The Politics of Institutional Choice: The Formation of the Russian State Duma 9781400823949

Citation preview

THE POLITICS OF INSTITUTIONAL CHOICE

THE POLITICS OF INSTITUTIONAL CHOICE T H E F O R M AT I O N O F T H E R U S S I A N S TAT E D U M A

Steven S. Smith and Thomas F. Remington

PRINCETON UNIVERSITY PRESS PRINCETON AND OXFORD

COPYRIGHT © 2001 BY PRINCETON UNIVERSITY PRESS PUBLISHED BY PRINCETON UNIVERSITY PRESS, 41 WILLIAM STREET, PRINCETON, NEW JERSEY 08540 IN THE UNITED KINGDOM: PRINCETON UNIVERSITY PRESS, 3 MARKET PLACE, WOODSTOCK, OXFORDSHIRE OX20 1SY ALL RIGHTS RESERVED Library of Congress Cataloging-in-Publication Data Smith, Steven S., 1953– The politics of institutional choice : the formation of the Russian State Duma/Steven S. Smith and Thomas F. Remington p. cm. Includes bibliographical references and index. ISBN 0-691-05736-2 (alk. paper) — ISBN 0-691-05737-0 (pbk.: alk. paper) 1. Russia (Federation). Federal'noe Sobranie. Gosudarstvennaia Duma. 2. Legislative bodies–Russia (Federation) 3. Representative government and representation–Russia (Federation) 4. Russia (Federation)–politics and government–1991– I. Remington, Thomas F., 1948– II. Title. JN6697.8 .S63 2001

328.47'072–dc21

00-044118

THIS BOOK HAS BEEN COMPOSED IN BASKERVILLE TYPEFACE THE PAPER USED IN THIS PUBLICATION MEETS THE MINIMUM REQUIREMENTS OF ANSI/NISO Z39.48-1992 (R1997) (PERMANENCE OF PAPER) WWW.PUP.PRINCETON.EDU PRINTED IN THE UNITED STATES OF AMERICA 1 3 5 7 9 10 8 6 4 2 1 3 5 7 9 10 8 6 4 2 (Pbk)

CONTENTS

LIST OF ILLUSTRATIONS vi LIST OF TABLES vii PREFACE ix LIST OF ABBREVIATIONS xi ONE

Choosing Legislative Institutions in Russia 3 TWO

Forming a Parliamentary Party System 27 THREE

Creating the Council of the Duma 54 FOUR

Setting a Framework for Party-Committee Relations 72 FIVE

Choosing an Electoral System 93 SIX

Party Discipline in the Russian Duma 116 SEVEN

Institutional Choice 137 Appendix on Data and Methods 161 REFERENCES 169 INDEX 177

ILLUSTRATIONS

1.1.

Efficient and Redistributive Institutional Choices 13

2.1.

Policy Positions of Party-List Candidates, 1993 31

6.1.

Policy Positions of Factions, 1994–1995 125

6.2.

Policy Positions of Factions, by Mode of Election, 1994–1995 125

6.3.

Party Support within Factions, 1994–1995 126

TABLES

2.1. 2.2. 2.3. 2.4. 2.5. 2.6. 2.7. 3.1.

3.2.

3.3.

3.4. 4.1. 4.2.

4.3. 4.4. 4.5. 4.6. 4.7. 4.8.

5.1.

1993 Party-List Election Results and 1994 Faction Membership 32 Faction Voting on Selected Faction Threshold Proposals, January 1994 41 The Effects of Electoral, Policy, and Partisan Influences on Position on Faction Thresholds, January 1994 43 Faction Voting on Selected Faction Threshold Proposals, March 1994 45 The Effects of Electoral, Policy, and Partisan Influences on Position on Faction Thresholds, March 1994 46 Faction Voting on the Imperative Mandate, March 1994, by Faction 49 The Effects of Electoral, Policy, and Partisan Influences on the Imperative Mandate Vote, February 1994 50 Multivariate Logit Estimates of the Effect of Policy Position and Mode of Election on Attitudes about Institutional Arrangements among Russian Candidates, 1993 and 1995 61 Multivariate Logit Estimates of the Effect of Policy Position, Mode of Election, and Faction on Attitudes about Institutional Arrangements among Russian Deputies, 1994–1995 63 Multivariate Logit Estimates of the Effect of Policy Position, Mode of Election, and Faction on Attitudes about Institutional Arrangements among Russian Deputies, 1996 66 Logit Estimates of the Effect of Faction on Attitudes about the Council of the Duma among Russian Deputies, 1996 70 Support for Majority Bloc Control of Committee Chairmanships, Party-List Candidates in 1993 78 The Effects of Electoral, Policy, and Faction Influences on Key Votes on the Committee System and Chairmanships, January 1994 81 Committee Assignments by Faction Membership, April 1994 82 Committee Chairs by Faction Affiliation 83 Committee Membership by Faction, January 1994 84 Factions’ Role in Committee Assignments in 1994 85 Deputies’ Reports on Faction-Committee Relations in 1994 88 The Effects of Electoral, Policy, and Faction Influences on Preference for Council of the Duma Voting Rights for Committee Chairs, Late 1994 89 Results of December 1993 Parliamentary Elections 101

viii

5.2. 5.3. 5.4.

5.5. 5.6.

6.1. 6.2. 6.3.

7.1. 7.2. 7.3. 7.4. 7.5. 7.6. A.1. A.2.

L I ST O F TA B L ES

Simulated Results under Alternative Electoral Systems, Based on 1993 Election Results 103 Differences between Party-List Vote Share and Candidate Vote Share, by District, 1993 104 March 1995 Votes on the Presidential Amendment and Its Final Passage, Duma Election Law, April 1995 Faction Membership 109 March 1995 Votes on the Presidential Amendment and Its Final Passage, Duma Election Law, SMD and Party-List Deputies 111 Logit Estimates of the Effect of Policy Preferences, Mode of Election, and Party on Votes on the Presidential Amendment and Final Passage, Duma Election Law 112 OLS Estimates of the Effects of Policy Preferences, Faction, and Mode of Election on Policy Positions, 1994–1995 127 OLS Estimates of the Effects of Policy Preferences, Mode of Election, and Faction on Faction Support, 1994–1995 130 OLS Estimates of the Effects of Policy Preferences, Faction, and Mode of Election on Faction Support, 1994–1995, Excluding Nonparty Factions 131 Relationships between Deputies’ Goals and Attitudes about Institutional Features of the Duma, 1994–1995 Duma 139 Factional Membership in the State Duma, 16 January 1996 143 Leadership Posts in the Sixth Duma, by Faction, January 1996 143 Factional Voting Support for Unified List of Committee Chairmanships, Second Duma, 23 January 1996 145 1999 Parliamentary Election Results and Seat Distribution in 2000 by Party 150 Distribution of Leadership Posts by Faction, January 2000 151 Faction Representation in the 1994–1995 Deputy Survey 162 Faction Representation in the 1996 Deputy Survey 163

PREFACE

T

HIS BOOK is the product of the collaboration of two scholars— one a specialist in U.S. congressional institutions and politics, the other a student of Soviet and Russian politics—who found they shared an interest in the early development of the Russian parliament. Our collaboration began in 1992 as we studied the operation of the transitional Supreme Soviet in Russia. The convening of the new State Duma in January 1994, following the crisis of 1993, presented a rare opportunity to witness the impact of politics on institutional choice at a moment when a new legislature was being founded. Aware that such choices often have a lasting effect but that the specific reasons they were made are easily forgotten with time, we sought to take advantage of this historic opening. We wanted to investigate the dynamics of institutional choice in a new legislative body before its new rules and procedures had been established, while it was still possible to learn why they were adopted. We reasoned that Russia made a difficult but illuminating case for interrogating models of legislative institutions, difficult because of the instability of constitutional forms over this decade and rewarding because many of the architects of these changes are accessible and articulate. Most models of institutional choice in legislative settings used in political science were developed in the American context, making it challenging to separate the effects of social context from pure political strategy. Such models may therefore be context-bound and thus limited in their portability, or so parsimonious that they do not yield useful theories about choices in real-world settings. The deep dislocations that Russia has experienced in its political and economic system over the past decade give us confidence that if analytic models generated in more stable political environments yield productive explanations of outcomes in Russia, the models must be reasonably robust. We are convinced that Russian politicians have the same sorts of interests that motivate politicians in elective environments anywhere, and we argue that while these cannot be reduced to a single master objective function, politicians’ aims nonetheless comprise a broadly similar menu in many different national settings. Moreover, we claim, the strategies politicians pursue in deciding among institutional alternatives are shaped by the nature of the goals that are most immediately affected. This means that our models need to be grounded in time and place if they are to be specified sufficiently to yield testable propositions. Therefore, in each chapter we explain the political implications of the alterna-

x

PREFACE

tives the deputies were considering, as the deputies saw them at the time, in the light of their multiple and sometimes competing interests. Our research has benefited greatly from the help of others. First and foremost, we would like to acknowledge the extraordinary contributions that Moshe Haspel made to the project, both as a research assistant and as a collaborator on several of the papers we have published. Moshe succeeded in solving many of the intractable problems associated with compiling and analyzing our data. We have also received crucial funding support from several organizations. This enabled us to visit Russia many times in order to talk with deputies and staff members of the parliament, and other close observers of Russian politics, and to conduct surveys among candidates running for the Duma in 1993 and 1995 and with elected deputies in 1994 and 1996. We wish to express our appreciation for the support of the National Council for Soviet and East European Research (currently the National Council for Eurasian and East European Research), the International Research and Exchanges Board (IREX), the National Science Foundation (SBR-9631221), the University of Minnesota Graduate School, and the University Research Fund at Emory. These institutions, which have been generous and tolerant of our scholarly needs, bear no responsibility for the findings or conclusions reported here. We also extend our appreciation to many individuals in Russia who have shared their insights with us over a number of years. Among them are past and present members of parliament, including Vladimir Podoprigora, Nikolai Medvedev, Viktor Sheinis, and Vladimir Isakov. We are indebted as well to the director of the Parliamentary Library, Irina Andreeva, and her staff, who have helped us greatly in finding documentary materials. Friends and colleagues in Russia have been hospitable and patient in explaining the dynamics of the political process. Among them, we wish to single out the late Alexander Sobyanin, a pioneering analyst of parliamentary voting in Russia; we also thank Mark Urnov and Nikolai Petrov, who have themselves played major advisory roles to legislative and executive bodies; and we thank several colleagues at the Institute of State and Law of the Academy of Sciences, especially Vladimir Lysenko, Yuri Vedeneev, and William Smirnov, well as as Professor Alexander M. Yakovlev, who served as the president’s liaison to parliament in its first two years. To Igor Bunin and Boris Makarenko of the Center for Political Technologies, and Mikhail Zakhvatkin and Georgeii Satarov of INDEM, we extend our thanks and appreciation for their partnership in this enterprise. Finally, we wish to note our debt to Stanley Bach of the Congressional Research Service, a good friend whose involvement in the CRS’s exchanges with Russia’s Parliamentary Library in the early 1990s led to the happy accident of this collaboration.

ABBREVIATIONS

The following abbreviations are used in the text and tables for parliamentary factions and deputy groups.

AIG APR BRNI CEC CPRF DM DPR KEDR LDPR LDUD12 PPwr NDR NRP OVR PDep PRES RC RMDR RW ROS RReg SPS STB WR Yabloko

Agro-Industrial Group Agrarian Party of Russia Future of Russia/New Names Central Electoral Commission Communist Party of the Russian Federation Dignity and Mercy Democratic Party of Russia Constructive-Ecological Movement “Cedar” Liberal-Democratic Party of Russia Liberal-Democratic Union of December 12 People’s Power Our Home Is Russia New Regional Policy Fatherland–All Russia People’s Deputy Party of Russian Unity and Accord Russia’s Choice Russian Movement for Democratic Reforms Russia’s Way Rossiia Russia’s Regions Union of Rightist Forces Stability Women of Russia Bloc “Yavlinsky-Boldyrev-Lukin”

THE POLITICS OF INSTITUTIONAL CHOICE

ONE CHOOSING LEGISLATIVE INSTITUTIONS IN RUSSIA

T

HE TUMULTUOUS developments in Russia since the late 1980s have offered a rare opportunity to observe the creation of new legislative institutions—not just once but through a sequence of changes in the Soviet Union to the creation of the Federal Assembly and its two houses. To be sure, Russia is not the only place where democratization is under way, but developments in Russia are well documented by press accounts, official documents, and scholarly studies. We exploit this opportunity to explain the emergence of key features of the new legislative process in the Russian Federation. In doing so, we test and qualify competing positive theories of legislative institutions. Specifically we examine the fit of electoral, policy, and partisan accounts of legislative behavior and institutions to the choice of an electoral system, a leadership structure, a committee system, and floor behavior in the State Duma. The choice of institutional arrangements within the Duma illustrates the contingent nature of rational action by Russian politicians, the importance of sequence in the emergence of institutional options and the relevance of politicians’ goals, and the distinct contributions of several goals to the explanation of legislators’ behavior. While we are sympathetic to the bias for parsimony reflected in current positive theories of legislative institutions, we are critical of approaches that attribute a single dominant goal to political actors. We demonstrate that over a very short period of time, Russian parliamentarians crafted a new legislature that did not represent a single logic but instead reflected their multiple political goals. The challenge is to provide an explanation of institutional choices made in a context in which it is easy to demonstrate that electoral, policy, and partisan interests may be at work. In fact, as we move from feature to feature of the new Duma, potential electoral, policy, and partisan bases for the choices made are readily identified. We argue that the certainty of the relationship between a goal and the institutional choice varies widely but systematically. Plausible connections between, say, electoral interests and the formation of a parliamentary party system are complicated by the existence of other factors that interact with the party system to produce outcomes with implications for electoral interests. These fac-

4

CHAPTER ONE

tors are often unknown and their effects not understood. In other situations, however, the relationship between personal political interest and institutional choice is readily comprehended. When politicians can directly relate their personal political interests to proposed institutional features, and can do so with little uncertainty, these direct interests will trump other interests that are less directly and less predictably related to their goals. Politicians simplify choices, we posit, taking account of the relative clarity and relevance of alternatives to their goals. The core features of the Russian Duma, we show, are the product of the members’ many separate choices, with each choice structured by the goal or goals most directly and clearly affected by it. In this chapter we review the recent history of Russian legislative institutions, first outlining the theoretical problems involved in explanations of institutional change and then narrowing our focus to a specific set of events and the competing explanations for them. In chapters 2–6 we develop and test propositions representing different explanations for the behavior of Russian parliamentarians. In chapter 7 we return to the lessons that can be drawn about theories of institutional choice from the early stages of democratization in Russia.

Russian Parliamentary Institutions, 1989–1999 In 1989 Mikhail Gorbachev inaugurated competitive elections to the USSR’s hierarchy of soviets.1 Under the Gorbachev reforms the process culminated in elections to a reformed Union-level parliamentary body, the Congress of People’s Deputies. Although their fairness and competitiveness varied across the USSR, the elections produced a set of reformminded deputies who formed, for the first time in Soviet history, an independent political caucus in the national legislature. Under Gorbachev’s plan the Congress elected a much smaller full-time working parliament, called the Supreme Soviet, which was assigned responsibility for deliberating on proposals for new legislation required by Gorbachev’s political and economic reform program. Many of the newly elected USSR deputies took their duties seriously, some also helping to form political coalitions in the reformed parliaments elected in the national republics of the USSR in 1990. In the Russian Republic, which also adopted a two-tiered legislative structure consisting of a Congress of People’s Deputies and a Supreme Soviet, a substantial body of reformminded deputies was elected in the Russia-wide elections of deputies held in March 1990. Many of these reformers, joined often by moderate bureaucratic nationalists eager to acquire greater power over Soviet assets on Russian Republic territory, fought the Union government for

L E GI S L AT I V E I N S TI T U TI O N S

5

greater republican autonomy. Through 1990 and 1991 the Union-level and Russian-level parliaments fought over jurisdictional rights and policy direction for the country. Gorbachev’s parliamentary institutions allowed him to control parliamentary decisions. But Gorbachev needed to gain even greater power to overcome resistance to his policies in the Communist Party, the vast Soviet bureaucracy, and many republican governments, so he demanded that the parliament amend the constitution to create a presidency and that it elect him to fill the post. Gorbachev’s effort to gain greater control over Union institutions provoked Boris Yeltsin and his supporters to create a parallel set of institutional arrangements in the Russian Republic that would give them the means to check the power of the central government over the republic. Gorbachev’s parliamentary and presidential reforms therefore failed to contain the “war of laws” that undermined the Soviet state, and indeed tended to exacerbate it. Gorbachev’s hopes to revitalize Union-level institutions came to an end following the August 1991 coup attempt, when a group of hardliners in the Union government tried to seize control and restore central power. Although the coup collapsed within days, Gorbachev’s power as USSR president declined rapidly after this, while Yeltsin’s political star rose. Yeltsin, who had been elected chairman of the Russian parliament in June 1990 and subsequently president of Russia in June 1991, rallied deputies in the Russian Congress and Supreme Soviet in resistance to the coup leaders. In turn, Yeltsin demanded that the Russian Congress grant him emergency powers to enact radical reforms on Russian territory. This the Congress did at the end of October 1991. Yeltsin’s success was temporary. The Russian Congress refused to grant his demands for a new constitution that would grant him, the president, greater control over the government. Negotiations over the constitution were complicated by sharp differences regarding the balance of power between the central government in Moscow and the powers of the constituent units of the Russian Federation, themselves differing in constitutional status between privileged ethnic republics and ordinary administrative territories. These cross-cutting divisions, which would prove to be lasting, produced an impasse over a new constitution, with Yeltsin and Congress refusing to accept each other’s constitutional proposals. The conflict between Yeltsin and Congress intensified as the effects of Egor Gaidar’s economic stabilization program, initiated in January 1992, began to be felt. The impasse was resolved in September 1993, when Yeltsin issued decrees dissolving the Congress and Supreme Soviet and declaring that elections were to be held in December to a new parliament that did not, constitutionally speaking, yet exist. He also ordered a referendum on a

6

CHAPTER ONE

draft constitution that would establish the new institutions. The constitutional draft was similar to one hammered out by a large and broadly representative constitutional assembly, convened by Yeltsin in the summer of 1993 as a forum bypassing the Congress’s own constitutional commission. The version of the constitution drafted by the assembly gave Yeltsin much but not all the power he demanded. Yeltsin’s decrees dissolving parliament met determined resistance from a portion of the deputies. Some blockaded themselves inside the parliament building—the White House—and refused to recognize Yeltsin’s decrees as legally binding. The chairman of the parliament, Ruslan Khasbulatov, once an ally of Yeltsin but later a bitter rival, led the resistance. Following a ten-day standoff a body of paramilitary forces, loosely allied with the parliamentary opposition, stormed the barricades and joined the opposition camp inside the White House. With the help of supporters on the streets, they attempted to overthrow the Yeltsin regime. But the White House was shelled by Yeltsin’s forces, killing some inside, and the resistance leaders were arrested. Thus the December 1993 vote for deputies to the new Federal Assembly and the vote on the proposed constitution were conducted in the shadow of the events of October 1993. Nonetheless the constitution was approved by a majority of those voting, and most major political forces agreed to abide by its provisions.2 The constitution provided for a strong presidency, but with elements of a parliamentary system as well. The president named the head of government, but parliament’s confirmation of the appointment was required. Similarly parliament could deny the government its confidence, forcing the president to choose between naming a new government or calling new parliamentary elections. Parliament could initiate impeachment proceedings. The president lacked the power to call a referendum. However, the president had the constitutional authority to make policy in the form of decrees without seeking parliamentary approval, so long as these did not contradict either federal law or the constitution. It was a constitutional scheme with a decidedly presidential tilt, but by no means so one-sided as many observers believed. The new Federal Assembly, created by the 1993 Constitution, is bicameral. Earlier parliamentary models—the Supreme Soviet structures in the past—had been nominally bicameral, but the high centralization of control over parliamentary agenda and voting exercised through the Presidium vitiated any meaningful separation of voting patterns in the two chambers. The Federal Assembly, however, provided for two chambers with different numbers of seats, different means of election, and without a common leadership or executive committee. The lower or popular chamber, the State Duma, consisted of 450 seats. Half of these were filled in territorial single-member district races by plurality vote. The other 225

L E GI S L AT I V E I N S TI T U TI O N S

7

were filled by proportional representation from candidates elected on party lists, subject to the requirement that to receive any seats a party must win at least 5 percent of the party-list vote. There is no compensatory seat rule; each set of 225 seats is filled independently. The upper house is named the Federation Council. It is composed of two representatives from each of Russia’s eighty-nine territorial constituents, called the “subjects of the federation,” regardless of population or territorial status. The transitional provisions governing the 1994–95 parliament required that the members of the Federation Council be elected in at-large, two-winner races in each subject of the federation. However, a 1995 law provided that in the future the Federation Council would be comprised of the heads of the executive and legislative branches of each territorial unit. The terms of office for members of the Federation Council were therefore determined by their terms as elected officials in their home regions. The Russian constitutional system adopted in 1993 is a hybrid in several respects. As in other semipresidential systems, it combines presidential with parliamentary forms of executive power. Shugart and Carey (1992) would treat the Russian case as a “president-parliamentary” system, which they consider inherently prone to stalemate and breakdown because the government must answer to both the president and the parliament. It is also a mixed system in its electoral law. Both the electoral system decreed into place by President Yeltsin in 1993 and the new law passed by the Duma in 1995 (see chapter 5 below) employed a mixed plurality-proportional method of electing deputies to the State Duma. Finally, the new constitution mixes unitary and federal elements. The leaders of the regions are automatically given seats in the upper house of parliament, yet the constitution fails to define any reserved sphere of rights for subcentral jurisdictions. Russia’s case offers us an almost unparalleled laboratory setting, therefore, for testing the effects of institutional variation on the choice of new representative structures, since political and social context can be held constant in order to assess the impact of institutionally determined influences on actors’ strategic choices.

Explaining Institutions Explaining the evolution of institutions is a central problem of the social sciences. Institutions are sets of rules that guide behavior. They provide guidance by informing individuals and groups of the likely consequences of their behavior. Institutions lend order to interactions between individuals and groups, and interact with other institutions. They generate patterns in behavior by creating incentives of varying strength.3

8

CHAPTER ONE

Institutional change is as complex as social life itself. Institutions are nested within other institutions. At times institutions change thoroughly and quickly, at other times incrementally, almost imperceptibly. And the layers of institutions can change at different times and rates. In the case of a legislative body, the institution is located within a larger institutional setting, a single chamber is located within a bicameral legislature, and a legislative party or committee is located within a chamber of that legislature. Change affecting that legislature may come in many degrees and sequences. A presidential system might be radically changed into a parliamentary system, a committee system might be reformed wholesale, or the rules and practices of individual committees might be changed in a piecemeal fashion. To make matters more complex, institutions are grounded in both formal and informal rules. Most political institutions combine formal and informal rules to give some order to decision-making processes. For a legislature, constitutions and statutes, as well as chamber, committee, and party rules, combine with accepted practices to lend predictability to decision making. Informal rules are adapted to formal ones; formal rules emerge where informal ones are judged inadequate; and recognized precedents, sometimes recorded, serve to fill the gap between discretionary informal practices and more rigid formal rules. Processes described as steps outlined in formal rules may miss critical informal features that create biases. In a similar way, a mere description of informal norms probably will not capture features of important formal rules, which easily fade out of immediate view if they go unquestioned by the actors. These complexities of institutions make a comprehensive theory about the choice of institutions an unrealistic objective but should not discourage the development of theory on more narrowly defined features of a set of institutional arrangements. That is our approach. Because the constitution was imposed in short order and before the convening of the new parliament, we consider the constitution as exogenous to the choices of parliamentary institutions that followed in 1994. We also limit our attention to four important features of the new Russian Duma—the electoral system by which its members are elected, the organization of a system of parliamentary parties, the new Council of the Duma through which the Duma sets its agenda, and the relationship between standing committees and parties. By nearly any measure, these are among the Duma’s most important institutional features. The four sets of institutional choices have several important characteristics. First, all four features were determined by members of the Duma at readily identified times and places. In two of the cases, the outcome represented clear discontinuous change; in the other cases, the party

L E GI S L AT I V E I N S TI T U TI O N S

9

and committee systems, the Duma largely continued an inherited practice with important modifications. In all four cases, the members of the Duma ultimately voted to retain, modify, or reject inherited practices. Consequently an explanation of choices about these institutional features must place emphasis on the members themselves. We can more safely build a theory on the foundation of individual choices of legislators than we could for many other forms of institutional change. Second, all four institutional features were created by formal rules. We are setting aside those informal institutional features that develop gradually, are less likely to be adopted purposely, and, as North (1990) put it, are likely to develop through spontaneous interaction. Undeniably the behavioral strategies produced by informal institutions may carry over from one formal institution to another, and so may be critical to understanding the way the formal institutions do or do not change. But our focus on formal institutions allows us to define the institutions chosen and the alternatives that were discarded. Third, these changes occurred within a broader context of rapid change in the institutional environment. Radical changes in constitutions and an influx of a large number of new legislators, stimulated or imposed by Gorbachev and Yeltsin, compelled legislators to confront the problem of creating or re-creating institutions. While legislators could borrow from past legislative practice in structuring new legislative institutions, they could not avoid making explicit choices about which institutions should be continued and which should be changed. The more normal process of institutional change, it is often argued, is for actors to make marginal changes, first in informal rules and then in formal rules, often without consciously seeking relevant information (North 1990, 8). Fourth, the self-conscious process of choosing new institutions meant that legislators were likely to seek relevant information to evaluate their options and make choices as best they could. This is not to argue that legislators had complete information or could avoid acting on partial theories, subjective models, or biases acquired over time. Rather, more than in many or even most situations, legislators and other close participants articulated arguments, generated data, noted historical experiences, published proposals for change, and consciously evaluated alternative sets of rules. Finally, it bears noting that we are focusing on the early stages of a new national legislature. Technically all members were new to the institution, although some had considerable experience in predecessor institutions. The politicians making choices about the institutional setting were not merely readopting a set of rules inherited from the distant past that could be defended, if need be, on the basis of tradition and long experience.

10

CHAPTER ONE

Clearly these characteristics of recent institutional change in Russian parliamentary institutions limit the degree to which we can generalize to all forms of institutional change. We will not attempt to do so. But we do think that our cases are of more than parochial interest. Minimally, legislative bodies that adopt formal rules to govern themselves are likely to have much in common with the experience in the Duma. The characteristics also approximate the conditions assumed by the new economics of institutions, a class of theories initially developed in economics but now pursued in political science as well.4 Explanation is founded in an account of the interaction of purposive individuals. Individuals pursue strategies to achieve a goal (wealth, for the economist) in a particular institutional context (rules). They do so by acting consistently with the interest or goal when confronted with alternative strategies. The collective choice is explained as a product of the individuals’ strategies. If those strategies are stable for the individuals involved, the collective choice represents an equilibrium that will remain in place until some important feature of the strategic context (the individuals, their goals, information, and so on) change. Beyond the concept of equilibrium, two central features of the new economics of institutions are property rights and transaction costs. Individuals seek property rights—rights to control and use valuable assets— to increase the probability and certainty of success. Rights often are embedded in formal rules. They are enforced by a variety of institutional processes. These external devices are complemented by efforts of individuals and groups to multiply and protect their assets. The use of valuable assets often involves mutually advantageous exchanges, which in turn are based on contracts or rules to help specify and enforce agreements about those exchanges. Efforts to measure, exercise, enforce, and protect rights and assets entail costs—transaction costs. Transaction costs, which vary widely but are nearly ubiquitous, reduce the value of property rights and exchange, and bend strategies away from the efficient ones predicted by neoclassical theory. Institutions, by adding credibility to property rights, providing ready-made monitoring and enforcement mechanisms, and standardizing exchanges, may greatly reduce transaction costs and enhance the net value of the exercise of property rights. Put differently, institutions greatly influence the value of property rights and the severity of transaction costs. Organizations of many kinds form in society as individuals and groups seek to protect rights and reduce transaction costs.5 The central proposition of the economics of institutions is that transaction costs are substantial in human affairs (Coase 1960). Communicating, negotiating, recruiting, hiring, monitoring, enforcing, and other activities are essential to any coordinated action—from everyday group

L E GI S L AT I V E I N S TI T U TI O N S

11

decisions about social activities to the maintenance of a core task for a firm or organization to the functioning of parliaments and other national political institutions. They must be addressed by groups engaged in some collective activity. In institutional settings in which many individuals interact on a recurring basis, formal rules and elaborate informal rules are likely to emerge. These can have long-term consequences. Because the strategies of individuals and groups reflect the relative costs of interacting with others, the strategies will also reflect the institutional arrangements for managing transaction costs. Over time, individuals and groups will develop a vested interest in those institutional arrangements; that is, once means for reducing transaction costs are invented, they tend to be used to manage new costs as they arise. Parliaments are complex packages of rules that embody rights and restrictions governing collective choice under a decision rule, usually majority rule. Many rules concern rights of participation in decision making. Rules define the set of players, spell out permissible moves, and specify the sequence of play (Bach 1989; Tsebelis 1990, 93–94). Other rules concern the allocation of tangible resources—staff, access to information, and so on. Some rights are allocated by stage in the legislative process, such as those defining when proposals may be offered or amendments considered. Others are allocated by policy jurisdiction. Some establish special privileges for elected or appointed leaders; others are allocated to parties or factions. Agenda-setting processes, veto powers, and other features that allocate rights affect the nature of the options for choice offered to members. All four institutional features studied here involve the parliamentary equivalent of property rights. Elections determine who gains status as a parliamentarian and whatever resources come with the office. Rules governing the registration and privileges of parliamentary parties establish a variety of rights of participation in the agenda-setting process, in committee and floor decision making, and in interchamber negotiations. Membership on the Council of the Duma determines who has a voice in setting the chamber’s agenda. And the committee system may allocate control over policy jurisdictions and a set of parliamentary privileges for committee leaders and members. Managing transaction costs is an ever present problem in parliaments. Parliaments, of course, are created to provide a means to reduce insurmountable transaction costs associated with collective action by a society. But as multimember bodies with formal equality of membership, most parliaments confront both severe transaction costs and important constraints on how transaction costs can be managed. Formal and informal rules of many kinds reduce the associated transaction costs and lend predictability to parliamentary processes, as do leaders, committees,

12

CHAPTER ONE

parties, and other organizational structures invented by legislators. Parliamentarians, having invested in particular forms of these institutional arrangements, tend to use them to solve new challenges to collective action as they arise.

New Directions in the Economics of Institutions Institutional change designed to allocate property rights or reduce transaction costs may have one of three consequences: It may increase the collective welfare of parliamentarians, advantage some actors at the expense of others, or merely advantage some actors relative to others (Knight 1992; Moe 1990; Tsebelis 1990). Enhancing collective welfare represents the classic problem of coordinating individual or group interests to improve collective benefits—that is, to improve aggregate efficiency. Outcomes that advantage some actors more than others are a natural product of competition among individuals to advance political causes—that is, to realize redistributional gains. Tsebelis (1990, 104–5) illustrates the distinction in a way similar to that depicted in Figure 1–1. Changes in the upper right improve the condition of both actors, while other changes advantage one actor at the expense of the other. In the upper right, a change might improve the welfare of one actor more than the other even when both actors are at least somewhat better off, as we illustrate. In fact, we can readily imagine actors choosing among alternative efficiency-improving institutions, with the outcome reflecting the strength of the contending forces. Thus efficiency-enhancing improvements may generate intense conflict. The distinction between efficiency-improving and redistributive institutional changes is important for understanding democratizing systems in formerly command economies. A major theme of arguments for democratic reform is its importance for economic reform. Whether aggregate well-being recovers quickly, if at all, from the initial drop in living standards caused by deflationary stabilization programs depends, in large measure, on the distributional effects of the initial economic policies and the political outcomes generated by those effects. The “J-curve” model of the effects of radical economic stabilization policies is widely accepted, but there is debate over its political implications.6 Many observers call attention to the possibility that the political polarization occasioned by high adjustment costs from structural reform may lead to a renewed authoritarian rule, or at least a reversal in initially liberal reform policies (Haggard and Kaufmann 1995; Przeworski 1991; Przeworski et al. 1995). But, as Hellman (1998) notes, it is the “early winners” from

L E GI S L AT I V E I N S TI T U TI O N S

13

Figure 1.1. Efficient and Redistributive Institutional Choices

initial liberalization, those who capture rent-seeking opportunities, who are likely to provide a base of support for the new regime. The open question is whether the early winners are strong enough to deflect the regime from further liberalization. The credibility of initial government policies, which depends on who controls the institutions of government, also matters. A survey of economic performance from postcommunist countries found that, in general, the more decisive the breakthrough to stabilization, the earlier and more successful the economic results (Aslund, Boone, and Johnson 1996). One reason for this outcome is that economic actors judge the credibility of the government’s commitment to maintaining its reform course. A government that might be persuaded to relax budget constraints and reverse privatization generates weaker incentives for various economic and political actors to accept the costs of adjusting to a liberal system. A credible governmental commitment to reform speeds the adjustment process (Johnson, Kouvelis, and Sinha 1997; Diermeier et al. 1997). Plainly the distributional consequences of early economic reform decisions often generate intense political conflict among policy makers, even where long-term, aggregate welfare is enhanced by stable liberalizing policies. Parliament, and other institutions of government, are subject to demands from three separate groups: those who seek to lock in their initial gains in control over economic assets, those who seek to ease the pain of adjustment, and those who seek to reverse the initial distributional consequences of liberalization. Because institutions may have effects on

14

CHAPTER ONE

policy choices, the intense conflicts engendered by liberalization spill into conflicts over institutional arrangements. Intense conflict about both policy and institutions surely characterizes the first few years of the new Russian Federation and its parliament. But even in Russia, where deep divisions about economic reform permeated politics in the early 1990s and where we would expect complexity in the political goals of politicians and the dimensionality of the issue space, we might find some efficiency-improving institutional choices by parliamentarians. Parliamentarians, after all, may have certain common interests (preserving a career in parliament, perhaps) that lead them to fashion institutions that promote those interests. Moreover, whatever the divisions among parliamentarians over their initial institutions, parliamentarians would be expected to adjust to those institutions and, maybe very quickly, make investments in pursuing their personal goals under those rules. Thus, with time, expected gains and costs from changing the rules may become more similar across the membership, such that no majority coalition can be made better off by a significant change and institutional arrangements appear to stabilize. Still there are many potential sources of instability for institutional preferences among parliamentarians: change in the parliamentarians, change in the parliamentarians’ goals or environment, change in the larger institutional context of the parliament, and so on. In real time, the factors may change at least as fast as the opportunities for change in the procedures and structures of a parliament. Predicting the contours of conflict over institutional arrangements, such as the procedure and internal structure of parliament, is clearly not easy. With shifting societal conflicts, an evolving party system, and the emergence of new political elites, specifying what parliamentarians want and what the shape of alignments among them will be is nearly impossible. Parliamentarians may have common interests (such as in the power or resources of the parliament) that generate efficient institutional choices at times, while divisions rooted in economic and social factors produce institutional choices with strong distributional consequences. Our study contributes to the theory of institutional change in two ways. First, we seek to elaborate a theory of redistributive politics. The new economics of institutions has emphasized efficiency. Efficiency improvement through the creation of institutions often takes the form of solving a collective-action problem, such as a prisoner’s dilemma, public goods problem, or a tragedy of the commons problem. Institutions help to solve the problem of coordinating the behavior of individuals who otherwise would behave in self-interested ways that do not maximize their individual or collective welfare. This subject is well-developed theoretically, and we leave a review of the literature to others (see Tsebelis 1990, 107–

L E GI S L AT I V E I N S TI T U TI O N S

15

10). Redistributive changes, more frequently the concern of political science than of economics, have not been subject to general theory (Moe 1990). Redistributive politics must be approached in terms of the specific actors, their goals and preferences, and the procedures for changing rules. Redistributive politics is precisely where the emerging theory on legislative institutions is problematic and contested, and where the new economics of institutions is in need of extension. We hope to clarify the difference between efficiency- or redistribution-seeking in the policy domain, where intense conflict among the winners and losers from reform determines the stakes of institutional choice, and the efficiency- or redistributional consequences for parliamentarians of institutional arrangements when other goals are at stake. Second, we seek to make progress toward a theory of institutional choice. Studies that fall under the heading of new institutionalism, the new economics of institutions, positive political theory, or social choice share an important weakness—the literature on the consequences of institutions is much more developed than the literature on the choice of institutions. The comparative statics design of this literature is best illustrated in political science in the literature on electoral systems. Variation in electoral systems is associated with variation in the party system and candidate strategies (Cox 1997). The theory and empirical analysis of the study of electoral systems is perhaps the most rigorous and thoroughly developed of any field of institutional studies. Yet the literature on the choice of electoral systems is remarkably thin (see Lijphart 1992; Bawn 1993; Geddes 1996; Remington and Smith 1996; Cox and Neto 1997). The working assumption of the scholarship in this field seems to be that the consequences shape the choices, and yet the choices themselves have been subject to systematic investigation in only a few systems.

Electoral, Policy, and Partisan Interests in the Duma Although wealth maximization is usually taken as the unproblematic goal of individuals, firms, and organizations in theories of the economy, goals have proven problematic in theories of politics. The value of service in a parliament to the parliamentarian may reside in many attributes of the institution. The policy-making process may produce valued legislation, provide the means to block implementation of unfriendly policies, or merely provide a way to advertise long-term policy goals. The contacts and visibility of membership may be valuable assets in the pursuit of higher office. The opportunity to participate, power, or the perquisites of office may be valued in themselves. In addition, the parliament’s or party’s success may be valued—perhaps for its own sake but

16

CHAPTER ONE

also for instrumental reasons. An effective parliament or parliamentary party might be viewed as essential to democracy, for example, or may be instrumental to the parliamentarian’s policy or electoral goals. In other legislative contexts these possibilities are reflected in a variety of contrasts—party-ideology-constituency, reelection-policy-power, and so on.7 The theoretical temptation is to posit that parliamentarians sum up the values of parliamentary service and base their strategies on the value of the alternative behaviors that might be pursued. Unfortunately such a general utility maximization model (equivalent to a theory of consumer tastes) yields no testable predictions about institutions. Utility must be given specific arguments. Moreover, such complex accounts of microlevel political motivation are an obstacle to theorizing about the macrolevel phenomena. Sociologist James Coleman (1990, 19) properly observes that a student of social systems must choose between a complex psychological view of the individual and a rich account of social systems. Given our current state of knowledge in the social sciences, the complex profile of the individual may yield insights into the individual decision maker but serves as an obstacle to aggregating individuals and assessing the outcome of individuals’ interaction. To build a viable theory, scholars of the American Congress have usually posited some specific goal as dominant and then generated propositions about the individual behavior and institutions. Predictably, different goals have produced contrasting theories. In fact, they typically are brought to bear on somewhat different aspects of legislative institutions, which has created some misunderstanding of the relationship between goals and institutions. Three general perspectives underpin the alternative approaches, which provide a useful guide to our study of institutional choice by Russian parliamentarians.8 A Policy Perspective The first model emphasizes variation in the distribution of policy preferences among the legislators as a driving force in institutional change. Institutions are created and changed as the players seek rules that facilitate the achievement of their established policy goals. In this case, the distribution of policy preferences among deputies of the Duma would shape their choice of parliamentary institutions. Spatially, legislators’ policy preferences and policy proposals can be treated as points in a unidimensional or multidimensional space. Legislators then favor policies close to their own preferences and choose institutional arrangements that facilitate the adoption of such policies.

L E GI S L AT I V E I N S TI T U TI O N S

17

The policy model is a distributional perspective that emphasizes the conflict over policy and process. Expectations for institutions turn on the precise nature of the conflict. A cohesive majority would be expected to choose a majoritarian governing system that restricts the rights of minority factions or parties. It might also establish a strong central leader who, on behalf of the majority, directs the activities of the committees and staff. In the absence of a cohesive, policy-based majority, deputies would be unwilling to create a strongly centralized or majoritarian decision-making process that could be used against them. The governing structure might limit the power of a central leader, ensure ample rights for minorities to influence the legislative agenda, and widely distribute resources such as committee chairmanships and staffing. The absence of a highly centralized decision-making process and the adoption of a party-based, power-sharing arrangement in the Council of the Duma suggests that the Duma lacked a cohesive majority bloc after the 1993 and 1995 elections. Theories of institutional choice founded on assumptions that actors attempt to maximize policy utility by manipulating institutional arrangements yield tidy predictions of institutional choices only to the extent to which they yield tidy predictions of policy choices. If the conditions under which the median voter theorem applies are met—that is, actors’ policy preferences are well ordered and single-peaked, and the preferences can be arrayed along a single dimension—then the equilibrium outcome of the social choice is identical with the median voter’s preference. The median voter, who determines the policy position of the majority, also determines the procedures under which policies are approved. Krehbiel (1991) labeled this relationship the principle of remote majoritarianism. In a multidimensional space, no single median voter may exist, majorities are unstable, and point predictions of policy and institutional choices are not possible. As Riker (1980) observed, the instability of majorities is transferred to the institutions themselves. Still, depending on the inherited rules and alignment of preferences in the multidimensional issue space, the range of likely policy and institutional choices is not unlimited and is often quite restricted. A warning is in order. The policy perspective should not be equated with spatial theory but rather should be viewed as a special application of spatial theory. In the typical spatial theory of legislative politics, preferences determine choice at the instant of decision and may be influenced by any number of political considerations—including but not limited to long-term policy commitments. Thus the policy model is consistent with the typical operationalization of preferences in tests of spatial theories, but the policy perspective should not be equated with spatial theory.

18

CHAPTER ONE

A Career/Reelection Perspective A second perspective rests on the assumption that legislators seek reelection (Mayhew 1974) or other political offices. The reelection or career perspective posits that legislators seek reelection and do so until they pursue a more highly valued office. They also seek institutional arrangements that enhance the prospects of reelection. Whether institutions are designed to meet the electoral needs of all or just some legislators depends on whether reelection is a positive- or zero-sum game for legislators. The Russian system is especially important for the examination of electoral incentives because of the mixed electoral system for the Duma. Fifty percent of the deputies are elected from party lists (PL) and 50 percent from single-member districts (SMD). The electoral fate of PL deputies is dependent on party—their party’s electoral success and their placement on the party list. Deputies affiliated with different parties are in direct competition with one another for reelection; all else being equal, an increase in the support for one party reduces the number of seats allocated to other parties. In contrast, the electoral fate of SMD deputies depends entirely on election outcomes in local districts. They do not run directly against one another and may prefer institutional arrangements that enhance every SMD deputy’s prospects for reelection. Thus the party list system tends to create a zero-sum, redistributive game in parliamentary politics, and the SMD system creates incentives for positive-sum, efficient institutional choices, that is, arrangements that enable all members to be made better off. It is noteworthy that in all-PL and all-SMD systems the opportunity to evaluate the effects of electoral incentives is far more limited than in mixed electoral systems. In most PL systems, the electoral incentives for legislators to adhere to a party line are so strong that there is little variance in individual behavior to observe and explain. In many SMD systems, two sources of variation appear to be significant. The first is the means of nomination. Differences in the party’s control of the nomination—placement on the ballot—affects the degree of independence from the party experienced by legislators. Party control of nominations not only varies across political systems but also may vary among parties within a party system, although this latter source of variation is likely to be limited over the long run. A second source of variation is in the political composition of districts. In American-styled systems, variation in local constituencies (or at least local political conditions for legislators) provides a basis for predicting differences in behavior, but all legislators confront the same general strategic problem of garnering plurality or majority support in a geographically limited district. Mixed systems provide far greater variation in the electoral context and therefore greater opportu-

L E GI S L AT I V E I N S TI T U TI O N S

19

nity to observe the consequences of the mode of election for preferences about parliamentary institutions. For expectations about the institutional implications of these two sets of deputies, we can borrow from the extensive literature on the subject. PL deputies should favor a system that is generally party-oriented and gives more power to central party leaders, who are needed to maintain and enhance party reputations and who have the leverage with PL deputies to gain compliant behavior. PL deputies see reelection at least partially as a zero-sum game among the parties. SMD deputies prefer freedom to associate with parties as they wish, freedom from the binding constraints of strong parties, and freedom to pursue issues of concern to local constituencies in a more decentralized decision-making process. They are more likely to see reelection as a positive-sum game for deputies. We look for a relationship between the participants’ mode of election and institutional preferences. A Party-Based Perspective A third class of influences on institutional choice derives from the collective benefits of party membership. Like other public goods, partisan benefits may be provided inefficiently if entrepreneurial party leaders do not assume the costs and gain the benefits associated with organizing and maintaining a party following. The literature on parties in the United States Congress suggests, however, that party leaders may produce policy and electoral benefits that their fellow party colleagues cannot gain individually (Shepsle and Weingast 1995, 170–74; Cox and McCubbins 1993). Party organizations and leaders may help members pursue policy and electoral goals more efficiently, such as by structuring the agenda to reduce the tension between goals when divisive issues arise. In the American case, however, it is argued that the electoral or policy interests within legislative parties are diverse. Consequently legislators will not create party organizations that are capable forcing them to behave in a manner inconsistent with their personal electoral or policy interests. In the study of many non-American parliamentary bodies, the individual legislator is ignored in favor of a strictly party-based theory of legislative politics. Near-perfect party discipline leads scholars to set aside individuals and to theorize directly about party strategy. Individuals are assumed, usually implicitly, to have compatible electoral and policy interests; individual differences can be ignored without significant consequences. For example, the theories of the formation of government coalitions generally treat the parliamentary party as a unitary actor and attribute political objectives or goals to the party (Laver 1999).

20

CHAPTER ONE

The Russian case requires that we retain the individual legislator as our unit of analysis. The system of parliamentary parties—or factions, as they are officially known—was decided, in part, by votes cast on the floor of the Duma during the first three months of its first session. Plainly we cannot take as given the existence of any particular mix of parliamentary factions when the deputies so explicitly, and individually, cast votes to shape that mix. We consider these choices in chapter 2. Still, partisan interests may require attention. In fact, we argue that Russian partisan interests are not fully decomposable into electoral and policy interests of individual legislators, for two reasons. First, if parliamentarians find themselves better off by pursuing goals collectively, they may be willing to sacrifice something in order to realize those gains. The gains from collective action create incentives for parliamentarians to accept collective electoral or policy strategies that are not entirely consistent with the strategies they would have pursued acting individually. Moreover, the strength of these incentives may vary among parliamentary factions. Evidence for varied partisan effects is scattered throughout the book and is more fully developed in chapter 6, where we directly consider variation in party discipline. Second, partisan interests are not entirely a function of the collective interests of the members of parliament associated with the parliamentary faction. Most Russian parliamentary factions are associated with electoral parties that predate the formation of parliamentary factions and have an organizational presence external to the parliament. Parliamentarians rely on these parties, to varying degrees, for a place on the ballot, a position on the party list, campaign resources, a network of political contacts, and so on. And many parties are led by prominent individuals who can promote the career of a parliamentarian inside or outside the parliament and politics. With resources of their own, acquired outside parliament, parties and their associated factions may be able to influence legislators’ strategies. Partisan effects are considered throughout this study. As for electoral and policy interests, the affected partisan interests usually are transparent for institutional choices examined. But their influence is likely to vary among factions and across the institutional choices confronting members of the Duma. Common Interests, Convergent Goals, and Efficient Institutions Policy, electoral, and partisan interests need not generate divisions among parliamentarians. As we have noted, common interests may generate a choice of institutions that make most or all parliamentarians better off. We must recognize this possibility. Parliamentarians may have readily settled on the internal procedures and structures of the Duma.

L E GI S L AT I V E I N S TI T U TI O N S

21

However, they did not, at least not on some of the most fundamental features of the new parliament. Still, the divisions that emerged may have rested on judgments about what was best for the parliament collectively, judgments that were distributed more or less randomly among the parliamentarians. For that reason we test the importance of policy, electoral, and partisan differences against the null hypothesis that none of those differences shaped preferences about the institutional choices considered by parliamentarians. Our working hypothesis is that policy, electoral, and partisan differences among parliamentarians structured their preferences, in varying ways, about the internal organization of the Duma. In asking whether legislators’ various goals lead to different strategies, we are questioning a working assumption of many political theories. A common approach to theorizing about politicians’ goals is to assume that the various goals are mutually supportive. In theories of the U.S. Congress, for example, it is convenient to assume that legislators’ electoral interests and personal policy views dictate the same strategies. A theory that posits that legislators’ maximize their chances of reelection is essentially the same as one that posits that legislators’ seek to minimize the distance between their policy ideal points and the collective outcomes. Similarly, in theorizing about the choice of institutions in four central and Eastern European systems (other than Russia), Geddes (1996) collapses all politicians’ goals into one—career advancement. She argues: In advancing this argument, I do not deny that political leaders represent the interests of constituents and prefer some policies to others. But for politicians considering institutional changes, interest in furthering their careers usually converges with interest in achieving the policy goals of their constituents. The same institutions that will improve their chances of winning elections will also improve their chances of achieving policy goals, since the greater the likelihood that they and their party allies will be elected, the greater the chance of passing the legislation they favor. (18)

We are sympathetic to this argument. Goal or incentive compatibility surely is a factor in self-selection for political careers, in the recruitment efforts of leaders and parties, and in the competitive electoral process. Liberals are more likely to run and to win in liberal districts, for example. Missing from this reasonable approach is the recognition that stable institutions—electoral rules, decision-making processes, and so on—facilitate the selection of politicians with internally consistent goals. Of course there is reason to be suspicious of this claim, even in stable systems. Fenno (1973) demonstrated, for example, that the specific goal matters in predicting the committee-related strategies of members of the U.S. Congress. More important, when institutions themselves are unstable or yet to be chosen—when the rules of the game are uncertain—

22

CHAPTER ONE

goal convergence is likely to be more problematic. Politicians may have difficulty determining whether their multiple goals, or the multiple interests placing demands on them, are compatible. And as they confront a series of choices about basic institutions, they may find that they must choose between valued objectives.

Layered Institutions, Evolving in Time The early development of Russian parliamentary institutions illustrates one way in which the political context—defined institutionally and temporally—allows actors with multiple and partially incompatible goals to behave rationally. We argue that single-goal theories cannot explain important features of choices about parliamentary institutions. While there are sound reasons for grounding theory on assumptions about the electoral and policy goals of legislators, our thesis is that the importance of each of these influences varies with political context. The thesis rests on several propositions about goal-oriented legislators. First, legislators are motivated by multiple political goals. Legislators not only often have strong personal policy preferences but they weigh their electoral, career, and influence interests in the balance as well. Therefore their behavior is a weighted function of their personal policy preferences, the preferences of electorally relevant actors (party officials, contributors, and constituents), and the preferences of career- or influencerelevant actors (party leaders, the president, and so on). Second, legislators are opportunists. Legislators do not find all their goals equally relevant to each of the policy or institutional choices they confront. Variation in goal relevance, transaction costs, and information across contexts leads legislators with complex goal profiles to pursue goals opportunistically. A single goal may often dominate other considerations in the context of a choice about a particular feature of parliamentary institutions. Third, the value of an institutional feature to legislators’ goals may vary with time. This is the result of two processes. One is that legislators’ investments in an institutional feature—a party or committee system, for example—accumulate over time. Sunk costs in a particular means of managing transaction costs reduce legislators’ willingness to adopt new arrangements. A second is that conditions change in unforeseen ways. For example, the partisan balance of forces within the parliament may shift so that the implications of a set of rules change. Fourth, legislators are locally, not globally, rational. Choices about institutional arrangements in the immediate environment may have consequences for a broader set of institutional arrangements, but legislators’

L E GI S L AT I V E I N S TI T U TI O N S

23

(and scholars’) ability to evaluate the institutionally and temporally distant consequences of a decision about immediate rules is limited. The quality of both the essential information and the theory required to evaluate that information is likely to deteriorate as legislators consider the more distant consequences of institutional alternatives. Goals relevant to immediate institutions will be given greater weight in legislators’ decision-making calculus. In short, legislators’ strategies are the product of their goals, the weight assigned to each goal, the relevance of particular choices to their goals, and their certainty about the consequences of their choices for their goals. As the weight, relevance, and certainty of a goal increase, the more important that goal is in a legislator’s choice. If these considerations vary systematically and similarly among legislators, as we argue they often do, then some institutional choices will reflect the effect of one goal while other choices will reflect another. More than one goal may be critical to an adequate explanation of an institution’s evolution, even over a short period. Our argument should not be confused with the “nested games” thesis of Tsebelis (1990). Tsebelis asserts that (scholarly) observers often err by defining the game and strategic options confronting decision makers too narrowly. By so doing, observers too readily label behavior nonrational and turn to other explanations of the decision makers’ behavior. Once the larger set of nested games and strategic options are properly specified, either by recognizing the multiple arenas in which actors participate or by discovering how actors changed the rules of the game, the rationality of the strategic choices becomes transparent. Critically, however, Tsebelis notes that perfect information about the larger set of nested games allows actors to assess their interests and produces redistributive outcomes. Imperfect information about the long-term consequences of institutions leads actors to hedge their bets and accept efficient institutions. Tsebelis’s argument is compelling, but our emphasis differs in an important way. Although we agree that good theory must account for the nesting or layering of “games” and institutions, Tsebelis limits his theoretical analysis and empirical examples to single-goal episodes. As we demonstrate in the following chapters, none of the usual political goals— policy, electoral, partisan—successfully explains preferences and strategies about all important parliamentary institutions. We explicitly account for multiple goals, which present decision makers with obstacles to rational choice whether they are embedded in a single game or in nested games. And we argue that the importance of the various political goals varies systematically and predictably.

24

CHAPTER ONE

If goal relevance varies from choice to choice, so will the nature of the relevant information. The strategic context of institutional choice must be defined for the specific choice to be made. For that reason we separate our discussions of the Duma’s choices about a party system, an agendasetting committee, the committee system, and the electoral law. Even then we find that specific decisions about each of these institutional features vary in goal relevance and informational context.

A Preview Most Russian parliamentarians readily articulate and conspicuously pursue policy, electoral, and partisan goals. The relevance of the goals and the certainty of their implications vary widely across the institutional choices parliamentarians confront in establishing a new parliament. Specifically, (1) the choice of an electoral system has direct and predictable consequences for their reelection and partisan goals but uncertain effects on policy goals; (2) the internal party and committee systems have implications for policy, electoral, and partisan goals; and (3) everyday voting choices have direct and predictable consequences for policy but uncertain effects on electoral goals. As a result, the choice of an electoral system is most likely to be shaped by legislators’ electoral goals, the choice of party and committee structures within the parliament are likely to be influenced by multiple goals, and everyday floor voting is most likely to reflect parliamentarians’ policy preferences. Thus institutions are not merely the projection of political goals at play in everyday policy making. In chapter 2 we analyze Russian deputies’ early decisions in creating a system of party factions with privileged status under Duma rules. Deputies debated at the outset the distribution of rights of representation to political factions and to other deputy associations that were not arms of electoral associations. One of the most contentious disputes was whether all small deputy groups would be allowed to enjoy the rights of party factions. Another was how and whether parties should be represented on the steering body of the parliament. The result was a generally partyoriented but nonmajoritarian arrangement, a system that has proven durable since its establishment in January 1994. Chapter 3 then details the creation of the Duma’s steering committee, called the Council of the Duma. Deputies decided not to give committee chairs voting rights on the Council, but also not to represent factions and groups proportionally. The Council was denied some of the gatekeeping and agenda powers enjoyed by its predecessor, the Presidium, but it was established as the central management institution for both

L E GI S L AT I V E I N S TI T U TI O N S

25

political agreements and housekeeping, and the chamber’s chairman was declared its agent. We discuss the creation of a system of standing committees in chapter 4. We analyze the chamber’s decisions as to how leadership positions were to be distributed and the relationship between party factions and committee work. We test alternative explanations for the finding that committees were given relatively weak jurisdictional rights vis-a`-vis the floor, the Council of the Duma, and the parties. In chapter 5 we assess the deputies’ choice of a new electoral law. Despite pressure from the president to increase the proportion of singlemember district seats in the Duma, and on the part of other forces to increase the share of party list seats, the deputies chose to retain the equal balance of list and SMD seats. In explaining this outcome, we weigh the relative importance of electoral, policy-based, and partisan factors. Chapter 6 returns to the issue of partisanship and asks how to account for the relatively high degree of cohesiveness we observe in floor voting. A simple preference-based explanation would hold that parties themselves add nothing to an explanation derived from knowledge about the homogeneity of ideological outlooks among members. An electoral explanation would suggest that deputies with similar electoral backgrounds would tend to vote alike. A third possibility is that party affiliation produces an additional, observable degree of cohesiveness that neither a purely preference-oriented or electoral/career rationale can offer. We test this possibility and offer reasons for the results we find. The final chapter extends our analysis of institutional choice in the State Duma to the new Dumas elected in 1995 and 1999. The new deputies were confronted with choice over whether to replicate the same set of governing arrangements as in 1994 or to replace them with new ones. For the most part, the deputies chose to continue the same arrangements—the Council of the Duma, the faction system with the same threshold for groups to acquire factional rights, an identical election system, comparable committee rules. We suggest reasons for the apparent stability of these early choices, summarize our general argument, and reflect on the implication of our findings for the future direction of theory about the development of legislative institutions.

NOTES 1. The literature on the political events of the 1989–93 period is extensive. See, for example, Hough 1997; Colton and Hough 1998; Lowenhardt 1995; Sakwa 1996; and White, Rose, and McAllister 1997.

26

CHAPTER ONE

2. The voting on the constitution was marred by a number of irregularities, some very serious. Inexplicably, the number of voters enrolled on the registration lists maintained by the Central Election Commission fell by two million between April 1993, when a referendum was held on approval for Yeltsin and his government, and the December vote. Whether coincidentally or not, the lower number of registered voters meant that the threshold for a valid result—50 percent of all registered voters—was thus substantially easier to reach. Vote tallying in the regions was left in the hands of officials controlled by regional executives, who had their own interests in the election (many were running for seats in the new upper house) and who were under heavy pressure from the presidential administration to ensure both a sufficient turnout and a favorable result on the constitutional referendum. All the ballots were destroyed shortly after the election, preventing any recount, and evidence, both circumstantial and statistical, suggested that fraud was at a high level in the voting. Still none of the major political parties called for a recount. 3. An accessible introduction to the extensive literature on institutions may be found in Shepsle and Bonchek 1997. 4. Eggertson (1990) provides a review of this literature. 5. It is the introduction of property rights and transaction costs in the new economics of institutions that has widened the scope of neoclassical economics. 6. The J-curve refers to the initial sharp drop in living standards (plotted on the Y-axis) as price controls are lifted, taxes raised, and government credits, subsidies, and spending cut. In principle, time (on the X-axis) brings about a gradual improvement in living standards as the economy recovers and growth is stimulated. For a comprehensive discussion, see Przeworski 1991. 7. Students of American politics have long sought to measure the effects of party, ideology, and constituency influences on the behavior of individual members of Congress. The list of relevant studies is long. See MacRae 1958, Clausen 1973, and Kingdon 1973. These studies reflected an attitudinal model common to behavioral studies in the 1950s, 1960s, and 1970s. The distinction between reelection, policy, and power originated with Fenno 1973. 8. Kaare Strom (1990) shares our general perspective on the goals of political actors. In his case the issue is the basis for predicting the formation of government coalitions in parliamentary systems. We owe our perspective to the work of Fenno (1973) on the motivation of members of the U.S. Congress and the shaping of institutional arrangements of the committees on which they serve.

TWO FORMING A PARLIAMENTARY PART Y SYSTEM

P

ARLIAMENTARY PARTIES are often taken for granted. Generally political science has treated parliamentary parties as mere byproducts of electoral systems and the parties they generate.1 The number and internal features of parliamentary parties are determined by the rules governing elections and the internal features of the “parent” electoral parties. Less frequently, the constitutional and societal bases for parties are emphasized, although the problem of disentangling the electoral, constitutional, and societal influences on the identity and number of electoral parties has proven difficult.2 But even when a more complex view of electoral parties is pursued, the nature of parliamentary parties is assumed to be a near-perfect reflection of those electoral parties. Mixed systems like Russia’s present special difficulties in attributing cause and effect in the relationship among system characteristics, electoral parties, and parliamentary parties. Cross-cutting incentives to create, multiply, or consolidate parties were generated by an electoral law that provided for a mixed party-list and single-member-district system and constitutional provisions that combined elements of both presidential and parliamentary systems. In 1993 and 1995 these factors generated multiple parties, only a few of which managed to win seats in the Duma. Many of the elected deputies were tightly bound to their electoral parties, whereas some deputies were elected with no or only weak party attachments and others were elected with multiple affiliations. How these deputies would fit in the new system of parliamentary parties was not predetermined. To the contrary, the deputies themselves would craft rules that would give shape to the Duma’s system of parliamentary parties. Our purpose in this chapter is to determine how electoral, policy, and partisan interests shaped the choices deputies made in 1994 about parliamentary parties.3 We first provide essential background on the Russian electoral and party systems. We then note existing theories of electoral and party systems and discuss the application of these theories to the Russian case. We emphasize that the sequence of events—parliamentary elections, creation of parliamentary institutions and factions, formation of electoral parties, and subsequent presidential and parliamentary elections—appeared to have been critical to the alternatives posed to deputies and the choices they made about the Duma’s system of factions. We then turn to the specific decisions—faction size and faction control over

28

CHAPTER TWO

membership—made in the early weeks of the new Duma in 1994. We find that distinct electoral, policy, and partisan interests shaped deputies’ choices. Policy and partisan interests were strongly reinforcing, and electoral interests shaped an important feature of the hybrid system.

The Emergence of Russian Parties The radical transformation of Russia in the five years preceding the 1993 parliamentary elections greatly influenced the formation of parties in 1993. Still visible behind Russia’s multiparty system were the dim outlines of a two-party system, both in the elections and in the organization of parliamentary parties. Beginning with the mobilization of radical antiestablishment sentiment in the 1989 and 1990 elections by democratic political activists, political organizations and parliamentary factions associated with market-oriented and liberal democratic values challenged Communists, ultra-nationalists, and representatives of the traditional bureaucratic power structures (Colton and Hough 1998; McFaul 1993; McFaul and Markov 1993; Tolz 1990; Brudny 1993; Chiesa and Northrop 1993; Kiernan 1993; Fish 1995). The 1990 elections for the Russian Congress of People’s Deputies pitted a coalition of democratic reformers, called Democratic Russia, against a bloc of socialists and nationalists, both in the national elections and in many local races. These elections, as in Soviet elections in the past, were conducted in single-member districts. The system, it appears, created incentives for reformers and their opponents to form coalitions large enough to win the elections in the districts, in most cases in the second, run-off round of voting. The bipolarity of the campaign carried over into the alignments of deputies in the Congress and the Supreme Soviet (Remington et al. 1994; Sobyanin 1994). The transitional parliament elected in 1990 allowed deputies to organize formally into groups, factions, and blocs. Under its rules, groups could be any bounded membership body, based, for example, on common regional, professional, or generational ties. Factions referred to groups sharing a common political program. A deputy could be a member of multiple groups but only of one faction. Blocs, in turn, were coalitions of three or more factions. More than a dozen factions formed and gained formal rights in floor procedure. In fact, the factions gained sufficient status that they established a council of factions and consulted on matters before the body. Despite the proliferation of factions, a strongly bimodal distribution of ideological positions among deputies can be detected in the Supreme Soviet and at most of the Congresses. By late 1992

PA RL I A ME N TARY PART Y SY S T EM

29

a reliable majority against Yeltsin’s reform program had emerged in the Congresses (Remington et al. 1994). The transitional parliament was unceremoniously disbanded but the early experience with parliamentary factions had several lasting consequences. First, the identity and even some of the membership of some factions carried over into the new parliament. With the exception of the Communists, these initial factions were not associated with established electoral parties. The rules of the parliament helped to establish the principle that deputies had the right to organize themselves. Second, the parliament experimented with rules that set a minimum membership requirement for factions and granted factions certain parliamentary privileges, such as speaking rights in plenary sessions. Third, factional representation played some role in structuring the commissions on the constitution and electoral law. Although the work of these commissions did not directly generate the 1993 constitution and electoral law, the debates and drafts greatly influenced the decisions of the Yeltsin camp. And, finally, several factions of the transitional parliament formed electoral parties to compete in the 1993 elections and gain seats in the new Duma. The proportional representation (PR) system imposed for the 1993 elections was explicitly designed to change the parliament’s composition but proved controversial among Yeltsin’s advisers. The proponents of PR noted that a PR system would increase the prospects of top democrats to be elected to the parliament. Democratic leaders were drawn largely from an urban political elite and were expected to have trouble recruiting quality candidates and mounting effective campaigns in many local districts. The proponents argued, moreover, that Communists probably were greatly advantaged in local contests because of their extensive local organization, their ability to nominate known personalities in nearly all corners of the country, and their influence over local election officials. Furthermore, they insisted, conducting PR on a nationwide basis, establishing stringent eligibility requirements for placement on the party ballot, and setting a reasonably high threshold of votes for party representation in the Duma would reduce the chances that parties with a strong regional or ethnic appeal would win seats. The group favoring singlemember-district (SMD) elections argued that the democrats could expect to do well in the elections and would benefit from the bonus in seats usually experienced by the favored party in SMD elections. They also hoped that the splintering among democratic forces that might be bred by a PR system could be avoided with SMD elections. The presidential decree providing for the December 1993 parliamentary elections provided for half of the 450 seats in the lower house,

30

CHAPTER TWO

the Duma, to be chosen on the basis of nationwide party-list voting and proportional representation and for the other half to be chosen from single-member districts.4 The two types of electoral mandate differed sharply. The PR component of the vote used a single, nationwide electoral district, with a threshold of 5 percent. It was a system designed to encourage both the creation and consolidation of national parties. Access to the party-list ballot was tightly controlled by party elites, although candidates were allowed to run on both the party-list and district ballots. Voters had no opportunity to register a preference for one candidate on the list over another in the party-list voting. The arrangement of candidates on the list, therefore, was fixed, which gave the leaders who structured the list a free hand in placing loyalists at the top. In the singlemember districts, by contrast, candidates’ party ties were weak. Indeed, many candidates responded to the voters’ antipathy toward parties by emphasizing their political independence (Colton and Hough 1998; White, Rose, and McAllister 1997). Plainly the newly elected PL and SMD deputies had widely varying campaign experiences. In such a setting, therefore, we would expect the type of electoral mandate to shape politicians’ preferences for certain parliamentary institutions. The parties offered candidates with very different sets of policy preferences. During the campaign of 1993 we conducted a survey of candidates appearing on party lists and district ballots (see the appendix for details on the survey and on related methods of inquiry). The survey included questions on contemporary policy issues. The principal component that was identified in a factor analysis clearly captured the left/right, or reform/anti-reform, dimension. A boxplot of the party-list candidates’ scores, organized by party, is shown in Figure 2.1 for every party for which there were more than ten party-list respondents. The shaded boxes contain the 50 percent of values falling between the 25th and 75th percentiles, and the lines that extend from the box indicate the highest and lowest values, excluding outliers. The bar within the box signifies the median for the faction. Clearly a polarized alignment pitted the large parties on the left—the Communist Party of the Russian Federation (CPRF), Vladimir Zhirinovsky’s Liberal-Democratic Party of Russia (LDPR), and the Agrarian Party of Russia (APR)—against a few large parties on the right—Russia’s Choice (RC), Bloc “Yavlinsky-BoldyrevLukin” (Yabloko), and the Party of Russian Unity and Accord (PRES). The party-list results are shown in Table 2.1. The table also lists the total seats for each parliamentary party, a number that includes SMD deputies who chose to affiliate with a parliamentary party. Unquestionably the PR system had the predicted effect—deputies from multiple parties were elected to the Duma. Moreover, a majority of the SMD depu-

PA RL I A ME N TARY PART Y SY S T EM

31

Figure 2.1. Policy Positions of Party-List Candidates, 1993 Source: Principal Component, Responses to Policy Questions, 1993 Candidate Survey.

ties eventually affiliated with a parliamentary party, although a sizable minority chose to form new parliamentary groups (New Regional Policy [NRP], Russia’s Way [RW], and the Liberal-Democratic Union of December 12 [LDUD12]). However, neither of the two major party blocs— listed in the table as reform-oriented parties and opposition parties— came close to taking a majority of seats in the Duma. The reformoriented parties comprised just over one-third of the Duma, and the opposition parties made up about 40 percent.5 The mixed PR/SMD system had several important consequences. Most obvious was that the PR system gave Zhirinovsky’s party, the LDPR, a strong position in the Duma. This nationalist party garnered 23 percent of the nationwide PR vote, although it fielded very few SMD candidates and won only one district seat. As Yeltsin strategists expected, the other major opposition parties, the Communists and the Agrarians, did much better in the SMD contests, winning 34 percent of the SMD seats but only about 20 percent of the PR vote. Democrats’ strength in parliament was evenly balanced between seats won in districts and on party lists; they won about 34 percent of the seats in both the PR and SMD contests. The manner in which the elected parties would organize themselves and the new Duma was an open question as the elections passed. In the autumn of 1993 President Yeltsin appointed a commission to study the issue and to prepare rules and an agenda for the new parliament. Guided by the draft constitution and the electoral decree, the commission proposed draft rules that placed parties at the center of the new Duma.

32

CHAPTER TWO

TABLE 2.1 1993 Party-List Election Results and 1994 Faction Membership Party Reform-Oriented Parties RC PRES Yabloko Centrist Parties DPR WR Opposition Parties APR CPRF LDPR

Party-List Party-List Voting Percentage Seats Reserved

Faction Seats in April 1994

15.51 6.73 7.86

40 18 20

73 30 28

5.52 8.13

14 21

15 23

7.99 12.40 22.92

21 32 59

55 45 64

Parties Failing to Reach the 5 Percent Threshold Civic Union 1.93 Future of Russia/New Names 1.25 DM 0.70 KEDR 0.76 RMDR 4.08 Factions Forming after Convening of the Duma NRP LDUD12 RW

0 0 0 0 0 66 26 14

Abbreviations: RC: Russia’s Choice; PRES: Party of Russian Unity and Accord; Yabloko: Bloc “Yavlinsky-Boldyrev-Lukin” (The names of its three leaders); CPRF: Communist Party of the Russian Federation; LDPR: Liberal-Democratic Party of Russia; KEDR: Constructive-Ecological Movement “Cedar”; RMDR: Russian Movement for Democratic Reforms; NRP: New Regional Policy; LDUD12: Liberal-Democratic Union of December 12.

Although the commission’s draft rules were modified in important ways before adoption by the Duma, the final version retained the principles of a party-dominated chamber. The term the president’s commission, and later the Duma, used for parliamentary parties was faction (or fraktsiia—equivalent to the German Fraktion), referring to the parliamentary organizations of electoral parties. The term was used by the transitional Supreme Soviet and Congress, whose deputies had been elected without electoral party ties. After 1993, when elections used a list system for national electoral parties, deputies

PA RL I A ME N TARY PART Y SY S T EM

33

granted automatic faction status to members of any political party or bloc that had won at least 5 percent of the party list vote and had thus gained parliamentary representation for its list candidates. Deputies elected in single-member districts could also join that faction whether or not they had been nominated and supported by that party. Deputies who did not affiliate with factions could form deputy “groups” that were independent of extraparliamentary parties. This option was particularly attractive to those deputies elected in single-member districts who chose to remain politically independent. But if members of such groups wanted to acquire the same rights and privileges for their groups that party factions enjoyed, they had to meet a minimum membership threshold. As a result, the Duma’s first session saw a debate over the vital procedural question of the threshold number of members that would be required for such a deputy group to be granted the same rights as those the parliamentary faction of an electoral party automatically held. Throughout this book we refer to both party factions and registered deputy groups as factions, although official parliamentary documents take pains to distinguish the two. The presidential commission initially proposed that the steering body for the Duma be a council comprising the leaders of the factions. The commission was strongly influenced by its members’ reaction to the autocratic practices of the chairman of the transitional Supreme Soviet and Congress, who exercised his power through the Presidium—an executive committee comprised of committee chairs that held broad agenda-setting powers and controlled the administrative apparatus of the parliament. The proposed rules provided for only limited powers for the Duma’s chair and did not create a central Presidium. Members of the commission envisioned the Duma’s new governing structures as horizontal—based on agreements among the factions—rather than vertical—exercised hierarchically through the chairman, the Presidium, and the committees. As it turned out, opposition factions, which played no part in planning the new parliament but were stronger than Yeltsin’s strategists had expected before the elections, deferred little to the commission’s recommendations. Indeed, the Duma’s newly elected chairman, Ivan Rybkin, had been a leader of the Communist faction in the old Supreme Soviet. By the end of its first six months the Duma had developed two distinct systems of operation: a strongly party-oriented system for leadership and coordination, and a committee-oriented system for the drafting of legislative details. The important question for us is this: Did the electoral, policy, or partisan interests of the deputies shape the key decisions about factions in early 1994?

34

CHAPTER TWO

Russian Parliamentary Parties in a Comparative Context The literature on comparative political systems offers some guidance on what to anticipate in the development of the Duma’s party system. However, the mixed electoral system muddies expectations for the developments in late 1993 and 1994. While Russia was not immune to the same general political principles operating in other political systems, the outcome for Russia was greatly influenced by a particular sequence of events. Research on political parties (in both stable and transitional settings) has focused on four factors that influence party systems: the electoral system, the arrangement of political institutions, the alignment of policy or ideological preferences in the electorate, and the degree of stability in the electoral system, institutions, and the electorate (Remington and Smith 1995).6 Previous research justifies the following expectations: • Proportional representation systems, as a general rule, produce more parties than single-member-district systems. • Other features of electoral systems are of critical importance, including district magnitude, one-stage versus two-stage elections, electoral thresholds, suffrage, and parties’ eligibility requirements. • The degree of party control over ballot access—nominations—influences the cohesiveness of parliamentary parties. • Factionalized party systems are associated with greater cohesiveness in parliamentary parties. • Consolidation of parties is encouraged by a single, elected, and powerful presidency and by holding presidential and parliamentary elections at the same, or nearly the same, time. • Parties form and position themselves in response to the distribution of voters across the one or more dimensions that organize political attitudes—unidimensional (unimodal or bimodal) alignments encourage a two-party system, whereas multidimensional and multimodal alignments encourage multiple parties. • A configuration of parties, laws, and institutions may tend to reproduce itself or to change, depending on the interest of the dominant coalition of political actors in retaining the existing system or overturning it. In sum, electoral law, voter preferences, constitutional arrangements, and the stability of the relations among these create powerful incentives for party behavior. The impact of these features on the direction of the development of parliamentary parties is not straightforward. We consider their influences on two issues that members of the new Duma would confront—

PA RL I A ME N TARY PART Y SY S T EM

35

the required size of parliamentary factions (and thus the number of factions) and deputies’ obligations to adhere to factional policy positions (imperative mandate). Electoral Incentives Historical precedents do not provide a solid foundation for predicting the manner in which the mixed PL/SMD electoral system was to be translated into a system of parliamentary factions.7 The multiple parties likely to exceed the 5 percent threshold were expected to be represented in the Duma with corresponding parliamentary factions. That much the commission seemed to settle at an early point. The more difficult issue was the creation of additional parliamentary parties that SMD deputies might seek to form. Some SMD candidates, after all, were running with multiple-party endorsements; others were running with no party endorsement at all. The most that could be predicted with confidence was that some hybrid arrangement combining party factions with groups for independents would be established. In the abstract, we would expect party-list deputies, whose nomination and election depended on electoral parties and their leaders, to support strong parliamentary factions, if for no other reason than the fact that strong factions would enhance the leaders’ influence. In contrast, SMD deputies would be expected to seek institutional arrangements that allow them to pursue the interests of their local districts, unfettered by the demands of strong factions and leadership. In practice, the deputies’ mode of election is likely to be relevant to some but not all features of the Duma’s faction system. To be apparent, the differences in mode of election had to generate differences in incentives that were relevant to the institutional choices that deputies actually confronted. The rule governing the threshold of group size for faction status had only indirect consequences for the relationship between a faction and its members. The primary consequence would concern the desire of independent deputies, almost all of whom were elected in districts, to organize factions of their own, which would allow them to avoid joining a party-based faction in order to enjoy the parliamentary advantages of faction membership. Thus a low threshold would serve their interests. Yet those SMD deputies who associated with an electoral party, some of whom actually ran on the party list as well, might be expected to join the party’s faction and share the party’s views about the faction system, whatever they might be. Party leaders and party-list deputies would be expected to favor a party-oriented system of factions, but there is no obvious reason why party-list deputies would support or oppose a high threshold for the registration of nonparty factions apart from their

36

CHAPTER TWO

general interest in preserving the rights exercised by party factions for as small a number of deputy groupings as possible. Deputies’ electoral interests would be more directly implicated by a decision regarding the imperative mandate.8 The imperative mandate refers to party-list deputies’ obligations to their party factions. If a partylist deputy refuses to join the party’s faction, leaves the party’s faction, or is excused from the faction by action of the faction, the deputy must vacate his or her seat in the Duma. To do otherwise, the argument goes, would undermine the electoral mandate granting a party a certain number of seats. This view would give a party faction the ability to expel a deputy from a faction, and thus from the Duma, for whatever reason the faction determined, including failure to support the faction’s policy positions. A credible contrary view was articulated in the March 1994 debate on the matter. The Duma’s own rules, by the time the issue was raised, allowed a deputy to choose a faction freely and even allowed a faction to “donate” members to another faction so that the second faction could reach the size threshold necessary for registration. These rules, it was argued, undermined a rigid interpretation of the electoral mandate. Furthermore, some insisted, all deputies—PL and SMD alike—should be equally free from undue pressure and to vote according to their own judgment. An imperative mandate would potentially affect party-list deputies directly and SMD deputies indirectly, but for neither set were the consequences simple or uniform. Party leaders might be expected to prefer an imperative mandate because it would increase their influence over rank-and-file list deputies. To the extent that they already exercised influence over party-list deputies, the party-list deputies could be expected to support the imperative mandate. But, it almost goes without saying, party-list deputies also might not like granting central party leaders the leverage that the imperative mandate would provide. SMD deputies would not be directly affected by the imperative mandate. It would not affect factions comprised of independent SMD deputies at all and could not be used to punish them for changing factions or opposing leaders’ policy positions. Yet SMD deputies, with their weaker party connections and stronger local orientation, might prefer weaker over stronger party leaders. And party-list deputies, beholden to party leaders, may not have any choice but to support the leaders’ positions. Thus the following propositions seem reasonable: Proposition 2.1. The relationship between mode of election and preferences on the threshold for faction status is weak.

PA RL I A ME N TARY PART Y SY S T EM

37

Proposition 2.2. The relationship between mode of election and preferences for the imperative mandate is strong, with opposition to the mandate coming more from SMD deputies than party-list deputies. Policy Incentives Policy differences were at the center of the dispute that led to the demise of the transitional parliament and the creation of a new constitution in 1993. The new constitution was designed to strengthen the president’s decree-making power in many areas of public policy and to enhance his bargaining power in areas that remained subject to legislation (Remington, Smith, Haspel 1998). The Yeltsin camp also hoped that the 1993 elections would yield a parliament that was, at a minimum, less strongly opposed to its reform program than was the transitional parliament. Moreover, the significance of the system of factions for policy was understood before the Duma convened. Most deputies and observers expected that an agenda-setting committee would be created for the Duma based, for the most part, on party or faction representation (see chapter 3). Consequently we would anticipate a strong relationship between the deputies’ policy preferences and their attitudes about a system of factions. The policy implications of the faction-size rule stemmed from the election results and views about how a strong Duma would serve policy goals. Haspel (1998) argues that deputies of the left saw the Duma as their primary vehicle for blocking President Yeltsin’s reform program. The members of the three large parties on the left—the CPRF, APR, and LDPR—viewed large, cohesive factions as the foundation for an opposition bloc that would have a plurality of deputies. The deputies may also have had in mind the past year’s debate over the constitution (Remington, forthcoming). The Communists had advocated drafts that provided for a strong Soviet-style parliament and a weak chief executive or none at all. Consequently deputies on the left were likely to view a low threshold as a potential threat to maintaining their large factions. Moreover, a low threshold would encourage the organization, representation, and procedural status of independent deputies, few of whom were expected to support the left. One would expect the policy significance of the imperative mandate to be related to the identity of the factions likely to benefit most by retaining political leverage over party-list deputies. As Table 2.1 indicates, the LDPR, situated to the left of the major left-right, or Communist-versusreformer, political axis, was the only party with a large number of deputies that included a very large percentage of party-list deputies. The CPRF also counted among its deputies about two-thirds party-list deputies.

38

CHAPTER TWO

Clearly a rule that granted factions effective control over the membership status of deputies would advantage the left over the right. Deputies of the left were therefore more likely to support the imperative mandate than were deputies of the right. Thus the following proposition is suggested: Proposition 2.3. Deputies of the left are more supportive than deputies of the right of a high-membership requirement for factions and of the imperative mandate. Partisan Incentives A complication throughout our analysis of the political foundations of preferences for institutional arrangements is the strong relationship between policy preferences and their partisan or factional affiliations (see Figure 2.1). Very little overlap exists in the policy positions of members of factions on the left and right. Moreover, the large factions associated with parties are polarized to the ends of the spectrum. The opposition factions—the CPRF, LDPR, and APR—had similar policy preferences on the left-right dimension. Plainly the “factional” interests of faction members are likely to be intimately connected with the deputies’ shared policy interests. Still, there are good reasons to posit that factional interests apart from shared policy goals may influence deputies’ attitudes about the structure of the Duma’s factional system. For example, for the factions associated with parties, leaders were simultaneously parliamentary faction and electoral party leaders. Party leaders would be motivated to consider the consequences of parliamentary faction activity for the party’s reputation and ability to attract quality candidates. Even common policy goals might require compromise among generally like-minded deputies, something that can be facilitated by faction meetings and leadership. Moreover, some party leaders would be motivated to fashion a system of factions that supported their pursuit of higher office. For all these reasons, factions and their leaders may pursue strategies that extend beyond, or are even incompatible with, policy goals. Finally, it is important to recognize the wide variation in the organizational strength of the parties associated with parliamentary factions. Even if we set aside the nonparty factions that formed after the faction-size rule was adopted, the relationships among party, faction, leader, and rank-and-file members ranged from strong and longstanding to weak and new. The CPRF and LDPR are distinctively strong-party factions, although for different reasons. Members of those party-based factions can be expected to have a clearer view of party interests and be influenced more effectively by party leaders.

PA RL I A ME N TARY PART Y SY S T EM

39

The CPRF was based on the established communist party structure inherited from the Soviet and transitional eras. No other party or faction could match the organizational resources of the CPRF, which continues to be the closest approximation to a European-style party, possessing a defined program, membership, constituency, and organization. Deputies of the CPRF received substantial electoral support from their party during the 1993 campaign. The APR combined partylike and trade association attributes but could call on a national network of collective farm administrators to mobilize electoral support. Yabloko, in contrast, was an ad hoc alliance of three parties with three prominent politicians who united in order to compete more effectively in the 1993 election campaign. Russia’s Choice (RC) had a prominent leader in Egor Gaidar, the former acting prime minister and a leading reformer, and could rely on a national network of activists, but they were concentrated in the major urban areas. Elsewhere the party had a thin organization, if any. The LDPR had a highly visible and charismatic leader in ultra-nationalist Vladimir Zhirinovsky, but it had little organization and virtually no district candidates, as we have noted. Zhirinovsky personally constructed his party’s list. Still other groups that competed and won party-list seats either were ad hoc coalitions constructed after Yeltsin’s decree, such as Women of Russia (WR) and the Party of Russian Unity and Accord (PRES), or were small parties formed around prominent leaders, as was the case with the Democratic Party of Russia (DPR). Thus we can expect to find intrafaction cohesiveness beyond the dictates of individuals’ electoral and policy interests and to find that strong parties would be expected to show greater cohesiveness on any issue relevant to faction interests. Rules affecting the parties’ parliamentary factions surely would be issues that affect factional interests. Proposition 2.4. Faction (or proto-faction) membership has an independent effect on voting on the faction size and imperative mandate rules. Proposition 2.5. Strong-party factions have greater cohesiveness than weak-party or nonparty factions in voting on the faction size and imperative mandate rules.

Duma Action on the Faction Size and Imperative Mandate Rules On the opening day of the new Duma, deputies approved the provisional rules proposed by the presidential commission. The rules provided that any electoral association that had elected deputies on a national party list could form a faction of deputies in the Duma, no matter how few members it had. Accordingly the smallest, the Democratic Party of Russia,

40

CHAPTER TWO

enjoyed the same formal status as did the largest, Russia’s Choice. With only a few exceptions, deputies elected on party lists joined a corresponding faction early in the session. Nearly all other Duma deputies, those elected from districts, either registered with one of the eight party factions that had elected party-list deputies or joined one of the other groups that eventually formed.9 But the Duma then revisited the issue and cast a series of votes on the faction-size rule. The Faction-Size Rule The presidential commission recommended that the threshold for faction registration be set at the size of the smallest party-list faction, which was fourteen seats. This proposal was challenged by Zhirinovsky, whose LDPR had the largest number of party-list deputies. He asserted that party-list deputies deserved a special status and argued in favor of a high threshold that recognized the electoral success of his and other parties: Factions were already elected by the population in 12 December. No one can ever deny the will of millions of voters. As far as there are independent deputies, they wish to form into groups. But naturally, these will be formations on regional characteristics, professional, even perhaps gender, but under no circumstances political, because the voters have already rated you as independent, party-less deputies, who do not join with any party in our country. If you join, it means that you are violating the will of your voters. And on the subject of the numbers. Here there were two numbers: some want fourteen, while six political parties proposed fifty. Some want to orient themselves on the parties that received the minimum of votes from the voters. But we propose [that we choose a standard] by the party that received the most votes of all.10

Zhirinovsky quickly declared that a threshold of fifty would be approved by his faction. That position was endorsed by leaders of the CPRF and the Agrarians. Gennady Ziuganov, the CPRF leader, predicted that a low threshold, as suggested in the temporary rules, would cause a split in some large factions, increasing the number of factions which in turn would paralyze the work of the Duma (Haspel 1998, 51). In contrast to the leaders of the opposition parties, the leaders of the reform factions and independent deputies argued in favor of a low threshold. A threshold of twelve was endorsed by Anatoli Shabad, a leader of Russia’s Choice, by Sergei Shakhrai of PRES, and by many independent deputies. One independent deputy protested the organization of any factions, insisting that any party-based structure violated the purpose of a parliamentary body.

PA RL I A ME N TARY PART Y SY S T EM

41

TABLE 2.2 Faction Voting on Selected Faction Threshold Proposals, January 1994 (percent in favor) Faction CPRF APR RW LDPR WR DPR NRP PRES LDUD12 Yabloko RC

Threshold of 50

Threshold of 14

Threshold of 35 (voted last)

97.6 95.9 100.0 100.0 94.7 100.0 18.6 0.0 4.8 18.2 1.6

0.0 0.0 12.5 0.0 0.0 0.0 62.1 96.6 95.2 65.0 97.0

97.7 94.0 100.0 100.0 100.0 100.0 60.3 13.8 4.8 57.9 3.1

Note: Based on the April 1994 faction membership; factions are listed in order of mean policy position, from anti- to pro-reform. See appendix. Source: Roll-call voting in the Duma.

In a series of votes on the first day of the session, January 11, 1994, deputies rejected both high and low alternatives and adopted a thirtyfive-member requirement for the registration of a faction. The outcomes for three of these alternatives are reported by eventual faction membership in Table 2.2. We include three groups that formed after these temporary rules were adopted: the NRP, Russia’s Way, and the LDU of December 12. Our measure of faction (or group) affiliation is membership as of April 1994, when most members had decided on their affiliation. For a small number of deputies, therefore, the influence of factional affiliation in our analysis reflects a prospective rather than actual influence, which should bias our estimates of partisan effects slightly downward and slightly magnify the effects of policy and electoral factors. Several features of the voting patterns deserve notice. First, factional cohesion is very high; in several cases, faction members voted in lockstep fashion. This is noteworthy because they were voting before the faction system was fully formed. (This result tends to support the inference that members joined those factions that were most hospitable to their own political interests.) These votes occurred in a sequence in which members voted alternately on size options that were high (fifty-, forty-five-, forty-member, and so on) and low (fourteen-, twenty-, twenty-five-member, and so on) before finding the lowest level—thirty-five members— that could still command the support of the high-threshold camp while also winning enough votes from the low-threshold camp to gain a majority. Once the thirty-five-member threshold won a majority, the voting stopped and that threshold was adopted as the rule.

42

CHAPTER TWO

Second, there is a clear left-right orientation to the voting. We have listed the factions in Table 2.2 in order of the mean policy position (based on a dimensional analysis of roll-call voting in the 1994–95 term). Left factions clearly supported the higher thresholds, but so did two centrist factions, Women of Russia and the Democratic Party of Russia, both highly dependent on party-list deputies for their membership. Thus, in addition to policy-based considerations, electoral effects are clearly evident. Our questions are the following: How are these electoral and policy considerations combined for particular factions? Are there distinctive factional effects above and beyond those that are clearly associated with members’ policy and electoral positions? To determine whether the interfaction differences were significant, independent of electoral and policy considerations, we report multivariate estimates from a logistical regression equation (Table 2.3). The dependent variables are the three votes listed in Table 2.2. In all models, party effects are captured by dummy variables, with the CPRF as the excluded category, so that the party coefficients measure differences from the CPRF. We also estimated a model that captures the joint effect of policy and electoral interests within factions by using a set of interaction terms for faction by policy position and mode of election, but factional cohesion tended to be so high in these votes that virtually no variance within factions was detected in this way. Therefore we do not report the model with individual interaction terms for factions. We also attempted to determine whether an aggregate measure of the left-right policy positions of individual single-member districts added any explanatory weight to the model. The row labeled “Constituency” in Table 2.3 notes the coefficients for this variable; none is strong or significant. The analysis excludes deputies who were not members of a faction in April 1994, just after these votes occurred.11 Additional measurement details are provided in the appendix. Unfortunately the deck is heavily stacked against finding faction effects. Several factions are small, which makes it difficult to find statistically significant differences between factions. There is no way to adequately adjust for this, but we can carefully scrutinize the size and signs of faction coefficients. Substantial policy effects are present, consistent with Proposition 2.3. Mode-of-election effects are also present, although they are stronger than Proposition 2.1 predicted. Factional effects are weaker than foreseen by Propositions 2.4 and 2.5, and are significant only in the case of the NRP (made up of SMD deputies), which was significantly more likely to oppose the high threshold levels and thus to reflect the judgment of “independent” SMD members. We found that PRES, which also had a sizable share of SMD members, tended to oppose the higher thresholds

PA RL I A ME N TARY PART Y SY S T EM

43

TABLE 2.3 The Effects of Electoral, Policy, and Partisan Influences on Position on Faction Thresholds, January 1994 (logit estimates) Independent Variables Faction APR RW LDPR WR DPR NRP PRES LDUD12 Yabloko RC Mode of election Policy position Constituency Constant N −2 log likelihood Chi-square

Threshold of 50 .01 11.52 7.57 −.59 8.59 −2.00* −9.44 −.22 .29 −.90 −3.34* 3.38* .11 2.01* 371 123.8 388.9*

Threshold of 14 Threshold of 35 (voted last) .34 −.47 −6.06 1.06 −7.49 1.44* 1.65 1.51 −.68 1.38 2.57* −3.24* .22 −2.25* 374 156.0 354.0*

−.62 8.55 6.72 −.31 8.00 −1.14* −2.17 −2.14 1.31* −1.92 −1.05* 2.47* .26 1.69* 372 166.7 322.9*

p < .05 Source: Roll-call voting in the Duma.

*

and support the lower ones (although, given the faction’s small size, the effect does not meet the criterion of statistical significance). Future members of Russia’s Way tended to vote their policy outlook rather than their electoral outlook, and to support high thresholds, an effect that appears to reflect a faction line. (Again, because the group was tiny, the effect is not statistically significant despite the sizable coefficient.) The LDPR and DPR, of course, also strongly support the high threshold— reflecting the fact that both were overwhelmingly made up of party-list deputies. Thus, even after controlling for the influence of electoral mandate type and policy position, we find some weak evidence for factionbased effects. In sum, in a manner consistent with our expectations, mode of election and policy had strong effects whereas factional membership had very modest independent effects on the threshold votes. In the immediate aftermath of the threshold decision, only one of the three independent groups met the adopted threshold, the New Regional Policy group, which soon attracted more than sixty members. Other proto-factions attempted to persuade the party factions with which they were most closely aligned to “lend” some members for the purpose of meeting the threshold, but

44

CHAPTER TWO

both groups were initially unsuccessful. By May 1994, after benefiting from some defections from other factions and the support of more independent deputies, the LDU of December 12 met the threshold and was officially registered. Critical to the registration of the LDU of December 12 was the defection of six deputies from the LDPR. The defectors, however, soon left, citing irreconcilable ideological differences with December 12, to form a group called Derzhava. December 12’s registration remained in force even after the six left the group. But only Zhirinovsky’s faction demanded the retraction of the registration. The Organization Committee, headed by Vladimir Bauer of the Russia’s Choice faction, rescinded the registration of the group on the grounds that at least two of the members’ signatures had been falsified; the December 12 group then appealed to the Council of the Duma. Members of the December 12 group charged that the Russia’s Choice faction was joining in the general denial of rights to members from single-member districts in favor of party-list deputies. The Council eventually allowed the December 12 group to retain its faction status. In March 1994 the deputies returned to the threshold issue as part of the floor debate over the adoption of a permanent set of standing rules. Proponents of a lower-threshold level of membership for groups again moved to decrease the thirty-five-member threshold. By wide majorities, the deputies rejected all alternatives and voted to retain the thirty-fivemember limit. The sequence of votes gives us another opportunity to examine voting alignments in the light of the deputies’ two months of experience with the system of factions and groups that had formed. To the two alternative threshold standards that were considered we add a third item, a proposal that would have allowed the Council of the Duma discretion in choosing which groups to register rather than setting a fixed membership minimum. Table 2.4 indicates the frequency of support for the three alternatives by faction. Clearly factions were less cohesive than in January, despite the fact that factions had more opportunity to organize and develop strategies. A quarter of the Communists were willing to support a lower threshold level, as were 30 percent of the LDPR members and one-third of the Agrarians. A substantial number of deputies on the left were willing to relax the thirty-five-member standard. Still, the left-right cleavage remains apparent. Large majorities of Russia’s Choice and Yabloko favored lowering the threshold, whereas large majorities of the Communists, Agrarians, and Liberal Democrats did not. The reform factions were generally more willing to give the Council of the Duma some discretion to decide whether to register groups.

PA RL I A ME N TARY PART Y SY S T EM

45

TABLE 2.4 Faction Voting on Selected Faction Threshold Proposals, March 1994 (percent in favor) Faction CPRF APR RW LDPR WR DPR NRP PRES LDUD12 Yabloko RC

Threshold of 30 Threshold of 12 25.7 34.3 50.0 30.0 25.7 0 7.7 30.0 94.4 66.7 78.3

0 22.2 87.5 5.1 0 0 25.7 25.0 100.0 75.0 74.2

Flexibility for Council of the Duma 20.6 61.8 16.7 18.2 9.1 0 54.1 80.0 94.7 66.7 87.8

Note: Based on the April 1994 faction membership; factions are listed in order of mean policy position, from anti- to pro-reform. See appendix. Source: Roll-call voting in the Duma.

Electoral considerations were also evident again. The groups that were made up of single-member district deputies (NRP, December 12, and Russia’s Way) favored lowering the threshold and giving the Council of the Duma some discretion to register groups. But note that members of Russia’s Way were now voting differently than in January. At that time they tended to vote with their friends on the left for a high threshold, but in March they voted for a low threshold. We are using the same measure of faction membership for both time periods, so it was the members’ positions that changed. Surely neither their electoral nor policy views had shifted. Their changed preferences over Duma institutions and, by extension, the changes in the voting patterns of other factions, presumably reflected the change in the relationship of the threshold issue to other institutional features of the Duma and the emergence of new issues onto the agenda. The pattern of voting exhibited in Table 2.4 leaves room for electoral, policy, and party effects. To see if the interfaction differences were significant, independent of electoral and policy considerations, we report multivariate estimates in Table 2.5. The dependent variables are the three votes listed in Table 2.4. In all models, party effects are captured by dummy variables, with the CPRF as the excluded category. The analysis excludes deputies who were not members of a faction in April 1994, just after these votes occurred. In the multivariate model, policy position loses some of its strength as a determinant of voting on the threshold issue compared to its strength

46

CHAPTER TWO

TABLE 2.5 The Effects of Electoral, Policy, and Partisan Influences on Position on Faction Thresholds March 1994 (logit estimates) Independent Variables Faction APR RW LDPR WR DPR NRP PRES LDUD12 Yabloko RC Mode of election Policy position Constituency Constant N −2 log likelihood Chi-square

Threshold of 30 −.48 .83 −1.83* −1.73 −6.29 −.99 −.97 2.27* 6.90 .36 .45 −.67* .13 −.43 218 196.8* 78.7*

Threshold of 12 .29 11.96 −1.11 −8.29 −8.64 −1.30 −10.79 9.37 9.77 −.34 .78 −1.57* .53 −1.14* 240 128.5 138.8*

Flexibility for Council of the Duma 2.03* −-.65 -.81 −1.70 −6.95 .07 5.28 1.50 -1.32 −.29 .11 −1.49* −.21 .17 218 215.7 116.4*

p < .05 Source: Roll-call voting in the Duma.

*

in January, but it remains statistically detectable on all three items. Election mode loses its significance entirely. We see stronger effects for faction than we did in January, however, even after electoral and policy factors are held constant. Variance for most factions is greater than that for the Communists, weakening the possible impact of faction membership. A negative sign for the policy variable indicates that members of a given faction opposed lowering the threshold from the current level of thirty-five; thus, as in January, the opposition factions are marked by negative coefficients. So, too, are those factions (such as the DPR and Women of Russia) that benefited from the higher threshold. And so is the one group (the NRP) that had succeeded in attracting the requisite number of members to have their group registered. Once in the club of factions, they had little incentive to give other groups a lower threshold for faction status. But the coefficient does not approach statistical significance once we control for electoral and policy positions, whereas several of the other faction variables (December 12, Yabloko, Agrarians) approach statistical significance. Thus the evidence for a strong faction/

PA RL I A ME N TARY PART Y SY S T EM

47

weak faction distinction is mixed—size, electoral composition, and policy interests also appear to affect the positions of faction members on this issue. It is notable that once policy and partisan variables are considered, the mode of election shows no effect on the threshold votes. This is in sharp contrast to the January voting, where electoral mode was a strong independent predictor of the likelihood of a vote on the threshold issue. This is strong presumptive evidence that timing matters—by March, deputies had come to calculate their preferences over institutions in more complicated ways that reflected the relationship between factions and the Council of the Duma. We also estimated equations for the three votes using interaction terms between faction, mode of election, and policy position, but found no significant effects within factions (data not shown). Factional differences reflected something other than a mechanical interaction between the general policy outlook of their members and the faction’s electoral composition. Probably influences such as faction size (a small, party-list dependent faction and a large SMD-based faction might both prefer a high-threshold level, all else being equal), satisfaction with the decisions made by the Council of the Duma, and the dimensionality of the policy space all weighed into the calculations of faction members about the threshold issue. To test for these effects, however, would entirely exhaust the already overextended degrees of freedom in the model; multicollinearity between faction, policy, and electoral categories already makes it impossible to separate faction effects from the interaction of policy and electoral effects. Therefore we can only speculate about the impact of other factors. Nonetheless, we note empirically that many deputies changed their positions between January and March, suggesting that decisions made in the intervening period about other institutional questions affected their preferences over faction strength. The Imperative Mandate Rule The defection of two LDPR deputies in February 1994 stimulated a debate about the imperative mandate. The imperative mandate concerned the obligations of party-list deputies to their faction (and party) leaders. Proponents of the imperative mandate argued that party-list seats in the Duma were the property of their parties. If a deputy resigned or was expelled from a faction, the deputy lost his or her right to Duma membership. In this view, party-list deputies would undermine the purpose of party-list elections if they failed to vote with their parliamentary faction. Zhirinovsky further argued that a faction had the right to expel a deputy and thus to remove his or her electoral mandate. Other deputies argued

48

CHAPTER TWO

that the stabilization of a party system required that deputies be discouraged from continuously forming and reforming factions. Emotions ran high in the debate (Haspel 1998, 62–64). Zhirinovsky went so far as to charge that the defecting deputies were guilty of criminal acts. Deputies recognized that in the factionalized Duma the outcome on this issue would affect the bargaining power of faction leaders and influence policy outcomes from time to time. The issue arose as an amendment both to the standing rules and to the Law on the Status of the Deputy. Under the amendment, party factions could remove from office any party-list deputy who violated party discipline. The expelled deputy’s seat would be filled by the next candidate on the list. The imperative mandate was endorsed initially by leaders of Russia’s Choice, the Communists, Agrarians, and DPR, in addition to Zhirinovsky, but was eventually rejected by a close margin. Many deputies, in responding to our questions in private interviews, indicated that they favored the rule initially but were persuaded by Zhirinovsky’s behavior at the founding congress of his party—when he was elected “leader for life” and the party agreed not to hold another congress for ten years—that the imperative mandate would give Zhirinovsky a disproportionate advantage in building a personal power base. Moreover, the imperative mandate seemed to advantage the group of large factions on the left of the political spectrum, including the LDPR, CPRF, and Agrarians. In the end, most deputies in Russia’s Choice and other democratic factions voted against the proposal, and so it was defeated. Unfortunately for our analysis, the issue was not resolved on a clean up-or-down vote but rather was decided as the Duma rejected package proposals fashioned by Zhirinovsky to attract the needed majority. These proposals combined the imperative mandate with provisions to register small factions and to allow the Council of the Duma to limit floor amendments. The faction vote with the largest number of voting deputies is provided in Table 2.6. Plainly the vote had a strong left-right cast, at least as structured by faction results. In a multivariate model (Table 2.7), the critical factors appear to be policy position and mode of election, with faction having little independent influence. Deputies on the left favored the imperative mandate; deputies on the right opposed it. Moreover, SMD deputies were substantially more opposed to the imperative mandate, which would strengthen the influence of faction leaders, than party-list deputies. Among the factions, the LDPR deputies showed more support for the imperative mandate than their policy and electoral interests would dictate, although the coefficient is not statistically significant. The vast majority of the future members of the December 12 group opposed the imperative mandate, although more favored the packaged motion than would be expected

PA RL I A ME N TARY PART Y SY S T EM

49

TABLE 2.6 Faction Voting on the Imperative Mandate, March 1994, by Faction (percent in favor) Faction CPRF APR RW LDPR WR DPR NRP PRES LDUD12 Yabloko RC

93.5 67.6 75.0 100.0 0 75.0 17.1 0 28.6 12.5 0

Note: Based on the April 1994 faction membership; factions are listed in order of mean policy position, from anti- to pro-reform. See appendix. Source: Roll-call voting in the Duma.

based on their policy and electoral circumstances, perhaps because of the lower-threshold provision included.12 It bears notice that factions have adopted a variety of stances on voting discipline. Some use a “solidarity vote” procedure, with either two-thirds or a simple majority vote within the faction needed to impose the expectation of a party-line vote by faction members. Some participants claim that the LDPR is distinctive in its application of the “fuehrer principle”— that Zhirinovsky can order compliant voting behavior from LDPR deputies on his own authority. Most factions seem to tolerate defections on matters subject to a solidarity vote with the stipulation that members carefully explain their position and that they not directly and actively oppose the faction. Nonvoting is often a convenient option for deputies in such cases. Members of nearly all factions claim that solidarity votes pertain only to “matters of principle.” Some factions appear to have adopted a solidarity-vote rule as a defensive response to similar rules adopted by other factions. We take up the issue of intrafaction cohesiveness again in chapter 6.

Conclusion A mix of political interests influenced deputies’ preferences about the faction threshold and the proposals for imperative mandate rules. The imperative mandate issue, which struck closer than the threshold issue to

50

CHAPTER TWO

TABLE 2.7 The Effects of Electoral, Policy, and Partisan Influences on the Imperative Mandate Vote, February 1994 (logit estimates) Independent Variables Faction APR RW LDPR WR DPR NRP PRES LDUD12 Yabloko RC Mode of election Policy position Constituency Constant N −2 log likelihood Chi-square

−1.12 .04 8.78 −11.72 −.66 −.40 −7.84 2.35* .19 −6.86 −1.25* 2.34* −.07 .04 233 129.3 193.6*

p < .05 Source: Roll-call voting in the Duma.

*

deputies’ electoral interests, generated differences between party-list and SMD deputies, even when controlling for faction membership and policy positions. The threshold issue more directly implicated the deputies’ policy and electoral interests, which were reflected in the voting on the various threshold proposals. Deputies created a faction system that provided a modest barrier to the formation of new deputy groups and put SMD and party-list deputies on equal footing regarding their ability to move from faction to faction without fear of immediately losing their electoral mandate. Party leaders did not gain a rule that would have provided them with a strong source of leverage with rank-and-file, party-list deputies. Once deputies had gained experience with the initial factional system, factional membership replaced electoral type as a predictor of the deputies’ interest in preserving the system when they voted on it again in March. The findings are largely consistent with our expectations that the directness of the connection between an institutional choice and the deputies’ goal determines the importance of that goal to deputies’ strategies. The findings are also consistent with the argument that timing matters. As institutional choices were made and the new arrangements experienced, the institutional preferences of many deputies evolved, as did the observed relationships between deputies’ goals and their institutional preferences.

NOTES

1. More recently, scholars have begun to explore the intra- and interparty dynamics of American congressional parties using formal models in order to determine the conditions under which members would use parties to resolve collective dilemmas such as producing majority-supported policy alternatives, suppressing unwelcome options, and maintaining control of the executive. See, for instance, Aldrich 1995; Binder 1996, 1997; Cox and McCubbins 1993; Kiewiet and McCubbins 1991; and Rohde 1991. Krehbiel (1991, 1998, 1999) explicitly challenges accounts that explain congressional outcomes as a product of the activity of congressional parties. 2. But see Laver 1999. 3. Here and elsewhere, we use the term party to simplify an already complicated set of distinctions between types of political organizations. The electoral law provided that registered “electoral associations” (izbiratel’noe ob”edinenie) that met a certain threshold of signatures could both nominate candidates in singlemember districts and run candidates on party lists. Some of the electoral associations that ran candidates in the 1993 elections did constitute themselves as parties, whereas others, such as Yabloko, constituted themselves as “blocs,” and still others as “movements.” For simplicity, we shall refer to all electoral associations as parties. 4. The presidential team designed certain provisions on implementation of the new constitution in ways intended to benefit Yeltsin’s political supporters. For example, the constitutional draft contained transitional provisions that required new parliamentary elections in two years, rather than in four, as was to be the case thereafter. The effect of the provision was to limit the risks attendant to the election of a new parliament and to make deputies face the electorate earlier than the president, whose five-year term, under the old constitution, was set to expire in June 1996. The rapid succession of elections may also have the effect of speeding the formation of electoral parties. Despite the obvious efforts of some Yeltsin supporters to take advantage of their government positions in the 1993 elections, Yeltsin himself refused to involve himself in party-building activities. This appeared to contribute to the fragmentation of democratic forces into multiple parties, which may have cost them both PR and SMD seats. Indeed, the president’s public posture and that of many senior members of his administration suggested that the outcome of the parliamentary elections was largely unimportant. The passage of the constitution appeared to be Yeltsin’s top priority, and that was largely a question of turnout. The upper house, the Council of the Federation, is composed of two deputies from each of the eighty-nine constituent members of the Russian Federation (but only eighty-eight in practice because Chechnya has not sent members to the Council of the Federation). Under the presidential decree for the 1993 elections, the deputies were selected on the basis of a single ballot, with the top two vote getters winning the seats. After the first two-year term, scheduled to expire in December 1995, the constitution provides merely that the two deputies from each region be selected from the legislative

52

CHAPTER TWO

and executive branches of regional governments. It does not specify the manner of the election nor does it set a length of term for Council deputies. 5. These results must be viewed in the context of the parties that failed to gain seats in the Duma. Yeltsin’s election decree set very high thresholds to registration of parties. Parties and electoral associations had to have been registered by the Ministry of Justice to form lists. They also had to collect one hundred thousand signatures nominating petitions, of which no more than 15 percent could be obtained in any one province. And they had to gather the signatures and submit them to the Central Electoral Commission (CEC) by November 6. On the electoral law’s effect on the campaign, see Urban 1994, 136–39. Only twentyone parties submitted lists to the CEC by the deadline, and only thirteen of these were certified as valid. Five of the thirteen parties competing for the list vote failed to clear the 5 percent hurdle. On the other hand, less than 9 percent of the voters cast party-list votes for parties that failed to enter parliament. This is a far smaller proportion of “wasted votes” than is typically found in plurality systems and even in many proportional systems (McGregor 1993). Overall, however, these features of the electoral law had a very substantial effect: Of dozens, if not hundreds, of organizations that might have attempted to field candidates, only eight succeeded in entering parliament through proportional representation. 6. The literature on these subjects is vast. Consult Cox 1997; Downs 1957; Duverger 1954; Epstein 1980; Iversen 1994; Huber and Inglehart 1995; Mezey 1979; Lijphart 1984; Lijphart and Grofman 1984, 1992; Lipset and Rokkan 1967; Loewenberg and Patterson 1979; Nohlen 1984; Shugart and Carey 1992; Stepan and Skach 1993; Mainwaring 1993; Katz 1997; Rae 1967; Sartori 1976; Taagepera and Shugart 1989. 7. The German experience with a mixed PR/SMD electoral system suggests that the sharp partisanship imported to the parliament with the PR deputies would overshadow the local orientation of the SMD deputies. 8. For more details on the imperative mandate debate, see Haspel 1998, 62–64. 9. All three were comprised mainly of SMD deputies. New Regional Policy had a heavy representation of state enterprise directors. Russia’s Way had a strongly nationalist and Communist bent. The LDU of December 12 (named after election day) had a predominantly reformist outlook and was close to Russia’s Choice on substantive policy questions, but its members sought to establish a separate political identity. Most of its members were young, professional, and concerned with maintaining strong district ties. By mid-1994 only seven deputies chose not to affiliate with a faction. 10. Quoted in Haspel 1998, 50, from the Duma’s Stenogramma, January 11, 1994). 11. We also have estimated these equations with a survey-based measure of deputies preferences (see appendix). Because the survey measure is fully independent of the roll-call record, it would be preferable as a measure to the rollcall-based measure reported in Table 2.3. Unfortunately, because the survey includes only a fraction of the deputies who voted each time, it yields a very small N (about 80 of the 250 or so deputies who voted each time), so we do not report the results for that measure here. We are confident, however, that our measure

PA RL I A ME N TARY PART Y SY S T EM

53

of policy preference based on members’ voting behavior produces closely equivalent results. The simple correlation between the measure of policy preferences revealed through the survey and the scores calculated by our dimensional analysis of voting behavior is .82. 12. Interaction effects for within-faction policy and mode of election were not found to be significant.

THREE CREATING THE COUNCIL OF THE DUMA

T

HE COUNCIL of the Duma serves as the chamber’s executive committee. The Council has the power to assign legislative issues to particular committees, formulate and propose the daily and longerterm agendas to the chamber, resolve disputes arising from the interpretation of the standing orders, and negotiate compromise agreements on controversial legislative issues. The Council is interesting for two reasons: First, it replaced the Presidium, the executive committee of the old Supreme Soviet, which was made up of central leaders and committee chairs and was the means for central control of the Supreme Soviet by the Communist Party. Second, the new Council is a nonmajoritarian institution, representing factions on a parity rather than proportional basis. Thus, in contrast to a body such as the Bureau of the French National Assembly, which in a similarly multiparty, mixed parliamentary-presidential setting nonetheless operates under a clearly majoritarian rule, the Russian parliament’s steering body is composed in such a way as to allow small party factions and those in the minority a share of power in governing the chamber equal to larger factions and those that might be in the majority coalition. In this chapter we examine the attitudes of Duma candidates and deputies regarding the Presidium and the Council of the Duma. Our findings corroborate our view that a single political goal is unlikely to account for the major features of the Russian Duma. Our findings also show important short-term variation in attitudes about institutional arrangements that appear to be associated with participants’ institutional choices and subsequent experience.

The Presidium and the Council of the Duma The Council of the Duma was seen by deputies as a sharp departure from previous practice. Russia has had two principal legislative systems since the wave of democratizing reforms launched in the late 1980s. In the 1990–93 “transitional” period, a two-tiered adaptation of the traditional communist model of the Supreme Soviet system was in place. These institutions featured a slightly modified Presidium as their principal steering organ or executive committee. The 1990–93 parliament was abruptly suppressed by President Yeltsin in September 1993, with decrees dissolv-

COUNCIL OF THE DUMA

55

ing parliament, annulling the deputies’ mandates, and requiring new elections to a new parliament in December 1993. Moreover, by decree he enacted a new electoral system for these elections. Finally, he mandated a referendum on a new constitution, drafted under his direction, to be held simultaneously with the parliamentary elections. The alternatives with which politicians in 1993 and 1994 were most familiar were those that had existed in the USSR and in legislative assemblies at the level of the Russian Republic. These options had been the subject of intense discussion for at least three years, as working groups of deputies and experts developed proposals for a new constitution, new electoral laws, and new models of legislative organization. Steering arrangements were among the central points of debate: Should the new parliament preserve the Soviet-era institution of a Presidium, or should it replace it with another structure? How much power should the chamber’s chairman have? How much power should be granted to political parties or factions through the steering body? In the communist era, the Presidium was charged with preparing the agenda of the Supreme Soviet for its periodic if brief sessions. Moreover, under Soviet law since the Stalin period, the Presidium was a full-time working organ empowered to enact any decisions falling under the jurisdiction of its parent Supreme Soviet in those periods when the Supreme Soviet was not in session (Minagawa 1985). The communist-era Supreme Soviet, of course, was hardly ever in session, so the Presidium, from the 1930s to the 1980s, had become a small parliament in itself. It exercised its power through a large and cohesive staff that in turn was closely tied to and firmly controlled by the staff of the Communist Party Central Committee. Furthermore, the Soviet state—under the Stalinist constitutional framework, which survived with minor modifications until the late 1980s—assigned the Presidium the status of a collective head of state. Its chairman, whose title was Chairman of the Supreme Soviet, was the equivalent of a state president. The Presidium model survived into the reform era. Between 1989 and 1991 the Soviet state under Gorbachev operated under a modified version of the Stalin model, and from 1990 to 1993 the Russian Republic, which constitutionally was a constituent member of the Soviet federation until that federal union dissolved in 1991, operated under an almost identical version of the same modified system. This system retained the Presidium as a steering body for both the Congress of People’s Deputies, which was elected in direct national territorial district races, and the Supreme Soviet, which was the working parliament elected by deputies of the Congress from among its own membership. The Presidium was composed of committee chairs and the chairs and deputy chairs of the chambers. As in the past, it was presided over by the

56

CHAPTER THREE

Chairman of the Supreme Soviet, who was elected to this position from among the membership of the Congress. Together with several deputy chairs, he exercised powers inherited from the older system: chairing the meetings of the Presidium, managing the central staff, overseeing the budget and material resources of the parliament, and chairing plenary sessions of the legislature. The chairman also had a privileged role in shaping the Presidium’s collective decisions regarding the parliament’s agenda, the assignment of draft legislation to committees, the nomination of deputies to committee leadership posts, and the management of legislative processes. Thus both the 1989–91 USSR-level legislative system and its Russian Republic equivalent, which existed between 1990 and 1993, were highly centralized in the way these powers were exercised. However, the Yeltsin-appointed commission that had been charged with drawing up an organizational scheme for the new Duma in the autumn of 1993 discarded the old Presidium model in favor of a council made up of the heads of the organized political factions (see Remington, forthcoming). In turn, this proposal was accepted by the newly elected deputies when they convened in January 1994. The new Council of the Duma differs sharply from the old Soviet Presidium. The Council features equal voting rights for the leader of each parliamentary faction. Committee chairs take part in Council meetings but lack voting rights. To a certain extent, the power-sharing Council is offset by majoritarianism in other areas: the majority-elected chairman, majoritarian rules in floor voting, the distribution of committee chairmanships in proportion to the parties’ shares of members, and the open rule on agenda and amendment rights on the floor.1 But the powers of the Council of the Duma are sufficiently great as to raise the question of why the Duma has adopted more egalitarian and consensual arrangements for its steering body. The puzzle is heightened because the Duma chose to preserve this institutional structure even after the 1995 parliamentary elections brought the Communists and their allies into a position to elect a Duma chairman of their choice and command almost a majority of votes. Although it was, by some accounts, in their power to force an alteration in steering arrangements that would have resulted in a more advantageous majoritarian framework, they refrained from changing the rules.

Alternative Perspectives on the Council of the Russian Duma Our purpose in this chapter is to determine whether policy, electoral, and partisan interests structured attitudes about the Presidium and the Council of the Duma. We consider the attitudes of Duma candidates and

COUNCIL OF THE DUMA

57

deputies at two crucial points in the Duma’s short history, the organization of the first modern Duma after the 1993 elections and that of the second Duma after the 1995 elections. The three sets of interests, introduced in chapter 1, imply somewhat different patterns of attitudes about the Presidium and the Council. Agenda Setting and Policy Goals Consider two extreme distributions of policy preferences. At one extreme is a highly polarized two-party system with one faction enjoying majority status. A perfectly polarized distribution entails a perfectly cohesive majority faction or coalition, whose members willingly join to structure their institution in a way that allows them to enact policies at minimum cost. Such an institution would be a majoritarian one that restricted the rights of the minority to influence the agenda and established a strong central leader who, on behalf of the majority, directed all activities of the parliament. At the other extreme, a multidimensional policy space, defined by several cross-cutting issues, leaves individuals’ preferences aligned differently from dimension to dimension. No single majority exists; coalitions change from issue to issue. The members of no single coalition or faction share similar policy preferences on enough issues to motivate them to manipulate institutional arrangements affecting all issues. In the absence of a cohesive, policy-based majority, parliamentarians are unwilling to create a strongly centralized or majoritarian agenda-setting process. Instead, they prefer equality in the formal rights of all members or groups to propose agenda items and have a voice in policy making. No cohesive majority was elected in 1993. In the context of the 1994 Duma, right-wing deputies and factions were slightly outnumbered by left-wing deputies and factions. For that reason, and perhaps because they had the experience under the Presidium fresh in mind, right-wing deputies are expected to be more supportive of replacing the Presidium model with a less centralized, less majoritarian agenda-setting process. These arguments lead to two propositions: Proposition 3.1. Individuals’ attitudes about the Presidium and Council of the Duma are correlated with their policy preferences, with deputies on the left more supportive than those on the right of a majoritarian agenda-setting process. Proposition 3.2. A majority party or coalition produces a majoritarian agenda-setting mechanism; the absence of a majority party or coalition produces a nonmajoritarian agenda-setting mechanism.

58

CHAPTER THREE

Agenda Setting and Electoral Goals Because electoral laws specify the rules under which elections are conducted and won, they shape the electoral strategies of politicians, parties, and factions. Those rules, it is well understood, influence the kind of parties that compete and often further detail the way parties’ candidates are chosen and placed on the ballot. But beyond election rules and parties, parliamentary institutions may enhance the resources and influence the legislative record of parliamentarians that may be useful for electoral purposes. The agenda-setting features of the Duma may have electoral consequences for deputies. For example, certain kinds of legislation may be advantaged or disadvantaged. Issues subject to debate may bare deep divisions or facilitate logrolling. Even so, these consequences are likely to have only indirect influences on the electoral prospects of individual deputies. Thus we begin with the expectation that agenda-setting mechanisms are likely to be evaluated more for their policy or partisan implications than their electoral implications. With that important expectation, we might nevertheless expect differences among deputies in their preferences about agenda-setting institutions based on their mode of election. In chapter 2 we noted that the electoral fate of PL deputies is dependent on the support of central party leaders and on their parties’ national reputations, whereas the electoral fate of SMD deputies has a weaker relationship to party and is likely to be more dependent on the deputies’ personal reputations in local districts. All things being equal, PL deputies should be more tolerant of a system that gives more power to central party leaders, who are needed to maintain and enhance party reputations and who have the leverage with PL deputies to gain compliant behavior. PL deputies see reelection at least partially as a zero-sum game among the parties. SMD deputies prefer the freedom to pursue individual agendas unrestricted by strong parties and leaders. They are more likely to see reelection as a positivesum game for deputies, one that can be aided by distributive legislation and logrolling. These arguments suggest two propositions: Proposition 3.3. Mode of election is less important than policy preferences in influencing attitudes about agenda-setting institutions. Proposition 3.4. PL deputies prefer a party-based system of agenda-setting; SMD deputies prefer a system based on nonpartisan principles, such as representation by committee.

COUNCIL OF THE DUMA

59

Agenda Setting and Partisan Goals Control of the parliament’s agenda can be instrumental in pursuing factions’ goals. By determining the legislation to be considered, the order of consideration, and the eligible alternatives, leaders seek to (1) promote legislation that best serves collective faction goals, implementing tradeoffs among the faction goals, if necessary, and (2) minimize the costs for individual deputies whose policy or electoral interests are not entirely compatible with faction strategies. Both elements are important. First, a faction’s collective policy and electoral goals may not be fully compatible. For example, a party’s policy commitments may not be consistent with maximizing its success in the next election. Agenda control may allow the leader to reduce the severity of the trade-offs that must be made, perhaps by suppressing issues or policy options that force a choice among the competing considerations. Second, because rank-and-file members’ goals depend on faction success, rank-and-file members are willing to make some personal sacrifice (policy, electoral) to realize gains from collective action by their factions. This may mean allowing faction leaders to pursue a legislative program that is not entirely compatible with their policy or electoral interests. But there is a limit to how much rank-and-file members are willing to sacrifice. A leader who imposes costs that are considered too high may be deposed. A leader, therefore, seeks to structure the agenda so that the pursuit of collective faction goals is reasonably compatible with individual goals. For example, members whose electoral prospects would be harmed if they were compelled to vote on a particular amendment might be protected by an agenda that prevents the amendment from being proposed. Moreover, a faction is likely to have an interest in reducing the advantage that opposing parties gain by controlling the agenda. Even when a faction’s members have diverse policy and electoral interests that make agenda manipulation difficult, the party still prefers not to grant an advantage to another faction or coalition. To be able to block the agendarelated strategies of other parties may be just as important to the policy and electoral interests of a faction’s membership as affirmative agenda manipulation by that party’s leaders. Rank-and-file deputies of the same faction will share preferences about agenda-setting institutions more than is suggested by their individual policy or electoral interests. To be sure, diversity in individual interests will limit leaders’ ability to manipulate the agenda on behalf of any one interest. But this is only the cost side of the equation. The benefits of a good faction or party reputation and majority status may be so substantial that

60

CHAPTER THREE

some modest, short-term sacrifice of policy or electoral interests will be worthwhile. Thus we can state a final proposition: Proposition 3.5. Deputies of the same faction will share preferences about agenda-setting institutions. Of course factions may or may not differ in their preferences about agenda-setting processes. The relative strength of the factions will determine whether a party prefers a majoritarian process or some other process for setting the parliamentary agenda. A majority party, or majority coalition of parties, prefers majoritarian processes that allow it to structure the agenda. A minority party prefers rules that limit manipulation of the agenda by majority rule. In the absence of a stable majority faction or coalition, perhaps because of the multidimensionality of the policy space or the great spread of multiple parties across a vast policy space, parties, in order to protect themselves, will prefer a process of mutual veto over the agenda. This is, of course, the conclusion reached in Proposition 3.2.

Duma Action on the Council of the Duma We administered four surveys to candidates running for the Duma in 1993 and 1995 and members of the elected Duma in 1994 and 1996 (see appendix). The 1993 survey of parliamentary candidates was conducted before the new Duma was organized and within months of the forcible dissolution of the transitional parliament of 1990–93. These responses could not have been influenced by knowledge of the institutional arrangements chosen in the coming months. The 1994 survey of deputies measures attitudes of the players that were present when the Council of the Duma arrangement was adopted in the first Duma. The 1995 candidate survey and the 1996 deputy survey allow us to measure changes in the alignment of policy and institutional preferences just before and after the formation of the second Duma. To measure respondents’ general attitudes about the old Presidium arrangement, we posed two questions about the Presidium model to our sample of parliamentary candidates: Did the respondent consider that the Presidium’s influence had been excessive, as much as proper, or insufficient, and did the respondent think that the new Federal Assembly needed a Presidium? We measure policy positions by scaling responses to questions about policy issues. The data were gathered at critical junctures in the early development of the Russian parliaments so that we can capture change

COUNCIL OF THE DUMA

61

TABLE 3.1 Multivariate Logit Estimates of the Effect of Policy Position and Mode of Election on Attitudes about Institutional Arrangements among Russian Candidates, 1993 and 1995 Dependent Variables 1993 Candidates Survey Rated the influence of the old Supreme Soviet Presidium as “excessive” Considered a presidium to be necessary for the new Duma Agreed that a majority party or bloc be allowed to name the chairs of committees Agreed that a majority party or bloc should be allowed to appoint the staff of the Duma Agreed that a majority party or bloc be allowed to set the agenda of the Duma 1995 Candidates Survey Agreed that parties should have an important role in organizing the parliament Agreed that a majority party or bloc be allowed to name the chairs of committees Agreed that a majority party or bloc should be allowed to appoint the staff of the Duma Agreed that a majority party or bloc be allowed to set the agenda of the Duma

Policy Position

Mode of Election

−1.33*

.24

−.56*

−.72*

.16

.02

.05

−.40†

−.06

−.09

.12

1.32*

−.18

.71*

−.20

.81*

−.20

.86*

p < .05; †: p < .10 Source: 1993 and 1995 candidate surveys.

*

in attitudes about institutional arrangements over the short history of the new Duma. We limit our report to multivariate estimates because of the strong relationships among policy positions, mode of election, and faction membership. Tables 3.1, 3.2, and 3.3 show logistic regression estimates for variables using the respondents’ policy dimension scores, mode of election, and party as covariates. For faction, the analysis is restricted to the two surveys of deputies because of difficulties associated with identifying the party and faction of candidates.2 Table 3.1 reports the estimates for the effects of policy scores and mode of election on preferences about institutions for our surveys of candidates in 1993 and 1995. Because many candidates ran without any party endorsement or with multiple party endorsements, we do not report estimates for party or faction effects for the candidates. Tables 3.2 and 3.3 report the estimates for the effects of policy scores, mode of

62

CHAPTER THREE

election, and faction on preferences about institutions for our surveys of deputies in the first new Duma (1994–95) and deputies in the second Duma (1996). We organize our discussion by the three sets of covariates and then summarize their implications for the propositions. Policy Position During the 1993 campaign, which immediately followed the demise of the transitional parliament, responses to questions about the Presidium divided sharply by policy position, consistent with Proposition 3.1. In 1993 parliamentary candidates of the reformist camp were far more likely than candidates from the opposite camp to disapprove of the way the old Presidium was used and to oppose adopting the Presidium model in the new parliament (Table 3.1). On questions about other details, and obviously less salient issues during the fall of 1993, little difference existed between PL and SMD candidates. Preferences on some of these other issues might depend on the eventual balance of forces in the Duma, something candidates could not know before the election results were published. So, for example, candidates held quite mixed views about whether a faction or coalition receiving a majority of seats in the new Duma should be able to name the chairman of the chamber, control the hiring of staff, name the chairs of committees, or set the content and order of the legislative proceedings. Elected deputies, not candidates, chose institutional arrangements, of course. By the time that we could interview deputies in the last month of 1994, the results of the 1993 elections were digested and many of the choices about institutional arrangements had been made, so the deputies’ responses reflected nearly a year’s experience with the adopted institutional arrangements. We asked respondents to comment on the power of the Council of the Duma and the influence of the Chairman of the Duma, and whether chairs of committees ought to have the right to vote in the Council of the Duma (Table 3.2). Deputies’ views of the power of the Council and its chairman, as well as of voting rights on the Council, were at best weakly related to their policy views in late 1994. Deputies expressed general satisfaction with the influence of the Council and its chairman. A plurality of 47 percent considered the Council’s power to be “about right,” and more than three-fourths of the respondents considered the chairman’s power to be about right. Deputies favoring a stronger or weaker Council or speaker were not disproportionately to the left or right on the policy spectrum (data not shown). Voting rights for committee chairs on the Council of the Duma was another contentious issue that lost its divisiveness. The system of equal faction representation on the Council, although denying

−1.48* (.63) −.33 (.88) −.23 (.58) −.87 (.67) −1.82 (1.21) −1.05 (.86) −1.18 (.85) 11.7 −.03

−.02 (.28) .27 (.38)

2

.26 .00

−.61 (.40) .42 (.50)

2 −.29 (.18) 1.72* (.44)

−.30* (.15) −.96* (.31)

1

−1.52* (.63) −3.00* (1.15) 1.34* (.67) −3.98* (1.19) −1.78† (1.05) −.04 (.96) .13 (.89) 76.0* .46

−.46 (.35) .12 (.38)

2

Faction Leaders

.87 (.93) 4.26* (1.23) 1.22 (1.00) 2.04* (.93) 2.33† (1.34) 1.85 (1.56) 1.66 (1.16) 46.9* 12.6* .21 .11

.29 (.33) 1.37* (.52)

2

Factions 1

−9.38 (28.3) −1.53 (1.19) .65 (.68) −1.69† (.96) −9.63 (57.3) −1.60 (1.15) −1.76 (1.16) 23.7* 24.3* −.03 .00

Speaker −.10 (.20) −.02 (.41)

1

17.0* .08

.44* (.16) 1.09* (.33)

1

.09 (.57) .04 (.87) −.46 (.61) −.15 (.65) .58 (.91) −.95 (.86) −1.16 (.89) 23.1* .09

.22 (.28) .97* (.38)

2

Committee Chairs on Council

* p < .05; †: p < .10 Note: The CPRF is the excluded category for the dummy variables for factions; constants not reported. Source: 1994 deputies survey.

Chi square P.R.E.

Yabloko

RC

PRES

NRP

LDPR

LDUD12

Faction APR

1.8 .00

.20 (.16) .20 (.32)

Policy position

Mode of election

1

The Council of the Duma

Independent Variables

Rated as Excessive the Influence of:

Dependent Variables

TABLE 3.2 Multivariate Logit Estimates of the Effect of Policy Position, Mode of Election, and Faction on Attitudes about Institutional Arrangements among Russian Deputies, 1994–1995 (standard errors in parentheses)

64

CHAPTER THREE

committee chairs a vote, undercuts majority rule. Voting rights for committee chairs would enhance the influence of large factions because chairmanships were distributed in proportion to the factions’ shares of seats. We asked deputies whether they believed that committee chairs should have a full vote in the Council. Responses were related to policy position (Table 3.2), although faction differences appear to be more important. These findings for the deputies of the 1994–95 Duma may seem puzzling in light of how strongly policy views structured attitudes about the Council’s predecessor institution, the Presidium, among candidates for the Duma in 1993. The absence of a majority faction or bloc in the aftermath of the 1993 parliamentary elections led to a scheme of equal voting rights for all factions. This power-sharing arrangement appears to have satisfied deputies who once differed widely in attitudes about previous institutions and still differed widely in policy views. The salience of policy preference-based considerations for views on parliamentary governance abated once a nonmajoritarian institutional framework, reflecting the absence of a majority for any one faction or bloc, had been put into place. General satisfaction with the Council may help us to explain its continuation after the 1995 elections, but the election outcome might lead us to predict some other arrangement. In fact, the widely held view that forces in opposition to Yeltsin gained control of the Duma might lead us to expect that a strongly majoritarian process would have been imposed by the new majority. Indeed, the new chairman of the chamber was viewed as more supportive of the left in his policy views and exercise of his formal powers, which seems to have revived policy-based differences in evaluations of the power of the chamber’s chairman and the power of the Council and factions. However, as we explained in chapter 2, even the 1995 elections did not generate a stable majority coalition in the Duma. Communists had a large plurality of seats (33 percent). However, even with the support of the Agrarian and People’s Power factions, the left had a bloc that fell just short of a majority. (People’s Power was a left-leaning deputies’ group composed mainly of single-member-district deputies.) When combined with Zhirinovsky’s LDPR or sometimes Yavlinksy’s Yabloko, a sizable majority opposed to Yeltsin existed in the post-1995 Duma, but this was not a cohesive bloc that agreed on a positive legislative program. The opposition majority spread across a vast space in the political spectrum and did not constitute a majority bloc with sufficient policy agreement to serve as a basis for organizing the Council of the Duma on majoritarian principles. Policy positions were not related to responses about institutional arrangements that we gathered from candidates in the 1995 elections. However, in 1996, after the Duma’s more strongly oppositional member-

COUNCIL OF THE DUMA

65

ship selected a Communist, Gennadii Seleznev, as speaker, substantial dissatisfaction with the speaker was expressed by reform-oriented deputies (Table 3.3). This relationship holds up even after faction membership is controlled. The shift toward a somewhat more majoritarian leadership system produced the predicted effect of sharpening policy-based differences in attitudes toward agenda-setting institutions. While preferences about changing the old Presidium structure were strongly related to policy preferences as the new Duma organized in late 1993, the Council of the Duma proved satisfactory to most deputies and survived the transition to a second Duma after the 1995 elections. The absence of a coherent majority bloc throughout the period is consistent with the power-sharing arrangement reflected in the composition and voting rules of the Council. Mode of Election For the candidates in the 1993 campaign and for the elected deputies of that Duma of 1994–95, attitudes about key features of the Council of the Duma reflect their different electoral needs, as Proposition 3.4 suggests.3 In 1993 SMD candidates, many of whom ran without affiliation to an electoral party, were much less eager to have a Presidium reestablished in the new Duma (Table 3.1). By late 1994 SMD deputies were as accepting as PL deputies of the influence of the Council and the Duma chairman, but they were far more critical of the role of parties and their leaders, much more opposed to rule by majority parties or blocs, and more supportive of giving committee chairs a formal voice on the Council (Table 3.2). And these differences between PL and SMD candidates and deputies continued into the next 1995 campaign and 1996 Duma (although party appears to account for many of the differences). In short, before the new governing arrangements were adopted, reformist and Communist candidates differed sharply in their preferences over steering institutions. Once they had been chosen—and neither camp was strongly advantaged over the other—electoral differences became more important in structuring attitudes. The PL-bias of the system needs explanation but must first be understood in greater detail. As we saw in chapter 2, to gain faction status in the Duma, a group must either have won representation in the Duma on the basis of party-list voting or register at least thirty-five deputies under its name. This threshold was set just low enough to win a majority’s consent and high enough to ensure that only a few organized factions could enjoy the benefits of official faction status. Equal representation in the Council of the Duma kept all factions satisfied while proportionalism in the distribution of committee chairmanships (see chapter 4) ensured that a stronger faction would not be greatly disadvantaged by the

7.1 .00

−.30 (.29) −.25 (.65)

−.55* (.22) .27 (.43) .46 (1.47) 1.49 (1.19) 2.23† (1.24) .80 (1.27) 2.70* (1.24) 1.64 (1.21) 18.3* .04

2

1

36.9* .10

−1.17* (.22) −.20 (.37)

1

2

−.38 (45.47) 9.48 (28.22) −.70 (47.85) 9.14 (28.22) 9.96 (28.22) 9.43 (28.22) 79.3* 25.9* .22 .18

−.55* (.20) 1.52* (.44)

1

.77 (.82) −.72 (1.24) .89 (.86) .54 (.88) 2.12* (.88) .30 (.91) 32.3* .13

−.33 (.27) .75 (.55)

2

Faction Leaders

9.03 (17.22) 7.62 (17.23) 8.44 (17.23) 8.70 (17.23) 10.33 (17.23) 7.93 (17.23) 61.7* 21.4* .27 .05

−.64* −.25 (.18) (.25) 1.26* .55 (.36) (.50)

1

Factions

Rated as Excessive the Influence of:

−.62* (.30) .46 (.53)

2

Speaker

* p < .05; †: p < .10 Note: The CPRF is the excluded category for the dummy variables for factions; constants not reported. Source: 1996 deputies survey.

Chi square P.R.E.

Yabloko

RReg

NDR

PPwr

LDPR

Faction APR

Mode of election

Policy position

Independent Variables

The Council of the Duma

Dependent Variables

1.0 .00

.06 (.16) .29 (.32)

1

1.99* (.66) −.31 (.62) .25 (.73) −.36 (.68) 1.40† (.75) −1.21 (.78) 26.4* .21

−.10 (.23) −.65 (.48)

2

Committee Chairs on Council

TABLE 3.3 Multivariate Logit Estimates of the Effect of Policy Position, Mode of Election, and Faction on Attitudes about Institutional Arrangements among Russian Deputies, 1996 (standard errors in parentheses)

COUNCIL OF THE DUMA

67

nonmajoritarian character of Council membership. Members of factions associated with strong parties, regardless of electoral category, would be predicted to support these arrangements. Members of factions associated with no party or weak parties (those failing to cross the 5 percent threshold) and members of nonparty factions (registered groups) would be less supportive because they face certain disadvantages under the system: Before the election their candidates cannot be sure that they will win a place in the governing body, and after the election their members face the danger that they will fail to retain thirty-five members and thus lose their registered status. Finally, deputies outside any faction or registered group are at the greatest disadvantage under this system and so might even be expected to oppose it: They do not have any chance of bidding on leadership posts; they lack a voice on the Council of the Duma; and they lack privileged floor rights and other perks of faction status. On the other hand, they still enjoy the public good that follows from the Duma’s nonmajoritarian structure, to wit, the party-provided hurdles to the rise of a despotic chairman. Three related factors may account for the PL-bias of the adopted system. First, PL deputies may not be the only ones interested in a factionoriented system. Some SMD deputies also ran on a party list and may want to run on a party list in the future, giving them incentives to be responsive to faction leaders’ demands and willing to join a chamber majority for a PL-biased system. Second, PL deputies organize and choose leaders before a new Duma convenes, giving them an edge over SMD deputies who do not meet until the Duma convenes. And third, the sequence of events in late 1993 and early 1994 advantaged PL deputies. Leaders whose parties were likely to exceed the threshold for representation in the Duma began devising parliamentary rules and organization in December 1993, even before the official election results were published, giving them an opportunity to present a well-developed plan for the organization of the new parliament before SMD deputies had a chance to consider it and perhaps organize to oppose it. Thus electoral motivations combined with the collective action problem and the sequence of events to produce institutional arrangements that appear to reflect the interests of PL deputies over SMD deputies. Mode of election continued to discriminate among candidates for the parliament in 1995 (Table 3.1), even when their policy positions were controlled. SMD candidates were less likely than PL candidates to be supportive of a strong faction role in parliament. And this relationship was again evident among elected deputies in 1996, although the strong relationship between faction and mode of election reduces the estimated effects of mode of election on attitudes about the role of factions (Table 3.3).

68

CHAPTER THREE

Party and Faction In 1993 candidates of the several parties differed widely in their central tendencies on questions about the Presidium and the role of factions in the Duma. The reform-oriented members of Russia’s Choice strongly disapproved of the Presidium model but were quite divided about majority faction or bloc control of key features of the Duma (data not shown). The nationalistic candidates of LDPR strongly approved of the Presidium model and generally favored a majoritarian Duma. Communists also approved of the Presidium model but opposed a majoritarian Duma. A plausible account of this pattern is that candidates of the parties were reacting to the experience with the Presidium and factions in the deposed parliament of the immediate past. Russia’s Choice candidates may have believed that the exclusion of reform-oriented deputies from positions of power within the Presidium system would be repeated if a similarly centralized system was adopted in the new Duma. LDPR candidates, whose new party was highly centralized under the leadership of Zhirinovsky, may have looked forward to the success of their party and leader, and desired a highly centralized, party-dominated Duma. Communists—whose previous party organization had been banned and whose leaders had been jailed—had reasons to be quite pessimistic about the parliamentary election outcomes and therefore may have been more skeptical than other parties about the implications of majoritarian arrangements for the new Duma. Whatever the explanation, the evidence from the candidates of 1993 is consistent with the proposition that preferences about policy and the Presidium were organized by the party blocs. For the sake of brevity, we forego a review of responses by faction in the 1994 deputy survey and 1995 candidate survey and turn to the 1996 survey of deputies (Table 3.3). After the 1995 elections, Our Home Is Russia (NDR, for its Russian initials) replaced Russia’s Choice as one of the large party factions in the Duma after Russia’s Choice failed to clear the 5 percent threshold. The strong approval of the role of the Council of the Duma, noted above, is again evident across the four large factions—Our Home Is Russia, Yabloko, LDPR, and CPRF. In contrast, the two groups made up of SMD members tended to rate the Council’s power as excessive. However, the power exercised by the new Communist chairman, Gennadii Seleznev, is not universally approved. Our Home and Yabloko deputies, along with those of other reform-oriented parties, frequently expressed dissatisfaction with the chairman’s power, while Communists never did. And on the question of voting rights for committee chairs on the Council of the Duma, Communists, who held a large

COUNCIL OF THE DUMA

69

share of the committee chairmanships, were considerably more supportive than were members of Yabloko or Our Home. As we indicated in chapter 2, parliamentary factions vary widely in their mix of PL and SMD members. Therefore, although the evidence is consistent with Proposition 3.5, that deputies of the same party will share similar attitudes about agenda-setting institutions, it is difficult in practice to separate the effects of faction from mode of election. Table 3.3 shows that the apparently strong effects of mode of election on attitudes about factions and their leaders dissipate when controlling for faction. But the results for faction are not so robust that we can be sure that faction is more important than mode of election.

Discussion Policy position and mode of election appear at times to structure views about important features of the Duma, albeit in diverse ways and at different times. In the 1993 candidate survey, reform-oriented candidates were more antagonistic to the Presidium model than candidates on the left, but SMD candidates were considerably more antagonistic to the Presidium model even when policy position is controlled. During the 1995 campaign, after nearly two years of experience with the new Duma, policy differences were not significantly related to attitudes about the chairman’s power and the place of parties or factions, but SMD candidates were far more dissatisfied than PL candidates with the chairman and party control of key components of the Duma. Party also contributes to the explanation of institutional preferences, as is evident in Tables 3.2 and 3.3. In both tables, the Communists are the missing category for the dummy variables representing the parties. In the 1994–95 Duma, party-based differences emerge in several places, even when policy position and mode of election are controlled. Perhaps most notable, December 12 and New Regional Policy, both factions composed of SMD deputies, stand out for their opposition to the role of factions and faction leaders in the Duma. These are factions deliberately organized by SMD deputies who seek to retain some independence of electoral parties and their parliamentary factions, so here the party effect is largely the equivalent of a mode-of-election effect. Similarly an entirely SMD-composed faction of the 1996 Duma, Russia’s Regions, proves distinctive for its dissatisfaction with the Council and faction leaders. Collinearity between faction and policy is evident, as in the several cases where the addition of the faction variables reduces the independent strength of the coefficient for policy position. For example, policy position has a strong relationship to attitudes about the Council among

70

CHAPTER THREE

TABLE 3.4 Logit Estimates of the Effect of Faction on Attitudes about the Council of the Duma among Russian Deputies, 1996 Logit Coefficient CPRF Missing category APR .50 LDPR 1.79 PPwr 2.32* NDR 1.34 RReg 3.09* Yabloko 2.20* * Chi square = 17.3

Mean Policy Position .91 .53 .36 .32 −.62 −.70 −.77

p < .05 Source: 1996 deputies survey.

*

deputies in 1996, but loses its punch when the faction variables are added. When the factions are considered alone (Table 3.4), factions whose mean policy positions are most distant from the Communists, who anchor the left side of the policy spectrum, show significant logit coefficients because Communists are treated as the missing category. In fact, in only one case—attitudes about the chairman in 1996—does policy position retain its explanatory power once the faction variables are introduced. Collectively the faction variables appear to do better than policy position when both are operating. Similarly, on certain items—such as whether faction leaders are considered to have undue influence within their factions—variance across factions appears to wash out any separate effect for electoral category. Factions’ institutional preferences thus divide along ideological lines on some issues, and according to their different electoral composition with respect to others. This finding is consistent with our conjecture that factions provide their members with a collective good—a value added—by enabling them to trade off electoral and policy-related pay-offs from their parliamentary activity. All five propositions appear to be supported by the early record of the Duma. Individuals’ attitudes about the Presidium and Council of the Duma are correlated with their policy preferences at the time the issue was first decided. No reliable majority faction or coalition emerged in the first two Dumas, and both Dumas maintained a nonmajoritarian agendasetting mechanism. Mode of election proved important with respect to the role of parties. PL deputies preferred strong parties and a party-based system of agenda setting; SMD deputies preferred weaker parties and an agenda-setting system based on nonpartisan principles and the representation of committee chairs. Some faction differences were apparent in

COUNCIL OF THE DUMA

71

attitudes about factions and their leaders, but the differences between factions in attitudes about the Council of the Duma were minor by the end of the first year.

Conclusion Preferences about institutional arrangements among deputies and candidates in the early Dumas appear to show important temporal and political patterns. The Presidium model was salient in the fall of 1993, but its successor, the Council of the Duma, quickly gained acceptance and proved quite uncontroversial by the end of the next year. Initially, right and left divided over the utility of the old Presidium model, and some SMD deputies were concerned that faction representation in the Council disadvantaged them. But by late 1994, when we interviewed deputies, general satisfaction with the Council’s exercise of power was expressed, although policy positions and mode of election continued to shape preferences about giving committee chairs voting rights on the Council well into the first Duma. These differences evaporated in the second Duma. Plainly the salience of institutional features rises and falls over time, and sometimes very rapidly. The relevance of institutional features to the policy, electoral, and partisan interests of politicians also varies over time. And policy, electoral, and partisan interests appear to play some role in the structuring of preferences about agenda-setting institutions in the Duma.

NOTES 1. Chapter 4, below, details the arrangement by which committee chairmanships were distributed among factions. 2. There are many complications. For example, only some SMD candidates have a party affiliation that is known at the time of the election; others run with no known party affiliation. Some run as both SMD and PL candidates. For the 1994 and 1996 surveys of deputies, we exclude deputies from factions with very little representation in our samples. In both cases, the missing category for the party dummy variables is the Communist Party. 3. Candidates who appeared on both party lists and district ballots are grouped with party-list candidates.

FOUR SETTING A FRAMEWORK FOR PARTY-COMMITTEE RELATIONS

I

N THIS CHAPTER, we go beyond the issue of faction and committee representation on the Council of the Duma, addressed in chapter 3, to consider the political foundations for the additional choices made about the committee system and faction-committee relations. The Duma had to establish a committee system, determine committee jurisdictions and leadership posts, and allocate committee seats and chairmanships to the members. Moreover, the factions had to determine their members’ relationships to committee work. As with other institutional choices, we can readily derive a set of expectations about deputies’ preferences from their electoral, policy, and partisan interests. We begin with some essential background and review our expectations about the alignment of institutional preferences among deputies. We draw on both survey data on candidates’ preferences about the committee system and roll-call voting on concrete proposals about the committee system in order to test our predictions. We discover that deputies were motivated by a mix of electoral, policy, and partisan interests and that no one of those interests counts for all key features of faction-committee relations. We point out as well that the institutions inherited from the past, combined with the timing of events in late 1993 and early 1994, influenced the formation of a committee system incorporating the new system of factions within the Duma.

Inherited Institutions, New Preferences A system of standing committees was a central organizational feature of the supreme soviets during the communist and transition periods. Before Gorbachev, the USSR Supreme Soviet was nominally bicameral and each chamber had a system of ten standing committees, which were exactly parallel in structure. The Supreme Soviet Gorbachev reformed in 1989 had fourteen joint committees, and each chamber had four legislative commissions, differing according to the chamber’s nominal representational functions. The Russian Republic’s Supreme Soviet in the 1990–93 period used a similar structure, with four legislative commissions in each chamber, plus twenty joint committees of the Supreme Soviet.1

PA R T Y- C O M M I T T E E R E L AT I O N S

73

Many deputies and staff had a great deal of experience with a committee system when the new Duma convened in January 1994. They had knowledge of committee structure and procedures; they were familiar with scheduling and floor procedures associated with committee legislation, including bill referral, successive readings, amending processes, and the circulation of drafts to multiple committees to obtain the “clearance” (soglasovanie) of other committees for particular pieces of legislation; and they had considerable experience with working groups of deputies, executive branch officials, and experts, formed under the auspices of committees, in the drafting and rewriting of bills. The principle of a standing committee system excited no controversy in the planning commission that designed preliminary rules before the Duma convened or later among deputies in the Duma. The Duma would have a large, complex agenda and need to fill voids in law left by the collapse of the communist system. A committee system would provide the essential division of labor while retaining the formal status of legislators as coequal members. The commission charged with designing preliminary rules for the Duma assumed that committees would be a central feature of the policy-making process. Still, several questions concerning committees were contentious, particularly those affecting the relationship between factions and committees. Inherited from the transitional parliament, and reiterated in interviews, was the expressed norm that “legislative” and “professional” matters were the prerogative of committees, whereas “politics” was the domain of factions. And as in other national parliaments and U.S. state legislatures, a strong committee system was considered essential to an influential parliament. But these sentiments left unresolved the problems of managing political conflict over legislation itself, distributing chairmanships and allocating memberships to committees, and determining the degree of control that the Council of the Duma and the membership at large would exercise over committee work. In a new decision-making process more thoroughly organized around parliamentary factions, deputies confronted these problems for the first time. Specifically deputies had to choose among institutional arrangements that would determine how centralized or decentralized the policy-making process of the Duma would be and that would influence the relationship between factions and committee members. Many party leaders and leading deputies were well aware of the committee systems developed in American and European parliaments. During the 1990–93 period, many Russians visited other parliaments, attended conferences on legislative institutions, and consulted with procedural experts from Western systems. Unfortunately the American and parliamentary systems provided only little guidance for deputies operating in the

74

CHAPTER FOUR

hybrid party and faction system that the 1993 electoral system generated. And the 1993 election outcomes seem to have eliminated the possibility that a cohesive majority coalition would be able to design and impose a system of committees that suited its mix of political interests. The committee system and faction-committee relations were in doubt in the month between the elections and the convening of the Duma. Deputies’ electoral, policy, and partisan interests were implicated in the choices to be made. On the electoral side, differences in outlooks between party-list and SMD deputies could be expected. Party-list members might reason that their electoral fortunes were tied to party and faction. Therefore they would seek to give their factions a strong hand in shaping the committee system, allocating committee assignments, and directing the behavior of committee members. SMD deputies, with an interest in emphasizing issues and pursuing activities related to reelection in geographically defined districts, would be less supportive of a faction-dominated system of committees. They would prefer a committee system that had a jurisdictional arrangement that recognized issues of local concern and that allowed them freely to pursue those issues in committee action and legislation. A system that exclusively reflected the national policy agendas of faction or party leaders, that handed committee chairmanships to faction elites, and that granted agenda control to faction leaders would not serve the disparate and more local interests of SMD deputies. Policy interests surely were involved. The intense conflict between the supporters and opponents of reform could be expected to generate a struggle for advantage in the new committee system. Preferences for a centralized decision-making process, for example, would be expected to depend on whether a cohesive majority coalition was likely to form and use a centralized process to its advantage (Smith and Deering 1990). Even with the election results in, deputies were uncertain about the policy and partisan preferences of many SMD deputies. Yet, in December 1993, it appeared that no cohesive majority coalition would exist. And with no majority ready to impose its will, no serious challenge to the basic outlines of the inherited legislative process was mounted. Our interviews and review of the published record did not uncover any serious consideration of significant change in the place of committee in the decisionmaking process of the parliament. Attention quickly turned to the distribution of committee chairmanships and other assignments. Timing and uncertainty were crucial to the developments. Because the parties that won seats were organized and represented by leaders, their parliamentary factions were advantaged by the possibility of conducting discussions about the allocations of committee chairmanships and rank-

PA R T Y- C O M M I T T E E R E L AT I O N S

75

and-file seats before the Duma convened. In fact, the sharp policy conflict was funneled through party faction leaders from the start. Discussions among faction leaders began immediately after the December 12 elections and continued until the Duma convened on January 11. The early involvement of faction leaders was crucial. We emphasized in previous chapters that parties have interests that predate and are external to those of the parliamentary factions. Consequently the policy interests of the factions’ members did not necessarily exhaust the partisan or factional interests pursued in the presession and early session discussions about committee allocations. For example, faction leaders may seek to create processes that increase their personal influence over their own rank-and-file faction members. Even if leaders act in the collective welfare of faction members, they may blend the electoral and policy interests of their faction colleagues in a way that advantages some and disadvantages others. That faction leaders took the initiative to design a process of allocating committee assignments meant that faction, or at least leadership, interests would influence the options considered. And because faction leaders have special leverage over party-list deputies and can bring to bear resources that are important to SMD deputies, they may be in a position to motivate allegiance to factions beyond what policy concerns would predict. Moreover, factions would not necessarily emphasize the same issues with equal intensity. While we know from previous work (see chapter 2) and from the behavior of deputies in the transitional parliament (Remington et al. 1994) that factions aligned neatly on the generalized left/ right, or reform/anti-reform, dimension, we also know that the factions differed in the degree to which particular policy issues were salient to them, reflecting both ideological and constituency interests. The Agrarians were most concerned about land rights, farm subsidies, and other agricultural matters; Russia’s Choice deputies cared strongly about advancing economic reform; Zhirinovsky had a keen interest in foreign affairs; the Communists were concerned about preserving state ownership of industry. These variations in the intensity of interest in specific policy problems are likely to have influenced faction leaders’ strategies in seeking to structure the committee system. Minimally they could provide a basis for arranging committee jurisdictions and assigning deputies to chairmanships and other committee posts. Positions on key questions, such as the means for selecting committee chairs, would depend on the balance of forces in the Duma. But the balance of forces remained in doubt as the Duma convened. Several days before the Duma was to convene, Russia’s Choice threatened to withdraw from the talks on the grounds that the demands of the Communists and

76

CHAPTER FOUR

the LDPR for committee posts had become unreasonable. Russia’s Choice wanted to control the committees responsible for economic legislation, the Communists were claiming the committees on defense and security, and Zhirinovsky wanted the foreign affairs committee (RFE/RL Daily Report, January 10, 1994). Presumably the Communists and LDPR were allying against Russia’s Choice (the largest faction but one whose strength fell far short of a majority) to deny it the right to control the economic committees. Because the actual balance of political forces in the new Duma was not clear at that point, it is likely that Russia’s Choice hoped that referring the decision over the distribution of committee chairmanships to the floor would yield a more favorable outcome than an interfactional bargain. The strategy of Russia’s Choice leaders changed when the voting in the first several days of the Duma’s session demonstrated conclusively that the reformers were outnumbered by their opponents. The clearest indication of this was the election of Ivan Rybkin to the chairmanship. Rybkin, a moderate Communist, was supported by the Communists, the Agrarians, and their allies and won by an extremely narrow margin. Following his election Russia’s Choice changed its strategy and supported institutional arrangements based on interfactional collective control of the chamber (Haspel 1998). Arrangements based on the preferences of a simple majority on the floor seemed less attractive than an arrangement reflecting proportionality among the factions. There is reason to suspect that the Communists, too, changed their tack in favor of an open floor debate on committee allocations once the balance of forces on the floor became clear. The following propositions, addressing different dimensions of the choices made, are suggested: Proposition 4.1. Party-list deputies are more supportive than SMD deputies of a centralized decision-making process and of faction control of committee positions and members. Proposition 4.2. Reform-oriented deputies prefer faction-based proportionality in the allocation of committee positions, whereas opposition deputies prefer a majoritarian procedure for the allocation of committee positions. Proposition 4.3. Faction leaders, more strongly than other deputies, favor faction control over committee members. Proposition 4.4. Variation in the intensity of factions’ interests across issues produces different emphases in demands for creating committee jurisdictions and positions and provides a basis for allocating jurisdictions and positions.

PA R T Y- C O M M I T T E E R E L AT I O N S

77

Duma Action on the Committee System The potential importance of the sequence of events for preferences about the committee system requires that we keep in mind the chronology of Duma actions in creating its committee system. Uncertainty about the election outcome—how many seats would be allocated to each party, if any—limited the ability of candidates in the 1993 campaign to foresee how their interests might be affected by the institutional choices to be made. During this period the planning commission made little progress in detailing the committee system and establishing procedures for allocating committee positions. Then, just a few days after the Duma convened in January 1994 the deputies confronted votes on a plan to establish committees and allocate committee positions. These votes provide some basis for evaluating Propositions 4.1 and 4.2. Propositions 4.3 and 4.4 require a more qualitative account. After reviewing these stages in the process, we turn to deputies’ attitudes about issues about the committee system that remained contentious after initial choices were made. Preliminary Preferences For most candidates in the 1993 campaign, the details of the committee system surely were not a concern. The work of the Yeltsin-appointed commission was not publicized, and it was difficult to predict what the real choices about a committee system would be. Of course candidates were well aware of the general context—Yeltsin had disbanded the transitional parliament, jailed the leaders of the opposition, banned the old communist party, and was proceeding to have his constitution ratified by the electorate. Moreover Yeltsin’s allies were exploiting their considerable advantages—access to money and the media—to achieve a favorable election outcome. If any prediction were warranted, it would be that Yeltsin would gain a more friendly parliament that would be able to proceed with economic and political reforms. Under the circumstances, many candidates might not have an opinion about features of the committee system and, among those who did, most would not have much confidence in their judgment about the best arrangements. If anything, reformers and SMD candidates might prefer a more party-oriented, majoritarian system than opposition and party-list candidates. The patterns in responses to our survey questions support this view.2 More than a quarter of all respondents, and one-third of party-list candidates, indicated “hard to say” to our question about whether a majority party or bloc should have the power to name committee chairs. We

78

CHAPTER FOUR

TABLE 4.1 Support for Majority Bloc Control of Committee Chairmanships, Party-List Candidates in 1993 Party CPRF LDPR APR DPR Yabloko PRES RC Total

Percent

N

8.0 73.9 56.3 16.1 0.0 12.5 58.8 38.5

25 23 16 6 9 8 17 104

certainly have reason to question whether other candidates had a firm opinion. Only weak relationships existed between attitude on majority control of chairmanships and the candidates’ mode of election or policy views (a Pearson r of −.11, p = .045, for mode of election; −.07, p = .196, for policy positions). Contrary to our expectation, somewhat more SMD deputies favored majority control, but only a minority of either SMD or party-list candidates favored majority control. Candidates on the reform side of the political spectrum were only slightly more supportive of majority control than opposition candidates. Party differences were more apparent (Table 4.1). Unfortunately we have party affiliation only for party-list candidates in the survey, so the Ns are very small and a full multivariate analysis is not possible. Candidates of Russia’s Choice, the party most closely connected to Yeltsin and led by reformer Egor Gaidar, favored majority control by a 3-to-2 margin. Communist candidates opposed majority control by a 9-to-1 margin. Candidates of Zhirinovsky’s LDPR favored majority control. They may have anticipated that their leader’s remarkable popularity would generate a large number of Duma seats and make it part of a majority coalition. We cannot readily account for the attitudes of Agrarian candidates. Their ideological cousins, the Communists, clearly feared majority control during the campaign but they would not have feared it after the election. Voting on the Chairmanship Package Agreement After the Duma convened in January and the composition of the chamber was more fully understood, faction leaders continued to discuss the structure and membership of committees that began operating in December. Over the weekend of January 15–16 they completed negotiations on a “package deal” and presented it to the chamber on Monday,

PA R T Y- C O M M I T T E E R E L AT I O N S

79

January 17. Then, in a series of votes, the Duma decided several crucial issues concerning the committee system. These votes allow us to examine the relationship between the revealed preferences of the deputies and their electoral, policy, and partisan interests. The package agreement represented a faction-based solution to the problem of distributing committee chairmanships. The agreement provided factions shares of committee chairmanships in proportion to their size and, critically, allowed factions to gain chairmanships and vice chairmanships for committees in which they had the most intense interest. Under the plan, a faction could bid for committee chairmanships by placing “chips” on each position it desired. Factions were given bidding rights in proportion to their size and then could allocate their chips as they pleased—spreading them widely across a large number of committees or concentrating them in particular committees and positions of greater importance to them.3 The interim organizational commission had recommended the creation of just eight committees for the Duma, but the number and jurisdictions of committees remained in play during the negotiations. Zhirinovsky demanded the chairmanship of the foreign relations committee, but nearly all other faction leaders did not trust him to handle the job properly. The leaders tacitly agreed to deny Zhirinovsky his demand and later created a committee on “geo-politics” so that an LDPR member could chair it. By the time the negotiations on the plan were complete, twenty-three standing committees were created. Jurisdictions were tangled, and some committees ended up with very small legislative workloads; however, faction leaders professed satisfaction with the resulting agreement and presented it to the floor on Monday, January 17. With this agreement, factions were staking a claim to some ownership rights over committees. In subsequent years they often referred to an implicit “gentleman’s agreement” to the effect that once a faction had been given the right to name a member to the chairmanship of that committee, it could remove or replace that member from the chairmanship position. As we shall see, this right proved largely unenforceable, but it has frequently been asserted as chairs have been removed and replaced. More important, faction leaders were asserting control over committees. This became apparent when the Duma agreed to the proposal to restrict membership on the Council of the Duma to faction leaders (see chapter 3). Committee chairs could attend Council sessions, as could any member of the Duma, but they would lack voting rights. The issue had been left unresolved before the weekend agreement, but once the grand bargain over committee chairmanships was confirmed by the floor, it was ruled that only faction leaders and not committee chairs

80

CHAPTER FOUR

would be voting members of the Council on the Duma. These decisions were consistent. They established an arrangement whereby factions rather than committees, and specifically faction leaders, would collectively manage the lower chamber of parliament. Other points, such as the rules regarding membership on committees, chairs’ powers over legislation, membership, subcommittee formation, and agenda setting, reflected a similar preference for collective party control rather than a decentralized committee government. The voting alignments on the package depended, of course, on the motions that were offered. In fact, the deputies were confronted with a negotiated package of many dimensions. The package outlined the number and jurisdiction of committees, a mechanism for assigning chairmanships, an implicit assignment of certain chairmanships to certain factions, and a separate mechanism for making other assignments to committees. No doubt the plan received mixed evaluations from many deputies. If some SMD deputies were concerned about the faction-dominated allocation mechanisms, many of them would be pleased that their factions would gain chairmanships important to them. If reformers were disheartened by the prospect of opposition deputies chairing key committees, they could take heart in their own opportunities to be assigned to committees of their choice. Consequently tests of the propositions outlined above are likely to be inconclusive. The intricate agreements among faction leaders in the package deal were reflected in the crosscutting patterns in the voting. Multivariate estimates of the effects of mode of election, policy position, and faction membership on voting on two key motions are shown in Table 4–2. The alignments are only partially consistent with our expectations. Consistent with our expectations, party-list deputies were more likely to support the package than SMD deputies by a 77-to-52 percent margin. Support for the package was greater among opposition deputies than among reform deputies. We expected the outnumbered reformers to be satisfied with a proportional outcome. Factional influences are also evident. Two factions, the Agrarians and the LDPR, differed in their mix of SMD and party-list deputies, and both shared an anti-reform policy stance with the CPRF faction, but both were noticeably more supportive of the package than was the larger CPRF, which is the missing category for the faction dummy variables. In fact, deputies of the large factions at the two ends of the policy spectrum—the CPRF and Russia’s Choice—were internally divided about the package. Perhaps some members of both factions were dissatisfied that the package deal allotted disproportionate gains in influence for the smaller factions.

PA R T Y- C O M M I T T E E R E L AT I O N S

81

TABLE 4.2 The Effects of Electoral, Policy, and Faction Influences on Key Votes on the Committee System and Chairmanships, January 1994 (logit estimates) Independent Variables Faction APR RW LDPR WR DPR NRP PRES LDUD12 Yabloko RC Mode of election Policy position Constituency Constant N −2 log likelihood Chi-square

Dependent Variables Reconsider Vote on Adopt List of Committee Committee Composition Chairs and Deputy Chairs −1.92* 9.45 −2.45* −.89 −.89 2.56* −9.07 1.59 .63 −8.39 1.43* .96* .34 −1.19* 302 222.02 166.89*

1.61* −10.44 2.84* 1.69 −.17 .01 8.09 −1.53 .66 7.93 −1.35* −.96 .36 2.10 341 216.31 97.85*

p < .05 Source: Roll−call voting in the Duma.

*

Composing the Committees Ultimately faction control of the distribution of committee chairmanships was achieved at the expense of unorganized deputies and independents who joined nonparty groups, the LDU of December 12 and Russia’s Way. As Table 4.3 shows, these deputies had no opportunity to claim leadership positions. Moreover, as we have seen, the democratic and leftist factions tacitly joined against Zhirinovsky, denying him control over the most important committees (Table 4.4). The Duma agreed to a rule whereby the composition of committees by faction would be roughly proportional to the strength of factions on the floor. This was understood to be an elastic guide rather than a rigid rule. The Agrarian faction saturated the Agriculture Committee with its members (as it had done in the USSR and Russian Republic Supreme Soviets), with little objection from other Duma members. By and large, however, a rough balance was observed in the factional makeup of committees, as shown in Table 4.5.

Faction APR CPRF LDPR NRP RC WR PRES Yabloko DPR LDUD12 RW No Faction 12.5 10.3 14.1 14.5 16.4 5.1 6.5 6.3 14.1 5.1 2.4 3.3

12.6 10.0 14.2 14.9 16.6 5.1 6.3 6.3 3.3 5.4 2.6 2.8

12.9 10.0 15.7 20.0 17.1 1.4 8.6 8.6 5.7 0 0 0

8.7 8.7 21.7 13.0 17.4 4.3 13.0 8.7 4.3 0 0 0

Percent of All Percent of All Deputy Percent of All Duma Committee Members Committee Chairs Percent of All Committee Deputies (N = 446) (N = 429) (N = 70) Chairs (N = 23)

TABLE 4.3 Committee Assignments by Faction Membership, April 1994

PA R T Y- C O M M I T T E E R E L AT I O N S

83

TABLE 4.4 Committee Chairs by Faction Affiliation Committee

Factional Affiliation of Chair

Legislation and Judicial-Legal Reform Labor and Social Support Health Protection Ecology Education, Culture, and Science Women, Family, and Youth Budget, Taxes, Banks, and Finances Economic Policy Property, Privatization, and Economic Activity Agrarian Issues Industry, Construction, Transport, and Energy Natural Resources and Utilization Defense Security International Affairs CIS Affairs and Ties with Countrymen Abroad Nationality Affairs Federation Affairs and Regional Policy Local Self-Government Issues Affairs of Public Associations and Religious Organizations Organization of the Work of the State Duma Information Policy and Communications Issues of Geopolitics

APR LDPR RC LDPR NRP WR Yabloko DPR NRP APR LDPR LDPR RC CPRF Yabloko PRES NRP PRES PRES CPRF RC RC LDPR

According to our survey, for the most part deputies themselves chose which committees they would join, but in many cases they cleared their choice with their factions. As Table 4.6 illustrates, members of partybased factions were far more likely to report that their faction coordinated committee assignments than were the independent factions. (There appears to be little difference in the behavior of SMD and PL members of the party factions in the way they coordinated their committee choices with their factions, although the numbers are too small to substantiate this claim statistically.) Faction Control of Committee Chairmanships If membership was largely decided by individual choice with some varying degree of consultation with the relevant faction, more difficult was the question of whether factions “owned” committee chairmanships.

1 2 2 6 4 1 4 2 3 0 8 2 3 3 5 1 0 1 0 1 2 2 8

2 1 1 0 1 1 4 2 2 21 0 4 1 1 1 1 0 1 3 2 1 1 1

Legislation and Judicial-Legal Reform Labor and Social Support Health Protection Ecology Education, Culture, and Science Women, Family, and Youth Budget, Taxes, Banks, and Finances Economic Policy Property, Privatization, and Economic Activity Agrarian Issues Industry, Construction, Transportation, and Energy Natural Recources and Utilization Defense Security International Affairs CIS Affairs and Ties with Countrymen Abroad Nationality Affairs Federation Affairs and Regional Policy Local Self-Government Issues Affairs of Public Associations and Religious Organizations Organization of the Work of the State Duma Information Policy and Communications Issues of Geopolitics 1 1 1 0

1 0 0 1 0 1 1 1 1 0 1 0 0 2 1 0 0 1 1 2 1 4 1

1 3 3 0 2 0 2 2 3 1 3 1 2 1 2 2 2 2 2 0 1 1 0

1 0 3 0 2 2 2 1 0 0 1 0 1 1 1 1 1 0 1 3 2 4 2

4 3 1 2 6 3 9 4 7 2 3 1 3 5 5 0 1 4 3

LDPR DPR CPRF WR RC

APR

Committee

TABLE 4.5 Committee Membership by Faction, January 1994

1 1 0 0

2 0 0 0 1 0 4 1 1 0 3 1 1 1 2 3 2 2 2 1 1 2 0

2 2 0 1 2 1 3 2 2 0 0 0 0 1 3 2 0 1 0

1 3 1 1

3 2 2 1 7 0 9 4 2 2 7 1 3 3 4 1 4 1 3

PRES Yabloko NRP

Source: 1994 deputies survey (see appendix).

39

32

15.6

15.4 9.1

25 59.4

48.7 30.8

None; I chose myself. I chose myself but cleared the decision with my faction. My faction recommended the committee. My faction made the final decision. N

CPRF

RC

Responses

33

6.1

18.2

33.3 42.2

APR

22

9.1

50 40.9

18

22.2 72.2

Faction LDPR Yabloko

16

81.3 12.5

LDUD12

TABLE 4.6 Factions’ Role in Committee Assignments in 1994 (in percent) Question: What role did your parliamentary faction [or group] have in marketing your committee assignment?

29

6.9

75.9 13.8

NRP

214

11.2

38.8

4.2 43.0

Total

86

CHAPTER FOUR

Faction leaders invoked the initial package agreement as a basis for the proposition that factions could remove chairs who were appointed by faction bidding. They argued that the agreement meant nothing unless factions could freely substitute a more loyal faction member for a chair who lost the confidence of his faction colleagues. The issue was an extension of the concept of an imperative mandate. The imperative mandate concerned the obligation of a party-list deputy to caucus with the party faction responsible for his or her election. Under the theory of the imperative mandate, if the deputy left the faction, for whatever reason, the deputy lost his or her seat in the Duma. As noted in chapter 2, the Duma rejected the theory and appeared to adopt the view that party-list and SMD deputies were on an equal footing in matters related to faction membership. Here, the faction leaders were making a parallel argument. If deputies owe their committee chairmanship to their faction, as was provided under the package agreement, then the faction must have the right to remove a disloyal chair. Here, too, however, the Duma turned away from the strong-faction system implied by the package deal. The question of faction ownership of chairmanships arose repeatedly in the course of the 1994–95 term, as committee chairs veered away from factional influence. In one case, a split developed in the DPR. The leader of the faction attempted to remove one of its members as chair of the Committee on Economy Policy, an initiative opposed by most of the faction members (RFE-RL Daily Report, October 14, 1994). Early the next year Russia’s Choice voted to recommend that one of its members, Vladimir Bauer, quit his position as head of the Organization Committee. Instead, Bauer (who had been elected in a single-member district and had had the support of his committee) chose to resign from the faction and retain his chairmanship (Segodnia, February 9, 14, 17, 1995). In July another Russia’s Choice committee chairman, Mikhail Poltoranin, left the faction and kept his post as chairman of the Information Committee. These confrontations were resolved without formal votes by the Duma, apparently because faction leaders anticipated an outcome similar to the imperative mandate decision. The episodes demonstrate that faction leaders favored strong faction powers, as anticipated. But the leaders had imperfect control of the membership of their factions when the issue of chairmanship property rights arose. Indeed, although deputies had accepted the package agreement hammered out by faction leaders early in the 1994 session, most appeared to oppose the enforcement powers that faction leaders inferred from the agreement. It is reasonable to speculate that faction leaders were able to set the agenda early in the session, having had a head start in December and facing a likely stalemate over the organization of

PA R T Y- C O M M I T T E E R E L AT I O N S

87

Duma committees. Once the Duma turned to substantive policy questions, rank-and-file deputies put the brakes on faction power, or at least the kind of faction power that faction leaders intended to exercise. Everyday Relations between Committee Members and Factions Differences among factions are addressed in chapters 2 and 5 at great length, but the consequences of these differences for faction-committee relations bear notice here. We observed above that the relationship between a faction and its committee contingents is likely to vary across factions and committees. Strong party factions, those with strong electoral parties and organizations, are likely to maintain greater influence over committee members than are weak factions. Our survey of deputies in late 1994 gives us a foundation for describing the variation. Table 4.7 indicates the percentage of respondents from each faction who reported that fellow faction-committee members coordinated their strategies “as a rule” and those who reported that the faction decided the policy positions of committee members “as a rule.” The sample size is not large enough to reliably indicate any variation in faction-committee relations across committees, but some inter-faction differences are apparent. In most factions, ranging across the policy spectrum, deputies indicated that coordination among faction committee members was the norm. Only in the LDU of December 12, a faction comprised entirely of independent deputies and formed after the Duma convened, was coordination limited to special situations. And only in the NRP, another faction comprised largely of independent deputies, did a substantial percentage of respondents report that such coordination almost never occurred. While the small Ns caution against putting too much weight on these findings, the patterns in the third column of Table 4.7 are consistent with the view that faction-committee relations vary systematically with the nature of the faction. The CPRF was distinctive in that it was the only faction for which a clear majority of its members reported that the norm was for the faction to dictate committee members’ policy positions. The CPRF, of course, had the strongest external party organization of the factions. The LDPR and Yabloko factions (and electoral parties) also registered high on external party strength and were led by prominent leaders known for their impatience with defectors. Evolving Preferences about Committees As time passed during the 1994–95 Duma, questions about the initial package agreement faded from view. A continuing concern was whether

75.0 48.5 68.2 51.7 75.0 37.5 66.7 53.8

As a Rule 25.0 48.5 31.8 27.6 25.0 50.0 27.8 38.5

Special Cases Only

Source: 1994 deputies survey (see appendix).

CPRF APR LDPR NRP PRES LDUD12 Yabloko RC

Faction 0.0 3.0 0.0 20.7 0.0 12.6 5.6 7.7

Almost Never

Faction Members of Committee Coordinated Policy Positions

62.5 36.4 50.0 41.4 37.5 37.5 50.0 41.0

As a Rule

37.5 60.6 31.8 34.5 62.5 25.0 44.4 51.3

Special Cases Only

0.0 3.0 18.2 24.1 0.0 37.6 5.6 7.7

Almost Never

Faction Decided Committee Members’ Policy Positions

TABLE 4.7 Deputies’ Reports on Faction-Committee Relations in 1994 (in percent)

PA R T Y- C O M M I T T E E R E L AT I O N S

89

TABLE 4.8 The Effects of Electoral, Policy, and Faction Influences on Preference for Council of the Duma Voting Rights for Committee Chairs, Late 1994 (logit estimates) Independent Variables Faction APR LDPR NRP PRES LDUD12 Yabloko RC Mode of election Policy position Constant N −2 log likelihood Chi-square

.03 −.43 −.80 .36 −1.42 −2.57* −2.64* −.91* .45 .06 159 191.0 23.3*

*

p < .05 Note: Factions with few deputies in the survey sample are not included. Source: 1994–95 deputies surveys.

the committee chairs should have full voting rights on the Council of the Duma. The Council, as noted in chapter 3, provided faction leaders equal voting rights but, while permitting committee chairs to attend meetings and to address the Council, withheld from the chairs full membership rights. The lack of voting rights for committee chairs gave faction leaders much stronger influence over the Duma’s agenda, while the chairs, naturally, believed that they should have such rights. Of course the political context had evolved. The implications of the package deal for the relationship of committee chairs to their factions became clearer—committee chairs gained in independence vis-a`-vis their home factions. In our survey conducted in late 1994, nearly a year after the Duma convened and more than nine months after the package agreement was adopted in January and the new rules in March, we asked a sample of deputies whether they agreed with the idea of granting voting rights to committee chairs on the Council of the Duma. Only 45 percent of the deputies agreed. Multivariate estimates of the effects of electoral, policy, and partisan considerations on deputies’ responses are provided in Table 4.8. Preferences on chairs’ voting rights in the Council were structured by faction membership and mode of election but not by policy positions, at least in a multivariate model.4 The issue tended to divide deputies of

90

CHAPTER FOUR

most factions, with the exceptions of the two large reform factions, Russia’s Choice and Yabloko. By 4-to-1 ratios, those factions opposed voting rights for chairs, yielding the distinctive coefficients in Table 4.8. We can only speculate about this pattern, but it is reasonable to suppose that members of those two factions, which were the only strongly proreform factions with committee chairs, realized that voting rights for chairs would bias the Council of the Duma more strongly against their interests. SMD deputies were more supportive of voting rights for chairs than were party-list deputies. This is consistent with our expectation that SMD deputies would be less supportive of a party-dominated process than would party-list deputies. Granting Council voting rights to chairs would dilute the influence of faction leaders and, correspondingly, increase the influence of committees, within which SMD and party-list deputies had more equal status. Still, the countervailing concerns of factional and policy interests for many SMD deputies limited the number of SMD deputies favoring voting rights—only 53 percent favored voting rights as opposed to 31 percent of party-list deputies.

Conclusion Although factions were unable to enforce their ownership rights to committees when internal divisions arose between factions and committee chairs, the package agreement itself was not renegotiated nor were any of the initial agreements over Duma rules regarding the rights of factions and committees in the Council of the Duma. The early procedural decisions proved to be durable. The Duma voted to approve the full set of Standing Orders at the first reading on February 17, 1994, despite protests from some that the rules gave disproportionate influence to parties or to one wing or another, and passed the final reading of the new rules on March 25. Individual factions were left to their own devices to resolve problems of enforcing factional discipline over committee chairs, but collectively faction leaders retained the right they had won to set the agenda of the chamber, to refer bills to committees, to set the timetable for the consideration of legislative business, and to work out basic political agreements to move legislation forward. In all these collective rights, the faction-based Council of the Duma departed significantly from the committee-based predecessor institution, the transition-era Presidium, which was made up of the chairs of the standing committees. In adopting and retaining these arrangements, deputies were crosspressured. Differences in the electoral interests, policy positions, and

PA R T Y- C O M M I T T E E R E L AT I O N S

91

factional interests surfaced during the debates and votes, as well as in our surveys. Party-list deputies were more supportive of a centralized decision-making process and of party control of committee positions and members than were SMD deputies. Reform-oriented deputies emerged less supportive than party-list deputies of voting rights for committee chairs on the Council of the Duma. Factional differences were apparent, even controlling for mode of election and policy position, at several stages. And faction leaders more strongly favored faction control over committee members than did other deputies. These findings support our argument that, in addition to general policy-based dispositions that motivated factions to seek to control committees, particular factions have special interests in particular policy jurisdictions and pursue strategies with those interests in mind. The sequence of events was important, too. The use of party-list, proportional representation for half the seats generated national parties and gave factional interests a head start in fashioning a committee system and allocating committee positions within the Duma. The inter-faction negotiations reflected both the relative strength of the left- and rightwing camps, and the ability of party-based factions and independents’ groups to attract members in the first days of the Duma’s session. The election of the chamber’s chair informed them of the balance of forces within the Duma. Subsequent negotiations reflected the absence of a cohesive majority bloc but still produced a package agreement for the organization of the Duma that represented the domination of faction interests. Only later, when the powers claimed by faction leaders were challenged, did the power of factions to organize the Duma become circumscribed. Committee chairs asserted a degree of independence from factions, although the relationship probably varied widely across factions and committees. There is little doubt that the formation of the Duma’s committee system reflected strategies influenced by electoral, policy, and partisan goals. No single goal appears to account for the choices made. The most important evidence is the strength of each factor in the multivariate estimates for voting on the package agreement. The assertion of faction ownership of committee chairmanships by leaders of left- and right-wing factions and of party-list–dominated factions and SMD-dominated factions demonstrated the independent interests of faction leaders. Further, the finding of electoral and partisan forces at work in shaping preferences about Council voting rights for committee chairs shows the continuing cross-pressures in deputies’ calculations about key features of the committee system.

NOTES 1. Including the Constitutional Commission. 2. See the appendix for details on the survey and the measure of policy preferences. 3. The package agreement also included a provision for allocating the top chamber posts. Because the Communists had already won the chairmanship of the chamber, the factions agreed that a Russia’s Choice member should be the first deputy chairman, responsible for managing the flow of legislative administration. 4. We also tested for policy preference effects and found none. The weak linear relationship between policy position and support for chairs’ voting rights masks a modest curvilinear relationship. Deputies on the moderate left of the policy spectrum were somewhat more supportive of voting rights than were other deputies.

FIVE CHOOSING AN ELECTORAL SYSTEM

A

N ESSENTIAL FEATURE of every legislative system is the mechanism for electing legislators. If other political systems are a measure, the early choices about electoral laws often have enduring consequences (Lijphart 1992; Hibbing and Patterson 1992; Nohlen 1984; Bawn 1993). Once adopted, decisions on such fundamental issues as the design of an electoral system can become self-perpetuating. As winners under the existing system seek to preserve it, it becomes less likely that an alternative system will be favored by the majorities or supermajorities required to enact it. The mixed electoral systems of postwar Germany and postcommunist Hungary, which are similar to the Russian system, illustrate the point. Both acted to lock in the initial advantages of the major political parties that negotiated the original agreement. In these systems, no coalition of parties opposed to these arrangements has been strong enough to overturn them. In this chapter we examine the choice of an electoral system for the Duma that was made by the Duma itself in 1995. Given an opportunity to alter the system that was put in place by presidential decree in 1993, the Duma chose to retain the system with few changes. That system, a hybrid of single-member-district and proportional representation systems, remained in place for the 1999 parliamentary elections. The 1995 electoral law provides an opportunity to determine whether the policy goals, electoral interests, or partisan concerns of Russian deputies structured their strategies on the electoral law.

Alternative Theories of Electoral Systems Emerging democracies adopt their initial constitutional and electoral systems in different ways. Those of postwar Germany and Japan followed military defeat and foreign occupation. Several European postcommunist states adopted new representative institutions through roundtable talks between incumbent communist rulers and challenger elites with a broad popular mandate (Jasiewicz 1993; Olson 1993). In contrast, the Russian electoral system of 1993 was the product neither of defeat in war nor a pact between old and new elites. Following the 1990–93 transitional period, during which Russia became an independent state upon

94

CHAPTER FIVE

the demise of the Soviet Union, Russia’s current representative institutions were decided mainly by pro-reform figures associated with President Yeltsin. Yeltsin instituted the electoral law governing the parliamentary elections of December 1993 by presidential decree. A new electoral law was approved in 1995 by the Duma and enacted into law following a protracted wrangle among the two chambers of parliament and the president. Although Russia differs from 1989 Hungary and postwar Germany in the way its electoral system was adopted, the system chosen, like theirs, is mixed. In Russia, the numbers of proportional and single-memberdistrict seats were equal and were filled through separate ballots in the December 1993 elections. The system was implemented by presidential decree after Boris Yeltsin disbanded the parliament controlled by opponents to his plans for economic and political reform and for a new constitution. Despite its tortuous legislative history, the 1995 law governing future Duma elections provided for an electoral system that differed little from that of 1993. At first glance, the 1993 electoral decree and the 1995 electoral law should be additional cases of institutional choice dictated by the implications of election results for control over policy-making institutions. Yeltsin and his closest lieutenants seemed motivated by the desire to proceed with constitutional and economic reforms. The president appeared to be free to decree an electoral system that maximized his allies’ numbers in the new parliament. And he consulted with talented specialists who could provide advice based on the results of the April 1993 referendum, among other things. In 1995, the lessons of 1993 and subsequent events could be incorporated in the calculations of the participants. On the surface, the model of choice suggested by Bawn (1993) should account for Russia’s 1993 and 1995 electoral systems. Bawn proposes a model based on three assumptions: (1) parties’ preferences for electoral systems are defined over policy outcomes; (2) parties make use of all available information about the electorates’ voting preferences; and (3) parties participating in the choice of electoral institutions know the rules and the preferences of the other parties involved in the choice of electoral system. The model, Bawn continues, yields the expectation that parties “act to bring about the electoral system that they expect to result in the most desirable policy outcomes.” Moreover, we would expect that, once established, a party’s “ranking of alternative electoral systems will not change unless exogenous events change the way alternative systems affect subsequent policy choices.” Bawn shows, for post–World War II West Germany, that these propositions correctly predict that the Social Democratic Party (SPD) would support a proportional representation system in 1949 and switch to the cur-

ELECTORAL SYSTEM

95

rent mixed PR/SMD system in 1953. In both systems, the total proportion of seats allocated to each party was determined by the voting on the second (party-list) ballot. But the mixed system allows voters to reward incumbents or to cast ballots based on candidates’ personal qualities without affecting the proportional distribution of seats in the popular chamber. In both the 1949 and 1953 laws, Bawn argues persuasively, the SPD chose not to maximize its own seats but rather to minimize the likelihood that its main rival, the Christian Democrats, would win an outright majority of seats, thus increasing its own chances of leading a coalition government and shaping policy. The result was a lasting electoral system backed by a coalition of the Social Democratic Party and its smaller partner, the Free Democratic Party. The institutional context within which an electoral system operates is bound to influence the participants’ views of that electoral system. Germany’s parliamentary system placed a premium on government control, which is accompanied by policy control. In Russia’s separation-ofpowers system, parliamentary electoral outcomes have less direct policy consequences but, nevertheless, might be viewed by deputies primarily in terms of implications for policy. The deep conflicts over economic and political reconstruction could be expected to influence deputies’ views of nearly any institutional feature that might give an advantage to reformers or to their opponents. A possibility Bawn did not anticipate is that individual deputies evaluate proposals for change in the electoral system with their personal political careers in mind. With half the deputies chosen from party lists and the other half from district elections in 1993, variation in elected deputies’ evaluations of alternative electoral schemes could be expected in 1995, when the law was reconsidered. By then, the deputies had had time to evaluate the results of the 1993 elections and listen to arguments about changing the system. In addition, parliamentary factions may pursue collective goals that are distinct from individuals’ goals—to enlarge their representation, establish favorable public images, and pursue legislative objectives. And their leaders may have goals of their own, goals that may be incompatible with electoral or policy objectives of some rank-and-file faction members, at least partially. This may be most likely in the early years of a democracy, when parties are just forming, charismatic leaders start new parties, some parties are comprised of a cadre of political friends, internal party rules and organizations are not yet elaborated, and the resources of rank-andfile members are limited. Therefore, a third possibility is that deputies’ preferences for electoral laws are influenced by central party and faction leaders.

96

CHAPTER FIVE

In this chapter we examine the electoral, policy, and party foundations of the 1995 election law. The contending propositions are as follows: Proposition 5.1. Strategies on the 1995 electoral system legislation were correlated with deputies’ policy preferences. Proposition 5.2. Deputies’ sought to preserve the mode of election under which they were elected to the parliament. Proposition 5.3. Strategies on the 1995 electoral system legislation were driven by the preferences of central party leaders. We proceed by (1) providing essential background on the choice and consequences of the 1993 electoral system; (2) drawing inferences about the effects of the 1993 system for party representation in the Duma; and (3) using those inferences to understand the strategies of the deputies in adopting an election law in 1995. A statistical analysis of support for the key proposals in the 1995 debate, backed by an account of the debate itself, will allow us to evaluate the three propositions.

The 1993 Electoral Decree Although figures close to the president had the final word, the 1993 electoral law was the outcome of a lengthy process of deliberation in which many political figures and specialists participated. Members of the old parliament, since January 1993, had been working on a new electoral law in anticipation of new parliamentary elections (Sobyanin and Sukhovol’skii 1993/1994; Urban 1994). The working group included a number of deputies and specialists who were not at all close to the Yeltsin entourage. The group’s proposal was the basis of President Yeltsin’s decree and made many changes in the old electoral and parliamentary systems besides reducing the number of seats. The proposal included the following provisions, all of which were incorporated in Yeltsin’s decree: 1. Mixed PR/SMD system. The proposal introduced a system of proportional representation in Russian parliamentary elections. Half the 450 seats in the Duma were to be allocated to candidates elected from party lists according to the share of the vote each party received in a single, Russia-wide electoral district. The other half were to be filled by the winners of contests in single-member districts elected on a first-past-the-post, plurality basis. The party-list and district voting were conducted on separate ballots. In addition, the plan called for a single nationwide district instead of multiple multimember districts for the PR vote. 2. Five percent party-list threshold. To gain party-list seats, a party had to receive more than 5 percent of the party-list popular vote.

ELECTORAL SYSTEM

97

3. Single-stage election. The new plan provided for single-stage, plurality elections in territorial districts without run-offs. 4. Lower voter-turnout threshold. The proposal set a threshold of 25 percent voter turnout in order for a territorial district election to be valid (down from the 50 percent set in the 1989 and 1990 elections). 5. New Ballot Form. On the new ballot, a voter would place a mark by the box representing the preferred candidate or party rather than crossing out all unwanted choices, as on the traditional Soviet-era ballot. 6. Selection of Deputies for the Upper House. A different method was chosen for electing the deputies of the upper house, the Federation Council. In this case, and for this election only, Yeltsin decreed that the upper chamber would be elected by direct popular election. The two candidates from every constituent territorial unit (subject) of the federation who received the most votes were declared deputies to the Federation Council from that region. The political context in which the Yeltsin team formulated the decree in the fall of 1993 had features that surely influenced the choice. Three related features of that context bear notice—the identity of the players making the decision, the sequence in the creation of parties and the electoral system, and uncertainty about support for proto-parties. The electoral decree was not the product of an explicit agreement— a pact—among the major organized political forces over the rules of the game, as was the case in several other transitions from communism in eastern Europe. Within broad limits of acceptability to the general public, Yeltsin was free to choose the rules of competition and impose them by decree. No formal vote or parliamentary coalition-building process was involved. Nevertheless the president, who was not a member of any party, had to contend with leaders of multiple reform parties and with faction leaders who indicated their groups’ intentions to compete in the forthcoming elections. The president, one may reasonably assume, sought a system that would maximize the number of reform-oriented deputies in the new Duma, whatever their party or factional affiliation. To be sure, the arrangement of political institutions under the new constitution meant that Yeltsin could make policy by issuing decrees as long as the parliament did not actively oppose him, as the old one had. His minimum goal, therefore, was a Duma that was not dominated by a majority opposition coalition. But since the constitutional draft he supported gave the Duma the power to vote no confidence in the government, he and his advisers must have preferred that the parties sympathetic to liberal reform would gain a majority.1 Thus we assume that he wished to maximize the number of pro-reform deputies in the new Duma and minimize the number of seats won by his opponents. Of course most

98

CHAPTER FIVE

Russian parties in the fall of 1993 were little more than parliamentary factions and elite associations; their local branches were embryonic at best. Only the Communists and Agrarians could draw on an inherited organization in many localities. Both Yeltsin’s team and his opponents, however, lacked reliable information about their prospects under various schemes. The referendum of April 1993 revealed patterns of support for and opposition to President Yeltsin and to his government’s reform policy. Elections of governors in a few regions during the year gave some indication of voters’ preferences. Neither these, however, nor the elections of deputies in 1989 and 1990 and the presidential election of 1991 gave any concrete evidence of the levels of local or national support for the current array of political parties, blocs, and movements. Moreover there was no experience using the particular set of electoral districts created for the December 1993 balloting. Finally, the results of the referendum were a weak basis even for predicting a nationwide PR result because of the further deterioration of economic and social conditions in the interim, the unpredictable public response to the October 1993 violence, and uncertainty about voter turnout. Thus, while Yeltsin and his advisers surely would have liked to maximize the number of friendly deputies in the Duma in open elections, they had a weak basis for determining the kind of system that would achieve their objective. Indeed, participants we interviewed in the spring of 1993 expressed great uncertainty about the prospects of the parties under any electoral system. Uncertainty was so great that very different systems were supported by various Yeltsin advisers. The contending advisers were quite aware of the trade-offs they faced. Proponents of a predominantly SMD system offered as their primary argument a classical electoral system analysis—the reformers, as the most popular group in the country, would benefit from the bonus typically associated with an SMD, plurality system. In their view the April referendum indicated that the Russian electorate was more pro-reform than anti-reform, which would allow a pro-reform party to pick up the support of voters who would not want to waste their votes by supporting minor parties, as would be the case in a PR system. They also pointed out the importance of responding to the demands of local elites for influence over the campaigns. And they thought, following Duverger’s Law (Duverger 1954), that SMD, plurality elections would breed a two-party system that reformers would dominate for the foreseeable future. Proponents of a predominantly PR system argued that old-line Communists so dominated local governments and electoral commissions that an SMD system was a recipe for disaster for reformers. The use of party lists would allow the reformers to control the selection of list candidates.

ELECTORAL SYSTEM

99

In addition, reform-oriented partisans were concentrated in the largest cities and thus, under an SMD system, would have a difficult time fielding candidates, let alone winning pluralities, in certain regions of the country. Besides, the proponents of this position doubted that the reformers could, in the time allowed, form a single party or bloc and benefit from the usual SMD bonus. Yeltsin had been refusing to lead such a party for some time. Since the reform forces were likely to compete for the same district votes, and Yeltsin preferred to remain above the partisan contest, a PR system would allow several smaller reform parties to win seats. Neither side could muster hard evidence for its position. The conflicting assessments of the Yeltsin camp’s political situation seemed to reinforce the more normative arguments for a mixed system proffered by Viktor Sheinis, a liberal deputy who chaired the working group drafting the new law. Sheinis and others espoused the value of the local representation that comes with an SMD system but strongly argued for a large PR component in the electoral system in order to provide an impetus to the development of Russian parties. Severe time constraints reduced the Yeltsin team’s ability to analyze the probabilities associated with each alternative. Yeltsin announced elections for December as part of his actions to disband the old parliament in September. Yet Yeltsin’s advisers, who had debated electoral system alternatives for many months, appear to have settled on the Sheinis plan by the late summer of 1993, before Yeltsin dissolved parliament. As best as we can determine, then, calculating politicians struggled with what they realized was inadequate information for making the choices they faced. The uncertainty, combined with the particular institutional context they hoped would prevail after the referendum on the constitution, yielded a strategy that spread the risks across the two systems of representation.

The 1993 Election Results The results of the December 1993 elections are shown in Table 5.1.2 Thirteen electoral organizations filed candidate lists. Eight of the thirteen succeeded in receiving more than 5 percent of the votes cast and therefore gained parliamentary seats. In addition, many deputies elected in single-member-district races formed parliamentary groups independent of the party-based factions. Three formed in the first week of proceedings. Other district deputies joined party factions. The table shows each party’s percentage of the national vote for party lists, the number of seats received as a result of the list balloting, and the number of district

100

CHAPTER FIVE

deputies who affiliated with each party as of April 1994. The last two columns show each parliamentary faction’s size in the Duma. The results in Table 5.1 support the conventional interpretation that no party or bloc won a majority in the Duma. The opposition, or antireform, parties—the Agrarians, Communists, and Liberal Democrats— won more seats than the democratic, or pro-reform, camp, comprising Russia’s Choice, PRES, and Yabloko. The opposition wing constituted about 40 percent of the Duma. Would the Yeltsin camp have been better off under another electoral system? In fact, the 1993 election results provided essential information for the players and shaped their preferences about election law, which would be revisited in the new parliament. The results provide a basis for characterizing the strategic context in which the 1995 debate on the election law was conducted. Specifically the results allow us to simulate the consequences of alternative electoral systems. We use district-level data on both party-list and district-candidate voting at the level of the local district. Data come from unpublished reports provided by members of the Central Electoral Commission.3 We know that similar data were used by at least some participants to evaluate the consequences of the mixed system used in 1993.4 More details on our methods are provided in the appendix. Comparing Alternative Electoral Systems Based on the outcomes of the party-list votes, we can project the results of the elections under several alternative electoral systems. As in other simulated election analyses, great caution must be exercised in interpreting the results (Bawn 1993). Quite likely, for example, the number, size, and composition of districts would have been different had Russia used the SMD system exclusively. Moreover, we cannot assume that voters would behave the same way in an exclusively PR or SMD system as they had in the mixed system. Furthermore, we cannot treat a simulation of election results as a simulation of legislative behavior, which is, of course, what concerned Yeltsin. The party discipline associated with a parliamentary system cannot be assumed in the presidential-parliamentary system of Russia. Consequently the electoral system itself is likely to have a strong effect on legislators’ behavior. But whether a highly party-oriented Duma would have proven to be in Yeltsin’s interest would have depended on the election results, precisely what was so difficult to predict in late 1993. With these considerations in mind, we turn to estimates of the consequences of alternative electoral systems for the distribution of deputies across factions and groups, illustrated in Table 5.2. The first column of

*

34 13 5 14 66 214

21 32 59 0 112 225

7.99 12.4 22.92 0 43.31 4.22 100

1 2 66 0 0 0 0 69

14 21 0 0 0 0 0 35

5.52 8.13 0 0.76 1.25 1.93 0.7 18.29

33 12 8 26 0 79

District Deputies Affiliated

40 18 20 0 0 78

List Seats Received

15.51 6.73 7.86 0 4.08 34.18

Party List Vote (%)

439

55 45 64 14 178

15 23 66 0 0 0 0 104

73 30 28 26 0 157

Total Seats (April)

97.56

12.22 10.00 14.22 3.11 39.56

3.33 5.11 14.67 0.00 0.00 0.00 0.00 23.11

16.22 6.67 6.22 5.78 0.00 34.89

As Percent of Duma (450 Seats)

Russia’s Way did not put up a list as electoral association; formed in first week as a group comprising SMD deputies.

Reform Parties RC PRES Yabloko LDUD12 RMDR Total Centrist Parties DPR WR NRP KEDR BRNI Civic Union DM Total Opposition Parties APR CPRF LDPR RW* Total Against all lists Total

Party

TABLE 5.1 Results of December 1993 Parliamentary Elections

102

CHAPTER FIVE

the table, reflecting the outcome of an all-SMD ballot, is based on the actual shares of district seats held by factions in April 1994 (it corresponds to the third column of Table 5.1). The last column, projecting the results of an all-PR election, applies the shares won by parties in the list voting (as adjusted by redistributing the votes cast for parties that did not clear the 5 percent hurdle to those that did) to a hypothetical parliament of 450 party-list seats. The middle column reports the seat shares that actually obtained in the half-and-half system that was used. The second column, simulating the outcome of a one-third PR–two-third SMD election, is based on the assumption that factions hold the same number of district seats that they actually hold while there are only 113 PR seats rather than 225. Thus the party-list vote is used to distribute only half as many seats as was actually the case. Of course different district sizes and boundaries would probably have yielded a different outcome. The fourth column reports the results of a simulated election in which the party-list balloting determines two-thirds of the seats. As in the other simulations, parties retain their current shares of the 225 district seats, but now there are 450 PR seats to allocate to parties based on the party-list vote. The simulation suggests that the repercussions of altering the proportion of PR seats are substantial. In an all-SMD system, the reform-oriented factions would have been better off than they were in the actual outcome. This is not because their total share differs appreciably from one system to another. While the three party-based factions in the reform camp gain a larger share of seats as the fraction of PR seats increases, the lost share of seats held by the district deputies in the independent pro-reform December 12 faction entirely offsets the gain. Instead, it is the balance between centrist and opposition factions under alternative systems that is critical for the reformers. Together, the centrist factions do much better under SMD-weighted systems because of the sizable number of independent deputies elected in districts who joined the New Regional Policy faction. In fact, the total percentage of centrist-faction deputies is twice as large under an all-SMD system as under an all-PR system. However, setting aside the NRP, the two small centrist parties—Women of Russia and the Democratic Party of Russia—gain a far larger share of seats with more proportionality. The overall percentage of opposition deputies would have been lower under an all-SMD system and higher under an all-PR system. This result is a function of the LDPR’s great success in party-list voting and its great weakness in district voting. Indeed, the LDPR’s party-list success far outweighs the Agrarians’ disproportionate success in district voting. Without the LDPR, the opposition bloc of Agrarians and Communists would improve its share only slightly as the degree of proportionality increases.5

ELECTORAL SYSTEM

103

TABLE 5.2 Simulated Results under Alternative Electoral Systems Based on 1993 Election Results (% of all seats in the duma) Alternative Electoral Systems Party Reform Parties RC PRES Yabloko LDUD12 Subtotal Centrist Parties DPR WR NRP Subtotal Opposition Parties APR CPRF LDPR RP Subtotal

100% SMD

One-third PR

50/50 SMD/PR

Two-thirds PL

100% PL

14.67 5.33 3.56 11.56 35.12

15.68 6.21 5.33 7.69 34.91

16.22 6.67 6.22 5.78 34.89

16.74 6.93 7.26 3.85 34.79

17.78 7.78 9.11 0.00 34.67

.004 .008 29.33 30.67

2.37 3.85 19.53 25.74

3.33 5.11 14.67 23.11

4.38 6.52 9.78 20.68

6.44 9.33 0.00 15.78

15.11 5.78 2.22 6.22 29.33

13.02 8.58 10.36 4.14 36.09

12.22 10.00 14.22 3.11 39.56

11.11 11.42 18.29 2.07 42.90

9.11 14.22 26.22 0.00 49.56

Thus the simulation of aggregate results shows that the choice of electoral system had significant implications for the reform forces. In the short run, the system that would have both maximized the percentage of reform deputies and minimized the percentage of opposition deputies was an all-SMD system. This is primarily the result of two factors—the strength of the LDPR in party-list voting and the large number of deputies elected in districts who chose not to join the party-based factions.6 District-Level Consequences of the 1993 Electoral System The aggregate pattern in Table 5.2 is not the full story. District-seat voting may have been affected by a number of factors beyond the party’s popularity, and these factors probably had differential effects on the parties. Parties differed in their ability to field strong district candidates and organize local campaigns. The character of interparty competition may have varied from district to district. The closely allied Agrarians and Communists might have been expected to coordinate their efforts in local districts and to try to avoid dividing the anti-Yeltsin vote.7 If so, the anti-Yeltsin parties

104

CHAPTER FIVE

TABLE 5.3 Differences between Party-List Vote Share and Candidate Vote Share, by District, 1993 Party

Mean

Standard Deviation

Number of Cases

APR CPRF LDPR RC Yabloko PRES

−9.307 −.761 12.285 −.563 −.169 −3.570

13.341 8.283 9.553 12.232 9.134 9.453

63 49 54 92 77 55

would have been advantaged in individual districts where they nominated a single candidate while the pro-Yeltsin parties nominated more than one. To determine the aggregate effect of these and other district-level factors, we examine for each party the mean difference between party-list and district voting for districts in which the party had a candidate for the district seat (Table 5.3). For Zhirinovsky and his Liberal Democrats, the party-list vote was almost always stronger than the district vote. The LDPR’s party-list vote averaged more than 12 percentage points more than the vote for the district candidate. So even where the LDPR managed to field candidates, its candidates did poorly relative to the party list. For the Agrarians, the vast majority of local candidates outpolled the party list in the same districts. For the other parties, the mean difference is quite close to zero, although there is substantial variation from district to district. Thus it appears that an exclusively PR system would have greatly advantaged Zhirinovsky and his Liberal Democrats, whereas an exclusively SMD system would have helped the Agrarians. Furthermore, as best as we can determine, neither Russia’s Choice, the large party closest to Yeltsin, nor the Communists would have gained a significantly greater proportion of seats by moving to a more exclusively PR or SMD system. However, if the Communists and Agrarians coordinated their candidate nomination efforts, it would be reasonable to argue that the AgrarianCommunist bloc, which later tended to behave as a voting bloc in the Duma, was collectively advantaged by SMD voting. Did the fragmentation of the parties undermine the district-level competitiveness of the democratic parties more than the opposition parties? This question can be answered only if we make the very strong assumption that voters for one of the pro-reform parties (Russia’s Choice, PRES, Yabloko, and the Russian Movement for Democratic Reform) would vote for any one of those parties before voting for any other party. With that

ELECTORAL SYSTEM

105

assumption, the answer is yes: The democratic parties undercut their opportunities for district-seat victories by competing against each other. More than one democratic candidate ran in 92 (54 percent) of the 170 districts in which at least one candidate ran under a democratic party label. In 22 of the 92 cases, the sum of the votes for the democratic wing’s candidates exceeded the number of votes for a winner who joined an opposition or centrist faction. In contrast, in only 39 (33 percent) of the 118 districts with at least one opposition party candidate (Communist, Agrarian, or LDPR) did more than one opposition candidate run. In only 9 of the 39 cases, the seat went to a democratic or centrist faction while the sum of the votes for the opposition bloc’s candidates exceeded the number of votes for the winner. Thus democratic parties exhibited a higher frequency of internecine competition with detrimental consequences than did opposition parties.8 Both reform and opposition factions gained members when winning candidates with no party identification joined factions in parliament. About a quarter of the victorious independents joined a reform faction (including the December 12 group), and about the same number joined an opposition faction (including Russia’s Way). But the largest number of independents joined the NRP, an avowedly centrist faction representing non-party SMD deputies. With the influx of independents, the democratic factions ended up with one-third of the district seats, and the opposition with a little less than that. Given the concerns of democrats about their prospects in district contests, the reform parties could not have been too disappointed with that result. But they probably could have been still more successful had they developed a coalition strategy as their opponents seem to have done. In sum, the parties most likely to be supportive of Yeltsin’s reform program suffered in two ways in 1993. First, under the electoral system decreed by Yeltsin, the large share of PR seats benefited a key opposition party (the LDPR) and, on net, cost the reformers seats in the Duma. Second, owing to their fragmentation and inability to coordinate in district races, the reformers lost district seats that they might have won.

The Choice of an Electoral System in 1995 In 1995 the Duma adopted a new electoral law to govern the elections scheduled for December 1995. The policy model predicts that the deputies of the Duma would prefer a system that produces desired policy outcomes. If the mixed system of 1993 represented a mixed strategy by the Yeltsin camp in the face of great uncertainty about the consequences

106

CHAPTER FIVE

of alternative systems for policy outcomes, the lessons of 1993 and intervening events would clarify the implications of each alternative. Deputies’ preferences, so determined, would, along with the decision rules of the Duma, determine the collective choice of a new electoral system by the Duma. Our task is to relate the 1993 results to the preferences of the major groups of deputies about electoral system alternatives and account for the collective choice of the Duma in light of the distribution of preferences. We begin by noting the preferences that would be dictated by the implications of 1993 election results for the policy outcomes. Given the 1993 results, policy considerations would have dictated polarized preferences for the electoral system on the part of reform and opposition deputies. As we have seen, analysis of roll-call voting in the Duma shows that a single left-right, economic reform dimension dominated Duma voting. Pro-reform deputies, as identified by that dimension, should have favored a system weighted heavily in favor of district seats so as to minimize the opposition’s share, while, for the same reasons, opposition deputies would be expected to favor a more heavily proportional system. Since neither bloc commanded a majority, centrists would determine the choice of the Duma and would want to remain pivotal on policy questions in the new Duma. The NRP was the largest centrist group and its mean on the left-right dimension is to the reform side of the overall Duma median. Thus it is reasonable to predict, on the basis of policy-oriented preferences for electoral systems, that the NRP would side with the reformers for a system more heavily weighted in favor of SMD seats. If the NRP joined with the reform factions on an electoral law, the coalition would be just a few votes short of a majority. On the other hand, a coalition of the major opposition factions with the independent factions falling just to their side of the median (Women of Russia and the DPR) would not by themselves constitute a majority. A handful of deputies without faction membership might determine the outcome. Yet, as it turned out, the Duma preserved the 1993 system: a 50–50 split of PR and SMD seats, no compensatory seats to ensure overall proportionality, a 5 percent PR threshold, and plurality district voting.9 What accounts for this outcome? In fact, Yeltsin proposed a plan that is consistent with our prediction that reform-oriented forces would benefit by a system favoring more SMD seats. He proposed an electoral law that would have increased the share of SMD seats to two-thirds (300 seats) and reduced the share of PR seats to one-third (150 seats). He also proposed that candidates be prohibited from running simultaneously in districts and on national party lists, a reform that would hurt parties lacking local branches. Most

ELECTORAL SYSTEM

107

deputies, however, including many in Russia’s Choice and other reformoriented factions, preferred to maintain the equal proportions of list and district seats. Evidently preferences defined over policy outcomes did not exclusively determine preferences for alternative electoral systems. Power- and Reelection-based Preferences Conceivably the interest of at least some deputies in protecting their chances of reelection, their personal power, or the organizational interests of their parties led them to favor an electoral law different from the one dictated by their policy interests. Research in other cases has suggested that, in defining their preferences for a new electoral law, legislators and their parties may be moved by considerations other than maximizing policy influence (Gunther 1989). In Russia, two considerations—party leaders’ interests in maintaining their personal influence and deputies’ interests in retaining their seats in parliament—appear to have been at play. Let us first consider the interests of party leaders. In early 1995 Russian parties were still in an embryonic state of organization and leaders of parliamentary factions tended to take the initiative in building party organizations. Russia’s Choice, PRES, CPRF, Agrarians, LDPR, DPR, and Yabloko were all led by the heads of their parliamentary factions. Even the parliamentary leaders of New Regional Policy, comprising independent deputies elected in districts in 1993, took steps in 1995 to organize a national electoral movement. The parliamentary leaders were able to use the resources of their parliamentary factions to coordinate communications with the regional branches, to take leadership positions in the party organizations, and to set party programs. More important, parliamentary party leaders exercised control over their electoral organizations by controlling placement of candidates on the party list. A large PR element in the electoral law served to maintain their control over candidates and the emerging electoral parties. The electoral interests of several small parliamentary factions also may have dictated preferences for electoral systems that varied from those predicted from policy interests. For instance, most members of Yabloko were elected on the party list. By adopting Yeltsin’s new proposal, they could expect that they would lose a number of party-list seats and reduce their relative influence in the Duma. Thus, even if their policy interests suggested a more heavily SMD system, their electoral interests dictated that at least the proportion of PR seats be maintained. At the other end of the political spectrum, the Agrarians had policybased reasons for supporting more PR seats but had electoral reasons to

108

CHAPTER FIVE

support the maintenance of a large proportion of SMD seats. Agrarian deputies could expect to be personally advantaged in a heavily SMD system. But this preference would pit them against the Communists and the LDPR, factions composed of more party-list than district deputies. A further consideration is that small parties benefited from the absence of a majority bloc, which could reasonably be attributed to the mixed electoral system. After the 1993 elections faction leaders settled on institutional arrangements that reduced the risk that any bloc would be disadvantaged. Since, as we have seen, chamber leadership was vested in an all-faction steering committee that was organized on a parity rather than a proportional basis, smaller factions might seek to preserve the existing multiparty, nonmajoritarian arrangements by perpetuating an electoral system that tended to reduce the chances that a cohesive majority bloc would emerge. Debate on the 1995 Law The debates over the 1995 electoral law provide a strong basis for determining the positions of the factions. In addition to the electronically tallied votes on the new electoral law, we have the record of debate on the bill as each faction offered its perspective during floor deliberation. The record supports the view that nonpolicy considerations weighed heavily in the positions taken by the Duma factions. In the debate in the second reading, the Duma voted both on individual amendments to the bill and for final approval of the amended bill. Table 5.4 reports the vote outcome by faction for the amendment the president offered that would have reduced the number of PR seats to 150 and increased the number of SMD seats to 300. The amendment failed. Table 5.4 also shows the vote on the final version of the bill, including the 225:225 seat provision. The president’s position won the support of only 143 deputies, while the final bill passed easily. Reform-oriented deputies split over the seatratio issue, but most preferred the Duma version in the final vote.10 Without the support of Russia’s Choice, Yeltsin’s plan had no chance of adoption. Yabloko and PRES, which might have favored the Yeltsin plan for policy reasons, behaved instead in a manner consistent with their electoral interests and supported continuation of a large PR component. The Communists, DPR, and LDPR preferred an all-PR system, consistent with both their policy and electoral interests, but, lacking a majority, indicated that they would settle for the 50–50 mix.11 Table 5.5 lists votes on the two alternatives according to the electoral category of the deputies. Deputies elected from the districts tended to favor the president’s proposal, whereas list deputies strongly opposed re-

1 0 0 0 1 20 14 0 36 25 25 21 143

APR LDPR DPR CPRF WR RC PRES Yabloko NRP STB ROS No faction Total

36 37 4 39 1 6 1 19 1 1 2 10 157

No 0 0 0 0 0 0 1 1 1 0 0 0 3

Abstain 14 18 4 7 19 28 6 7 1 11 10 17 142

Not Voting 50 5 8 43 21 25 13 24 10 9 11 21 240

Yes 0 4 0 0 0 4 5 0 17 14 11 10 65

No 0 0 0 0 0 4 0 0 2 2 0 0 8

Abstain

1 46 0 3 0 21 4 3 10 12 15 17 132

Not Voting

Vote on Final Passage: 225 SMD/225 PR

Note: Deputies are listed by their April 1995 faction affiliations. Two new factions were registered by then, Stability (STB) and Rossiia (ROS). Deputies who were members of unregistered groups are counted under “No faction.”

Yes

Party Faction

Vote on Presidential Amendment: 300 SMD/150 PR

TABLE 5.4 March 1995 Votes on the Presidential Amendment and Its Final Passage, Duma Election Law, April 1995 Faction Membership

110

CHAPTER FIVE

ducing the number of PR seats. In the final vote, however, even the SMD deputies supported the Duma’s bill over the possibility that parliament might fail to adopt a law and the elections would be run under another presidential decree. The evidence indicates that deputies’ own electoral interests strongly influenced their preferences for an electoral system. Other considerations weighed into the balance as well. During the debate (Biulleten’ 1995), several faction representatives spoke strongly in favor of the current system for its beneficial consequences for governance of the country and of the Duma on the grounds that a large share of PR seats tended to foster the development of parties, which contributed to the stabilization of political conflict in the country and to efficient decision making in the Duma. L. N. Zavadskaia, speaking for the Women of Russia faction, commented that “if we cut off parties and political movements from elections today, and make elections solely single-member district, we will never form a political system, the basis of which is parties.” Moreover, she noted, in the Duma, “we can pass many political decisions only because the Duma is politically structured. If we dissolve into singlemember districts, we will have to spend three or four or five times longer reaching decisions than at present.” The Women of Russia, it should be noted, included few district deputies. V. B. Isakov, chair of the Legislation Committee and a prominent representative of the Agrarian Party, observed that the half-and-half system is the “golden mean” and agreed with Zavadskaia that it tends to contribute to the development of a multiparty system in the country, which is essential to “reanimating” a political system in Russia. The golden mean for Agrarians may have involved compromising their conflicting policy and electoral interests. Table 5.6 summarizes the estimates of the effects of policy preferences, mode of election, and party on the two key votes during the election law debate. In both cases, policy and electoral interests appear to have structured voting, with little independent influence for parties. A generous interpretation would add that some deputies were concerned about the implications of the electoral law for building a viable party system. But clearly a theory of strategic choice, based solely on preferences defined over policy outcomes, cannot account for the Duma’s action on the electoral system legislation in 1995. Deputies’ personal career and electoral interests played a critical role in shaping the outcome.

Conclusion A theory of strategic choice based solely on preferences defined over policy outcomes cannot account for the adoption of the 1993 Russian electoral system or its 1995 successor. Clearly uncertainty and miscalcula-

Yes 119 24

Party Faction

SMD Deputies PL Deputies

55 102

No 1 2

Abstain 46 96

Not Voting

Vote on Presidential Amendment: 300 SMD/150 PR

104 136

Yes

58 7

No

5 3

Abstain

54 78

Not Voting

Vote on Final Passage: 225 SMD/225 PR

TABLE 5.5 March 1995 Votes on the Presidential Amendment and Its Final Passage, Duma Election Law, SMD and Party-List Deputies

112

CHAPTER FIVE

TABLE 5.6 Logit Estimates of the Effect of Policy Preferences, Mode of Election, and Party on Votes on the Presidential Amendment and Final Passage, Duma Election Law (standard error in parentheses) Dependent Variables Independent Variables Constant Policy preference Mode of election

Vote on Presidential Amendment: 300 SMD/150 PR −9.09 (42.20) −1.13* (.42) −1.32* (.62)

Vote on Final Passage: 225 SMD/225 PR 8.70 (38.94) 1.10* (.36) 2.33* (.64)

Faction No faction membership APR LDPR DPR WR RC PRES Yabloko NRP STB ROS Number voting Chi square −2 log likelihood

9.94 (42.20) 7.16 (42.21) −.15 (60.56) −.62 (141.62) 10.40 (42.22) 9.44 (42.21) 11.58 (42.22) −2.39 (74.08) 13.00 (42.21) 12.72 (42.21) 11.86 (42.20) 300 307.6 (p = .0000) 107.555

p < .05 Note: The CPRF is the dropped party category.

*

−8.19 (38.93) 1.04 (52.84) −11.50 (38.93) −.06 (102.91) −.03 (69.09) −6.08 (38.94) −7.95 (38.93) 2.28 (65.10) −9.47 (38.93) −9.07 (38.93) −9.38 (38.93) 305 151.7 (p = .0000) 164.248

ELECTORAL SYSTEM

113

tion were important influences. Uncertainty was reduced once the first election occurred and its results were examined. But even then, the choice of an electoral system reflected preferences based on more than policy outcomes. The deputies of the Duma who debated the 1995 electoral law faced a trade-off between divergent goals—maximizing the expected number of seats in parliament for political forces sharing their policy preferences, maximizing the chances of their own reelection, and, for the leaders of parliamentary parties, maximizing their personal power. The result was an electoral system much like the initial system. Indeed, the choice of an initial system proved to be self-replicating as the party factions that gained parliamentary representation chose to preserve the existing system. The battle over the election law indicates that politicians operate with mixed motives in choosing institutions. A mix of political goals, uncertainty about the consequences of choices, and the specific arrangement of political institutions greatly influenced the choice of an electoral system. The mixed system implemented by the Yeltsin camp for the 1993 electoral system appears to have been consistent with a policy-based model of choice, at least under the circumstances that the Yeltsin camp encountered in September and October 1993. Uncertain about the consequences of more exclusively PR and SMD systems but confident about the arrangement of policy-making institutions, they adopted a system of half party-list seats and half district seats for the more powerful house of the new Federal Assembly. But the 1993 Russian experience also suggests that extreme care must be taken in applying the framework to systems where the quality of the information available to the designers about the consequences of their choices is likely to be poor. It is reasonable to speculate that the inability to make informed choices is common to parties involved in structuring electoral systems in the early stages of new democracies. Undeveloped parties, the absence of a relevant electoral record, rapidly changing conditions, and other factors make estimates regarding the consequences of alternative electoral choices quite unreliable. Once a system is chosen and initial elections held, adjustments in the system can be made with some confidence about their consequences for control of policy-making institutions. Still, in 1995, after uncertainty about the consequences of alternative electoral systems was reduced, the policy-based model did not perform well. Incumbent deputies and factions had acquired a vested interest in the electoral system that brought them to office and some of them moved to reinstate that system at the expense of their policy interests. In fact, the president’s closest allies on economic reform appear to have chosen to preserve their own seats at the risk of electing fewer like-minded deputies in the next election. Only a model that incorporates both the reelection and ideological interests of deputies can explain their behavior.

NOTES 1. Under the constitution, the Duma’s powers to defeat a government through a no-confidence vote are limited. The Duma may make such a vote once with no further action required. A second vote of no confidence within three months, however, requires that the president either appoint a new government or dissolve the Duma and hold new elections. Both the president and parliament thus face a certain risk in the event that the Duma votes no confidence in the government. Moreover, the constitution specifies that the president must be guided in his decisions both by the constitution and current legislation. Opposition deputies in the Duma subsequently treated this provision as a useful loophole allowing them to attempt to pass laws that constrained presidential decreemaking power. One may assume that, after the violence of October 1993, both Yeltsin and his opponents would be reluctant to provoke new confrontations. 2. The data in Table 5.1 are drawn from several sources. Final official figures on national vote shares and the corresponding seats assigned for the parties are reported in Biulleten’ Tsentral’noi Izbiratel’noi Komissii Rossiiskoi Federatsii [The bulletin of the Central Electoral Commission] (Moscow), no. 1 (12) (1994): 67. Figures on the outcomes in district races by party are drawn from our data set. Figures on the seats held by each faction are taken from a handbook published by a Moscow research firm (Belonuchkin 1994, 79). 3. See appendix. 4. Other studies of the 1993 elections rely on voting data aggregated to the eighty-nine subjects of the federation (republics, oblasts, krai, and the two large cities) or on opinion surveys (Clem and Craumer 1993; Sobyanin, Gel’man, and Kaiunov 1994; Slider, Gimpel’son, and Chugrov 1994; Wyman et al. 1994; Whitefield and Evans 1994; Hough 1994). Such data do not allow us directly to evaluate alternative electoral systems. 5. The LDPR did rather poorly in district races. LDPR-affiliated candidates ran in some sixty districts but only won in five of them. The other major factions—Russia’s Choice, the Communists, and the Agrarians—each took about one-fifth of the district seats they contested. 6. We must emphasize that the simulation concerns the short term. The large number of independent deputies elected in districts may not continue as parties develop district-level organizations. 7. For a discussion of the leftist parties’ strategies in the 1993 election, see Davidheiser 1998. 8. Interestingly, if there were prior agreements among the opposition parties to avoid competing against one another in district races, they seem also to have included Liberal Democrats. In twenty-three districts, Communists and Agrarians ran against each other. In twenty-two, Liberal Democrats ran against one of the other two party’s candidates. In twelve, candidates nominated by all three parties were running. 9. The final version of the law made one concession to the powerful regional elites represented in the parliament’s upper house, the Council of the Federation, who strongly resisted an electoral law that favored Moscow-based, Russia-

ELECTORAL SYSTEM

115

wide political parties over regional and local interests. Although preserving the single, federal electoral district for tallying the total votes each party received on the PR ballot and determining its share of the 225 PR seats, the law required each party to create regional sub-lists to facilitate representation for regionally based politicians. A party could place no more than twelve candidates on the central (i.e., all-Russian) portion of its list. All other candidates received seats from that party’s list depending on the share of the party’s vote won in the given region for the sub-list on which they ran. The law allowed the parties complete freedom in devising their regional sub-lists: In theory, a party could create one or two regional sub-lists for all candidates other than those designated as its twelve “central” candidates. Thus this provision is only a minor modification over the previous system. The electoral law had an unusually complex legislative history. It was originally passed by the Duma in the first reading on 24 November 1994. At this reading the Duma rejected the president’s proposed system of 300 PR seats and 150 SMD seats and adopted instead the allotment of 225 seats for both. The Duma passed the bill in the second, substantive, reading on 15 March 1995. Two important votes from that stage are reported here in Tables 5.4–5.7. The bill then passed the third reading on 24 March and went to the Council of the Federation, which rejected it on 12 April. The Duma then passed it again, making only minimal revisions, on 21 April. The Council rejected it again on 4 May. The Duma then overrode the Council on 11 May and sent the bill to the president for his signature. The president, however, vetoed the bill on 23 May. The Duma, unable to win enough votes to override the president, formed a conciliation commission comprising members of the two chambers and representatives of the president. The commission reported back a version of the bill making only minor changes in the Duma’s version, and the Duma passed it on 9 June. The Council of the Federation rejected the bill again on 14 June but immediately reversed itself and passed the bill on 15 June. The president then signed it into law on 21 June. The law was published in Rossiiskaia gazeta on 28 June 1995. 10. The strong presidential powers granted Yeltsin under the constitution influenced deputies’ calculations. Deputies were aware that if parliament was unable to pass a new electoral law, the alternative was for the president to promulgate one by decree. As time passed, some deputies were more willing to compromise with Yeltsin under the pressure of this possibility; others were more inclined to support the Duma’s version of the law even if they disagreed with some of its provisions, so as to forestall a presidential decree. 11. The LDPR opposed the final version of the bill in the second reading by almost unanimously refusing to vote on it, but in later stages it strongly supported the 225:225 system over the alternatives urged by the president and the Council of the Federation.

SIX PART Y DISCIPLINE IN THE RUSSIAN DUMA

I

N PREVIOUS CHAPTERS, we reported that deputies’ mode of election and their policy preferences shaped their preferences about electoral law, the old Presidium, the power of parties and their leaders within the Duma, and the relationship between committees and parties. At times, more purely partisan considerations seemed to have an effect. These observations were broadly consistent with the argument made in chapter 1. No single political goal accounts for the choice of prominent features of Russian parliamentary institutions. When the implications of institutions are relevant to a goal and are fairly certain, the behavior of a significant number of deputies appears to reflect the importance of that goal. In this chapter our primary purpose is to contrast deputies’ strategies in structuring institutions with their roll-call voting behavior more generally. Specifically we consider the sources of variation in faction support on the floor of the Duma. If studies of the U.S. Congress are any measure, the electoral consequences of typical floor votes are quite indirect and uncertain for most legislators (Kingdon 1973). If studies of European parliaments are the gauge, party trumps other considerations most of the time. Thus the effects of mode of election might be substantially weaker in floor voting than in some of the institutional choices we have examined. Policy and faction effects might be stronger. A secondary purpose is to explore institutional effects, as opposed to the choice of institutions, in the parliamentary laboratory that Russia’s electoral system created. We seek evidence of differences between district-mandate and list-mandate members in the degree to which they exhibit partisan loyalty in floor voting. We then ask whether the differences are associated with voting-behavior patterns even when controlling for membership in partisan factions. If they are, then we are justified in concluding that incentives related to electoral mandate type outweigh the influence of policy-based motivations and intrafaction coordination in influencing the degree to which members agree with their factions in voting. If they are not, then we may infer that either homogeneity of preferences over policy or explicit measures by parties to reinforce faction loyalty—or both—better explain levels of cohesiveness in deputies’ floor behavior.

PA RT Y D IS C I PL I N E

117

Electoral, Policy, and Partisan Incentives in Parliamentary Voting Among the most studied and still controversial questions about legislative behavior is the role of electoral, policy, and partisan incentives. The debate continues in the literature on the American Congress. The mixed electoral system and the varied character of the new Russian parties provide variation in incentives for faction support that is seldom seen in established Western democracies. Our working hypothesis is that electoral influences on floor voting are limited in Russia’s early democracy. Particularly for deputies in singlemember districts, the effect of a vote on one’s chances of reelection is likely to be small. Information about individual deputies is not widely available to the electorate. In fact, during the 1994–95 Duma the official documents of the Duma were difficult to find except in a few locations in Moscow. Consequently the electoral consequences of voting and nonvoting would be difficult to calculate for the typical deputy, while the factional and policy consequences would be more direct. We would expect the latter to play a greater role in structuring the voting decisions of deputies. Electoral and Institutional Incentives Among the well-established propositions of political science is that institutional and electoral contexts strongly influence the way legislators’ goals of controlling government and gaining reelection translate into voting behavior. Broadly speaking, the literature has identified five classes of institutional and electoral influences on voting. 1. Characteristics of the party system may affect cohesiveness of voting by party members in parliament. Two such factors have been identified. First, the competitiveness of the party system, both in the electoral arena and in parliament, directly affects cohesiveness of party voting: The more competitive the party system, the higher the level of party loyalty and cohesion in voting in the legislature. The closer the margin between keeping and losing majority status, empirical studies in a number of settings have found, the greater the degree of discipline (Bowler, Farrell, and Katz 1999). The second factor is the degree of central control exercised by party headquarters over the behavior of individual members of the party faction in parliament. The centralization and organizational rules of a party, the pattern of its organizational development, may give its parliamentary faction greater independence of action or may instead hold individual members of the parliamentary faction closely in line.

118

CHAPTER SIX

Katz and Mair (1993) find that, over recent decades, while central party headquarters in European party systems continue to dominate the parties in absolute terms, parliamentary parties are acquiring greater staff resources over time, and, in all likelihood, greater influence within their parties. Generally scholars find that the greater the degree of centralization of power in the central party offices, the greater the level of discipline in voting. 2. A second set of influences concerns differences between the left and the right. Studies in many democratic legislatures have found that socialist and Communist parties exhibit higher levels of voting discipline than do their colleagues on the right. This has been shown, for example, in the relatively new setting of the European Parliament, where party groups are formed from national parliamentary factions and leftist groups generally exhibit higher levels of normative support for party discipline (Bowler and Farrell 1999, 218); and in comparative surveys of national parliaments, where the norm of party loyalty is more strongly expressed by parties of the left (Loewenberg and Mans 1988). 3. A third factor concerns the relationship of the parties in parliament to the government’s survival. The more that government portfolios, policy benefits, electoral support, and other valued goods depend on the support of a parliamentary party or coalition for a sitting government, the greater the incentive for members of the majority party to maintain voting discipline in order to preserve or obtain those benefits. The logic of parliamentary government rests on the salience of costs to individual members of parliament from elections that are separate from purely policy-based incentives, so that a government can blackmail its partisans in parliament to keep it in office on pain of dissolution and early elections (Huber 1996) even when its parliamentary co-partisans oppose it on strictly policy grounds. Indeed, historically the emergence of party government is credited with inducing disciplined parliamentary voting and coherent electoral competition among parties (Cox 1987). Much of the debate over the relative merits of presidential and parliamentary systems for stability-endangering confrontation between executive and legislative branches revolves around the nature and strength of incentives for partisan support of the executive in a presidential system (Shugart and Carey 1992; Mainwaring 1993). In other words, where the power of the executive does not depend on maintaining a parliamentary majority, parliamentary party members may express their opposition to their own government’s policies with impunity. 4. A fourth set of influences is predicated on the assumption that the voting behavior of members of parliament is learned; as the legislature grows in organizational efficiency and capacity, members are socialized

PA RT Y D IS C I PL I N E

119

to accept role norms conducive to the effective functioning of the legislative institution (Agh 1999). One problem with this theory is that members appear to determine their own behavioral patterns without requiring much tutelage as they assess the relative importance of the institutional and electoral incentives they face. For example, Montgomery (1995) finds that newly elected deputies in a parliament using a mixed electoral system quickly oriented their representative focus according to their electoral mandate. Thus differing levels of party loyalty across countries and parties may reflect calculation more than learning. 5. The fifth set of considerations is drawn from the effects of the electoral system. For the most part, deputies elected in PR systems are more likely to exhibit cohesive voting than are their counterparts elected in single-member-district election systems. Party-list candidates and legislators must satisfy their party and its leaders to gain and maintain placement on the list; SMD candidates may have to gain party support for nomination, but ultimately they must appeal to a geographically defined constituency. Many European systems combine parliamentary and partylist systems and produce powerful incentives for strong, cohesive parties, whereas many American systems combine presidential and single-member-district systems to produce loosely organized, incohesive parties (see Epstein 1980; Rae 1971; Sartori 1976; Taagepera and Shugart 1989; Shugart and Carey 1992).1 The relative importance of these factors is difficult to sort out. Few studies weigh them using multivariate analysis across countries or within particular systems; most authors treat party discipline as part of a systemic package of electoral and institutional features. For that reason the Russian case offers us unique advantages for testing the importance of these different factors in a single country where other influences are constant. We begin by considering the basic difference between SMD and partylist members. The proposition about the influence of electoral constraints on parliamentary behavior generated by this literature is straightforward: Party-list systems generate greater discipline in parliamentary parties than do SMD systems. The reasonable, though untested, proposition for the mixed system of Russia is stated as follows: Proposition 6.1. Party-list deputies will be more supportive of faction positions than will SMD deputies. Unfortunately the existing literature on electoral systems and their consequences for party systems and parliaments gives little attention to mixed-mandate systems (Lijphart 1990, 1994; Lijphart and Grofman 1984; Bawn 1993; Remington and Smith 1996). Only in Germany is there a mixed electoral system old enough to have generated a literature on

120

CHAPTER SIX

mandate types, but it has not done so. Research on the German parliament links mandate type with attention to constituency service and other role orientations but not to behavior. Aberbach, Putnam, and Rockman (1981) found that there were some differences in role orientations among German Bundestag members associated with the difference between list and district mandate types. Another study (Lancaster and Patterson 1990) found that orientations toward the provision of distributive benefits for constituents varied systematically between PL and SMD members.2 Differences between PL and SMD members are given little or no attention in detailed accounts of Bundestag politics. Moreover, literature on the German parliament has treated mandate type as ancillary to other factors affecting legislative behavior, notably party and ideology (Patzelt 1993, 1994; Herzog et al. 1990; von Beyme 1986). German deputies of both mandate types are considered to combine territorial with partisan representative orientations.3 Beyond the effects of mode of election, deputies’ relationships with their parliamentary parties may vary according to their personal political circumstances. Students of SMD systems have concentrated on the effect of electoral marginality and constituency preferences on the behavior of legislators (Deckard 1976; Fiorina 1974; Sullivan and Uslaner 1978). The argument is that the greater the threat to reelection, the more voting behavior is shaped by electoral strategies and the less it is influenced by personal views and party leaders. Constituency influence also is a natural subject for students of representative government and is usually considered in juxtaposition to personal ideology and party influences. Evidence for the marginality thesis and constituency influence has been mixed. Party-list deputies vary in “safety” by virtue of their place on the list. Some are close to the cut-off point for membership in the parliament. Others are guaranteed election so long as their party cleared the 5 percent threshold. Unfortunately the causal connection between list placement and party loyalty is unclear. While individuals expected to be loyal to party leaders might be rewarded with high rankings, individuals seeking higher rankings might be expected to seek to prove their fealty to leaders with loyal backing while in parliament. Existing literature offers little help in choosing between these alternatives. About the best that can be predicted is that party-list deputies are loyal to leaders responsible for their list placement regardless of the personal placement. We do not fully explore the possible calculations of deputies, but we address the more obvious possibilities in the following propositions: Proposition 6.2. Among SMD deputies, electoral marginality is negatively related to support for parliamentary faction positions.

PA RT Y D IS C I PL I N E

121

Proposition 6.3. Among party-list deputies, ranking on a party list is positively related to support for parliamentary faction positions. Policy Incentives Parties in the 1994–95 Duma were embryonic. In the December 1993 elections, party preferences among voters were weak, most party organizations were rudimentary, and candidates running in some districts sometimes avoided affiliation with a national party (Colton and Hough 1998; White, Rose, and McAllister 1996; Sakwa 1995; McFaul and Petrov 1995; McAllister and White 1995). Most observers, and surely most deputies, expected electoral parties to continue to emerge, reform, and even disintegrate over the next few years. The relationship between electoral parties and parliamentary factions was quite variable and often weak. Deputies who belonged to a party’s parliamentary faction were not necessarily members of that party. Members of parliamentary factions showed a considerable amount of fickleness in their affiliations. In fact, many deputies changed their faction memberships according to their judgments about the political compatibility and attractiveness of a faction. During the course of that two-year Duma, nearly one-quarter of the members changed factions. Some changed affiliations more than once. Furthermore Russia’s president-parliamentary type of constitution (Shugart and Carey 1992, 26) offers relatively weak incentives either for faction cohesiveness or for the formation of a majority coalition. The president is popularly elected and serves a fixed term, so the presidential selection is not dependent on parliamentary majorities. And the president wields decree-making power that allows him to govern as he sees fit as long as decrees are consistent with the constitution and established law (Remington and Smith 1998). The Russian constitution provides for a government headed by a prime minister, who is subject to confirmation by parliament, and includes a no-confidence-vote procedure by parliament in the government. However, the constitution gives the president the option to dissolve parliament and force new parliamentary elections upon parliament’s refusal to confirm the president’s nominee for prime minister or the adoption of a no-confidence resolution. Threats to control over the government that bond partisans together in many parliamentary systems have not done so in Russia. Consequently it is quite reasonable to suppose that the personal policy preferences were important influences on deputies’ everyday voting decisions, perhaps even for party-list deputies. The empirical proposition is the following:

122

CHAPTER SIX

Proposition 6.4. Deputies’ support for parliamentary faction positions is correlated with their personal policy preferences. Partisan Incentives Nonetheless membership in a faction was valuable because the Duma’s rules gave registered factions certain collective privileges that benefited members. In addition to voting membership on the Council of the Duma, the Duma’s rules gave faction leaders and leaders of registered deputy groups other valuable privileges as well, including office space, secretarial assistance, and other material perquisites, as well as priority rights to recognition in floor debates. Thus a deputy has considerable incentive to join either a faction or a registered deputy group. By the same token, the rules tend to encourage groups to behave as if they were party factions. Nevertheless parliamentary factions varied from groups of temporary convenience with no external counterpart to genuine factions with considerable organizational structure throughout Russia. On the weak end of the spectrum were several informal parliamentary groups by SMD deputies who lacked any meaningful attachment to an electoral party during the 1993 campaign. Reacting to the dominance of the Duma’s leadership and organization by the parties who had fielded lists and quickly formed factions, many of the single-member-district deputies formed their own deputy groups. Interviewed members of such groups cited several goals in doing so. All expressed the view that the parties were taking over the Duma at the expense of the ability of district deputies to represent the constituency and legislative interests of their voters. They also sought to claim representation for themselves on the Council of the Duma. In the spring of 1995, after the splintering of the December 12 and NRP factions, two more deputy groups formed—Russia and Stability—and attracted a sufficient number of members to win registration. The history of these factions suggests that political independence and district mandates were essential to the formation of these groups but that many of their members sought politically compatible and electorally viable partisan associations in order to run for reelection. Plainly deputies’ relationships to faction and party varied widely. For the SMD deputies of the nonparty groups, “party loyalty” is nothing but group loyalty. With little or no connection to electoral parties, such deputies are expected to exhibit less loyalty to their parliamentary faction than deputies whose factions are associated with parties that can generate incentives for supporting the positions of central leaders. We have suggested that for parliamentary factions with external ties to electoral parties, stronger parties may be associated with higher party

PA RT Y D IS C I PL I N E

123

cohesiveness within parliament. The CPRF, Agrarians, LDPR, and Yabloko had extensive extraparliamentary assets. The Communists had inherited much of the CPSU’s organizational network, including, for better or worse, name recognition.4 The Agrarian Party represented the amalgam of three complementary organizational structures: a party, an occupational association, and a trade union of agricultural workers. The LDPR had a leader familiar to the entire country (Zhirinovsky had come in third in the Russian presidential election in 1991) as well as a highly disciplined organization; his party elected Zhirinovsky “leader for life” at its party congress in spring 1994. And Yabloko offered a well-known leader who could claim to be the only serious democratically oriented alternative to the current government.5 These observations suggest this proposition: Proposition 6.5. The electoral strength of party organizations and leaders associated with parliamentary factions is positively related to the party discipline exhibited by faction members.

Data We pursue two measures of floor-voting behavior—a measure of deputies’ general policy position and a measure of faction support. The first is the principal component of a factor analysis of all floor votes in the 1994–95 Duma (Remington and Smith 1998). We label this variable “policy position.” The second is a measure of support for faction or party positions similar to party unity scores that are calculated for members of the U.S. Congress. In that case we narrow the analysis to votes that divide the parties and seek to determine whether differences between factions are more important than differences within factions in support for faction positions. In both cases we estimate the effects of policy preferences, mode of election, and faction or party membership. We label this variable “faction support.” Methods for identifying party votes and measuring faction support are discussed in the appendix. The independent variables are faction membership, mode of election, policy preferences, electoral marginality, and party-list placement. Faction membership is based on the faction membership at the time the term expired in December 1995.6 Mode of election is the means by which a deputy gained a seat in the Duma, as in other chapters. To capture policy preferences, we use the responses of 162 deputies in a 1995 survey to a battery of twenty questions about contemporary policy issues. We use the principal component to scale deputies (Remington and Smith 1998). We label this variable “policy preference.”

124

CHAPTER SIX

For SMD deputies, marginality is measured as the incumbent’s margin of victory over the second-place candidate. PL deputies cannot be given a single PL ranking because they could be listed on national or regional lists. Rather than devising separate variables for national and regional lists or combining lists in some arbitrary manner, we use a single dummy variable to indicate whether a deputy was in the top twelve spots on the national list. If a party’s vote exceeded the 5 percent threshold for party representative in the Duma, it was guaranteed a minimum of twelve seats, so differences among the top twelve were not meaningful.

Findings The key variables of the empirical propositions—mode of election, electoral marginality and list placement, policy preference, and faction— are so strongly correlated with one another that inferences about their independent effects are difficult to draw. To see this, and to aid in the interpretation of multivariate estimates reported in the next section, we begin with an overview of the electoral foundations, policy locations, and cohesiveness of the parliamentary factions. Four features of parliamentary factions in 1994–95 deserve special notice. First, on the whole, factions behaved in remarkably disciplined ways on the floor. In Figure 6.1 we array the factions on the political spectrum from the left to the right, according to their mean policy positions. The shaded boxes contain the 50 percent of values falling between the 25th and 75th percentiles, and the lines that extend from the box to the highest and lowest values, excluding extreme outliers. The bar within the box indicates the median for the faction. Plainly, with important exceptions, the factions show great internal cohesiveness and inter-faction polarization. There is no overlap between the large factions on the left and those on the right. This pattern means that separating policy and party effects will be difficult. Second, the important exceptions are all factions not associated with an electoral party—New Regional Policy, Russia, and Stability. These factions show moderate mean policy positions and large variances. We would expect weak “party” effects for members of these factions. Third, factions differ widely in their numbers of SMD and party-list deputies. Figure 6.2 illustrates the policy positions of the two types of deputies for each faction. The three nonparty factions occupying the middle of the spectrum are comprised primarily of SMD deputies. The memberships of the DPR and LDPR are almost entirely from their party lists. These patterns complicate the statistical problem of separating the

PA RT Y D IS C I PL I N E

125

Figure 6.1. Policy Positions of Factions, 1994–1995 Source: Roll-call record of the Duma.

Figure 6.2. Policy Positions of Factions, by Mode of Election, 1994–1995 Key: SMD Deputies = white bars; PL Deputies = black bars. Source: Roll-call record of the Duma.

effect of deputies’ mode of election from the effect of party on party support. Fourth, two of our empirical propositions receive support from an inspection of Figure 6.3, which reports party support for the factions arranged by policy position, as in the previous two figures. Levels of “party” support are substantially lower for the three nonparty factions than for the other factions. The Communists and LDPR show the greatest levels of faction support, consistent with the proposition that external party strength facilitates intraparliamentary party cohesiveness.

126

CHAPTER SIX

Figure 6.3. Party Support within Factions, 1994–1995 Key: SMD Deputies = white bars; PL Deputies = black bars. Source: See appendix.

And fifth, the expectation that SMD deputies are less supportive of their factions than are party-list deputies is not clearly confirmed in Figure 6.3. SMD deputies were somewhat less supportive for several factions—the LDPR, Women of Russia, Russia, PRES, and Yabloko. But for larger factions the results are unclear. For the CPRF, SMD deputies show slightly greater support than party-list deputies. For Russia’s Choice, the difference is minimal. Policy Positions In Table 6.1, we report ordinary least squares (OLS) estimates of the effects of policy preferences as measured by our survey, faction, and within-faction mode of election on policy positions, as measured by voting behavior. The model is estimated only for those respondents in the survey who provided their names and whose preference scores can be compared with their voting records. These two measures—survey-based policy preferences and voting-based measures of policy positions—are strongly correlated (at .82), so the deck is stacked against finding strong independent effects for faction and mode of election. Nevertheless, as the estimates in column 1 indicate, faction membership is significant. The missing faction category is the CPRF, which means that the faction coefficients measure difference from the communists. Mode of election is statistically significant for only small factions—PRES and Russia.7 Because they lacked any connection to an electoral party, the NRP, Stability, and Russia factions might best be excluded from a “party” effects model. The estimates for an alternative specification are provided in column 2. These estimates are calculated by excluding members of the three

PA RT Y D IS C I PL I N E

127

TABLE 6.1 OLS Estimates of the Effects of Policy Preferences, Faction, and Mode of Election on Policy Positions, 1994–1995 Independent Variables

1

2

Constant −1.07* 1.47* Policy preferences .12* −.01 Faction Dummy Variables APR −.20 −.31* LDPR −.18 −.34 DPR −1.23* −1.44* WR † † RC −2.39* −2.73* PRES † † Yabloko −2.17* −2.48* NRP −.93* excluded STB −1.94* excluded ROS −1.30* excluded Faction-Mode of Election Interaction Terms APR −.03 −.07 LDPR −.40 −.32 CPRF −.15 −.17* WR −.71 −1.00* RC −.12 −.12 PRES −2.30* −2.57* Yabloko −.09 −.03 STB † excluded ROS −1.07* excluded 2 Adjusted R .91 .97 N 128 101 p < .05 † Dropped owing to multicollinearity.

*

“nonparty” factions and the associated variables. These estimates show no effect of policy preferences even though the simple correlation between preferences and positions is .83 for the remaining deputies. In a multiparty parliament such as the Duma, members typically exhibit higher intraparty homogeneity in policy preferences than would be the case where there are fewer parties. In fact we find that members of the same factions tend to share similar policy preferences, even when we use an exogenous measure of policy outlooks, one not affected by the choice and framing of issues on which deputies cast votes. That we nonetheless detect some faction-specific effects on voting behavior, but no within-faction effects for different electoral mandate types, is therefore significant. It provides empirical confirmation of our supposition that factions enable members to coordinate their voting positions in

128

CHAPTER SIX

ways that accommodate their different electoral and policy interests. To be sure, there is probably considerable measurement error in both our measures of policy preferences (as opposed to the electoral mandate variable, which is captured precisely), making it difficult to be precise about the relative weight of pure policy preference or factional membership as influences on voting. However, it is notable that mode of election has only a limited relationship to within-faction variation in levels of voting cohesion. Faction Support Bivariate relationships between faction support and the independent variables provide some support for the propositions. Excluding deputies who were not affiliated with a faction, faction support was correlated at .40 (p = .000) with mode of election (party list = 1; SMD = 0), indicating that party-list deputies had higher faction support scores. For party-list deputies who were faction members, party support is slightly higher for deputies who were placed high on the party lists—correlations of .13 (p = .008) for placement on the 1993 lists and .08 (p = .054) for placement on the 1995 lists. For SMD deputies who were faction members, however, faction support and margin of victory in 1993 were correlated at -.15 (p = .02)—more marginal SMD deputies were marginally more supportive of their factions, which is inconsistent with our expectation. Consistent with our observation about the high levels of faction cohesiveness in the leftwing factions, faction support was correlated with policy preferences at .24 (p = .005) and policy positions at .25 (p = .000). Unfortunately the bivariate relationships may be quite misleading because of the unequal distribution of deputy types across the factions and the collinearity among the variables. In Table 6.2 we report a series of estimates of the effects of the electoral, policy, and party variables on faction support. The set of dummy variables for the factions excludes the CPRF, which is the most cohesive faction, so their coefficients can be taken as a measure of differences between each faction and the CPRF. The other sets of variables are interaction terms between faction, on the one hand, and mode of election and policy preference (or position), on the other hand. The coefficients for the interaction terms gauge the within-faction effects of mode of election and policy on faction support. The modest relationships between electoral marginality, list placement, and faction support do not survive in the multivariate analysis and so the corresponding variables are not reported in the table. Estimates for both policy preferences, the preferred measure, and policy positions are included so that we can see estimates with the larger N associated with the policy positions variable.

PA RT Y D IS C I PL I N E

129

The following major inferences may be drawn from Table 6.2: • The special cohesiveness of the CPRF faction is reconfirmed by the estimates in Equation 1. Every other faction shows significantly lower levels of faction support. • Mode of election has the predicted effect for all but two factions, although the effect is very small (Equation 3) and is overwhelmed by interfaction differences when they are taken into account (Equation 2). • Policy preferences are weakly related to faction support (Equation 4). For the large left-wing factions (CPRF, Agrarians, LDPR), the more leftleaning deputies exhibit greater faction support; for the large right-wing faction (Russia’s Choice) and the large moderate faction (NRP), the more right-leaning deputies exhibit greater faction support. The very strong correlations between faction and policy preferences makes it difficult for either set of variables to retain statistical significance when they are both in the equation (Equation 5). Note that the N is smaller for the equations with policy preferences because they are based on a sample survey rather than the full membership. • Substituting policy positions for preferences allows a larger N, increases the statistical significance of a coefficient of a given size, and alters the inferences for policy effects somewhat (Equations 6–8). The within-faction policy effects on faction support are stronger and more frequently retain their significance when controlling for faction. • Allowing for the possibility that mode of election and policy positions determined faction membership, the effects of mode of election and policy positions are estimated without controlling for faction (Equation 8). Mode of election shows a moderately strong relationship, in the predicted direction, to faction support within the factions of the left but not for the factions of the right. Policy position retains the bifurcated pattern for the largest factions—faction support is strongest from the left within the left-wing factions and strongest from the right for the right-wing factions. Overall the equation does not predict party support quite as well as equations that include faction. The statistical evidence indicates that faction, policy preference/position, and mode of election have significant effects, in descending order of importance, on levels of faction support observed among deputies of the Duma. Interfaction differences outweigh policy differences among deputies in explaining deputies’ support for their own factions’ policy positions. These findings are consistent with the analysis of general policy positions. As for general policy positions, the inclusion of deputies from the nonparty factions may distort our picture of faction effects on faction support among deputies of factions associated with electoral parties. To check this possibility, we reestimate several of the equations by excluding the

130

CHAPTER SIX

TABLE 6.2 OLS Estimates of the Effects of Policy Preferences, Mode of Election, and Faction on Faction Support, 1994–1995 1

2

3

4

5

6

7

8

Constant .91* .92* .76* .78* .88* .72* .90* Faction Dummy Variables1 APR −.10* −.11* −.08* −.12* * LDPR −.03 −.04 −.01 −.06* * * DPR −.11 −.12 −.07 −.11* WR −.03* −.06 −.06 −.05 RC −.04* −.05* −.03 −.14 PRES −.16* −.19* −.11 −.21* Yabloko −.09* −.12* −.05 −.17* NRP −.25* −.26* −.23* −.24* STB −.20* −.20* −.19* −.19* * * * ROS −.21 −.22 −.18 −.20* 2 Faction Interaction with Mode of Election APR −.002 .05* .03* LDPR .004 .12* .08* * CPRF −.02 .14 .05* * WR .01 .11 .10* RC .004 .11* −.01 PRES .03 −.001* .01 Yabloko .03 −.07* .01 STB −.02 −.07* −.02 ROS .01 .14* −.01 Faction Interaction with Policy Preferences (4 and 5) and Policy Positions (6,7,8) APR .05* .03 .08* .03* .07* LDPR .05* .01 .17* .04* .10* * .03 .19* DPR −.14 † .17 * * CPRF .09 .02 .13 .01 .12* WR −.05 † −.22* .04 .12* RC −.07* −.01 −.11* −.09 −.12* PRES .01 † −.03* −.07* −.03 * Yabloko −.04 .01 −.10* .10* −.11* NRP −.12* −.02 −.04 .01 −.03 STB .03 −.02 −.01 −.01 −.01 ROS .08 .02 .02 .01 .01 N 366 366 366 117 117 366 366 366 Adj. R2 .75 .74 .34 .37 .75 .62 .76 .66 Source: Roll-call record of the Duma. * p < .05 † Dropped owing to multicollinearity. 1 The CPRF is the dropped category. 2 The DPR and NRP are not included; the DPR had no SMD deputies, the NRP had no PL deputies.

PA RT Y D IS C I PL I N E

131

TABLE 6.3 OLS Estimates of the Effects of Policy Preferences, Faction, and Mode of Election on Faction Support, 1994–1995, Excluding Nonparty Factions Constant .91* Faction Dummy Variables APR −.14* LDPR −.08* DPR −.12* WR −.10 RC −.16* PRES −.23* Yabloko −.19* Faction Interaction with Mode of Election APR .01 LDPR .01 CPRF −.02 WR .03 RC −.001 PRES −.002 Yabloko .001 Faction Interaction with Policy Position APR .03* DPR .05* LDPR .03 CPRF .002 WR .05 RC −.09 PRES −.07 Yabloko −.09* Adjusted R2 .51 N 262 *

p < .05.

nonparty faction deputies and associated variables. The results are reported in Table 6.3. The same general pattern is apparent—large faction effects, small and bifurcated policy effects, and little or no mode-of-election effects.

Problems of Interpretation Recognizing the fundamental importance of a competitive party system to the development of a democratic polity in Russia, the framers of the 1993 constitution and electoral law preserved the traditional candidate-

132

CHAPTER SIX

centered, single-member-district electoral system with which Soviet voters had long experience. To this territorially oriented, personalized representative nexus they joined a proportional component with the largest possible district magnitude—a single nationwide district. Given the disparity between the two types of mandate, we expected to find sizable differences in the voting behavior in the Duma between deputies of different electoral categories. We have not found major within-faction differences in party support between PL and SMD deputies. The relationship between mode of election and party support is more complicated. Substantial differences among factions in level of average loyalty or cohesiveness were related to the overall mix of PL and SMD deputies within them. This suggests that both electoral and party forces influenced deputies’ voting behavior in 1994–95. In each faction, deputies behaved as if they had balanced electoral and party goals, and coordinated on a level of cohesiveness acceptable to both categories of members. Many SMD deputies chose not to affiliate with electoral parties or to join party-connected factions once in the Duma. These deputies formed their own parliamentary factions and even showed substantial support for the majority positions of their faction colleagues. However, they did dispose of their factions and create new ones when it suited them, and they generally showed lower levels of support for faction positions than did their colleagues in the party-connected factions. We conjectured that if members of Russia’s parliamentary factions differed systematically between SMD and PL categories in the degree of voting cohesion regardless of partisan affiliation, it meant that electorally oriented pressures tended to outweigh partisan influences on deputies’ calculations about their voting stances. We have not found consistent differences between list and district members regardless of party affiliation, so we are confident in positing that party-based influences are present. To be sure, we cannot rule out the possibility that group effects are a simple reflection of the degree to which members of the same party factions are homogeneous in their preferences over policy. Given the evidence from other observations that factions do meet regularly to coordinate their voting positions on particular legislative issues, we would surmise that faction organizations do enjoy some success in keeping deviation from the faction line within accepted boundaries for both list and district members. It is clear that the relationship between faction membership and policy preferences is a complicated one. Policy differences no doubt play an important role in politicians’ decisions to associate with one party or another, and influence the choice of faction among incumbent deputies. The entire structure of electoral parties and factions is shaped by

PA RT Y D IS C I PL I N E

133

the policy preferences. Still, the weak relationship between policy preferences and party support within factions indicates an independent faction effect on deputies’ behavior. The small number of parliamentary factions limits our ability to model more comprehensively the possible influences on differences across factions in their ability to stimulate discipline among members. For example, we noted above that party factions differ in their access to organizational resources useful in electoral campaigns: a prominent leader, a national organization, and a popular policy appeal. We might imagine that factions associated with electorally attractive parties would do a better job of maintaining their members’ loyalties. The small number of cases (and lack of reliable evidence on electoral-parliamentary party connections) makes systematic analysis of this possibility difficult at this time. All factions with potentially strong external ties—Russia’s Choice, CPRF, LDPR, Yabloko, and the Agrarians—are also distant from the center and were relatively successful in the party-list voting in 1993.

Conclusion The weak effects of mode of election and strong effects of faction and policy preferences that we have observed are consistent with the general theory outlined in chapter 1. In Russia’s early democracy, the floor-voting record of deputies is not likely to have great electoral significance, which leaves room for factional and personal policy preferences to be influential in the voting decision. Unlike basic institutional choices such as the Duma’s party structure, its agenda-setting mechanisms, and electoral law, floor voting in 1994–95 did not appear to be greatly influenced by deputies’ electoral context. We must be careful not to overstate the case against electoral influences on floor voting. In the first place, election outcomes, determined by local voters in the case of SMD deputies, determined who was casting the votes. To a large degree, then, constituency preferences influenced the identity and preferences of the deputies and the nature of the factions they joined. Moreover, we have not directly measured constituency preferences and so cannot contrast their influence on voting decisions with the influences of personal policy preferences and faction. What we can say is limited to deputies’ mode of election. We simply have not found strong effects that the cross-national comparisons of parliaments suggest we should. Empirical theorists propose that the competition for power by selfinterested political parties can supply public goods to young democracies, such as the stabilization of initial constitutional arrangements and

134

CHAPTER SIX

the provision of electoral accountability of rulers to ruled (Katz 1986; Lawson 1993; Mainwaring and Scully 1995; Aldrich 1995; Powell 1989). In the absence of an agreed-on external agency to enforce commitment to constitutional rules of engagement, mutual rivalry creates incentives for parties to monitor the behavior of others and pay the cost of enforcing adherence to democratic procedures. Under these conditions, party competition yields increasing returns for a young democracy. The literature also suggests that the system-level conditions generated by a competitive party system entail trade-offs: Clarity and efficiency of electoral choice cannot be maximized simultaneously with representativeness and inclusiveness (Shugart and Carey 1992; Powell 1989). The degree to which a particular party system tends to yield one set of effects rather than another is influenced by institutional arrangements, particularly constitutional rules regarding parliamentary-executive relations, parliamentary rules governing the balance of rights between majority and opposition parties, and the character of electoral mandates held by deputies. The evidence presented here suggests that Russia’s parliamentary parties of 1994–95 had developed a capacity to coordinate the floor behavior of their members even if they were not yet national electoral organizations with articulated structures and established electoral constituencies. The deputies’ levels of cohesiveness on votes over which parties differ strike us as relatively high, given the absence of any party competition before 1993. Faction leaders, after all, possess very weak instruments for maintaining party discipline.8 As Kiewiet and McCubbins (1991, 44) comment, we might better consider what party leaders can do to influence their members’ behavior than what they cannot do. To what degree party factions in the Duma reflected the preferences of their voters in adopting their positions on legislative issues is a question well beyond the scope of this book. But they presented the voters with reasonably coherent alternative policy packages embodied in the behavior of their deputies in parliament.

NOTES

1. In addition, as Epstein (1980) and others have noted, some party-specific features of the relationship between external and legislative parties will influence the behavior of parliamentarians. For a review of the literature on parliamentary parties and voting behavior, see Brady and Bullock 1985 and Collie 1985. Strong intraparty cohesiveness has not been a mystery in parliamentary and party-list systems, so students of these systems give it little attention and instead focus on

PA RT Y D IS C I PL I N E

135

the social cleavages underlying the party system and the composition of government coalitions (Lipset and Rokkan 1967; Sartori 1976; Cox 1987; Suleiman 1986). Studies of presidential and SMD systems include many on party cohesiveness and individual party loyalty. In the case of the U.S. Congress, the work of Brady and his colleagues (Cooper and Brady 1981; Brady, Brody, and Epstein 1989) has demonstrated how changes in the heterogeneity of party constituencies alters the cohesiveness of the legislative parties. Party loyalty is a by-product of the correspondence between a legislator’s own constituency and the constituencies of party colleagues. Theories of “conditional party government” in the United States emphasize the processes of selection and self-selection through which candidates for public office affiliate themselves with ideologically differentiated party organizations (Aldrich 1994; Rohde 1991) and their consequences for policy goal-maximizing behavior in Congress (Kiewiet and McCubbins 1991; Cox and McCubbins 1993). 2. Lancaster (1986) argued that members of parliaments in which there is a clearly defined territorial link between mandate and constituency have a stronger incentive to provide distributive benefits to voters than have members in multimember districts. 3. For background on the German system, see Saalfeld (1990). As Schu¨ttemeyer (1994) explains, the theory of a Fraktionenparlament—a parliament dominated by party groups—still holds sway. Also see Montgomery (1995) on the postcommunist Hungarian system. Montgomery found that mandate type was strongly associated with the ways parliamentarians handled their representative responsibilities, affecting the way they made use of staff and office resources, allocated time, introduced inquiries and proposed agenda items on the floor, and perceived relevant constituencies. All varied systematically by electoral category. 4. The CPRF formed as a reaction to Mikhail Gorbachev’s liberalizing reforms, and continued as a force of opposition to President Yeltsin’s policies of market reform and openness to the West. Although the party vehemently denounced the strong presidential powers Yeltsin demanded in 1990–93, it made a critical decision in the fall of 1993 to enter the parliamentary arena under the constitutional and electoral rules decreed into effect by Yeltsin despite Yeltsin’s forcible suppression of the radical opposition in October 1993. The party was rewarded with a respectable share of the list vote and a substantial share of influence in the Duma. 5. These assets stood the parties in good stead in the 1995 election. The Communists proved by far the most successful in the voting, winning 23 percent of the list vote; Zhirinovsky’s LDPR came in second, with 11 percent; Yabloko won roughly the same share of the vote in 1995 (7 percent) as in 1993 (7.9 percent). Only the Agrarians fell below the 5 percent threshold for representation, winning only 3.8 percent. However, this share was more than that won by all but nine of the forty-three parties running. We therefore expect that because of the resources the external organizations of these four parties possessed, members of their corresponding parliamentary factions should vote more cohesively than members of other factions and groups. 6. We have also examined the support for faction leaders’ positions as an indicator of party support. Identifying leaders is problematic, and nonvoting by lead-

136

CHAPTER SIX

ers makes the calculation of leadership support scores difficult for some factions. Overall levels of leadership support are lower than our reported measure of party support, although the factions’ means rank similarly. And, in general, the estimates for party support scores, as reported in the text, and leadership support scores are in the same direction but are weaker for the leadership support scores. The results suggest that there is greater measurement error for the leadership support scores than for the party support scores reported in the text. For background on leadership support scores, see Cox and McCubbins 1993 and Kiewiet and McCubbins 1991. 7. The NRP is excluded from the set of interaction terms because it had no party-list members. 8. The Duma itself deprived faction leaders of the most drastic sanction—that expulsion from a party faction would entail loss of the parliamentary mandate— when they rejected the “imperative mandate” rule.

SEVEN INSTITUTIONAL CHOICE

I

N CHAPTER 1 we argued that legislators are motivated by policy, electoral, and partisan goals and pursue those goals opportunistically. Legislators do not find all their goals equally relevant to each of the policy or institutional choices they confront. Variation in goal relevance, transaction costs, and information across contexts leads legislators with complex goal profiles to apply goals opportunistically. A single goal may often dominate other goals in the context of a choice about a particular feature of parliamentary institutions. In this way legislators are locally, not globally, rational. The interactions among institutions and strategies usually become more complex as the consequences of interests become more remote, temporally or institutionally, and lead legislators to emphasize the immediate consequences of the institutional choices they face. Moreover, because goal relevance, transaction costs, and information can vary over time, the preferences of legislators about an institutional feature can change, sometimes rapidly. Our observations of the choices made by legislators in the new Duma during the 1994–95 period are consistent with this theoretical account. In this chapter we summarize those observations. Then we extend our coverage to more recent developments in the institutional features of the Duma and note the continuing influence of multiple goals on the institutional choices of the deputies. We conclude with a discussion of the implications of our study for theories of legislative institutions.

Summary of Findings Our consideration of institutional choice by Russian legislators has focused on the issues in dispute. It is worth noting again that several features of the Duma were inherited from the Union and Russian Supreme Soviets with no discussion. The absence of controversy about these features of the Duma reflects the participants’ views that goal achievement was not substantially undermined by the inherited institutional forms. It also prevents us from testing competing theories about individual-level foundations for choices about those institutional features. Two examples of carryover features illustrate our point. One such feature was a system of standing committees, along with a set of chairs and

138

CHAPTER SEVEN

deputy chairs for each committee. The planning commission of the fall of 1993 and the new Duma never questioned the need for standing committees. To be sure, how committee posts were allocated was a controversial matter, as was the relationship between committees and other key features of the new parliament. But the desirability of committees went uncontested. Similarly the way the Duma approves its daily agenda, considers amendments, allocates time for debate, and votes were inherited from the predecessor Supreme Soviets (for further details, see Remington forthcoming). The Russian Supreme Soviet had adapted floor procedure to a multifaction system, and the new Duma borrowed the process with no serious discussion. This means that we could examine the fault lines among deputies that surfaced over the use of these features by partisans of various kinds, but we could not directly test competing perspectives about the origin of these features. Because the contemporary use and original intentions of an institutional feature may differ, we must be cautious in drawing inferences about the origin of the system from the 1994 decisions about how to fill committee positions. The major controversial features of the Duma’s institutional arrangements have been our subjects. We summarize the effects of differences in policy preferences, mode of election, and faction affiliation on preferences about institutional arrangements in Table 7.1. The table reports our results for both survey responses about institutional arrangements and floor votes on proposals. A check mark (✓) indicates a strong relationship in multivariate estimates. An asterisk (*) in the column for faction effects indicates that just a few factions showed distinctive institutional preferences, controlling for policy preferences and mode of election. Sometimes these faction effects are large, but they usually involve only one or two distinctive factions. Plainly no single goal or interest—no single, consistent logic—underpinned the essential institutional features of the Duma chosen by deputies in the chamber’s formative period. If we had begun with a single-goal theory of institutional choice and then went searching for corroborating evidence, we could have uncovered considerable circumstantial evidence for our account of institutional choice, whichever of the three goals we had chosen. But the record of the early experience in the new Duma indicates that sometimes one, sometimes another, and sometimes more than one goal determines attitudes and behavior about the institutional options posed to deputies. Policy differences—the pro-reform, anti-reform dimension that has been the centerpiece of Russian politics in the 1990s—proved critical in shaping views of Duma’s faction system. In comparison with opponents of reform, reformers were significantly less supportive of the old centralized Presidium model for an executive committee, less supportive of a

*

Note: Strong relationship: ✓; modest relationship: .

Party-Faction System Faction threshold, January 1994 (Vote) Faction threshold, March 1994 (Vote) Imperative mandate, February 1994 (Vote) Influence of factions, deputies’ attitudes, January 1995 (Survey) Influence of faction leaders, deputies’ attitudes, January 1995 (Survey) Party role, deputies’ attitudes, January 1995 (Survey) Candidates’ attitudes, November 1995 (Survey) Council of the Duma, Chair of Duma Presidium, candidates’ attitudes, November 1993 (Survey) Voting rights on the Council of the Duma, March 1994 (Vote) Influence of the chair, deputies’ attitudes, January 1995 (Survey) Influence of the Council, deputies’ attitudes, January 1995 (Survey) Committee System Majority control of committees, candidates’ attitudes, January 1993 (Survey) Committee composition, January 1994 (Vote) Election of committee chairs and chamber vice chairs, January 1994 (Vote) Faction control of committee memberships, deputies’ attitudes, January 1995 (Survey) Electoral System Mix of SMD and PL Seats, June 1995 (Vote)

Institutional Preferences or Evaluations (data source)

✓

✓ ✓

✓

✓

✓

✓ ✓

✓ ✓

✓ ✓

✓

✓

✓ ✓ ✓

*

*

✓

*

*

✓

✓

*

*

*

*

Policy Preference Mode of Election Faction Membership

TABLE 7.1 Relationships between Deputies’ Goals and Attitudes about Institutional Features of the Duma, 1994–1995 Duma

140

CHAPTER SEVEN

high threshold for faction registration, and eventually less supportive of the imperative mandate. Moreover, reformers were less supportive of the committee plan proposed by faction leaders. And reformers proved more supportive of shifting the balance between SMD and party-list seats in favor of more SMD seats. In the instances of a substantial policy effect on institutional preferences, the reformers considered large and strong factions and a centralized decision-making process governed by faction leaders to be disadvantageous. The causal connection between policy interests, which were highly polarized, and preferences about the system of factions was a direct product of the 1993 election outcomes and was obvious to everyone. The election yielded an opposition bloc of factions that was measurably larger than the reform bloc. High thresholds, disciplined factions, and a centralized decision-making process would advantage the opposition bloc and risk the re-creation of an opposition-controlled parliament. While the policy divide reaching into the new Duma could hardly have been wider, it proved not to be the only basis for defining self-interest on the part of the newly elected deputies. The mixed electoral system imposed by presidential decree created significant differences in electoral environments among deputies. To be sure, many deputies in the CPRF who won district seats were just as tightly bound to the party as party-list deputies; for most other SMD deputies, however, the ties that bound them to party and faction were quite loose. Many deputies ran with no party endorsement or with multiple endorsements in 1993. And many SMD deputies won their races with much less than a majority of the total vote. Some SMD deputies, many of whom were intellectuals, were even strongly antiparty, as we frequently see in new democracies. In these ways, vast differences in electoral experiences and, perhaps, electoral futures were brought into the Duma. These differences in electoral context were much larger, and probably of potentially far greater significance, than the differences in electoral marginality found among American legislators in their single-member districts or in list placement found among Western European parliamentarians in party-list systems. As is frequently observed in cross-national comparisons, party-list deputies are more motivated than SMD deputies to support their national party leaders and tolerate faction discipline; SMD deputies are more oriented toward their local districts. Strong national or local party control of nominations (placement on the ballot) may counter these tendencies, and do in some political systems, but, in Russia, party control of district ballot access was mixed and limited, while parties exercised strong control of party lists. Mode of election differentiated deputies on many questions of institutional choice in the new Duma. In comparison with party-list deputies, SMD deputies were more likely to oppose disciplined parties, a high

INSTITUTIONAL CHOICE

141

threshold for faction recognition, and the imperative mandate. In this respect, SMD deputies had tendencies similar to those found among reformers. Still, as we have demonstrated, deputies’ mode of election had strong independent effects on attitudes about party-related institutions. And mode of election determined other institutional preferences that showed no or little relationship to policy preferences. SMD deputies were significantly more likely to favor voting rights on the Council of the Duma for committee chairs, more likely to oppose approval of the slate of committee chairs and chamber vice chairs produced by faction leaders, and much less likely to favor faction control of committee memberships for rank-and-file deputies. Not evident in Table 7.1 are the shifts in preferences we detected. For example, before the 1993 elections reformers’ preferences for more majoritarian governing rules were affected by their overly optimistic expectations of a strong showing at the polls. Afterward they tended to prefer protective arrangements through proportional and parity rights of representation on the Duma’s steering body and in the distribution of committee chairmanships. In this case, estimates of the consequences of certain institutional options changed as legislators updated their views of the consequences of the institutional alternatives. The particular sequence of events was critical to their choices. Similarly we found that deputies’ votes on the issue of the minimum size required for the registration of deputy groups shifted somewhat between the first day of the new Duma’s proceedings in January and a second round of voting on this issue three months later. Since this simple rule has significant consequences for the distribution of power in the chamber, we anticipated that deputies would need to make trade-offs among their electoral, ideological, and factional interests. We found that while all three sets of effects were present and statistically detectable both on the first day and again three months later, policy-related and electoral influences were relatively stronger in January than in March, and factional effects grew during the period. This is consistent with an account suggesting that factional influences for deputies are a product of faction leaders’ and members’ efforts to coordinate voting positions within their factions in ways that minimize the losses and maximize the benefits to members of voting with their fellow faction members.

Developments since 1995 An additional perspective on goals and institutional preferences can be gained by comparing the formation of governing arrangements in the 1994 Duma with the analogous steps taken by the new Dumas that convened in 1996 and 2000. If the governing arrangements chosen by the

142

CHAPTER SEVEN

Duma’s deputies in early 1994 reflected purely conjunctural influences—such as lingering fear that President Yeltsin would force parliament to disband again if it organized itself differently or some passing political impulse in the body politic—then we would expect the new members elected in 1995 and 1999 to organize their work differently. If the distribution of members’ policy, electoral, and partisan interests did not change too greatly, however, then we might expect members to adopt similar arrangements with relatively little controversy. Such would be the case, for example, if, following the election, there was still no strong and cohesive ideological majority. Then, in view of the fact that the election law the deputies passed in 1995 preserved the half-SMD, half–party-list composition of the chamber by electoral mandate, we would anticipate that the interaction of policy and electoral concerns would be roughly the same as in the first Duma. Under those conditions, with little change in the relative balance and distribution of the motivating goals for deputies’ calculations about the effects of alternative institutional arrangements, we would expect them to preserve the old arrangements. The 1996 Duma This indeed is what happened in 1996. In an election in which fortythree electoral organizations put up lists of candidates for the PR ballot, only four received more than the minimum requirement of 5 percent of the list vote. All four had roots in the outgoing Duma: the Communists, the LDPR, Yabloko, and Our Home Is Russia, the pro-government party of Prime Minister Chernomyrdin, elements of which were based on two deputy groups that formed in the Duma in the spring of 1995. These four parties were automatically entitled to form factions under the old rules. The rest of the deputies could form groups, including SMD deputies from parties that had fought unsuccessfully to win party-list seats and from groups formed by independents. The groups represented distinct policy directions—communist, nationalist, democratic opposition, and pro-government pragmatism. Individually none of the factions in the new Duma possessed a majority of the chamber. The Communists were by far the largest faction in the Duma, but they commanded a majority of votes in the Duma only by forming voting coalitions with other factions and groups (see Table 7.2). Besides the four party factions that formed on the basis of their list representation, another three deputy groups attracted a sufficient number of members to meet the thirty-five-member threshold for registration. Two of these (Agrarians and People’s Power) corresponded to electoral associations that had party lists in 1995 but failed to meet the 5 percent threshold. Those of their candidates who succeeded in winning

INSTITUTIONAL CHOICE

143

TABLE 7.2 Factional Membership in the State Duma, 16 January 1996 Faction

Percent of Vote Won in PL Ballot

PL Seats Received

SMD Deputies Affiliated

Total Seats

Share of Seats (percent)

22.7 11.4 10.3 7.0 — — — —

95 50 45 31 0 2 2 0

54 1 20 15 41 33 35 26

149 51 65 46 41 35 37 26

33.1 11.3 14.4 10.2 9.1 7.8 8.2 5.8

CPRF LDPR NDR Yabloko RReg APR PPwr Unaffiliated

Note: The party-list mandate deputies who joined the Agrarian and People’s Power groups were “lent” to these groups by the Communist faction, along with a number of single-member-district deputies. The purpose was explicitly to bring these two groups, ideological allies of the Communists, up to the level of membership required for registration. In this way the Communist faction gained two friendly, and dependent, votes on the Council of the Duma.

TABLE 7.3 Leadership Posts in the Sixth Duma, by Faction, January 1996 Chair of Duma

Deputy Chairs of Duma

Committee Chairs

Deputy Committee Chairs

Faction

N

%

N

%

N

%

N

%

CPRF LDPR NDR Yabloko RReg APR PPwr Unaffiliated Total

1

100.0

1 1 1

20.0 20.0 20.0

1

20.0

1

20.0

9 4 3 5 3 2 2

32.1 14.3 10.7 17.9 10.7 7.1 7.1

5

100.0

28

100.0

17 24 24 19 8 8 6 2 108

15.7 22.2 22.2 17.6 7.4 7.4 5.6 1.9 100.0

1

100.0

district seats then attracted enough like-minded colleagues and “borrowed” members from the Communists that they could register. The seventh faction, Russia’s Regions, was a group made up of independents, generally sympathetic to the executive branch but unwilling to join one of the party factions. As in 1994, the deputies of the new Duma were polarized in their policy positions. Factions differed sharply by ideological preferences. The mean score of the Communists, was 1.01 and the mean score of Yabloko was −.85, on a scale that varies from +2.00 to –2.00. The median deputy was slightly skewed to the left of the mean, a little more than was

144

CHAPTER SEVEN

true in the previous Duma. Thus, in a highly divided Duma, the policy camps were still relatively balanced, tilting somewhat more to the left than had been the case in 1994, but not by much. We would anticipate, therefore, that there would be little pressure to alter the inherited rules concerning the minimum threshold for registration of groups; to change the procedure for distributing committee and other leadership positions; to alter the powers, composition, or means of forming the Council of the Duma; to change the system of factional rights in such a way as to favor either single-party control, to weaken factions substantially, or to overturn the electoral system. The first vote taken by the newly convened Duma deputies was on a motion to govern their proceedings by retaining the Standing Orders adopted in the previous Duma. The motion passed by an overwhelming margin. Deputies did not revisit the contentious threshold size issue or question how the Council of the Duma would be structured. Instead, their first item of business was to choose a chair for the chamber. Then, once Gennadii Seleznev, a member of the Communist faction, was elected chairman, faction leaders negotiated over the distribution of leadership posts. First they decided to give each faction a deputy chairmanship of the chamber.1 Then they worked out a preliminary agreement to assign committee chairmanships to factions. As before, committee chairmanships were distributed in rough proportionality to factional strength, weighted by factions’ policy-based interest in controlling a particular jurisdiction. To make the task of assigning committee chairmanships to factions easier (albeit among a smaller number of factions), they increased the number of committees from twentythree to twenty-eight. As a result, committee jurisdictions overlapped even more than in the previous Duma. Next the deputy chairmanships of the committees were distributed proportionally. In all, around onethird of the members received some kind of leadership position (see Table 7.3). As in 1994, nonparty deputies objected strenuously to the package agreement on the grounds that it favored the interests of factions and party-list deputies over the interests of independents and deputies elected in single-member districts. The Russia’s Regions group initially refused to accept the agreement and to delegate members to occupy the two committee chairmanships reserved for it. Several deputies considered sending an appeal to the Constitutional Court that the distribution of committee chairmanships by agreement among faction leaders violated their constitutional rights.2 After additional bargaining, however, when it was evident that the package agreement was unlikely to be overturned either by the Duma or the Court, and the faction leaders agreed to give Russia’s Regions two additional committee chairmanships, the

INSTITUTIONAL CHOICE

145

TABLE 7.4 Factional Voting Support for Unified List of Committee Chairmanships, Second Duma, 23 January 1996 Faction CPRF APR PPwr RReg Yabloko LDPR NDR Unaffiliated

Percent Voting Yea 91 83 58 43 74 81 82 37

group withdrew its objections.3 The revised list then passed on January 23 by a wide margin (345 votes to 8) and with large majorities in most factions (Table 7.4). The Communists refrained from asserting majority control over the chamber, although they took advantage of the rules on group registration and Council of the Duma membership to multiply their points of leverage by ensuring that two allied groups gained registration as factions. But they chose not to alter Duma rules to permit a majority coalition to control committee chairmanships or the Council. Instead, in allocating committee chairmanships, they preserved the principle of proportionality, and in the make-up of the Council of the Duma they retained the parity principle. Why did the Communists choose not to form a dominant coalition and create a majority-ruled leadership system? The answer may lie in any of several considerations. They may have been reluctant to move too hastily to confront Yeltsin and give him a pretext for suspending the constitution or canceling presidential elections. Or, more likely, differences among members of the Communist faction, or between the Communists and their frequent coalition partners, the Agrarian and Popular Power groups, made many members of a potential majority coalition unwilling to concede control to the Communist faction’s leadership. When they devised the package agreement in January 1996, the faction leaders agreed to revisit the arrangement two years later. As in the previous Duma, the informal nature of the agreement led to conflicts when committee chairmen left their committees and factions attempted to assert their right of ownership of the committee. For example, when the progovernment Our Home Is Russia group expelled the wayward Lev Rokhlin, chairman of the Defense Committee, from the faction in September 1997, the group expected that the other factions would allow it to replace Ro-

146

CHAPTER SEVEN

khlin as chairman with another faction member.4 The Communists balked at this. Gennadii Ziuganov, the Communists’ leader, declared that he felt no obligation to honor the interfactional package agreement on distributing committee chairmanships.5 The Communists did, however, agree to allow Our Home to name a replacement for Alexander Shokhin as first deputy chairman of the Duma when Shokhin voluntarily left this position to become chairman of the Our Home faction.6 Only in May 1998 did the Communists agree to allow Our Home to appoint a new chairman of the Defense Committee with one of their faction members, as part of a side agreement under which the Our Home faction agreed to support a Communist, Oleg Mironov, as human rights commissioner. The precedents set by these early decisions demonstrate that the institution of distributing committee chairmanships in a package in proportion to the strength of factions in the chamber is a purely informal and largely unenforceable arrangement. Factional property rights in leadership positions are contingent on the consent of other factions and frequently must be renegotiated through piecemeal deals involving two or three leadership positions simultaneously. More firmly established is the rule of inter-factional consultations on major political decisions. Future Dumas may certainly adopt different governing arrangements. If the constitutional relationship between president and government evolves toward a more parliamentary system, in which the government is formed from a victorious majority coalition in parliament, the Duma may well shift toward a more majoritarian arrangement for exercising internal power. Until the external environment prompts deputies to reconsider the party-dominated, power-sharing arrangements that they have devised, however, the rules set in place by the first Duma in January 1995 are likely to be self-perpetuating. In January 1998 the Duma passed a new version of the Standing Orders, incorporating numerous minor changes. The new text incorporated a number of amendments that had been proposed but did not alter the institutional arrangements discussed here in any significant way. The thirty-five-member threshold for registration of groups remained; the rule under which the Council of the Duma was composed equally of the heads of the party factions and registered groups was retained; the powers of the Council of the Duma to shape and propose the agenda were also left unchanged; committees still have no formal power to bottle up legislation. Numerous minor details were added to clarify bill procedure. Through the life of the second Duma, however, none of the Duma’s essential institutional features was overturned. When the working group on the standing orders (made up, in accordance with the parity rule, of one representative of each faction) presented its proposal for a new revision of the rules to the floor, the chair

INSTITUTIONAL CHOICE

147

observed that the principle of parity of factions and groups had been a difficult issue and had prompted lengthy discussion. Ultimately, she explained, “the working group proceeded from the premise that the liquidation of the equality of rights of factions and deputy groups would lead to the violation of the fragile political equilibrium that we have achieved, and would affect the internal life of the chamber and the legislative process.”7 Following debate, all the recommendations of the working group were adopted by a large margin (331 yeas, 4 nays, 0 abstentions). The evidence from our 1996 member survey sheds some light on the reasons for this apparent state of equilibrium. Despite continuing deep divisions over ideological issues, we found (as we did in our 1994 survey) relatively high areas of consensus over many institutional features. One such feature has been the proper balance between committee and party power. Generally faction leaders and strong partisans in the chamber supported giving party leaders a vote on the Council of the Duma and denying it to committee chairs, but nonparty members and singlemember-district deputies were almost as likely to oppose giving committee chairs voting rights on the Council. The Standing Orders adopted in 1994 denied committee chairs a vote, and that rule was retained in the second Duma. We found that, overall, only 35 percent of deputies would advocate giving committee chairs full voting rights in the Council. This was not an issue that divided left from right. Voting rights were supported by only 30 percent of Communists and LDPR deputies, 28 percent of NDR members, 14 percent of Yabloko members—but 71 percent of Agrarians. The gap between Communists and Agrarians on this point clearly reflects the nonparty status of the Agrarians in the current Duma. The question of the degree of influence the factions have in the Duma was also divisive but does not coincide with the left-right cleavage. Asked whether political factions have too much power, about the right amount, or too little, more than half the respondents were satisfied with the present situation. Not surprisingly, Communists were especially satisfied (97 percent chose this response), but so, too, were two-thirds each of the LDPR and Yabloko. Majorities of NDR, Agrarian, and Russia’s Regions members thought factions had too much influence. In short, the deputies split rather evenly according to their degree of party organization and the electoral resources the parties offer the factions. Deputies in factions tied to electorally viable national parties were satisfied with the party-oriented institutional system. Likewise, 56 percent of all deputies said that the power exercised by the Council of the Duma was about the right amount. This view was shared by majorities of factions with varying policy perspectives—the Communists, Agrarians, NDR, LDPR, and Yabloko. The People’s Power

148

CHAPTER SEVEN

and Russia’s Regions groups were less favorable—40 percent of Russia’s Regions and 24 percent of People’s Power believe it has too much power. But, overall, 10 percent of deputies would give it still more power, a figure that is roughly the same across all factions. The same division of opinion between party-oriented and independent-dominated factions is apparent here. Divisions varied greatly on the issue of the chairman’s power. Overall, 56 percent of all deputies believed that the chairman of the Duma had about the right amount of power. Communists were a good deal more satisfied—88 percent thought his power the right amount and the rest thought it insufficient. Agrarians and People’s Power members were satisfied as well: three-fourths of Agrarians thought it appropriate and the rest would have given him more; 70 percent of People’s Power deputies were satisfied and some would have given him more (three, or 17.6 percent). Other factions were divided between those who considered the power about right and those who thought it too much. NDR deputies were evenly divided—45 percent thought it appropriate and 45 percent thought it too much. Even more negative evaluations came from other factions: 35 percent of Yabloko deputies thought it appropriate and 54 percent thought it excessive; and 70 percent of the Russia’s Regions group considered it excessive. Chairman Seleznev did not have the degree of confidence that Rybkin elicited after his first year; our survey found much higher levels of satisfaction with Rybkin’s powers in 1994. The data suggest that, both in behavior and in preferences, the politics of institutional structure divided party-oriented factions from independents much more than left from right. These results are consistent with the interpretation of the Duma as an assembly whose operating rules and organizational structure have been created under the influence of a set of overlapping member goals, some of which coincide under certain circumstances but may diverge at other points, depending on the implications of the arrangement for particular interests of members and the degree of knowledge members have about how these institutions will affect them. The 2000 Duma When the new corps of deputies elected in December 1999 convened in January 2000, the political situation had changed significantly. President Yeltsin’s resignation on 31 December 1999 automatically made Prime Minister Vladimir Putin acting president. Putin’s rapid ascent in late 1999 was reflected in the electoral success of the party associated with him, Unity, which rivaled the Communist Party in the list vote in December. Moreover, Putin himself took an active part in forming the new gov-

INSTITUTIONAL CHOICE

149

erning arrangements in the 2000 Duma. The composition of the Duma following the December elections, combined with Putin’s intervention, led to striking changes in the way leadership positions were distributed in the new Duma. Table 7.5 indicates the distribution of party list and single-memberdistrict seats by party in December and the translation of these results into the factions and deputy groups that formed in the Duma. As in the 1993 and 1995 elections, the winning parties could be classified according to four ideological tendencies: leftist, reformist, nationalist, and pragmatic pro-government. The latter category has come to be known as the “party of power,” and various political organizations have attempted to occupy this niche in the spectrum. In 1999 the Kremlinsupported party called “Unity” ran successfully as the pro-government party, benefiting from Putin’s endorsement. A rival “party of power,” the Fatherland–All Russia bloc, had enjoyed early popularity in the race but then faded as Unity rose on the strength of its association with the popular prime minister. Together, however, these two parties won over 36 percent of the party-list vote. The Communists won approximately the same share of the party-list vote as in 1995 and about the same number of single-member-district races. But whereas in 1995 the successful parties had benefited enormously from the fact that fully half the party-list vote had been cast for parties that failed to clear the 5 percent threshold, in 1999 there were far fewer wasted votes (fewer than 19 percent), so that the electoral bonus was far less. As a result, the Communists wound up with only 124 seats overall rather than 149. As before, the leaders of the winning parties began negotiating over the distribution of leadership positions as soon as the votes were counted and before the new Duma convened. Using a modified form of the auction system employed in 1994 and 1996, they attempted to reach a package agreement over which faction would claim the chairmanship of the Duma and the distribution of the chairmanships of the committees that could be presented to the floor for approval. However, the negotiations reached an impasse and remained unresolved as the Duma opened on January 18, 2000. Vladimir Putin intervened in the bargaining, directing the Unity faction to negotiate privately with the Communist faction. They did so, bringing into the bargain the largest of the registered independents’ groups, a group of single-member-district deputies called “People’s Deputy.” These three factions together commanded a bare minimum majority of 228 deputies. This was sufficient to force the other factions and groups either to join them on the terms dictated to them or to forfeit any possibility of receiving a leadership post. Quickly Zhirinovsky’s LDPR faction, the Agro-Industrial Group, and Russia’s Regions

24.29 — 23.32 13.33 5.98 — 8.52 5.93 — — 81.37

CPRF AIG Unity OVR LDPR PDep SPS Yabloko RReg Independent Totals

67 — 64 36 17 — 24 17 — — 225

List Seats Won 55 2 9 32 2 — 5 5 — —

SMD Seats Won 124 2 73 68 19 — 29 22 — —

89 41 81 45 17 58 32 21 38 19 441

20.2 9.3 18.4 10.2 3.9 13.2 7.3 4.8 8.6 4.3 100.0

Total Seats Won No. Duma Faction Duma Seat in Election Members (February) Share (%)

Source: Central Election Commission reports; RFE/RL reports.

List Vote (%)

Party or Deputy Group

TABLE 7.5 1999 Parliamentary Election Results and Seat Distribution in 2000 by Party

INSTITUTIONAL CHOICE

151

TABLE 7.6 Distribution of Leadership Posts by Faction, January 2000 Faction

No. of Seats In % Committee Chairs (No.) In %

CPRF AIG Unity OVR LDPR PDep SPS Yabloko RReg Independent Total

89 41 81 45 17 58 32 21 38 19 441

20.2 9.3 18.4 10.2 3.9 13.2 7.3 4.8 8.6 4.3 100.0

10 2 7 1 1 5 1 0 1 0 28

35.7 7.1 25.0 3.6 3.6 17.9 3.6 0 3.6 0 100.0

Source: Press reports.

joined the majority. The Fatherland–All Russia faction, however, headed by the influential former prime minister Evgenii Primakov, joined with the Union of Rightist Forces (SPS) and Yabloko to voice their protest against the arrangement. These three factions then declared a boycott of Duma proceedings and walked out of the session. The Duma then voted on the election of Communist Gennadii Seleznev as chairman of the Duma. The motion passed with 294 votes. The opposition alliance’s boycott lasted until February 9, when they returned to work, agreeing to accept the meager positions that had been reserved for them in return for vague promises from the majority to give expedited consideration to agenda items they considered important. This muscular display of majoritarian power by the three largest factions resulted in a lopsided distribution of committee positions, as Table 7.6 indicates. The Communists gained the speaker’s position and ten committee chairmanships. Unity won the first deputy chairmanship and seven committee chairmanships. People’s Deputy received five committee chairmanships. The LDPR, Russia’s Regions, Agro-Industrial Group, OVR, and SPS were assigned one committee chairmanship each, and Yabloko none. Compared with the previous two Dumas, the new allocation of leadership posts was far more majoritarian and far less proportional. Many observers were disturbed that this arrangement excluded Putin’s nominal allies in the reform camp from chairmanships. Particularly surprising was the treatment of SPS, which had loyally supported Putin over the war in Chechnia and considered Putin an ally. While we can only speculate about the motives of Vladimir Putin and the Unity faction in forming this alliance with the Communists, two considerations appear relevant. One effect of the agreement was to neutralize Evgenii Primakov, and this may well have been its immediate objective.

152

CHAPTER SEVEN

Primakov had been widely mentioned as a likely candidate for the speaker’s job, which would have given him visibility and a modest share of political influence in national politics. By turning the speakership over to Gennadii Seleznev, the Communist who had held the job over the 1996–99 Duma term, Putin ensured that Primakov would not become an influential figure in the Duma. A second consideration is that the Communists gained less than had initially appeared to be the case. According to press reports, they had made claims to the defense, security, and legislation committees, but received none of them. They failed to obtain the two committees with greatest influence over the economy, the Budget Committee and the Banking Committee. Nor did they obtain either the committee on property and privatization or on foreign affairs. In fact, none of the committees considered most powerful went to the Communists. Similarly, the Communists lost several battles over committee jurisdictions. One case occurred when the Communist chair of the Economic Policy Committee demanded jurisdiction for his committee over banking and Central Bank regulation and monetary policy, and some rights in the area of production-sharing agreements. The Council of the Duma, however, rejected all these demands.8 Another case concerned the relationship between the jurisdictions of the Legislation Committee and the Committee on State Organization. Anatolii Luk’ianov of the CPRF had chaired the Legislation Committee in the previous Duma and had made active use of its constitutional jurisdiction to sponsor a number of antipresidential constitutional amendments. Now Luk’ianov had been assigned the chairmanship of a committee with a broad but vague mandate, the Committee on State Organization. He and the Communists demanded that his new committee be assigned jurisdiction over constitutional legislation. However, the Unity faction opposed this change and proposed assigning constitutional revision to the Legislation Committee (which was headed by an SPS deputy). In floor voting, the Communists were defeated.9 This vote again suggested that the alliance of Unity with the Communists at the beginning of the Duma’s work was tactical and might not serve as a precedent for voting alignments on other issues. Thus, although the Unity-Communist alliance enabled Putin to neutralize Primakov and OVR by giving the Communists seemingly important leadership posts, it left the Communists in positions where they could not threaten either Putin’s policy or his power. The coalition that commanded a bare winning majority in January in the distribution of committee portfolios did not become a stable voting coalition over more substantive issues. This situation stood in sharp contrast to the previous Dumas, when the Communists were almost invariably on the winning side in Duma votes on major legislation. The balance of power appeared to reside with Unity. Unity’s early influence in 2000

INSTITUTIONAL CHOICE

153

derived more from its access to executive power and its central location in the political spectrum. By mid-2000, Unity did not appear to be likely to join with the Communists on many major policy matters. Indeed, Putin’s promise to pursue land reform legislation put him and Unity at odds with the Communists on perhaps the most fundamental economic issue facing the country. Even with the remarkable change in the composition of the Duma following the 1999 elections, the Duma did not alter any of the basic decision-making processes that we have reviewed in this book. The powers and composition of the Duma’s steering body remained the same as in the two previous Dumas, including the rule that each faction, regardless of size, is represented on the Council by a single leader. The thirtyfive-member threshold rule for determining which deputy groups will be registered and receive the rights and benefits of faction status remained in force, and had similar effects as before in that three groups—AgroIndustrial, People’s Deputy, and Russia’s Regions—representing leftist, rightist, pro-government, and centrist tendencies—quickly formed. Factions continued to demonstrate reasonably high levels of voting cohesion over matters concerning factional prerogatives, and the election law remained unchanged. The great difference lay in the changed relationship between the Duma and the government. Putin’s manipulation of the coalition strategy of the Unity faction may indicate the Kremlin’s intent to construct a more reliable majority in the Duma than existed between 1994 and 1999. A coalition in the Duma supporting the government would tend to bring the political system slightly closer to one in which the government reflected the balance of party forces in the Duma. It would alter the balance of policy-making power in the political system by strengthening the government’s accountability to parliament. Such a trend would require that the president form a government with a strong base of party support in the Duma and would alter deputies’ calculations about the relative benefit and risk of supporting the government in floor voting. In particular, it would strengthen the attractiveness to deputies of forming a crossparty majority which advanced the policy interests of the government. A more parliamentary form of government would speed the evolution of the Duma in a majoritarian direction.

Implications for Theories of Legislative Institutions Recent theories of legislative institutions, most applied to the U.S. Congress, treat legislative procedures and structures as endogenous to the goals of legislators. In particular, institutions are viewed as solutions to collective action or transaction problems, the nature of which depends

154

CHAPTER SEVEN

on the goals of the legislators. Different assumptions about the goals of legislators have yielded different expectations for institutional arrangements and different emphases on the various components of modern congressional institutions. Three theories of congressional institutions, all explicitly grounded in the new economics of institutions, have been particularly important. In Weingast and Marshall’s (1988) distributional model, legislators seeking reelection confront a problem enforcing logrolling agreements about distribution. A committee system and supporting rules provide a solution. In Krehbiel’s (1991) information model, legislators seeking desirable policy outcomes confront a problem of motivating some members to provide the information. A committee system that creates incentives for committee members to gather and disseminate information is the solution. In Cox and McCubbins’s (1993) parties-as-cartels model, legislators seeking reelection confront a problem of maintaining or enhancing their party’s reputation. Party organization and leadership attend to the collective activities of parties that influence their reputation. In each case, legislators are better off, in a manner defined by their personal goals, by cooperating in the creation and maintenance of institutions. We might add other perspectives that do not explicitly derive from the economics of institutions tradition but offer alternative perspectives to the first three. For example, Dodd (1977) argues that legislators seek individual power in a powerful institution. They decentralize decisionmaking processes in order to spread power (or, at least, committee-based power) as widely as possible. Over time, increasing decentralization degrades Congress’s power, so legislators must confront the problem of reforming the decision-making process to recentralize power. The party and committee system represent a balance for legislators seeking personal power in a powerful institution. And we should add to our list the conditional party government perspective (Rohde 1991), which treats congressional parties as policy coalitions. The effectiveness and importance of parties depends on their internal cohesiveness of the coalitions and the degree to which the two coalitions are polarized. In this view, legislators are motivated by their policy preferences to create, energize, and maybe limit parties. In a context of cohesive and polarized parties, the majority party exploits agenda control mechanisms that it creates to generate outcomes biased in its favor. Referring to the first three perspectives, Shepsle and Weingast (1995, 22–23) observe that “nothing inherent in the logic of these approaches makes them antithetical . . . Congress is a multifaceted organization, one that is unlikely to be understood in terms of a single principle.” We agree with this assessment and would extend it to the other perspectives. The

INSTITUTIONAL CHOICE

155

question is the nature of the synthesis that is likely to emerge. Shepsle and Weingast provide some guidance. We suggest a somewhat different approach. Shepsle and Weingast (1995, 24–32) begin by identifying the “essential” features of the various theories. The distributive theory depends on heterogeneous policy interests and a corresponding multidimensional policy space. The information theory turns on imperfect information that is diverse and highly valued. The parties-as-cartels approach emphasizes the need for party-based coordination to manage the party-relevant externalities of committee actions. These essential features, they claim, can be woven into an integrated account of congressional institutions. In a legislature composed of members with heterogeneous interests, arrayed in a multidimensional space, a committee system could simultaneously address the problems of exchange and information. A party-based legislature might provide for exchanges of information and policy among fellow partisans with heterogeneous interests and a need for information. Having made these observations, Shepsle and Weingast (1995, 31) then offer the following guidance: Central to any effort at resolving the theoretical compatibility among these . . . approaches is a theoretical model that melds the gains from exchange framework with the informational and party-based approaches in a multidimensional setting. Developing such a model entails solving a series of difficult nested testing problems that we are not about to pose, let alone resolve, here. But a comprehensive, spatially organized theory in many dimensions is clearly required if we are to assess the relative importance of distributive, information, and partisan impact.

We view the options somewhat differently. We do not have the confidence that the theories can be melded. In fact we are not sure that an integrated theory of the development of Congress or legislative institutions is likely to be developed. Instead we think that our general approach, one that recognizes the compatibility of the approaches but does not attempt to transform the accounts into a common metric, is more promising, at least for now. A missing ingredient in Shepsle and Weingast’s recommendation is consideration of the microfoundations of the contending theories. It is the endogeneity of institutions to legislators’ goals—and the connection between a particular goal and particular institutional features—that has made these accounts so persuasive and important. For example, the distributive model differs from the information model not only regarding the heterogeneity of interests and dimensionality but also with respect to the goal posited for legislators. The distributive model begins with

156

CHAPTER SEVEN

the assumption that electorally motivated legislators seek benefits for their district or state constituencies; the information model begins with legislators’ concern about policy outcomes, however that concern is motivated. The parties-as-cartels theory begins with the concerns of electorally motivated legislators about their parties’ reputations—policy is entirely instrumental for legislators assumed to be exclusively concerned about reelection. The goal posited for legislators is essential to each of the approaches. In the parties-as-cartels account, for example, committee decisions have externalities defined in terms of their consequences for a party’s reputation. Reputation matters because it affects legislators’ reelection prospects. Just why constituents weigh party reputations independently of their own representatives’ behavior is beyond the scope of the theory. That they do gives legislators a reason to see that those externalities are managed. It is precisely the public-goods character of the party reputation that generates the collective-action problem that leaders solve. Issues and committees with little effect on a party’s reputation are of little concern to leaders. The importance of the argument stems from the (implicit) claim that party reputation has electoral consequences for legislators independent of what the legislators would otherwise do, such as voting in a manner consistent with constituency preferences. And this is precisely what Cox and McCubbins (1993) demonstrate empirically. Election outcomes for incumbents are strongly correlated with incumbents’ past electoral performance (fit to the district), but a dummy for party (interpreted as party advantage or disadvantage) shows a significant independent effect in all recent congressional elections. It is the independent impact of party reputation on the reelection chances of legislators that leads them to be concerned about reputation. Cox and McCubbins make the opposite argument. As they state: “We have focused solely on collective dilemmas that entail electoral inefficiencies. Another perspective on parties might focus instead on collective dilemmas entailing policy inefficiencies. . . . For the purposes of our discussion here, however, the differences between these two views are inconsequential (1993, 134).” Their assertion is difficult to square with their argument about the party record as a collective good. To square them, they must argue that the solution to electoral inefficiencies is not fundamentally different than the solution to policy inefficiencies. But that seems unlikely and, in any event, requires demonstration. Electoral inefficiencies—such as those owing to inadequate attention to the party reputation—need not vary perfectly with policy inefficiencies. For example, policy inefficiencies can arise in any policy arena, whatever its relevance to party reputation. If so, party leaders would be expected to have wideranging interests and involvement without regard to reputational con-

INSTITUTIONAL CHOICE

157

cerns. In fact it is not clear why party leaders would be the best solution to policy inefficiencies unless additional strong assumptions are made about the distribution of policy preferences. The translation of the reelection goal into a policy goal is also made by Krehbiel (1991) in elaborating his information theory. Somewhat loosely, he argues that the heterogeneity in constituents’ preferences “translates” into diversity in legislators’ preferences (77). In the informational account, the induced policy preferences (or, more precisely, preferences for policy outcomes) create a need for information and collectively form a single dimension along which majorities form to create and control committees. If instead legislators were concerned about reelection, the dimensionality of their constituents’ policy interests would determine the policy space and it would not be clear what majority, if any, could control committees. Moreover, constituents’ policy interests would dictate the degree of control over committee actions that the legislators would deem desirable. As Weingast and Marshall (1988) claim, many constituencies might have little interest in the actions of many committees. Shepsle and Weingast (1995) offer their guidance without consideration of the difficulty of translating electoral interests into the ideal points of spatial theory. The implicit assumption is that policy and electoral considerations can be translated into a common metric. This might be accomplished theoretically by noting that many political considerations, including electoral ones, shape a legislator’s preference at the point of decision (about a policy or institution). Preferences, then, reflect electoral considerations, among other things, and, by assumption, capture all that we need to know about legislators’ political motivations. Policy commitments (say, long-term ideological positions) are mixed with all other influences to define preferences exogenously to the theory. The interplay between policy commitments and other considerations is set aside by assumption. Such an approach strips away much of what is interesting about institutional choice. As a theoretical matter, the consequentialist arguments of positive theories are constructed for the purpose of demonstrating the implications of players’ goals for the institutions they construct. If we can predict identical institutions from different goals, we will have made progress. But, to date, the differences in emphases (parties, committees, leaders, etc.) and predictions for individual behavior persuade us of the value of retaining distinct individual-level assumptions. In fact, as an empirical matter, the observation that legislators with similar electoral (policy) interests sometimes have different policy (electoral) interests appears to help us explain variation in institutional preferences.

158

CHAPTER SEVEN

However convenient the marriage of electoral and policy goals might be in the American case, it is far less useful in the case of Russia. In the former, relatively weak electoral and legislative parties in the United States allow a legislator to freely adjust his or her strategies to electoral conditions, personal views, and so on. A universal single-member-district system places all legislators in broadly the same electoral context, and variation is found primarily in the competitiveness of districts. Most legislators seek reelection. And, broadly speaking, legislators’ policy positions are matched with their districts’ policy preferences. These features have made the American case more interesting than many European parliamentary systems in which strong parties create little within-party variation in legislators’ behavior, and party goals, rather than individual goals, become the starting point for theory about policy and institutional choices. In addition, these features have meant that there is plentiful circumstantial evidence for theories built on either policy or electoral goals. In the Russian case, the mixed electoral system creates critical complications for theory. While we might treat SMD deputies’ policy preferences as imputed from constituency preferences, as is suggested for members of Congress, we are still left with the other half of the Duma’s membership, chosen on the basis of placement of party lists. PL deputies might be assumed to take the policy positions of their parties (their “constituency”), and we could build a theory of a Duma comprised of the two stylized types of legislators. We must confess that this simplification has great appeal. Unfortunately our evidence shows substantial variation in the policy preferences expressed in surveys and revealed in voting behavior for SMD and PL deputies of the same party. Moreover, there is variation across factions in the variance of expressed and revealed policy preferences. Thus, at least for the Russian Duma, folding electoral interests into policy interests would shroud demonstrably important sources of variation in institutional preferences as we are trying to understand a potentially pivotal stage in the institution’s development. Shepsle and Weingast are right, of course, in noting the possible compatibility of the distributive, informational, and partisan accounts of Congress. And we recognize that we cannot assess the relative importance of distributive, informational, and partisan considerations without an integrated framework. Our hunch is that a more general theory of legislative institutions will be grounded in a more complex account of the interaction between the electorate, parties, candidates, and legislators. In such a theory, candidates and legislators will have their own political goals, the pursuit of which is promoted or constrained by the necessity of appealing to an electorate under a variety of electoral systems. In designing

INSTITUTIONAL CHOICE

159

their legislative institutions, legislators will weigh sometimes conflicting policy, electoral, power, and partisan goals that they bring with them from the typically complex electoral environment. Retaining the complexity of the goals imported with politicians into the legislature, it seems to us, is more likely to yield satisfactory theory. In fact our understanding of legislative institutions has been enhanced by developing distinct theories. Each theory has exploited an individuallevel perspective to generate expectations about institutional arrangements. Simplifying assumptions about individuals’ goals have proven essential to the exploration of their implications for institutions. More complex assumptions about goals would have yielded few propositions about the strategies of individuals and their preferences for institutional arrangements. Moreover, we do not fully understand the implications of even single goals for legislators’ preferences about their institutions. Fair criticisms of the distributive, informational, and partisan accounts have surfaced, and the task of sorting through them and reformulating the arguments has hardly begun. It seems unwise to make integration a top priority when clarification and correction in the original theories remain to be done. Moreover, it is not obvious to us that a single theory is the likely outcome of our efforts to understand processes of institutional development or change. No single process underlies geological change, plant growth, or organic evolution, and it seems unlikely that we will discover a single process of change in political, or even legislative, institutions. In explaining the pattern of development in a particular legislature, we probably will have to draw on several theories. Variation in sunlight, temperature, moisture, nutrients, and other factors influence biological development, but, for any specified period of time, variation in one or two of them is likely to play a more important role in development than the others. And each factor is likely to be more important for some facets of development than for others. In the case of political institutions, our task is to understand these variable effects.

Conclusion The special features of the Russian system have given us leverage on important questions generated by positive theories of legislative institutions. We have demonstrated the importance of Russian deputies’ multiple goals for the early development of the Duma’s most significant institutional features—its party system, committee system, and leadership structures—as well as the electoral system. And we have shown that partisan forces were sufficiently distinct from deputies’ personal electoral

160

CHAPTER SEVEN

and policy goals to have a significant independent effect on the choice of legislative institutions. No single logic produced the new Duma. It reflects a composite of choices made on the basis of a variety of political goals. The multiple goals of Russian deputies pose a theoretical challenge to political science. We have advanced the proposition that legislators do not find all their goals equally relevant to each of the policy or institutional choices they confront. Variation in goal relevance, transaction costs, and information across contexts leads legislators with complex goal profiles to apply goals opportunistically. A single goal may often dominate other goals in the context of a choice about a particular feature of parliamentary institutions. In this way, legislators are locally, not globally, rational. The interactions among institutions and strategies usually become increasingly complex as the consequences of interests become more remote, temporally or institutionally, and leads legislators to emphasize the immediate consequences of the institutional choices they face. Moreover, because goal relevance, transaction costs, and information can vary over time, the preferences of legislators about an institutional feature can change over time.

NOTES 1. Yabloko and the Agrarians objected to the arrangement, but a position was held open for Yabloko and was later filled by a Yabloko deputy. The package passed on the floor by a vote of 359 to 56 with one abstention (RFE/RL Newsline, 19 January 1996). 2. Ibid., 23 January 1996. 3. Segodnia, 24 January 1996. 4. RFE/RL Newsline, 10 September 1997; Segodnia, 5, 6 September 1997. 5. Segodnia, 11 September 1997. 6. Shokhin was not required to give up the deputy chairmanship of the Duma on becoming chairman of the Our Home faction. But he was apparently observing another provision of the gentleman’s agreement among factions, to the effect that a faction leader would not simultaneously take a leadership position in the chamber or a committee. It was very much in the interest of both Shokhin and the Our Home faction to uphold this rule, for fear of upsetting the entire web of interfactional agreements on power sharing. See Segodnia, 5 September 1997. 7. Gosudarstvennaia Duma: Stenogramma Zasedanii, Biulleten’, no. 149, 22 January 1998. 8. Segodnia, 11 February 2000. 9. Polit.ru web site, 18 February 2000.

APPENDIX DATA AND METHODS

I

N THIS APPENDIX we describe the survey, roll-call voting, and election data used in the study. In addition, we describe the dimensional analysis of the roll-call data that was performed to create policy position scores for deputies.

Survey Data We conducted four surveys, two of candidates for the Duma and two of elected deputies. The candidate surveys were conducted in late 1993 and late 1995. The deputy surveys were conducted in late 1994 to early 1995 and in 1996. 1993 Candidate Survey This survey was conducted in October and November 1993 by the Center for Political Technologies, in cooperation with Postfaktum News Agency and Russia’s leading survey research institute, VTsIOM (All-Russian Center for the Study of Public Opinion). A total of 420 candidates for the Duma were interviewed: 248 SMD candidates and 172 list candidates, of whom 30 also ran in districts. Thirty-one regions of the Russian Federation are represented, as are at least 68 electoral districts (several respondents refused to indicate their district). Forty-eight respondents were from the city of Moscow, and 36 were from St. Petersburg. The following regions are also represented: Altai krai, Arkhangel’sk, Briansk, Volgograd, Voronezh, Irkutsk, Karachaevo-Cherkessia, Krasnodar krai, Krasnoyarsk krai, Mordovia, Moscow oblast, Nizhnii Novgorod, Novosibirsk, Omsk, Perm, Rostov, Samara, Saratov, Sverdlovsk, Smolensk, Stavropol, Tatarstan, Tula, Udmurtia, Ulyanovsk, Khabarovsk, Cheliabinsk, and Chuvashia. Party representation was as follows: 25 respondents are from Russia’s Choice (includes 8 SMDs); 11 from PRES (includes 1 SMD); 13 from Yabloko (includes 1 SMD); 11 from Civic Union (includes 2 SMDs); 8 from Women of Russia (includes 1 SMD); 7 from RDDR; 27 from the Agrarian Party (includes 8 SMDs); 25 from the Communist Party (includes 1 SMD); 10 from DPR (includes 1 SMD); 31 from LDPR (includes 7 SMDs); and 3 from Future of Russia/New Names.

162

APPENDIX

TABLE A.1 Faction Representation in the 1994–95 Deputy Survey Faction

SMD

Party List

Total

RC Yabloko LDUD12 PRES NRP LDPR APR CPRF Other Total

22 7 15 2 29 6 22 12 10 125

17 11 1 6 0 16 11 20 7 89

39 18 16 8 29 22 33 32 17 214

1994 Deputy Survey The 1994–95 deputy survey was conducted in December 1994 and January 1995. A total of 214 deputies of the State Duma were interviewed by the Center for Political Technologies with interviewers from VTsIOM. The sample includes 125 SMD deputies and 89 party-list deputies. The faction composition of the sample is indicated in Table A-1. 1995 Candidate Survey The 1995 candidate survey was conducted in October and December 1995, by the Center for Political Technologies with VTsIOM interviewers. The candidates interviewed numbered 203; 102 were running in SMDs, 40 were on party lists only, and 61 ran in both districts and on party-lists. The following regions were represented (number of districts, number of respondents): Bashkortostan (5, 13); Altai krai (2, 15); Krasnodar krai (2, 16); Belgorod oblast (1, 1); Kemerovo (4, 15); Leningrad oblast (1, 1); Nizhnii Novgorod (2, 2); Omsk (2, 18); Penza (2, 19); Pskov (1, 1); Rostov (6, 12); Samara (5, 23); Sverdlovsk (1, 1); Tula (3, 12); and Cheliabinsk (5, 14). Party representation is shown by the number of survey respondents on a party list: Russia’s Choice (15), Our Home Is Russia (8), Yabloko (3), Ivan Rybkin Bloc (2), Women of Russia (5), Congress of Russian Communities (3), Agrarian Party of Russia (11), Communist Party of the Russian Federation (12), LDPR (7), Derzhava (1), and others (29)—a total of 101 party-list deputies.

APPENDIX

163

TABLE A.2 Faction Representation in the 1996 Deputy Survey Faction

SMD

Party List

Total

CPRF NDR LDPR Yabloko APR PPwr RReg None Total

16 11 1 11 18 17 19 1 94

17 18 25 17 3 0 1 0 81

33 29 26 28 21 17 20 1 175

1996 Deputy Survey The 1996 deputy survey was conducted in March and April 1996 by the Center for Political Technologies with VTsIOM interviewers. There were 175 deputies interviewed. Table A-2 lists the sample’s composition by party and mode of election.

Roll-Call Voting Data The data used to analyze Duma voting alignments come from the Moscow think tank, INDEM. From 1994 to 1997 INDEM sold to subscribers a series of proprietary software packages for the analysis of voting patterns in the Russian Duma, entitled “INDEM-Duma.” These products incorporated the results of a multidimensional scaling procedure with the raw vote files obtained from the Secretariat of the State Duma, together with background information on deputies and annotations on individual voting items to permit a user to map the locations in twodimensional space of individual members of the Duma for given time periods. We have extracted the raw vote data files from these programs and matched them to lists of individual deputies to create one set of data files listing all the electronically recorded votes by deputy, and another listing factional and committee affiliations and electoral backgrounds for each Duma member. The INDEM program also allowed us to identify particular key measures, such as those discussed in chapters 2, 4, and 5, and to analyze separately for those votes the alignments by deputy faction, electoral-mandate type, and policy preference.

164

APPENDIX

The Duma does not make public the electronically recorded voting records from its proceedings. Roll-call votes (poimennye golosovaniia), which are published in the stenographic report of the Duma proceedings, constitute only a small minority of all recorded votes. Duma deputies, however, cast votes on an enormous number of items (in 1994–95, 5,548 electronically recorded votes were cast), and all except secret votes are conducted electronically. Nearly all the Duma’s decisions, routine and otherwise, are made by twenty-second electronic voting. The electronic records are stored in the Duma’s data archives, and the Secretariat makes them available to members of the Duma. Under the Duma’s Standing Orders, however, they may be made available to others only with the Duma’s consent. We selected all votes in which voting alignments were significantly related to faction affiliation (see below) and subjected them to principal components analysis to produce our own scale scores on the major voting dimensions for individual deputies and groups of deputies. The dimensional analysis consistently identified a dominant voting cleavage, anchored by Communists at one end and liberal reformers at the other. The issues that loaded most strongly on this dimension included property rights measures, especially property in land; questions of privatization; relations with the West; and presidential power. We treat this voting dimension as a reflection of the deep division among Russian politicians between those adhering to Communist and nationalist positions and those preferring that Russia develop along the lines of a market-oriented, liberal democratic society. In 1994 this dimension accounted for 23 percent of the variance in voting alignments. Subsequently the dimension became less sharply articulated but was still the strongest dimension observed, and it continued to differentiate leftists from reformists in the Duma. Therefore we use a scale score based on revealed policy preferences, as indicated in votes across a broad range of issues, as a measure of underlying ideological predispositions. We also supplement vote-based measures of policy preferences with scores based on a dimensional analysis of the responses to the policy questions asked in our survey of deputies. However, the N in that case is much smaller, rendering the problem of establishing statistically significant relationships between policy preferences and our other variables even more difficult. Unfortunately the analysis of roll-call behavior in Russia’s Duma is not methodologically straightforward. Nonvoting is common and a major source of measurement error. Faction-switching adds another source of error for any analysis of the Duma’s two-year session. And identifying “party votes” is more problematic in a multiparty parliament than in twoparty systems.

APPENDIX

165

Nonvoting is a viable option for Russian deputies, as it is for deputies in a number of other parliaments. It is distinguished from abstention, which deputies may also choose, and also differs from simply not being registered in attendance for a given session. Any variant other than “yea” is equivalent to “nay,” since 226 votes—an absolute majority of 450 deputies—are required to approve a motion on a law. Thus many deputies express their opposition to a measure by refraining from voting for it. Both nay votes and abstentions, which require actively casting a vote, are rare. On the other hand, nonvoting may simply be the consequence of a deputy’s absence from the session, so it is unwise to code all nonvoting as “nay.” For that reason, neither coding nonvotes always as “nays” nor as a mean value between “yea” and “nay” is an entirely plausible solution. Nonetheless, giving missing data the mean value probably errs on the conservative side, since such a procedure will tend to mute differences between supporters and opponents of measures. This is the approach we have taken here. We identify party votes in order to calculate party support on those issues that involve some minimal degree of differences among parties. In most studies of the U.S. Congress, analysts accept the Congressional Quarterly’s definition of a party vote as one in which a majority of Democrats oppose a majority of Republicans. Such a definition is not useful in a multiparty parliament such as the Duma and cannot be generalized beyond systems with stable majority and opposition coalitions. In this study we estimated the linear relationship between deputies’ yea-nay votes and a set of dummy variables for the factions. A vote was included as a party vote if the multiple r was 0.5 or higher. Of the 5,548 votes initially available, 2,810 (50.6 percent) were retained under this decision rule.1 The definition of factional affiliation is also problematic. Almost onequarter of the members of the Duma changed factions at least once over the 1994–95 term. Two new factions were created in 1995 (“Stability” and “Russia”), and both were created with an eye to the impending 1995 parliamentary elections, according to participants we interviewed and press accounts at the time. They took members from factions such as New Regional Policy and Russia’s Choice. What is the relevant faction, then, for judging members’ loyalty? Rather than prejudge whether the faction they entered initially, based on their 1993 election and judgments over policy positions at that time, or the faction they joined later, calculating their policy and electoral interests for the future, should be taken as their “true” affiliation, we calculated loyalty in both ways, measuring the similarity of a member’s voting record to the faction he or she originally joined (“initial party”) and to that of the faction to which the individual belonged as the Duma’s term expired (“exit party”). The

166

APPENDIX

differences appear to have little effect on the inferences we draw about the relationship between electoral mandate and party loyalty, although they obviously influence how we calculate party loyalty. In chapter 6 we use measures based on “exit party” except where noted otherwise.

Election and Candidate Party Affiliation Data The Central Electoral Commission (CEC) never published a final official report of the results of voting for each candidate and party in each SMD electoral district for 1993. It did publish the official results for the partylist voting nationwide and by region, and a list of winning SMD candidates (see the CEC’s bulletin, Biulleten’ Tsentral’noi Izbiratel’noi Komissii Rossiiskoi Federatsii [1994], no. 1: 67). Therefore the data set we constructed is based on two reports, given to the authors, which came from members of the CEC. One lists the results of the voting in 205 singlemember electoral districts where valid elections were held; the other lists the results of voting for party lists in each electoral district. These were combined with information on the party affiliation of district candidates listed in Rossiiskaia gazeta on 30 November 1993, which indicates the way each candidate was nominated. About half the candidates were nominated by national electoral associations, which we are calling “parties.” The organization that nominated the candidate was noted in each candidate’s listing and was used in the data set as the indication of the candidate’s party identification. Some shrinkage occurred in the number of candidates running between the time the newspaper published the list and election day; a few dozen candidates withdrew. The number of district candidates in our data set is 1,623. The data may contain errors. In fact there were a number of charges of election fraud in the December 1993 voting. One widely discussed theory is that the CEC allowed some regional officials (regional governors were directly in charge of tallying and reporting votes from election districts in their regions) to manipulate the outcomes of the elections. Yeltsin was pressuring governors to ensure at least a 50 percent turnout to validate the constitutional referendum, which was his principal concern. Some governors, then, according to this theory, inflated turnout figures in their provinces, getting themselves elected in their own races for the upper chamber and distributing the “surplus” votes among various opposition parties in the party-list votes for the Duma. Since the CEC never allowed independent observers access to the ballots or the raw data and never published a full tally of votes, one cannot dismiss these charges. We discuss these charges of fraud in greater detail in Remington and Smith 1995.

APPENDIX

167

To identify the party affiliation of district candidates, we use the listing of candidates published in the Russian press in November 1993, which indicated the electoral organization that nominated candidates (Rossiiskaia gazeta, 30 November 1993). Candidates for district seats could be nominated in one of two ways. Registered electoral associations—those with the right to put up national candidate lists—could nominate candidates, or individuals could qualify if they could collect the required number of valid signatures (at least 1 percent of the voters of the district) on petitions. A total of 1,586 candidates ran for district seats, and half of them were nominated by national electoral associations. There were 1,717 candidates placed on party lists by processes governed by party rules or leaders. And some candidates were placed on both party lists and district ballots. Candidates winning district seats and qualifying for seats through their party lists could choose which kind of seat to take in the Duma. Nearly all of them chose to assume the district seat and free another list seat for a lower-ranked member of their party. Unfortunately the weakness of the ties between district candidates and political parties complicates comparison between votes for candidates and votes for party lists in each district. Even where candidates had a clear partisan identity, party names were not printed on the ballots used for the territorial districts. Voters had to know or figure out which candidates were associated with which parties; some polling stations posted a list of candidates’ party affiliations on the wall; in other cases, voters asked the parties’ observers stationed at the polling place, election officials, or friends and family members for help in determining candidates’ party affiliation. Personal observations at several polling places in Moscow on election day indicate that many voters wished to know the party labels of candidates and were frustrated that this information was not provided on the ballot form itself. Some district winners lacked a party affiliation as candidates but later joined a parliamentary faction. In such cases, we use the faction (parliamentary party) with which elected deputies registered in the Duma as an indicator of a candidate’s party. But a number of deputies not nominated by or affiliated with the registered electoral parties decided to form their own parliamentary factions in order to gain the advantages conferred on registered groups. They then competed in parliament for influence with the factions that formed out of electoral parties (Remington and Smith 1995). Still other district winners decided to change their partisan affiliation after the election. Some of the party switchers had been nominated by parties that had failed to clear the 5 percent threshold; others simply decided to move to a parliamentary faction more congenial to their political position. In fact members of party factions continue to flow in and out of factions, although not in great numbers.

168

APPENDIX

To determine the party affiliations of SMD deputies, therefore, we use figures for faction membership as of April 1994, that is, after the initial phase of Duma organization was over. Since party affiliation for Russian deputies is often far looser than in other European parliaments, no method for counting the parties’ victories in districts is entirely satisfactory. These data are taken from G. V. Belonuchkin 1994.

NOTES 1. We also have separated the votes into categories in order to take into account any effects associated with the parliamentary nature of the issue. We identified votes taken during the debate over the agenda, those on substantive motions, and those where the president had taken a definite position (usually by vetoing a bill) and the parliament was called upon to respond. Also, the Duma allows the chair to put the same motion several times in succession to a vote if and when the chair judges that a few more votes might be found to pass a measure for which a majority is barely lacking. In order not to bias our estimates by such repeated votes, we selected only the last in such series as indicators of members’ actual voting preferences. Moreover, we excluded prodecural votes.

REFERENCES

Aberbach, Joel D., Robert D. Putnam and Bert A. Rockman. 1981. Bureaucrats and Politicians in Western Democracies. Cambridge, Mass.: Harvard University Press. Agh, Attila. 1999. “The Parliamentarization of the East Central European Parties: Party Discipline in the Hungarian Parliament, 1990–1996.” In S. Bowler, D. M. Farrell and R. S. Katz, eds., Party Discipline and Parliamentary Government, 167– 88. Columbus: Ohio State University Press. Aldrich, John H. 1995. Why Parties? The Origin and Transformation of Political Parties in America. Chicago: University of Chicago Press. ———. 1994. “Rational Choice Theory and the Study of American Politics.” In Lawrence C. Dodd and Calvin Jillson, eds., The Dynamics of American Politics, 208–33. Boulder, Colo.: Westview. Bach, Stanley. 1989. “The Appeal of Order: The Senate’s Compliance with Its Own Rules.” Paper delivered at the annual meeting of the Midwest Political Science Association, Chicago, Illinois. Bawn, Kathleen. 1993. “The Logic of Institutional Preferences: German Electoral Law as a Social Choice Outcome.” American Journal of Political Science 37:965– 89. Belonuchkin, G. V. 1994. Federal’noe sobranie: Spravochnik. Moscow: Panorama. 18 April. Binder, Sarah A. 1996. “The Partisan Basis of Procedural Choice: Allocating Parliamentary Rights in the House, 1789–1990.” American Political Science Review 90 (1): 8–20. ———. 1997. Minority Rights, Majority Rule: Partisanship and the Development of Congress. Cambridge: Cambridge University Press. Biulleten’. 1995. Gosudarstvennaia Duma: Stenogramma zasedanii. No. 91 (15 March). Bowler, Shaun, and David M. Farrell. 1999. “Parties and Party Discipline within the European Parliament: A Norms-Based Approach.” In S. Bowler, D. M. Farrell, and R. S. Katz, Party Discipline and Party Government, 208–26.. Columbus: Ohio State University Press. Bowler, Shaun, David M. Farrell, and Richard S. Katz, eds. 1999. Party Discipline and Party Government. Columbus: Ohio State University Press. Brady, David W., and Charles S. Bullock III. 1985. “Party and Factional Organization in Legislatures.” In Gerhardt Loewenberg, Samuel C. Patterson, and Malcolm Jewell, eds., Handbook of Legislative Research. Cambridge, Mass.: Harvard University Press. Brady, David W., Richard Brody, and David Epstein. 1989. “Heterogeneous Parties and Political Organization: The U.S. Senate, 1880–1920.” Legislative Studies Quarterly 14:205–24. Brudny, Yitzhak M. 1993. “The Dynamics of ‘Democratic Russia,’ 1990–1993.” Post-Soviet Affairs 9:141–70.

170

REFERENCES

Chiesa, Giulietto, with Douglas Taylor Northrop. 1993. Transition to Democracy: Political Change in the Soviet Union, 1987–1991. Hanover: Dartmouth College, University Press of New England. Clausen, Aage. 1973. How Congressmen Decide. New York: St. Martin’s. Clem, Ralph S., and Peter R. Craumer. 1993. “The Geography of the April 25 (1993) Russian Referendum.” Post-Soviet Geography 34:481–96. Coase, Ronald. 1960. “The Problem of Social Cost.” Journal of Law and Economics 3:1–44. Coleman, James S. 1990. Foundations of Social Theory. Cambridge, Mass.: Belknap Press of Harvard University Press. Collie, Melissa P. 1985. “Voting Behavior in Legislatures.” In Gerhard Loewenberg, Samuel C. Patterson, and Malcolm E. Jewell, eds. Handbook of Legislative Research. Cambridge, Mass.: Harvard University Press. Colton, Timothy J., and Jerry F. Hough, eds. 1998. Growing Pains: Russian Democracy and the Election of 1993. Washington, D.C.: Brookings Institution. Cooper, Joseph, and David W. Brady. 1981. “Institutional Context and Leadership Style: The House from Cannon to Rayburn.” American Political Science Review 75:411–25. Cox, Gary W. 1987. The Efficient Secret: The Cabinet and the Development of Political Parties in Victorian England. Cambridge: Cambridge University Press. ———. 1997. Making Votes Count: Strategic Coordination in the World’s Electoral Systems. New York: Cambridge University Press. Cox, Gary W., and Mathew D. McCubbins. 1993. Legislative Leviathan: Party Government in the House. Berkeley: University of California Press. Cox, Gary W., and Amorim-Octavio Neto. 1997. “Electoral Institutions, Cleavage Structures, and the Number of Parties.” American Journal of Political Science 41 (1): 149–74. Davidheiser, Evelyn. 1998. “Right and Left in the Hard Opposition.” In Timothy J. Colton and Jerry F. Hough, eds., Growing Pains: Russian Democracy and the Election of 1993, 177–209. Washington, D.C.: Brookings Institution Press. Deckard, Barbara. 1976. “Electoral Marginality and Party Loyalty in House RollCall Voting.” American Journal of Political Science 20:469–82. Diermeier, Daniel, Joel M. Ericson, Timothy Frye, and Steven Lewis. 1997. “Credible Commitment and Property Rights: The Role of Strategic Interaction between Political and Economic Actors.” In David L. Weimer, ed., The Political Economy of Property Rights: Institutional Change and Credibility in the Reform of Centrally Planned Economies, 20–42. Cambridge: Cambridge University Press. Dodd, Lawrence C. 1977. “Congress and the Quest for Power.” In Lawrence C. Dodd and Bruce I. Oppenheimer, eds. Congress Reconsidered, 269–307. New York: Praeger. Downs, Anthony. 1957. An Economic Theory of Democracy. New York: Harper. Duverger, Maurice. 1954. Political Parties: Their Organization and Activity in the Modern State. Translated by Barbara and Robert North. New York: Wiley. Gerald M. Easter. 1997. “Preference for Presidentialism: Postcommunist Regime Change in Russia and the NIS.” World Politics 49 (2): 184–211.

REFERENCES

171

Eggertson, Thrainn. 1990. Economic Behavior and Institutions. Cambridge: Cambridge University Press. Epstein, Leon D. 1980. Political Parties in Western Democracies. New Brunswick, N.J.: Transaction Books. Fenno, Richard F. Jr. 1973. Congressmen in Committees. Boston: Little, Brown. Fiorina, Morris. 1974. Representatives, Rolls Calls, and Constituencies. Lexington, Mass.: Heath. Fish, M. Stephen. 1995. Democracy from Scratch: Opposition and Regime in the New Russian Revolution. Princeton, N.J.: Princeton University Press. Geddes, Barbara. 1996. “Initiation of New Democratic Institutions in Eastern Europe and Latin America.” In Arend Lijphart and Carlos H. Waisman, eds., Institutional Design in New Democracies, 15–41. Boulder, Colo.: Westview. Gunther, Richard. 1989. “Electoral Laws, Party Systems, and Elites: The Case of Spain.” American Political Science Review 83:835–58. Haggard, Stephan, and Robert R. Kaufman. 1995. The Political Economy of Democratic Transitions. Princeton, N.J.: Princeton University Press. Haspel, Moshe. 1998. “Should Party in Parliament Be Weak or Strong? The Rules Debate in the Russian State Duma.” In John Loewenhardt, ed., Party Politics in Post-Communist Russia, 178–200. London: Frank Cass. Herzog, Dietrich, Hilke Rebenstorf, Camilla Werner, and Bernhard Wessels. 1990. Abgeordnete und Bu¨rger: Ergebnisse einer Befragung der Mitglieder des 11. Deutschen Bundestages und der Bevo¨lkerung. Opladen: Westdeutscher Verlag. Hibbing, John R., and Samuel C. Patterson. 1992. “A Democratic Legislature in the Making: The Historic Hungarian Elections of 1990.” Comparative Political Studies 24:430–54. Hough, Jerry F. 1994. “The Russian Election of 1993: Public Attitudes toward Economic Reform and Democratization.” Post-Soviet Affairs 10:1–37. ———. 1997. Democratization and Revolution in the USSR, 1985–1991. Washington, D.C. : Brookings Institution. Huber, John D. 1996. “The Vote of Confidence in Parliamentary Democracies.” American Political Science Review 90:269–82. Huber, John D., and Ronald Inglehart. 1995. “Expert Interpretations of Party Space and Party Locations in 42 Societies.” Party Politics 1:73–111. Iversen, Torben. 1994. “The Logics of Electoral Politics: Spatial, Directional, and Mobilizational Effects.” Comparative Political Studies 27:155–89. Jasiewicz, Krzysztof. 1993. “Structures of Representation.” In Stephen White, Judy Batt, and Paul G. Lewis, eds., Developments in East European Politics, 124–46. Durham, N.C.: Duke University Press. Johnson, Simon, Panos Kouvelis, and Vikas Sinha. 1997. “On Reform Intensity under Uncertainty.” Journal of Comparative Economics 25:297–321. Katz, Richard S. 1986. “Party Government: A Rationalistic Conception.” In Francis G. Castles and Rudolf Wildenmann, eds., Visions and Realities of Party Government. Berlin: Walter de Gruyter. ———. 1997. Democracy and Elections. New York: Oxford University Press.

172

REFERENCES

Katz, Richard S., and Peter Mair. 1993. “The Evolution of Party Organizations in Europe: The Three Faces of Party Organization.” American Review of Politics 14:593–618. Kiernan, Brendan. 1993. The End of Soviet Politics: Elections, Legislatures, and the Demise of the Communist Party. Boulder, Colo.: Westview. Kiewiet, D Roderick, and Mathew D. McCubbins. 1991. The Logic of Delegation: Congressional Parties and the Appropriations Process. Chicago: University of Chicago Press. Kingdon, John W. 1973. Congressmen’s Voting Decisions. New York: Harper and Row. Knight, Jack. 1992. Institutions and Social Conflict. Cambridge: Cambridge University Press. Krehbiel, Keith. 1991. Information and Legislative Organization. Ann Arbor: University of Michigan Press. ———. 1998. Pivotal Politics: A Theory of U.S. Lawmaking. Chicago: University of Chicago Press. ———. 1999. “Paradoxes of Parties in Congress.” Legislative Studies Quarterly 24 (1): 31–64. Lancaster, Thomas D. 1986. “Electoral Structures and Pork Barrel Politics.” International Political Science Review 7:67–81. Lancaster, Thomas D., and W. David Patterson. 1990. “Comparative Pork Barrel Politics: Perceptions from the West German Bundestag.” Comparative Political Studies 22:458–77. Laver, Michael. 1999. “Divided Parties, Divided Government.” Legislative Studies Quarterly 24 (1): 5–29. Lawson, Stephanie. 1993. “Conceptual Issues in the Comparative Study of Regime Change and Democratization.” Comparative Politics 25:183–206. Lijphart, Arend. 1990. “The Political Consequences of Electoral Laws, 1945–85.” American Political Science Review 84:481–96. ———. 1992. “Democratization and Constitutional Choices in Czecho-slovakia, Hungary, and Poland, 1989–91.” Journal of Theoretical Politics 4:207–23. ———. 1994. Electoral Systems and Party Systems: A Study of Twenty-seven Democracies, 1945–1990. New York: Oxford University Press. ———. 1984. Democracies: Patterns of Majoritarian and Consensus Government in Twenty-one Countries. New Haven: Yale University Press. Lijphart, Arend, and Bernard Grofman, eds. 1984. Choosing an Electoral System: Issues and Alternatives. New York: Praeger. Linz, Juan J. 1994. “Presidential or Parliamentary Democracy: Does It Make a Difference?” In Juan J. Linz and Arturo Valenzuela, eds., The Failure of Presidential Democracy. Baltimore: Johns Hopkins University Press. Linz, Juan J., and Alfred Stepan. 1996. Problems of Democratic Transition and Consolidation: Southern Europe, South America, and Post-Communist Europe. Baltimore: Johns Hopkins University Press. Lipset, S. M., and Stein Rokkan. 1967. “Cleavage Structures, Party Systems, and Voter Alignments: An Introduction.” In Seymour Martin Lipset and Stein Rokkan, eds., Party Systems and Voter Alignments. New York: Free Press.

REFERENCES

173

Loewenberg, Gerhard, and Thomas C. Mans. 1988. “Individual and Structural Influences on the Perception of Legislative Norms in Three European Parliaments.” American Journal of Political Science 32 (1): 155–77. Loewenberg, Gerhard, and Samuel Patterson. 1979. Comparing Legislatures. Boston: Little, Brown. Lo¨wenhardt, John. 1995. The Reincarnation of Russia: Struggling with the Legacy of Communism, 1990–1994. Durham, N.C.: Duke University Press. MacRae, Duncan. 1958. Dimensions of Congressional Voting. Berkeley: University of California Press. Mainwaring, Scott. 1993. “Presidentialism, Multipartyism, and Democracy.” Comparative Political Studies 26:198–228. Mainwaring, Scott, and Timothy R. Scully. 1995, “Introduction: Party Systems in Latin America.” In Scott Mainwaring and Timothy R. Scully, eds., Building Democratic Institutions: Party Systems in Latin America, 1–34. Stanford: Stanford University Press. Mayhew, David R. 1974. Congress: The Electoral Connection. New Haven: Yale University Press. McAllister, Ian, and Stephen White. 1995. “Democracy, Political Parties, and Party Formation in Postcommunist Russia.” Party Politics 1:49–72. McFaul, Michael. 1993. “Russian Centrism and Revolutionary Transitions.” PostSoviet Affairs 9 (3): 196–222. McFaul, Michael, and Sergei Markov. 1993. The Troubled Birth of Russian Democracy: Parties, Personalities, and Programs. Stanford: Hoover Institution. McFaul, Michael, and Nikolai Petrov, eds. 1995. Previewing Russia’s 1995 Parliamentary Elections. Moscow: Carnegie Moscow Center. McGregor, James. 1993. “How Electoral Laws Shape Eastern Europe’s Parliaments.” RFE/RL Research Report 2 (4): 11–18. Mezey, Michael L. 1979. Comparative Legislatures. Durham, N.C.: Duke University Press. Minagawa, Shugo. 1985. Supreme Soviet Organs. Nagoya, Japan: University of Nagoya Press. Moe, Terry M. 1990. “Political Institutions: The Neglected Side of the Story.” Journal of Law, Economics, and Organization 6:213–53. Montgomery, Kathleen. 1995. “Crafting Representation in a New Democracy: The Case of Hungary.” Ph.D. dissertation, Emory University, Atlanta, Georgia. Nohlen, Dieter. 1984. “Changes and Choices in Electoral Systems.” In Arend Lijphart and Bernard Grofman, eds., Choosing an Electoral System: Issues and Alternatives, 217–24. New York: Praeger. North, Douglass C. 1990. Institutions, Institutional Change, and Economic Performance. New York: Cambridge University Press. Olson, David M. 1993. “Political Parties and Party Systems in Regime Transformation: Inner Transition in the New Democracies of Central Europe.” American Review of Politics 14:619–58. Patzelt, Werner J. 1993. Abgeordnete und Repra¨sentation: Amtsversta¨ndnis und Wahlkreisarbeit. Passau: Wissenschaftsverlag Rothe.

174

REFERENCES

Patzelt, Werner J. 1994. “Legislators in New Parliaments: The Case of East Germany.” In Attila Agh, ed., Emergence of East Central European Parliaments: The First Steps, 270–88. Budapest: Hungarian Center of Democracy Studies. Powell Jr., G. Bingham. 1989. “Constitutional Design and Citizen Electoral Control.” Journal of Theoretical Politics 1 (2): 107–30. Przeworski, Adam. 1991. Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America. New York: Cambridge University Press. Przeworski, Adam, et al. 1995. Sustainable Democracy. New York: Cambridge University Press. Rae, Douglas W. 1967. The Political Consequences of Electoral Laws. New Haven: Yale University Press. ———. The Political Consequences of Electoral Laws. Rev. ed. New Haven: Yale University Press. Remington, Thomas F. 1996. “Menage a` trois: The End of Soviet Parliamentarism.” In Jeffrey W. Hahn, ed., Democratization in Russia: The Development of Legislative Institutions, 106–39.. Armonk, N.Y.: M. E. Sharpe. ———. Forthcoming. Political Representation in a Transitional Regime: The Evolution of Parliamentary Institutions in Russia, 1989–1999. New Haven: Yale University Press. Remington, Thomas F., and Steven S. Smith. 1995. “The Development of Parliamentary Parties in Russia.” Legislative Studies Quarterly 20:457–89. ———. 1996. “Political Goals, Institutional Context, and the Choice of an Electoral System: The Russian Parliamentary Election Law.” American Journal of Political Science 40:1253–79. ———. 1998. “Theories of Legislative Institutions and the Organization of the Russian Duma.” American Journal of Political Science 42:545–72. Remington, Thomas F., Steven S. Smith, and Moshe Haspel. 1998. “Electoral Institutions and Party Cohesion in the Russian Duma.” Journal of Politics 60:417–39. Remington, Thomas F., Steven S. Smith, D. Roderick Kiewiet, and Moshe Haspel. 1994. “Transitional Institutions and Parliamentary Alignments in Russia, 1990– 1993.” In Thomas F. Remington, ed., Parliaments in Transition: The New Legislative Politics in the Former USSR and Eastern Europe, 159–80. Boulder, Colo.: Westview. Riker, William H. 1980. “Implications of the Disequilibrium of Majority Rule for the Study of Institutions.” American Political Science Review 74: 432–46. ———. 1995. “The Experience of Creating Institutions: The Framing of the United States Constitution.” In Jack Knight and Itai Sened, eds., Explaining Social Institutions, 121–44. Ann Arbor: University of Michigan Press. Roeder, Philip G. 1994. “Varieties of Post-Soviet Authoritarian Regimes.” PostSoviet Affairs 10:61–101. Rohde, David W. 1991. Parties and Leaders in the Postreform House. Chicago: University of Chicago Press. Saalfeld, Thomas. 1990. “The West Germany Bundestag after 40 Years.” In Philip Norton, ed., Parliaments in Western Europe, 68–89. London: Frank Cass. Sakwa, Richard. 1995. “The Development of the Russian Party System: Did the Elections Change Anything?” In Peter Lentini, ed., Elections and Political Order

REFERENCES

175

in Russia: the Implications of the 1993 Elections to the Federal Assembly, 169–201. Budapest: Central European University Press. ———. 1996. Russian Politics and Society. London: Routledge. Sartori, Giovanni. 1976. Parties and Party Systems: A Framework for Analysis. Vol. 1. Cambridge: Cambridge University Press. Schu¨ttemeyer, Suzanne S. 1994. “Hierarchy and Efficiency in the Bundestag: The German Answer for Institutionalizing Parliament.” In Gary W. Copeland and Samuel C. Patterson, eds., Parliaments in the Modern World: Changing Institutions, 29–58. Ann Arbor: University of Michigan Press. Shepsle, Kenneth A., and Mark S. Bonchek. 1997. Analyzing Politics: Rationality, Behavior, and Institutions. New York: W. W. Norton. Shepsle, Kenneth A., and Barry R. Weingast. 1995. “Positive Theories of Congressional Institutions.” In Kenneth A. Shepsle and Barry R. Weingast, eds., Positive Theories of Congressional Institutions, 5–35. Ann Arbor: University of Michigan Press. Shugart, Matthew S., and John M. Carey. 1992. Presidents and Assemblies: Constitutional Design and Electoral Dynamics. Cambridge: Cambridge University Press. Slider, Darrell, Vladimir Gimpel’son, and Sergei Chugrov. 1994. “Political Tendencies in Russia’s Regions: Evidence from the 1993 Parliamentary Elections.” Slavic Review 53:711–32. Smith, Steven S., and Christopher J. Deering. 1990. Committees in Congress. Washington, D.C.: CQ Press. Sobyanin, Alexander. 1994. “V korolevstve krivykh zerkal.” Segodnia (10 March). Sobyanin, Alexander, and V. Sukhovol’skii. 1993/1994. “Itogi vyborov 12-ogo dekabria 1993 goda i budushchii federal’nyi zakon o vyborakh.” Konstitutsionnoe pravo: vostochnoevropeiskoe obozrenie 2, no. 4; 3, no. 1: 2–10. Sobyanin, Alexander, E. Gel’man, and O. Kaiunov. 1994. “The Political Climate of Russia’s Regions: Voters and Deputies, 1991–1993.” The Soviet and Post-Soviet Review 21:63–84. Stepan, Alfred, and Cindy Skach. 1993. “Constitutional Frameworks and Democratic Consolidation: Parliamentarism versus Presidentialism.” World Politics 46 (1): 1–22. Strom, Kaare. 1990. Minority Government and Majority Rule. Cambridge: Cambridge University Press. Suleiman, Ezra N., ed. 1986. Parliaments and Parliamentarians in Democratic Politics. New York: Holmes and Meier. Sullivan, John L., and Eric M. Uslaner. 1978. “Congressional Behavior and Electoral Marginality.” American Journal of Political Science. 22:536–54. Taagepera, Rein, and Matthew S. Shugart. 1989. Seats and Votes: The Effects and Determinants of Electoral Systems. New Haven: Yale University Press. Tolz, Vera. 1990. The USSR’s Emerging Multiparty System. Washington, D.C.: Center for Strategic and International Studies. Tsebelis, George. 1990. Nested Games: Rational Choice in Comparative Politics. Berkeley: University of California Press. Urban, Michael. 1994. “December 1993 as a Replication of Late-Soviet Electoral Practices.” Post-Soviet Affairs 10:127–58.

176

REFERENCES

Von Beyme, Klaus. 1986. “The Role of Deputies in West Germany.” In Ezra N. Suleiman, ed., In Parliaments and Parliamentarians in Democratic Politics. New York: Holmes and Meier. Weingast, Barry R., and William Marshall. 1988. “The Industrial Organization of Congress.” Journal of Political Economy 96:132–63. White, Stephen, Richard Rose, and Ian McAllister. 1997. How Russia Votes. Chatham, N.J.: Chatham House. Whitefield, Stephen, and Geoffrey Evans. 1994. “The Russian Election of 1993: Public Opinion and the Transition Experience.” Post-Soviet Affairs 10:38–60. Wyman, Matthew, Bill Miller, Stephen White, and Paul Heywood. 1994. “The Russian Elections of December 1993.” Electoral Studies 13:254–71.

INDEX Aberbach, Joel, 120 Agh, Attila, 119 Agrarian Party of Russia (APR), 30, 31, 32, 37, 38, 39, 41, 43–50, 63, 64, 66, 70, 75, 76, 78, 80–85, 88, 89, 98, 100–105, 107– 10, 112, 114nn. 5 and 8, 123, 125, 126, 127, 129–131, 135n.5, 142, 143, 147, 148, 160n.1, 161 Agro-Industrial Group (AIG), 149, 150, 151, 153 AIG. See Agro-Industrial Group Aldrich, John, 51n.1, 134, 134n. 1 APR. See Agrarian Party of Russia Bach, Stanley, 11 Bauer, Vladimir, 44, 86 Bawn, Kathleen, 15, 93, 94, 95, 100, 119 Belonuchkin, G. V., 114n.2, 168 Binder, Sarah, 51n.1 Bloc “Yavlinsky-Boldyrev-Lukin” (Yabloko), See Yabloko Bonchek, Mark, 26n.3 Bowler, Shaun, 117, 118 Brady, David, 134n.1 BRNI. See Future of Russia/New Names Brody, Richard, 134n.1 Brudny, Yitzhak, 28 Bullock, Charles, 7, 134n.1 Carey, John, 7, 52n.6, 118, 119, 121, 134 chairman (speaker) of the Duma, 64–65 Chiesa, Giulietto, 28 Chugrov, Sergei, 114n.4 Clausen, Aage, 26n.7 Clem, Ralph, 114n.4 Coase, Ronald, 10 Coleman, James, 16 Collie, Melissa, 134n.1 Colton, Timothy, 25n.1, 28, 30, 121 committee chairs, 17, 33, 55, 56, 62, 64, 65, 70, 77, 81, 89–91, 144, 145, 147, 151 committees, standing, 25, 72, 73, 79, 137–38 Communist Party of the Russian Federation (CPRF), 30, 32, 37, 38, 39, 40, 41, 42, 44, 45, 48, 49, 64, 68–70, 75, 76, 78, 80, 82, 84, 85, 87, 88, 92n.3, 98, 100– 105, 107–9, 114nn. 5 and 8, 123, 125–31,

135 nn. 4 and 5, 140, 143, 147, 149–52, 161 Communist Party of the Soviet Union (CPSU), 5 Congress of People’s Deputies, Russian, Soviet Federated Socialist Republic, 4, 5, 28, 55 Congress of People’s Deputies, Union of Soviet Socialist Republics, 4 Congress, United States, 16, 19, 21–22, 26n.8, 116, 117, 123, 134n.1, 153 Constitution, Russian, 6–8, 37, 51n.4, 55, 121 Constructive-Ecological Movement “Cedar” (KEDR), 31, 32 Cooper, Joseph, 134n.1 Council of the Duma, 8, 11, 17, 24–25, 44, 45, 47, 51n.4, 54, 56, 57, 62, 64, 65, 70, 71, 73, 89–91, 114n.9, 122, 139, 144–47, 153 Cox, Gary, 15, 19, 51n.1, 52n.6, 118, 134– 35n.1, 136n.6, 154, 156 CPRF. See Communist Party of the Russian Federation CPSU. See Communist Party of the Soviet Union Craumer, Peter, 114n.4 Deckard, Barbara, 120 Deering, Christopher, 74 Democratic Party of Russia (DPR), 31, 32, 39, 41- 43, 45, 46, 48–50, 78, 81-84, 86, 101–3, 106–9, 112, 124–27, 130, 131, 161 deputies, party-list, 18–19, 27, 30, 31, 35– 37, 39, 42, 47, 48, 50, 58, 62, 65, 67, 69, 70, 71 nn. 2 and 3, 74, 76, 77, 78, 80, 83, 86, 90, 91, 95, 96, 100, 102, 103, 107, 110, 111, 118, 120, 121, 124, 126, 132, 139, 140, 158, 161, 162 deputies, single-member district, 18–19, 25, 27, 28, 29, 31, 33, 35–37, 39, 42, 45, 48, 50, 51n.3, 52n.7, 58, 62, 65, 67–71, 71n.2, 74, 76–78, 80, 83, 86, 90, 91, 93, 94, 96, 98–100, 102, 104, 106, 107, 108, 110–13, 117, 118, 120, 122, 124, 126, 128, 132, 139, 140, 158, 161, 166 deputy groups, 33, 122 Diermeier, Daniel, 13

178 Dignity and Mercy (DM), 32 DM. See Dignity and Mercy Dodd, Lawrence, 154 Downs, Anthony, 52n.6 DPR. See Democratic Party of Russia Duverger, Maurice, 52n.6, 98 efficient change, 12, 14, 134 Eggertson, Thrainn, 26n.4 elections, parliamentary, 1993: 29–30, 51n.4, 55, 74, 95–99, 105 election law, 27, 29, 34, 58, 83, 84, 86, 105, 107, 108, 113, 115n.9; nominations, 18, 20, 30, 34, 35 Epstein, Leon, 52n.6, 119, 134n.1 Evans, Geoffrey, 114n.4 factions, 17, 20, 24, 27, 28–30, 32, 33, 35– 47, 47–50, 54, 56, 57, 59, 60, 64, 67–70, 72–75, 80, 83, 85, 86, 90, 91, 95, 105, 107, 117, 121–24, 128, 129, 132–34, 139, 144, 146, 147, 153, 167 faction size. See threshold, faction Farrell, David, 117, 118 Fatherland-All Russia (OVR), 150–152 Federal Assembly, 6, 60 Federation Council, 7, 97 Fenno, Richard, 21, 26 nn. 7 and 8 Fiorina, Morris, 120 Fish, Stephen, 28 fraktsiia, 32. See also parties; factions Future of Russia/New Names (BRNI), 32, 101, 161 Gaidar, Egor, 5, 39, 78 Geddes, Barbara, 15, 21 Gel’man, E., 114n.4 Germany, 93–95, 119–20, 135n.3 Gimpel’son, Vladimir, 114n.4 goals, general, 3, 9–10, 15–16, 20–23, 91, 159 goals, deputies’: partisan, 24, 95, 158; policy, 15, 16, 19, 21, 23–24, 38, 157–58, 160; reelection, 16, 19, 23–24, 59, 157– 58, 160 Gorbachev, Mikhail, 4–5, 9, 72, 134n.4 Grofman, Bernard, 52n.6, 119 Gunther, Richard, 107 Haggard, Stephan, 12 Haspel, Moshe, 37, 40, 48, 52 nn. 8 and 10, 76 Hellman, Joel, 12 Herzog, Dietrich, 120

INDEX

Hibbing, John, 93 Hough, Jerry, 25n.1, 28, 30, 114n.4, 120 Huber, John, 52n.6, 118 Hungary, 93–94 imperative mandate, 35, 47–49, 86, 136n.8, 139–41 Inglehart, Ronald, 52n.6 institutional change, 4, 8–10, 12 institutional preferences, 14, 19, 69, 70, 73, 141 Iverson, Torben, 52n.6 Jasiewicz, Krzysztof, 93 J-curve, 12–13, 26n.6 Johnson, Simon, 13 Kaiunov, O., 114n.4 Katz, Richard, 52n.6, 117, 118, 134 Kaufmann, Robert, 12 KEDR. See Constructive-Ecological Movement “Cedar” Khasbulatov, Rulsan, 6 Kiernan, Brendan, 28 Kiewiet, Roderick, 51n.1, 134, 134n.1, 136n.6 Kingdon, John, 26n.7, 116 Knight, Jack, 12 Kouvelis, Danos, 13 Krehbiel, Keith, 17, 51n.1, 154, 157 Lancaster, Thomas, 126, 134n.2 Laver, Michael, 19, 51n.2 Lawson, Stephanie, 134 LDPR. See Liberal-Democratic Party of Russia LDUD12. See Liberal-Democratic Union of December 12 Liberal-Democratic Party of Russia (LDPR), 30, 31, 32, 37, 38, 39, 41, 43– 50, 63, 64, 66, 68, 70, 76, 78, 80–85, 87– 89, 100–105, 107–9, 112, 114n.8, 115n.11, 123–127, 129, 130, 131, 135n.5, 142, 143, 147, 149, 151, 161 Liberal-Democratic Union of December 12 (LDUD12), 31, 32, 41, 43–46, 49, 50, 52n.9, 63, 81, 82, 85, 87–89, 101–3, 122 Lijphart, Arend, 15, 52n.6, 93, 119 Lipset, S. M., 52n.6, 134–35n.1 Loewenberg, Gerhard, 52n.6, 118 Lo¨wenhardt, John, 25n.1 MacRae, Duncan, 26n.7 Mainwaring, Scott, 52n.6, 118, 134

INDEX

Mair, Peter, 118 Mans, Thomas, 118 Markov, Sergei, 28 Marshall, William, 154, 157 Mayhew, David, 18 McAllister, Ian, 25n.1, 30, 121 McCubbins, Mathew, 19, 134n.1, 136n.6, 154, 156 McFaul, Michael, 28, 121 McGregor, James, 52n.5 Minagawa, Shugo, 55 mode of election, 19, 35–39, 43, 45, 47, 48, 50, 53n.12, 58, 61, 63, 66, 67, 69, 70, 78, 80, 81, 89, 110, 112, 116, 123, 124, 126, 128, 129, 130, 132, 133, 138, 140–41 Moe, Terry, 12, 15 Montgomery, Kathleen, 119, 134n.3 Narod. See People’s Power NDR. See Our Home Is Russia Neto, Amorim-Octavio, 15 new economics of institutions, 10, 14, 15 new institutionalism, 1–24 New Regional Policy (NRP), 31, 32, 41, 43, 45, 46, 49, 50, 52n.9, 63, 69, 81–85, 87, 88, 101–3, 105–7, 109, 112, 122, 124–27, 129, 130, 136n.7, 165 Nohlen, Dieter, 52n.6, 93 nonvoting, by deputies, 49, 165 North, Douglass C., 9 Northrup, Douglas, 28 NRP. See New Regional Policy Olson, David, 93 OVR. See Fatherland-All Russia Our Home Is Russia (NDR), 60, 68, 69, 70, 142, 143, 145–48, 160n.6, 162 package agreement, 78–80, 86, 87, 89, 90, 91, 92n.3, 144 parties, 8, 18, 20, 24, 25, 27, 28–35, 37, 39– 40, 51n.3, 52n.5, 58, 59, 65, 67–69, 103, 107, 121, 122. See also factions Party of Russian Unity and Accord (PRES), 30, 31, 32, 39, 40, 41, 43, 45, 46, 49, 50, 63, 78, 81–84, 88, 89, 100, 101, 103, 104, 107–9, 112, 125–27, 139, 131, 161 party voting, 123–26, 128–31 Patterson, Samuel, 52n.6, 93, 120 Patzelt, Werner, 120 PDep. See People’s Deputy People’s Deputy (PDep), 149, 150, 151, 153 People’s Power (Narod), 64, 66, 70, 142, 143, 148

179 Petrov, Nikolai, 121 policy positions, 35, 38, 42, 45–48, 50, 61– 64, 66, 69, 81, 88–90, 123, 126, 128–30, 158 Poltoranin, Mikhail, 86 positive political theory, 15 Powell, Bingham, 134 PRES. See Party of Russian Unity and Accord presidential commission on organizing parliament, 1993; 31, 32, 33, 40, 56, 77, 92n.1 Presidium, 24, 33, 54–57, 60, 62, 64, 65, 68–71, 90, 116, 139 proportional representation, 7, 29–31, 34, 51n.4, 52n.7, 91, 93, 95, 96, 98–100, 102, 103, 105–8, 112, 113, 115n.9, 119 Przeworski, Adam, 12, 26n.6 Putnam, Robert, 120 Rae, Douglas, 52n.6, 119 RC. See Russia’s Choice redistributional change, 12, 14, 15, 18 Remington, Thomas F., 15, 28, 29, 34, 37, 56, 75, 119, 122, 123, 138, 166 Riker, William, 17 RMDR. See Russian Movement for Democratic Reforms Rockman, Bert, 120 Rohde, David, 51n.1, 134n.1, 154 Rokhlin, Lev, 145, 146 ROS. See Rossiia Rose, Richard, 25n.1, 30, 121 Rossiia (ROS), 109, 112, 122, 125, 126, 130, 134n.1 RReg. See Russia’s Regions rules, 8–11, 22, 27, 29, 31, 33, 39, 44, 48, 56, 60, 80, 90 Russian Movement for Democratic Reforms (RMDR), 32, 101, 104, 161 Russia’s Choice (RC), 31, 32, 39-41, 43–50, 63, 68, 75, 76, 78, 80–86, 88–90, 92n.3, 100, 101, 103, 104, 107–9, 112, 114n.4, 125, 127, 129–31, 161, 165 Russia’s Regions (RReg), 66, 70, 143, 144, 147–51, 153 Russia’s Way (RW), 31, 32, 41, 43, 45, 46, 49, 50, 52n.9, 81, 82, 101, 105 RW. See Russia’s Way Rybkin, Ivan, 33, 76, 148, 162 Saalfeld, Thomas, 134n.3 Sakwa, Richard, 25n.1, 121 Sartori, Giovanni, 52n.6, 119, 134–35n.1

180 Schu¨ttemeyer, Suzanne, 134n.3 Scully, Timothy, 134 Seleznev, Gennadii, 65, 68, 144, 148, 151, 152 Shabad, Anatoli, 40 Shepsle, Kenneth, 19, 26n.3, 154, 155, 157, 158 Shugart, Matthew, 7, 52n.6, 118, 119, 121, 134 Sinha, Vikas, 13 Skach, Cindy, 52n.6 Slider, Darrell, 114n.4 Smith, Steven S., 15, 34, 37, 74, 119, 121, 123, 166 Sobyanin, Alexander, 28, 96, 114n.4 social choice, 14, 15 speaker of the Duma. See chairman of the Duma SPS. See Union of Rightist Forces Stability (STB), 109, 112, 122, 125–27, 130, 165 standing orders, 44, 90, 144, 146, 147, 164 STB. See Stability Stepan, Alfred, 52n.6 Strom, Kaare, 26n.8 Suleiman, Ezra, 134–35n.1 Sullivan, John, 120 Supreme Soviet, Russian Federated Soviet Socialist Republic, 4, 5, 28, 32, 33 Supreme Soviet, Union Soviet Socialist Republics, 5, 55, 72 survey, candidates for the Duma, 1993: 30, 60, 61, 69, 161 survey, candidates for the Duma, 1995: 60, 61, 123, 162 survey, Duma deputies, 1994: 60, 71n.2, 87, 89, 147 survey, Duma deputies, 1996: 60, 71n.2, 163

INDEX

Taagepera, Rein, 52n.6, 119 threshold, faction, 7, 29, 30, 32, 33, 34, 35, 36–37, 40, 41, 43–46, 49, 50, 51n.3, 65, 67, 68, 96, 106, 120, 124, 139, 150, 142, 167. See also factions Tolz, Vera, 28 transaction costs, 10–11, 22, 158 Tsebelis, George, 11–12, 14, 23 Union of Rightist Forces (SPS), 150, 151 Urban, Michael, 96 Uslaner, Eric, 120 von Beyme, Klaus, 120 Weingast, Barry, 19, 154, 155, 157, 158 White, Stephen, 25n.1, 30, 121 Whitefield, Stephen, 114n.4 Women of Russia (WR), 32, 39, 41, 42, 43, 45, 46, 49, 50, 81, 82, 84, 101–3, 106, 109, 110, 112, 125, 127, 130, 131, 161, 162 WR. See Women of Russia Wyman, Matthew, 114n.4 Yabloko, 30, 31, 32, 39, 41, 43–46, 49–50, 51n.3, 63, 64, 66, 68–70, 78, 81–85, 87– 90, 100, 101, 103, 104, 107–9, 112, 123, 125–27, 130, 131, 135n.5, 142, 143, 147, 148, 150, 151, 160n.1, 161 Yavlinsky, Gregory, 30 Yeltsin, Boris, 5–7, 9, 26n.2, 29, 31, 33, 37, 51n.4, 52n.5, 54, 64, 77, 94, 96–99, 106, 134n.4, 145 Zavadskaia, L. N., 110 Zhirinovsky, Vladimir, 30, 31, 39, 40, 44, 47, 48, 64, 68, 75, 76, 79, 91, 104, 123, 149