THE POLITICAL PSYCHOLOGY OF INEQUALITY

To begin with, we should acknowledge that the parsimonious economic calculus posited in much of the political economy literature does not provide a straightforward basis for predicting preferences about redistribution. It is entirely plausible, of course, that the have-nots in a given society will believe that they stand to gain from measures that transfer income or assets from the haves. But a number of considerations may lead them in other directions. They may be unaware of the size of the gaps in wealth, or they may believe that these gaps are justifiable rewards for effort and skill. Even if they know and care about inequality, they may doubt that the government is capable of redressing distributive injustices, or they may care even more about other, non-economic issues.

Findings from opinion surveys and other comparative studies provide considerable evidence on the cognitive and normative impediments to strong demands for redistribution. Survey research in both the United States and other Organisation for Economic Co-operation and Development (OECD) countries show that publics were generally unaware that income gaps had grown wider since the 1970s (Bartels Reference Bartels2008; Kenworth and McCall Reference Kenworth and McCall2008). In many developed countries, moreover, the salience of income redistribution appears to have been muted by “post-industrial values” (Inglehart Reference Inglehart1997), by racial prejudices, or by religious beliefs that cut across class lines. Scheve and Stasavage (Reference Scheve, Stasavage, Shapiro, Swenson and Donno2008) provide evidence from the United States and Western Europe showing, in effect, that Marx was right, or at least on the right track: religiosity can function as an emotional substitute for relief that might otherwise be provided by social protections against life-cycle and economic risks.

Concerns about the unequal distribution of income can, of course, also be highly salient in some circumstances. But as noted, even when this is the case, doubts about government capability may deter support for progressive policies. Feldman (Reference Feldman2003) reaches this conclusion with survey research from the United States; and although survey evidence from developing countries is more limited, other studies show that this factor may be at work there as well. For example, Mares (Reference Mares2005) shows that labor demands for social protection is contingent on the capacity of the state to deliver. And Soifer (Reference Soifer2009), building on Boix (Reference Boix2003), shows that economic inequality is destabilizing only in countries where the state capability is relatively high.

Cultural beliefs derived from different national experiences, moreover, are also likely to shape beliefs about the role government should play in reducing inequality. For example, Corneo and Gruner (Reference Corneo and Gruner2002) show that support for redistribution is stronger in post-communist countries than in Western democracies, a likely reflection of the socialist legacy. Among Western democracies, support for redistribution also goes in a direction consistent with common understandings of cultural tolerance for inequality: support is weakest in the United States and strongest in Norway, with Australia, Canada, New Zealand, and West Germany following in that order.

Beyond these factors, we need to consider that the inter-temporal bases of preferences about redistribution. Both expectations about opportunities and the actual experience of upward mobility do appear to affect the way people think about inequality. Again, however, these effects are not entirely straightforward.

Benabou and Ok's (Reference Benabou and Ok2001) well-known POUM (prospect of upward mobility) hypothesis is frequently cited as an explanation for weak support for progressive taxation in the United States, and comparative evidence from other OECD countries provides support for this proposition (Alesina, Di Tella, and MacColloch Reference Alesina, Di Tella and MacColloch2004). Survey research from Latin America and other developing countries shows similar results: positive expectations are associated with higher levels of personal happiness and greater tolerance for existing levels of inequality (Graham and Pettinato Reference Graham and Pettinato2001; Kaufman Reference Kaufman2009). The puzzle for comparative analysis is that these beliefs appear to have only a limited connection with underlying social realities. American society is not in fact more mobile than other wealthy countries, and in Latin America, some of the highest percentages of optimism are found in highly stratified societies (Graham and Pettinato Reference Graham and Pettinato2001). This disconnect between expectations and actual patterns of social mobility does not invalidate the POUM hypothesis, but it does leave unresolved the question of where and how these expectations are formed in the first place.

The actual experience of upward mobility, finally, poses a different kind of paradox: the frustration that arises from the fact that, above a fairly low level of income, people tend to evaluate their well being less in terms of absolute improvements in their own situations than in terms of comparison with others. Surveys summarized in a recent study of the Inter-American Development Bank (Lora Reference Lora2008) document dissatisfactions that appear to arise from chasing an ever-receding horizon of consumption standards set by wealthier reference groups. Their findings are consistent with the results of panel studies conducted in Peru and Russia. These identify a large group of “achievers” who—despite significant improvements in income—remain frustrated by the impossibility of keeping up with the Joneses (Graham and Pettinato Reference Graham and Pettinato2001). One important implication is that class resentments, and possible demands for redistribution, may be especially pronounced among upper-middle-class individuals who would be most inclined to measure their achievements against that of the rich.

In sum, it seems unlikely that we can generate a highly parsimonious economic model of the way individuals perceive and evaluate the distribution of income in their society. A wide variety of considerations—both economic and non-economic—will enter the way people calculate their interests and form normative judgments, and there is a good chance that these considerations will differ quite substantially across societies and over time.

BRINGING POLITICAL SOCIOLOGY BACK IN

Insights from political sociology can help advance our understanding of reactions to inequality. In this respect, I highlight two important issues that might guide further research: the intersection of economic interests with other group affiliations, and the dynamics of change in political institutions.

First, we can expect that reactions to economic inequality will depend in part on the way individuals—the core units of political economy approaches—are embedded in a broader framework of relationships to neighborhoods, ethnic and religious communities, or partisan groups. Although research into the nature of these ties is hardly new, the implications for the way they affect the redistributive calculus and behavior of individuals warrants further investigation, especially in developing countries. A core hypothesis is that the salience of redistributive issues depends on the extent to which perceptions of economic interests overlap with ethnic, religious, geographic, or other non-economic identities.

Identities are socially constructed, and their meanings and importance to individuals is subject to changing circumstances. The strength of such identities, moreover, can be diluted by crosscutting loyalties that weaken attachments to particular social groups. But ongoing patterns of exclusion or opportunity hoarding can also harden inter-group boundaries into “durable” inequalities (Tilly Reference Tilly1998). Where inter-group boundaries are clearly demarcated and exit is costly, it follows that the most intense pressure for redistribution will derive less from the hierarchical ordering of individual shares in income or wealth than from the “horizontal” inequalities among such groups (Stewart Reference Stewart2000). Individual members are more likely to link their own prospects for material well being to the economic and political position of their group. They are more likely to be aware of changes in these positions and more available to political entrepreneurs who seek to exploit solidarities and resentments to win votes or mobilize collective action.

In recent decades, one important source of inter-group conflict has stemmed from the uneven territorial impact of globalization. Where new sources of trade and investment have widened the gap between rich and poor regions, geographic divisions have provided important foundations for political protest over distributional inequities. In some cases, such conflicts can be mitigated by internal migration, economic transfers, or power sharing. But inter-regional rivalries can also be especially intense when geographic differences overlap with other, ethnic or religious sources of horizontal inequality. Examples abound: the bitter conflict between Bolivia's lowland and highland regions and struggles over the control of increasingly valuable petroleum resources in Iraq and Nigeria are but a few of these.

It is important to emphasize that these regional rivalries, as well as other inter-group conflicts, involve struggles over political power as well as over access to economic resources. The stakes for competing regional leaders include not only access to material benefits for themselves and their supporters, but the division of authority between central governments and sub-national units as well the projection of regional authority at the national level.

Contestation over de jure political authority is also central to Acemoglu and Robinson's (Reference Acemoglu and Robinson2006) model of democratic transitions. The establishment of democratic institutions is a means through which elites can commit credibly to the continuation of redistributive policies over time. The more general implication is that the dynamics of institutional change should be a core feature of a research agenda on the politics of inequality. The distributive stakes are high, Acemoglu and Robinson argue, because the power acquired through political institutions—whether regime type, constitutional design, or other de jure rules of the game—increases the odds that the groups with advantages today will continue to have them tomorrow.

Several lines of research follow from this point. Boix (Reference Boix2003) and Acemoglu and Robinson (Reference Acemoglu and Robinson2006) focus on how conflicts over the distribution of wealth between groups or individuals affect demands for changes in political institutions. But the causal arrows can go in the opposite direction as well: precisely because of the inter-temporal implications of institutions, exogenous changes in the political rules of the game can all trigger intense inter-group conflicts over prospective changes in the distribution of wealth.

Institutional changes can be set into motion by a variety of factors that are initially independent of distributive interests. For example, world wars and decolonization had a profound impact on definitions of citizenship, group identities, and intensified contestation over economic and political power (Horowitz Reference Horowitz1985). The collapse of the Soviet empire and the reconfiguration of territorial boundaries had similar effects. In these cases and others, major institutional transformations upset the previous inter-group equilibrium and set fierce distributive struggles in motion.

Paradoxically, in many developing countries, the third wave of democratic transitions has had a weaker impact on distributive politics than might have been anticipated initially (Kaufman Reference Kaufman2009). There are many reasons for this, including the economic constraints, institutional legacy of past welfare systems, or the weakness of left parties (Haggard and Kaufman Reference Haggard and Kaufman2008; Huber et al. Reference Huber, Nielsen, Pribble and Stephens2006). But the intensity of distributive conflicts—and corresponding threats to political stability—may also depend on the strength of group ties and the distribution of power among them. In countries such as Bolivia, South Africa, and Malaysia, for example, power is divided between economic elites that control the main sources of wealth and political leaders representing poor ethnic or racial majorities. Such arrangements are not likely to constitute a stable equilibrium, at least not without some reduction in existing levels of inequality. But conflicts over redistribution will, of course, be fraught—especially if economic assets are relatively immobile (Boix Reference Boix2003, Reference Boix2008).