed in the preceding four sections of the paper. These arguments differ with respect to the form of inequality that is the source of concern, along three dimensions: (a) the economic variable(s) whose unequal distribution is at issue; (b) the relevant distributional entity, i.e., whether economic class inequality or ethnic group inequality is most salient; and (c) what part(s) of the unequal distribution are the most problematic, i.e., is the problem primarily poverty at the lower end, or privilege at the upper end, or bipolarization; or is the entire distribution implicated?

Table 4.1 sets out the results of my analysis in the form of a matrix, with a row for each of the ten arguments for reducing inequality and three columns in which the form(s) of inequality relevant for each argument are shown. Inequality factors printed in bold font are more significant than those that are not. The economic variable whose distribution is most often at issue is clearly wealth, which figures in nine of the ten arguments; but each of the other three variables is relevant for roughly half of the arguments. Economic class differences are salient to every argument; but ethnic group differences are also salient in six of them. The configuration of inequality that is most often implicated is privilege. Poverty, as well as the entire distribution, each figure in three arguments; bipolarization figures in just one.

Table 4.1 Results of the analysis

In showing the form(s) of economic inequality that are most relevant for each argument, the inequality factors shown in the cells of Table 4.1 also point to the kinds of inequality reducing policies that would be needed to improve societal well-being in the way suggested by the argument. Quite different policies are required to reduce privilege than to reduce poverty; different policies are called for to reduce class inequality than to reduce group inequality; and of course the appropriate policy for reducing economic inequality will also depend on which economic variable(s) are primarily at issue. A careful examination of alternative policies to achieve the objectives expressed by each of the arguments for reducing economic inequality is beyond the scope of this paper. What I have tried to show here is simply that optimal choice of policies to reduce economic inequality should be informed by specification of the particular societal objectives that one is trying to achieve through the reduction of inequality.