1. Introduction
What are the benefits of legislative activity? Formal and empirical research in legislative politics typically assumes that the main benefits are policy-oriented (Krehbiel Reference Krehbiel1998; Tsebelis Reference Tsebelis2002; Martin Reference Martin2004; Cox and McCubbins Reference Cox and McCubbins2005). In other words, by initiating legislation, Members of Parliament (MPs) move the status quo of the policy closer to their ideal point. Other important benefits are electoral and career benefits: by initiating legislation, MPs send signals about their quality and performance to the electorate and party leaders, who in turn reward them (Bräuninger and Debus Reference Bräuninger and Debus2009; Loewen et al. Reference Loewen, Koop, Settle and Fowler2014; Gratton et al. Reference Gratton, Guiso, Michelacci and Morelli2015; Bräuninger et al. Reference Bräuninger, Debus and Wüst2017). However, previous research has largely neglected another potential motivation for legislative activity: private benefits.
In this research note, we investigate the causal effect of initiating legislation on private benefits. We define private benefits as payments and gifts received from private actors, such as private individuals, organizations and companies. We estimate the causal effect of legislative activity on private benefits by exploiting a natural experiment in New Zealand, where a list of MPs is drawn at regular intervals of time through a lottery system and is given the possibility of introducing bills. Our findings suggest that initiating legislation affects the likelihood of receiving private benefits. We find that this effect is driven by those MPs who propose bills that are successfully passed in parliament. We conclude that private benefits depend on legislative performance.
We make sense of this finding by arguing that private actors allocate benefits to active and successful parliamentarians. Not only does legislative activity signal information on the parliamentarians' performance to the electorate and the party, but also to private actors. If a parliamentarian initiates bills that are adopted, the probability that he/she gets re-elected and becomes an influential policy-maker in future increases (Loewen et al. Reference Loewen, Koop, Settle and Fowler2014; Gratton et al. Reference Gratton, Guiso, Michelacci and Morelli2015; Horiuchi and John Reference Horiuchi and John2016; Williams and Indridason Reference Williams and Indridason2018). Hence, private actors allocate benefits to successful legislators to improve their reputation, access information and potentially influence future policy outcomes.
Political scientists have already studied (private) members' bills lotteries in Canada (Loewen et al. Reference Loewen, Koop, Settle and Fowler2014), the United Kingdom (UK) (Horiuchi and John Reference Horiuchi and John2016) and, more recently, New Zealand itself (Williams and Indridason Reference Williams and Indridason2018). Yet, to our knowledge, this is the first study to investigate the effect of legislative activity on private benefits. The high frequency with which ballots take place and the disclosure of private interests make New Zealand the perfect case where to study this effect.
2. Study design
We first describe the New Zealand ballot and explain why the ballot represents a valid natural experiment. Then, we outline our estimation strategy and briefly discuss the data. A detailed discussion of the New Zealand ballot as well as of the data and the measurement of the variables can be found in the Appendix.
MPs in New Zealand that are not a member of the cabinet can propose members' bills. The parliament usually debates members' bills on every second Wednesday of the month. The Order Paper, which contains the parliament's agenda, includes a limited number of members' bills to be discussed in the first reading. According to Standing Order 281, this limit is currently set to eight bills. If the parliament completes the first reading of a member's bill, a slot in the Order Paper becomes available. Then, the parliament selects a new member's bill via a ballot (i.e., a lottery).Footnote 1 To participate in the ballot, parliamentarians submit a notice of their proposal to the Clerk. The parliamentarians can only submit one proposal to the ballot. The bills that are not drawn are kept for the following ballots, until they are drawn in the lottery, the member withdraws them or the parliament term ends (New Zealand Parliament 2017). The legislative process in New Zealand usually takes three readings, but most members' bills fail in the first reading and do not become law. In addition, the government can veto proposals that have substantial budgetary implications. Detailed information on the members' bill ballot can be found in the Appendix.
We use this natural experiment, namely the random selection of members' bills in the New Zealand parliament, to estimate the causal effect of proposing a bill on private benefits. Our estimation strategy focuses on examining whether being drawn in the ballot increases the chances for the MP to receive private benefits. The New Zealand ballot is especially suited to study the link between legislative activity and private benefits. First, little self-selection takes place. The majority of MPs who do not hold ministerial positions enter regularly the ballot every year: in New Zealand every time roughly 80 MPs enter the ballot.Footnote 2 Second, attrition levels are non-existent, namely in the period under analysis no MP is drawn and then fails to propose legislation. This is the case also in the UK (Horiuchi and John Reference Horiuchi and John2016), but we suspect it is especially the case in New Zealand, where, differently from the UK (House of Commons 20102016), MPs need to present a complete draft to enter the ballot (since 2011) (New Zealand Parliament 20092016). Third, differently from similar procedures in place in other countries, such as the UK and Canada, the New Zealand ballot takes place more than once a year. Hence, our data exhibits more observations and higher variation.
Our study uses publicly available data on the members' bills in New Zealand. We focus on the period from 2009 until 2016, which includes 47 ballots. On average each ballot selects two bills from 80 entries. Our dataset includes 3056 entries in the ballot and 117 drawn members' bills. These drawn bills were proposed by MPs from the Labour party (51), the National party (27), the Greens (20), New Zealand First (11), the Maori Party (5), ACT (2) and Mana (1). We gathered data on whether the bill was drawn in the ballot; whether the bill passed and, if not, at which stage the bill was rejected; the dates of bill proposal and adoption or rejection. Out of the 117 bills, 16 were formally adopted. The topics of the bills vary substantially and include areas such as finance, economic affairs, interior affair, environment, health and education.
We rely on unique data on the private benefits of New Zealand MPs. We measure private benefits by using the data from the Register of Pecuniary and Other Specified Interests of Members of Parliament, where MPs have to provide information on financial benefits outside the parliament. We focus on two types of benefits: gifts and payments for activities. Examples of gifts are trips and hospitality offered to MPs by companies and associations and examples of payments for activities are consulting fees.Footnote 3 In the main analysis, we use an ordinal variable which takes value: 0 if the MP does not receive any gifts or payments in the same year of the ballot; value 1 if the MP receives either gifts or payments in that year; value 2 if the MP receives both gifts and payments in that year. In the Appendix, we use alternative measures of private benefits.
The regression analysis controls for a series of confounding factors: whether the MP acts as a chair of a parliamentary committee, as a measure for seniority; the ministerial experience of the MP, as a measure of the legislative experience; whether the MP was elected through the electorate or party vote (Williams and Indridason Reference Williams and Indridason2018), as a measure of the MP's electoral incentives. The Appendix discusses in more detail data and measurement issues and includes descriptive statistics (see Table A1--A3 in the Appendix).
3. Results
We analyze the data in two steps. First, we investigate whether being drawn in the ballot is associated with receiving private benefits. Then, we examine whether the legislative outcome influences private benefits. Before analyzing the data, we run a balance test to confirm that the ballot selects bill proposals randomly (see Table A4 in the Appendix).
Table 1 shows the results of the ordered logistic regression model, which includes control variables. We analyze the relationship between a bill being drawn and the private benefits of the MP that proposed the bill. We control for the MP's party, the legislative period in which the ballot takes place and other control variables, such as the MP's seniority (whether he/she is a committee chair), whether he/she was elected via list or constituency and his/her previous cabinet experience. The full regression output can be found in Table A5 in the Appendix. We find a statistically significant and positive relationship between being drawn in the ballot and receiving private benefits. The probability of receiving either gifts or payments increases by 7 percent after being drawn. The Appendix includes robustness checks.
Table 1. The Effect of Bill Drawn on Private Benefits (Same Year - Ordinal)
Clustered standard errors in parentheses ***p < 0.01, **p < 0.05, *p < 0.1
In Models 4 and 5, we examine the effect of initiating legislation on different subsets of the data. We first remove those proposals which are drawn and rejected and then we remove those which are drawn and adopted. We find a statistically significant difference between those MPs who successfully passed a bill and those MPs whose bill was not drawn in the first place. Yet, no difference is present between those MPs whose bill was drawn but then rejected and those MPs whose bill was not drawn in the first place. We run a series of robustness checks in the Appendix.
Next, we test whether the success of the legislative proposal influences private benefits. We want to understand whether MPs are rewarded for proposing legislation that is adopted in the legislative process. Hence, we focus exclusively on the drawn bills and tests whether private benefits depend on whether the bill passes or not. The legislative success is likely to be correlated with a number of variables related to the seniority and experience of parliamentarians. We address this potential omitted variable bias by including the control variables discussed in the previous section.
Table 2 examines the effect of passing a bill on private benefits (Table A6 shows the full outputs of the regression model). The results show that legislative success affects the private benefits of MPs. This effect is robust to different specifications and measures of the outcome variable. If we focus on the successful bills, the predicted probability of receiving either gifts or payments increases by 36 percent. In the Appendix we run some robustness tests.
Table 2. The Effect of Bill Passed on Private Benefits (Same Year – Ordinal)
Clustered standard errors in parentheses ***p < 0.01, **p < 0.05, *p < 0.1
Existing theoretical work on the role of signaling in legislative politics provides a rationale for these findings (Caselli and Morelli Reference Caselli and Morelli2004; Mattozzi and Merlo Reference Mattozzi and Merlo2008; Gratton et al. Reference Gratton, Guiso, Michelacci and Morelli2015; Buisseret and Bernhardt Reference Buisseret and Bernhardt2017). Private actors have an incentive to allocate benefits to those MPs who are better able to influence policy-making. The private market updates its beliefs on the capacity of an MP to influence policy-making in future in two ways: by looking at his/her re-election probability and by looking at his/her probability to be a key decision-maker in the future. In other words, the market allocates private benefits to those MPs who are more likely to stay in the parliament and to be key players in the foreseeable future.
The probability of re-election depends, among other things, on whether the MP proposes legislation. Williams and Indridason (Reference Williams and Indridason2018) provide evidence for this in New Zealand. We argue that, instead, the market updates the probability of an MP to be a key decision-maker in the future based on his/her capacity to successfully pass legislation. Those MPs able to pass legislation in the current parliament exhibit the quality and the support from their parties needed to take an active role in the parliament in future. As a result, where an MP's bill is drawn and is successfully passed, both updates are positive: that MP will be more likely to be re-elected and will be more likely to play a central role in the parliament in future. That is why we find a statistically significant difference between those MPs who successfully passed a bill and those MPs whose bill was not drawn in the first instance.Footnote 4
Instead, if an MP's bill is drawn but is rejected by the parliament, his/her probability of being re-elected will increase, but his/her probability to influence policy-making in future will be lower. In other words, the two updates go in the opposite direction, balancing each other out. That is why we find no difference between those MPs whose bill was drawn but then rejected and those MPs whose bill was not drawn in the first place.
In conclusion, our findings suggest that private actors target active and successful parliamentarians. These parliamentarians are both more likely to be re-elected and more likely to be influential in the future. In contrast, parliamentarians that unsuccessfully propose a bill have a higher re-election probability, but are not necessarily more likely to become an influential decision-maker. Hence, private actors have fewer incentives to allocate benefits to this group.
4. Conclusion
By proposing (successful) legislation, MPs can increase their approval rate in their electorate (Williams and Indridason Reference Williams and Indridason2018), their chances of being re-elected (Loewen et al. Reference Loewen, Koop, Settle and Fowler2014) and making a career inside the parliament (Horiuchi and John Reference Horiuchi and John2016). In this study, we find that MPs also receive immediate private benefits, as a result for being more active and successful in the parliament. These benefits depend on whether the MP manages to push the bill through. Hence, we conclude that in parliament there are clear rewards in place for MPs to be more active.
These findings suggest that the market allocates private benefits to those MPs who are more likely to influence policy-making in future, which depends on the MP's probability to be re-elected and to be key decision-makers in the future. As a consequence, the private market rewards those MPs who propose bills that are successfully passed, because they are more likely to be re-elected and to be key policy-makers in future.
Future research may investigate the effect of legislative activity on parliamentarians' private benefits by using data on the monetary value of benefits. We believe it would be important to understand whether these benefits change the behavior of politicians, the quality of legislation or even encourage the entry of different type of politicians.
Supplementary Material
The supplementary material for this article can be found at https://doi.org/10.1017/psrm.2018.66
Acknowledgments
This work was supported by the European Research Council (advanced grant 694583).
