Showing papers in "American Political Science Review in 1994"

Domestic political audiences and the escalation of international disputes

Abstract: International crises are modeled as a political “war of attrition” in which state leaders choose at each moment whether to attack, back down, or escalate. A leader who backs down suffers audience costs that increase as the public confrontation proceeds. Equilibrium analysis shows how audience costs enable leaders to learn an adversary's true preferences concerning settlement versus war and thus whether and when attack is rational. The model also generates strong comparative statics results, mainly on the question of which side is most likely to back down. Publicly observable measures of relative military capabilities and relative interests prove to have no direct effect once a crisis begins. Instead, relative audience costs matter: the side with a stronger domestic audience (e.g., a democracy) is always less likely to back down than the side less able to generate audience costs (a nondemocracy). More broadly, the analysis suggests that democracies should be able to signal their intentions to other states more credibly and clearly than authoritarian states can, perhaps ameliorating the security dilemma between democratic states.

Shortcuts versus encyclopedias: information and voting behavior in california insurance reform elections

Abstract: Voters in mass elections are notorious for their apparent lack of information about relevant political matters. While some scholars argue that an electorate of well-informed voters is necessary for the production of responsive electoral outcomes, others argue that apparently ignorant voters will suffice because they can adapt their behavior to the complexity of electoral choice. To evaluate the validity of these arguments, I develop and analyze a survey of California voters who faced five complicated insurance reform ballot initiatives. I find that access to a particular class of widely available information shortcuts allowed badly informed voters to emulate the behavior of relatively well informed voters. This finding is suggestive of the conditions under which voters who lack encyclopedic information about the content of electoral debates can nevertheless use information shortcuts to vote as though they were well informed.

Collective Identity Formation and the International State

Abstract: The neorealist-neoliberal debate about the possibilities for collective action in international relations has been based on a shared commitment to Mancur Olson's rationalist definition of the problem as one of getting exogenously given egoists to cooperate. Treating this assumption as a de facto hypothesis about world politics, I articulate the rival claim that interaction at the systemic level changes state identities and interests. The causes of state egoism do not justify always treating it as given. Insights from critical international relations and integration theories suggest how collective identity among states could emerge endogenously at the systemic level. Such a process would generate cooperation that neither neorealists nor neoliberals expect and help transform systemic anarchy into an “international state”—a transnational structure of political authority that might undermine territorial democracy. I show how broadening systemic theory beyond rationalist concerns can help it to explain structural change in world politics.

Repression of Human Rights to Personal Integrity in the 1980s: A Global Analysis.

Abstract: This crossnational study seeks to explain variations in governmental repression of human rights to personal integrity (state terrorism) in a 153-country sample during the eighties. We outline theoretical perspectives on this topic and subject them to empirical tests using a technique appropriate for our pooled cross-sectional time-series design, namely, ordinary least squares with robust standard errors and a lagged dependent variable. We find democracy and participation in civil or international war to have substantively important and statistically significant effects on repression. The effects of economic development and population size are more modest. The hypothesis linking leftist regime types to abuse of personal integrity rights receives some support. We find no reliable evidence that population growth, British cultural influence, military control, or economic growth affect levels of repression. We conclude by considering the implications of our findings for scholars and practitioners concerned with the prevention of personal integrity abuse.

Divided Government, Fiscal Institutions, and Budget Deficits: Evidence from the States.

Abstract: Does partisan control of American state government have systematic effects on state spending and taxing levels? Does divided control affect the government's ability to make hard decisions? Do institutional rules like legal deficit carryover restrictions matter? Using a formal model of fiscal policy to guide empirical analysis of data covering the American states from 1968 to 1987, we conclude that (1) aggregate state budget totals are driven by different factors under Democrats and Republicans, the net result being that Democrats target spending (and taxes) to higher shares of state-level personal income; (2) divided government is less able to react to revenue shocks that lead to budget deficits, particularly where different parties control each chamber of the legislature; and (3) unified party governments with restricted ability to carry deficits into the next fiscal year (outside the South) have sharper reactions to negative revenue shocks than those without restrictions.

Comparative Democracy: the Economic Development Thesis

Abstract: In comparative politics, an established finding—that economic development fosters democratic performance—has recently come under challenge. We counter this challenge with a dynamic pooled time series analysis of a major, but neglected data set from 131 nations. The final generalized least squares-autoregressive moving averages estimates (N = 2,096) appear robust and indicate strong economic development effects, dependent in part on the nation's position in the world system. For the first time, rather hard evidence is offered on the causal relationship between economics and democracy. According to Granger tests, economic development “causes” democracy, but democracy does not “cause” economic development. Overall, the various tests would seem to advance sharply the modeling of democratic performance.

Democracy and the Peaceful Settlement of International Conflict.

Abstract: The research reported here develops an explanation for the often-noted absence of international war between democratic states. This explanation is derived from a theoretical rationale centered on universal democratic norms for reconciling competing values and interests. I argue that democratic states locked in disputes are better equipped than others with the means for diffusing conflict situations at an early stage before they have an opportunity to escalate to military violence. Not only is this explanatory logic consistent with the published findings on democracy and war, but it also entails the novel empirical proposition that disputes between democracies are more amenable than are other disputes to peaceful settlements, the hypothesis I examine here. Analyses of contemporary interstate disputes reveal that even when potentially confounding factors are controlled, democratic opponents are significantly more likely to reach peaceful settlements than other types of disputants.

Electoral competition with informed and uninformed voters

Abstract: I present a model of electoral competition in which candidates raise campaign contributions by choosing policies that benefit interest groups and then expend those contributions to influence voters who are uninformed about the policies. Informed voters, however, vote based on those policies, so candidates face a trade-off between choosing a policy to generate funds to attract the uninformed vote and choosing a policy to attract the informed vote. Electoral equilibria are characterized for two categories of policies: particularistic and collective. In the case of particularistic policies, the equilibrium policies of the candidates are separated if the proportion of uninformed voters is sufficiently high, and the degree of separation is an increasing function of that proportion. The model is extended to include the public financing of elections and incumbency advantages. For the case of collective policies, the candidates locate at the median of the ideal points of the informed voters, and contributions are zero.

Does Attack Advertising Demobilize the Electorate

Abstract: J/47Te address the effects of negative campaign advertising on turnout. Using a unique jfjJ \ experimental design in which advertising tone is manipulated within the identical X V t audiovisual context, we find that exposure to negative advertisements dropped intentions to vote by 5%. We then replicate this result through an aggregate-level analysis of turnout and campaign tone in the 1992 Senate elections. Finally, we show that the demobilizing effects of negative campaigns are accompanied by a weakened sense of political efficacy. Voters who watch negative advertisements become more cynical about the responsiveness of public officials and the electoral process. It is generally taken for granted that political campaigns boost citizens' involvement-their interest in the election, awareness of and information about current issues, and sense that individual opinions matter. Since Lazarsfeld's pioneering work (Berelson, Lazarsfeld, and McPhee 1954; Lazarsfeld, Berelson, and Gaudet 1948), it has been thought that campaign activity in connection with recurring elections enables parties and candidates to mobilize their likely constituents and "recharge" their partisan sentiments. Voter turnout is thus considered to increase directly with "the level of political stimulation to which the electorate is subjected" (Campbell et al. 1966, 42; Patterson and Caldeira 1983). The argument that campaigns are inherently "stimulating" experiences can be questioned on a variety of grounds. American campaigns have changed dra

Civic Culture and Democracy: The Question of Causal Relationships.

Abstract: A causal model of relationships between structural properties of states, civic culture attitudes of the general public, and change in level of democracy is tested with cross-national data. The model permits inferences about the possibility of unidirectional or reciprocal causation between civic culture attitudes and democracy, controlling for macrosocietal variables such as economic development, income inequality, and subcultural pluralism. Most civic culture attitudes do not have any significant impact on change in democracy. One of them, interpersonal trust, appears clearly to be an effect rather than a cause of democracy. The exception is the percentage of the general public that prefers gradual reform of society instead of revolutionary change or intransigent defense of the status quo. Support for gradual reform has a positive impact on change in democracy, and it is unrelated to a country's years of continuous democracy—findings that support the hypothesis of a unidirectional civic culture effect on democracy.

The Determinants of Industry Political Activity, 1978–1986.

Abstract: C ampaign contributions made by corporate political action committees (PACs) are increasingly important-and controversial-in U.S. elections. "Influence purchasing" is often alleged by pundits, and campaign finance reform is continually debated in Congress. In the academy, social scientists have analyzed the allocation of interest group monies and have looked for possible impacts on legislation. Few, however, have tried to explain the total contributions of economic interest groups. We shall analyze variations in contributions across industries and over time with an empirical model based on organizationally constrained profit-maximizing behavior. We assume that corporate PAC contributions in an industry are raised to enhance industry profits subject to the constraints of organization costs and freeriding incentives. Our empirical work covers 124 industries over five election cycles from 1978 through 1986. Both the costs and benefits of political activity vary across industries. We argue that the benefits of political action are determined by direct contact with the government as a regulator or purchaser of industry output, by government's ability to ameliorate adverse market conditions, and by the industry's ability to solve collective action problems without government assistance. The costs of industry political action arise mainly from collective action problems, because effective political activity often requires concerted action by the constituent firms in each industry. We also make a methodological contribution. Previous studies use statistical techniques that suffer from sample selectivity bias or else force the probability of acting politically to be explained by the same set of coefficients as is the amount of action taken. We explain and employ a technique that accounts for both problems. The results are consistent with the theses that industries follow investment-oriented goals in deciding the amount of political action to undertake, and that collective action problems are important constraints on industry political activity. We can explain between 60% and 80% of the variation in contributions in our sample. First, we review existing work on corporate political activity, particularly papers examining patterns of aggregate PAC spending by industry. Then we present an organization-cost-constrained industryprofit-maximizing model, based on a comparison between an idealized zero-organization-cost industry and the more realistic case where organizing collective action is difficult. The independent variables used in the analysis are also discussed and defined. Next, we discuss the dependent variable and total industry contributions and explain our statistical method. Then the results are presented, followed by a discussion of industry structure and political activity. To make clear how our results can explain different contribution patterns, we give examples using five actual industries.

Power and Influence in State Legislative Policymaking: The Interaction of Gender and Position in Committee Hearing Debates.

Abstract: ere is an implicit assumption behind advocating for more minorities or women in elected office, namely, that these officials will bring a new power and influence to their underrepre.5 sented groups. However, for women, this idealized viewpoint ignores the social dynamics that subordinate women's words and actions even in "well-balanced" male and female group interactions. Using transcribed verbatim transcripts of 12 state legislative committee hearings, this research analyzes the conversational dynamics of committee members, witnesses, chairs, and sponsors. Sex differences among committee members are highly significant, even after accountingfor political factors and structural features of the hearing. In addition, male and female chairs do not conduct hearings in the same way, and these differences affect the behavior of witnesses and committee members. The findings suggest that as the proportion of women increases in a legislative body, men become more verbally aggressive and controlling of the hearing. Women legislators may be seriously disadvantaged and unable to participate equally in legislative policymaking in committee hearings.

Economic security and value change

Abstract: Confirming Inglehart's prediction (1971) of an intergenerational shift toward postmaterialist values, a time series analysis controlling for the joint effects of inflation and unemployment demonstrates that there is a statistically significant trend toward postmaterialism in all eight West European countries for which data are available over the past two decades. Evidence from the 1981–83 and 1990–91 World Values Surveys indicates that this value shift occurs in any society that has experienced sufficient economic growth in recent decades so that the preadult experiences of younger birth cohorts were significantly more secure than those of older cohorts. Large intergenerational differences tend to be found in societies that have experienced rapid growth in gross national product per capita, and are negligible in societies that have had little or no growth. Accordingly, postmaterialism increased in 18 of the 20 societies on five continents for which we have comparable data over the past decade.

Racial Threat and Partisan Identification

Abstract: Power theory views relationships between groups as a function of their competitive positions in political, economic, and social arenas. In contexts where the threat posed by a minority group is high, the dominant group's response is predicted to be more hostile than in contexts where that threat is low. A pooled time series analysis of voter registration data for Louisiana parishes for 1975-90 provides support for the operation of the threat mechanism. Higher black concentrations are associated with declines in the percentage of white registered voters who are Democrats and an increase in the percentage who are Republicans. Consistent with the expectations of power theory, this relationship is conditioned by the social status of the parish. T he past three decades have witnessed two dramatic trends in southern politics. First, beginning with the Voting Rights Act of 1965, there has been an extraordinary increase in political participation among southern blacks. This mobilization has been directed almost entirely to the Democratic party. Second, commencing at about the same time, a steep decline has occurred in the identification of southern whites with the Democratic party. This decline has been accompanied by a limited growth in Republican identification in the region and a large increase in the number of southern whites who identify as Independents. The end product of these trends has been twofold. On the one hand, the once solidly Democratic South no longer can be said to have a majority party (Black and Black 1987, 232). On the other hand, the Democratic party that once championed racial segregation in the region has come increasingly to depend upon black voters for its

Toward a Theory of Constitutional Amendment

Abstract: Constitutional design proceeds under the assumption that institutions have predictable consequences, but modern political science has not pursued the empirical verification of these predicted consequences with much vigor. I shall attempt to link the theoretical premises underlying one important aspect of constitutional design, the amendment process, with the empirical patterns revealed by a systematic, comparative study of constitutions. An examination of all amendments in the 50 American states since 1776 reveals patterns that are then confirmed using data from 32 national constitutions. The interaction of the two key variables affecting amendment rate can be described by an equation that generates predicted amendment rates close to those found in the cross-national empirical analysis. A constitution's length measured in number of words, the difficulty of an amendment process, and the rate of amendment turn out to have interlocking consequences that illuminate principles of constitutional design.

On the Idea of the Moral Economy

Abstract: The moral economic school, which has flourished among anthropologists, economic historians, and classicists, has received only limited attention from political scientists. This is perplexing, since at its core is to be found an intersection of debates over rational choice theory, the character of modern and premarket societies, and the normative standing of the market—in other words, over issues of formidable importance to our discipline. I seek to correct that neglect by mapping out and critically analyzing the moral economists' conception of modernity, their critique of the economic approach to human behavior and institutions, and their attempt to formulate an Aristotelian theory of the economy. These projects, though flawed, together are more than rich enough to provide fertile ground for political scientists and philosophers. I conclude with a discussion of the moral economists' effort to develop a normative theory of the economy together with a related critique of the market.

Divided government in the american states: a byproduct of legislative professionalism?

Abstract: S since World War II, divided government has become increasingly common in the American states. A significant component of the increase is the deterioration of Republican fortunes in state legislatures: after the 1990 and 1992 elections, for example, only five state legislatures were controlled by the Republicans. I shall examine the hypothesis that the professionalization of state legislatures makes legislative service more attractive to Democratic candidates and less attractive to Republican candidates, because full-time legislative service is incompatible with another career, and Democrats, on average, have less lucrative career opportunities than Republicans. Statistical analysis of post-World War II legislative elections outside the South is consistent with the argument: other things being equal, every ten-thousand-dollar increase in real biennial legislative compensation is associated with approximately a 1% increase in Democratic legislators.

Strategic Voting Equilibria under the Single Nontransferable Vote.

Abstract: Previous investigations of strategic voting equilibria in mass electorates have looked only at single-member districts. I shall investigate such equilibria in multimember districts operating under the single nontransferable vote system. What appear to be the most natural equilibria conform to the M + 1 rule, according to which strategic voting in M-seat districts produces exactly M + 1 vote-getting candidates in equilibrium, any others having their support totally undercut. This result provides the beginnings of a formal underpinning for Reed's recent extension of Duverger's Law to the Japanese case. The model also generates specific and empirically testable hypotheses concerning the exceptions to the M + 1 rule that one ought to expect in equilibrium. I test these hypotheses with Japanese data. Finally, the model also reveals a type of strategic voting that is specific to multimember districts. I use Japanese data again to explore the empirical importance of this kind of strategic voting.

Issues, candidate image, and priming: the use of private polls in kennedy's 1960 presidential campaign

Abstract: Interpretations of electoral campaigns have pointed to two mutually exclusive strategies: candidates are expected to focus either on policy issues or on personal image. We argue, however, that social psychologists' notion of priming offers an empirically grounded and theoretically plausible campaign strategy for treating image and issues as interconnected strategic concerns. Based on both quantitative and historical analysis of John F. Kennedy's 1960 presidential campaign, we find that the candidate's policy positions were related to results from his private public opinion polls. Archival and interview evidence suggests that Kennedy deliberately used these popular issues to shape the electorate's standards for evaluating his personal attributes (rather than to win over utility-maximizing voters). We conclude that the study of priming offers one important approach to reintegrating research on candidate strategy and voter behavior.

Bayesian inference for comparative research

Abstract: RV egression analysis in comparative research suffers from two distinct problems of statistical inference. First, because the data constitute all the available observations from a population, conventional inference based on the long-run behavior of a repeatable data mechanism is not appropriate. Second, the small and collinear data sets of comparative research yield imprecise estimates of the effects of explanatory variables. We describe a Bayesian approach to statistical inference that provides a unified solution to these two problems. This approach is illustrated in a comparative analysis of unionization. A widely used tool in quantitative comparative politics is the linear regression model, estimating relationships between institutional and economic variables observed for a set of countries constituting some population of interest. A brief survey of recent comparative research finds no shortage of work of this type. For instance, Remmer (1991) identifies economic crisis as a source of electoral volatility in Latin American countries in the 1980s; Robertson (1990) presents evidence that institutions moderating the transaction costs of negotiations between capital and labor reduce industrial conflict; and Golden (1993) finds that union wage moderation in industrialized democracies results from coordinated collective bargaining. Although the methodological advantages are well documented (Jackman 1985), we identify two distinct problems of statistical inference endemic to this type of quantitative comparative research. First, because the data constitute all the available observations from a population, conventional inference based on the long-run behavior of some repeatable data mechanism is not appropriate. Second, because data sets tend to be small and collinear, the effects of explanatory variables are estimated imprecisely. In most applications the first problem is generally ignored and the second is solved by imposing restrictions on the regression model to obtain "sensible" results. The Bayesian model of statistical inference provides a unified solution to these two distinct problems of quantitative comparative research. Two features of Bayesian inference are important in this context. First, probability is conceived subjectively as characterizing a researcher's uncertainty about the parameters of a statistical model. This subjective probability concept seems particularly useful in comparative settings, where the data are typically convenience samples not generated by a known probability mechanism such as random sampling. Second, Bayesian inference allows the inclusion of other information, in addition to the quantitative sample information. Again, this seems useful in the comparative context where rich historical material is commonly available, sparking ideas for researchers, but formally discarded in the final analysis. In the Bayesian approach, this historical material can be formally incorporated into the analysis. We begin by detailing why the nonstochastic and weak data typical of comparative research are problematic for conventional linear regression analysis. We next introduce some ideas about Bayesian inference and describe Bayesian regression analysis. These ideas are applied to a comparative analysis of unionization-the subject of a recent exchange in the American Political Science Review between Michael Wallerstein and John Stephens (Stephens and Wallerstein 1991; Wallerstein 1989). The Bayesian regressions we present are supplemented with a sensitivity analysis that investigates how our conclusions depend upon the sample data and our prior beliefs. Although we survey some ideas in Bayesian statistics, a full introduction to the Bayesian approach to statistical inference is beyond our scope here. (Good book-length treatments are Pollard 1986, Lee 1989 and, at a slightly higher level, Press 1989.) Our intention here is to spotlight problematic but ignored issues of statistical inference in a common area of application. The Bayesian alternative that we present represents just one way forward to address these inferential problems.

Enhancing democracy through legislative redistricting

Abstract: We demonstrate the surprising benefits of legislative redistricting (including partisan gerrymandering) for American representative democracy. In so doing, our analysis resolves two long-standing controversies in American politics. First, whereas some scholars believe that redistricting reduces electoral responsiveness by protecting incumbents, others, that the relationship is spurious, we demonstrate that both sides are wrong: redistricting increases responsiveness. Second, while some researchers believe that gerrymandering dramatically increases partisan bias and others deny this effect, we show both sides are in a sense correct. Gerrymandering biases electoral systems in favor of the party that controls the redistricting as compared to what would have happened if the other party controlled it, but any type of redistricting reduces partisan bias as compared to an electoral system without redistricting. Incorrect conclusions in both literatures resulted from misjudging the enormous uncertainties present during redistricting periods, making simplified assumptions about the redistricters' goals, and using inferior statistical methods.

The Reverse Coattails Effect: Local Party Organization in the 1989 Brazilian Presidential Election

Abstract: Increasingly, it is said that the main determinants of electoral outcomes are class, ethnicity, and religion and that local political organizations occupy only marginal roles in national elections. I assess the effects of local party organizations in the presidential election of 1989 in Brazil. Given the long hiatus in competitive politics, the absence of any parties linked to the country's previous democratic experience, and the weakness of citizen identification with political parties, Brazil should be a textbook example of the collapse of local political organizations. The presidential candidates, however, acted as if party endorsements mattered, and in the context of Brazilian politics, it was rational for municipal mayors to trade blocs of votes for future local benefits. Applying a series of increasingly complex models to the vote shares of the leading candidates, I show that all candidates did significantly better in municipalities where the mayor represented their party. I also show that spatial factors affect the tactics of local politicians, and I distinguish charismatic from purely organizational components of support.

Reassessing mass support for political and economic change in the former ussr

Abstract: Cet article reexamine les conclusions de Finifter et Mickiewicz (1992) a partir de donnees collectees en Russie, Ukraine et Lituanie entre 1990 et 1992. L'analyse montre une relation importante entre les orientations pour les reformes politiques et economiques. Ceux qui preferent des reformes politiques d'une nature democratique sont egalement partisans d'une reduction de l'intervention de l'Etat dans son role de garant du bien-etre social. De plus, les conclusions, en totale contradiction avec celles de Finifter et al., montrent que les sovietiques ayant recus une education sont plus susceptibles de preferer une responsabilite individuelle en matiere de bien-etre, que ne le sont les personnes moins eduquees

Individual and contextual variations in political candidate appraisal

Abstract: Tn this note we elaborate on the conditions under which on-line and memory-based strategies of political candidate evaluation can be implemented. We suggest that the structure of information . may be an important contextual variable affecting the voter's choice of these strategies. In addition, we propose that citizens with less political sophistication are particularly sensitive to structural differences in the political information environment. We use an experimental design that manipulates the information-processing context to test these ideas. Our results suggest that the context in which information is presented plays a critical role in moderating the influence of individual differences in political sophistication.

Peace, War, and the Balance of Power

Abstract: Every possible relation between the distribution of power and the likelihood of war has been defended somewhere in the literature on international politics: war is least likely if power is distributed equally, war is least likely if power is distributed unequally, and the distribution of power has no effect on the likelihood of war. I try to settle this dispute by examining the effect of expectations about the outcome of war on the choice between war and negotiation. I argue that each of these mutually contradictory propositions can be derived from some plausible set of premises and thus that which one is correct depends on which set of premises best describes a situation. The most important factors affecting the relation between the distribution of power and the likelihood of war are (1) whether the terms of a compromise agreement that might be accepted in lieu of war affect the distribution of power between the antagonists and therefore the probability that the agreement will be enforced and (2) how many states' interests will be affected by the outcome.

Popular influence on supreme court decisions

Abstract: In their 1993 article in this Review, William Mishler and Reginald Sheehan reported evidence of both direct and indirect impacts of public opinion on Supreme Court decisions. Helmut Norpoth and Jeffrey Segal offer a methodological critique and in their own reanalysis of the data find, contrary to Mishler and Sheehan, no evidence for a direct path of influence from public opinion to Court decisions. Instead, they find an abrupt-permanent shift of judicial behavior consistent with an indirect model of influence whereby popularly elected presidents, through new appointments, affect the ideological complexion of the Court. In response, Mishler and Sheehan defend the direct public opinion linkage originally noted, at both individual and aggregate level; respond to the methodological critique; and offer further statistical analysis to support the aggregate linkages.

The tower of babel as a coordination game: political linguistics in ghana

Abstract: G od's punishment for the audacity of building the Tower of Babel sets up a neat n-person coordination game for the world's scattered peoples. Individuals from each language group may want to communicate directly with members from other language groups, but they have to agree upon a common language to do so.' In world history, the Babelian curse has been negotiated in a variety of ways. The most common solution is based on simple market conditions: high levels of contact between peoples of different language groups induce new speech forms, permitting intergroup communication. Over a long expanse (e.g., of a river or through a subcontinent), linguists have discovered language chains. People at point A communicate effectively but haltingly with neighbors at point B, who themselves communicate with their neighbors at point C, and so on; but people at point A cannot communicate at all with people at point N. Dialect chains are not the only solution; often pidgins or lingua francas help overcome the Babelian curse.2 All these solutions are based on the microdynamics of

Report from the Laboratory: The Influence of Institutions on Political Elites' Democratic Values in Germany

Abstract: The unification of Germany revives several questions about the future of Germany's democracy. Given the socialist-authoritarian background, how supportive are East Germany's elites of liberal democratic rights? Has the socialist-democratic experience instilled into elites a social egalitarian conception of democracies? In what ways, if at all, do elites support direct democracy procedures? I examine political elites' conceptions of democracies in the united Germany in 1991, using a survey of 168 parliamentarians from the united parliament in Berlin. I find that the socialist and parliamentary institutions in the East and the West, respectively, have substantially influenced elites' conceptions of democracies in Germany, leading to a value divergence across the East-West boundary. Yet the findings also suggest that a partial value convergence in terms of liberal democratic rights among postwar elites has taken place. The results support an institutional learning theory, but they also suggest that support for liberal democratic values has been diffused into East Germany.