Electoral Cycles and Party Switching: Opportunistic ...

Timing Matters: Incentives for Party Switching and Stages of Parliamentary Cycles Carol Mershon Department of Politics University of Virginia Charlottesville, VA 22904-4787 USA tel: 434-924-7875 fax: 434-924-3359 e-mail: [email protected] Olga Shvetsova Department of Political Science Binghamton University Binghamton, NY 13902-6000 tel: 607-777-4230 fax: 607-777-2675 e-mail: [email protected]

July 2008

Chapter 8 in William B. Heller and Carol Mershon, eds. Political Parties and Legislative Party Switching. Under contract, Palgrave Macmillan.

We thank Nina Barzachka, Rado Iliev, and Adriana Stojan for research assistance, and Will Heller and other members of the Party Switching Research Group for rich discussions of party and legislative politics. We are also grateful to John Aldrich, Jim Caporaso, Mikhail Filippov, Tim Nokken, Lucio Renno, and Christian Stecker for their helpful comments.

0

Timing Matters: Incentives for Party Switching and Stages of Parliamentary Cycles 1

Introduction Virtually all political scientists maintain that democratic political life depends on political

parties. Precisely because parties are fundamental to democracy, we and the other contributors to this book probe changes of party affiliation among elected politicians. This chapter is distinctive in that it links the phenomenon of party switching to the stage in the parliamentary cycle. Our goal here is two-fold. On the one hand, our research is the first to explore whether and how patterns in switching behavior differ systematically across distinct stages of the legislative term—for instance, one stage devoted to committee assignments, early in the term, and another at the end of the term, dominated by the view of elections on the horizon (Mershon and Shvetsova 2005, 2008a, 2008b; Shvetsova 2004; cf. Desposato, this volume). On the other hand, linking changes of party affiliation to the time when benefits are available provides direct confirmation for the main theoretical premise shared by all of this volume’s contributors, namely, that politicians switch parties for opportunistic reasons. In a nutshell, we seek to gain leverage on the question of why legislators switch by asking when they switch. We examine politicians’ choices and changes of party labels during the legislative term, in different stages of the parliamentary cycle. As we define it, the parliamentary cycle includes legislative stages and also the electoral stage that occurs before or during the official campaign for the next legislature and that, assuming backwards induction, affects behavior in the term. The next section elaborates on this conception of the parliamentary cycle and hypothesizes that the cycle leaves its imprint on switching behavior. Third, we outline our research design. Fourth, we assess our hypotheses against data from our two primary country-terms, the 1996-2001 Italian Chamber of Deputies and the 1993-1995 Russian Duma, and a total of 35 subsidiary country1

Comment [CAM1]: Will & Carol check if this is too close to Intro or Ch 2).

terms in Italy, Romania, the UK, and the US. The last part draws out the implications of our study.

2

The Parliamentary Cycle: Stages, Motives, and Behavior Like other contributors to this book, we assume that legislators switch parties in strategic

efforts to attain desired goals. A distinguished tradition holds that parties and politicians are motivated by office, policy, and votes (e.g., Strøm 1990; Müller and Strøm 1999). Hence we assume that members of parliament (MPs) may switch partisan affiliation in order to obtain parliamentary offices and privileges, achieve preferred policy outcomes, or position for reelection. All of these motives may prove relevant to some degree (cf. Strøm 1990) in each decision to change a party, and thus none should be ignored. Yet we add that their relative salience differs across specific periods of time, depending on what dominates the parliamentary agenda and hence which payoffs are most prominent, immediate, and available (Mershon and Shvetsova 2005, 2008a). In our approach, individuals can change their choice of parties as they pursue goals that are specific to the stage in the parliamentary cycle. The different stages of the parliamentary cycle hold out different mixes of incentives to MPs, make some incentives more prominent at some times than others, and thus highlight different motivations. Positing variations in the salience of incentives and motivations over time, we distinguish types of switching and expect variations in the frequency of switching according to stage in the parliamentary cycle. Figure 1 identifies the stages within the parliamentary cycle where we expect distinctive rewards to switching and distinctive motivations for switching. We assume that the period of greatest salience of each motivation can be discerned at the outset, on the basis of legislative records, for example. By associating changes of partisanship with the periods when they occur, and by classifying periods according to the type of rewards decided, we appraise the relative 2

impact of the aims of office (including legislative perks), policy, and (re-)election. (Figure 1 about here.) What we call stage A (for Affiliation) marks the transition from the popular vote to the taking up of legislative seats in the first legislative session. In most systems, MPs have the opportunity to alter the label from when they announce (party) group affiliation for the legislative session.1 We posit that MPs are motivated primarily by perks during stage A, for they respond to the availability of goods tied to membership in one parliamentary group or another, as shaped by legislative rules. To be sure, policy motives are not absent during this stage, since office serves as an instrument to affect policy (cf. Laver and Shepsle 1996). Stage B (for Benefits) is when committee seats, committee chairs, and other legislative posts (and, in parliamentary systems, executive portfolios) are allocated. We posit that this division of positions of power heightens office-seeking goals for MPs. Again, policy does not disappear as a concern, for office permits the pursuit of policy. But when offices are up for grabs, MPs’ choices to switch or stay with their original party are driven chiefly by interest in office. In stage C (for Policy Control), the legislative agenda focuses most heavily on policy domains relevant to a broad range of issues. Whereas some analysts, mindful of the executive’s role in policymaking, might paint an entire term as akin to a continuous stage C shaded gray to capture the parliament’s overall contribution to policy decisions, we see a non-continuous sequence of shorter, near-black, phases of concentrated attention to the most important policy domains. Given MPs’ heightened focus on policy in this non-continuous sequence of phases, switching in stage C should occur so as to shape policy choices and secure agenda control. Stage E (for Election) closes the cycle in Figure 1.2 In this stage, electoral motivations

Comment [CAM2]: Does last sentence of note 2 sit oddly with last sentence of footnote 1? Will, what is your perspective? (I’m probably beyond being able to see the issue clearly. Olga says: no problem, take it out!)

should come to the fore and switching should chiefly aim at pre-electoral positioning. Figure 1

3

simplifies reality, and in some institutional contexts stage E may recur during the cycle, as nonparliamentary elections are held during the term. Although their own seats are not immediately at stake, MPs learn from opinion polls, which proliferate before elections, and from voting returns, which reveal whether a switch is likely to improve their re-election chances. One stage, D (for Dormant), does not appear in Figure 1. Stage D is simply all periods other than stages A, B, C, and E—a residual set of intervals between the active stages. We expect no distinctive switching by politicians driven by office, policy, or re-election—that is, we expect no switching—to occur during D, since it is more advantageous to switch in active stages. All stages except D can overlap or coincide. Given the general premise that MPs are goal-minded, the general expectation is that they switch when their goals are best served by such action. The expectation that timing matters leads to more specific testable hypotheses. 

Hypothesis 1: The rate of switching should vary over the course of the parliamentary cycle.

Switching should be relatively frequent in stages A, B, C and E. 

Hypothesis 2: Attributes of switches should vary by stage. Office-driven switching should

predominate in stages A and B, policy-driven switching should predominate in stage C, and votedriven switching, in stage E. Thus, independents should switch in stages A and B. MPs who, by virtue of seniority or other distinction, are most able to strongly influence policy should switch, Comment [CAM3]: Will, Carol is not quite satisfied with this; if you can think of something, thanks!

if they switch, in stage C. MPs who expect switching to serve as a boon in the next election should move in stage E.3 

Hypothesis 3: Stage C switches should aim for clear policy effect. Those switches identifiable

by timing as policy-driven should seek the center of the policy space, grab agenda advantage, or break the government. Otherwise, the switches would not change policy outcomes.

4



Corollary to Hypothesis 3: New parliamentary groups formed in stage C should locate at the

center of the policy space. Assuming policy goals, political entrepreneurs should found new groups in the most advantageous position in policy space: the center (cf., e.g., McKelvey and Schofield 1987). 

Hypothesis 0: Null. No or very little variation between stage D and the active stages should

emerge in rates of switching and attributes of switches.

3

Research Design and Methods Hypotheses in hand, we address our strategy for testing them. First, like others in this

field, we define a switch as “any recorded change in party affiliation on the part of a politician holding or competing for elective office” (Heller and Mershon 2005b, 4; cf. Introduction, this volume). Second, we choose 1996-2001 Italy and 1993-1995 Russia as primary country-term cases, thus affording a rare comparison of two very different countries with strikingly similar electoral institutions. Although no two electoral systems are ever fully identical, for the terms we study in depth Italy and Russia both used hybrid laws, combining proportional representation (PR), thresholds for PR, and plurality in single-member districts (SMDs). Other institutional similarities included powerful sub-national governments (though Italy is not federal) and nonconcurrence between parliamentary and other major elections.4 These similarities are essential to discerning the imprint of timing in the parliamentary cycle on patterns of switching, as formal institutions should strongly affect political actors’ choices in partisan competition. Given the institutional parallels, we would be able to reject our hypotheses with confidence if differences in switching behavior appeared (unless some feature of rules differed too). Similarities in switching would equally clearly corroborate our reasoning. Third, we deliberately select for intensive study the legislative term in Italy and in Russia 5

that qualifies as that term with the highest number of switches for any term to date. Specifically, about one-fourth of Italian MPs switched parties at least once during the 1996-2001 term, with the total number of switches in the 630-member Chamber standing at 277. Almost one-third of the MPs in the Russian Duma switched at least once from 1993 to 1995, and switches totaled 342 for the 450-member house. Critics might argue that this choice is problematic, since in the first post-Communist Duma legislators might have still been learning to read and respond to institutional incentives (Kaminski 2002; Kunicová and Remington 2005). One might also imagine that the Italian deputies under scrutiny were still adapting to a new institutional and partisan environment, especially since the share of neophytes in the 1996-2001 Chamber was relatively high (Verzichelli 1996; Zucchini 2001, 172). Yet we see these possibilities as sources of strength in our design: If we find that, even during the relatively uncertain first Duma and 1996-2001 Chamber, switching varies according to type of incentive dominant in distinct stages, then we are likely to find elsewhere that switching varies by stage in the parliamentary cycle. Whereas practical concerns lead us to restrict the in-depth inquiry to strategically chosen terms,5 we extend the inquiry to 1963-72 Italy, Romania, the United Kingdom, and the United States so as to introduce greater variation in electoral laws, executive-legislative relations, and territorial dispersion of power. Whereas the UK and the US use plurality electoral laws, Romania uses closed-list PR and Italy before 1993 used open-list PR for elections to the lower house. The two SMD systems differ in regime type, as do the two PR systems. Table 1 also shows variations across countries and across time in powers allocated to sub-national governments. [Table 1 about here.] Moving beyond the parameters highlighted in Table 1, the five countries present variation in regime longevity, institutionalization of the party system, and rules on legislative term length.

6

To adopt Huntington’s (1991) well-known scheme, the US and UK belong to the first wave of democratization, and post-fascist Italy to the second wave. Romania and Russia democratized after the third wave had started in the mid-1970s. Reflecting the timing of democratization, the Romanian and Russian party systems are the most weakly institutionalized of those studied.6 For the spans covered, the UK and US party systems are the most strongly institutionalized, and the Italian evinces variation, with great institutionalization in 1963-1972 and diminished institutionalization in 1996-2001. Although the US is the only country here with fixed terms, Romania has not yet seen an early parliamentary dissolution. Finally, central to our approach to demarcating the boundaries of stages in the parliamentary cycle is the idea that the time of greatest salience of each motivation that induces switching can be readily identified according to the parliamentary record, institutional prescriptions, and events forcing decisions on a key policy dimension. Table 2 illustrates how we move from the general principle to specific definitions of stages. For our primary country-terms, we identify stage A (Affiliation) as the interval between election day and the last day that MPs have to announce parliamentary group membership; and we locate stage B (Benefits) from the last day of announcement of group membership to the day that the distribution of all legislative offices (and, in the Italian parliamentary system, executive portfolios) is completed. Stage B can cover multiple sub-stages in a single term, involving, e.g., multiple cabinets. For primary cases only, we decompose Stage C (Control of policy) into multiple stretches of time so as to isolate the sub-stages of the most active, controversial policy bargaining, focusing on domains with special weight: budgetary questions and, as applicable, security and foreign policy and constitutional matters (cf. Lijphart 1984; Laver and Shepsle 1996).7 We set exogenous criteria for locating the periods of most intense policy bargaining on these dimensions, attending to such

7

steps as the executive’s transmittal of the annual budget bill to the legislature. [Table 2 about here.] Operationalization of stage D (for Dormant) is simple: whatever spans do not qualify as either A or B or C or E become, by default, stage D. Approaches to measuring stage E (Elections) can be several. One set of boundaries would be the start of registration for the ballot for the next legislative elections and election day itself. Yet this approach is inadequate for tapping strategic behavior in anticipation of decisions on lists and candidacies. Given the evident capacity of Italian and Russian MPs to anticipate electoral competition—well in advance of scheduled parliamentary elections—we use a cut-off date to start stage E, specific to practices in each national setting. As the rightmost column of Table 2 indicates, a key distinction between the primary and subsidiary country-term cases regards the degree of detail in coding stages. For subsidiary cases, we simply code the first two months of the term, starting with election day, as stage A. For subsidiary cases, too, the first month that the legislature is officially in session becomes stage B; in parliamentary systems, B includes any month(s) in which cabinet form and fall. Lacking particulars on types or intensity of legislative activity, we treat any month that the legislature is in session as entering Stage C. The six months before the next national legislative elections comprise the simplified stage E in subsidiary cases. For both primary and subsidiary cases, D is a residual stage. With this operationalization of stages in subsidiary cases, we may underestimate D relative to primary cases and more generally may introduce error, which works against generating findings favorable to our hypotheses. Whatever support does appear for our logic is thus reinforced.

4

Empirical Analysis of Switching For the primary cases of the 1996-2001 Italian and 1993-1995 Russian lower houses, we 8

first investigate the frequency of switching by aggregated stage, distinguishing MPs by mode of election, and next disaggregate stages. We then move to the subsidiary terms in Italy, Romania, the UK, and the US. 4.1

Switching by aggregated stage in 1996-2001 Italy and 1993-1995 Russia Table 3 compares switching in Russia and Italy by aggregated stage in the parliamentary

cycle. We standardize the measure of frequency of switching by examining mean weekly switches per 100 MPs. Note first, as the top row of the table displays, that the opening weeks of the term, when newly elected deputies must choose parliamentary group affiliation (stage A), exhibit the highest aggregate rate of switching in both Italy and Russia. And in both hybrid electoral systems, it is above all SMD MPs who choose a parliamentary group different from their electoral label. [Table 3 about here.] Yet in stage A the Russian switching rate is over 15 times the Italian rate among all MPs, and the Russian rate is over 25 times the Italian among SMD MPs. The startling contrast stems from the behavior of Russian independents, who won roughly two-thirds of Russian SMDs in 1993, reflecting not only the electoral rules and weakly institutionalized party system but also the compressed campaign, a product of Yeltsin’s decision to call the parliamentary elections on short notice and in conjunction with the constitutional referendum. These independents, having earned SMD seats, engaged in party shopping and hopping in the Russian stage A. Although in Italy the ballot structure inhibited independents from entering the Chamber, all mobile SMD MPs in stage A were center-leftists who in the 1996 race mounted broad appeals and who, once in Parliament, opted for the motley Mixed Group instead of joining the legislative group of the largest party sponsoring their list (Di Virgilio 1997, 2002).8

9

When office benefits are allocated (stage B, on the second row), mean weekly switches per 100 MPs are roughly similar for the full lower house across Russia and Italy. The occurrence of the event of switching in stage B is almost three times higher among SMD legislators in Russia than among PR ones; independents in particular rushed to move, for joining parties made them eligible for internal legislative office. In Italy, the proportion is flipped, with PR MPs switching about three times as often as their colleagues. This pattern largely traced to the choices of a subset of PR MPs who split from their parliamentary party so as to compete for seats on the Committee for Constitutional Reform, as detailed below. Switching in the aggregated policy control stage (C) is over five times more frequent in Russia than in Italy. The rate of switching in proximity to Italian sub-national elections (the aggregated stage E) nearly matches that registered in the Italian policy stage (C). Switching in the Russian electoral stage is double that of the Italian E but under half that of the Russian C. Mean weekly switches per 100 MPs drop to their lowest rate in the full house for Italy and Russia in the aggregated stage D (dormant), after initial affiliations are announced, when neither benefits are allocated nor major questions of policy control dominate the agenda nor elections loom. In both cases, switching in D dies down. By these data, the null hypothesis fails. Overall, Table 2 allows for a preliminary assessment of our hypotheses. The basic notion that the rate of switching varies by stage (Hypothesis 1) finds support. Among MPs elected on party labels, switching rates per aggregated stage look roughly alike across Russian and Italian MPs regardless of mandate. It is thus the propensity of the Russian independents to jump parliamentary groups that underpins the statistically significant differences observed so far. Indeed, the fact that Russian independents switch often in stages A and B aligns with Hypothesis 2. Yet Russian independents are also prone to switch in the aggregated policy stage C. Further

10

inspection of the first two hypotheses calls for scrutiny of sub-stages in these primary cases, just as evaluation of Hypothesis 3 requires isolating the most intense sub-stages of policy bargaining. 4.2

Switching disaggregated by sub-stage Figures 2A and 2B portray mean weekly switching rates for all MPs and SMD MPs in

each of the sub-stages, ordered chronologically, in the 1993-1995 Duma and the 1996-2001 Chamber. The behavior of PR MPs, as the remaining category, can be imputed. The error bars show standard errors in mean weekly switches per 100 MPs for each sub-stage, and sub-stages with relatively high rates are tagged. [Figures 2A and 2B about here] Looking first at Figure 2A, the frequency of switching among Russian SMD MPs in the Duma’s first affiliation stage (stage A.1) is an obvious outlier. This is, by a very long shot, the highest rate of switching in any sub-stage for either subgroup of deputies (SMD or PR) in the two legislatures. During the Russian A.1, which lasted only one week, SMD (but not PR) deputies were asked to self-identify as to factional membership. Whereas any party qualifying for PR seats automatically received official Duma faction status, additional groups of SMD MPs could be formed and could obtain the same rights enjoyed by factions, as long as the group met the minimum size of 35. Seizing the opportunity to affiliate, SMD legislators switched with abandon, and formed an entirely new faction during A.1. PR MPs could not create new groups during A.1, which ended when the Duma shifted its focus to internal institutional matters, committee assignments, and other office-related votes (stage B, benefits). Only after the extant groups had successfully monopolized committee and leadership posts did they write rules permitting new groups to form, with the approval in March 1994 of the law on registration of Duma factions. This law opened Russia’s second affiliation stage (A.2) and enabled PR MPs to

11

switch. Even in A.2, however, SMD MPs—in particular, those elected as independents—evinced a higher rate of switching. Switching in the single benefits stage was not as great as in the two affiliation stages, but MPs elected as independents provided most of the moves in B as well. The second hypothesis thus finds additional corroboration. Legislators’ policy disputes dealing with the second major campaign of the Chechen War (classified as stage C.3, near the midterm) generated the second highest mean weekly switches per 100 MPs in the 1993-1995 Duma. Relatively high rates of switching appeared in two other policy stages we define, one involving the first Chechen War and subsuming legislative deliberations on the 1995 budget (C.2) and the other comprised of legislators’ efforts to contribute to managing the Budennovsk hostage crisis (C.4). These stage C switches created new factions and redefined agenda setting, consistent with the third hypothesis. Moreover, as the term unfolded, a number of SMD MPs originally elected as independents emerged as notables within the Duma, able to exert strong policy influence, so that their moves may also be interpreted as supporting the third hypothesis. As Figure 2B exhibits, the rate of Italian switching is highest in a policy stage near the midterm. In October 1998, during stage C.6, the Prodi government made a vote on the 1999 budget a matter of confidence, lost, and resigned. Communist Refounding (RC), not in the executive but until then routinely in its legislative majority, split on the confidence vote. The dissidents entered the Mixed Group, unable to form a separate legislative party due to rules on minimum size. The pro-government majority of the RC moderated and established a distinct parliamentary group with a new name. The RC’s fission and its effect comport with the third hypothesis and fit the corollary as well. The second-highest peak in Italian switching arose in a benefits stage, B.3, when seats

12

and leadership posts on the large Bicameral Committee on Constitutional Reform were allocated. The Prodi government made constitutional reform one of its top priorities. Since the choice of rules carried profound consequences for policy outcomes (e.g., Riker 1982), appointments to the Bicameral Committee were intrinsically linked to high-stakes policy. The bulk of switches in B.3 involved MPs from the Christian Democratic Center-United Democratic Christians (CCD-CDU), the leftmost group in the center-right bloc. The moves were made en masse, as the CDU portion of what began as the unified CCD-CDU group split off and entered the Mixed Group; the CDU contained an unusually large share of PR MPs. The day after the CDU bolted, its leader was named to the Bicameral Committee, with a seat on two of its four subcommittees. Events in stage B.3 march with our hypotheses 2 (on office and policy motives, bound together, and on notables) and 3 (agenda advantage). Relatively high rates of switching distinguished the run-up to and aftermath of the May sub-national elections, stage E.3. To contest the 1998 elections, Senator and former President Cossiga founded a new centrist party (the UDR, Union of Democrats for the Republic), which attracted MPs from the former CDU and CCD, as well as some with center-left origins. As the campaign unfolded and overlapped with a policy stage, Cossiga announced the UDR’s stance on budget legislation and constitutional questions, pursued tightly linked electoral and policy aims, and sought the center of the Italian policy space. His initiatives and the switches he spurred support the second and third hypotheses and the corollary on the location of new groups. After the peak in policy stage C.6, a string of stages with relatively pronounced switching rates appeared. The single week classified as D.5 occurred in October 1998, between decisions on benefits in the D’Alema I cabinet (which replaced the ill-fated Prodi I), and the resumption of committee handling of the 1999 budget; all but one of the switchers in D.5 moved from the UDR

13

to the Mixed Group. Stage D.6 preceded the electoral cycle defined by the June 1999 subnational and European Parliament elections (E.4). In phase D.6, preparing for the 1999 elections, MP and former Premier Prodi launched the center-left Democrats-Olive Tree (Dem-U). Additional moves in stage E.4 enlarged the Dem-U, so that it satisfied the minimum size requirement (20 MPs) for a separate parliamentary group. Other switches in the wake of the 1999 elections, still in E.4, created the Democratic Union for Europe (UDEur) as an organized component of the Mixed Group and as the successor to the UDR, also led by Cossiga. Chamber membership in the UDEur swelled sufficiently during policy stage C.8 to qualify it as a legislative group, and stage C.8 witnessed debates on finance so fierce that two days after the 2000 budget bill won approval, the government resigned. Together, these episodes feature moves responding to notables’ initiatives and coinciding with policy controversy and electoral campaigns. Given the timing of D.5 and D.6 relative to benefit, policy control, and electoral phases, switching in these two sub-stages of the generally calm D does not upset our hypotheses and even buttresses them. In both Italy and Russia, we uncover evidence on vote-driven switching that aligns with our reasoning. The rate of switching in the Russian re-election stage E is roughly the same as that in the two Italian non-parliamentary electoral stages just discussed. We cannot observe a parliamentary re-election stage for Italy as we do for Russia, since the President dissolved the Italian Parliament before the date of the 2001 election had been set and before the mandated day could be identified for depositing candidate names and party lists. We can apply the 90-day rule of thumb used for non-parliamentary elections, however, and count back from the earliest date of those in question. Not a single MP moved during the pre-electoral sub-stage so defined; one moved a few days before this sub-stage started. Thus, in Italy, jockeying for electoral advantage

14

occurred in conjunction with sub-national and supranational elections (cf. Heller and Mershon 2005). And since Italian MPs synchronized vote-seeking switching to non-legislative elections, they had tested the performance of their new electoral vehicles by the time the next parliamentary elections were held. In Russia, instead, relatively many MPs switched immediately before the parliamentary campaign. We attribute this contrast in timing to differences in institutions, the degree of party system institutionalization, the age of the democratic regimes, and the information available to the players in electoral politics. We have adduced evidence in favor of all three hypotheses.9 Yet a more robust appraisal of our hypotheses would extend to varied institutional settings and would include party systems commonly viewed as relatively stable. We thus now turn to our subsidiary country-terms to probe the more general purchase of our reasoning. 4.3

Comment [CAM4]: Will, does this subheader work for you? It is the best Olga and I can devise.

Broadening the range of comparison: subsidiary cases Table 4 examines the frequency of switching in stages of the parliamentary cycle across a

total of 35 country-terms in Italy, Romania, the UK, and the US. Not a single switch occurred in stage A (affiliation) in Italy, Romania, and UK, whereas the standardized measure of mean monthly switches per 100 MPs is 0.051 for the A stages in the US. The switchers in the US stage A were almost all elected as independents or on a minor party label; having won their seats, they joined one of the two major parties, with which they may or may not have had a prior affiliation.10 We see this pattern as testament to the impact of electoral systems and modes of candidate selection. It is only in the US that the use of a primary system in SMDs opens a side door through which a few independents enter the lower house, as, for example, challengers who have lost in primaries run—and win—as independents in a three-way race in the general election. In the other three countries, independents’ access to legislative office is extremely

15

restricted: In the UK, party organizations choose candidates for SMD races; and in 1963-1972 Italy and in Romania, party-list PR gives party leaders great control over candidate selection (cf. Birnir 2004, esp. 144-146, on Romania; Wertman 1977 on Italy). Hence the contrast emerges between, on the one hand, the completely quiescent stage A in 1963-1972 Italy, Romania, and the UK, and, on the other, the US, where the relatively few independents and minor party candidates who win tend to join one of the major parties in stage A so as to avail themselves of the benefits of party affiliation. We do not wish to exaggerate the contrast. Yet the finding here reinforces the lesson drawn earlier from the hybrid electoral systems that switching in stage A is shaped by election rules. The independents permitted in SMD races in Russia switched often in stage A; and the Italian SMD candidates who were forced by law to link to a specific PR party list but whose campaigns stressed the broader themes of the center-left bloc tended to switch to the generic Mixed Group in stage A. [Table 4 about here.] Institutional influences also underpin cross-national differences in switching rates for other stages in the parliamentary cycle. In particular, we attribute the variations in switching in stage B (benefits) to variations in electoral laws, legislative institutions, and regime type. First consider the two PR systems here. The parliamentary regime, 1963-1972 Italy, is distinguished by a relatively high switching rate in the aggregated benefits stage. The Italian datum for stage B also enters into a policy (C) stage and rests on a single month—indeed, the single day in 1969 that marked the split of the briefly-reunified Socialists into Italian Socialist Party (PSI) and Italian Social Democratic Party (PSDI). The split was precipitated by sharp intraparty disputes over government policy and ruptured the ruling coalition (Di Scala 1988, esp. 161-165). Politicians of the PSI and PSDI merged in 1996 and split in 1969, confident that their

16

implementation and reversal of the fusion carried no adverse consequences for their survival, given PR, and for their long-term participation in the executive, given DC dominance of the party system (Mershon 1994, 2002). In semi-presidential Romania, stages B and C also overlap at times, and it is only during the overlap that any B stage switching occurs. Switches to independent status predominated in this span; in fact, the vice-chair of a ruling party became an independent amid disagreements over the budget that triggered a change in government (BBC 1998). Note that MPs’ behavior was constrained by the standing orders of the Romanian Chamber, which stipulate that MPs cannot enter a parliamentary party that has not already earned seats in elections (Chiva 2007, 203-204; Parlamentul României 2008b). Moves to join (or found) splinter parties or start-ups are impossible under these rules. Switching in the benefits stage for the two SMD systems speaks to the impact of legislative institutions and regime design. In our span of study, British politicians never jumped party when offices in one-party majority executives were decided or when posts on relatively weak legislative committees were distributed. In the US presidential system, executive positions are obviously not at stake in the legislative stage B. Instead, seats and leadership offices on the most specialized, most powerful set of legislative committees in the world are allocated in the US benefits stage. This stage engenders the highest mean monthly switches per 100 legislators of any stage in the US. Granted, the mean here stands well below that recorded for Italian and Romanian legislators in some other aggregated stages. The fact remains that if US legislators switch, they are likely to switch when either office benefits are awarded, in stage B, or the perks of party affiliation are available, in A.11 The index of mean monthly switches per 100 MPs per stage reaches its maximum value

17

in our dataset in the aggregated Romanian stage C (policy). This average in turn includes the single greatest monthly observation of switches per 100 MPs within our dataset, 12 that for June 2001, when the Party for Social Democracy in Romania (PDSR) merged with the Social Democratic Party of Romania (PSDR) to create the Party of Social Democracy (PSD). The merger stemmed not just from policy proximity (as the names suggest) but also from institutional incentives and constraints. In part as a response to the summer 2000 increase of the PR threshold (from 3 to 5 percent), the PDSR and PSDR agreed upon an electoral alliance for the November 2000 parliamentary elections. The official fusion represented the culmination of the electoral alliance, a culmination delayed by internal party deliberations on the merger (Lovatt and Lovatt 2001; Project on Political Transformation, 2008). When we draw within-nation comparisons, switching rates are highest in the aggregated stage C for Romania and the UK, and reach their second-highest value for Italy. These withinnation maxima reflect the founding of new parliamentary groups, discussed further below. Suffice it to emphasize for now that this pattern supports the third hypothesis, since the new parties aimed for an impact on policy in stage C. The Romanian PSD styled itself a natural “party of government” (Lovatt and Lovatt 2001). The Italian Socialist Party of Proletarian Unity (PSIUP) split from the PSI in a 1964 policy stage to protest Socialist participation in government. The PSI and PSDI merged in a 1966 policy stage so as to participate more effectively in coalition government, and split in an overlapping B and C stage in 1969, given recurring policy disagreements (Di Scala 1988). The British Social Democratic Party (SDP), initiated in a 1981 policy stage, sought to pull votes from both Labour and Conservatives and to identify a middle ground in programmatic terms as well (e.g., Crewe and King 1995, esp. Ch. 23).

18

As the bottom row of Table 4 exhibits, in stage E the least institutionalized party system, Romania, witnesses the most switching, which echoes the finding for the primary cases in Italy and Russia. Once more, we do not wish to overstate our claims. Yet we infer from this evidence that legislators are more likely to maneuver for votes by defecting to a new party or by adopting independent status in the run-up to parliamentary elections in party systems where voters lack firm attachments to party labels and where the ideological distinctions among parties are relatively blurry. The relatively weak institutionalization of the Romanian and Russian party systems, then, underlies the flurry of switching that both exhibit in anticipation of near-at-hand parliamentary elections. For no country here do mean switches per 100 MPs in the aggregated stage D bottom out at zero. All, however, evince limited switching in D relative to the rate observed in at least a few of the active stages. It turns out that stage D is indeed relatively dormant, contrary to the null hypothesis and in line with the first hypothesis and our overall reasoning. The evidence from the 35 subsidiary country-terms backs all three hypotheses, and also points to the importance of peaks in switching marking the formation of new parliamentary parties. The corollary to the third hypothesis holds that parties established in policy stages should locate at the center of the policy space. We now assess evidence on the corollary from both primary and subsidiary country-term cases. 4.4

Switching to found new parliamentary groups As Table 5 summarizes, of the total of fifteen parliamentary parties created over the

course of the 37 terms comprehended in our study, twelve (80 percent) were launched in a policy stage, C. This count includes the Italian UDR, established during an overlapping policy and electoral stage. Excluded from the dozen is the Italian Dem-U, which qualified as a

19

parliamentary group in an electoral stage, E. As detailed earlier, however, the timing of that party’s founding fell soon after the midterm peak in switching in an Italian stage C. Of the thirteen new players in legislative decision-making just discussed, all but two—the PSIUP of 1964 and the PSI of 1969—were either centrist or more center-leaning than their predecessor party or parties.13 [Table 5 about here.] Consideration of the remaining two parliamentary parties here bolsters the evidence in favor of the corollary. Two new groups in the Russian lower house were created as independents sought to reap the rewards of group affiliation during the tumultuous stage A in 1993-1995 Russia. Both entrants originating in an affiliation stage, when office aims are assumed to dominate, located on the periphery of the policy space. In accordance with the corollary to the third hypothesis, then, when legislative parties were founded during a policy stage, they tended adopt centrist or center-leaning policy positions. Parliamentary groups established outside policy stages located away from the center. Table 5 also presents suggestive evidence on the impact of institutions. The only instances of fission captured in our data occur in the Italian Chamber, which for the 1963-1972 span (and from the end of World War II to 1993) featured a highly permissive form of PR and from 1993 to 2005 used a hybrid system deliberately designed to preserve small parties (e.g., Gambetta 2001). The products of party fission in Italy could realistically hope to survive, given the electoral rules. Romania witnessed only one fusion over the eight years studied, reflecting the constraints imposed by the legislative standing orders and the incentives to coalesce under the 2000 electoral laws. The UK experienced one start-up, the Social Democrats (SDP), and one fusion, the Liberal Democrats. Under SMD rules, political entrepreneurs should strive to avoid

20

fission; and in the case of the SDP, they succeeded, in that the party drew legislators with multiple affiliations (not only Labour MPs but also, in the 1979-1983 term alone, two MPs with Independent Labour status, one Conservative MP, and one Conservative Peer; see Butler and Butler 2000, 248-249; Crewe and King 1995, 114, 478). The utter absence of new parties in the 1950-2000 US Congress may be interpreted as a consequence not only of the constraints of SMD competition but also of the need for parties to mount nationwide presidential contests. The relative abundance of start-ups in the two systems using hybrid electoral laws also deserves comment. Under the Italian variant of hybrid laws, even small units in electoral competition could hope to earn legislative seats, due both to the mechanics of the linkage between SMD and PR races and to the parties’ strategic construction of electoral cartels and strategic nomination of some candidates from small parties in SMDs (e.g., D’Alimonte and Bartolini 2002; Di Virgilio 1997; 2002).14 The unlinked PR and SMD tiers in the Russian hybrid laws slowed the process of party system institutionalization and invited political entrepreneurs to create and change parties with ease (cf. Shvetsova 2003; 2005).

5

Conclusion The key to the analytical approach here is the notion that if legislators change party

affiliations, the timing of their switch reveals its chief motivation: When they switch tells us why they switch. We assume that legislators are strategic actors who view party affiliation as a strategic choice—as a means to an end. In this perspective, any choice to change parties signifies that a new allegiance has become more advantageous than the old one, given changed opportunity structures. Starting from this premise, we expect politicians’ behavior to track and respond to evolving opportunities. To check our expectations, we introduce a partitioning of the parliamentary cycle into distinct stages that are defined by different legislative activities and hence by different rewards available to MPs. We do not assert that legislators in all legislatures 21

switch automatically as a different type of payoffs becomes most salient in the parliamentary cycle. After all, sticking with the party of origin may well be the most advantageous strategy for many politicians. Acknowledging this, we argue that if MPs switch, they should do so when their move is most likely to bring them a particular type of payoff; and the type, like the switch, depends on time. Attention to the timing of switches brings us to findings that are novel in the literatures on party switching, legislative politics, and political parties. For our two primary country-term cases, 1996-2001 Italy and 1993-1995 Russia, we find that switching patterns are tied to the stages in the parliamentary cycle. We identify in our evidence significant differences in switching rates across stages of the cycle and heightened switching for perks, office, policy advantage, agenda grabbing, and pre-electoral jockeying at distinctive stages. Phases of relative calm in the legislative agenda induce relatively low rates of switching. Differences in the timing of vote-driven switching across the two cases can be attributed to differences in the degree of party system institutionalization and thus the information held by elites and voters. In the more institutionalized Italian system, electoral positioning via switching occurs in anticipation of and, to a lesser extent, after sub-national elections, as MPs adjust expectations about voters (cf. Heller and Mershon 2005). In Russia, pre-electoral positioning occurs during national election campaigns. Driving home the import of party system institutionalization is the rough resemblance, displayed in Table 3, in partisan Russian MPs (regardless of mandate) and those in Italy. The commonalities across the two systems with different levels of party system institutionalization provide compelling support for our overarching theoretical framework. In selecting the two primary country-terms, we deliberately restrict variation in electoral institutions. To submit our hypotheses to a stiffer test, we extend the inquiry to an additional 35

22

subsidiary country-terms, guided by the criterion of maximizing institutional variation. Our comparison of 1963-1972 Italy, Romania, the UK, and the US yields findings that largely comport with the hypotheses we advance. In accordance with the first hypothesis, the rate of switching varies across the stages of the parliamentary cycle. Moreover, institutional variations across countries shape which stage evinces the highest switching rate within each country. The subsidiary cases also furnish support for the second hypothesis. For example, independents (where they win legislative seats, as influenced by electoral laws) tend to move above all in affiliation and benefits stages. Switches in stage C feed into new parties that are centrist or center-leaning. This evidence aligns with the third hypothesis and its corollary. In addition, the broad comparisons we draw permit a tentative inference. PR means more parties among which to switch and, indirectly, creates more opportunities to switch in parliamentary and semi-presidential systems, since it makes coalition governments more likely and coalitions do not last as long as single-party majorities (see, among many others, Laver and Schofield 1990). To put the point differently, SMD systems tend to have lower switching rates, both because fewer parties exist and because switching tends to be less advantageous. In a twoparty system resulting from SMD rules, switches necessarily occur between the minority and the majority. Switching is polarizing and visible, and yet unlikely to be effective—to allow switchers to redirect the policy agenda or win votes. This inference, though provisional, is confirmed in a large-N analysis controlling for multiple institutional effects (Mershon and Shvetsova 2008b). A venerable tradition in political science—indeed, virtually all of the profession—holds that institutions matter. Our study too conveys the import and impact of institutions. What is distinctive here is the emphasis on the timing of changes of party affiliation, which sheds new light on the role of institutions in structuring the strategic choices of elected legislators, choices

23

that to a greater or lesser extent alter the legislative party system in which legislators operate. Taking timing into account enriches research on party affiliation and partisan competition. Some phases of the parliamentary cycle offer legislators relatively many stimuli to reconsidering choice of partisanship. In other phases, such stimuli are in low supply. Not just the quantity but also the quality of stimuli counts: Distinct categories of activity dominating discrete stages of the parliamentary cycle elicit a particularly strong focus on particular goals on the part of legislators. Temporal stages leave their imprint on changes of partisanship in legislatures. Timing matters for switching, parties, and legislative party systems.

24

Figure 1. Switching Behavior during the Parliamentary Cycle.

Pre-cycle campaign

Stage A: (Affiliation) Reaffirm affiliation with the electoral party

Party system canStage be congruent B: (Benefits) of committees, two Determination consecutive

Taking up seats and in the or not announcing group membershipcampaigns

government participation, portfolios

ts Allocating parliamentary/cabinet offices

Stage C: (Control of policy) Advantage in agenda control

Parties in elections Active policymaking

Campaign

Stage E: (Elections) Coalition formation to ensure personal reelection

Positioning for advantage in the next election

25

Figures 2A & 2B. Mean Weekly Switches per 100 MPs (All and SMD), by Sequential SubStage in Term, 1993-95 Russian Duma and 1996-2001 Italian Chamber (SEs in error bars).

Mean Weekly Switches per 100 MPs

Mean Weekly Switches per 100 MPs

A.1 off chart*

10

2.5

C.3

8 6 4 C.2

A.2

C.4

E

2

0 0 4 8 12 16 Sub-stages in Russian Term, Ordered Chronologically

*A.1 is outlier: mean/100 all MPs = 27.8 mean/100 SMDs = 55.6 Vertical axis truncated so as to clarify scale for other stages.

mean/100 all MPs mean/100 SMDs

C.6

B.3

2.0 D.5

C.8

1.5 1.0 E.3

D.6 E.4

A

0.5 0.0

0 8 16 24 32 Sub-stages in Italian Term, Ordered Chronologically

mean/100 all MPs mean/100 SMDs

26

Figures 3A & 3B. Weekly Switches per 100 MPs, Sub-stage Mean Weekly Switches per 100, and Differences from Sub-stage Means, Russia 1993-95 and Italy 1996-2001.

Weekly Switches per 100 MPs

Weekly Switches per 100 MPs

A.1 off chart*

10 A.2

C.2

weekly switch/100 sub-stage mean diff from mean

C.3

5

0 *A.1 is outlier (weekly switch/100=sub-stage mean=27.8). Vertical axis truncated so as to clarify scale for other stages.

-5

-10 0

26 52 78 104 Weeks in Legislative Term, Russia

4

130

weekly switch/100 sub-stage mean diff from mean

3 B.3

E.3

C.6 E.4

2

C.8

D.5

1

D.6

0 -1 0

65 130 195 Weeks in Legislative Term, Italy

260

27

Table 1. Case Selection: Countries and Legislative Terms. Electoral Laws List PR Regime Type Plurality Hybrid Open Closed Parliamentary UK 1970-2005 Italy 1963-72 Italy 1996-2001 Presidential US 1950-2000 Russia 1993-95 Semi-presidential Romania 1996-2004 Note: Federal and regionalized systems appear in italics. Primary cases, where operationalization of stages is most finely grained, appear in boldface. Number of terms studied in subsidiary cases = 2 Italy, 2 Romania, 6 UK, 25 US.

28

Table 2. Operationalizing Stages of the Parliamentary Cycle. Operationalization Operationalized Stage Stage Concept shaped by: Italy 1996-2001

L

Affiliation: A MPs take up seats and announce group affiliation

electoral laws affecting independents; legislative rules on groups and on start of session

E

Benefits: Legislative and relative strength of G B executive offices are president v. premier; allocated number of cabinets per I legislative term S L A T I V

Control of policy: Policy rules on introducing, C making dominates considering, and agenda approving bills, incl. committee role

E D Dormant: All other than definition of other stages A, B, C, and E (this is residual stage) E Re-election: MPs rules on dissolution L E position for advantage in of legislature before E next legislative election elections; rules on C (includes nonregistration for ballot; T parliamentary elections rules on election timing; O in some contexts) duration of term R A L Note: All stages save D can overlap or coincide.

election day to last day MPs must state group membership

with >1 cabinets, B recurs; day groups announced to day legislative and 1st executive payoffs completed; day Nth cabinet falls to day N+1 cabinet named legislative record, from day executive sends budget to house to day bill passed; day constitutional bill presented to day committee dissolved as indicated by other stages E not observed; parliament dissolved (many non-parliamentary elections in 5-year term; 90 days before, 30 days after election day)

Russia 1993-1995

Subsidiary

election day to day of group selection for SMD MPs; for PR, 5 weeks after rules specified on group selection given strong presidency, no governing coalition formed; period of allocation of legislative committee posts as reflected in legislative records legislative record, from first vote on finance until finance moves off the agenda; on war: as indicated by key events that open and close policy episode as indicated by other stages

1st 2 months, starting election day

day marking 90 days before election to end of ballot registration 30 days before election

for committees: 1st month that legislature in session; for cabinet(s) in parliamentary systems, any month in which cabinet(s) negotiated any month that legislature is officially in session, as shown in legislative records

as indicated by other stages 6 months before election day

(no non-parliamentary elections included)

29

Table 3. Mean Weekly Switches per 100 MPs by Type of Stage and by MPs’ Mode of Election, 1996-2001 Italian Chamber and 1993-1995 Russian Duma. Italian Chamber Russian Duma Aggregate N All Aggregate All SMD Stage Weeks MPs SMD PR N weeks MPs All Party¹ Indep¹ PR A 4 0.36 0.43 0.18 6 5.56 11.04 0.90 15.01 0.07 B 9 0.35 0.25 0.67 12 0.30 0.44 0.23 0.52 0.15 C 109 0.20 0.18 0.27 26 1.09 1.62 0.52 2.03 0.56 D 95 0.10 0.11 0.08 53 0.14 0.18 0.10 0.20 0.10 E 94² 0.21 0.21 0.27 8² 0.42 0.50 0.51 0.47 0.33 Key to stages: A: Affiliation B: Benefits C: Control of policy E: Electoral D: Dormant, i.e. all periods other than A, B, C, and E. Sources: For dates and numbers of switches: Camera dei Deputati 2000b; Duma Statistical Services (INDEM) 2000. For detailed documentation needed to code stages: Camera dei Deputati 2008b; INDEM 2000; Istituto Cattaneo 2008; Ministero dell’Interno 2008; Pasquino 1999. Tukey test for differences of means (all MPs) across Russian stages: p < 0.005; Tukey test for differences of means (all MPs) across Italian dormant and active stages (given overlapping of active stages): p < 0.10. ¹ Independents are elected as a subset of SMD MPs only in Russia, as discussed in the text. ² The only observable electoral stages in Italy pertain to non-parliamentary elections, whereas the only electoral stage in Russia is associated with the campaign for the subsequent parliament.

30

Table 4. Mean Monthly Switches per 100 MPs by Type of Stage, Electoral System, and Regime, 35 Country-Terms. S List PR SMD t Open, Parliam Closed, Semi-Pres Parliamentary Presidential a Italy 1963-1972 Romania 1996-2004 UK 1979-2005 US 1950-2000 g Mean/100 T Mean/100 T Mean/100 T Mean/100 T e MPs Mos MPs Mos MPs Mos MPs Mos A 0.000 (0.000) 4 0.000 (0.000) 4 0.000 (0.000) 18 0.051 (0.134) 50 B 0.933 (3.604) 16 0.172 (0.237) 7 0.000 (0.000) 13 0.069 (0.141) 50 C 0.367 (2.118) 99 0.982 (5.348) 77 0.032 (0.149) 247 0.008 (0.041) 450 D 0.025 (0.080) 19 0.143 (0.553) 19 0.019 (0.098) 108 0.011 (0.050) 100 E 0.011 (0.042) 14 0.258 (0.352) 14 0.022 (0.062) 56 0.003 (0.025) 175 Key to stages: A: Affiliation B: Benefits C: Control of policy E: Electoral D: Dormant, i.e. all periods other than A, B, C, and E. Note: Numbers between parentheses are standard deviations. Sources: For numbers and dates of switches: Ballot Access News 2001a, 2001b; Butler and Butler 2006, 112-114; Camera dei Deputati 2008a; Timothy Nokken, personal communications, June 2006 and July 2008; Parlamentul României 2008c; US Congressional Biographical Directory 2008. For stages: Butler and Butler 2006; Camera dei Deputati 2008a; CQ Weekly (various dates, e.g., 11/19/83, 2403; 12/8/84, 3051; 2/2/85, 177; 9/14/85, 1827; 2/17/90, 437; 11/12/94, 3222; 12/3/94, 3427; 12/23/95, 3863; 11/28/98, 3218; 1/16/99, 156-157); Keesing’s Record, various dates; Mershon 2002, 212; Parlamentul României 2008a; UK Parliament 2008; US House of Representatives, Office of the Clerk 2008; US House of Representatives, Office of the Majority Leader 2008.

31

Table 5. Switching to Create New Parliamentary Parties, by Type of Stage, Electoral System, and Regime, 37 Country-Terms. Stage C

Non-C

Open, Parl Italy 63-72

List PR Closed, Semi-Pres Romania

PSIUP¹ PSU² PSI¹ and PSDI¹

PSD²

SMD Parl UK

Pres US

Hybrid Parl Semi-Pres Italy 96-01 Russia

SDP³ No new party LibDem²

UDR³ GC¹ UDEur³

Ross³ Stab³

No new party

DemU³

LDS³ N-96³

¹ Fission ² Fusion ³ Start-up Note: Entries are acronyms of parties, ordered chronologically by type of stage and by country. Acronyms in italics indicate a new party that is either centrist or more center-leaning than its predecessor(s). Sources: Butler 2000, 248-249; Camera dei Deputati 2008a, 2008b; Crewe and King 1995, 114, 478, passim; Di Scala 1988; INDEM 2000; Lovatt and Lovatt 2001; Parlamentul României 2008a, 2008c. Key to party acronyms (year founded): DemU Democrats-Olive Tree (1999) GC Communist Group (1998) LDS Liberal Democratic Union of December 12 (1994) LibDem Liberal Democratic Party (1988) N-96 New Regional Policy – Duma 96 (1993) PSD Social Democratic Party (2001) PSDI Italian Social Democratic Party (1969) PSI Italian Socialist Party (1969) PSIUP Italian Socialist Party of Proletarian Unity (1964) PSU Unified Socialist Party, Italy (1966; name changes not recorded here) Ross Russia (1995) SDP Social Democratic Party (1981) Stab Stability (1995) UDR Union of Democrats for the Republic (1998) UDEur Democratic Union for Europe (1999)

32

BBC [British Broadcasting Corporation]. 1998. Friday, 13 March, 1998, 14:31 GMT

Romania's coalition Democrats urge formation of new government. http://news.bbc.co.uk/2/hi/world/monitoring/65158.stm . Last accessed July 25, 2008. Di Scala, Spencer M. 1988. Renewing Italian Socialism: From Nenni to Craxi. New York: Oxford University Press. Heller, William B. and Carol Mershon. 2003. “Switching in Parliamentary Parties: Exits and Entries in Parliamentary Groups in the Italian Chamber of Deputies, 1996-2001.” Unpublished manuscript. University of Virginia. INDEM (Information and Analysis System). 2000. INDEM-Statistics 2000 (Full version). Moscow: INDEM Foundation. . Shvetsova, Olga. 2003. "Resolving the Problem of Pre Election Coordination: The 1999 Parliamentary Election as Elite Presidential 'Primary,'" in V. Hesli and W. Reisinger, eds., Elections, Parties and the Future of Russia. Cambridge: Cambridge University Press. Shvetsova, Olga. 2005. "Compromising a Long Lasting Transitional Formula," in Josep Colomer, ed., The Handbook of Electoral System Design. St. Martin Press.

Wertman, Douglas. 1977. “The Italian Electoral Process.” In Howard R. Penniman, ed. Italy at the Polls: The Parliamentary Elections of 1976. Washington, DC: American Enterprise Institute. New references / alphabetize and complete later. NEED FIRST NAMES OF REFS (NOT CPS STYLE) RA has worked on REFS – see below Project on Political Transformation and the Electoral Process in Post-Communist Europe, University of Essex. 2008. “Romania: Legislation.” Available at:

http://www2.essex.ac.uk/elect/database/indexCountry.asp?country=ROMANIA&opt=leg . Last accessed July 9, 2008.

Marina Popescu. The parliamentary and presidential elections in Romania, November 2000 . Electoral Studies Volume 22, Issue 2, June 2003, Pages 325-335 Annual chronologies available at www.istcattaneo.org/archivi;

33

Ministry of Interior information at http://cedweb.mininterno.it; http://www.mininterno.it/sezioni/ attivita/Le%20elezioni/s_000000056.htm, http://www.mininterno.it/sezioni/attivita/ Le%20elezioni/s_000000056.htm United States Congressional Biographical Directory. 2008. Available at: http://bioguide.congress.gov/biosearch/biosearch.asp . Last accessed July 9, 2008. E Euro Constl Review http://www.law.nyu.edu/eecr/vol10num1/features/interpreting.html Interpreting an Electoral Setback—Romania 2000 Alina Mungiu-Pippidi and Sorin Ionita. Winter 2001 Birnir, Jóhanna Kristín. 2004. “Institutionalization of the Party System.” In Henry F. Carey, ed. Romania since 1989: Politics, Economics, Society. Lanham, MD: Lexington Books. Butler, David and Gareth Butler. 2006. British Political Facts since 1979. New York: Palgrave Macmillan.

Camera dei Deputati. 2008a. Legislature precedenti. http://legislature.camera.it/ . Last accessed July 7, 2008. Camera dei Deputati. 2008b. XIII [Tredicesima] legislatura. http://leg13.camera.it/ . Last accessed July 9, 2008. Chiva, Cristina. 2007. The Institutionalization of Post-Communist Parliaments: Hungary and Romania in Comparative Perspective. Parliamentary Affairs 60 (2): 187-211. Crewe, Ivor and Anthony King. 1995. SDP: The Birth, Life, and Death of the Social Democratic Party. Oxford: Oxford University Press. Gallagher, Tom. 2005. Communism, the Failure of Democratic Reform, and the Theft of a Nation. New York: NYU Press.

34

Keesing's Record of World Events. various dates. www.keesings.com. Last accessed July 3, 2008. Lovatt, Catherine. 2000a. “Apathy and Confusion: Low Voter Turnout Blemishes Local Elections.” Central Europe Review, 2 (23), June 12. http://www.cereview.org/00/23/lovatt23.html . Last accessed July 4, 2008. Lovatt, Catherine. 2000b. “Think Global, Vote Local.” Central Europe Review, 2 (17), May 2. http://www.ce-review.org/00/17/lovatt17.html . Last accessed July 4, 2008. Lovatt, Catherine and David Lovatt. 2001. “PSD: A New Party of Government.” Central Europe Review 3 (23), June 25. http://www.pecina.cz/files/www.cereview.org/01/23/romanianews23.html . Last accessed July 10, 2008. Parlamentul României, Camera Deputaţilor. 2008a. Activitate parlamentară. http://www.dcep.ro/pls/steno/steno.home?idl=2 . Last accessed June 28, 2008. Parlamentul României, Camera Deputaţilor. 2008b. “Organizarea şi funcţionarea camerei deputaţilor: Grupurile parlamentare.” http://www.cdep.ro/pls/dic/site.page?id=134&idl=1. Last accessed July 10, 2008. Parlamentul României, Camera Deputaţilor. 2008c. Prima pagina. http://www.cdep.ro/ . Last accessed June 28, 2008. United Kingdom Parliament. 2008. “List of Previous Commons Recess Dates.” http: www.parliament.uk/faq/recess.ccfm . Last accessed June 13, 2008. U.S. House of Representatives, Office of the Clerk. 2008. “Official House Information: House Floor Proceedings.” http://clerk.ouse.gov/index.html Last accessed June 30, 2008. U.S. House of Representatives, Office of the Majority Leader. 2008. “U.S. House Vote Schedule.” http://majorityleader.house.gov/docUploads/2008-CALENDAR.pdf Last

35

accessed June 30, 2008.

References Àgh, A. (1999). The Parliamentarization of the East Central European Parties. In S. Bowler, D. M. Farrell & R. S. Katz (Eds.), Party Discipline and Parliamentary Government. Columbus: Ohio State University Press. Aldrich, J. H. (1995). Why Parties? The Origin and Transformation of Political Parties in America. Chicago: University of Chicago Press. Beck, N., & J. N. Katz. (1995). What to Do (and Not to Do) with Time-Series-Cross-Section Data in Comparative Politics. American Political Science Review 89: 634-647. Beck, N., J. N. Katz, & R. Tucker. (1998). Taking Time Seriously: Time-Series-Cross-Section Analysis with a Binary Dependent Variable. American Journal of Political Science 42: 1260-1288. Canon, D. T., & D. J. Sousa. (1992). Party System Change and Political Career Structures in the U.S. Congress. Legislative Studies Quarterly 17 (3):347-363. CQ Weekly [Online]. (Various dates.) Available at http://library.cqpress.com/cqweekly. Corbetta, P., & A. M. L. Parisi (Eds.). (1997). A domanda risponde. Il cambiamento del voto degli italiani nelle elezioni del 1994 e del 1996. Bologna: Il Mulino. Cox, G. W., &. M. D. McCubbins. (2006). Setting the Agenda: Responsible Party Government in the U.S. House of Representatives. Cambridge: Cambridge University Press. D’Alimonte, R., & S. Bartolini (Eds.). (2002). Maggioritario finalmente? La transizione elettorale 1994-2001. Bologna: Il Mulino. Di Virgilio, A. (1997). Le alleanze elettorali. Identità partitiche e logiche coalizionali. In R.

36

D’Alimonte & S. Bartolini (Eds.), Maggioritario per caso. Le elezioni politiche del 1996. Bologna: Il Mulino. Di Virgilio, A. (2002). L’offerta elettorale. La politica delle alleanze si istituzionalizza. In R. D’Alimonte & S. Bartolini (Eds.), Maggioritario finalmente? La transizione elettorale 1994-2001. Bologna: Il Mulino. Desposato, S. W. 2006. Parties for Rent? Careerism, Ideology, and Party Switching in Brazil's Chamber of Deputies. American Journal of Political Science. add Gillespie, E. (2005, April 12.) Frontline: Karl Rove: The Architect. PBS [online]. Available: http://www.pbs.org/wgbh/pages/frontline/shows/architect/interviews/gillespie.html. Heller, W. B., & C. Mershon. (2004, July). Theoretical and Empirical Models of Party Switching. Paper presented at the First Workshop of the Party Switching Research Group, Dublin, Ireland. Heller, W. B., & C. Mershon. (2005). Party Switching in the Italian Chamber of Deputies, 19962001. Journal of Politics 67(2):536-559. Hickey, J. G. (2003, October 27). Gillespie Focuses On Growth of GOP. Insight on the News [online]. Available: http://www.findarticles.com/p/articles/mi_m1571/is_2003_Oct_27/ai_109128676?pi=cal Kaminski, M. (2002). Do Parties Benefit from Electoral Manipulation? Electoral Laws and Heresthetics in Poland, 1989-93. Journal of Theoretical Politics, 14(3), 325-358. Kunicová, J. & T. F. Remington. (2005, April). The Effect of Electoral Rules on Distributive Voting: Some Evidence from the Russian State Duma, 1994-2003. Presented at the Annual Meetings of the Midwest Political Science Association, Chicago, IL. Laver, M. & K. Benoit. (2003). The Evolution of Party Systems between Elections. American

37

Journal of Political Science 47(2): 215-233. Laver, M. & K. Shepsle. (1996). Making and Breaking Governments: Cabinets and Legislatures in Parliamentary Democracies. Cambridge: Cambridge University Press. Mainwaring, S. P. (1999). Rethinking Party Systems in the Third Wave of Democratization: The Case of Brazil. Stanford: Stanford University Press. Mainwaring and Scully Mainwaring and Torcal. McElroy, G. M. (2003, April). Party Switching in the European Parliament: Why Bother? Paper prepared for delivery at Annual Meetings of the Midwest Political Science Association. Mershon, Carol. 2002. Costs of Coalition Mershon and Shvetsova 2005 – Cville paper,). Mershon, Carol, and Olga Shvetsova. 2008a. "Parliamentary Cycles and Party Switching in Legislatures." Comparative Political Studies 41 (1):99-127. Mershon, Carol, and Olga Shvetsova. 2008b. "Party Switching in Sitting Parliaments and the Midterm Effect." Presented at the Annual Joint Sessions of the European Consortium for Political Research, Rennes, France. Miskin, S. (2003). Politician Overboard: Jumping the Party Ship. Research Paper, Parliament of Australia. Available: http://www.aph.gov.au/library/pubs/rp/2002-03/03RP04.htm . Moser, R. G. (2001). Unexpected Outcomes: Electoral Systems, Political Parties, and Representation in Russia. Pittsburgh: University of Pittsburgh Press. T. P. Nokken & K. T. Poole. (2004.) Congressional Party Defection in American History. Legislative Studies Quarterly, 29(4), 545-568. Pasquino, G. (1999). Autopsia della Bicamerale. In D. Hine & S. Vassallo (Eds.), Politica in Italia, Edizione 99. Bologna: Il Mulino. Remington, T., & S. S. Smith. (1998). Theories of Legislative Institutions and the Organization of the Russian Duma. American Journal of Political Science 42: 545-572.

38

Riker, W. H. (1982). Liberalism against Populism: A Confrontation between the Theory of Democracy and the Theory of Social Choice. San Francisco: W.H. Freeman. United Press International (UPI). (1984, October 3). Campaign Notes: County Official Wins Florida Runoff Election. New York Times [Online]. Available: http://query.nytimes.com/gst/fullpage.html?res=9C0CE0DB123BF930A35753C1A9629482 60&n=Top%2fReference%2fTimes%20Topics%2fOrganizations%2fR%2fRepublican%20P arty . United States Congress. (2005). Biographical Directory of the United States Congress, 17742005. Washington, DC: U.S. G.P.O. Verzichelli, L. (1996). I gruppi parlamentari dopo il 1994. Fluidità e riaggregazioni. Rivista italiana di scienza politica 26(2), 391-413. Yoshinaka, A. (2005). House Party Switchers and Committee Assignments: Who Gets ‘What, When, How? Legislative Studies Quarterly, 30(3): 391-406. Zucchini, F. (2001). La commissione affari costituzionali: Gli effetti paralizzanti del mutamento. In G. Capano & M. Giuliani (Eds.), Parlamento e processo legislativo in Italia. Continuità e mutamento. Bologna: Il Mulino References / below are refs done by RA – need to change & to “and” and some more stuff Àgh, Attila. (1999). The Parliamentarization of the East Central European Parties: Party Discipline in the Hungarian Parliament, 1990-1996.". In Shaun Bowler, David. M. Farrell & Richard S. Katz (Eds.), Party Discipline and Parliamentary Government. 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.

39

Alina Mungiu-Pippidi and Sorin Ionita. Interpreting an Electoral Setback—Romania 2000

East European Constitutional Review 10(1)

http://www.law.nyu.edu/eecr/vol10num1/features/interpreting.html . Winter 2001 Annual chronologies available at www.istcattaneo.org/archivi;

Ballot Access News. 2001a. “Candidates Elected to the House, 1902-2000, Who Weren’t Nominees of the Democratic or Republican Parties.” Vol. 17 (No. 3, June 1). Available at http://web.archive.org/web/20030417164754/http://www.ballot-access.org/2001/0601.html#17 . Last accessed July 22, 2008. Ballot Access News. 2001b. ““Minor Party Candidates Elected to the House, 1902-2000, Who Were Nominated by Their Own Minor Party, Plus Major Party.” Vol. 17 (No. 3, June 1). Available at http://web.archive.org/web/20030417164754/http://www.ballotaccess.org/2001/0601.html#17 . Last accessed July 22, 2008. Budge, Ian, Hans-Dieter Klingemann, Andrea Volkens, Judith Bara, and Eric Tannenbaum. 2001. Mapping Policy Preferences. Estimates for Parties, Electors and Governments, 1945-1998. Oxford: Oxford University Press. Beck, Nathaniel., & Jonathan N. Katz. (1995). What to Do (and Not to Do) with Time-SeriesCross-Section Data in Comparative Politics. American Political Science Review 89 (3): 634647. Beck, Nathaniel, Jonathan N. Katz, & Richard Tucker. (1998). Taking Time Seriously: TimeSeries-Cross-Section Analysis with a Binary Dependent Variable. American Journal of Political Science 42 (4): 1260-1288. Birnir, Jóhanna Kristín. 2004. “Institutionalization of the Party System.” In Henry F. Carey, ed. Romania since 1989: Politics, Economics, Society. Lanham, MD: Lexington Books. Butler, David and Gareth Butler. 2006. British Political Facts since 1979. New York: Palgrave Macmillan.

Camera dei Deputati. 2008a. Legislature precedenti. http://legislature.camera.it/ . Last accessed July 7, 2008.

40

Camera dei Deputati. 2008b. XIII [Tredicesima] legislatura. http://leg13.camera.it/ . Last accessed July 9, 2008. Canon, David. T., & David J. Sousa. (1992). Party System Change and Political Career Structures in the U.S. Congress. Legislative Studies Quarterly 17 (3):347-363. Chiva, Cristina. 2007. The Institutionalisation of Post-Communist Parliaments: Hungary and Romania in Comparative Perspective. Parliamentary Affairs 60 (2): 187-211. CQ Weekly [Online]. (Various dates.) Available at http://library.cqpress.com/cqweekly. Corbetta, Piergiogio., & Arturo M. L. Parisi (Eds.). (1997). A domanda risponde. Il cambiamento del voto degli italiani nelle elezioni del 1994 e del 1996. Bologna: Il Mulino. Cox, Garu W., &. Mathew D. McCubbins. (2005). Setting the Agenda: Responsible Party Government in the U.S. House of Representatives. Cambridge: Cambridge University Press. Crewe, Ivor and Anthony King. 1995. SDP: The Birth, Life, and Death of the Social Democratic Party. Oxford: Oxford University Press. D’Alimonte, Roberto., & Stefano. Bartolini (Eds.). (2002). Maggioritario finalmente? La transizione elettorale 1994-2001. Bologna: Il Mulino. Di Virgilio, Aldo. (1997). Le alleanze elettorali. Identità partitiche e logiche coalizionali. In Roberto D’Alimonte & Stefano Bartolini (Eds.), Maggioritario per caso. Le elezioni politiche del 1996. Bologna: Il Mulino. Di Virgilio, Aldo (2002). L’offerta elettorale. La politica delle alleanze si istituzionalizza. In R. D’Alimonte & S. Bartolini (Eds.), Maggioritario finalmente? La transizione elettorale 1994-2001. Bologna: Il Mulino. Desposato, Scott W. 2006. Parties for Rent? Careerism, Ideology, and Party Switching in Brazil's Chamber of Deputies. American Journal of Political Science. 50 (1): 62-80

41

Comment [LBF5]: Cannot find rest of name. All publications seem to be AML or Arturo ML

Gallagher, Tom. 2005. Modern Romania: The End of Communism, the Failure of Democratic Reform, and the Theft of a Nation. New York: NYU Press. Gillespie, Ed (2005, April 12.) Frontline: Karl Rove: The Architect. PBS [online]. Available: http://www.pbs.org/wgbh/pages/frontline/shows/architect/interviews/gillespie.html. Heller, William. B., & Carol Mershon. (2004, July). Theoretical and Empirical Models of Party Switching. Paper presented at the First Workshop of the Party Switching Research Group, Dublin, Ireland. Heller, William B., & Carol Mershon. (2005). Party Switching in the Italian Chamber of Deputies, 1996-2001. Journal of Politics 67(2):536-559. *Hickey, Jennifer G. (2003, October 27). Gillespie Focuses On Growth of GOP. Insight on the News

[online].

Available:

http://www.findarticles.com/p/articles/mi_m1571/is_2003_Oct_27/ai_109128676?pi=cal Kaminski, Marek. (2002). Do Parties Benefit from Electoral Manipulation? Electoral Laws and Heresthetics in Poland, 1989-93. Journal of Theoretical Politics, 14(3), 325-358. Keesing's Record of World Events. various dates. www.keesings.com. Last accessed July 3, 2008. Kunicová, Jana. & Thomas F. Remington. (2005, April). The Effect of Electoral Rules on Distributive Voting: Some Evidence from the Russian State Duma, 1994-2003. Presented at Comment [LBF6]: Yale Conference on Distributive Politics?

the Annual Meetings of the Midwest Political Science Association, Chicago, IL. Laver, Michael. & Kenneth. Benoit. (2003). The Evolution of Party Systems between Elections. American Journal of Political Science 47(2): 215-233. Laver, M. & K. Shepsle. (1996). Making and Breaking Governments: Cabinets and Legislatures in Parliamentary Democracies. Cambridge: Cambridge University Press.

42

Lovatt, Catherine. 2000a. “Apathy and Confusion: Low Voter Turnout Blemishes Local Elections.” Central Europe Review, 2 (23), June 12. http://www.cereview.org/00/23/lovatt23.html . Last accessed July 4, 2008. Lovatt, Catherine. 2000b. “Think Global, Vote Local.” Central Europe Review, 2 (17), May 2. http://www.ce-review.org/00/17/lovatt17.html . Last accessed July 4, 2008. Lovatt, Catherine and David Lovatt. 2001. “PSD: A New Party of Government.” Central Europe Review 3 (23), June 25. http://www.pecina.cz/files/www.cereview.org/01/23/romanianews23.html . Last accessed July 10, 2008. Mainwaring, Scott. P. (1999). Rethinking Party Systems in the Third Wave of Democratization: The Case of Brazil. Stanford: Stanford University Press. Mainwaring, Scott and Timothy Scully, (1995) Building Democratic Institutions: Party Systems in Latin America. Stanford: Stanford University Press. Comment [LBF7]: I cannot find this citation

Mainwaring, Scott and Torcal Mariano. (2006) Ed. Richard Katz McElroy, Gail M. (2003, April). Party Switching in the European Parliament: Why Bother? Paper prepared for delivery at Annual Meetings of the Midwest Political Science Association. Mershon, Carol. 2002. Costs of Coalition. Stanford: Stanford University Press Mershon and Shvetsova 2005 – Cville paper,). Carol Mershon and Olga Shvetsova. "Electoral Cycles and Party Switching: Opportunistic Partisan Realignment in Legislatures." Paper

Comment [LBF8]: Not sure if this is the right citation

proposed for presentation at the 2005 Meetings of the International Studies Association. Mershon, Carol, and Olga Shvetsova. 2008a. "Parliamentary Cycles and Party Switching in Legislatures." Comparative Political Studies 41 (1):99-127. Mershon, Carol, and Olga Shvetsova. 2008b. "Party Switching in Sitting Parliaments and the Midterm Effect." Presented at the Annual Joint Sessions of the European Consortium for Political Research, Rennes, France. Ministry of Interior information at http://cedweb.mininterno.it;

43

http://www.mininterno.it/sezioni/ attivita/Le%20elezioni/s_000000056.htm, http://www.mininterno.it/sezioni/attivita/ Le%20elezioni/s_000000056.htm

Comment [LBF9]: These links don’t work

Miskin, Sarah (2003). Politician Overboard: Jumping the Party Ship. Research Paper, Parliament of Australia. Available: http://www.aph.gov.au/library/pubs/rp/2002-03/03RP04.htm . Moser, Robert. G. (2001). Unexpected Outcomes: Electoral Systems, Political Parties, and Representation in Russia. Pittsburgh: University of Pittsburgh Press. Timothy. P. Nokken & Keith. T. Poole. (2004.) Congressional Party Defection in American History. Legislative Studies Quarterly, 29(4), 545-568. Parlamentul României, Camera Deputaţilor. 2008a. Activitate parlamentară. Comment [LBF10]: Link does not work

http://www.dcep.ro/pls/steno/steno.home?idl=2 . Last accessed June 28, 2008. Parlamentul României, Camera Deputaţilor. 2008b. “Organizarea şi funcţionarea camerei deputaţilor: Grupurile parlamentare.” http://www.cdep.ro/pls/dic/site.page?id=134&idl=1. Last accessed July 10, 2008. Parlamentul României, Camera Deputaţilor. 2008c. Prima pagina. http://www.cdep.ro/ . Last accessed June 28, 2008. Pasquino, Gianfranco. (1999). Autopsia della Bicamerale. In David Hine & Salvatore Vassallo (Eds.), Politica in Italia, Edizione 99. Bologna: Il Mulino.

Popescu, Marina. The parliamentary and presidential elections in Romania, November 2000 . Electoral Studies Volume 22, Issue 2, June 2003, 325-335 Project on Political Transformation and the Electoral Process in Post-Communist Europe, University of Essex. 2008. “Romania: Legislation.” Available at:

http://www2.essex.ac.uk/elect/database/indexCountry.asp?country=ROMANIA&opt=leg . Last accessed July 9, 2008.

44

Remington, Thomas, & Steven S. Smith. (1998). Theories of Legislative Institutions and the Organization of the Russian Duma. American Journal of Political Science 42 (2): 545-572. Riker, William H. (1982). Liberalism against Populism: A Confrontation between the Theory of Democracy and the Theory of Social Choice. San Francisco: W.H. Freeman. United Kingdom Parliament. 2008. “List of Previous Commons Recess Dates.” http: http://www.parliament.uk/faq/recess.cfm. Last accessed June 13, 2008. United Press International (UPI). (1984, October 3). Campaign Notes: County Official Wins Florida

Runoff

Election.

New

York

Times

[Online].

Available:

http://query.nytimes.com/gst/fullpage.html?res=9C0CE0DB123BF930A35753C1A9629482 60&n=Top%2fReference%2fTimes%20Topics%2fOrganizations%2fR%2fRepublican%20P arty . United States Congress. (2005). Biographical Directory of the United States Congress, 17742005. Washington, DC: U.S. G.P.O. United States Congressional Biographical Directory. 2008. Available at: http://bioguide.congress.gov/biosearch/biosearch.asp . Last accessed July 9, 2008.

United States House of Representatives, Office of the Clerk. 2008. “Official House Information: House Floor Proceedings.” http://clerk.house.gov/index.html , Last Accessed June 30. United States House of Representatives, Office of the Majority Leader. 2008. “U.S. House Vote Schedule.” http://majorityleader.house.gov/docUploads/2008-CALENDAR.pdf , Last accessed June 30 2008. Verzichelli, Luca (1996). I gruppi parlamentari dopo il 1994. Fluidità e riaggregazioni. Rivista italiana di scienza politica 26(2), 391-413. Yoshinaka, Antoine (2005). House Party Switchers and Committee Assignments: Who Gets

45

‘What, When, How? Legislative Studies Quarterly, 30(3): 391-406. Zucchini, Francesco. (2001). La commissione affari costituzionali: Gli effetti paralizzanti del mutamento. In Giliberto Capano & Marco Giuliani (Eds.), Parlamento e processo legislativo in Italia. Continuità e mutamento. Bologna: Il Mulino.

46

47

Endnotes 1

Here too any Independents might choose to join an organized legislative group. Some

legislatures permit an Independent group. Note that many scholars have omitted what we call stage A from analytical scrutiny, for research has traditionally focused either on the election or on the legislature after it has convened for its first session. 2

Institutions determine whether the official electoral campaign occurs near the end of an

ongoing term or after the legislature is dissolved. We focus on behavior during the term. 3

MPs who foresee defeat should also move in E. Testing this notion requires fine-grained data

on MPs’ electoral performance and so lies beyond the paper’s scope. 4

Given these similarities, smaller institutional differences do not undermine the case selection.

Whereas Russia permitted independents in SMDs, Italy from 1993 to 2005 required that every SMD candidate be linked to at least one PR party list. In Italy, not Russia, SMD wins were compensated in the PR tier (Katz 1994; Moser 2001). Russia had equal shares of SMD and PR MPs, and in Italy about 75% of MPs came from SMDs. PR thresholds were 5% in Russia and 4% in Italy. 5

Rich parliamentary documentation underpins the detailed analysis performed for primary cases.

6

With Mainwaring (1999; cf. Mainwaring and Scully 1995; Mainwaring and Torcal 2006), we

define a highly institutionalized party system as one displaying stable patterns of party competition, strong linkages between parties and voters, substantial legitimacy for parties, and solid party organizations. 7

We do not examine security policy in Italy, since internal warfare appears only in Russia.

8

According to the Chamber’s internal rules (Arts. 14 and 15), the Mixed Group enjoys much the

same rights and privileges as do party groups.

48

9

The evidence also points to midterm peak in switching, which we explore in Mershon and

Shvetsova (2008a, 116-118) and Mershon and Shvetsova 2008b. 10

To rephrase, we count as US switchers those who win election on something other than a major

party label and then affiliate with one of the two major parties before January 3 of odd years, when Congress is convened and roll-call votes can be held. We thus identify switchers whose behavior is not captured in the authoritative Poole-Rosenthal roll-call vote database (available at www.voteview.com; cf. Nokken this volume; Nokken and Poole 2004), which highlights the capacity of our theoretical approach to bridge what is often a disconnect between studying elections and studying Congress in US politics. All the same, since we focus on choices made while politicians hold national legislative office (not, e.g., non-elective posts), our data on the US omit a few switchers examined by Nokken in this volume. 11

candidates elected as independents or on minor party labels account for 11 of 12 switches in

the US stage A and 13 of 16 switches in stage B. Recall that stages A and B overlap but do not coincide for the US and for most other country-term cases as well. 12

This statement holds true when the Russian weekly observations are converted to a monthly

format, despite the extremely mobile Russian independents in stage A. 13

To place parliamentary parties in policy space, we rely on expert judgments (Birnir 2004;

Chiva 2007; Crewe and King 1995; Di Scala 1988; Lovatt and Lovatt 2001) and, as available, roll-call voting data (on Italy, see sources cited in Heller and Mershon this volume, Ch. Zz; on Russia, the Duma statistical service [INDEM]). The well-known Manifesto data (e.g., Budge et al. 2001) are inappropriate for our purposes because they pertain to units in electoral competition, not parliamentary groups founded at particular stages during a term.

49

14

Moreover, small parties could hope to survive in electoral contests using pure PR, such as EP

elections. During the 1996-2001 national legislative term, Italy’s electoral arenas were structured by a total of seven distinct electoral systems (Corbetta and Parisi 1997, 14).

50