From parliamentary pay to party funding: The acceptability of informal institutions in advanced democracies

Authors


  • [Correction added on 6 August 2015, after original online publication: The copyright line was updated.]

Abstract

While direct state funding of political parties has been a prominent theme in cross-national research over the last decade, we still know little about party strategies to access state resources that are not explicitly earmarked for partisan usage. This article looks at one widespread but often overlooked informal party practice: the ‘taxing’ of MP salaries – that is, the regular transfer of fixed salary shares to party coffers. Building on notions of informal institutions developed in work on new democracies, the theoretical approach specifies factors that shape the acceptability of this legally non-enforceable intra-organisational practice. It is tested through a selection model applied to a unique dataset covering 124 parties across 19 advanced democracies. Controlling for a range of party- and institutional-level variables, it is found that the presence of a taxing rule and the collection of demanding tax shares are more common in leftist parties (high internal acceptability) and in systems in which the penetration of state institutions by political parties is intense (high external acceptability).

Introduction

While the expansion of formal state funding and its implications for the working of political parties has been widely analysed in cross-national research, covering both old and new democracies (e.g., Katz & Mair 1995a; Biezen 2003a; Birnir 2005; Casas-Zamora 2005; Scarrow 2006, 2007, 2011; Roper & Ikstens 2008; Koss 2011), informal party strategies for accessing state resources in advanced democracies are much less subject to cross-national, comparative research (see, for exceptions, Peters 2004; Biezen & Kopecký 2007; Nassmacher 2009; Kopecký et al. 2012). Research on new democracies, in contrast, takes compliance with formal-legal rules less as a given and puts a stronger emphasis on the study of informal rules not sanctioned by the state (Grzymala-Busse & Jones Luong 2002; Grzymala-Busse 2003, 2008; Helmke & Levitsky 2004, 2006). Consequently, this literature has a lot to tell us about the study of informal practices at work in old democracies.

Drawing on these two literatures, this article explores a widely neglected informal strategy of parties to access state resources in the context of advanced democracies. It looks at whether and how much rent political parties can extract from their parliamentarians' salaries. Many parties have internal rules that oblige their members who enter public office on a party ticket to regularly donate a specific share of their salary to party coffers and thereby access state resources indirectly (e.g., Tsatsos 1992; Nassmacher 2001, 2009; Biezen 2003b). Country studies have long indicated that this phenomenon is widespread in advanced democracies. However, we not only lack a theoretical approach accounting for the ‘taxing’ of national parliamentarians across a broader cross-national scale that integrates organisation-centred and system-centred perspectives. We also lack information on the basic nature of financial redistribution from national parliamentarians to party, becoming manifest in the presence of a taxing rule as adopted by parties and the share of MP salaries collected on the basis of such a rule (the level of tax), across advanced democracies in the first place. Opening the ‘black box’ of intra-party politics across a wide range of parties and democracies based on so far unexplored data, this article addresses both gaps and thereby adds to our understanding of the informal dimension of party-state relations – a theme high on the research agenda, about which we still know fairly little.

Our analysis can draw on a new dataset covering (a) whether parties collect such taxes, and (b) how much they take from MPs' parliamentary salaries across 19 advanced democracies. Of the 124 parties covered in our analysis, 79 (63.7 per cent) collect party taxes from their MPs, spreading across 17 of the 19 democracies we examined. The tax shares range from 0.5 to 75 per cent of MPs' base salary, with an average tax share of 13.3 per cent. A brief look at the German parties included in our study further indicates that the contribution of party taxes to the overall income of parties (including local, regional and national levels) is significant. In 2008, contributions ranged from 6.6 per cent (Christian Social Union) up to 20.2 per cent (the Greens).1 With the monthly purchasing power adjusted salary of a German MP sitting in the Bundestag at the time being around €6,800, these requirements constitute monthly payments of approximately €450 and €1,380, respectively.2 This highlights not only that many parties have been able to introduce rather far-reaching taxing practices vis-À-vis their MPs, but that attention needs to be given to parties' varying capacity to exploit their representatives' salary as an additional source of party funding.3

Practices that access state resources informally have existed since parties started to operate in democratic institutions (Burnell & Ware 2009) and are often – in formal terms – legal. For citizens, however, legality is not necessarily a sufficient condition for considering practices acceptable (McAllister 2000; Allen & Birch 2012), which highlights the broader relevance of our study. Disapproval of ‘partisan politics’ and calls for de-politicisation have intensified dramatically in recent years (Dalton & Wattenberg 2002; Mair 2007) and, with them, attempts to regulate the use of resources party representatives in parliament or government can access when performing institutional roles (Allen 2011). This latter option does not exist when it comes to the ‘taxing’ of politicians' private salaries, which raises the question under which conditions this practice is considered acceptable and why.

Addressing this puzzle, we theorise conditions for the acceptability of taxing practices within parties as organisations and within distinct systemic contexts. Using Helmke and Levitsky's conceptualisation of informal rules that operate in the context of effective formal institutions without formally violating them (Helmke & Levitsky 2006: 15) allows us to integrate claims about intra-organisational and systemic conditions that affect parties' capacity to extract funding from MP salaries, which can be derived from classical approaches to (formal) party organisation (Duverger 1964; Katz & Mair 1995b, 2009; Biezen 2004). From an organisational perspective, internal acceptability is relevant simply because parties cannot rely on formal (i.e., legally enforceable) sanctions to assure their MPs' compliance with these informal rules. Simultaneously, normative tensions revealed by this practice raise questions around its external acceptability, which brings us back to the growing hostility against ‘partisan politics’. Motives for why democratic states have introduced MP salaries include the aim to attract qualified office-holders but also to attract representatives from all corners of society. They were not intended to constitute an additional source of party income. Without violating the law, the practice – when going beyond minor transfers – can thus work against the purposes of providing parliamentary pay. In some countries such as Germany, the practice was further reviewed by the courts as a potential violation of parliamentarians' free mandate (Nassmacher 2001), which points to a tension between MP autonomy (that salaries also ought to assure) and parties' attempts to control their representatives.4 Most fundamentally, parliamentary parties' ability to top up their funding by drawing on their representatives' salaries intensifies the inequality between parliamentary and extra-parliamentary parties. It thereby violates norms of equal competition similar to the informal partisan usage of parliamentary resources such as MP allowances or office staff (Gauja 2010; Birch 2011).

In the following we present two hypotheses on the acceptability of taxing practices. Having justified the case selection, described our dataset and operationalised the variables, we run a selection model which shows that leftism (shaping the intra-organisational acceptability of the practice) and informal party access to the state (shaping its external acceptability) are the only variables that significantly affect parties' likelihood of having a taxing rule and their capacity to collect high taxes, substantiating our theoretical account. We conclude with a discussion of our findings' broader repercussions and their implications for future research.

The acceptability of informal party finance strategies

Previous research has approached salary transfers from parliamentarians to their parties as a form of ‘indirect state funding’ (Biezen 2003b). The notion of ‘indirectness’ is important since it stresses the gap between the resource's origin (the state), the initial recipient of the salary (the parliamentarian) and the eventual benefactor (the party organisation). A ‘taxing regime’ or a ‘party tax’ bridges this gap through an intra-organisational rule that obliges MPs to regularly donate a specified share of their public salary to party coffers (e.g., national or regional headquarters).5 These rules can be mentioned in a party's statutes, but usually their content – especially the ‘tax share’ – is communicated in internal and less publicly visible exchanges. Compliance is not legally enforceable, which is why this practice can be conceptualised as an informal rule, communicated and enforced outside officially sanctioned channels (Grzymala-Busse & Jones Luong 2002: 534; Helmke & Levitsky 2004: 727). Parties have an interest in collecting high salary contributions from MPs, while the latter have an interest in keeping their private income to themselves, which creates a conflict of interest between the two. Given parties' increasing difficulties in attracting talented personnel, the question of under which conditions taxing practices are acceptable to MPs arises. This suggests that while parties can put pressure on their MPs, compliance is more a matter of internal acceptability in the context of a particular party than the expression of a hierarchy between a party and its MPs. The need for internal acceptability is accompanied by the need for external acceptability when further considering the nature of ‘taxing regimes’ as accommodating informal institutions. These set of rules, as defined by Helmke and Levitsky (2006: 15), operate in the context of effective formal institutions. Yet while being legal they risk altering the substantive effects of these rules. As argued earlier, a considerable lowering of MP salaries through financial transfers from MPs to their party makes these positions less attractive, both for highly qualified personnel (who earn better in other professions) and for candidates from less privileged backgrounds, which inevitably feeds into the nature of parliamentary representation.

Drawing on Duverger (1964: 198), the internal acceptability of taxing practices from the MP's perspective is predominantly a product of his or her relationship with the party organisation, which is shaped by three, closely tied mechanisms. First, the heavier the burdens are that a party imposes on prospective office-holders – be it in terms of parliamentary discipline or financial contributions – the more selective its recruitment strategy is likely to be. The further parties are on the left, the more they tend to recruit candidates favourable towards redistribution and the more accepting its MPs can be expected to be towards high salary transfers to party coffers (Duverger 1964: 198–200). Second, mass parties as compared to cadre parties, have more elaborate extra-parliamentary structures and are therefore considered to be less deferential towards parliament and its representatives by Duverger (1964: 190), echoed by existing studies showing the extra-parliamentary arm of (predominantly leftist) mass parties to be particularly strong in terms of resources (Gibson & Harmel 1998: 467–468).6 Third, this position is normatively underpinned especially in leftist parties, where MPs are considered as delegates of their party, whose ‘status as members of the party's -inner circle- takes precedence over their status as members of parliament’ (Duverger 1964: 202). The stronger this notion is held, the higher the appropriate salary share to be ‘returned’ to the party. Taking these arguments together suggests a positive relationship between leftism and taxing practice:

  • H1 (the internal acceptability hypothesis): The further left a party is positioned ideologically, the more likely its MPs will accept a taxing rule and the higher the salary share its MPs will contribute to party coffers.

While intra-party dynamics are important to understand why the obligations resulting from intra-organisational taxing regimes are acceptable to MPs, the systemic environment can also make demanding taxing practices more or less acceptable – that is, increase acceptability externally. Building on Katz and Mair's (1995b) cartel party thesis, Biezen (2004: 701) introduced the notion of ‘parties as public utilities’ describing a trend towards the intensifying usage of state resources by political parties, resources that democracies started to provide, either directly (in the form of party funding) but also indirectly (by attaching increasingly generous resources to institutional roles). The increase of MP salaries paid by the state across the range of long-lived democracies has allowed politics to become an independent profession over the last decades (Best & Cotta 2000). Parties act as gatekeepers to this profession, while using their MPs' private salaries as a resource (Katz & Mair 2009).

At the same time, however, democracies still vary considerably in the extent to which parties penetrate state institutions and access various types of resources formally and informally (e.g., Biezen & Kopecký 2007; Kopecký et al. 2012), which leads us to an earlier concept – the German Parteienstaat (Hennis 1998) and the Italian Partitocrazia (Calise 1997) – introduced to describe democratic regimes in which this penetration is particularly intense.7 In such regimes, the connections between parties and state are numerous and (leaving illegal practices aside) considered legitimate to assure the functioning of these polities. Italy, Germany, but also Austria and Belgium, have served as prime examples (e.g., Calise 1997; Hennis 1998; Peters 2006). In such settings, the usage of parliamentary salaries by political parties can be conceptualised as one of many ways through which parties access state resources to maintain themselves as the core actors in charge of the democratic process (reflecting Biezen's conception of parties as public utilities) and the collection of party taxes is likely to be a legitimate, general practice among parties. This means that in systems where party-state penetration is generally high, parties are better placed to extract regular and considerable contributions from their MPs – irrespective of parties' internal dispositions and despite the tension between taxing practices and the purpose for which salaries are, in principle, provided (Helmke & Levitsky 2006: 15). In regimes where the mutual autonomy of state institutions and parties is considered normatively desirable and has been maintained to a wider extent, party taxes are less acceptable.

  • H2 (the external acceptability hypothesis): In political systems in which parties' informal access to state resources is broad (i.e., party-state penetration is intense), parties are more likely to have taxing rules and collect higher salary shares than in systems in which this access is limited.

Data collection and taxing practices

We evaluate our approach using original data on the taxing practices adopted by political parties operating in 19 Western democracies. Data collection targeted parties in 15 democracies in Europe, plus the four Anglo-Saxon democracies New Zealand, Canada, Australia and the United States. As such, individual parties act as the units of observation for our analysis. To assure the basic homogeneity of the cases covered and thereby to assure a meaningful application of the relevant concepts and measures presented below, we restricted the analysis to parties that operate in political systems, whose democratic institutions are long-lived (established before or right after the Second World War) and that are characterised by institutionalised political parties and by party systems in which the left–right dimension constitutes a major axis of competition. While Spain or Portugal who democratised in the 1970s, for instance, by now meet the latter criteria, in-depth studies on the evolution of parties suggest that their formation and development in these new Southern democracies (as in Central and Eastern Europe) followed different patterns in terms of organisation than those we find in more long-lived democracies (e.g., Biezen 2003a; 2005); democracies that were the reference point of those classical works on party organisation upon which the theoretical frame of this article rests. Studying an intra-organisational practice, we thus opted for a relatively exclusive approach when initially specifying the set of democracies to study.8 We included all parties in these countries that held seats in the lower house of their national parliament between early 2008, when the data collection process started, and the end of 2010, thereby covering all parties that had potential access to the salaries of national parliamentarians as an income source.

The data collection drew on the following primary sources: party constitutions and finance regulations, over 50 face-to-face semi-structured interviews with party officials and party experts, an email survey targeting party head offices, complemented by news reports and the existing (in many cases fairly detailed) country literature. The interviews were essential since as insightful as formal documents are, we found that a party tax can exist informally as an expectation that is communicated to office-holders once they enter office and only specified in internal documentation of the party executive or parliamentary party, which might not be accessible to outsiders. It may also be communicated to candidates in a pledge that they sign after nomination and is kept by the leadership (potentially to be used as a ‘reminder’). The triangulation of different data sources helped to assure the accuracy of data and to avoid interviewee bias (in particular with regard to those parties that did not document in their constitution that parliamentary salaries indeed were one of their sources of finance) and thereby allowed us to put together the most reliable and detailed picture of taxing practices. Of the 141 possible parties that could have been included in the analysis, we were able to collect data on the taxing practices of 124 parties.9

We identified whether a party has established an organisational rule that obliges national MPs to regularly donate a specific share of their salary or not. Almost without exception, whenever a tax has been established on some office-holders' salaries or allowances, this rule includes national MPs. Accordingly, the focus on national MPs as ‘tax payers’ gives us a reasonably precise picture of which parties have a tax (coded 1) and which do not (coded 0), and forms the basis for our first dependent variable Presence of Tax.

The analytical focus on the ‘tax payer’ was critical since tax payer and collector are not necessarily located at the same governmental level.10 Thus, we could not simply look at what the national executive, as tax collector, took from national MPs. In particular, parties operating in federal systems have powerful regional party executives, so taking only the national headquarters into account would not lead to an appropriate estimate of the tax burden imposed on national MPs. To assure comparability, our second dependent variable Tax Share refers to the combined percentages of the average base salary11 paid by national parliamentarians as collected by the national and the regional organisational units in a party: in unitary systems the only tax collector is usually the national party,12 whereas in federal systems contributions from national MPs often go to the regional level or to both regional and national level.13 Our operationalisation thereby avoids an underestimation of a party's capacity to draw on national MPs' salaries.14 Tax share can range from no required contributions (coded 0 per cent) to contributions constituting MPs' full base salary (coded 100 per cent).

Table 1 summarises the taxing practices of parties by the country in which they operate. It highlights the proportion of parties that have established an internal taxing practice vis-À-vis their MPs as well as the mean extent of these taxing rules. Of the 124 parties covered by our analysis – that is, parties for which we were able to gather information about their taxing practices – as many as 79 (63.7 per cent) expect their national MPs to provide regular fixed donations to party coffers. In many cases, parties also expect highly significant contributions. While 40 parties (out of 79) take a share below 10 per cent of their MPs' base salary, 31 parties request a share between 10–30 per cent, and the remaining eight take more than 30 per cent. The average tax that these parties demand is as high as 13.3 per cent of their MPs' base salary. With the average purchasing power adjusted salary for the elected representatives of those 79 parties being around €5,800 per month, the 13.3 per cent requirement constitutes monthly payments of over €770.15 Consequently, through taxing rules, parties are able to informally redirect significant amounts of money away from MPs' private accounts to party coffers, stressing the suitability of an approach focusing on sources of their acceptability, internally and externally.

Table 1. Taxing practices on the salaries of national MPs in advanced democracies
 Number of partiesNumber of parties in analysisNumber of parties in analysis with taxing rule (%)Mean percentage of tax share of parties in analysis with taxing rule
Australia442 (50.0)7.8
Austria555 (100.0)12.8
Belgium141211 (91.7)11.9
Canada440 (0.0)N/A
Denmark883 (37.5)14.6
Finland883 (37.5)2.3
France866 (100.0)22.9
Germany666 (100.0)10.7
Iceland542 (50.0)2.7
Ireland664 (66.7)19.0
Italy111010 (100.0)22.9
Luxemburg644 (100.0)15.2
Netherlands1096 (66.7)16.3
New Zealand762 (33.3)7.0
Norway772 (28.6)5.0
Sweden873 (42.9)7.3
Switzerland1177 (100.0)8.8
United Kingdom1193 (33.3)3.3
United States220 (0.0)N/A
Total14112479 (63.7)13.3

Operationalising explanatory and control variables

Testing the internal acceptability hypothesis (H1) requires us to specify a party's ideological orientation; we use party positions on the left–right scale as specified by Benoit and Laver (2006).16 Since we use this variable (Ideological Position) as a proxy for the general orientation of MPs, the left–right position averaging party ideology across the full range of issues covered in the study is the most appropriate choice of measure.17 Expert survey data has proved to be reliable in estimating party positions and is consequently widely used in comparative studies (Lowe et al. 2011). The values assigned to parties can range from 0 to 20; the higher the value the further the party is located on the right. The mean Ideological Position of the 124 parties in our sample is 10.4.18

To test the external acceptability hypothesis (H2), we need a convincing proxy for parties' leeway to access state resources informally, which is separate from the practice we examine. A good measure for informal party access, concerning an arena clearly distinct from the parliamentary one, is provided by a series of recent studies that provide rankings of patronage across a wider range of democracies (e.g., Kopecký & Scherlis 2008; Kopecký et al. 2012). Party patronage is not necessarily illegal (as corruption is), but its legitimacy is often questioned since it clashes with the idea that non-elected positions should be distributed on the basis of merit only. If patronage practices are nonetheless widespread, this implies considerable leeway for parties in a system to access state resources, also informally. Accordingly, the relative access of parties to non-elected positions in public and semi-public life (e.g., in the civil service, governing boards, public sector companies) – that is, the scope of party appointment power in the state apparatus (Kopecký & Scherlis 2008: 356) – serves as our proxy of informal party control. We distinguished countries in which party access to state apparatus is extensive (coded 1) from those in which it is limited (coded 0).19 Seven (36.8 per cent) countries qualified as democracies with high levels of party-state access accordingly.20

To control for rivalling explanations we introduce five additional variables in our models that follow theoretical rationales distinct from our approach on the acceptability of taxing regimes as accommodative informal rules: three affect parties' institutional resource access, two affect parties' capacity to ensure compliance with taxing rules. Starting with the former, the likelihood that a party taxes national MPs' salaries is linked to the party being represented in the national parliament in the first place. The longer a party is nationally represented, the more likely it is that a taxing rule exists. We measure the Duration of National Representation in number of years since the party won its first MP, prior to 2011 and divided by ten.21 Moving on, the level of parliamentary pay can be expected to increase the likelihood of taxing national MPs and the level of the tax on MPs positively. An MP who earns well can be more easily asked for a regular contribution and for significant contributions than an MP whose pay is comparatively meagre. To measure this variable, we used the average pay of rank-and-file MPs, divided by 1,000, in the respective currency of a country and adjusted these figures by taking into account the differences in the purchasing power between countries, hence yielding purchasing power parity adjusted salaries in euros (Parliamentary Pay PPP). Finally, the extent to which parties have access to formal state funding should affect the extent to which they bother to collect party taxes, a demanding intra-organisational task. Especially, the debate on party funding reform tends to assume that the more generous formal resource access parties have, the less pressure there is for them to exploit resources in informal ways (e.g., Nassmacher 2009; Koss 2011). To measure Formal State Funding we combine two indicators developed by Bischoff (2006) in a cross-national study on new party entry in 21 democracies, which categorise democracies along the same logic on the basis of two four-point scales capturing the formal regulation of resource access. The first indicator measures how easily parties can access direct party funding in a system, ranging from a barrier of less than 1 per cent of the vote is required to receive such funding (1) to no funding provision at all (4). The second measures how far parties can profit from free television time, ranging from provision of free television time to all parties (1) to no free television time at all (4).22 Combined in one index, ranging from 2 to 8, they provide information on the relative availability of direct and indirect formal state funding. The higher the score a country receives – that is, the scarcer Formal State Funding – the stronger the incentives are for parties that operate in those more restrictive environments to establish a taxing practice and to collect high taxes from MPs.

Moving to the two control variables that can be expected to affect parties' capacity to assure compliance with taxing rules, we further include a variable capturing the size of the parliamentary group. The collection of MPs' contributions constitutes a demanding task and resistance is likely to grow with the level of tax. Office-holders need to be chased up by party personnel and reminded of their duty. The costs to implement such a rule increase with the number of office-holders who are taxed. Parliamentary Group Size is measured as a party's most recent absolute number of seats in the lower house of parliament between early 2008 and late 2010 (the period of data collection).23 Finally, we control for the incentives the electoral system generates for pursuing personal vote strategies distinct from party reputation and support. If parties can widely determine a candidate's fate at elections, they should find it easier to impose organisational rules such as party taxes. To capture Personal Vote Incentives, we classified electoral systems along the degree of control party leaders can exercise over access to their party's label by controlling party endorsements and ballot rank. This was done in line with the classification of electoral systems by Carey and Shugart (1995). We distinguish between systems in which party leaders present a fixed ballot that voters cannot disturb (0), systems in which party leaders present a ballot which voters can disturb (1) and systems in which party leaders do not control access to ballots or rank (2).

Model choice

The two-step nature of the taxing practice as adopted by a party is best represented by a selection model where the decision to tax represents the selection phase and the level of taxation represents the outcome stage. On the one hand, modelling the level of taxation alone would bias the results due to a censored sample as the latter would not include parties that did not tax. Using the selection model, we avoid this problem as all parties, whether they have established an internal taxing practice vis-À-vis their MPs or not, are included. On the other hand, if we modelled the level of taxation and gave a value of 0 to those parties not taxing their MPs, we would treat the increase from 0 to a higher unit in the same way as a transition from one level of taxation to the next level of taxation. The decision to tax (moving from 0 per cent to a higher level), however, is qualitatively different from moving from one level of tax to another even if these transitions are characterised by equal percentage changes.

Heckman selection models are available means of estimating models with data where there is an interval level value for the dependent variable (Y) when the value for another variable (Z) is 1, and have been widely used in research projects where a two-step approach has been necessary (e.g., Achen 1986; Timpone 1998). Treating taxation as a two-step approach, Heckman selection models allow us to specify the factors underlying the presence of a taxing regime and driving the level of tax, respectively. Moreover, Heckman selection models allow us to account for the hierarchical nature of the data structure. Using a multivariate setting to account for taxing practices as adopted by parties with some of the explanatory variables being country-level rather than party-level variables, it is essential to account for the contextual variation embedded within the countries in which the parties operate. A Heckman selection model that treats countries as cluster variables does exactly that, addressing the hierarchical structure of data by adjusting the standard errors accordingly.24

Findings and discussion

We have already shown that the establishment of an institutionalised taxing regime is indeed a very widespread practice across parties in advanced democracies, and that this taxation is not of a symbolic nature but represents rather high proportions of MPs' salaries. We now turn our focus on the variation in parties' ability to both establish a taxing regime vis-À-vis their MPs and to collect high levels of rent from them.

As shown by Table 2, the findings confirm both our internal acceptability hypothesis (H1) – the further left a party is positioned ideologically, the more likely its MPs accept a taxing rule and the higher the salary share its MPs contribute to party coffers; measured through Ideological Position – as well as our external acceptability hypothesis (H2) – in political systems where parties' informal access to state resources is broad, parties are more likely to have a taxing rule and collect higher salary shares than in systems in which this access is limited; measured through Party–State Access. With respect to the former, negative and statistically significant coefficients of −0.24 and −1.23 indicate that right-wing parties are less able to obtain MPs' cooperation for introducing taxing regimes and collecting high levels of tax than their left-wing counterparts. Operating in systems that allow parties greater access to informal state resources, however, increases all parties' leverage vis-À-vis their MPs regarding party taxing as shown by positive and statistically significant coefficients of 3.40 and 5.11. These findings lend further support to conceptualising party taxes as informal practices, which are expressions of interaction between MPs and their party rather than forms of one-sided dependency of either actor.

Table 2. Explaining the presence of tax and tax share
 Heckman
Presence of tax
  1. Notes:-Standard errors in parentheses. ** p < 0.01; * p < 0.05.
Parliamentary group size0.00 (0.00)
Ideological position−0.24** (0.05)
Parliamentary pay PPP (*€1,000)−0.09 (0.15)
Party–state access3.40** (0.62)
Personal vote incentives0.30 (0.24)
Formal state funding−0.10 (0.13)
Representation (*10 years)−0.06 (0.04)
Constant3.23** (1.17)
 Tax share
Parliamentary group size−0.05** (0.01)
Ideological position−1.23** (0.38)
Parliamentary pay PPP (*€1,000)2.21** (0.50)
Party–state access5.11* (2.42)
Personal vote incentives0.93 (2.57)
Formal state funding−0.44 (0.54)
Constant12.69 (6.89)
Number of observations124
Censored observations45
Uncensored observations79
Number of countries19
Log likelihood−346

Two of our control variables also have significant effects in the expected direction, increasing the confidence in our findings overall. One affects parties' capacity to ensure compliance with taxing rules (Parliamentary Group Size), and one parties' institutional resource access (Parliamentary Pay PPP). Parties' capacity to assure compliance with taxing rules declines as parliamentary groups become bigger, with chasing up more office-holders to remind them of their duty increasing the implementation costs of such rules. As a result, it is not surprising that parties with more MPs do in fact reduce their representatives' resistance to the taxing rules by collecting lower tax shares (as shown by the negative coefficient of −0.05). Higher MP salaries, however, enable parties to collect higher taxes (as shown by the positive coefficient of 2.21) as their MPs benefit more from the public office they obtained under their party's label. Consequently, the theoretical rationales as captured by our control variables (capacity to assure compliance; institutional resource access) are also important to account for taxing practices next to factors driving their acceptability. At the same time, however, they only affect the level of tax. In contrast, the variables driving the acceptability of taxing practices – Ideological Position and Party–State Access – are significant in both stages and the only ones that affect the more fundamental stage (the likelihood that a party taxes MP salaries).

To illustrate the real-world meaning of these effects, we have produced predicted values for both the taxing regime and the level of taxation (Table 3). Taking a closer look at the likelihood of a party to have established an institutionalised taxing rule vis-À-vis its MPs, Ideological Position (H1) and Party–State Access (H2) stand out. As Table 3 demonstrates, the probability of adopting a taxing regime declines from 96 to 36 per cent as parties move from left to right on the ideological spectrum. This suggests that left-wing MPs are indeed more accepting of such a redistribution of their salary. At the same time, however, all parties – irrespective of their internal dispositions – are better equipped to establish a taxing practice when operating in countries where connections between parties and state are more extensive and considered legitimate. Whereas almost all parties (i.e., 98 per cent) that operate in democracies in which party–state access is intense are predicted to have an institutionalised taxing practice, only a minority of parties (i.e., 40 per cent) operating in more restrictive contexts are predicted to do so. The latter parties are in weaker positions in their interaction with their MPs as the mutual autonomy of state institutions and parties is appreciated more in such countries.

Table 3. Predicted values for presence of tax and tax share
Explanatory variablePresence of taxTax share (%)
  1. Notes:-Presence of tax is the average predicted value for presence of taxing regime. Tax share is the average predicted value for level of tax.
Full sample0.638.5
Ideological position  
Leftist (<5)0.9620.4
Centre-left (5–10)0.7012.1
Centre-right (10–15)0.544.9
Rightist (>15)0.362.0
Party–state access  
Limited0.404.2
Extensive0.9815.0
Parliamentary pay PPP  
Lowest quartile (<€4,000) 4.6
€4,000–€6,000 6.3
€6,000-€8,000 10.3
Highest quartile (>€8,000) 18.4
Parliamentary group size  
Lowest quartile (<8) 9.0
8–18 11.4
18–43 8.4
Highest quartile (>43) 5.3

Given the centrality of both the internal and external acceptability of taxing practices, we further looked at how many parties are predicted to have institutionalised taxing rules when different combinations of internal and external favourability for such rules are present. In fact, supportive internal and external conditions versus restrictive internal and external conditions cover almost the full range of the possible likelihood of taxation. Whereas all left-wing parties in systems with broad party–state access are predicted to collect regular fixed donations from their MPs, only 7 per cent of right-wing parties in countries where parties' informal access to state resources is limited are predicted to do so. These findings lend further support to conceptualising party taxes as informal practices, with their presence shaped by the mutual acceptability of such a practice by the specific parties and their MPs.

Moreover, we find that operating in countries where parties' access to the state is high (i.e., taxing practice being externally acceptable) tends to be necessary for right-wing parties to establish a taxing rule, and leftist ideology (i.e., taxing practice being internally acceptable) tends to be sufficient for a party to establish a taxing rule irrespective of whether it operates in a system where the penetration of state institutions by political parties is intense or weak. While as many as 90 per cent of right-wing parties in democracies with broad party–state access are predicted to require fixed regular donations from their MPs, as few as 7 per cent of right-wing parties operating in more restrictive environments are. At the same time, the proportions of left-wing parties that are predicted to be able to establish an institutionalised taxing practice are very similar in the two contexts. Whereas all left-wing parties are expected to tax their MPs in democracies in which party–state access is broad, the vast majority of them (92 per cent) are predicted to do so also when the favourable external context is not present. Unlike their right-wing counterparts, leftist parties find the internal acceptability of taxing their MPs strong enough to establish a taxing practice in less favourable and, thus, more restrictive systemic contexts.

The second stage, tax share, is again affected by Ideological Position and Party–State Access, but also by Parliamentary Group Size and Parliamentary Pay PPP. These two latter characteristics highlight that parties' capacity to ensure compliance and their access to institutional resources, respectively, add further depth to our understanding why some parties are able to obtain MPs' consent for higher levels of taxation.

In line with our approach, it is the leftist parties (H1) and those operating in countries where their informal access to state resources is extensive (H2) that are predicted to collect higher levels of tax. As parties move from left to right, the percentage of taxation is predicted to decrease by 18.4 percentage points (from 20.4 to 2.0 per cent), whereas the predicted tax share of only 4.2 per cent for parties operating in countries where the penetration of state institutions by political parties is weak rises to as high as 15.0 per cent for parties that operate in more favourable systemic contexts. Both the internal and external acceptability of party taxing, measured through Ideological Position and Party–State Access, respectively, noticeably shape how much rent parties can extract from their MPs.

Moving on to the control variables, it is parties with fewer MPs who are able to collect higher proportions of their MPs' salaries (capacity to ensure compliance) as well as parties that operate in states where parliamentary salaries are higher (institutional resource access) that are predicted to benefit from higher levels of taxation. The effect of moving from the lowest to highest size of the parliamentary group produces a 3.7 per cent reduction in the predicted level of taxation (from 9 to 5.3 per cent), while the predicted tax share rises from 4.6 per cent for parties that operate in countries with lowest parliamentary salaries to 18.4 per cent for those whose MPs benefit from the highest salaries. These effects on Tax Share, albeit significant, are, however, of more limited scope than those brought about by changes in characteristics that shape the acceptability of taxing. This further supports our conceptual interpretation of party taxing as an informal process that is predominantly shaped by the mutual acceptability of it as seen by both national parties and their MPs.

Conclusions

Linking Helmke and Levitsky's (2006: 15) conceptualisation of ‘accommodating informal institutions’ developed in research on new democracies with theoretical approaches on (formal) party organisation in long-lived democracies (Duverger 1964; Katz & Mair 1995b; 2009; Biezen 2004), this article has presented a framework accounting for the diversity of ‘taxing regimes’ across parties; focusing on variables driving the acceptability of the taxing practices rather than conceptualising the practice as a form of one-sided dependency between party and MP. The centrality of acceptability originates in the practice's informal nature: in the lack of formal enforcement mechanisms on the side of the party and in the discrepancy between the practice and the purpose for which parliamentary salaries are provided (e.g., attracting high quality representatives and representatives from diverse backgrounds). The former aspect highlights the importance of internal, the latter external acceptability. Building on the notion of informality, in terms of internal acceptability, we argued that leftist party ideology facilitates the establishment of a taxing rule and the collection of high salary shares. In terms of external acceptability, we expected that taxing practices will be more acceptable in political systems in which parties' informal access to state resources is generally broad than in settings in which the separation between parties and state apparatus is more pronounced, irrespective of parties' ideological dispositions.

Both variables, contributing to the acceptability of the practice, indeed affected the presence and level of tax in the expected manner, despite controlling for rivalling influences that followed distinct theoretical rationales – namely variables that shape parties' institutional resource access or their relative ability to assure MPs' compliance. While the level of parliamentary pay had a significant positive impact and the size of the parliamentary group had a significant negative impact on tax share, they did not have an impact on the likelihood of a taxing rule being established in the first place. Direct party funding, the personal vote incentive generated by the electoral system, or the period a party held national seats, had no significant effects. In essence, while not all rivalling factors were insignificant, the theoretical perspective we proposed – linking party-level and systemic perspectives – appeared as more suitable to account for the nature of this, so far understudied, party finance strategy. Our findings, thereby, highlight the fruitfulness of borrowing conceptual tools from research on new democracies to explore so far understudied areas of advanced democracies (Grzymala-Busse & Jones Luong 2002; Helmke & Levitsky 2004, 2006) in a research area where ‘conceptual travel’ more often goes the other way round (Svåsand 2013).

Our findings further contribute to the prominent party finance literature (Scarrow 2007) by accounting for a finance strategy that – leaving qualitative work aside – has been widely neglected in comparative studies. It highlights the need to go beyond the study of formal channels through which parties access state resources and to also explore the practices through which parties access resources informally. Practices of accessing funding indirectly (Biezen 2003b; Biezen & Kopecký 2007) can be, but are not necessarily, illegal. They can exist in the context of formal regulations thanks to incomplete or ambiguous rules without violating the latter, and thus highlight the difficulty of establishing regulation that is able to prevent the resources that are attached to institutional roles from being used for party-political purposes (Birch 2011: 32). Future research faces the challenge of providing a fuller picture of how parties exploit these grey zones that democratic settings inevitably provide. Such efforts ought to include how the informal strategies that parties employ to access resources have changed over time, addressing prominent debates on party change and, more specifically, party organisational decline.

Acknowledgements

This research was supported by the British Academy (SG-04570) and the Economic and Social Research Council (RES-239-25-0032). We also thank Jonathan Barry, Caitlin Milazzo and Filippo Tronconi for their input, as well as the three anonymous referees and the editors of EJPR for very detailed and thoughtful comments. Any remaining errors are the authors' responsibility.

Appendix: Appendix: Country-level explanatory variables

Coding of the country-level explanatory variables

 Formal state fundingParliamentary pay PPP (€)Party–state accessPersonal vote incentives
Australia65,170No1
Austria57,090Yes0
Belgium54,620Yes1
Canada58,190No0
Denmark24,330No1
Finland54,430No2
France55,780Yes0
Germany36,840Yes0
Iceland82,690No1
Ireland66,450No1
Italy410,510Yes0
Luxemburg65,000Yes1
Netherlands36,580No1
New Zealand75,340No0
Norway54,580No0
Sweden64,560No1
Switzerland52,150Yes1
United Kingdom75,920No0
United States811,020No2

Notes

  1. 1

    In 2008, the Christian Democrats, Social Democrats, Liberals and Left Party earned 12.2, 13.2, 7.3 and 9.5 per cent, respectively, from party taxes. See, for detailed figures, Bundestagsdrucksache 17/3610 (Deutscher Bundestag 2010).

  2. 2

    Data on MP salaries has been obtained from national parliaments' public salary reports or through information requests from the Public Relations Offices of national parliaments when public reports were not available. In order to calculate purchasing power adjusted salaries, the purchasing power parities (PPP) for private consumption were used (OECD 2013).

  3. 3

    The only other large-N study of taxing practices published to date focuses on the payments from the Members of the European Parliament (MEPs) (Bolleyer et al. 2013). It covers fewer parties and only those that operate within the European Union. More importantly, in methodological terms, an extensive body of empirical literature on the European elections has shown those to be ‘second-order’ (e.g., with opposition parties as well as the smaller and niche parties tending to benefit from the different electoral context at the expense of the governmental and mainstream parties). This has distinct implications for the political make-up of the European Parliament, different from those that ‘first-order’ elections have on national representation (e.g., Marsh 2005). In theoretical terms, the paper on MEP taxing presented a party-centred approach, reflecting its focus on MEPs who as parliamentarians operate outside their national political systems. The role of the national context was neither theorised, nor central to the findings. We remedy this by integrating party-centred perspectives with systemic perspectives (both of which co-exist in the literature), which is why a focus on national taxing practices in conjunction with a complete coverage of parliamentary parties (as far as feasible) across the democracies studied was crucial for this study.

  4. 4

    For detailed information on the various motives for introducing parliamentary pay, see case studies covering the historical development of a political class (Borchert & Zeiss 2003) and of the increasing professionalisation of parliamentarians (Best & Cotta 2000) in advanced democracies.

  5. 5

    Note that the criteria of regular and specified exclude any one-off payments driven by individual preferences or will that may be decoupled from the nature of intra-party dynamics (e.g., unregulated donations of MPs to their constituency organisations, which are widespread, even in parties without any established taxing practice as conceptualised here).

  6. 6

    This line of argument does not suggest that all mass parties are leftist, as the example of Christian democratic parties illustrates. But it suggests that the further on the left a party is located in terms of ideology, the more likely its structure resembles the mass party model rather than the cadre model.

  7. 7

    Translations into English such as ‘party state’ (which is the term that seems to be used most often in the English party literature and denotes a particular type of democratic system, see Puhle 2002) are reminiscent of (and sometimes considered as equivalent to) the concept of the ‘one party state’ (see, e.g., Lewis 2006). Therefore, when referring to the original concept we use the German and Italian originals. The variable in our analysis is labelled ‘party-state access’ to capture the core aspect of these types of democratic regimes relevant to our argument.

  8. 8

    While most of the countries covered are ‘Northern European’, we do not think this biases our findings, considering the wide geographical scope (that includes all long-lived Westminster democracies), and that France and Italy show the theoretically expected taxing patterns on the systemic and party levels.

  9. 9

    The Duncan Index of Dissimilarity (Duncan & Duncan 1955) was used to compare the distribution of parties within the full population of 141 applicable parties and our sample of 124 parties across the 19 democracies. With the index value remaining as low as 0.07, we can be confident that our operational sample provides a very representative account of parties' taxing practices. See the online appendix for the list of missing parties.

  10. 10

    Examples are the Social Democrats in Germany and Austria.

  11. 11

    Sometimes MP salaries increase with seniority. In these cases, the average salary is taken as a reference.

  12. 12

    This is naturally not the case with regard to regional parties whose highest level is the regional level, such as in the Scottish National Party. The main point here is that in unitary systems the highest organisational layer in a party is in charge. In Belgium, with its split party system, MPs only pay to either Flemish or Walloon parties.

  13. 13

    Austria, Australia, Germany and Switzerland are the main examples. If regional parties take different shares from ‘their’ MPs, we took an average.

  14. 14

    In a few parties, MPs make additional payments to their constituency organisation. However, these vary from locality to locality and do not necessarily qualify as a tax (i.e., they might not exist in certain areas or occur ad hoc depending on the goodwill of the actual MP). Furthermore, they are miniscule compared to MPs' salary transfers to other levels and have therefore been left out.

  15. 15

    For example, the average purchasing power adjusted monthly contributions that the French and Austrian parties (all of whom have established the taxing rule) ask from their MPs are €1,320 and €900, respectively.

  16. 16

    Benoit and Laver's (2006) data was not available for parties that emerged after 2006. When these parties were formed as an ideologically similar splinter party or a merger of existing parties, we applied the Benoit and Laver (2006) measurement, respectively, by relying on the ideological position of its mother party or the average ideological position of the merging parties. To ensure that no bias is introduced, we ran parallel Heckman models with and without parties that emerged after 2006. Findings were not significantly affected.

  17. 17

    In supplementary analyses, we consider whether party family belonging relates to parties' taxing practices. These analyses are available as part of the online appendix.

  18. 18

    See the online appendix for further descriptive information on the party-level explanatory variables.

  19. 19

    Country scores are obtained from the recent studies by Kopecký and Scherlis (2008) and Kopecký et al. (2012). Luxembourg, Switzerland, Finland, New Zealand and Australia were not covered by the two studies. They were classified along the definition of patronage by Kopecký et al. (2012), on the basis of existing in-depth studies that cover the practices in these five countries (e.g., Tsatsos 1992; M¨ller 2000; Detterbeck 2002; Peters 2004).

  20. 20

    See the Appendix for the detailed coding of the country-level explanatory variables.

  21. 21

    Information on when parties won their first MP was obtained from the websites of national parliaments.

  22. 22

    Country scores are obtained from Bischoff (2006). Switzerland, Iceland and Luxembourg are classified along the same lines based on www.idea.int/parties/finance/db/country_view.cfm and www.sgi-network.org/ index.php?page=indicator_quali&indicator=S1_2. For details on the two initial indicators, see Bischoff (2006).

  23. 23

    Naturally the size of parliamentary representation sometimes varies from election to election, yet the changes are usually not of such a kind to change the prevalent dynamics in a parliamentary group. The scores for each party were obtained from the websites of national parliaments.

  24. 24

    Although there are available alternatives for analysing variation within hierarchical data structures, multilevel regression models and general linear models would be unnecessarily complex procedures and they struggle to incorporate the two-step approach fundamental to modelling party taxes. In addition, their main advantage of providing cross-level interactions is not focused on in the context of this article.

Ancillary