Abstract
Why would a large incumbent party, that can by itself muster more than 45 per cent of electoral support, add seven insignificant political parties to its electoral list, thus providing them with a free ticket into the Assembly, state-sponsored financing for the next four years, and an independent deputy club in the Assembly? More importantly, why would the incumbent party, as a consequence of this deliberate decision, end up without a parliamentary majority? In this article, I discuss less frequently mentioned aspects of institutional design that can help us understand why the incumbent party makes such decisions and why a party system in a hybrid regime has a large number of parties. I offer qualitative evidence from postcommunist Serbia (1990–2018). The first two aspects refer to electoral design (but not the electoral formula or district magnitude): a larger number of parties under the incumbent party’s control provides stronger political influence in the Republican Electoral Commission and at polling stations on election day. In hybrid regimes, the incumbent party is interested in this influence because it can use it to arrange electoral fraud. The third aspect relates to parliamentary design: more parties in the Serbian Assembly under the incumbent party’s control secure more minutes on the floor for the incumbent party during parliamentary debate.
Introduction
At the 2016 snap Serbian parliamentary elections, the electoral list of the Serbian Progressive Party (SNS), headed by Aleksandar Vučić, won 48.25 per cent of the popular vote. This gave it 52.4 per cent of the parliamentary seats (131 seats out of 250). However, in the end, the SNS did not take up all of them—it had to renounce 36 seats in favor of the seven smaller parties that subsequently agreed to form a governing coalition with the SNS. Four of these seven parties formed deputy clubs 1 in the Serbian Assembly, which gave them relative autonomy in parliamentary work vis-à-vis the SNS.
The seven small parties did not appear in the Assembly out of the blue: the SNS brought them into the Assembly by inviting them to join their electoral list. 2 At the time the elections were held, the SNS was the incumbent party, and Aleksandar Vučić was by far the most popular Serbian politician. Vučić and the SNS could have probably won the 48.25 per cent all by themselves. Why would an incumbent party, that could muster over 45 per cent of electoral support on its own, 3 add seven insignificant political parties to its electoral list, thus providing them with a free ticket into the Assembly, state-sponsored financing for the next four years, 4 and an independent deputy club?
The puzzle does not end here. Since the Serbian electoral system includes a 5 per cent electoral threshold, these seven parties would have never been able to make it into the Assembly without this substantial help from the SNS. 5 The move to bring them in appeared particularly risky given the fact that because of the agreed seat distribution, the SNS itself ended up 28 seats short of an absolute majority (126 seats). If these seven parties had not been invited onto the SNS-led list, the SNS would have kept the 36 seats for itself. With only 98 parliamentary seats, the SNS was forced to form a coalition with minor and much smaller political parties, making its governing majority vulnerable to blackmail.
This is not a novelty in Serbian political party life. The practice was started by the Democratic Party (DS), which put together several such electoral lists in 2007-2012 when it was the incumbent party. In 2008, the DS-led electoral list was composed of five additional smaller political parties. In 2012, the DS produced another similar list of five smaller coalition partners. This looks like a large party working against its own interests. Why would an incumbent party take such a risky and potentially counterproductive move?
The quick answer is that this represents an investment strategy aimed at maintaining the hybrid regime. The strategy implies that the incumbent is prepared to renounce a part of its influence and power (in terms of giving up parliamentary seats or other patronage positions), while small and insignificant political parties return the favour by participating in electoral fraud and other informal activities that are important in providing democratic legitimacy for the incumbent party.
The discussion of this strategy will reveal several things. First, it will emphasize the role of so-called phantom political parties and decoy oppositions in hybrid regimes. 6 Such parties play an important role in helping the incumbent party rig the elections and ensure its domination of parliament, while maintaining a democratic façade. This discussion is also relevant to similar cases, since most recent reports on the state of democracy around the world indicate that the number of non-democratic (hybrid and autocratic) regimes is on the rise. 7
Second, this discussion extends the debate on how political parties can be considered to be relevant. It will demonstrate that the number of relevant parties is much higher than usually appears at first glance (measured by several conventional indices) because decoy and phantom parties that never make it into parliament or stand absolutely no chance of winning more than 0.5 per cent of the vote on their own can be regarded as relevant. (In Sartori’s view, only parliamentary parties are relevant. 8 )
A note on case selection. I selected Serbia as a case study because certain aspects and practices I discuss here cannot be found elsewhere. As will be seen, the role of phantom political parties (which may or may not be part of the opposition) in the Central Electoral Commission, polling stations, and on the parliamentary floor appears to be a specific peculiarity of Serbia, but could soon surface in other hybrid regimes. 9 I do not want to claim that these are the only aspects that deserve attention, but rather that the informal dynamics within these institutions are difficult to find and observe in other countries. Granted, Serbia is also an excellent example of vote buying, repression, and electoral hacking, but in these regards it offers few new insights, which is why I do not discuss these instances in this article.
The rest of the article is organized as follows. In the next two sections, I place the discussion within the theoretical context of informal institutions and hybrid regimes. The following sections give a detailed account of how the Serbian incumbent party benefits from a large number of smaller and oppositional parties in the electoral and parliamentary process. Finally, I offer a short conclusion.
Theoretical Framework and Conventional Approach
Before I begin, I want to place the debate within the broader theoretical context of research on informal institutions and hybrid regimes. Informal rules sometimes “complete or fill gaps in formal institutions, and operate parallel to formal institutions in regulating political behaviour.” 10 Informal rules are pronounced in hybrid regimes and flawed democracies that have recently also been described in terms of competitive or electoral authoritarianism, or counterfeit democracies. 11 For one thing, such regimes have to look like democracies, so they embrace an ostensibly democratic institutional design, underneath which completely opposite (nondemocratic) practices can be observed. As Puddington states, such regimes are based on the illusion of democracy. 12 This informal institutional scheme resembles the concept of layering, as advanced by Mahoney and Thelen, under which new rules exist parallel to existing ones, “thereby changing the ways in which the original rules structure behaviour.” 13
Semi-authoritarian systems hold elections with the participation of the opposition (hence the adjectives “competitive” and “electoral”), but the electoral process is so rigged that the opposition cannot win. Electoral fraud and vote-rigging are crucial aspects of such regimes 14 (though not the only identifying factors 15 ). Counterfeit democracies use various means—electoral bribery, electoral hacking, fake news production, ballot-box stuffing 16 —to accomplish this goal. One of the administrative institutions that plays a key role in this process is the Electoral Commission. Most recent interest in this institution can be recognized from a normative perspective in which commissions are supposed to “play watchdog,” warranting the integrity of the electoral process in a democracy. 17 In hybrid regimes, by contrast, electoral commissions are recognized as a part of a broader corrupt electoral scheme and are understood as rubber-stamp agents and another channel of influence for the incumbent party, 18 but they rarely receive detailed attention in the current literature on such regimes. 19 Even when the commission is not a government body appointed by the executive, 20 incumbents can still exert substantial influence to make the commission help them rig elections. Serbia is a compelling case because such influence is not only direct (for instance, via appointments and bribes), but is rather achieved with the help of phantom and decoy political parties who participate in the Serbian Electoral Commission and supply polling workers who conduct elections on the ground. I devote a significant part of this article to the work of the Serbian Electoral Commission and polling workers precisely because the rigging scheme that can be observed here supplies new mechanisms (including decoy parties) that are of value for a theoretical conceptualization of the electoral process in hybrid regimes.
Decoy parties are not only relevant to the functioning of electoral commissions, but are also part of a broader electoral and parliamentary scheme. To manipulate the electoral process to prolong its stay in office, the incumbent party has to produce a broad coalition of agents and a complex informal web of policy concessions and spoils that are given out to the “minor” or seemingly irrelevant political parties, some of which are unable to clear the electoral threshold on their own. 21 If such parties do not exist, the incumbent will help to create them, thus increasing the total number of incumbent and opposition parties. 22 Therefore, political parties in such regimes often do not emerge as genuine societal movements but are rather created by the incumbent party and are kept under its control by co-optative and clientelistic arrangements. 23 Co-optation is, therefore, one of the key components in such regimes—to mimic democracy (electoral competition and parliamentarism), an incumbent party needs to incorporate as many obedient and controllable agents as possible in the “democratic” process.
When Is a Political Party Relevant?
There is yet another aspect I would like to introduce. The discussion about how small parties help the incumbent maintain the hybrid system casts light on how we understand and are to count the number and the relevance of political parties in a party system. As shown in some empirical research, the change in number of parties affects the fluidity of the party system, which then affects the nature of the regime. 24 In 2018, there were 118 registered political parties in Serbia. 25 Most of these are phantom or decoy parties, and many exist only on paper. They have neither staff, nor an organization (party structure), nor voters. Their offices are usually empty, with nobody to answer the phone. Sometimes they do not even have websites where one can find out about their broader leadership and party organs (main committee, executive committee, party assembly, etc.). Some small parties become active only when elections are called. Their electoral success is so low (between 0.1 and 0.5 per cent of the votes) that some of them receive fewer votes than the number of signatures required to be able to participate in elections (see Table 1). They disappear from the public eye shortly after the elections, albeit remaining officially active until the next elections.
Electoral Lists That Received Fewer Votes Than the Required Number of Signatures for the Candidacy
Are they all relevant? They are, I claim. However, conventional methods of measuring their relevance may suggest otherwise. Consider, for example, the Taagepera & Laakso index of effective parties applied to Serbia. Figure 1 displays the effective number of parliamentary parties (ENPP) in Serbia in 1990–2016, and effective number of elective parties (ENEP) in 2000–2016, 26 but also the number of deputy clubs (measured by no index) for 1990–2016. The indices assume that the effective number of parliamentary and elective parties in 2016 was fewer than five, and has never been more than seven. Granted, such indices are created to avoid the trouble of counting all parties that participate in elections. 27 However, such indices may be misleading as they may fail to fully capture the role of such parties in the system, notably for the incumbent in maintaining a democratic façade. For example, the ENPP index assumes that parties which are not represented in parliament are not relevant. This is what we learned from Sartori—such parties can neither blackmail anyone nor form a governing coalition. 28 (Under Giovanni Sartori’s view, a relevant political party is one that has a coalition and/or blackmailing capacity. 29 ) Yet, I claim that hybrid regimes such as Serbia exemplify dynamics that neither the index nor Sartori’s classification is fully able to capture. Figure 1 confirms this claim: while the ENPP and ENEP remained more or less stable over three decades, the number of deputy clubs in parliament rose sharply on two occasions—in 1993 and after 2000. There must be an explanation as to why the number of deputy clubs keeps rising, but ENPP and ENEP do not help us answer this.
ENPP, ENEP, and Deputy clubs in Serbia 1990–2016
To overcome the insufficiencies of the index-based measurement, we could draw on Sartori’s idea of focussing on qualitative rather than quantitative properties of parties and party systems. 30 In hybrid systems, there can be a number of political parties that do not enter parliament but are nevertheless essential to the incumbent in ensuring the continuation of the status quo. Such parties may be highly relevant for a hybrid regime’s incumbent party when it comes to staging elections and dominating debate on the parliamentary floor, as we shall see below.
As already mentioned, the creation and maintenance of such parties constitutes a significant part of the political investment the incumbent is prepared to make in order to keep the system going. The incumbent does this by bringing them into national or local assemblies, from which the parties then receive public funds—a practice seen in most Balkan and East European countries. 31 Another way to co-opt a political party is to invite one to participate in some level of the public administration, even if they are not represented in parliament. This is called party patronage, a system that permits political parties to staff public sector institutions (public administration, utility and public enterprises) with their rank and file in order to extract resources to ensure their position. 32 Be that as it may, the informality here ensures that such favours are returned. Once minor political parties accept the incumbent’s offer to be included in the party patronage system, they are obligated to assist the incumbent in various informal activities, the most important being rigging the elections and helping ensure the incumbent’s dominance on the parliamentary floor. Such parties may be small, non-parliamentary, or irrelevant under Rae’s and Taagepera & Laakso’s indexes, and even under Sartori’s criteria, but they are indeed crucial for the incumbent in maintaining a democratic façade. They should all be counted as relevant.
The Republic Electoral Commission
I begin with the Republican Electoral Commission of Serbia. Its main task is to conduct the electoral process by eliminating irregularities. Each political party can participate in the Serbian Assembly elections (alone or in coalition) by submitting an electoral list to the Republican Electoral Commission. The Electoral Commission is empowered to make critical decisions about the electoral process. It decides on the number of electoral lists, declares the outcome of the elections, and adjudicates on complaints about the electoral process. All these decisions are made by simple majority rule.
To participate in the Serbian Assembly elections, each electoral list is required to collect and submit ten thousand signatures. 33 Signatures are usually given by voters who are already members of or support the list. But it is the Commission that decides if the signatures are valid, and accepts (or turns down) the list by a majority vote. Complaints about electoral process are decided in an identical manner. The most frequent complaints refer to various violations of voting procedure at polling stations. 34 When party representatives file a complaint with the Commission, the Commission usually rejects it if it does not fit the political interests of the Commission’s majority, those who won the elections. For example, after the 2018 Belgrade elections, the City Electoral Commission rejected 196 complaints about electoral irregularities, and accepted only one. 35 Therefore, whoever wants to ensure that they will not be handicapped by the Commission’s decisions will have to make up or control a majority in the Commission.
The commission does administrative jobs but it is composed of politicians, not of administrative professionals. It has 16 standing members who are the representatives of the parties and lists that already have their representatives in the Assembly based on the outcome of the previous elections. However, as the new electoral process begins and as new electoral lists are accepted, the Commission adds new members. When the Commission accepts a new electoral list, this list can nominate a new representative to the Commission, who then becomes a full member during the electoral process (leaving the Commission only if their list does not clear the threshold). Each new member’s vote has the same weight as the vote of any other member. The commission can literally disqualify an electoral list from electoral competition. If the party representatives that make the commission’s majority collude to disqualify a competitor, the competitor’s list cannot participate.
Not only are new participants handicapped by such a design but the incumbent party can be handicapped, too. If the incumbent does not control the majority in the commission, its decision-making mechanism can put them at a disadvantage. Imagine a party system of four political parties—A, B, C, and D. Suppose the ruling party A enjoys the support of 60 per cent of the electorate. Under this institutional arrangement, this party will have only one representative in the Commission. The oppositional parties B, C, and D are supported by the remaining 40 per cent of the electorate. They will receive three representatives in the Commission, thus outnumbering the incumbent by 3:1. The opposition represents a minority of the electorate, but it can decisively determine the electoral outcome by outvoting the incumbent in the Commission by 3 to 1.
How can the incumbent party avoid this outcome? To ensure that it has a majority in the Commission, the ruling party A either has to accept more parties (say, B and C) into the government, or it may help to create a number of smaller parties that will participate in the elections and become members of the Commission even if they do not get into government and do not even clear the threshold. It is this calculation that leads the ruling party to create and help minor parties participate in the elections. The incumbent party provides small parties with resources such as money, poll workers, and signatures to be able to run for parliamentary posts. These parties, when they become members of the Commission, return the favor by voting in agreement with the ruling party when it is necessary to eliminate somebody from the electoral process, or declare an electoral outcome that will be in the incumbent party’s political interest.
Can we be surer about this connection between the “insignificant” decoy parties and the incumbent party in Serbia? We observe that insignificant parties receive fewer votes than the number of signatures or poll workers they register for the election. The people who gave signatures or applied to volunteer as poll workers are apparently not voters or members of these lists, but of some others. Smaller parties received them from some other organization with a much higher mobilizing capacity. Consider Table 1, which supplies details about ballots and signatures for two electoral lists from the 2016 parliamentary elections.
The two lists—Republican Party and Dialogue—received far fewer votes than the required number of signatures they collected from their members or supporters before the elections. 36 To collect ten thousand signatures, a party needs to have quite a strong organization that will mobilize voters and supporters for this political action. As mentioned above, the parties that submitted these lists are inactive between elections. The only way to collect the required signatures for submission to the electoral list would be to receive help from the ruling party, which controls the required manpower and can lend it to minor parties. For instance, in 2017, the Serbian Progressive Party (SNS) had a membership of around seven hundred thousand people, which far exceeds the average party membership in Western democracies’ party systems. 37 This practically inexhaustible resource pool was enough to produce signatures for its own and an additional hypothetical sixty-nine electoral lists.
Poll Workers at Polling Stations
We have all seen film footage from the 2018 Russian presidential elections in which poll workers stuff ballot boxes with fake ballots, thus producing another landslide electoral victory for Vladimir Putin, with a 77 per cent majority. 38 For hybrid regimes, the control of the ballot box (ballot-box stuffing, ballot-counting) is essential, and is seen as the last resort. 39 Under the Serbian legislation, the count takes place at the polling station immediately after the elections are over (8 p.m.). In the 2016 parliamentary elections, Serbia had 8,387 polling stations. The Serbian administration does not have professional staff for electoral process, but rather asks each electoral list to supply up to two poll workers (so-called controllers) for this job at each polling station. Each list thus has to produce up to 16,774 volunteers who oversee the electoral process, guard the ballot box, and count the ballots at the end of the election day. 40 Polling station staff are thus composed of political party supporters who want to serve as poll workers during the election day.
Now there comes another puzzle. Each electoral list is supposed to mobilize and select poll workers from the people who support the list and would vote for it. (This is why they volunteer to guard the ballots for the list they support.) Yet, in 2016, there were five electoral lists that received fewer votes than the number of poll workers they ostensibly mobilized for the job (Table 2).
Electoral lists that received fewer votes than the required number of poll workers for the candidacy
Since these parties in fact do not exist, it is likely that poll workers were selected from among the ruling party’s members. In this case, the SNS could again lend its membership, thus producing poll workers for its own and an additional forty electoral lists. Again, why would the ruling party do something like this? The reason is to have tighter control of the electoral process on the ground. Imagine again the situation with four parties (A-D), in which the ruling party A has only two poll workers, and the oppositional parties have six poll workers altogether. Under these circumstances, the incumbent is unable to rig the count. Imagine now that the incumbent helped create another sixteen electoral lists, supplying additional thirty-two poll workers per polling station. In this case, the ruling party controls thirty-four poll workers, whereas the opposition has only six. When the ballots are counted in a 34:6 environment, it is physically easier for the ruling party to manipulate the count and produce electoral fraud.
The advantage such a ratio provides is ballot stuffing and recording of complaints. When party A controls thirty-four out of forty poll workers, they can more easily stuff ballot boxes without being noticed. Secondly, any irregularity, to be processed by the Republican Electoral Commission and Administrative Court at a later point, must be recorded in the polling station register. Permission for this is given by the polling station committee presidents. For example, if a voter turns up to the polling station without ID and is permitted to vote, any member of the committee can file a complaint about this irregularity. But if the president of the polling station committee does not accept it and does not enter it in the record, the complaint will not be registered and will be considered neither by the Commission nor by the Court. The majority poll workers can easily scare minority poll workers away from the polling station if they dare to report irregularities. The same can occur in securing storage of ballots and tabulating the results. Such types of irregularities have been a regular practice in the Serbian elections after 2012. 41 CRTA, an NGO that monitors elections in Serbia, reported such irregularities at 8 per cent of the polling stations in the 2018 Belgrade elections. 42
Serbian Assembly: The Rules for Forming Parliamentary Groups
Since the Serbian incumbent party wants to have as many parliamentary parties under its control as possible, it is prepared to add some of them to its own electoral list, even if their contribution to electoral results is negligent. Most of these parties could barely win 0.5 per cent of the votes on their own. 43 Given the electoral threshold of 5%, a coalition of ten such small parties would make sense. But why would any party that can win more than 40 per cent on its own bother to have such parties in a “puller” coalition? 44 What creates such an interest? To understand this part of the argument, it is essential to differentiate between a parliamentary political party and a deputy club. These may not be the same. When a party wins seats in the Serbian Assembly, it can form a deputy club. It is important to form a deputy club in the Serbian Assembly, because the House Rules Act recognizes only deputy clubs, not parliamentary political parties.
A deputy club must include at least five deputies (members of parliament). It does not really matter if they are from the same or different parties. (Members of such clubs may even be members of no party.) Sometimes one parliamentary party forms a single deputy club; but if several parties entered the Assembly on a single electoral list as a coalition, they, as a rule, will split off into several deputy clubs. For example, the DOS coalition that won the December 2000 parliamentary election was made up of 18 political parties. It began as a single deputy club, but within three years, the club split into seventeen separate deputy clubs. 45 The Serbian Progressive Party ran for the 2016 elections with an electoral list that contained eight political parties altogether. Five of these parties formed separate deputy clubs in the Assembly after the elections. Therefore, one electoral list will produce at least one deputy club, but usually produces much more than one. 46 In 1990–2016, the number of deputy clubs clearly dominates the number of electoral lists that cleared the 5 per cent threshold, and is rising (Figure 2). 47
Electoral Lists That Cleared and Deputy Clubs in Serbian Assembly (1990–2018)
It is possible to pin down the precise historical moment when the trend for deputy clubs to grow emerged in the Serbian Assembly—this was 10 February 1994. What happened on that day? In the December 1993 parliamentary elections, the SPS won 36.6 per cent of the votes, which gave it 123 seats in the Assembly (almost 50 per cent). To form a cabinet, the SPS needed the support of at least three more deputies (to get to 126 votes). At first, nobody was prepared to provide these three extra votes—until six deputies from a small opposition party called New Democracy (ND), led by Dušan Mihajlović, did so.
ND had entered the Assembly on the electoral list of DEPOS (Democratic Movement of Serbia, which was itself a coalition of three political parties). The list was dominated by the Serbian Renewal Movement (SPO), which was the largest opposition party at the time. ND had only six deputies in the Assembly. The House Rules Act at the time required ten deputies to form a deputy club. The six ND deputies wanted to support the SPS-led cabinet, but did not want to belong to the SPS deputy club. They did not want to belong to any oppositional deputy club either. They wanted their own club in the Assembly, because a separate club gives visibility and enables autonomous performance on the parliamentary floor. (Recall that the Assembly Acts recognize only deputy clubs, not political parties as parliamentary agents.) With their own separate deputy club, the ND deputies could participate in general discussions as members of ND, not as members of some larger deputy club or coalition. It was also in the SPS’s interest to help New Democracy get its own deputy club—with a separate ND club, the SPS-led cabinet would have the support of two deputy clubs rather than only one during general parliamentary discussion.
But how could they form a separate deputy club with only six deputies when the Act required ten? On 10 February 1994, the new parliamentary majority changed the House Rules Act, which now required five rather than ten deputies to form a separate deputy club. New Democracy was now able to form its own deputy club, after which the new SPS-led cabinet was voted in on 18 March 1994. 48
In 1993, the number of deputy clubs rose because of the institutional redesign, but then fell in 1997 as a consequence of the electoral boycott. 49 After 2000, the rise in the number of deputy clubs is a consequence of the opportunity to speak and to dominate the debate during the first reading. The more deputy clubs there are in the Assembly, the more time is provided for the first reading debate. The incumbent party may control a majority of the seats in the Assembly (and will thus muster majority support for any bill), but may still be dominated by the opposition during this debate in terms of the number of minutes allotted. The only way for the incumbent to counter this is to have more deputy clubs under its control in the Assembly. I address this problem in the next section.
Three Types of Parliamentary Debate
The House Rules Act (2012) of the Serbian Assembly recognizes three types of debate for each bill after the proponent of the bill (usually, the minister) presents it to the floor. I focus here only on the type that is most popular with the public—the general debate (first reading), which is usually how each session opens. 50
As mentioned above, the Act recognizes only deputy clubs and the general debate is reserved only for the heads of deputy clubs, not parliamentary political party leaders. The rules for the first reading allocates each deputy club an equal amount of time for the debate (20 minutes) irrespective of the club’s size. The total time allocated to this debate varies. Its length depends on the number of deputy clubs in the Assembly. The more deputy clubs, the longer the first reading debate.
To see why it is important for the incumbent to control as many parties and deputy clubs as possible, consider again the hypothetical example above. The four parliamentary parties, A, B, C, and D, share the Assembly’s 100 seats. A is the ruling party and occupies 60 seats, while B, C, and D are the parliamentary opposition, occupying 40 seats. Suppose that each party forms only one deputy club, meaning there are four deputy clubs in the Assembly altogether. Since each deputy club’s head is allocated 20 minutes for discussion, the total length of the opening debate will be 80 minutes. But the incumbent party will occupy only a quarter of the debate, whereas the opposition will dominate three quarters of the debating time. What can the incumbent party, which holds 60 per cent of the seats, do to reverse this disadvantage? Hypothetically, party A can split into 30 different deputy clubs (provided each club has 5 deputies). This radically changes both the length and the composition of the debate. The total length of the debate is now 660 minutes (rather than 80), and 91 per cent of the debate will now be dominated by party A, which controls 30 deputy clubs, whose heads can talk for 20 minutes each. The three opposition parties will be almost unnoticeable in this debate, allocated a mere 9 per cent of the debate time.
Let me now give a real example from the Serbian Assembly. (Due to the lack of space, I focus only on the 2016 parliamentary elections.) The Assembly elected in April 2016 consisted of 16 deputy clubs, meaning that the general debate’s total length could be up to 320 minutes. The Serbian Progressive Party (SNS) was the strongest party (98 deputies), but its head could talk for only 20 minutes. However, the SNS was elected on an electoral list together with seven other parties, four of which formed their own deputy clubs—the Social Democratic Party of Serbia, the Party of the United Pensioners, the Coalition of Socialists and Farmers, and New Serbia. 51 None of these parties had ever participated in elections by themselves. We can now understand the interest of the SNS, as the largest party, in bringing minor parties into the Assembly. Put simply, it needs more minutes on the floor. By introducing four smaller parties into the Assembly, each of which can form a separate deputy club, the SNS can control a larger chunk of the debating time. Rather than having only 20 minutes, the SNS now controls 80 minutes in the general debate. 52
This is not all. The SNS is not alone in the Serbian Government, but is in coalition with two other parties. The Socialist Party of Serbia and United Serbia ran on a separate electoral list as a coalition, but formed two separate deputy clubs in the Assembly after the 2016 elections. They all act in concert. So, altogether, the ruling coalition is backed by six deputy clubs, and occupies 120 minutes of the general debate in the Assembly.
This is still not all. The SNS forms a ruling coalition at the local and city levels with several parties that have seats in the Serbian Assembly. For instance, in 2016, the Serbian Radical Party (from which the SNS splintered in 2008) entered a ruling coalition with the SNS in the city hall of Niš (the third largest city in Serbia, with a budget of €85 million for 2018). The same year, the League of Social-Democrats of Vojvodina formed a ruling coalition with the SNS in the city hall of Novi Sad (the second largest Serbian city, with a budget of €207 million for 2018), while the Liberal Democratic Party is in a ruling coalition with the SNS in the municipality of Vračar (one of the richest Belgrade municipalities, with a budget of €4.2 million for 2018). The Alliance of Vojvodina Hungarians made a ruling coalition with the SNS in the province of Vojvodina’s Assembly (with a budget of €580 million for 2018). These political parties, even if formally in opposition in the Serbian Assembly, are a part of the SNS’s informal incumbent network at other levels, and frequently support the SNS verbally during debate (although they do not always vote for the Government’s bill, as their vote is unnecessary in any case). Therefore, another three “oppositional” deputy clubs 53 add another 60 minutes, which now makes a total of 180 minutes for the SNS-led coalition. In contrast, the whole opposition, divided among seven deputy clubs, occupies 140 minutes of the general debate (Table 3).
The Number of Deputy Clubs in the 2016 Serbian Assembly
Note: The depicted distribution of deputies is from mid-June 2018.
The total number of deputies does not always add up to 250 (the number of seats in the Assembly) because some deputies are not members of any club.
The façade opposition parties in the Serbian Assembly are not only coalition partners with the incumbent SNS at the local level, but confirm their allegiance to the incumbent either by regularly praising SNS-leader and Serbian President Aleksandar Vučić, or by entertaining fake debates with Vučić, or by criticizing everyone but Aleksandar Vučić.
For example, in August 2018, Čedomir Jovanović, the leader of the Liberal Democratic Party, formally in opposition in the Serbian Assembly, criticized the rest of the opposition for “being backward for not understanding how progressive Aleksandar Vučić is.” 54 Jovanović often praises Vučić for building a European Serbia and offers him help by pointing out that he (Vučić) cannot implement the requirements of the EU enlargement policy alone. 55 Nenad Čanak from the Alliance of Vojvodina Social-Democrats said in 2013 that Aleksandar Vučić was “a serious politician and a man who keeps his promise.” 56 The only “spirited” debate between Čanak and Vučić in the Serbian Assembly has been a quibble over the African turtledove hunting ban in 2015, a minor policy issue in the government public policy. 57 Vojislav Šešelj from the Serbian Radical Party (from which Vučić’s SNS splintered in 2008) frequently entertains fierce debates with the Serbian cabinet’s ministers and cabinets’ public policy, but never criticizes the hybrid democratic design and never personally attacks Vučić himself.
Cooperative opposition leaders are critical for the maintenance of the regime, and the incumbent attempts to keep such people in the game, often at high costs. For example, Vojislav Šešelj was tried for war crimes in the former Yugoslavia by the Hague Tribunal in 2003–2015. On 11 April 2018, he was sentenced to ten years for hate speech that implied “instigating deportation, persecution (forcible displacement) and other inhumane acts (forcible transfer) as crimes against humanity.” 58 Although Serbian electoral law explicitly forbids persons with criminal records to become or remain deputies in the Serbian Assembly, the SNS-controlled majority in the Assembly has refused to strip Vojislav Šešelj of his mandate. The Republican Electoral Commission even allowed him to run in the 2020 parliamentary elections.
Conclusions
The article has looked into several practices from the Serbian electoral and political party system that are relevant not only for Serbia but also for other similar (hybrid) regimes around the world. I discussed the phenomena of decoy and phantom political parties in helping the incumbent party dominate the Serbian Electoral Commission, polling stations, and debate in the Serbian Assembly. The central claim is that such parties are essential for the incumbent party to maintain a façade democracy. The incumbent is, therefore, prepared to invest some resources to create and maintain such parties. The investment can be measured by the limited number of parliamentary seats and public sector positions the incumbent is prepared to give up and hand over to the leaders of these parties.
I also discussed the issue of why the number of political parties and deputy clubs keep increasing as a consequence of such practices. The Serbian experience shows that a political party can still be regarded as relevant even though it would never be able to clear the threshold, nor win more than 1 per cent by itself, nor be part of a cabinet, nor blackmail anyone. The number of small and “irrelevant” parties increases because of the incumbent party’s preference for controlling the system by a whole range of smaller parties that exist only on paper. To use Gandhi and Przeworski’s phrase, a larger number of such cooperative political parties helps the incumbent party stay in office. 59 These small parties are relevant not on account of their parliamentary strength (which is almost nil), but because they can be an instrument of government in a broader sense. Such parties can be valuable partners for anyone, but especially incumbent parties. They may remain outside of the cabinet, and even outside of the assembly, and yet be regarded as relevant.
Does this mean that Sartori’s definition of relevance must be modified? Probably not. Sartori had in mind political parties in consolidated democracies, though he discussed the party systems of hybrid or nondemocratic regimes (such as PRI-led Mexico or communist Poland). Since in such regimes elections—which may be competitive, but unfree and unfair—produce different dynamics in party competition, the criteria for a party to be relevant are different. Therefore, to Sartori’s definition that a political party is relevant only if it can “determine over time, and at some point in time, at least one of the possible governmental majorities,” 60 we could add that, in flawed democracies, relevant parties are those that can help already existing majorities accomplish some other critical political goals, such as control of the electoral process and domination of the parliamentary floor.
Why have the factors discussed here not been considered in other research? One reason is that they may be country-specific and cannot be covered by statistical analysis. As I mentioned, I have not observed such practices in other hybrid regimes. The second reason is more important. The list of factors that are usually used in research to determine the level of fragmentation—district magnitude, electoral threshold, assembly size, etc.—are those that can be easily quantified. The same goes for the type of electoral system, federalism, social and ethnic cleavages, or legacy of conflict, all of which can be expressed in the form of a dummy variable. In contrast, it is hard to quantify the determinants I discuss in this article. Reliable data about decoy and phantom political parties are usually hard to obtain. How do we know for sure if decoy parties’ membership exists only on paper, or if real people are involved? Its financing is also non-transparent. Therefore, we can never be sure if the political party is created and sponsored by the incumbent, or emerged spontaneously, if a party cooperates with the incumbent party out of conviction, or because it was created by the incumbent to help it stage the elections. Some of these “facts” may remain unknown and difficult to measure, but the existence of such parties will nevertheless affect the party system’s dynamics and our ideas about party relevance. If we cannot be certain about the nature of phantom and decoy political parties, the most we can do is to measure the number of signatures, seats, and public sector positions the incumbent is willing to lend or give up for such parties to function.
