The Autocratic Middle Class: How State Dependency Reduces the Demand for Democracy

By Bryn Rosenfeld

How middle-class economic dependence on the state impedes democratization and contributes to authoritarian resilience

Conventional wisdom holds that the rising middle classes are a force for democracy. Yet in post-Soviet countries like Russia, where the middle class has grown rapidly, authoritarianism is deepening. Challenging a basic tenet of democratization theory, Bryn Rosenfeld shows how the middle classes can actually be a source of support for autocracy and authoritarian resilience, and reveals why development and economic growth do not necessarily lead to greater democracy.

In pursuit of development, authoritarian states often employ large swaths of the middle class in state administration, the government budget sector, and state enterprises. Drawing on attitudinal surveys, unique data on protest behavior, and extensive fieldwork in the post-Soviet region, Rosenfeld documents how the failure of the middle class to gain economic autonomy from the state stymies support for political change, and how state economic engagement reduces middle-class demands for democracy and weakens prodemocratic coalitions.

The Autocratic Middle Class makes a vital contribution to the study of democratization, showing how dependence on the state weakens the incentives of key societal actors to prefer and pursue democracy.

• Language: English
• Publisher: Princeton University Press
• Release date: Dec 1, 2020
• ISBN: 9780691209777

Book preview

1

The Autocratic Middle Class

Thus it is manifest that the best political community is formed by citizens of the middle class, and that those states are likely to be well-administered in which the middle class is large, and stronger if possible than both the other classes, or at any rate than either singly; for the addition of the middle class turns the scale, and prevents either of the extremes from being dominant.

—ARISTOTLE, THE POLITICS¹

MUCH HAS BEEN expected of the middle class. In two thousand years of scholarship on democratization—from Aristotle in the fourth century to Acemoglu and Robinson today—two views of the middle class have dominated. On the one hand, the middle class has been seen as a source of social stability. In this view, the middle class favors durable economic institutions and maintenance of the political status quo. Politically and socially, it acts as a conservative force. By moderating the conflicting redistributive demands of rich and poor, it stabilizes existing regimes. On the other hand, the middle class has been cast as an agent of political change and democratization. In scholarship on the the first wave of democratization in the nineteenth century, the second wave of democratizations following World War II, and the third wave of democratic transitions from the mid-1970s through the collapse of communism, the middle classes have been ascribed a critical role (see, e.g., Lipset 1960; Huntington 1991; Acemoglu & Robinson 2006; Ekiert 2010). To date, no comparative scholarship has sought to reconcile these contrasting views.

This is a matter of some urgency. Civic revolution has increasingly become the norm in recent episodes of democratization. The very fact of mass mobilization during transition, as well as its character, appears to have lasting impact on the subsequent quality of democracy (Haggard & Kaufman 2016). Although these popular insurrections are often labeled as middle-class revolutions, our understanding of the constituencies involved as well as their motivations remains limited. Moreover, while the role of the middle classes in mass uprisings against authoritarianism has varied across time and space, no clear consensus has emerged to explain these patterns (e.g., Koo 1991; Jones 1998; Shin 1999; Rueschemeyer, Stephens & Stephens 1992, 272). Recent protests around the world demonstrate the critical importance of understanding how citizens form democratic preferences under autocracy, and thereby the sources of bottom-up pressure for political change.

In Eastern Europe and the former Soviet Union, the collapse of communism inaugurated a period of market reforms, promising social transformation, prosperity, and democracy. The growth of a new, market-based middle class was widely expected to buttress these twin economic and political transitions. The transition from state socialism ended the institution of universal state employment and broadened the scope of economic activities neither initiated nor directly controlled by the state. While these processes reduced the size and influence of the state, they did so unevenly and in unexpected ways, leaving the state as an influential patron of the region’s educated, professional classes.

At the same time, the early post-communist years saw a staggering rise in unemployment, inflation, and poverty across the region. Initially, at least, these conditions produced a sharp differentiation of incomes and the immiseration of the communist-era middle class (Ekiert 2010, 112). During the 1990s, inequality rose most rapidly in the former Soviet states, where market reforms proceeded more slowly than in Eastern Europe, and were least complete (Hellman 1998). Economic recovery took off only a decade later, around 2000, when the region’s economies joined the high growth trajectory of other emerging markets. During this recovery, the post-Soviet middle classes have grown markedly, yet not only, or even primarily, because small business, entrepreneurship, and private enterprise have flourished. Instead, the rise of the middle class has been largely tied to the state. Today, it is clear that market reforms have failed to replace the old state-dependent middle classes of the communist era with an independent, entrepreneurial middle class. Moreover, a new state-dependent middle class has emerged, only deepening existing cleavages.

This book seeks to expand our grasp of authoritarian resilience and bottom-up pressures for democratization in states where economic growth is increasing the size of the middle class. Contrary to the conventional expectation, I ask why and under what conditions growth of the middle class may not increase popular pressure on regimes to democratize. To answer this question, I turn to a wide array of survey data on the political preferences and behaviors of the middle classes in the post-communist countries as well as qualitative evidence collected during nine months of fieldwork in the region. In contrast to existing scholarship, this book emphasizes that a variety of development strategies can drive an expansion of the middle class. These strategies differ in their effect on the formation of democratic constituencies. There are multiple pathways to the middle class, and not all of them, I argue, lead to greater support for democracy.

In this book, I tell the tale of two middle classes, with differing degrees of state dependency, that are nonetheless treated as one homogenous group by existing theories. The story focuses on contexts that combine autocracy with an economically interventionist state. These conditions are not rare. Pervasive public sectors are a feature of politics in the Middle East, Africa, Asia, Latin America, as well as the former Soviet Union (e.g., Hertog 2013; Diamond 1987; Weyland 1996; Gervasoni 2010; Oliveros 2016; Darden 2008). These conditions are widespread in resource states, but they are also present in post-socialist countries both with (e.g., Azerbaijan) and without oil (e.g., Belarus).

The post-Soviet middle classes typify the type of divided middle class that develops under autocratic state institutions and extensive state economic engagement. Following the collapse of communism, an important new cleavage appeared between the old, state-dependent middle class and the nascent middle class of the new market-based economy (see also, e.g., Ekiert 2010). Thanks to statist economic policies and a decade of economic recovery, the state-dependent sectors of the middle class in the former Soviet Union are growing. Civil servants and other government budget-sector employees are also increasingly well paid. Where formal wages remain low, the state has turned a blind eye to bribe-taking and graft by those on its payroll, ensuring that corruption helps to compensate. The upshot is that lucrative white-collar positions in public-sector enterprises, banks, and financial service firms as well as the state administration have become among the clearest pathways to the middle class across the former Soviet states. As society has grown wealthier, even state teachers and doctors find it increasingly lucrative to exploit their official positions (INDEM 2005). This is the new state-dependent middle class of post-communist quasi-capitalism.

The process of middle class formation now transpiring in the post-Soviet states recalls what Barrington Moore called the feudalization of the bourgeoisie. While a number of influential approaches to democratization—among them modernization theory, its values-based variants, and redistributive theories—expect the middle class to be a force for democracy,² an earlier macro-sociological literature understood that when the middle class maintains close ties with the state it can be a deeply conservative political force. This book problematizes recent theories of democratization, which largely ignore the implications of how growth of the middle class is achieved. Indeed, this book’s central argument is that failure of the middle class to gain economic autonomy from the state stymies support for political change and contributes to authoritarian resilience.

This study fills an important gap in the literature. While the middle classes have often been seen as a linchpin in successful democratic coalitions, very little work to date has provided a detailed empirical examination of this group’s political preferences. As Ansell & Samuels (2014, 46) write, Scholars have paid insufficient attention to the concrete interests of the … middle classes in the study of regime change. The distinctive contribution of this book is to show concretely, at the individual level and for a broad set of cases, how state-led development produces middle classes that are beholden to autocratic regimes. The findings thus contribute to our understanding of why some countries are democratic, while others are not. They also speak to long-standing debates in comparative politics about the role of class actors in democratic transition and participation in contentious politics, as well as a newer literature on the rise of state capitalism in contemporary autocracies. In addition, this research advances our understanding of the consequences of state-directed development. While this subject has thus far been studied primarily at the institutional and macro-sociological level (Gerschenkron 1962; Bellin 2002; Kohli 2004; Wengle 2015), this book focuses on how statist strategies of development structure the incentives of individual actors to side with autocracy. It also provides new insights into the persistent puzzle of development without democratization.

This introductory chapter is organized as follows. I first lay out my argument in brief. I then situate it within existing scholarly debates. My focus is at the micro-level, on how state dependency shapes the incentives faced by the middle classes to oppose the extension of democratic institutions. I next provide an overview of my empirical strategy. I then conclude with an outline of the book’s remaining chapters.

Argument in Brief: The Autocratic Middle Class

The question of how society’s class structure affects the prospects for democracy has long been a core preoccupation in the study of politics. One of the most enduring arguments in this literature holds that growth of the middle class gives rise to democratization. This view is echoed in many canonical approaches to democratization, including political economy’s redistributive theories, modernization theory, and its values-based variants. These theories expect the middle class—defined variously in terms of income, education, or occupation—to play a key role in democratic transition.

Yet remarkably little systematic research examines middle-class attitudes toward democracy in autocratic settings. As the developing world’s middle classes have grown rapidly over the past two decades, largely in nondemocracies, this gap has become even more significant. Will these growing middle classes turn the scale, enhancing their countries’ prospects for democracy? Or standing the conventional logic on its head: Might it be the case that certain modes of state-supported middle-class growth, in fact, delay democratization?

This book argues that a middle class whose status depends on public employment for an authoritarian state is often antithetical to democracy. Drawing lessons from the post-communist countries, it sheds light on how the economic institutions of state employment benefit autocrats, helping them to secure the support of key middle-class constituencies. While existing theories expect that growing middle classes will confront old networks of patronage and privilege, they ignore crucial variation in the extent to which the middle classes are stakeholders in existing autocratic systems. Indeed, in many contemporary autocracies, expanding opportunities within the state sector—within the state bureaucracy, public institutions, and state-owned enterprises—supply the principal avenues of mobility into the middle class and drive its numerical expansion.

Under autocracy, an expansive public sector can dampen democratic demands. How a person gains and maintains their place in the middle class affects their expected benefit from democracy and willingness to challenge the political status quo. The distinction between the middle classes of the state and private sectors is thus real, politically significant, and becomes the locus for divergent economic and political interests in autocratic systems. This distinction forms the basis for access to privileges, benefits, and at times even differing status before the law. Where the state supplies the principal avenues of social mobility into the middle class, a self-conscious, democratizing middle class fails to form. Material incentives based on employment status and workplace mobilization all militate against the expression of democratic demands. Rather, selective incentives are used to mobilize growing middle classes in support of autocracy.

I connect the middle classes’ reticence about democracy to the power of authoritarians to bestow or withhold benefits that are contingent and could be diminished or disrupted by change in the political control of the state. Some of these benefits are formal, like jobs, employment protections, shorter working hours, access to healthcare, and loans on advantageous terms from state banks. Other benefits are informal, though rooted in an official position, like opportunities to solicit bribes and kickbacks and preferential treatment by other state institutions. In turn, the power of state selective incentives varies with an individual’s exit options. These options are limited when an extensive public sector employs a preponderance of the middle class, crowding out private-sector alternatives. Limited exit options deepen dependence on the state. By tying future benefits to regime continuity, state economic engagement that concentrates rents in the public sector creates its own middle-class constituencies, encouraging opposition to democratization and dividing potential democratic coalitions.

If we survey the developing world today, it is possible to tell a tale of two middle classes: one whose economic opportunities and life chances are owed directly to the state, and the other whose economic livelihood is less directly dependent. These categories roughly correspond to two pathways into the middle class: one through the state, the other through the private sector. While the latter group may fear predation by a large authoritarian state, the former group has the state to thank for its upward mobility and middle-class status. White-collar workers at state and quasi-state enterprises, teachers, doctors, and other public-sector professionals, as well as those employed in the state administration, all typify what I call the state middle class. These groups comprise a middle class that, I argue, helps to stabilize autocracy and has been largely ignored by existing theories.

While reticence about democracy often goes hand in hand with greater satisfaction with the status quo and support for the incumbent autocrat, this need not and, indeed, is not always the case. Even where the middle class expresses grievances with the status quo or is not generally supportive of the current regime, it will remain unsupportive of democratization if it associates transition with diminished life chances. The latter case describes Ukraine after the Orange Revolution. The former case describes Russia during the post-election protests preceding Putin’s third term as President. Each of these cases as well as cross-national evidence shows that reticence about democracy is not merely a function of support for a particular autocrat or satisfaction with the status quo.

This study adopts a sociological definition of middle class emphasizing human and social capital. The definition, described in detail in the next chapter, captures the distinction between highly educated white-collar and professional strata versus less educated routine and manual laborers. This approach is notably distinct from a normative view of the middle class as a carrier of democracy, as synonymous with the capital-owning bourgeoisie, or as an exclusively income-based category. The middle class thus defined most closely resembles the middle class of educated professionals in modernization theory and its values-based variants (e.g., Lipset 1959; Welzel & Inglehart 2008). It also shares with Collier & Collier’s (2002) middle sector an emphasis on members of a broad range of occupational groups between the working class and economic elite. At the same time, it is broader than Moore’s (1966) bourgeoisie and empirically has above-median income like the middle class of elite competition theories (Ansell & Samuels 2008). This approach facilitates careful comparison with these other theories and a sound basis for empirical investigation. Education and occupation are theoretically well-grounded determinates of class and socioeconomic status (SES), readily available in most cross-national surveys, well correlated with each other empirically, and much more reliably measured than income—for which they are often used as proxies.³

Scope of the Argument

The argument just summarized is not specific to particular types of autocracy, with the potential exception of rightist dictatorships.⁴ The cross-national sample I employ in chapter 3 covers virtually the full spectrum of authoritarian regime types from highly repressive (Uzbekistan) to wavering on the cusp of democracy (Georgia). Moreover, the cases I examine in greater depth in subsequent empirical chapters range from hegemonic single party (Kazakhstan) to relatively pluralistic competitive authoritarianism⁵ (Ukraine). Although there are no right-wing dictatorships among the cases considered in this book, such regimes are probably beyond the scope of my theory. If faced with the choice between substantial neoliberal economic reforms under a rightest military junta and a populist democrat advocating leftist policies, the intertwining of economic and political considerations regarding transition would likely lead the public-sector middle class to side with democracy. This choice set describes several transitions to democracy in Latin America, including in Chile, Brazil, and Argentina.

That said, the argument’s intended scope is not limited to the post-communist region. Rather than hinge on historical legacies of communism, in line with Bellin (2000), I expect the argument to travel to other late-developing autocracies with high levels of state economic engagement and to apply whether the logic of the state is developmental or patrimonial. Over the past two decades, according to World Bank criteria, the middle class expanded rapidly across the developing world. At the same time, a wave of market backlash and nationalization has returned to the fore the notion of state control over strategic sectors and ideas implicit in the developmental state tradition.

Although many studies, including this one, show that economic and political reforms are intertwined in the minds of post-communist citizens, the book’s central argument does not hinge primarily on the association of democratization with neoliberalism. Even where neoliberal reforms are not expected to accompany the triumph of democratic forces, middle-class groups dependent on the state may be reticent about democracy for at least two additional reasons: (1) they may fear that being associated with the preceding regime will bode poorly for them following transition, and (2) they may expect that rising attention to the rule of law under democracy will reduce opportunities to earn informal rents.

Note: *Wavered on the cusp of democracy during the period under study. †Excluded from survey samples.

Descriptive Data on the Post-Communist Region

Table 1.1 lists the post-communist countries that I examine in this book and breaks them down by regime type. Throughout the text, I use the term post-communist to refer collectively to all of the countries in table 1.1 and post-Soviet or formerly Soviet to refer to the fifteen republics that once comprised the Union of Soviet Socialist Republics (USSR). Of these, my primary interest is in the nondemocracies, plus states that wavered on the cusp of democracy during this period (e.g., Georgia and Ukraine). The three Baltic States (Latvia, Lithuania, and Estonia) and Moldova, which democratized earlier and maintained stable democracies over this period, are thus excluded when I refer to the post-Soviet autocracies. Due to the absence of reliable survey data, Turkmenistan is not included in this book’s empirical analyses.

This device does not support SVG

FIGURE 1.1. Public employment in 27 post-communist countries by regime type. Data source: European Bank for Reconstruction and Development (EBRD) Life in Transition Survey (LiTS), 2006 (all post-communist countries).⁶

Figure 1.1 provides descriptive statistics on public employment in the nondemocracies of the former Soviet Union and the democracies of Central and Eastern Europe. Recalling that these countries once shared the institution of universal state employment, it is not difficult to see why state employees in the autocracies might be worried about democratic transition.⁷ As the first bar on the far left of the plot shows, the public sector is a major employer in the post-Soviet autocracies. Even after fifteen years of privatization, public employment continued to comprise just under half of total employment in the nondemocracies at the time these data were collected in 2006. In the democracies, meanwhile, it comprised only about a third of total employment. Turning to the right side of the figure, we see that the post-Soviet middle classes remain more heavily dependent on the state than other groups. Indeed, the public sector employs roughly seven in ten middle-class individuals (71%) across the post-Soviet countries (vs. 58% in the democracies). The share of working middle class employed in the state sector is highest in Azerbaijan (84%), Belarus (81%), and Uzbekistan (79%), and considerably lower (though still a clear majority) in Russia (59%). These figures make a strong case for the importance of public sector institutions in shaping the life chances and career prospects of the post-Soviet middle class.

Finally, it is worth noting that in terms of household expenditures, state employees in the former Soviet countries are better off than others. In fact, median household expenditures in the autocracies are 17 percent higher for state employees than for those employed outside the state sector. Given that the official wage gap favors the private sector in many former Soviet countries, these expenditure data strongly suggest that state employees in the autocracies benefit from informal rents (e.g., Sharunina 2013; Gorodnichenko & Peter 2007). By contrast, equivalent data from the democracies suggest virtually identical median household expenditures by state and non-state employees.⁸ As long as the public-sector middle class remains large and enjoys access to illicit income, it will have fewer incentives to favor democratization and associated political transparency.

The Middle Class and Democratization

This argument, though intuitive, is at odds with much of what has been written on the politics of the middle class. As the middle classes expand across the developing world, observers have frequently assumed that they will demand better governance and support democratization. What do existing theories tell us about the mechanism through which development, by enlarging the middle class, will lead to democratization? This section surveys existing explanations for the political preferences of the middle class and relates the book’s theoretical framework to several newer approaches and recent literature on authoritarian resilience.

Conventional wisdom holds that one of the signal developments en route to liberal democracy in Western Europe was the rise of the middle class. That reading of history continues to shape observers’ expectations about the likely political influence of growing middle classes in the developing world today. Fukuyama (2014) provides one account: In nineteenth-century Europe, wages began to rise, improving workers’ standards of living. At the same time, demand increased for skilled professionals like engineers, accountants, and lawyers. As a result, workers became more affluent, came to own property, and became more highly educated. They also became more politically conservative: more likely to vote for political parties that could protect their privileges rather than ones pushing to overturn the status quo (415).⁹ By the second half of the twentieth century, the median voter had become a middle class individual with a stake in the existing system (417). This process is said to have shored up democracy in Europe and is thought to be happening across many parts of the developing world today.

However, the conditions greeting the rise of the middle class were notably different in the early democratizers, and especially in England (Landes 1998). First, the state apparatus, as van de Walle (2014) has observed, was much smaller historically. Government expenditure as a share of national income was a mere fraction of what it is in the modern era. Second, while autocracy is the dominant regime type among developing countries today, England had been experimenting with the institutions of democracy since the first half of the 19th century. Accordingly, the status quo bias exhibited by Europe’s middle classes as they became more politically influential favored increasingly democratic political systems. The post-war expansion of the public sector only reinforced western Europe’s democratic middle-class constituency.

Moreover, growth of the middle class has not always enhanced the prospects for liberal democracy, even in Europe. Although the English case is often taken as paradigmatic of the first-wave democracies, other first-wave cases suggest a more complex view. As an earlier sociological literature highlights, monarchic France and absolutist Germany exemplify cases where the rising middle classes actually reinforced authoritarianism. Moore (1966), for instance, argues that the French bourgeoisie’s dependence on the crown delayed the emergence of full democracy. And Dahrendorf (1967) observes that Germany’s state-directed industrialization resulted in the co-optation of the upwardly mobile middle class. In both cases, the absence of an independent economic base apart from the state limited rising groups’ interest in making political claims favoring democratization.

Why then do existing theories expect the prospects for democracy to improve with growth of the middle class, which occurs under autocracy? This book’s framework can be distinguished from influential arguments about the political role of the middle class in two broad strands of literature: modernization theory and its values-based variants and political economy’s redistributive theories of democratization.

The Middle Class in Modernization Theory

Classical modernization theory posits a pivotal role for the middle class in economic development and democratization. As articulated by Lipset (1960), and echoed by other influential scholars (e.g., Dahl 1971; Huntington 1991), economic development changes the class structure of society, improving the chances for democratization and democratic stability. At the individual level, scholars have long linked rising educational levels and greater occupational specialization with democratic values (Lipset 1960) and the resources to participate effectively in democratic politics (Almond & Verba 1963; Inkeles 1969). Education has been associated with rising political efficacy, greater political trust, and a growing commitment to tolerance, freedom, and democracy.¹⁰ At the same time, changes in the nature of work are thought to have political consequences. Democracy’s chances are said to improve as the share of jobs that place a premium on independent thinking and endow individuals with the skills for political activism grows (Lipset 1960).¹¹

More recent modernization arguments build explicitly on this earlier preoccupation with values (e.g., Inglehart 1977; Inglehart 1990; Abramson & Inglehart 1995; Inglehart & Baker 2000; Welzel & Inglehart 2008). Effective democracy, Welzel & Inglehart (2008) argue, requires a social base of support to ensure that formal democratic institutions are honored in practice. Economic development makes democracy a stable political outcome, because it is accompanied by a deepening of self-expression values and human empowerment (see also Welzel 2013). In short, modernization theory proposes that human capital formation paves the way for democracy.

Like these theories, I also emphasize education and occupation. Indeed, the post-communist middle classes are the product of rising educational levels, first under communism and then with the expansion of higher education following independence. They are likewise the product of rising managerial and professional opportunities—again, first in the communist era and then with the expansion of the white-collar workforce in the post-independence period. Importantly, however, modernization theory and its values-based variants cannot easily explain why individuals, who are similar in terms of income, education, and occupation hold different views of democracy depending on their relation to the state.

In contrast to my own framework, modernization theory ignores the intervening role of the state through public enterprises and organizations. By ignoring how state economic institutions shape citizens’ incentives, these theories over-predict homogeneity in the political preferences of the middle class. My approach foregrounds precisely these interests and occupational settings, revealing that the rise of a state-supported middle class may actually reinforce authoritarianism and delay democratization. Moreover, while modernization theory expects education to increase support for democracy, I show in chapter 3 that higher levels of education do not invariably make the middle class more democratic.¹²

This book thus also speaks to debates on the relationship between education and democratic values, suggesting that it depends crucially on political and institutional context. As Campbell (2013, 37) writes, of the various ways to measure SES, education is typically the most predictive and robust; other factors must ‘survive’ being in the statistical ring with [this] 800-pound gorilla. Indeed, my findings suggest that the additional explanatory power of income is quite limited after accounting for the effect on democratic preferences of human and social capital. I also find, however, that education alone, and in particular higher education under autocracy, does not consistently give rise to greater support for democracy (see also Wang 2018). Post-Soviet education—during a period when many governments experimented with democracy and institutions of higher education took on the task of preparing democratic citizens—has a weak positive effect on democracy support. By contrast, Soviet education—with its ideological and explicitly antidemocratic character—has if anything a negative effect.

Most importantly, education is closely tied to occupation, which this book shows decisively shapes preferences toward the political economic system. In the statistical ring, I find that state employment not only survives but triumphs over the 800-pound gorilla of education in analyses of democratic attitudes, pro-democracy protest participation, and voting. Thus, whether educated in the Soviet period or the product of post-Soviet higher education, the state middle class shares the same low level of support for democracy as the working class. These findings imply that to understand whether the educated will prefer democracy, we must also consider how state economic institutions affect their life chances. They may also be interpreted as implying that as autocratic countries promote tertiary education in pursuit of development, they will likely need to balance these policies with good jobs, good benefits, and other perks that keep educated groups satisfied.

The Middle Class in Redistributive Theories of Democratization

A second broad strand of literature focuses on the economic position of the middle class. Whereas modernization theories highlight the importance of a cluster of socio-structural variables for predicting political preferences and values (e.g., resources, education, and occupational status), the defining feature of the middle class in redistributivist theories is individual income. From the perspective of these theories, growth of the middle class is favorable to attaining and maintaining democracy, because it lessens the redistributive consequences of democratization (Boix 2003; Acemoglu & Robinson 2006). Indeed, the middle class is pivotal: whether it sides with the rich or the poor determines the fate of democracy. As long as the middle class is small relative to the mass of poor below it, fear of losing resources to the poor will keep the middle classes from favoring political liberalization. Thus until the middle class is large, these theories expect its representatives to act like elites and oppose democratization.

Although redistributivist theories are concerned primarily with macro-level relationships, they are built on microfoundations that are open to critique. First, the underlying individual-level mechanics of these theories follow from the Meltzer & Richard (1981) model, and assume that one’s place in the income distribution drives political preferences. Whether a person is rich or poor relative to others in society determines his/her expected benefit from democracy. Classic redistributivist theories assume that democracy makes society more equal, meaning that the poorer one is relative to others the more one stands to gain by democratization. Overlooked by these theories is whether an individual’s income and expected future benefits are tied to the state and thus likely to be diminished or disrupted by a transition to