Whose bread I don’t eat,his song I don’t sing? MPs’ outside earnings and dissenting voting behaviour

Abstract

The question whether politicians’ outside earnings affect their legislative behaviour is part of a lively debate about the quality of representative democracy. However, moonlighting effects on vote defections by members of parliament (MPs) have remained underexposed yet. Based on Competing Principals Theory, it is argued that, owing to a higher degree of career-related independence, MPs with high outside earnings can be less effectively disciplined by their party and, therefore, show higher probabilities to vote against the party line. This proposition is tested quantitatively using logistic panel regressions against a new dataset of more than 115,000 individual votes in the German Bundestag (2013–2017). Empirically, the results corroborate the theoretical expectations and are robust against different specifications. The findings have important implications for our understanding of the link between politicians’ career paths and their political behaviour.

Keywords

outside earnings parliament party discipline party unity roll-call votes

Introduction

In March 2019, Florian Post, a member of parliament (MP) of the Social Democrats (SPD), lost his seat in the prestigious Bundestag committee for economic affairs and energy. Even though his party had to withdraw one MP for arithmetic reasons, its chief whip revealed to a news magazine that for committee assignments, the party group leadership takes, amongst others, an MP’s voting behaviour into consideration (Spiegel Online, 2019). Post was regarded as an opponent of the SPD party group leader at that time and repeatedly voted against the party line (Süddeutsche Zeitung, 2019).

Opposing one’s own party guarantees media attention and can foster an MP’s popularity in the electorate (Campbell et al., 2019; Rowlands and Vander Wielen, 2019; Wagner et al., 2020). However, although visible sanctions are comparatively rare (Bailer, 2018), the anecdote above supports the conventional wisdom that vote defections can harm an MP’s career prospects. If this is true, then how an MP will act in parliament might, conversely, be influenced by the career ambitions and prospects he or she has. However, it seems reasonable to assume that the importance of office-seeking- vis-à-vis other goals such as vote- or policy-seeking (Strøm, 1997) is not equally weighted by every MP. An MP that already holds an important office in parliament or government, for example, has more to lose and a disproportionately higher incentive to toe the party line in order to secure his/her career-related achievements than an MP that doesn’t expect to be promoted to offices (Benedetto and Hix, 2007; Delius et al., 2013).

But what about MPs that pursue a successful career with high earnings besides their parliamentary mandate, e.g. by being a lawyer or running an own agricultural business? Perceived as leading to ‘greed, shirking and conflicts of interest’ (Geys, 2013: 470), outside earnings are mostly a cause of lurid headlines – although citizens do not respond negatively to all moonlighting activities even-handedly (Campbell and Cowley, 2015). Whereas existing research has repeatedly analysed whether outside earnings influence parliamentary effort (Geys and Mause, 2013), moonlighting effects on MPs’ vote defections have not been systematically assessed yet – although such effects could have considerable consequences for voters’ perceptions of parties, policy outputs and the stability of governments. In this study, I will fill this research gap as follows: After a literature review, it is argued that MPs with high outside earnings have a higher propensity to vote against the party line since, due to a higher level of career-related independence, they can be less effectively disciplined by their party. Preceded by a discussion of the study design, the results of logistic panel regressions reveal that, as hypothesized, high outside earnings are significantly related with more votes against an MP’s own party. The conclusion discusses the implications of the findings.

State of research: Party unity and moonlighting

A vast literature on party unity at different levels of observation has already led to a better understanding of the phenomenon. Studies on the party group level usually assess the impact of characteristics of the party group and/or the motion (e.g. Bergmann et al., 2016; Ceron, 2015; Close, 2018; Close et al., 2019; Kam, 2011; Rehmert, 2020; Saalfeld, 1995; Stecker, 2015) – supplemented by institutional variables in cross-country studies (e.g. Carey, 2007; Coman, 2015; Shomer, 2016, 2017; Sieberer, 2006). Conversely, individual-level analyses focus primarily on MP characteristics to explain their varying propensity to defect. Significant predictors of open dissent are especially an MP’s mandate and candidacy type, seniority, legislative and executive offices, electoral vulnerability, occupational background, programmatic self-positioning, local politics engagement and gender (André et al., 2015; Benedetto and Hix, 2007; Bhattacharya and Papageorgiou, 2019; Cowley and Childs, 2003; Degner and Leuffen, 2016; Gherghina and Chiru, 2014; Grimmer and Powell, 2013; Heuwieser, 2018; Ohmura, 2014; Sieberer, 2010; Sieberer and Ohmura, 2019; Slapin et al., 2018; Tavits, 2009, 2011; Willumsen and Öhberg, 2017; Zittel and Nyhuis, 2019). Many of those studies explicitly or implicitly show that MPs’ career-related goals in the broader sense influence legislative behaviour. However, an empirical study that examines the role of outside earnings for an MP’s probability to defect is still lacking.

Nevertheless, MP moonlighting has already been a subject of scientific interest. To start with, many studies investigate which MP-related factors are associated with higher or lower outside earnings (see Supplement material 1). Their results show that earnings seem to be high when MPs do not have much to lose and low when their political career is at stake. Hence, there might be a trade-off between an MP’s career ambitions inside and outside parliament.

Concerning behavioural consequences of moonlighting, it is frequently found that more outside jobs or higher earnings decrease parliamentary effort (Arnold et al., 2014, Eggers and Hainmüller, 2009; Fedele and Naticchioni, 2015; Gagliarducci et al., 2010; Hurka et al., 2018b, Staat and Kuehnhanss, 2017). Moreover, Geys and Mause (2016) show that British MPs who leave parliament by choice in order to pursue an extra-parliamentary career shift their work balance from parliamentary to their outside activities already in their last term. According to Mickler (2018), MPs’ prior occupation and their outside activities affect Bundestag committee assignments. Taken together, there is some evidence that MPs’ extra-parliamentary activities influence their parliamentary work. However, only a few studies relate MPs’ outside earnings to their voting behaviour (Geys and Mause, 2013): Given that moonlighting MPs have a lower propensity to vote for a stricter regulation of outside activities in the U.S. (Rosenson, 2007) but not in the UK (Johnston et al., 1997), the role of financial self-interest for MPs’ voting behaviour is ambiguous. Couch et al. (1992) show that public higher education expenditure is higher in those U.S. colleges having legislators on their payroll. However, we still miss evidence regarding the role of moonlighting in situations when an MP disagrees with his/her party group and has to decide whether to toe the party line or not.

The argument: Impact of outside earnings on dissenting voting behaviour

Competing Principals Theory (Carey, 2007) provides an explanation of why MPs differ in their propensity to vote against the party line. According to this specification of the principal-agent framework, MPs gear their legislative behaviour towards principals that control resources they aim to obtain. The main goals of MPs include re-selection, re-election, promotion to influential and/or well-paid positions and the implementation of their desired policy objectives (Strøm, 1997). Based on that, large parts of the literature discuss whether MPs aim more at their party or their voters in order to reach those goals, respectively. Subsequently, we will discuss how outside earnings could change the relationship between MPs and those two principals and whether they create an additional principal influencing their legislative behaviour.

In Germany’s two-tier electoral system, district MPs have to be re-selected by local party branches and then re-elected by constituency voters. Accordingly, district MPs are expected to pay particular attention to constituency interests. If those contradict the interests of the parliamentary party group, district MPs could be more inclined than list MPs to vote against the party line (e.g. Sieberer, 2010). Contrarily, list MPs depend on regional party branches to obtain promising list positions and their party’s regional election result to get re-elected. Consequently, as they depend solely on their party to keep their mandate, list MPs are expected to vote in accordance with the party line more often than district MPs. However, by and large, list MPs do not get punished after dissenting votes by being detained from promising list positions (Baumann et al., 2017; Kauder et al., 2017). Additionally, the candidacy mode and re-election prospects influence the electoral pressures lasting upon an MP as well (e.g. Ohmura, 2014, Sieberer and Ohmura, 2019).

In order to reach their office-seeking aspirations, MPs depend solely on their party. The national party leadership and its parliamentary party group decide on the allocation of executive and parliamentary leadership positions, respectively. In order to be promoted to prestigious offices, MPs have to align closely to the party’s interests, which also includes toeing the party line in parliament. Previous literature shows that party (group) leaders use office assignments to reward allegiant MPs or to punish disloyal ones (e.g. Cox and McCubbins, 1993). Consequently, especially for list, but also for district MPs, parties are important principals. Accordingly, party discipline, i.e. leadership-induced unity, is a crucial pathway to the empirically high rates of voting unity in parliamentary democracies (Carey, 2007; Sieberer, 2006).

However, the incentive structure for toeing the party line is not the same for all MPs since the composition and influence of their principals differ. Moving beyond the debate whether the constituency’s or the party’s interests matter more for particular MPs, we argue that MPs with high outside earnings are less dependent from both their voter and party principals.

First, outside earnings are expected to relieve MPs from electoral pressures. If they fail to reach re-nomination or re-election in the constituency and/or the party list, they will have something to fall back on. In a survey, two German MPs answered the following regarding the arrangements they made for a possible loss of their mandate (Kreiner 2007: 265):

To say: ‘The company is still running.’ That was the most meaningful [precaution].

One should not adhere to a politician’s chair, and one does not know how things will come. [My own company] was simply a means of precaution […]. (own translations)

Hence, defections as a means of signalling dedication to constituency preferences or interests are less likely for moonlighting MPs, due to lower electoral pressures.

Second, however, it is unlikely that this will bind moonlighting MPs more closely to their party principal. In order to reach prestigious offices, progressively ambitious MPs usually gear their legislative behaviour towards the party leadership that helps them reaching their office-related goals (Meserve et al., 2009; Schlesinger, 1966; Sieberer and Müller, 2017). Of course, most MPs would not, according to surveys, refuse an offer to be promoted to higher office (Würfel, 2018) – regardless of outside earnings. Thus, when facing the choice between their party’s position and their own preferences, many of them have an incentive to act in accordance with their party. However, the costs of disciplinary sanctions at the expense of their career vary among MPs (Slapin et al., 2018) and appear to be lower for moonlighting ones. MPs that, for instance, run their own agricultural company prior to and during their mandate are at no point dependent on their party to earn their living. Additionally, MPs with high outside earnings are not limited to the political sphere when they aim to advance their career. In a survey of German MPs, some of them explicitly named financial independence and autonomy of decision towards their party as the main reasons for continuing their job in addition to their mandate (Kreiner, 2007: 265) – observable in the following MP statement:

My [outside] job was, first of all, a guarantee for my own independence. I have sometimes hinted at, tinkered with the idea of voting against the party line. [Finally], I have voted against the party line for seven or eleven times in roll-call votes. (own translation)

The party group leadership has a toolbox of disciplinary sanctions to threaten with or to actually enforce in order to ensure unity, e.g. withdrawal of MPs from or non-promotion to a committee chair position, which guarantees them privileges like extra salary, media presence, attractive business trips and office space (Bailer, 2018; Patzelt, 2003). However, those sanctions do not harm moonlighting MPs’ career as much as MPs without something to fall back on (Kauder et al., 2017; Van Vonno, 2019). In contrast, MPs without a well-paid sideline job mostly live, in Weberian terms, ‘from’ politics. MP surveys suggest that the often-stated ‘flying splice’ from politics into leadership positions of big companies or organizations is usually subject to former top positions in politics, especially to ministers. Social decline, at least a perceived one, after losing their seat is far from being impossible for some MPs. Not all of them can easily return to their former occupation – if they had one (Byrne and Theakston, 2016; Edinger and Schwarz, 2009; Kreiner, 2007). Thus, MPs without high outside earnings strongly depend on their party as their ‘primary career facilitator before and within parliament’ (Ohmura et al., 2018: 169): on the one hand, to keep their mandate – probably by moving up the party list in times of decreasing party success –, on the other hand to be promoted to or kept in leadership positions. An open dissent with their party would pose them at a higher career-related risk than their moonlighting colleagues.

Since the effectiveness of party discipline is expected to be lower for them, moonlighting MPs are more than others free to represent their own policy beliefs or interests. This effect could even be amplified by the possible influence of ‘business interests’ as a further principal for MPs that earn the greater part of their living outside parliament. However, research on whether outside interests actually exert a measurable influence on MPs’ political behaviour is scarce and has, by now, produced at most indirect evidence for such an effect (Geys and Mause, 2013). Outside interests could imaginably pull MPs away from the mainstream position of their party and, given the lower party discipline, motivate them to break the party line. Such an effect is not compelling, though, since, for instance, MPs with outside earnings from companies are mostly members of market-friendly parties anyway (Hurka et al., 2018a).

To conclude, we expect outside earnings to loosen the pressures of party discipline. This makes it easier for MPs to take a more independent stance from the party group leadership. Regarding their legislative voting behaviour, this leads, all else being equal, to the following empirically testable (and correlative) hypothesis:

MPs with high outside earnings have a higher probability to vote against the party line than those with no or negligible outside earnings.

Study design

The hypothesis is tested against a new dataset of roll-call votes in the German Bundestag during the 18th legislative term (2013–2017). The Bundestag is regarded as one of the most powerful parliaments in Europe (Sieberer, 2011) and shows – compared to other parliamentary systems – a roughly average level of intra-party dissent (Sieberer, 2006). On the one hand, powerful party groups insist on the adherence to unity. On the other hand, especially the salient votes show enough variance of dissenting voting behaviour in need of explanation. Taken together, this makes the Bundestag – called a ‘party group parliament’ (Ismayr, 2012) – a suitable case to test a hypothesis based on the effectiveness of party discipline. The analysis is limited to roll-call votes because they are the only source of recorded individual-level voting behaviour in Germany. However, most of the votes are non-recorded ones (Sieberer et al., 2020; for a thorough discussion of the case selection, see Supplement material 3).

Since the argument offers individual-level propositions and predictors on both the MP and vote level will be included in the model, the unit of observation is one voting decision by one MP. Based on 651 MPs (including resigned and succeeding ones) and 202 roll-call votes, there are (without absences) about 115,000 observations to be analysed. Nevertheless, there are MPs defecting from their party more often and voting decisions being more contentious within a party group than others (for descriptive figures, see Supplement material 4). Thus, the observations are not independent from each other but clustered within MPs and votes since they represent multiple voting decisions for each MP. Consequently, we fit panel regression models with cluster-robust standard errors on the MP level.¹

Logistic regressions are estimated since the dependent variable is dichotomous. It measures whether an MP votes against the party line (value 1) or not (value 0). According to the definition in most studies, defection takes place if an MP differs in his/her voting behaviour from the majority of his/her party group, i.e. in one of the following three settings:

an MP votes ‘yes’ when the party majority votes ‘no’ or ‘abstention’,

an MP votes ‘no’ when the party majority votes ‘yes’ or ‘abstention’,

an MP votes ‘abstention’ when the party majority votes ‘yes’ or ‘no’.

Absences, i.e. when an MP does not vote at all, are not regarded as deviations and coded as missing observations (see Supplement material 5 for a discussion of the treatment of abstentions and absences). Non-whipped votes are excluded from the main analysis because there is no party discipline enforced whose effect should be evaluated here.²

German MPs are allowed to pursue outside activities and receive sideline earnings as long as the exercise of their parliamentary mandate takes centre stage (§ 44a Members of the Bundestag Act). Earnings are published not by the exact amount but a ten-level scheme ranging from level 1 (1,000 to 3,500 Euro) to level 10 (above 250,000 Euro).³ The classification allows the MPs to declare one-time, monthly or yearly earnings. To make those self-declarations comparable, the independent variable is measured by applying a two-step procedure: At first, for each MP all declared earnings are summed up on a yearly basis using the lower bound of the respective level indication (similarly Arnold et al., 2014).⁴ Thus, MPs’ minimum earnings are measured. As Figure 1a shows, in about 80 percent of the MP-years, no outside earnings are reported. The percentage of MPs reaching yearly levels 1 to 3 (and higher) as defined by the Bundestag regulations is rather small (5.9, 1.9 and 2.6 percent, respectively).

Figure 1.

Distribution of raw outside earnings per year.

Figure 1b shows that, on the one hand, the distribution of the yearly earnings is strongly right-skewed, and on the other hand, the earnings’ range is rather large. Hence, the findings could depend on – and could be biased through – the inclusion of few MPs with exceptional earnings. Additionally, raising one’s yearly earnings from 100,000 at another 100,000 Euro might not have the same behavioural effect as raising the income from 0 to 100,000 Euro. Though, as we expect a nonlinear effect, the raw earnings per year are non-linearly transformed in a rather straightforward way.⁵ Previously, it was argued that MPs with high outside earnings defect more frequently because they have something to fall back on that makes party discipline less effective. If earnings exceed their parliamentary salary, MPs obviously do not earn their living from their legislative career. The (monthly weighted) average of an MP’s basic salary (without further allowances) in the 18th legislative term was 107,573.76 Euro per year (Figure 1b, upper dashed line) – which is rather high in a European comparison (Mause, 2014). A correspondingly coded dichotomous variable reveals that 2.2 percent of the MPs have higher earnings outside than inside parliament. For these MPs, the hypothesized effect should appear without any qualification. However, it seems implausible that an MP with slightly lower outside earnings than his/her salary behaves the same way as an MP without any outside earnings. Even earnings half as much as the legislative salary (Figure 1b, lower dashed line) help MPs to make their living even if their political career comes to a sudden end. Therefore, it is likely that those MPs have, although less pronounced, a higher propensity to defect than non-moonlighting ones as well. For the sake of comparison, I code three dummy variables: 1) earnings higher than the legislative salary per year, 2) earnings between 50 and 100 percent and 3) earnings above 0, but less than half of legislative salary. Thus, the baseline category for interpretation consists of MPs without any outside earnings.

Other possible predictors of dissent have to be controlled to ensure that our results do not display spurious correlations. To start with, holding an executive or legislative office is a powerful predictor of party-compliant behaviour (recently for German MPs Bhattacharya and Papageorgiou, 2019; Zittel and Nyhuis, 2019). Concerning electoral pressures, there is a rich literature on a ‘mandate divide’ between directly and list-elected MPs concerning their voting behaviour (e.g. Sieberer, 2010) and other aspects of legislative work, like committee assignments (Stratmann and Baur, 2002, but Manow, 2013). Some scholars identify a behavioural divide between different candidacy modes (Ohmura, 2014; Stoffel, 2014) which is controlled for as well.⁶ Additionally, parliamentary experience might boost an MP’s self-confidence and therefore favour defections (for such a result, see Willumsen and Öhberg, 2017). It is also controlled for a ‘last period effect’ (Bailer and Ohmura, 2018; Willumsen and Goetz, 2017), i.e. the abandonment of a candidacy for the following election. To keep possible election cycle effects (Willumsen and Öhberg, 2017) constant, a dummy variable for votes in the election year 2017 is included. Finally, it is controlled for MPs’ demographic characteristics (age, sex) and party membership. The controls on the MP level are measured on a daily (e.g. offices) or yearly (e.g. seniority) basis.

Since model 2 comprises different types of legislative motions, it includes some additional controls on the vote level. First, an interaction term between the origin of the motion (own/other party group) and the government/opposition status is included since it can be expected that government parties are most united in voting down oppositional motions, but more discordant when government bills (as compromises among the coalition partners) come to the floor. For opposition MPs, the opposite pattern can be expected. Second, a noticeable share of all legislative votes concern foreign deployments of the German armed forces. Christian Democrats and the socialist Left show a rather clear programmatic stance for or against those missions, respectively, whereas Social Democrats and Greens present themselves programmatically less determined, with the likely result that vote defections by Red-Green MPs are more easily tolerated by their respective parties than defections in other legislative votes.

For details about the operationalization and the data sources see Supplement material 6.

Results

Table 1 shows the regression results, presented in logits. Model 1 is restricted to votes on government bills (nearly one third of all individual votes) which usually represent the publicly visible cases of dissent. In these salient decisions, presumably rather strong party discipline is enforced. Consequently, strategic incentives like subordinating to party discipline play a more important role for individual voting behaviour than policy incentives (Bräuninger et al., 2016). If the hypothesis cannot be confirmed in model 1, then it is likely to be rejected in model 2 consisting of all whipped legislative decisions as well since in less salient votes, disciplinary pressure is presumably lower.

Table 1.

Results of the logistic panel regression analysis.

	Model 1 (votes on gov. bills)	Model 2 (all votes)
Outside earnings: above 100% of legislative salary	1.043*** (0.304)	1.005*** (0.266)
Outside earnings: between 50% and 100% of legislative salary	0.786⁺ (0.471)	0.574 (0.446)
Outside earnings: above 0%, below 50% of legislative salary	0.064 (0.178)	0.006 (0.176)
Outside earnings: none	reference category
Outside activities: number (log)	−0.274 (0.212)	−0.415* (0.202)
Parliamentary office	−1.071*** (0.248)	−1.194*** (0.223)
Executive office	−3.463*** (0.734)	−4.106*** (0.748)
Parliamentary experience (years)	0.031* (0.013)	0.037** (0.014)
Direct candidacy only	−0.051 (0.236)	−0.058 (0.245)
List candidacy only	−0.364 (0.305)	−0.468 (0.304)
Direct mandate	0.103 (0.191)	0.150 (0.189)
No candidacy 2017	−0.200 (0.239)	−0.030 (0.134)
Age (years)	−0.002 (0.007)	0.004 (0.007)
Gender (male = 1)	0.030 (0.072)	0.000 (0.078)
Election year (votes in 2017 = 1)	0.579*** (0.116)	0.128⁺ (0.075)
Social Democratic MP	1.000*** (0.180)
Green MP	0.341 (0.240)
Socialist MP	0.059 (0.238)
Christian Democratic MP	reference category
Government MP		−1.877*** (0.167)
Own motion		−4.370*** (0.983)
Own motion × Government MP (interaction term)		5.603*** (0.982)
Bundeswehr deployment		−1.998*** (0.197)
Bundeswehr deployment × Red-Green (interaction term)		2.660*** (0.260)
Red-Green		0.825*** (0.161)
Constant	−4.572*** (0.404)	−4.470*** (0.346)
Wald Chi²	115.21	586.06
N (MP voting decisions)	33,689	115,442

Displayed are logged odds (logits) and cluster-robust standard errors in brackets.

Levels of significance: ***p < 0.001; **p < 0.01; *p < 0.05; ⁺p < 0.10.

For both samples, the coefficient of the key independent variable has the expected positive sign and shows a statistically highly significant effect (p < 0.001): MPs who obtain higher outside earnings than their legislative salary have, ceteris paribus, a higher probability to vote against the party line than MPs without outside earnings. For MPs earning between 50 and 100 percent of their legislative salary outside parliament, a similar positive tendency is shown in the estimates, although the effect is, as expected, weaker and not statistically significant at levels suitable for the large sample. No statistically significant difference in their voting behaviour is found between MPs with comparatively low and those without any outside earnings as well. Thus, our hypothesis is supported by the data. The results can be interpreted as outside earnings having only an effect on legislative behaviour if they are high enough to insulate an MP from the negative career-related consequences of disciplinary measures.

With regard to the control variables, the number of outside activities does only decrease the probability of dissent significantly in the full sample. That a high number of (different) outside activities alone does not go along with more vote defections is not surprising. It usually represents a constellation where an MP has many small arrangements (like paid speeches or book contracts) with a variety of sponsors that lead mostly to one-time payments rather than to a persistent source of income. Like in other studies, MPs with legislative and, even more, executive offices have a significantly lower probability to openly oppose their own party. In contrast, a greater parliamentary experience correlates with more frequent vote defections. In both samples, neither a mandate nor a candidacy divide could be detected. Additionally, MPs do not defect more often in their last term in parliament. In turn, MPs have a higher probability to vote against the party line in votes on (above-average salient) government bills in the year of the 2017 federal election than in the years before. Finally, MPs’ age and gender do not correlate with vote defections. In the full sample, the predictors on the vote level have the expected significant effects which shows that characteristics of the vote (differentiated between party camps) are worth to be included in the model (see also Stecker, 2015).

The logits displayed in Table 1 are not directly interpretable with regard to the substantive effect size. Therefore, Figure 2 shows how the probability of dissent changes when MPs obtain different levels of outside earnings. All controls are set to their observed values as suggested by Hanmer and Kalkan (2013) for models with dummy variables. According to the average marginal effects (AMEs), the probability of casting a vote against the party line regarding government bills rises, ceteris paribus, at about 2.5 percentage points if an MP has outside earnings above his/her legislative salary. Concerning all legislative votes, the AME of high outside earnings is noticeably smaller (probability increase of 1.6 percentage points) but still statistically significant. Since the overall share of votes against one’s own party is rather low, the AMEs show rather substantive effects. In contrast, as the regression coefficients already suggested, outside earnings below an MP’s legislative salary do not have a statistically significant effect on the probability of defection.

Figure 2.

Average marginal effects of outside earnings on dissenting voting behaviour.

Additionally performed robustness checks indicate that the conclusions drawn above are neither contingent upon the measurement of the independent variable, the controls regarding electoral pressures, the estimation strategy, the unit of observation nor upon the inclusion of particular MPs (see Supplement material 7 for full results).

Since the theoretical argument rests on the effectiveness of party discipline, only whipped votes were included in the main regression models. However, in three roll-call votes in the election period under study, MPs were explicitly released from party discipline when deciding on morality policy issues. As a final analytical step, MPs’ voting behaviour in those ‘free votes’ will be analysed. First, t-tests on the average ‘defection rates’ between MPs with high outside earnings and those without do not show statistically significant group differences. Second, high outside earnings are not associated with a significantly higher probability of ‘vote defection’ in multivariate logistic regression analyses of each of the ‘free votes’ (see Supplement material 2 for full results). To conclude, high outside earnings go along with more vote defections when party discipline is enforced whereas moonlighting MPs’ voting behaviour does not differ from their colleagues when this instrument of leadership-induced unity is suspended. Hence, the comparison of whipped and non-whipped votes underscores the discipline-based interpretation of the outside earnings effect derived in the theory section.

Conclusion

The aim of this study was to investigate empirically whether outside earnings are associated with legislative behaviour, precisely with MPs’ propensity to vote against the party line. A brief sketch of the literature revealed that the determinants of dissenting voting behaviour are a rather well-studied topic in legislative research. However, the role of moonlighting therein has neither been thoroughly theorized nor empirically examined yet. Based on insights from Competing Principals Theory, it was argued that, owing to a higher degree of financial and career-related independence, MPs with high outside earnings are less dependent on electoral pressures, but first and foremost less effectively disciplined by their party. Hence, it was hypothesized that MPs with high outside earnings have, everything else being equal, a higher probability to vote against the party line than their colleagues. Methodologically, first, a new dataset containing MPs’ outside activities and earnings as well as other characteristics and their voting behaviour in the 18th legislative term of the German Bundestag (2013–2017) was compiled. Second, in order to account for clustering of legislative dissent both on the MP and the vote level, the hypothesis was tested quantitatively using logistic panel regressions and considering a battery of controls on both levels against more than 115,000 individual voting decisions. Empirically, the results corroborate the hypothesis of a significantly higher probability of dissent for MPs that obtain the bulk of their earnings outside parliament.

This paper’s headline, the slightly adapted proverb ‘Whose bread I don’t eat, his song I don’t (always) sing?’, hints at the causal mechanism that is likely to be at work behind the observed relationship. Nevertheless, despite the theoretical considerations about why outside earnings could loosen the pressures of party discipline and thus could make MPs more free to vote against the party line, the behavioural hypothesis tested in this paper remains a correlative one. Whereas the results reveal an empirical relationship between moonlighting and vote defections, the study design is not able to ultimately verify the presented causal mechanism based on party discipline. Likewise, is does not prove that ‘business interests’ are causally responsible for the observed higher defection rates of moonlighting MPs. On the one hand, the strength of party discipline towards individual MPs could not be directly measured and set into relation with their voting behaviour. On the other hand, it cannot be ruled out definitely that non-observed MP-related factors (e.g. character traits like ego or drive) account for both high outside earnings and high defection rates which could thus bias the results.⁷ Apart from that – and that is why the proverb in the paper’s title ends with a question mark – we must not forget that MPs’ decision making process is more complex than a simple ‘bread-song question’ and votes against the party line remain a rare (yet potentially pivotal) phenomenon, even for MPs with high outside earnings. Voting with, not against one’s own party is the default pattern for most, if not all, MPs.

This study adds a further nuance to the fast-growing literature on the influence of MP characteristics, particularly career-related ones, on their legislative behaviour. In normative terms, its results are ambiguous: On the one hand, the maintenance of party unity is crucial for responsible party government in parliamentary systems since it connects the citizens’ voting decisions to identifiable blocs in parliament which, in turn, the government depends on (Bowler et al., 1999). If this chain of delegation (Müller, 2000) is broken due to high-earning, independent-minded MPs, then the accountability of elected parties and, consecutively, governments could be at stake. On the other hand, voting against unpopular government bills is rather popular in the electorate which appreciates independent-minded and acting MPs instead of ‘lock-step partisan behaviour’ (Campbell et al., 2019: 109). Insofar it remains open for future studies to examine if this valence effect of dissent outweighs the (predominantly) negative image of MPs’ outside activities in a direct comparison. Moreover, while this study has shown that moonlighting correlates with a more independent voting behaviour, it remains to be investigated whether moonlighting, in turn, results in a greater dependence on the MPs’ outside interests. Finally, the findings of this first study on that topic have to be replicated for longer observation periods and in other contexts. That might be other parliamentary or presidential systems like the U.S. where party unity has in part other determinants and, institutionally driven, a lower baseline level (Carey, 2007). Knowing what drives MPs’ legislative behaviour is indispensable for a better understanding and assessment of the functioning of every representative democracy.

Supplemental material

Supplemental Material, sj-pdf-1-ppq-10.1177_1354068820976971 - Whose bread I don’t eat, his song I don’t sing? MPs’ outside earnings and dissenting voting behaviour

Supplemental Material, sj-pdf-1-ppq-10.1177_1354068820976971 for Whose bread I don’t eat, his song I don’t sing? MPs’ outside earnings and dissenting voting behaviour by Philipp Mai in Party Politics

Footnotes

Acknowledgements

Previous versions of this paper were presented at the 2019 ECPR General Conference in Wroclaw and the 2019 Comparative Legislative Studies Standing Group Meeting of the German Political Science Association (DVPW) in Frankfurt/Main. I thank the participants, the two anonymous reviewers of Party Politics as well as Reimut Zohlnhöfer and Georg Wenzelburger for their helpful and constructive comments.

Author note

Philipp Mai is also affiliated to the Department of Social Sciences, TU Kaiserslautern, Germany.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Philipp Mai

Supplemental material

Supplemental material for this article is available online.

Notes

Author biography

Philipp Mai is a researcher and lecturer at the Institute of Political Science, University of Heidelberg and at the Department of Social Sciences, TU Kaiserslautern. His research interests include legislative behaviour and German politics.

References

André

Depauw

Beyens

(2015) Party loyalty and electoral dealignment. Party Politics 21(6): 970–981.

Arnold

Kauder

Potrafke

(2014) Outside earnings, absence, and activity: evidence from German parliamentarians. European Journal of Political Economy 36: 147–157.

Bailer

(2018) To use the whip or not: whether and when party group leaders use disciplinary measures to achieve voting unity. International Political Science Review 39(2): 163–177.

Bailer

Ohmura

(2018) Exploring, maintaining, and disengaging – the three phases of a legislator’s life. Legislative Studies Quarterly 43(3): 493–520.

Bauer-Blaschkowski

Mai

(2019) Von ‘Abweichlern’ und ‘Überzeugungstätern’. Eine Analyse des Abstimmungsverhaltens im 18. Deutschen Bundestag. In: Zohlnhöfer

Saalfeld

(eds) Zwischen Stillstand, Politikwandel und Krisenmanagement: Eine Bilanz der Regierung Merkel 2013–2017. Wiesbaden: Springer VS, pp. 219–256.

Baumann

Debus

Klingelhöfer

(2017) Keeping one’s seat: the competitiveness of MP renomination in mixed-member electoral systems. The Journal of Politics 79(3): 979–994.

Becker

Peichl

Rincke

(2009) Politicians’ outside earnings and electoral competition. Public Choice 140(3–4): 379–394.

Benedetto

Hix

(2007) The rejected, the ejected, and the dejected: explaining government rebels in the 2001–2005 British House of Commons. Comparative Political Studies 40(7): 755–781.

Bergmann

Bailer

Ohmura

, et al. (2016) Namentliche Abstimmungen im Bundestag 1949 bis 2013: Befunde aus einem neuen Datensatz. Zeitschrift für Parlamentsfragen 47(1): 26–50.

10.

Bhattacharya

Papageorgiou

(2019) Are backbenchers fighting back? Intra-party contestation in German parliament debates on the Greek crisis. Parliamentary Affairs 72(2): 425–444.

11.

Bowler

Farrell

Katz

(1999) Party cohesion, party discipline, and parliaments. In: Bowler

Farrell

Katz

(eds) Party Discipline and Parliamentary Government. Columbus, OH: Ohio State University Press, pp. 3–22.

12.

Bräuninger

Müller

Stecker

(2016) Modeling preferences using roll call votes in parliamentary systems. Political Analysis 24(2): 189–210.

13.

Byrne

Theakston

(2016) Leaving the house: the experience of former members of parliament who left the House of Commons in 2010. Parliamentary Affairs 69(3): 686–707.

14.

Campbell

Cowley

(2015) Attitudes to moonlighting politicians: evidence from the United Kingdom. Journal of Experimental Political Science 2(1): 63–72.

15.

Campbell

Cowley

Vivyan

, et al. (2019) Legislator dissent as a valence signal. British Journal of Political Science 49(1): 105–125.

16.

Carey

(2007) Competing principals, political institutions, and party unity in legislative voting. American Journal of Political Science 51(1): 92–107.

17.

Ceron

(2015) Brave rebels stay home: assessing the effect of intra-party ideological heterogeneity and party whip on roll-call votes. Party Politics 21(2): 246–258.

18.

(2018) Parliamentary party loyalty and party family: The missing link? Party Politics 24(2): 209–219.

19.

Gherghina

Sierens

(2019) Prompting legislative agreement and loyalty: What role for intra-party democracy? Parliamentary Affairs 72(2): 387–405.

20.

Coman

(2015) Institutions and vote unity in parliaments: evidence from 33 national chambers. The Journal of Legislative Studies 21(3): 360–389.

21.

Couch

Atkinson

Shughart

(1992) Ethics laws and the outside earnings of politicians: the case of Alabama’s ‘legislator-educators’. Public Choice 73(2): 135–145.

22.

Cowley

Childs

(2003) Too spineless to rebel? New Labour’s women MPs. British Journal of Political Science 33(3): 345–365.

23.

Cox

McCubbins

(1993) Legislative Leviathan. Party Government in the House. Berkeley, CA: University of California Press.

24.

Degner

Leuffen

(2016) Keynes, Friedman, or Monnet? Explaining parliamentary voting behaviour on fiscal aid for Euro area member states. West European Politics 39(6): 1139–1159.

25.

Delius

Koß

Stecker

(2013) ‘Ich erkenne also Fraktionsdisziplin grundsätzlich auch an…’ – Innerfraktioneller Dissens in der SPD-Fraktion der Grossen Koalition 2005 bis 2009. Zeitschrift für Parlamentsfragen 44(3): 546–566.

26.

Edinger

Schwarz

(2009) Leben nach dem Mandat. Eine Studie zu ehemaligen Abgeordneten. Available at: http://www.sowi.ruhr-uni-bochum.de/mam/content/pw1/sfbheft35.pdf (accessed 25 January 2020).

27.

Eggers

Hainmüller

(2009) MPs for sale? Returns to office in postwar British politics. American Political Science Review 103(4): 513–533.

28.

Engler

Dümig

(2017) Political parties and MPs’ morality policy voting behaviour: evidence from Germany. Parliamentary Affairs 70(3): 548–568.

29.

Fedele

Naticchioni

(2015) Moonlighting politicians: motivation matters! German Economic Review 17(2): 127–156.

30.

Gagliarducci

Nannicini

Naticchioni

(2010) Moonlighting politicians. Journal of Public Economics 94(9–10): 688–699.

31.

Geys

(2013) Election cycles in MPs’ outside interests? The UK House of Commons, 2005–2010. Political Studies 61(2): 462–472.

32.

Geys

Mause

(2012) Delegation, accountability and legislator moonlighting: agency problems in Germany. German Politics 21(3): 255–273.

33.

Geys

Mause

(2013) Moonlighting politicians: a survey and research agenda. The Journal of Legislative Studies 19(1): 76–97.

34.

Geys

Mause

(2016) The limits of electoral control: evidence from last-term politicians. Legislative Studies Quarterly 41(4): 873–898.

35.

Gherghina

Chiru

(2014) Determinants of legislative voting loyalty under different electoral systems: evidence from Romania. International Political Science Review 35(5): 523–541.

36.

Grimmer

Powell

(2013) Congressmen in exile: the politics and consequences of involuntary committee removal. The Journal of Politics 75(4): 907–920.

37.

Hanmer

Kalkan

(2013) Behind the curve: clarifying the best approach to calculating predicted probabilities and marginal effects from limited dependent variable models. American Journal of Political Science 57(1): 263–277.

38.

Heuwieser

(2018) Submissive lobby fodder or assertive political actors? Party loyalty of career politicians in the UK House of Commons, 2005–2015. Legislative Studies Quarterly 43(2): 305–341.

39.

Hurka

Daniel

Obholzer

(2018a) Determinants of moonlighting in the European Parliament. The Journal of Legislative Studies 24(1): 127–147.

40.

Hurka

Obholzer

Daniel

(2018b) When time is money: sideline jobs, ancillary income and legislative effort. Journal of European Public Policy 25(5): 651–669.

41.

Ismayr

(2012) Der Deutsche Bundestag. Wiesbaden: Springer VS.

42.

Johnston

Pattie

Rossiter

(1997) Sleaze, constituency and dissent: voting on Nolan in the House of Commons. Area 29(1): 20–33.

43.

Kam

(2011) Party Discipline and Parliamentary Politics. Cambridge: Cambridge University Press.

44.

Kauder

Potrafke

(2019) Conservative politicians and voting on same-sex marriage. German Economic Review 20(4): 600–617.

45.

Kauder

Potrafke

Riem

(2017) Do parties punish MPs for voting against the party line? CESifo Economic Studies 63(3): 317–332.

46.

Kreiner

(2007) Amt auf Zeit. Eine explorative Studie zum beruflichen und politischen Verbleib ehemaliger Bundestagsabgeordneter. Zeitschrift für Parlamentsfragen 38(2): 261–276.

47.

Manow

(2013) Mixed rules, different roles? An analysis of the typical pathways into the Bundestag and of MPs’ parliamentary behaviour. The Journal of Legislative Studies 19(3): 287–308.

48.

Mause

(2014) Self-serving legislators? An analysis of the salary-setting institutions of 27 EU parliaments. Constitutional Political Economy 25(2): 154–176.

49.

Meserve

Pemstein

Bernhard

(2009) Political ambition and legislative behavior in the European Parliament. The Journal of Politics 71(3): 1015–1032.

50.

Mickler

(2018) Who gets what and why? Committee assignments in the German Bundestag. West European Politics 41(2): 517–539.

51.

Müller

(2000) Political parties in parliamentary democracies: making delegation and accountability work. European Journal of Political Research 37(3): 309–333.

52.

Ohmura

(2014) When your name is on the list, it is time to party: the candidacy divide in a mixed-member proportional system. Representation 50(1): 69–82.

53.

Ohmura

Bailer

Meißner

, et al. (2018) Party animals, career changers and other pathways into parliament. West European Politics 41(1): 169–195.

54.

Patzelt

(2003) Party cohesion and party discipline in German parliaments. The Journal of Legislative Studies 9(4): 102–115.

55.

Rehmert

(2020) Candidacy eligibility criteria and party unity. Comparative Political Studies 53(8): 1298–1325.

56.

Rosenson

(2007) Explaining legislators’ position on outside income limits: voting on honoria ceilings in the U.S. Senate, 1981–1983. Public Choice 133(1–2): 111–128.

57.

Rowlands

Vander Wielen

(2019) Time-dependent legislative behavior and election outcomes. Party Politics. Epub ahead of print 22 August 2019. DOI: 10.1177/1354068819868059.

58.

Saalfeld

(1995) Parteisoldaten und Rebellen. Eine Untersuchung zur Geschlossenheit der Fraktionen im Deutschen Bundestag (1949–1990). Opladen: Leske+Budrich.

59.

Schlesinger

(1966) Ambition and Politics. Chicago, IL: Rand McNally.

60.

Shomer

(2016) The electoral environment and legislator dissent. Comparative Politics 48(4): 557–575.

61.

Shomer

(2017) The conditional effect of electoral systems and intraparty candidate selection processes on parties’ behavior. Legislative Studies Quarterly 42(1): 63–96.

62.

Sieberer

(2006) Party unity in parliamentary democracies: a comparative analysis. The Journal of Legislative Studies 12(2): 150–178.

63.

Sieberer

(2010) Behavioral consequences of mixed electoral systems: deviating voting behavior of district and list MPs in the German Bundestag. Electoral Studies 29(3): 484–496.

64.

Sieberer

(2011) The institutional power of Western European parliaments: a multidimensional analysis. West European Politics 34(4): 731–754.

65.

Sieberer

Müller

(2017) Aiming higher: the consequences of progressive ambition among MPs in European parliaments. European Political Science Review 9(1): 27–50.

66.

Sieberer

Ohmura

(2019) Mandate type, electoral safety, and defections from the party line: the conditional mandate divide in the German Bundestag, 1949–2013. Party Politics. Epub ahead of print 21 October 2019. DOI: 10.1177/1354068819881637.

67.

Sieberer

Saalfeld

Ohmura

, et al. (2020) Roll-call votes in the German Bundestag: a new dataset, 1949–2013. British Journal of Political Science 50(3): 1137–1145.

68.

Slapin

Kirkland

Lazzaro

, et al. (2018) Ideology, grandstanding, and strategic party disloyalty in the British parliament. American Political Science Review 112(1): 15–30.

69.

Spiegel Online (2019) Nahles will Kritiker kaltstellen. Available at: https://www.spiegel.de/politik/deutschland/spd-andrea-nahles-will-kritiker-florian-post-kaltstellen-a-1257947.html (accessed 25 January 2020).

70.

Staat

Kuehnhanss

(2017) Outside earnings, electoral systems and legislative effort in the European parliament. Journal of Common Market Studies 55(2): 368–386.

71.

Stecker

(2015) How effects on party unity vary across votes. Party Politics 21(5): 791–802.

72.

Stoffel

(2014) MP behavior in mixed-member electoral systems. Electoral Studies 35: 78–87.

73.

Stratmann

Baur

(2002) Plurality rule, proportional representation, and the German Bundestag: How incentives to pork-barrel differ across electoral systems. American Journal of Political Science 46(3): 506–514.

74.

Strøm

(1997) Rules, reasons and routines: legislative roles in parliamentary democracies. The Journal of Legislative Studies 3(1): 151–174.

75.

Süddeutsche Zeitung (2019) SPD straft Florian Post ab. Available at: https://www.sueddeutsche.de/muenchen/florian-post-spd-bundestag-strafe-1.4379323 (accessed 25 January 2020).

76.

Tavits

(2009) The making of mavericks. Local loyalties and party defection. Comparative Political Studies 42(6): 793–815.

77.

Tavits

(2011) Power within parties: the strength of the local party and MP independence in postcommunist Europe. American Journal of Political Science 55(4): 922–935.

78.

Van Vonno

CMC

(2019) Achieving party unity in the Netherlands: representatives’ sequential decision-making mechanisms at three levels of Dutch government. Party Politics 25(5): 664–678.

79.

Wagner

Glinitzer

Vivyan

(2020) Costly signals: voter responses to parliamentary dissent in Austria, Britain, and Germany. Legislative Studies Quarterly. Epub ahead of print 22 January 2020. DOI: 10.1111/lsq.12274.

80.

Wenzelburger

Fehrenz

(2018) Die Union und die ‘Ehe für alle’. Bestimmungsfaktoren des Abstimmungsverhaltens in der CDU/CSU-Fraktion im Bundestag. Zeitschrift für Parlamentsfragen 49(3): 512–530.

81.

Willumsen

Goetz

(2017) Set free? Impending retirement and legislative behaviour in the UK. Parliamentary Affairs 70(2): 254–279.

82.

Willumsen

Öhberg

(2017) Toe the line, break the whip: explaining floor dissent in parliamentary democracies. West European Politics 40(4): 688–716.

83.

Würfel

(2018) Life after the Bundestag: an analysis of the post-parliamentary careers of German MPs. German Politics 27(3): 295–316.

84.

Zittel

Nyhuis

(2019) Two faces of party unity: roll-call behavior and vote explanations in the German Bundestag. Parliamentary Affairs 72(2): 406–424.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.86 MB