Overview of Comparative Politics
Abstract and Keywords
This article discusses several crucial questions that comparative political scientists address. These questions also form part of the basis of the current volume. The article first studies the theory and methods used in gathering data and evidence, and then focuses on the concepts of states, state formation, and political consent. Political regimes, political conflict, mass political mobilization, and political instability are other topics examined in this article. The latter portion of the article is devoted to determining how political demands are processed and viewing governance using a comparative perspective.
Keywords: comparative political scientists, theory, states, political consent, state formation, comparative perspective, mass political mobilization, political regimes, political conflict, political instability
Why do authoritarian states democratize? What accounts for the contours, dynamics, and ideologies of the nation state? Under what conditions do civil wars and revolutions erupt? Why is political representation channeled through political parties in contemporary democracies? Why do some parties run on policy programs, others on patronage? Can citizens use elections and courts to hold governments accountable?
These are some of the crucial questions that the subfield of comparative politics addresses. And they are the questions, among others, around which we organized the Oxford Handbook of Comparative Politics (Boix and Stokes 2007). We asked a set of top scholars in comparative politics to write critical surveys of areas of scholarship in which they are expert. We assembled the volume with two guiding principles. First, we were committed to the possibility (and desirability) of generating a systematic body of theoretical knowledge about politics. The discipline advances, we believe, through theoretical discovery and innovation. Second, we embraced a catholic approach to comparative methodology.
(p. 544) 1 Comparative Politics as (Empirical) Political Science
In the last few decades, the discipline of comparative politics has experienced three main and defining changes: in its object of enquiry; in the methods it now deploys to gather data and test its empirical findings; and in the assumptions (about human and political behavior) it employs to build any theoretical propositions. In doing so, comparative politics has come of age, becoming a key contributor to empirical (as opposed to normative or philosophical) political enquiry. For organizational and administrative reasons, comparative politics is likely to remain a separate field in the discipline and in US departments (where most of today’s political research takes place) in the near future. But from an epistemological point of view, comparative politics is turning into a true science of politics—in the same way economic theory replaced the study of national economies at some point in the past.
Most graduate students in comparative politics who studied in leading departments in the 1960s through the 1980s were trained to conduct research in a single region or country. Indeed, the very term comparative was in most cases misleading. Comparative politics frequently entailed not making comparisons but studying the politics of a foreign country. This methodological choice came hand in hand with an epistemological one. The researcher had to show a deep understanding and a detailed analysis of the political intricacies of a particular polity. That descriptive work often came at the expense of any of the theoretical ambitions that had populated most classical political thinkers from Aristotle to Mill. With slight exaggeration one could think of that state of affairs as the State Department approach to comparative politics, where one scholar staffs the “Japan desk,” another the “Chile desk,” and so on. Of course there were extremely important exceptions. Almond and Verba’s The Civic Culture compared citizens’ attitudes in five countries. Barrington Moore’s Social Origins of Democracy and Dictatorship embarked in a parallel examination of the political and economic evolution of great powers since the early modern period.
The first way in which the field of comparative politics has changed has been epistemological. Even without abandoning the study of particular cases or countries, most comparatists have endorsed the construction and testing of causal theoretical models as the central task of the field. That shift in the object of research has had many progenitors. In part it came from realizing the limits of writing single case studies: Looking at one observation point in a plane will never tell us what forces brought it there. In part it was fed by a few yet very influential comparative pieces written by some modernization and political-development scholars (such as, again, Almond, Lipset, Moore, Rokkan, or Verba). Finally, the gradual introduction of statistical techniques and the mere exercise of data gathering spurred a general interest in cross-national comparisons.
Jointly with a growing acceptance among most researchers about the need to develop broad, general propositions about politics, comparative politics has also (p. 545) embraced the use of standard scientific practices to provisionally validate any theoretical claims. Large-n cross-national studies are now a prominent feature in the (comparative) study of politics—something that would have been hard to predict circa 1970s or even 1980s. But more important than the size of data-sets, what characterizes comparatists today (and rightly so, in our opinion) is a concern for a research design that makes it possible to test theoretical propositions. Provided there are enough degrees of freedom, this should be possible (at least in principle) to accomplish even with very few (but well-chosen) data points or cases. Interestingly, this growing consensus has come with an equally increasing and valuable skepticism about how much it can be accomplished by employing quasi-experimental methods of the kind comparatists usually employ. (In this, comparative politics is not alone: For good or ill, the debate over proper instrumentation has also taken over empirical economists. We will deal over this skepticism in more detail at the end of this chapter.)
In addition to a growing acceptance of the endeavors of theory-building and theory-testing through standard scientific procedures, the scientific enquiry of comparative politics has also shifted in the last decades or, one may say, over the course of the last three generations of scholars devoted to this field, in a third and probably more controversial way—namely, in the way in which theory is built. Probably influenced by the then-dominant approaches of structural sociology and Marxism, in the past comparatists relied on systemic, broad explanations to explain political outcomes. Just think of the initial theories of political modernization, the first articles relating democracy to development or the work on party formation laid out in the 1960s. Today, theory-building very often proceeds (or, perhaps more modestly, claims to proceed) from “microfoundations;” that is, it starts from the individual, and her interests and beliefs, to then make predictions about aggregate outcomes. We find this to constitute an advance in political science. Making us think hard about the final unit of analysis of the model, that is, about each individual (and his motives and actions), allows us to have theories that are more transparent (i.e. where one can truly probe the consistency and plausibility of assumptions) and easier to falsify.
At this point it is important, however, to pause to stress that embracing the principle of methodological individualism does not necessarily mean accepting a purely instrumental or rationalist model of human action. Nor does it mean that the interests and preferences of individuals are not shaped by social and political forces. Recent work in comparative politics has stressed that partisan, ethnic, national, and class identities are in important ways inculcated in individuals by parties, states, and other political actors. As is well known, our increasing reliance on microfoundations has been triggered to a considerable degree by an influx of mathematical and game-theoretic tools and by the influence of economic models in the discipline. But as Moon discussed in the Greenstein–Polsby handbook thirty years ago, models built on propositions about how individual actors will behave under certain circumstances may well employ a variegated set of assumptions about the interests and beliefs of the actors themselves. In fact, his claim (and our hunch) is that the only way to show that rationalistic assumptions do not work is to build models that are populated by intentional actors (with goals that are not strictly instrumental) and that these models perform (p. 546) better than those developed by rational choice theorists. To sum up, building theories of intentional actors and constructing models of (strictly) rationalist individuals are two different enterprises. The latter needs the former but the reverse is not true. Realizing that difference should save for all of us what has been a considerable source of conflict and confusion.
The growing emphasis on building broadly valid theoretical propositions (the first transformation of the field) together with the growing appreciation of the role of individuals and their motives (the third change of the field) have had a beneficial effect on comparative politics and, by definition, on political science. In a way, they have moved the study of politics much closer to our historical predecessors in the discipline. Classical political thinkers, from Aristotle and Machiavelli to Hobbes, Locke, and Rousseau, made an effort to construct a set of theoretical propositions that could explain political life; that is, the foundations of political obligation and its consequences. Even Tocqueville’s celebrated study of America has weathered the passage of time due to the universal theoretical implications it develops in the discussion of a single case. What’s more, each classical political theory started with a particular conception of human nature. With different tools and with a different data-set (for one, we have some information about how real democracies work in practice), all these different (micro) models are, in the end, grounded on specific assumptions about human behavior. These assumptions are still deeply contested in comparative politics: They span from a purely instrumental conception of political actors intent on securing survival and maximizing power to a notion of individuals that may consent to particular structures contingent on others cooperating to, finally, visions of politics that appeal to the inherent sociability of humans. This contestation is unavoidable and healthy. Our guess is that as we move to build intentional models of politics, it should become easier to adjudicate among different points of departure.
2 States, State Formation, and Political Consent
The foundations of power and the sources of political obligation are without much doubt the two main building blocks of any theoretical inquiry in politics. Hence it is not surprising that contractarian theorists paid considerable attention to the mechanisms underlying the formation of states—although they did so mainly for normative reasons. From an empirical or positive point of view, the effort to build theories of state formation happened much later in time. When they appeared, they divided into neoclassical models, which stressed the construction of a coercive structure as part of a voluntary agreement between individuals specialized in coercion and individuals in need of protection, and Marxian models, which portrayed the state as an invention of an elite intent on the exploitation of the masses. Today, mostly as a result of neoinstitutionalist contributions from authors such as North and Olson, (p. 547) the formation of states is seen as a historical turning point in which those agents who specialize in the exercise of violence acquire the right incentives to shift from plundering a population of producers to protecting them from other plunderers. In other words, the founders of states were mostly bandits who, under the proper material and military circumstances, had the incentive to pacify and control a given territory and population in a systematic and orderly manner.
That last theoretical insight has had important empirical implications. It can be actually related to specific historical circumstances that began to take place 10,000 to 8,000 years ago, when agriculture was invented and states followed suit. And it probably explains why stateless societies, where permanent protection mechanisms are absent, have fared much worse in economic terms than human communities governed by state structures.
Still, the theoretical and empirical underdevelopment of the current theories of state formation calls for further scholarly research. Let us just mention two potential avenues of analysis. In the first place, neoinstitutionalists have reduced the foundation of states to a single cause: the transformation of bandits into lords. Yet that conclusion does not seem convincing on both historical and formal grounds. States may form (from a strictly logical point of view) and indeed were formed whenever some producers decided to join forces (that is, whenever they decided to accept a common, binding authority) to respond in some coordinated fashion to (internal or external) plunderers. In fact, in the absence of this second formative path, it seems impossible to explain why noncoercive types have been successful at constructing and maintaining states—a democratic transition, for example, should be seen as an instance of state formation since the problem of political obligation reappears, now in a new light (i.e. with different subjects and sovereigns), as political authority is transferred from one or a few to many. In the second place, theories of state formation have offered some plausible conjectures on the impact of states on economic growth (indeed most of them were mainly built to explain the latter). But they still have little to say about the distributive and social consequences of the emergence of political authority.1
The monopoly of coercion and authority has grown exponentially in both scope and scale in the last half-millennium—it is probably appropriate to think about the modern state as a very different species from the ancient state. In an essay that is reproduced in this volume, Hendrik Spruyt provides a bird’s-eye overview of recent contributions to our understanding of modern state formation, an area of research that has grown substantially in the last three decades. He reviews the ways in which the modern state, with its absolute claims of sovereignty over a particular territory and population, formed and displaced all other forms of governance. This change came in response to a shift in war technology, the growth of commercial capitalism, and new ideas about legitimate government. Spruyt also examines several influential and still-unsettled debates about what caused the emergence of distinct types of constitutional and administrative regimes in the modern period. Most studies of state-building have focused on Europe in the modern period. The recent (p. 548) emergence of independent states outside of Europe in the last centuries is not adequately explained by these accounts. As Spruyt notes, state formation in the twentieth century allows us to evaluate the extent to which the international system, the economy, and the colonial legacy affect how sovereignty and legitimacy have expanded across the globe.
As conveyed by the dominant, Weberian definition of state, political authority relies both on coercion and on some form of legitimization. Accordingly, comparative politics has also considered the ideological dimensions of state formation and of intrastate identity conflict.2 Research on the ideological underpinnings of the modern state has focused on the formation of national identities. Although the debate on the sources, structure, and dimensions of national adscription and nationalism is far from settled, the scholarly community now sees nations as mostly “modern constructs,” i.e. as separate communities to whom individuals perceive themselves as belonging with particular moral attachments and political obligations. Here a, or perhaps the, fundamental point of contention among scholars results from disagreements about the ultimate trigger that caused the appearance of national sentiments. For some, nationalism is the functional response in the political sphere to the demands of industrialization. For others, it derives from the decline of traditional mindsets and the emergence of print capitalism and media markets. Yet for others, nationalism arose in modern times as a response to an upsetting of traditional hierarchies which led, in turn, to a reinterpretation of everyone’s political position as one of belonging to a nation of equals.3
3 Political Regimes and Transitions
Given the democratic revolution of the past quarter-century, it is scarcely surprising that democracy has been a central concern—perhaps the central concern—of comparative politics.
Over the last fifty years, democratization theory has developed several, at times overlapping, at times contradictory, insights and models. Empirically, there seems to be a strong correlation between levels of development and democracy (Lipset 1959; Przeworski and Limongi 1997; Boix and Stokes 2003). Theoretically, the initial structural explanations (Moore 1966) gave way to game-theory models stripped of any sociological foundations—either employed in a metaphorical way, e.g. O’Donnell, Schmitter, and Whitehead (1986), or in a strict, analytical manner, e.g. Przeworski (1990). Those two theoretical approaches have recently been combined to point out why the political consequences of different constitutional institutions account for (p. 549) the social and economic underpinnings of democratic regimes (e.g. Boix 2003). In a chapter reproduced in this volume, Barbara Geddes reviews these theories. She claims that we still have few firm and uncontested conclusions about democracy’s causes and that our empirical results are less robust than one would like, changing with the sample of countries studied, the timeframe considered, and the specification of estimations. The problem is not an absence of theory. Our theories of democratization have become increasingly sophisticated and explicit. Rather, Geddes suggests, the problem may lie in the heterogeneity of the explanandum, democratization. Transitions from absolutist monarchy to constitutional monarchy or to republics may be fundamentally different than transitions from modern military dictatorship to mass democracy. Separating these distinct phenomena, analyzing them—and, more to the point, developing distinct theories of them—is the key, in her view, to gaining firmer knowledge of why countries democratize.
Although behavioral studies and the use of mass surveys spearheaded the transformation of political science into a more comparative and hence more scientific endeavor, the examination of opinion surveys has been somewhat of a laggard in the field of democratization studies. In the volume we edited, Welzel and Inglehart partly correct that imbalance. Reporting findings from a series of recent cross-national studies that have analyzed the effects of mass beliefs on the broadest possible basis, they show that the process of socioeconomic modernization nurtures liberal mass orientations which have, in turn, a positive effect on democracy.
With the exception of Hobbes, the relationship between civic culture and political regimes has been one of the central preoccupations of all modern political theorists. Embracing the new methods that characterized the self-consciously empirical political science that emerged after the Second World War, Almond and Verba in the 1960s tackled this secular concern in their highly influential book on civic culture. Yet this attempt to put the study of the relationship on solid empirical grounds proved unsuccessful.4 The problem with this research agenda had less to do with the (still) very contentious notion of culture than with the ways in which researchers categorized democracy and political culture. They entertained too limited a conception of democracy, restricted to the institutional mechanisms that determine governance at the national level. They thus disregarded the vast number of democratic practices that operate at the local level and in intermediate social bodies. They defined political culture, in turn, as a set of beliefs and dispositions toward certain political objects. But this notion proved to be unsatisfactory: The role that these beliefs and attitudes played in sustaining democratic life and practices was unclear; their origins remained unknown; and, from a purely empirical point, there was no clear proof that democratic stability was bolstered by a particular democratic culture. Nonetheless, it was precisely at the time when the political culture approach had gone down a “degenerative path” that researchers rescued the concept of culture and hence the problem of its political effects by stressing its eminently relational nature. In the late 1980s, Gambetta put trust back into the research agenda. Several researchers emphasized the (p. 550) need to understand interpersonal networks to explain particular behavior. Coleman drew on game-theoretic concepts to develop the notion of social capital. And Putnam then transformed our way of understanding governance and culture in his famous study of Italian regional politics. This new approach is still in its infancy: We know little (both theoretically and empirically) about the mechanisms that go from social capital to good governance, and next to nothing about the dynamics that create, sustain, or deplete civic virtue. And some of us may doubt that trust, as opposed to an engaged skepticism, is the appropriate posture of citizens in democratic polities. But the new approach may well be putting us in the right path to “untangle the complex relationship between democracy and civic culture” (Sabetti 2007, 357).
More than thirty years ago, Juan Linz wrote a highly influential piece on dictatorships for the Handbook of Political Science, edited by Fred Greenstein and Nelson Polsby. Linz’s approach was mostly conceptual and sociological and drew on the literature on totalitarianism and authoritarianism that had been developing since the Second World War. Nondemocratic regimes, according to Linz, could be defined by their degree of internal pluralism, their ideology, and the level of political mobilization which they demanded of their subject populations. Typological approaches have limited empirical purchase, however. Realizing that no single principle can accommodate all the variety of autocracies in place, researchers responded by building a sprawling and mostly ad hoc list of types, such as military dictatorships, traditional absolutist monarchies, one-party states, totalitarian and post-totalitarian systems, parliamentary democracies, city-state oligarchies, “sultanistic” principalities, and so on. These ideal types have turned out to be scarcely informative about the mechanisms through which autocracies work. In this scientific tradition, researchers describe the traits of each type—in other words, they engage in the process of tallying the most frequent elements of each ideal model. But they hardly explain the mechanisms through which power is maintained and the consequences those different institutional structures may have on political stability, citizen compliance, and economic development.
After a period of relative neglect (perhaps a result of the third wave of democratization), the literature on dictatorships has experienced a notable transformation in recent years. Several political economists have started to abandon the strict typological tradition and have instead examined the incentives and mechanisms that structure power and the process of governance in authoritarian systems. Wintrobe (2007), for example, in an essay in the volume we edited, offers an account of dictatorships that starts from rationalist assumptions. To rule, dictators have to combine some degree of repression with the construction of political loyalty. Given the two variables—repression and loyalty—and the objective functions dictators may have, Wintrobe distinguishes between tinpot dictators (who maximize consumption and minimize repression levels), totalitarian dictators (intent on maximizing power), tyrants (who repress without achieving much “loyalty”), and timocrats (who invest in creating loyalty and gaining their citizens’ love). Besides Wintrobe’s work, other scholars have made important contributions to our understanding of electoral autocracies (Magalone 2006), the use of parties and legislatures in authoritarian regimes (Geddes 1999; Wright 2008; Gandhi 2008), or the dynamics of conflict within authoritarian elites (Svolik 2007).
(p. 551) 4 Political Instability, Political Conflict
In a vibrant and by now classic debate, scholars argued first that revolutions occur exclusively as a result of social and economic modernization (Moore, Skocpol, Huntington). More recently, an influential line of argument, brought forth by Goldstone, has framed revolutions as the outbreak that follows a Malthusian imbalance between a growing population and its environment. In an important essay published in the Oxford Handbook of Comparative Politics, Steve Pincus claims that the necessary prerequisite for revolution was always state modernization. State modernization programs simultaneously bring new social groups and new regions into direct contact with the state, and legitimize ideologies of change. These two developments create a social basis and a language on which to build revolutionary movements. Revolutions lead to very different political outcomes. In part following in the steps of Barrington Moore, Jr., Pincus argues that revolutions lead to open, democratic regimes when the state relies on merchant communities and foreign trade. Absent the latter, however, revolutions typically result in the imposition of an authoritarian regime.
In the last decades reality has put civil wars in a central place on the agenda of comparatists. Research on the sources of modern political violence (in the form of civil wars and guerrilla warfare) has gone through several theoretical turns since its inception as a comparative endeavor almost fifty years ago. Modernization scholars explained rebellions as a function of economic inequality (Russett 1964; Paige 1975; Midlarsky 1988; Muller 1985), the impact of social and economic development, and the status and political claims of particular social groups (Huntington 1968; Wolf 1969; Gurr 1973). That strand of enquiry was joined by a second line of research relating violent conflict to ethnic nationalism and the distribution of resources along ethnic lines (Horowitz 1985; Connor 1994). In recent years almost all scholars have de-emphasized the role of economic factors, existing social grievances, or political ideologies in igniting violent conflicts, to stress instead the context of economic and political opportunities in which potential rebels may decide to engage in violent action. Collier and Hoeffler (2004) link the emergence of rebellious activities to the availability of both finance—namely, abundant natural resources—and potential recruits—individuals with reduced prospects of material advancement through peaceful activities. Fearon and Laitin hypothesize that civil wars happen in “fragile states with limited administrative control of their peripheries” (2003, 88). Writing from a different angle, rooted in the examination of the micrologic of violence deployed in civil wars, Kalyvas downplays the presence of single, sociologically unique motivations and describes civil wars as “imperfect, mulilayered, and fluid aggregations of highly complex, partially overlapping, diverse, and localized civil wars with pronounced differences from region to region and valley to valley” (Kalyvas 2006, 371). In the volume we edited, Kalyvas insists as well that war-driven conditions are themselves likely to shape the outcomes of interest: Much changes as civil wars unfold, (p. 552) including the distribution of populations, the preferences of key actors, and the value of resources over which combatants seek control. The exploration of political conflict has also generated an important literature on contentious politics (episodic public collective action) and social movements (sustained challenge to holders of power). Modernization and the spread of democracy spawned the invention of social movements. Yet at the same time, the time and location of social movements (that is, their interaction with political institutions, society, and cultural practices) determined the form in which they emerged (Tarrow and Tilly 2007; Lichbach and deVries 2007).
5 Mass Political Mobilization
Modern democracies are representative democracies. As such they are also party democracies: Political representatives generally coordinate in stable organizations for the purposes of contesting elections and governing. In a chapter reproduced in this volume, Herbert Kitschelt offers of a broad review of the questions that scholars ask about party systems and the way they answer them. Why do democracies feature parties in the first place, as almost all do? Why do many parties compete in some democracies whereas in others competition is restricted to two major parties (or two major ones and a minor one)? Why do some parties compete with the currency of programs, others with valence issues, and still others with clientelism and patronage? Why are elections perennially close in some systems, lopsided in others? Kitschelt reviews the measures that scholars find helpful in answering these questions—party-system fractionalization, the effective number of parties, electoral volatility, and cleavages. The problems afflicting party politics are regionally specific: Whereas scholars of advanced industrial systems worry, as Kitschelt notes, about the decay of party–voter linkages, scholars of new democracies worry about whether such linkages will ever take shape.
Beyond the systemic or functional elements of parties and party systems, and starting with Lipset and Rokkan’s (1967) seminal contribution, researchers have also devoted considerable efforts to understanding why parties formed in the way they did; why and how parties and party systems developed in Western Europe and North America from rather loose networks of politicians, catering to small and strictly delimited electorates, in the early nineteenth century to mass-based, well-organized electoral machines in the twentieth century; and why the number and ideologies of parties varied across countries. As shown in Boix (2007), the nature of parties and party systems can be traced to the underlying structures of preferences, which could be either uni- or multidimensional. But these preferences or political dimensions were mobilized as a function of several additional key factors: the parties’ beliefs (p. 553) about which electoral strategy would maximize their chances of winning, and the electoral institutions that mediate between voters’ choices and the distribution of seats in national parliaments. (These electoral institutions, as shown in Boix 1999, were themselves the product of strategic action by parties.) In a way, that chapter may be read as a response to two types of dominant approaches in the discipline: those institutionalist models that describe political outcomes as equilibria and that, somehow trapped in static applications of game theory, hardly reflect on the origins of the institutions that they claim constrain political actors; and those narratives that stress the contingency and path dependency of all political phenomena while refusing to impose any theoretical structure on them. By contrast, we think it should be possible to build historical accounts in which we reveal (1) how political actors make strategic choices according to a general set of assumptions about their beliefs and interests and (2) how their choices in turn shape the choice set of future political actors.
One of the central contentions of the comparative work done in the 1960s was that partisan attachments and party systems had remained frozen since the advent of democracy in the West. Yet in the last forty years party–voter linkages have substantially thawed (Wren and McElwain 2007). Economic growth, the decline of class differences, and the emergence of postmaterialist values lie in part behind this transformation. In the wake of changes in the electorate and its preferences, it took party bureaucracies some time to adjust. Taking advantage of the slow rate of adjustment of the older parties, new parties sprang up to lure away dissatisfied voters.
Yet party dealignment and electoral volatility have not diminished, even after new parties that should have stabilized the electoral market have entered these party systems. Therefore, to explain continued volatility, we must look beyond changes in the structure of voter preferences. Weakening party–voter ties must be put in the context of a shift in the educational level of the population and new technologies (radios and TV). As parties became less important as informational shortcuts, politics has grown more candidate centered and party elites have been able to pursue electoral campaigns without relying on the old party machinery. If Wren and McElwain are right, our old models of, and intuitions about, party-centered democracy should give way to a more “Americanized” notion of democracies, where personal candidacies and television campaigns determine how politicians are elected and policy made.
In the last two decades, democracy has become the dominant system of government across the world, both as a normative ideal and as a fact. But not all nominal democracies generate accountable, clean governments. In a chapter reproduced in this volume, Susan Stokes addresses one of the possible causes of malfunctioning democracies by looking at the practices, causes, and consequences of clientelism. Clientelism, or the “proffering of material goods [by the patron] in return for electoral support [by the client],” was a hot topic of research in the 1960s and 1970s, buoyed by the emergence of new nations. Shaped by a sociological approach, researchers at that time explained clientelism as a practice underpinned by a set (p. 554) of norms of reciprocity. Yet, as Stokes claims, clientelism must rather be seen as a game in which patrons and clients behave strategically and in which they understand that, given certain external conditions (such as a certain level of development and the organizational conditions that allow for the effective monitoring of the other side), they are better off sustaining a pattern of exchange over the long run. Such a theoretical account then allows us to make predictions, which are beginning to be tested empirically, about the institutions underpinning clientelistic practices, the electoral strategies pursued by patrons, and the potential economic and political effects of clientelism: whether it depresses economic development and political competition.
Political activism has also spawned a large body of research. In her chapter in the Oxford Handbook of Comparative Politics, Pippa Norris (2007) reviews the social and psychological model of participation developed by Verba and Nie, as well as the critiques generated from a rational choice perspective. She then examines how key developments in the research community and the political world have affected the ways in which we evaluate this subfield. She notes a growing interest in the role of institutions in shaping participation in general and turnout in particular. Echoing Wren and McElwain, she draws our attention to changes in party membership, which was widespread and hence instrumental in many advanced democracies but has progressively shrunk, with consequences that are still widely debated among scholars. The constructs of trust and social capital, pioneered by Coleman and Putnam, are also relevant to our expectations about levels of participation. Norris also identifies causeoriented forms of activism as a distinct type of participation, activism that includes demonstrations and protests, consumer politics, professional interest groups, and more diffuse “new” social movements and transnational advocacy networks. All of these, she notes, have expanded and in a way marginalized the more institutionalized, party- and union-based mechanisms of participation that dominated in the past.
6 Processing Political Demands
In the magisterial five-volume Handbook of Political Science published thirty years ago by Greenstein and Polsby, the term accountability appears not once. The term representation appears sporadically and, outside of the volume on political theory, only a handful of times. Thirty years later, accountability has emerged as an organizing concept in comparative politics, with representation not far behind.
In democracies, how do citizens’ preferences get translated into demands for one public policy over another? If everyone in a society had the same preferences, the problem would not be a problem at all. But never is this the case. And scholarship on preference aggregation must come to grips with social choice theory, which should lead us to doubt that citizens in any setting in which politics is multidimensional can (p. 555) evince any stable set of policy preferences. The dominant strains of research, some of which come to grips with the social choice challenge and others of which ignore it, include examinations of the congruence (between preferences and outcomes) of various sorts (Powell 2007). One kind of congruence study looks at the fit between constituents’ preferences and the issue positions of their representatives. Another looks at the fit between electoral outcomes and the allocation of elected offices, treating citizens’ policy preferences as though they were fully expressed by their votes. Another sort of congruence study examines the coherence of issue positions among co-partisans, both political elites and citizens who identify with parties, and tends to find a good deal more coherence among the former than among the latter. Yet another deals with the congruence between electoral platforms and campaign promises, and government policy. Given the role institutions play in aggregating preferences, a big chunk of political science is again devoted to the study of the former—we say “again” because institutions were a key part of the study of politics until the sociological and survey revolution that gripped the discipline in the 1930s and 1940s. Neoinstitutional scholars have focused their attention on electoral rules, executives, legislatures, federalism and, more recently, the judiciary.
Since Duverger’s seminal work, research on electoral rules has focused on the ways (mechanical and psychological) in which electoral systems affect the voting behavior of electors and, as a result, the election of candidates, the structure of parties and party systems, and the politics of coalition-building in democracies (Duverger 1954; Cox 1997; Taagapera 2007).
The existing work on executives and legislatures has centered on two broad topics. First, what is the effect of a constitutional structure based on the separation of powers? Second, what determines the patterns of coalition-making in governments? In the volume we edited, Samuels (2007) reviews what we know about the impact of the separation of powers on accountability. The conventional view in the United States is that a separation of powers is so central to democratic accountability that this separation is nearly definitional of democracy. Samuels evaluates this proposition empirically. His own research and that of other authors which he reviews address questions of accountability and representation, as well as the effects of a separation of powers on the policy process and on regime stability. Among his central findings is that presidentialism has several deleterious effects; a separation of executive from legislative powers increases the chances for policy deadlock and for the breakdown of democracy. In turn, Strom and Nyblade (2007) critically assess the literature on coalition-making, particularly regarding the formation of governments in parliamentary democracies. Drawing on neoinstitutionalism and, more specifically, on the transaction costs literature, they show how the costs of negotiation and the demands of the electorate, interested in monitoring parties’ performance, reduce cycles and push politicians to strike relatively stable pacts. They note that theories of coalition formation began with William Riker’s application of the “size principle,” which predicted that parties would try to minimize the number of actors in a coalition. Although influential theoretically, this approach proved to be rather unsatisfactory empirically. In response, Strom and Nyblade relax Riker’s fundamental assumptions (p. 556) about payoffs, about the role of information, and about the effects of decision rules and institutions, to reach a much richer theory, and one that fits the data more closely.
As discussed by Pablo Beramendi in a thought-provoking chapter, we know far less than we should about federalism. Our theories on the origins of federalism are still sketchy—security threats, the level of heterogeneity, and the evolution of the world economy (in terms of its level of integration) shape the extent of decentralization in a critical manner. The study of the consequences of federalism is slightly more advanced. The relationship between democracy and federalism seems to be conditional, as far as we know, on the particular internal structure of federalism. The effects on the economy of having a federal structure, in turn, depend on how the federal institutions allocate power and responsibilities between the center and regional governments.
The study of the judiciary was traditionally reserved to legal scholars. In their chapter in the Oxford Handbook of Comparative Politics, John Ferejohn, Frances Rosenbluth, and Charles Shipan (2007) remind us why that is not the case any more. After examining the conditions under which courts attain independence, especially from executives but also from legislatures (an independence which O’Donnell and others consider a necessary condition for vertical accountability), they also explain other aspects of cross-national variation, such as why courts everywhere are not enabled to carry out judicial review and why courts are sometimes more active in the legislative process, other times less.
Assessing judicial independence, as these authors acknowledge, is not always straightforward. They advocate two measures: the frequency with which courts reverse governments, and the frequency with which they reverse governments that nationalize parts of the economy (or attempt to do so). The authors note that a drawback of either approach is that courts, which seek (among other objectives) not to have their decisions reversed, may rule against governments only when they anticipate not being reversed, in which case these measures would tend to overestimate their independence. Another difficulty is that courts may rule in favor of governments when they find governments’ actions to be lawful or when they spontaneously agree with governments’ actions. Hence, whereas rulings against governments probably indicate independence, rulings in their favor are less certain indications of dependence (see Helmke 2002; 2005).
7 Governance in Comparative Perspective
It was in the 1970s, that is, about two decades after comparative politics started to develop causal, testable theories, that political scientists ventured in a systematic way (p. 557) into the effects of politics on economic outcomes (and vice versa). Part of that growing interest in political economics started with the analysis by political sociologists of voting and, particularly, of economic voting. The first models presented a simple rule of thumb that voters could—and did—apply when deciding whether to vote for incumbents: If the economy had performed well on their watch, retain them; if it hadn’t, turn them out. Recent scholarly developments place economic voting in institutional contexts and present more nuanced stories about what voters need to know to carry off “simple” economic voting. Raymond Duch’s chapter in our volume reflects and advances this new agenda. Duch (2007) develops a series of propositions about how varying institutional contexts, coalition governments, and informational settings will mediate between economic conditions and voters’ appraisal of them. Factors that Duch suggests will influence economic voting include party-system size, the size of government, coalition governments, trade openness, and the relative strength of governing and opposition parties in the legislature.
At the same time that a few scholars studied how voters react to economic conditions, other researchers began exploring how politicians affect the economy (and therefore voting decisions). After Nordhaus published a seminal paper in 1975 on electoral business cycles, the scholarly literature has evolved in three (complementary) directions. A first set of studies has examined the impact of electoral cycles on the economy. Scholars now tend to agree that the presence of politically induced economic cycles is rather irregular. But, of course, this opens up an important question (particularly from the point of view of democratic representation): Why should voters accept policy manipulation and leave governments unpunished? In our volume, James Alt and Shanna Rose (2007) argue that political business cycles must be understood as a particular instance of the broader phenomenon of political accountability in democratic regimes. Political business cycles are not merely the result of a signaling game in which politicians try to build their reputation as competent policy-makers. Rather, the manipulation of economic policy and outcomes is an inevitable result of voters’ willingness to accept the transfer of some rents to politicians in exchange for the election of competent policy-makers. A second set of studies has focused on the effect that parties, mostly as congealed preferences, may have on macroeconomic policies. Here scholars detect some, generally mild and mostly transitory, effect on macroeconomic factors: Left-wing governments tend to get lower unemployment than right-wing governments for a while, at the cost of permanently higher (and even accelerating) inflation (Alesina et al. 1997). Yet these effects are mostly conditional on the institutional setup (of central banks and wage bargaining institutions) in which governments operate: This last insight has generated the third set of works in the political economics literature (Hall and Franzese 1998; Alvarez, Garrett, and Lange 1991).
After the first papers and books on the topic were written within the framework of modernization theory, welfare state scholars moved to assess the impact of power politics (through parties and unions) on the construction of different types of welfare states. That class-based orientation, however, had limited validity beyond some archetypical cases with high levels of union mobilization and strong left-wing parties.
(p. 558) Accordingly, researchers switched to explore the impact of cross-class coalitions—hence dwelling on the role of middle classes, agricultural producers, and employers. In doing so, they have shifted our attention from the pure redistributive components of the welfare state, which were the keystone of pure class-based, power politics accounts, to social policies as insurance tools that address the problem of risk and volatility in the economy. Related to this change in perspective, welfare state scholars have progressively spent more time mulling over the impact of the international economy on social policy. Two path-breaking pieces by Cameron and Katzenstein showing economic openness and the welfare state to be positively correlated have been followed by an exciting scholarly debate that has alternatively related the result to a governmental response to higher risk (due to more economic volatility in open economies), denied the correlation completely, or called for models that take openness and social policy as jointly determined. As Carnes and Mares (2007) discuss in the essay we edited, the welfare state literature has indeed traveled a long way from its inception. Yet it still has a very exciting research agenda ahead of it: First, it should become truly global and extend the insights (and problems) of a field built around Europe and North America to the whole world; second, it should offer analytical models that combine the different parameters of the successive generations of research in the area; third, it should take seriously the preferences and beliefs of voters across the world (and the cultural differences we observe about the proper role of the state); and, finally, it ought to integrate the consequences of welfare states (something about which we know much less than we should) with the forces that erect them.
Whether the transition to democracy in many developing countries in recent decades has meant a shift to accountable, effective government is a question that has concerned many scholars of comparative politics. Although both the number of researchers and the theories on the topic have multiplied considerably, we still know little about the relationship between growth and political regimes. We know that policy and performance vary considerably across democracies. Poor democracies show lower growth rates and worse public policies than rich democracies. In a nutshell, in spite of having formal mechanisms that should have increased political accountability and the welfare of the population in poor democracies, the provision of public goods and economic performance remain thoroughly deficient in those countries. In our edited volume, Keefer (2007) claims that, since the key parameters of democracy and redistribution (inequality and the struggle for political control between elites and nonelites) cannot explain that outcome (since low development and democratization are cast as contradictory), it must be political market imperfections that explain the failure of governments to deliver in democracies. In young, poor democracies, politicians lack the credibility to run campaigns that promise the delivery of universal benefits and public goods. Accordingly, they shift to building personal networks and delivering particular goods. This type of electoral connection, compounded by low levels of information among voters, who can scarcely monitor politicians, results in extreme levels of corruption and bad governance.
(p. 559) The promise of economic voting was that voters would be able to use economic conditions as a measure of the success or failure of governments; the anticipation of being thus measured would induce politicians to improve economic conditions on their watch. Economic voting would enforce accountability. Yet, as José María Maravall shows in his contribution to our edited volume, “in parliamentary democracies losses of office by prime ministers depend in one half of the cases on decisions by politicians, not by voters” (2007, 935). This fact would not be so dire if prime ministers were removed from office by colleagues who anticipated bad electoral outcomes—if, as Maravall puts it, “voters and politicians … share the same criteria for punishing prime ministers.” But they do not. Whereas prime ministers are more likely to be turned out by voters when economic times are bad, they are more likely to be turned out by their colleagues when economic times are good. Hence politicians who hold their comrades to account seem to practice a reverse kind of “economic voting.” Maravall’s chapter cautions us against excessive optimism regarding democracy, accountability, and economic voting. If (as economic voting implies) officeholders who produce bad economic outcomes will face the wrath of voters, why would they ever risk a costly transition to a liberalized economy? Whether asked in the context of post-Communist countries undertaking a “leap to the market” or in developing countries elsewhere in the world under pressure to move away from statist policies, the question has preoccupied comparative politics and political economy for more than a decade. Reviewing the literature on economic transitions in Eastern Europe, Timothy Frye identifies a number of factors, from the quality of domestic governance to membership in the European Union, that make governments more likely to undertake reforms and then stick with them (Frye 2007). Yet serious gaps remain in our understanding of the determinants of market reforms, including what role is played by institutional legacies from the past, and by contemporary social institutions—networks, business associations, reputational mechanisms—state institutions—courts, bureaucracies, legislatures—and the interaction of the two.
8 Theory and Methods
The questions posed above and others that our contributors raise are too complex, and too important, to restrict ourselves to one or another methodology in our attempts to answer them. It is not that, methodologically speaking, “anything goes;” some research designs and methods for gathering and analyzing evidence are not fruitful. But the contributors to the Oxford Handbook of Comparative Politics explain the advantages and pitfalls of a wide range of techniques deployed by compara-tists, from econometric analysis of cross-national data-sets and observational data to extended stints of fieldwork. They employ a variegated tool kit to make sense of political processes and outcomes.
(p. 560) The volume we edited includes several studies that take stock of what we would lose should the traditional comparative enterprise, with its emphasis on close knowledge of the language, history, and culture of a country or region, be abandoned altogether, and should the activity supporting that approach, the extended period of work in the field, be lost along with it. John Gerring (2007) contends that neither case studies nor large-n comparisons are an unalloyed good; rather, both entail tradeoffs, and we are therefore well advised as a discipline to retain both approaches in our collective repertoire. Where case studies are good for building theory and developing insights, Gerring argues, large-n research is good for confirming or refuting theory. Where case studies offer internal validity, large-n studies offer external validity. Where case studies allow scholars to explore causal mechanisms, large-n comparisons allow them to identify causal effects. Elisabeth Wood’s chapter alerts us to what we are in danger of losing should we as a profession give up on field research (Wood 2007). To the rhetorical question, “Why ever leave one’s office?” she gives several answers. Interacting personally with subjects in their own setting may be the only way to get a handle on many crucial research questions, such as which of many potential political identities subjects embrace and what their self-defined interests are. Fieldwork is not without perils, Wood explains, both intellectual and personal. Interview subjects may be evasive and even strategically dissimulating; field researchers may have strong personal reactions, positive or negative, to their subjects, reactions that may then color their conclusions; and fieldwork is a lonely endeavor, with predictable highs and lows. Wood suggests strategies for dealing with these pitfalls. James Mahoney and Celso Villegas (2007) discuss another variant of qualitative research: comparative historical studies. The aims of this research differ from those of cross-national studies, they contend. Comparative historical scholars “ask questions about the causes of major outcomes in particular cases,” and hence seek to explain “each and every case that falls within their argument’s scope.” By contrast, large-n researchers “are concerned with generalizing about average causal effects for large populations and … do not ordinarily seek to explain specific outcomes in particular cases.”
Robert Franzese’s chapter defends, in turn, large-n, quantitative techniques against some of the critiques that other contributors level against them (Franzese 2007). Comparative political scientists, like empirically oriented sociologists and economists, are bedeviled by four problems: a trade-off between quantity and quality in the collection of data; multicausality; context-conditionality, that is, the fact all the effects of our variables are conditional on other variables; and endogeneity. Yet as Franzese argues, these obstacles, which are in fact inherent to our trade, should not lead us to dodge quantitative strategies of research. On the contrary, a simple back-of-the-envelope calculation shows that the plausible loss of precision involved in measuring large numbers of observations does not justify retreating to qualitative studies of a few cases—even if we attain very precise knowledge about small samples, they fail to yield robust inferences. Similarly, the presence of multiple and conditional causality cannot be solved easily by case studies (although good process tracing may alleviate these problems). Finally, qualitative case-study research does not necessarily escape (p. 561) problems of endogeneity. To move from correlational analysis to causal propositions, Franzese contends, we need to employ more sophisticated techniques, such as variable instrumentation, matching, or vector autoregression. But even these techniques are not sufficient. Here we would like to add that, influenced by a few macroeconomists and political economists, part of the discipline seems on the verge of uncritically embracing the use of instrumentation to deflect all the critiques that are leveraged against any work on the grounds that the latter suffer from endogeneity. It turns out that there are very few, if any, instruments that are truly exogenous—basically, geography. Their use has extraordinary theoretical implications that researchers have either hardly thought about (for example, that weather determines regime, in a sort of Montesquieuian manner) or simply dodge (when they posit that the instrument is simply a statistical artefact with no theoretical value on its own and then insist that it is the right one to substitute for the variable of interest). Thus, we want to stress with Franzese that only theory-building can truly help us in reducing the problem of endogenous causation.
Adam Przeworski (2007) offers a less optimistic perspective on observational research, large-n or otherwise. Observational studies, ones that do not (and cannot fully) ensure that the cases we compare are matched in all respects other than the “treatment,” cannot deal adequately with problems of endogeneity. “We need to study the causes of effects,” he writes, “as well as the effects of causes.” Some covariates (traits a unit has prior to the application of a treatment) are unobserved. These unobserved covariates may determine both the likelihood of a unit’s being subjected to the treatment and the likelihood of its evincing the effect. Because these covariates are unobserved, we cannot test the proposition that they, rather than the treatment or putative cause, are actually responsible for the effect.
Przeworski discusses traditional as well as more novel approaches to dealing with endogeneity, but his chapter leans toward pessimism. “To identify causal effects we need assumptions and some of these assumptions are untestable.” His chapter will be must reading for comparatists as they assess the promise and limitations of observational versus experimental or quasi-experimental designs.
But perhaps the mood of the chapter is more pessimistic than it need be. Theory should help us distinguish cases in which endogeneity is plausibly present from ones in which it is not. One way of reading Przeworski’s chapter is that a crucial research task is to shift key covariates from the unobserved to the observed category. This task is implied by a hypothetical example that Przeworski offers. A researcher wishes to assess the impact of governing regime on economic growth. Future leaders of some countries study at universities where they become pro-democratic and learn how to manage economies, whereas others study at universities that make them prodictatorial and teach them nothing about economic management. Both kinds return home to become leaders and govern their societies and economies in the manner consistent with their training. It therefore appears that democracy produces economic growth. The training of leaders is a variable that we cannot observe systematically, in Przeworski’s view. But there is a difference between unobserved and unobservable. It is not obvious to us why this variable could never be systematically observed, should (p. 562) our theory—and, perhaps, our close, case-study-informed knowledge—tell us that we should worry about it.
Whether one studies a large or small number of cases, and whether one employs econometrics or other techniques, Robert Bates’s chapter in the Oxford Handbook of Comparative Politics argues that one should do theoretically sophisticated work informed by game theory (Bates 2007). Indeed, the use of game-theoretical models, of varying degrees of formalization, is a strong recent trend in comparative politics. Illustrating his methodological claims with his recent research on the politics of coffee production and commercialization, Bates offers a comprehensive strategy for comparative study. The first step of research is apprehension: a detailed study and understanding of a particular time and place. Verstehen is then followed by explanation: The researcher apportions the things she knows “between causes or consequences” and attempts to develop “lines of logic to link them.” In Bates’s view, the explanatory drive should begin with the assumption (or principle) of rationality and use game theory to impose a structure on the phenomena we observe. The structure of the game allows us to push from the particular to the construction of broader theories, themselves susceptible of validation. The construction of theoretical explanations must then be subject to the test of confirmation: This implies progressively moving from small-n comparisons to much larger data-sets in which researchers can evaluate their theories against a broad set of alternatives and controls.
An analysis of the methodological foundations of contemporary political research would be incomplete without an exploration of the role of rationalist assumptions in the discipline. Eleanor Ostrom (2007) takes as her point of departure the proposition that “the theory of collective action is the central subject of political science” and that the problem of collective action is rooted in a social dilemma (or, in game theory terms, a prisoner’s dilemma) in which, as is well known, rational individuals in pursuit of their optimal outcome may end up not cooperating even if it was in their interest to do so. Ostrom assesses the first generation of studies of collective action, which stress the structural conditions (number of players, type of benefits, heterogeneity of players, the degree of communication among them, and the iteration of games) that may increase the likelihood of achieving cooperation. She finds these studies wanting. Ostrom recognizes that the rationalist model only explains part of human behavior. Hence she calls for a shift toward a theory of boundedly rational, norm-based human behavior. Instead of positing a rationalistic individual, we should consider agents who are inherently living in a situation of informational uncertainty and who structure their actions, adopt their norms of behavior, and acquire their knowledge from the social and institutional context in which they live. In this broader theory of human behavior, humans are “adaptive creatures who attempt to do as well as they can given the constraints of the situations in which they find themselves (or the ones that they seek out).” They “learn norms, heuristics, and full analytical strategies from one another, from feedback from the world, and from their own capacity to engage in self-reflection. They are capable of designing new tools—including institutions—that can change the structure of the worlds they face for good or evil purposes. They adopt both short-term and long-term perspectives dependent on the structure of (p. 563) opportunities they face.” All in all, Ostrom’s approach encompasses a broader range of types of human action—from individuals who are fully “rational” (normally in those environments in which they live in repetitive, highly competitive situations) to “sociological agents” (whose behavior simply follows shared social norms). To some extent, the discipline seems to come full circle with this contribution: moving from cultural approaches under the aegis of modernization theory to the rationalist assumptions of institutionalist scholars and now back to a richer (perhaps looser but certainly closer to the way our classical thinkers thought about human nature) understanding of human agency. This journey has not been useless. On the contrary, as we traveled from one point to the other we have learned that a good theory of politics must be based on solid microfoundations; that is, on a plausible characterization of the interests, beliefs, and actions of individuals.
9 Looking Ahead
When we pause to take stock of the evolution of empirical political science in the last three decades, we think we should be pleased with the progress we have made as a community of scholars. The discipline has moved forward substantially in modeling certain political outcomes. We can offer the first inklings of theories of state formation and democratization. We are in a position to understand how power is sustained and exercised in dictatorships. We have fruitful models of the impact of institutions in preference aggregation and electoral behavior. We are beginning to tackle with some analytical precision the problem of political accountability. As we argued at the beginning of the chapter, we think this progress has been brought about by three conceptual engines: a renewed commitment to solve the central theoretical problems already debated by our forefathers; the decision to harness our empirical work (be it cases or cross-country regressions) to standard scientific practices; and a commitment to build “transparent” theoretical propositions; that is, propositions based on assumptions which can in turn be debated and from which we can derive results in a logical way.
There may be a hundred different things that need to be changed, improved, or altogether invented in our discipline. We will only stress one here. Many of our most successful models are simply equilibria. The gains from that type of methodological choice are clear. But we still know little about the ways in which political institutions, social practices, norms, and arrays of political interests originate and collapse. History was important in the broad, sociological literature written a few decades ago. Yet the way in which it was tackled was messy or unsystematic. Institutionalists altogether abandoned historical work. We think that, with the new tools we have in our hands, the right time has come to deal with that question again. To some extent, given the (p. 564) problems of endogenous causation we are confronted with, engaging in this type of work is now becoming inevitable.
Alesina, A. et al. 1997. Political Cycles and the Macroeconomy. Cambridge, Mass.: MIT Press.Find this resource:
Alt, J. E. and Rose, S. S. 2007. Context-conditional political budget cycles. Ch. 34 in Boix and Stokes 2007.Find this resource:
Alvarez, R. M., Garrett, G., and Lange, P. 1991. Government partisanship, labor organization and macroeconomic performance, 1967–1984. American Political Science Review, 85: 539–56.Find this resource:
Anderson, B. 1983. Imagined Communities: Reflections on the Origin and Spread of Nationalism. New York: Verso.Find this resource:
Bates, R. H. 2007. From case studies to social science: a strategy for political research. Ch. 7 in Boix and Stokes 2007.Find this resource:
Boix, C. 1999. Setting the rules of the game: the choice of electoral systems in advanced democracies. American Political Science Review, 93: 609–24.Find this resource:
——2003. Democracy and Redistribution. New York: Cambridge University Press.Find this resource:
——2007. The emergence of parties and party systems. Ch. 21 in Boix and Stokes 2007.Find this resource:
——2009. Some thoughts on the origins of inequality. Unpublished manuscript, Princeton University.Find this resource:
——and Stokes, S. 2003. Endogenous democratization. World Politics, 55: 517–49.Find this resource:
——(eds.) 2007. Oxford Handbook of Comparative Politics. New York: Oxford University Press.Find this resource:
Carnes, M. and Mares, I. 2007. The welfare state in global perspective. Ch. 35 in Boix and Stokes 2007.Find this resource:
Collier, P. and Hoeffler, A. 2004. Greed and grievance in civil war. Oxford Economic Papers, 56: 563–95.Find this resource:
Connor, W. 1994. Ethnonationalism: The Quest for Understanding. Princeton, NJ: Princeton University Press.Find this resource:
Cox, G. 1997. Making Votes Count. New York: Cambridge University Press.Find this resource:
Duch, R. 2007. Comparative studies of the economy and the vote. Ch. 33 in Boix and Stokes 2007.Find this resource:
Duverger, M. 1954. Political Parties. New York: Wiley.Find this resource:
Fearon, J. and Laitin, D. 2003. Ethnicity, insurgency, and civil war. American Political Science Review, 97: 75–90.Find this resource:
Ferejohn, J., Rosenbluth, F., and Shipan, C. 2007. Comparative judicial politics. Ch. 30 in Boix and Stokes 2007.Find this resource:
Franzese, R. J. 2007. Multicausality, context-conditionality, and endogeneity. Ch. 2 in Boix and Stokes 2007.Find this resource:
Frye, T. 2007. Economic transformation and comparative politics. Ch. 38 in Boix and Stokes 2007.Find this resource:
Gandhi, J. 2008. Political Institutions under Dictatorship. New York: Cambridge University Press.Find this resource:
Geddes, B. 1999. Authoritarian breakdown: empirical test of a game theoretic argument. Unpublished manuscript, UCLA.Find this resource:
——2007. What causes democratization? Ch. 14 in Boix and Stokes 2007.Find this resource:
(p. 565) Gellner, E. 1983. Nations and Nationalism. Ithaca, NY: Cornell University Press.Find this resource:
Gerring, J. 2007. The case study: what it is and what it does. Ch. 4 in Boix and Stokes 2007.Find this resource:
Greenfeld, L. and Eastwood, J. 2007. National identity. Ch. 11 in Boix and Stokes 2007.Find this resource:
Gurr, T. 1973. The revolution–social change nexus. Comparative Politics, 5: 359–92.Find this resource:
Hall, P. A. and Franzese, R. J., Jr. 1998. Mixed signals: central bank independence, coordinated wage bargaining, and European Monetary Union, International Organization, 52: 505–35.Find this resource:
Hardin, R. 2007. Compliance, consent, and legitimacy. Ch. 10 in Boix and Stokes 2007.Find this resource:
Helmke, G. 2001. The logic of strategic defection: judicial decision-making in Argentina under dictatorship and democracy. American Political Science Review, 96: 291–330.Find this resource:
——2005. Courts under Constraints: Judges, Generals, and Presidents in Argentina. Cambridge: Cambridge University Press.Find this resource:
Horowitz, D. L. 1985. Ethnic Groups in Conflict. Los Angeles: University of California Press.Find this resource:
Huntington, S. 1968. Political Order in Changing Societies. New Haven, Conn.: Yale University Press.Find this resource:
Kalyvas, S. 2006. The Logic of Violence in Civil War. New York: Cambridge University Press.Find this resource:
——2007. Civil wars. Ch. 18 in Boix and Stokes 2007.Find this resource:
Keefer, P. 2007. The poor performance of poor democracies Ch. 36 in Boix and Stokes 2007.Find this resource:
Lichbach, M. I. and deVries, H. G. E. 2007. Mechanisms of globalized protest movements. Ch. 20 in Boix and Stokes 2007.Find this resource:
Lipset, S. M. 1959. Some social requisites of democracy: economic development and political legitimacy. American Political Science Review, 53: 69–105.Find this resource:
——and Rokkan, S. 1967. Cleavage structures, party systems, and voter alignments. Pp. 1–56 in Party Systems and Voter Alignments, ed. S. M. Lipset and S. Rokkan. New York: Free Press.Find this resource:
Magalone, B. 2006. Voting for Autocracy: Hegemonic Party Survival and its Demise in Mexico. New York: Cambridge University Press.Find this resource:
Mahoney, J. and Villegas, C. M. 2007. Historical enquiry and comparative politics. Ch. 3 in Boix and Stokes 2007.Find this resource:
Maravall, J. M. 2007. Accountability and the survival of governments. Ch. 37 in Boix and Stokes 2007.Find this resource:
Midlarsky, M. I. 1988. Rulers and the ruled: patterned inequality and the onset of mass political violence. American Political Science Review, 82: 491–509.Find this resource:
Moore, B. 1966. Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World. Boston: Beacon.Find this resource:
Muller, E. N. 1985. Income inequality, regime repressiveness, and political violence. American Sociological Review, 50: 47–61.Find this resource:
Nordhaus, W. 1975. The political business cycle. Review of Economic Studies, 42: 169–90.Find this resource:
Norris, P. 2007. Political activism: new challenges, new opportunities. Ch. 26 in Boix and Stokes 2007.Find this resource:
O’Donnell, G., Schmitter, P. C., and Whitehead, L. (eds.) 1986. Transitions from Authoritarian Rule: Comparative Perspectives, Volume 4: Tentative Conclusions and Uncertain Democracies. Baltimore: Johns Hopkins University Press.Find this resource:
Ostrom, E. 2007. Collective action theory. Ch. 8 in Boix and Stokes 2007.Find this resource:
Paige, J. M. 1975. Agrarian Revolution. New York: Free Press.Find this resource:
Powell, G. B. 2007. Aggregating and representing political preferences. Ch. 27 in Boix and Stokes 2007.Find this resource:
Przeworski, A. 1990. Democracy and the Market. New York: Cambridge University Press.Find this resource:
——2007. Is the science of comparative politics possible? Ch. 6 in Boix and Stokes 2007.Find this resource:
——and Limongi, F. 1997. Modernization: theories and facts. World Politics, 49: 155–83.Find this resource:
(p. 566) Russett, B. M. 1964. Inequality and instability. World Politics, 16: 442–54.Find this resource:
Sabetti, F. 2007. Democracy and civic culture. Ch. 15 in Boix and Stokes 2007.Find this resource:
Samuels, D. 2007. Separation of powers. Ch. 27 in Boix and Stokes 2007.Find this resource:
Strom, K., and Nyblade, B. 2007. Coalition theory and government formation. Ch. 32 in Boix and Stokes 2007.Find this resource:
Svolik, M. 2007. Power-sharing and leadership dynamics in authoritarian regimes. Unpublished manuscript, UIUC.Find this resource:
Taagapera, R. 2007. Electoral systems. Ch. 28 in Boix and Stokes 2007.Find this resource:
Tarrow, S. and Tilly, C. 2007. Contentious politics and social movements. Ch. 19 in Boix and Stokes 2007.Find this resource:
Wintrobe, R. 2007. Dictatorship: analytical approaches. Ch. 16 in Boix and Stokes 2007.Find this resource:
Wolf, E. R. 1969. Peasant Wars in the Twentieth Century. New York: Harper and Row.Find this resource:
Wood, E. J. 2007. Field research. Ch. 5 in Boix and Stokes 2007.Find this resource:
Wren, A. and McElwain, K. M. 2007. Voters and parties. Ch. 23 in Boix and Stokes 2007.Find this resource:
Wright, J. 2008. Do authoritarian institutions constrain? How legislatures impact economic growth and foreign aid effectiveness. American Journal of Political Science, 52.Find this resource: